Modelling Survival Data in Medical Research PDF

Statistics
Third
Edition Texts in Statistical Science
Modelling Survival Data in Medical Research describes the mod-
elling approach to the analysis of survival data using a wide range of
Modelling
Modelling Survival Data in Medical Research

examples from biomedical research.
Well known for its nontechnical style, this third edition contains new
Survival Data in
chapters on frailty models and their applications, competing risks,
non-proportional hazards, and dependent censoring. It also de-
scribes techniques for modelling the occurrence of multiple events
and event history analysis.
Earlier chapters are now expanded to include new material on a num-
ber of topics, including measures of predictive ability and flexible
parametric models. Many new data sets and examples are included
Medical Research
to illustrate how these techniques are used in modelling survival data.
Features
Third Edition
• Presents an accessible introduction to statistical methods for
handling survival data
• Includes modern statistical techniques for survival analysis and
key references
• Contains real data examples with many new data sets
• Provides additional data sets that can be used for coursework
Bibliographic notes and suggestions for further reading are provided
at the end of each chapter.
Additional data sets to obtain a fuller appreciation of the methodology,
or to be used as student exercises, are provided in the appendix.
All data sets used in this book are also available in electronic format
online.
This book is an invaluable resource for statisticians in the pharma-
ceutical industry, professionals in medical research institutes, scien-
tists and clinicians who are analysing their own data, and students
David Collett
Collett
taking undergraduate or postgraduate courses in survival analysis.
K12670
w w w. c rc p r e s s . c o m
Modelling
Survival Data in
Medical Research
Third Edition
CHAPMAN & HALL/CRC
Texts in Statistical Science Series
Series Editors
Francesca Dominici, Harvard School of Public Health, USA
Julian J. Faraway, University of Bath, UK
Martin Tanner, Northwestern University, USA
Jim Zidek, University of British Columbia, Canada
Statistical Theory: A Concise Introduction Modelling Binary Data, Second Edition

F. Abramovich and Y. Ritov D. Collett
Practical Multivariate Analysis, Fifth Edition Modelling Survival Data in Medical Research,
A. Afifi, S. May, and V.A. Clark Third Edition
Practical Statistics for Medical Research D. Collett
D.G. Altman Introduction to Statistical Methods for
Interpreting Data: A First Course Clinical Trials
in Statistics T.D. Cook and D.L. DeMets
A.J.B. Anderson Applied Statistics: Principles and Examples
Introduction to Probability with R D.R. Cox and E.J. Snell
K. Baclawski Multivariate Survival Analysis and Competing
Linear Algebra and Matrix Analysis for Risks
Statistics M. Crowder
S. Banerjee and A. Roy Statistical Analysis of Reliability Data
Analysis of Categorical Data with R M.J. Crowder, A.C. Kimber,
C. R. Bilder and T. M. Loughin T.J. Sweeting, and R.L. Smith
Statistical Methods for SPC and TQM An Introduction to Generalized
D. Bissell Linear Models, Third Edition
A.J. Dobson and A.G. Barnett
Introduction to Probability
J. K. Blitzstein and J. Hwang Nonlinear Time Series: Theory, Methods, and
Applications with R Examples
Bayesian Methods for Data Analysis,
R. Douc, E. Moulines, and D.S. Stoffer
Third Edition
B.P. Carlin and T.A. Louis Introduction to Optimization Methods and
Their Applications in Statistics
Second Edition
B.S. Everitt
R. Caulcutt
Extending the Linear Model with R:
The Analysis of Time Series: An Introduction,
Generalized Linear, Mixed Effects and
Sixth Edition
Nonparametric Regression Models
C. Chatfield
J.J. Faraway
Introduction to Multivariate Analysis
Linear Models with R, Second Edition
C. Chatfield and A.J. Collins
J.J. Faraway
Problem Solving: A Statistician’s Guide,
A Course in Large Sample Theory
Second Edition
T.S. Ferguson
C. Chatfield
Multivariate Statistics: A Practical
Statistics for Technology: A Course in Applied
Approach
Statistics, Third Edition
B. Flury and H. Riedwyl
C. Chatfield
Readings in Decision Analysis
Bayesian Ideas and Data Analysis: An
S. French
Introduction for Scientists and Statisticians
R. Christensen, W. Johnson, A. Branscum,
and T.E. Hanson
Markov Chain Monte Carlo: Design and Analysis of Experiments with SAS
Stochastic Simulation for Bayesian Inference, J. Lawson
Second Edition A Course in Categorical Data Analysis
D. Gamerman and H.F. Lopes T. Leonard
Bayesian Data Analysis, Third Edition Statistics for Accountants
A. Gelman, J.B. Carlin, H.S. Stern, D.B. Dunson, S. Letchford
A. Vehtari, and D.B. Rubin
Introduction to the Theory of Statistical
Multivariate Analysis of Variance and Inference
Repeated Measures: A Practical Approach for H. Liero and S. Zwanzig
Behavioural Scientists
D.J. Hand and C.C. Taylor Statistical Theory, Fourth Edition
B.W. Lindgren
Practical Longitudinal Data Analysis
D.J. Hand and M. Crowder Stationary Stochastic Processes: Theory and
Applications
Logistic Regression Models G. Lindgren
J.M. Hilbe
The BUGS Book: A Practical Introduction to
Richly Parameterized Linear Models: Bayesian Analysis
Additive, Time Series, and Spatial Models D. Lunn, C. Jackson, N. Best, A. Thomas, and
Using Random Effects D. Spiegelhalter
J.S. Hodges
Introduction to General and Generalized
Statistics for Epidemiology Linear Models
N.P. Jewell H. Madsen and P. Thyregod
Stochastic Processes: An Introduction, Time Series Analysis
Second Edition H. Madsen
P.W. Jones and P. Smith
Pólya Urn Models
The Theory of Linear Models H. Mahmoud
B. Jørgensen
Randomization, Bootstrap and Monte Carlo
Principles of Uncertainty Methods in Biology, Third Edition
J.B. Kadane B.F.J. Manly
Graphics for Statistics and Data Analysis with R Introduction to Randomized Controlled
K.J. Keen Clinical Trials, Second Edition
Mathematical Statistics J.N.S. Matthews
K. Knight Statistical Methods in Agriculture and
Introduction to Multivariate Analysis: Experimental Biology, Second Edition
Linear and Nonlinear Modeling R. Mead, R.N. Curnow, and A.M. Hasted
S. Konishi Statistics in Engineering: A Practical Approach
Nonparametric Methods in Statistics with SAS A.V. Metcalfe
Applications Statistical Inference: An Integrated Approach,
O. Korosteleva Second Edition
Modeling and Analysis of Stochastic Systems, H. S. Migon, D. Gamerman, and
Second Edition F. Louzada
V.G. Kulkarni Beyond ANOVA: Basics of Applied Statistics
Exercises and Solutions in Biostatistical Theory R.G. Miller, Jr.
L.L. Kupper, B.H. Neelon, and S.M. O’Brien
A Primer on Linear Models
Exercises and Solutions in Statistical Theory J.F. Monahan
L.L. Kupper, B.H. Neelon, and S.M. O’Brien
Applied Stochastic Modelling, Second Edition
Design and Analysis of Experiments with R B.J.T. Morgan
J. Lawson
Elements of Simulation
B.J.T. Morgan
Probability: Methods and Measurement Applied Nonparametric Statistical Methods,
A. O’Hagan Fourth Edition
Introduction to Statistical Limit Theory P. Sprent and N.C. Smeeton
A.M. Polansky Data Driven Statistical Methods
Applied Bayesian Forecasting and Time Series P. Sprent
Analysis Generalized Linear Mixed Models:
A. Pole, M. West, and J. Harrison Modern Concepts, Methods and Applications
Statistics in Research and Development, W. W. Stroup
Time Series: Modeling, Computation, and Survival Analysis Using S: Analysis of
Inference Time-to-Event Data
R. Prado and M. West M. Tableman and J.S. Kim
Introduction to Statistical Process Control Applied Categorical and Count Data Analysis
P. Qiu W. Tang, H. He, and X.M. Tu
Sampling Methodologies with Applications Elementary Applications of Probability Theory,
P.S.R.S. Rao Second Edition
A First Course in Linear Model Theory H.C. Tuckwell
N. Ravishanker and D.K. Dey Introduction to Statistical Inference and Its
Essential Statistics, Fourth Edition Applications with R
D.A.G. Rees M.W. Trosset
Stochastic Modeling and Mathematical Understanding Advanced Statistical Methods

Statistics: A Text for Statisticians and P.H. Westfall and K.S.S. Henning
Quantitative Scientists Statistical Process Control: Theory and
F.J. Samaniego Practice, Third Edition
Statistical Methods for Spatial Data Analysis G.B. Wetherill and D.W. Brown
O. Schabenberger and C.A. Gotway Generalized Additive Models:
Bayesian Networks: With Examples in R An Introduction with R
M. Scutari and J.-B. Denis S. Wood
Large Sample Methods in Statistics Epidemiology: Study Design and

P.K. Sen and J. da Motta Singer Data Analysis, Third Edition
M. Woodward
Decision Analysis: A Bayesian Approach
J.Q. Smith Practical Data Analysis for Designed
Experiments
Analysis of Failure and Survival Data B.S. Yandell
P. J. Smith
Applied Statistics: Handbook of GENSTAT
Analyses
E.J. Snell and H. Simpson
Texts in Statistical Science
Modelling
Survival Data in
Medical Research
Third Edition
David Collett
NHS Blood and Transplant
Bristol, UK
First edition published in 1994 by Chapman and Hall.
Second edition published in 2003 by Chapman and Hall/CRC.
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
© 2015 by Taylor & Francis Group, LLC

CRC Press is an imprint of Taylor & Francis Group, an Informa business
No claim to original U.S. Government works

Version Date: 20150505
International Standard Book Number-13: 978-1-4987-3169-0 (eBook - PDF)
This book contains information obtained from authentic and highly regarded sources. Reasonable efforts
have been made to publish reliable data and information, but the author and publisher cannot assume
responsibility for the validity of all materials or the consequences of their use. The authors and publishers
have attempted to trace the copyright holders of all material reproduced in this publication and apologize to
copyright holders if permission to publish in this form has not been obtained. If any copyright material has
not been acknowledged please write and let us know so we may rectify in any future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmit-
ted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented,
including photocopying, microfilming, and recording, or in any information storage or retrieval system,
without written permission from the publishers.
For permission to photocopy or use material electronically from this work, please access www.copyright.
com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood
Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and
registration for a variety of users. For organizations that have been granted a photocopy license by the CCC,
a separate system of payment has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used
only for identification and explanation without intent to infringe.
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com
and the CRC Press Web site at

http://www.crcpress.com
Contents
Preface xv
1 Survival analysis 1
1.1 Special features of survival data 1
1.1.1 Censoring 2
1.1.2 Independent censoring 3
1.1.3 Study time and patient time 3
1.2 Some examples 5
1.3 Survivor, hazard and cumulative hazard functions 10
1.3.1 The survivor function 10
1.3.2 The hazard function 12
1.3.3 The cumulative hazard function 13
1.4 Computer software for survival analysis 14
1.5 Further reading 15
2 Some non-parametric procedures 17

2.1 Estimating the survivor function 17
2.1.1 Life-table estimate of the survivor function 19
2.1.2 Kaplan-Meier estimate of the survivor function 21
2.1.3 Nelson-Aalen estimate of the survivor function 24
2.2 Standard error of the estimated survivor function 25
2.2.1 Standard error of the Kaplan-Meier estimate 26
2.2.2 Standard error of other estimates 27
2.2.3 Confidence intervals for values of the survivor function 28
2.3 Estimating the hazard function 31
2.3.1 Life-table estimate of the hazard function 31
2.3.2 Kaplan-Meier type estimate 32
2.3.3 Estimating the cumulative hazard function 35
2.4 Estimating the median and percentiles of survival times 36
2.5 Confidence intervals for the median and percentiles 38
2.6 Comparison of two groups of survival data 40
2.6.1 Hypothesis testing 41
2.6.2 The log-rank test 44
2.6.3 The Wilcoxon test 48
2.6.4 Comparison of the log-rank and Wilcoxon tests 49
vii
viii CONTENTS
2.7 Comparison of three or more groups of survival data 50
2.8 Stratified tests 52
2.9 Log-rank test for trend 54
3 The Cox regression model 57

3.1 Modelling the hazard function 57
3.1.1 A model for the comparison of two groups 58
3.1.2 The general proportional hazards model 59
3.2 The linear component of the model 60
3.2.1 Including a variate 61
3.2.2 Including a factor 61
3.2.3 Including an interaction 62
3.2.4 Including a mixed term 63
3.3 Fitting the Cox regression model 65
3.3.1 Likelihood function for the model 67
3.3.2 Treatment of ties 69
3.3.3 The Newton-Raphson procedure 71
3.4 Confidence intervals and hypothesis tests 72
3.4.1 Confidence intervals for hazard ratios 73
3.4.2 Two examples 73
3.5 Comparing alternative models 76
3.5.1 The statistic −2 log L̂ 77
3.5.2 Comparing nested models 78
3.6 Strategy for model selection 83
3.6.1 Variable selection procedures 84
3.7 Variable selection using the lasso 90
3.7.1 The lasso in Cox regression modelling 91
3.7.2 Data preparation 92
3.8 Non-linear terms 95
3.8.1 Testing for non-linearity 96
3.8.2 Modelling non-linearity 97
3.8.3 Fractional polynomials 98
3.9 Interpretation of parameter estimates 99
3.9.1 Models with a variate 99
3.9.2 Models with a factor 100
3.9.3 Models with combinations of terms 104
3.10 Estimating the hazard and survivor functions 107
3.10.1 The special case of no covariates 110
3.10.2 Some approximations to estimates of baseline functions 110
3.11 Risk adjusted survivor function 116
3.11.1 Risk adjusted survivor function for groups of individuals 117
3.12 Explained variation in the Cox regression model 120
3.12.1 Measures of explained variation 122
3.12.2 Measures of predictive ability 123
CONTENTS ix
3.12.3 Model validation 124
3.13 Proportional hazards and the log-rank test 125
4 Model checking in the Cox regression model 131

4.1 Residuals for the Cox regression model 131
4.1.1 Cox-Snell residuals 132
4.1.2 Modified Cox-Snell residuals 133
4.1.3 Martingale residuals 135
4.1.4 Deviance residuals 136
4.1.5 Schoenfeld residuals 137
4.1.6 Score residuals 138
4.2 Assessment of model fit 142
4.2.1 Plots based on the Cox-Snell residuals 142
4.2.2 Plots based on the martingale and deviance residuals 145
4.2.3 Checking the functional form of covariates 147
4.3 Identification of influential observations 152
4.3.1 Influence of observations on a parameter estimate 153
4.3.2 Influence of observations on the set of parameter
estimates 155
4.3.3 Treatment of influential observations 158
4.4 Testing the assumption of proportional hazards 160
4.4.1 The log-cumulative hazard plot 161
4.4.2 Use of Schoenfeld residuals 163
4.4.3 Tests for non-proportional hazards 164
4.4.4 Adding a time-dependent variable 166
4.5 Recommendations 168
5 Parametric proportional hazards models 171

5.1 Models for the hazard function 171
5.1.1 The exponential distribution 172
5.1.2 The Weibull distribution 173
5.2 Assessing the suitability of a parametric model 177
5.3 Fitting a parametric model to a single sample 178
5.3.1 Likelihood function for randomly censored data 180
5.4 Fitting exponential and Weibull models 181
5.4.1 Fitting the exponential distribution 182
5.4.2 Fitting the Weibull distribution 186
5.4.3 Standard error of a percentile of the Weibull distribution 188
5.5 A model for the comparison of two groups 192
5.5.1 The log-cumulative hazard plot 192
5.5.2 Fitting the model 194
5.6 The Weibull proportional hazards model 199
x CONTENTS
5.6.2 Standard error of a percentile in the Weibull model 201
5.6.3 Log-linear form of the model 203
5.6.4 Exploratory analyses 205
5.7 Comparing alternative Weibull models 208
5.8 Explained variation in the Weibull model 215
5.9 The Gompertz proportional hazards model 216
5.10 Model choice 218
6 Accelerated failure time and other parametric models 221

6.1 Probability distributions for survival data 221
6.1.1 The log-logistic distribution 222
6.1.2 The lognormal distribution 222
6.1.3 The gamma distribution 224
6.1.4 The inverse Gaussian distribution 225
6.2 Exploratory analyses 225
6.3 Accelerated failure model for two groups 227
6.3.1 Comparison with the proportional hazards model 228
6.3.2 The percentile-percentile plot 231
6.4 The general accelerated failure time model 232
6.4.1 Log-linear form of the accelerated failure time model 234
6.5 Parametric accelerated failure time models 236
6.5.1 The Weibull accelerated failure time model 236
6.5.2 The log-logistic accelerated failure time model 239
6.5.3 The lognormal accelerated failure time model 240
6.5.4 Summary 241
6.6 Fitting and comparing accelerated failure time models 243
6.7 The proportional odds model 250
6.7.1 The log-logistic proportional odds model 253
6.8 Some other distributions for survival data 255
6.9 Flexible parametric models 256
6.9.1 The Royston and Parmar model 259
6.9.2 Number and position of the knots 262
6.9.4 Proportional odds models 266
6.10 Modelling cure rates 268
6.11 Effect of covariate adjustment 270
7 Model checking in parametric models 275

7.1 Residuals for parametric models 275
7.1.1 Standardised residuals 275
7.1.2 Cox-Snell residuals 276
7.1.3 Martingale residuals 277
7.1.4 Deviance residuals 277
CONTENTS xi
7.1.5 Score residuals 277
7.2 Residuals for particular parametric models 278
7.2.1 Weibull distribution 279
7.2.2 Log-logistic distribution 279
7.2.3 Lognormal distribution 280
7.2.4 Analysis of residuals 280
7.3 Comparing observed and fitted survivor functions 284
7.4 Identification of influential observations 287
7.4.1 Influence of observations on a parameter estimate 287
7.4.2 Influence of observations on the set of parameter
estimates 288
7.5 Testing proportional hazards in the Weibull model 291
8 Time-dependent variables 295

8.1 Types of time-dependent variables 295
8.2 A model with time-dependent variables 296
8.2.1 Fitting the Cox model 297
8.2.2 Estimation of baseline hazard and survivor functions 300
8.3 Model comparison and validation 302
8.3.1 Comparison of treatments 303
8.3.2 Assessing model adequacy 303
8.4 Some applications of time-dependent variables 304
8.5 Three examples 306
8.6 Counting process format 316
9 Interval-censored survival data 319

9.1 Modelling interval-censored survival data 319
9.2 Modelling the recurrence probability in the follow-up period 322
9.3 Modelling the recurrence probability at different times 325
9.4 Arbitrarily interval-censored survival data 332
9.4.1 Modelling arbitrarily interval-censored data 332
9.4.2 Proportional hazards model for the survivor function 334
9.4.3 Choice of the step times 337
9.5 Parametric models for interval-censored data 342
9.6 Discussion 343
10 Frailty models 345

10.1 Introduction to frailty 345
10.1.1 Random effects 346
10.1.2 Individual frailty 346
10.1.3 Shared frailty 347
10.2 Modelling individual frailty 348
xii CONTENTS
10.2.1 Frailty distributions 349
10.2.2 Observable survivor and hazard functions 351
10.3 The gamma frailty distribution 352
10.3.1 Impact of frailty on an observable hazard function 353
10.3.2 Impact of frailty on an observable hazard ratio 354
10.4 Fitting parametric frailty models 356
10.4.1 Gamma frailty 357
10.5 Fitting semi-parametric frailty models 363
10.5.1 Lognormal frailty effects 363
10.5.2 Gamma frailty effects 365
10.6 Comparing models with frailty 366
10.6.1 Testing for the presence of frailty 366
10.7 The shared frailty model 372
10.7.1 Fitting the shared frailty model 373
10.7.2 Comparing shared frailty models 374
10.8 Some other aspects of frailty modelling 377
10.8.1 Model checking 377
10.8.2 Correlated frailty models 378
10.8.3 Dependence measures 378
10.8.4 Numerical problems in model fitting 378
11 Non-proportional hazards and institutional comparisons 381

11.1 Non-proportional hazards 381
11.2 Stratified proportional hazards models 383
11.2.1 Non-proportional hazards between treatments 385
11.3 Restricted mean survival 389
11.3.1 Use of pseudo-values 391
11.4 Institutional comparisons 393
11.4.1 Interval estimate for the RAFR 397
11.4.2 Use of the Poisson regression model 400
11.4.3 Random institution effects 402
12 Competing risks 405

12.1 Introduction to competing risks 405
12.2 Summarising competing risks data 406
12.2.1 Kaplan-Meier estimate of survivor function 407
12.3 Hazard and cumulative incidence functions 409
12.3.1 Cause-specific hazard function 409
12.3.2 Cause-specific cumulative incidence function 410
12.3.3 Some other functions of interest 413
12.4 Modelling cause-specific hazards 414
12.4.1 Likelihood functions for competing risks models 415
12.4.2 Parametric models for cumulative incidence functions 418
CONTENTS xiii
12.5 Modelling cause-specific incidence 419
12.5.1 The Fine and Gray competing risks model 419
12.6 Model checking 422
13 Multiple events and event history modelling 429

13.1 Introduction to counting processes 429
13.1.1 Modelling the intensity function 430
13.1.2 Survival data as a counting process 431
13.1.3 Survival data in the counting process format 433
13.1.4 Robust estimation of the variance-covariance matrix 434
13.2 Modelling recurrent event data 435
13.2.1 The Anderson and Gill model 436
13.2.2 The Prentice, Williams and Peterson model 437
13.3 Multiple events 443
13.3.1 The Wei, Lin and Weissfeld model 443
13.4 Event history analysis 448
13.4.1 Models for event history analysis 449
14 Dependent censoring 457

14.1 Identifying dependent censoring 457
14.2 Sensitivity to dependent censoring 458
14.2.1 A sensitivity analysis 459
14.2.2 Impact of dependent censoring 461
14.3 Modelling with dependent censoring 463
14.3.1 Cox regression model with dependent censoring 464
15 Sample size requirements for a survival study 471

15.1 Distinguishing between two treatment groups 471
15.2 Calculating the required number of deaths 472
15.2.1 Derivation of the required number of deaths 474
15.3 Calculating the required number of patients 479
15.3.1 Derivation of the required number of patients 480
15.3.2 An approximate procedure 483
A Maximum likelihood estimation 487

A.1 Inference about a single unknown parameter 487
A.2 Inference about a vector of unknown parameters 489
xiv CONTENTS
B Additional data sets 491
B.1 Chronic active hepatitis 491
B.2 Recurrence of bladder cancer 492
B.3 Survival of black ducks 492
B.4 Bone marrow transplantation 495
B.5 Chronic granulomatous disease 495
Bibliography 499
Index of Examples 521

Preface
This book describes and illustrates the modelling approach to the analysis
of survival data, using a wide range of examples from biomedical research.
My experience in presenting many lectures and courses on this subject, at
both introductory and advanced levels, as well as in providing advice on the
analysis of survival data, has had a big influence on its content. The result
is a comprehensive practical account of survival analysis at an intermediate
level, which I hope will continue to meet the needs of statisticians in the phar-
maceutical industry or medical research institutes, scientists and clinicians
who are analysing their own data, and students following undergraduate or
postgraduate courses in survival analysis.
In preparing this new edition, my aim has been to incorporate extensions
to the basic models that dramatically increase their scope, while updating
the text to take account of the wider availability of computer software for
implementing these techniques. This edition therefore contains new chapters
covering frailty models, non-proportional hazards, competing risks, multiple
events, event history analysis and dependent censoring. Additional material
on variable selection, non-linear models, measures of explained variation and
flexible parametric models has also been included in earlier chapters.
The main part of the book is formed by Chapters 1 to 7. After an intro-
duction to survival analysis in Chapter 1, Chapter 2 describes methods for
summarising survival data, and for comparing two or more groups of survival
times. The modelling approach is introduced in Chapter 3, where the Cox
regression model is presented in detail. This is followed by a chapter that de-
scribes methods for checking the adequacy of a fitted model. Parametric pro-
portional hazards models are covered in Chapter 5, with an emphasis on the
Weibull model for survival data. Chapter 6 describes parametric accelerated
failure time models, including a detailed account of their log-linear represen-
tation that is used in most computer software packages. Flexible parametric
models are also described and illustrated in this chapter, while model-checking
diagnostics for parametric models are presented in Chapter 7.
The remaining chapters describe a number of extensions to the basic mod-
els. The use of time-dependent variables is covered in Chapter 8, and the
analysis of interval-censored data is considered in Chapter 9. Frailty mod-
els that allow differences between individuals, or groups of individuals, to be
modelled using random effects, are described in Chapter 10. Chapter 11 sum-
marises techniques that can be used when the assumption of proportional
xv
xvi PREFACE
hazards cannot be made, and shows how these models can be used in compar-
ing survival outcomes across a number of institutions. Competing risk mod-
els that accommodate different causes of death are presented in Chapter 12,
while extensions of the Cox regression model to cope with multiple events of
the same or different types, including event history analysis, are described in
Chapter 13. Chapter 14 summarises methods for analysing data when there
is dependent censoring, and Chapter 15 shows how to determine the sample
size requirements of a study where the outcome variable is a survival time.
All of the techniques that have been described can be implemented in
many software packages for survival analysis, including the freeware package
R. However, sufficient methodological details have been included to convey a
sound understanding of the techniques and the assumptions on which they are
based, and to help in adapting the methodology to deal with non-standard
problems. Some examples in the earlier chapters are based on fewer observa-
tions than would normally be encountered in medical research programmes.
This enables the methods of analysis to be illustrated more easily, as well as
allowing tabular presentations of the results to be compared with output ob-
tained from computer software. Some additional data sets that may be used
to obtain a fuller appreciation of the methodology, or as student exercises, are
given in an Appendix. All of the data sets used in this book are available in
electronic form from the publisher’s web site at http://www.crcpress.com/.
In writing this book, I have assumed that the reader has a basic knowledge
of statistical methods, and has some familiarity with linear regression analysis.
Matrix algebra is used on occasions, but an understanding of linear algebra is
not an essential requirement. Bibliographic notes and suggestions for further
reading are given at the end of each chapter, but so as not to interrupt the
flow, references in the text itself have been kept to a minimum. Some sections
contain more mathematical details than others, and these have been denoted
with an asterisk. These sections can be omitted without loss of continuity.
I am indebted to Doug Altman, Alan Kimber, Mike Patefield, Anne White-
head and John Whitehead for their help in the preparation of the current and
earlier editions of the book, and to NHS Blood and Transplant for permission
to use data from the UK Transplant Registry in a number of the examples.
I also thank James Gallagher and staff of the Statistical Services Centre,
University of Reading, and my colleagues in the Statistics and Clinical Stud-
ies section of NHS Blood and Transplant, for giving me the opportunity to
rehearse the new material through courses and seminars. I am particularly
grateful to all those who took the trouble to let me know about errors in ear-
lier editions. Although these have been corrected, I would be very pleased to
be informed (d.collett@btinternet.com) of any further errors, ambiguities
and omissions in this edition. Finally, I would like to thank my wife Janet for
her support and encouragement over the period that this book was written.
David Collett
September, 2014
Chapter 1
Survival analysis
Survival analysis is the phrase used to describe the analysis of data in the
form of times from a well-defined time origin until the occurrence of some
particular event or end-point. In medical research, the time origin will often
correspond to the recruitment of an individual into an experimental study,
such as a clinical trial to compare two or more treatments. This in turn may
coincide with the diagnosis of a particular condition, the commencement of
a treatment regimen or the occurrence of some adverse event. If the end-
point is the death of a patient, the resulting data are literally survival times.
However, data of a similar form can be obtained when the end-point is not
fatal, such as the relief of pain, or the recurrence of symptoms. In this case, the
observations are often referred to as time to event data, and the methods for
analysing survival data that are presented in this book apply equally to data
on the time to these end-points. The methods can also be used in the analysis
of data from other application areas, such as the survival times of animals in
an experimental study, the time taken by an individual to complete a task in
a psychological experiment, the storage times of seeds held in a seed bank or
the lifetimes of industrial or electronic components. The focus of this book is
on the application of survival analysis to data arising from medical research,
and for this reason much of the general discussion will be phrased in terms of
the survival time of an individual patient from entry to a study until death.
1.1 Special features of survival data

We must first consider the reasons why survival data are not amenable to
standard statistical procedures used in data analysis. One reason is that sur-
vival data are generally not symmetrically distributed. Typically, a histogram
constructed from the survival times of a group of similar individuals will tend
to be positively skewed, that is, the histogram will have a longer ‘tail’ to the
right of the interval that contains the largest number of observations. As a
consequence, it will not be reasonable to assume that data of this type have
a normal distribution. This difficulty could be resolved by first transforming
the data to give a more symmetric distribution, for example by taking log-
arithms. However, a more satisfactory approach is to adopt an alternative
distributional model for the original data.
1
2 SURVIVAL ANALYSIS
The main feature of survival data that renders standard methods inappro-
priate is that survival times are frequently censored. Censoring is described in
the next section.
1.1.1 Censoring
The survival time of an individual is said to be censored when the end-point
of interest has not been observed for that individual. This may be because the
data from a study are to be analysed at a point in time when some individuals
are still alive. Alternatively, the survival status of an individual at the time
of the analysis might not be known because that individual has been lost
to follow-up. As an example, suppose that after being recruited to a clinical
trial, a patient moves to another part of the country, or to a different country,
and can no longer be traced. The only information available on the survival
experience of that patient is the last date on which he or she was known to
be alive. This date may well be the last time that the patient reported to a
clinic for a regular check-up.
An actual survival time can also be regarded as censored when death is
from a cause that is known to be unrelated to the treatment. However, it can
be difficult to be sure that the death is not related to a particular treatment
that the patient is receiving. For example, consider a patient in a clinical
trial to compare alternative therapies for prostatic cancer who experiences a
fatal road traffic accident. The accident could have resulted from an attack
of dizziness, which might be a side effect of the treatment to which that
patient has been assigned. If so, the death is not unrelated to the treatment.
In circumstances such as these, the survival time until death from all causes,
or the time to death from causes other than the primary condition for which
the patient is being treated, might also be subjected to a survival analysis.
In each of these situations, a patient who entered a study at time t0 dies
at time t0 + t. However, t is unknown, either because the individual is still
alive or because he or she has been lost to follow-up. If the individual was
last known to be alive at time t0 + c, the time c is called a censored survival
time. This censoring occurs after the individual has been entered into a study,
that is, to the right of the last known survival time, and is therefore known as
right censoring. The right-censored survival time is then less than the actual,
but unknown, survival time. Right censoring that occurs when the observation
period of a study ends is often termed administrative censoring.
Another form of censoring is left censoring, which is encountered when the
actual survival time of an individual is less than that observed. To illustrate
this form of censoring, consider a study in which interest centres on the time
to recurrence of a particular cancer following surgical removal of the primary
tumour. Three months after their operation, the patients are examined to
determine if the cancer has recurred. At this time, some of the patients may
be found to have a recurrence. For such patients, the actual time to recurrence
is less than three months, and the recurrence times of these patients is left-
SPECIAL FEATURES OF SURVIVAL DATA 3
censored. Left censoring occurs far less commonly than right censoring, and
so the emphasis of this book will be on the analysis of right-censored survival
data.
Yet another type of censoring is interval censoring. Here, individuals are
known to have experienced an event within an interval of time. Consider again
the example concerning the time to recurrence of a tumour used in the above
discussion of left censoring. If a patient is observed to be free of the disease
at three months, but is found to have had a recurrence when examined six
months after surgery, the actual recurrence time of that patient is known to
be between three months and six months. The observed recurrence time is
then said to be interval-censored. We will return to interval censoring later,
in Chapter 9.
1.1.2 Independent censoring

An important assumption that will be made in the analysis of censored sur-
vival data is that the actual survival time of an individual, t, does not depend
on any mechanism that causes that individual’s survival time to be censored at
time c, where c < t. Such censoring is termed independent or non-informative
censoring. This means that if we consider a group of individuals who all have
the same values of relevant prognostic variables, an individual whose survival
time is censored at time c must be representative of all other individuals in
that group who have survived to that time. A patient whose survival time is
censored will be representative of those at risk at the censoring time if the
censoring process operates randomly. Similarly, when survival data are to be
analysed at a predetermined point in calendar time, or at a fixed interval of
time after the time origin for each patient, the prognosis for individuals who
are still alive can be taken to be independent of the censoring, so long as
the time of analysis is specified before the data are examined. However, this
assumption cannot be made if, for example, the survival time of an individual
is censored through treatment being withdrawn as a result of a deterioration
in their physical condition. This type of censoring is known as dependent or
informative censoring. The methods of survival analysis presented in most
chapters of this book are only valid under the assumption of independent cen-
soring, but techniques that enable account to be taken of dependent censoring
will be described in Chapter 14.
1.1.3 Study time and patient time

In a typical study, patients are not all recruited at exactly the same time, but
accrue over a period of months or even years. After recruitment, patients are
followed up until they die, or until a point in calendar time that marks the end
of the study, when the data are analysed. Although the actual survival times
will be observed for a number of patients, after recruitment some patients may
be lost to follow-up, while others will still be alive at the end of the study.
4 SURVIVAL ANALYSIS
The calendar time period in which an individual is in the study is known as
the study time.
The study time for eight individuals in a clinical trial is illustrated diagram-
matically in Figure 1.1, in which the time of entry to the study is represented
by a ‘•’.
1 D
2 L
3 A
Patient
4 D
5 D
6 A
7 L
8 D
End of recruitment End of study

Study time
Figure 1.1 Study time for eight patients in a survival study.
This figure shows that individuals 1, 4, 5 and 8 die (D) during the course
of the study, individuals 2 and 7 are lost to follow-up (L), and individuals 3
and 6 are still alive (A) at the end of the observation period.
As far as each patient is concerned, the trial begins at some time t0 .
The corresponding survival times for the eight individuals depicted in Fig-
ure 1.1 are shown in order in Figure 1.2. The period of time that a patient
spends in the study, measured from that patient’s time origin, is often re-
ferred to as patient time. The period of time from the time origin to the
death of a patient (D) is then the survival time, and this is recorded for in-
dividuals 1, 4, 5 and 8. The survival times of the remaining individuals are
right-censored (C).
In practice, the actual data recorded will be the date on which each indi-
vidual enters the study, and the date on which each individual dies or was last
known to be alive. The survival time in days, weeks or months, whichever is
the most appropriate, can then be calculated. Most computer software pack-
ages for survival analysis have facilities for performing this calculation from
input data in the form of dates.
SOME EXAMPLES 5
5 D
7 C
8 D
Patient
1 D
2 C
4 D
6 C
3 C
Patient time
Figure 1.2 Patient time for eight patients in a survival study.
1.2 Some examples

In this section, the essential features of survival data are illustrated through a
number of examples. Data from these examples will then be used to illustrate
some of the statistical techniques presented in subsequent chapters.
Example 1.1 Time to discontinuation of the use of an IUD

In trials involving contraceptives, prevention of pregnancy is an obvious cri-
terion for acceptability. However, modern contraceptives have very low fail-
ure rates, and so the occurrence of bleeding disturbances, such as amenor-
rhoea (the prolonged absence of bleeding), irregular or prolonged bleeding,
become important in the evaluation of a particular method of contraception.
To promote research into methods for analysing menstrual bleeding data from
women in contraceptive trials, the World Health Organisation made available
data from clinical trials involving a number of different types of contraceptive
(WHO, 1987). Part of this data set relates to the time from which a woman
commences use of a particular method until discontinuation, with the discon-
tinuation reason being recorded when known. The data in Table 1.1 refer to
the number of weeks from the commencement of use of a particular type of
intrauterine device (IUD), known as the Multiload 250, until discontinuation
because of menstrual bleeding problems. Data are given for 18 women, all of
whom were aged between 18 and 35 years and who had experienced two pre-
vious pregnancies. Discontinuation times that are censored are labelled with
an asterisk.
In this example, the time origin corresponds to the first day in which a
woman uses the IUD, and the end-point is discontinuation because of bleed-
6 SURVIVAL ANALYSIS
Table 1.1 Time in weeks to discontinuation of the use of an IUD.

10 13* 18* 19 23* 30 36 38* 54*
56* 59 75 93 97 104* 107 107* 107*
* Censored discontinuation times.
ing problems. Some women in the study ceased using the IUD because of the
desire for pregnancy, or because they had no further need for a contracep-
tive, while others were simply lost to follow-up. These reasons account for the
censored discontinuation times of 13, 18, 23, 38, 54 and 56 weeks. The study
protocol called for the menstrual bleeding experience of each woman to be
documented for a period of two years from the time origin. For practical rea-
sons, each woman could not be examined exactly two years after recruitment
to determine if they were still using the IUD, and this is why there are three
discontinuation times greater than 104 weeks that are right-censored.
One objective in an analysis of these data would be to summarise the
distribution of discontinuation times. We might then wish to estimate the
median time to discontinuation of the IUD, or the probability that a woman
will stop using the device after a given period of time. Indeed, a graph of this
estimated probability, as a function of time, will provide a useful summary of
the observed data.
Example 1.2 Prognosis for women with breast cancer

Breast cancer is one of the most common forms of cancer occurring in women
living in the Western world. However, the biological behaviour of the tumour
is often unpredictable, and a number of studies have focussed on whether the
tumour is likely to have metastasised, or spread, to other organs in the body.
Around 80% of women presenting with primary breast cancer are likely to
have tumours that have already metastasised to other sites. If these patients
could be identified, adjunctive treatment could be focussed on them, while the
remaining 20% could be reassured that their disease is surgically curable.
The aim of an investigation carried out at the Middlesex Hospital, docu-
mented in Leathem and Brooks (1987), was to evaluate a histochemical marker
that discriminates between primary breast cancer that has metastasised and
that which has not. The marker under study was a lectin from the albumin
gland of the Roman snail, Helix pomatia, known as Helix pomatia agglutinin, or
HPA. The marker binds to those breast cancer cells associated with metastasis
to local lymph nodes, and the HPA stained cells can be identified by micro-
scopic examination. In order to investigate whether HPA staining can be used
to predict the survival experience of women who present with breast cancer, a
retrospective study was carried out, based on the records of women who had
received surgical treatment for breast cancer. Sections of the tumours of these
women were treated with HPA and each tumour was subsequently classified
as being positively or negatively stained, positive staining corresponding to a
tumour with the potential for metastasis. The study was concluded in July
SOME EXAMPLES 7
1987, when the survival times of those women who had died of breast cancer
were calculated. For those women whose survival status in July 1987 was un-
known, the time from surgery to the date on which they were last known to
be alive is regarded as a censored survival time. The survival times of women
who had died from causes other than breast cancer are also regarded as right-
censored. The data given in Table 1.2 refer to the survival times in months of
women who had received a simple or radical mastectomy to treat a tumour of
Grade II, III or IV, between January 1969 and December 1971. In the table,
the survival times of each woman are classified according to whether their
tumour was positively or negatively stained.
Table 1.2 Survival times of women with tumours

that were negatively or positively stained with HPA.
Negative staining Positive staining
23 5 68
47 8 71
69 10 76*
70* 13 105*
71* 18 107*
100* 24 109*
101* 26 113
148 26 116*
181 31 118
198* 35 143
208* 40 154*
212* 41 162*
224* 48 188*
50 212*
59 217*
61 225*
* Censored survival times.
In the analysis of the data from this study, we will be particularly interested
in whether or not there is a difference in the survival experience of the two
groups of women. If there were evidence that those women with negative
HPA staining tended to live longer after surgery than those with positive
staining, we would conclude that the prognosis for a breast cancer patient
was dependent on the result of the staining procedure.
Example 1.3 Survival of multiple myeloma patients

Multiple myeloma is a malignant disease characterised by the accumulation
of abnormal plasma cells, a type of white blood cell, in the bone marrow.
The proliferation of the abnormal plasma cells within the bone causes pain
and the destruction of bone tissue. Patients with multiple myeloma also ex-
perience anaemia, haemorrhages, recurrent infections and weakness. Unless
treated, the condition is invariably fatal. The aim of a study carried out at
the Medical Center of the University of West Virginia, USA, was to examine
8 SURVIVAL ANALYSIS
the association between the values of certain explanatory variables or covari-
ates and the survival time of patients. In the study, the primary response
variable was the time, in months, from diagnosis until death from multiple
myeloma.
The data in Table 1.3, which were obtained from Krall, Uthoff and Harley
(1975), relate to 48 patients, all of whom were aged between 50 and 80 years.
Some of these patients had not died by the time that the study was completed,
and so these individuals contribute right-censored survival times. The coding
of the survival status of an individual in the table is such that zero denotes a
censored observation and unity death from multiple myeloma.
At the time of diagnosis, the values of a number of explanatory variables
were recorded for each patient. These included the age of the patient in years,
their sex (1 = male, 2 = female), the levels of blood urea nitrogen (Bun),
serum calcium (Ca) and haemoglobin (Hb), the percentage of plasma cells in
the bone marrow (Pcells) and an indicator variable (Protein) that denotes
whether or not the Bence-Jones protein was present in the urine (0 = absent,
1 = present).
The main aim of an analysis of these data would be to investigate the effect
of the risk factors Bun, Ca, Hb, Pcells and Protein on the survival time of the
multiple myeloma patients. The effects of these risk factors may be modified
by the age or sex of a patient, and so the extent to which the relationship
between survival and the important risk factors is consistent for each sex and
for each of a number of age groups will also need to be studied.
Example 1.4 Comparison of two treatments for prostatic cancer

A randomised controlled clinical trial to compare treatments for prostatic can-
cer was begun in 1967 by the Veteran’s Administration Cooperative Urological
Research Group. The trial was double-blind and two of the treatments used
in the study were a placebo and 1.0 mg of diethylstilbestrol (DES). The treat-
ments were administered daily by mouth. The time origin of the study is the
date on which a patient was randomised to a treatment, and the end-point is
the death of the patient from prostatic cancer.
The full data set is given in Andrews and Herzberg (1985), but the data
used in this example are from patients presenting with Stage III cancer, that
is, patients for whom there was evidence of a local extension of the tumour
beyond the prostatic capsule, but without elevated serum prostatic acid phos-
phatase. Furthermore, the patients were those who had no history of cardio-
vascular disease, had a normal ECG result at trial entry, and who were not
confined to bed during the daytime. In addition to recording the survival time
of each patient in the study, information was recorded on a number of other
prognostic factors. These included the age of the patient at trial entry, their
serum haemoglobin level in gm/100 ml, the size of their primary tumour in
cm2 and the value of a combined index of tumour stage and grade. This index
is known as the Gleason index; the more advanced the tumour, the greater
the value of the index.
SOME EXAMPLES 9
Table 1.3 Survival times of patients in a study on multiple myeloma.

Patient Survival Status Age Sex Bun Ca Hb Pcells Protein
number time
1 13 1 66 1 25 10 14.6 18 1
2 52 0 66 1 13 11 12.0 100 0
3 6 1 53 2 15 13 11.4 33 1
4 40 1 69 1 10 10 10.2 30 1
5 10 1 65 1 20 10 13.2 66 0
6 7 0 57 2 12 8 9.9 45 0
7 66 1 52 1 21 10 12.8 11 1
8 10 0 60 1 41 9 14.0 70 1
9 10 1 70 1 37 12 7.5 47 0
10 14 1 70 1 40 11 10.6 27 0
11 16 1 68 1 39 10 11.2 41 0
12 4 1 50 2 172 9 10.1 46 1
13 65 1 59 1 28 9 6.6 66 0
14 5 1 60 1 13 10 9.7 25 0
15 11 0 66 2 25 9 8.8 23 0
16 10 1 51 2 12 9 9.6 80 0
17 15 0 55 1 14 9 13.0 8 0
18 5 1 67 2 26 8 10.4 49 0
19 76 0 60 1 12 12 14.0 9 0
20 56 0 66 1 18 11 12.5 90 0
21 88 1 63 1 21 9 14.0 42 1
22 24 1 67 1 10 10 12.4 44 0
23 51 1 60 2 10 10 10.1 45 1
24 4 1 74 1 48 9 6.5 54 0
25 40 0 72 1 57 9 12.8 28 1
26 8 1 55 1 53 12 8.2 55 0
27 18 1 51 1 12 15 14.4 100 0
28 5 1 70 2 130 8 10.2 23 0
29 16 1 53 1 17 9 10.0 28 0
30 50 1 74 1 37 13 7.7 11 1
31 40 1 70 2 14 9 5.0 22 0
32 1 1 67 1 165 10 9.4 90 0
33 36 1 63 1 40 9 11.0 16 1
34 5 1 77 1 23 8 9.0 29 0
35 10 1 61 1 13 10 14.0 19 0
36 91 1 58 2 27 11 11.0 26 1
37 18 0 69 2 21 10 10.8 33 0
38 1 1 57 1 20 9 5.1 100 1
39 18 0 59 2 21 10 13.0 100 0
40 6 1 61 2 11 10 5.1 100 0
41 1 1 75 1 56 12 11.3 18 0
42 23 1 56 2 20 9 14.6 3 0
43 15 1 62 2 21 10 8.8 5 0
44 18 1 60 2 18 9 7.5 85 1
45 12 0 71 2 46 9 4.9 62 0
46 12 1 60 2 6 10 5.5 25 0
47 17 1 65 2 28 8 7.5 8 0
48 3 0 59 1 90 10 10.2 6 1
10 SURVIVAL ANALYSIS
Table 1.4 gives the data recorded for 38 patients, where the survival times
are given in months. The survival times of patients who died from other
causes, or who were lost during the follow-up process, are regarded as censored.
A variable associated with the status of an individual at the end of the study
takes the value unity if the patient has died from prostatic cancer, and zero if
the survival time is right-censored. The variable associated with the treatment
group takes the value 2 when an individual is treated with DES and unity if
an individual is on the placebo treatment.
The main aim of this study is to determine the extent of any evidence that
patients treated with DES survive longer than those treated with the placebo.
Since the data on which this example is based are from a randomised trial, one
might expect that the distributions of the prognostic factors, that is the age
of patient, serum haemoglobin level, size of tumour and Gleason index, will
be similar over the patients in each of the two treatment groups. However, it
would not be wise to rely on this assumption. For example, it could turn out
that patients in the placebo group had larger tumours on average than those
in the group treated with DES. If patients with large tumours have a poorer
prognosis than those with small tumours, the size of the treatment effect would
be overestimated, unless proper account was taken of the size of the tumour
in the analysis. Consequently, it will first be necessary to determine if any of
the covariates are related to survival time. If so, the effect of these variables
will need to be allowed for when comparing the survival experiences of the
patients in the two treatment groups.
1.3 Survivor, hazard and cumulative hazard functions

In summarising survival data, there are three functions of central interest,
namely the survivor function, the hazard function, and the cumulative hazard
function. These functions are therefore defined in this first chapter.
1.3.1 The survivor function

The actual survival time of an individual, t, can be regarded as the observed
value of a variable, T , that can take any non-negative value. The different val-
ues that T can take have a probability distribution, and we call T the random
variable associated with the survival time. Now suppose that this random vari-
able has a probability distribution with underlying probability density function
f (t). The distribution function of T is then given by
∫ t
F (t) = P(T < t) = f (u) du, (1.1)
0
and represents the probability that the survival time is less than some value t.
This function is also called the cumulative incidence function, since it sum-
marises the cumulative probability of death occurring before time t.
SURVIVOR, HAZARD AND CUMULATIVE HAZARD FUNCTIONS 11
Table 1.4 Survival times of prostatic cancer patients in a clinical trial to compare
two treatments.
Patient Treatment Survival Status Age Serum Size of Gleason
number time haem. tumour index
1 1 65 0 67 13.4 34 8
2 2 61 0 60 14.6 4 10
3 2 60 0 77 15.6 3 8
4 1 58 0 64 16.2 6 9
5 2 51 0 65 14.1 21 9
6 1 51 0 61 13.5 8 8
7 1 14 1 73 12.4 18 11
8 1 43 0 60 13.6 7 9
9 2 16 0 73 13.8 8 9
10 1 52 0 73 11.7 5 9
11 1 59 0 77 12.0 7 10
12 2 55 0 74 14.3 7 10
13 2 68 0 71 14.5 19 9
14 2 51 0 65 14.4 10 9
15 1 2 0 76 10.7 8 9
16 1 67 0 70 14.7 7 9
17 2 66 0 70 16.0 8 9
18 2 66 0 70 14.5 15 11
19 2 28 0 75 13.7 19 10
20 2 50 1 68 12.0 20 11
21 1 69 1 60 16.1 26 9
22 1 67 0 71 15.6 8 8
23 2 65 0 51 11.8 2 6
24 1 24 0 71 13.7 10 9
25 2 45 0 72 11.0 4 8
26 2 64 0 74 14.2 4 6
27 1 61 0 75 13.7 10 12
28 1 26 1 72 15.3 37 11
29 1 42 1 57 13.9 24 12
30 2 57 0 72 14.6 8 10
31 2 70 0 72 13.8 3 9
32 2 5 0 74 15.1 3 9
33 2 54 0 51 15.8 7 8
34 1 36 1 72 16.4 4 9
35 2 70 0 71 13.6 2 10
36 2 67 0 73 13.8 7 8
37 1 23 0 68 12.5 2 8
38 1 62 0 63 13.2 3 8
The survivor function, S(t), is defined to be the probability that the sur-
vival time is greater than or equal to t, and so from Equation (1.1),
S(t) = P(T > t) = 1 − F (t). (1.2)
The survivor function can therefore be used to represent the probability that
an individual survives beyond any given time.
1.3.2 The hazard function

The hazard function is widely used to express the risk or hazard of an event
such as death occurring at some time t. This function is obtained from the
probability that an individual dies at time t, conditional on he or she having
survived to that time. For a formal definition of the hazard function, consider
the probability that the random variable associated with an individual’s sur-
vival time, T , lies between t and t + δt, conditional on T being greater than or
equal to t, written P(t 6 T < t + δt | T > t). This conditional probability is
then expressed as a probability per unit time by dividing by the time interval,
δt, to give a rate. The hazard function, h(t), is then the limiting value of this
quantity, as δt tends to zero, so that
{ }
P(t 6 T < t + δt | T > t)
h(t) = lim . (1.3)
δt→0 δt
The function h(t) is also referred to as the hazard rate, the instantaneous death
rate, the intensity rate or the force of mortality.
From the definition of the hazard function in Equation (1.3), h(t) is the
event rate at time t, conditional on the event not having occurred before t.
Specifically, if the survival time is measured in days, h(t) is the approximate
probability that an individual, who is at risk of the event occurring at the
start of day t, experiences the event during that day. The hazard function at
time t can also be regarded as the expected number of events experienced by
an individual in unit time, given that the event has not occurred before then,
and assuming that the hazard is constant over that time period.
The definition of the hazard function in Equation (1.3) leads to some use-
ful relationships between the survivor and hazard functions. According to a
standard result from probability theory, the probability of an event A, condi-
tional on the occurrence of an event B, is given by P(A | B) = P(AB)/P(B),
where P(AB) is the probability of the joint occurrence of A and B. Using this
result, the conditional probability in the definition of the hazard function in
Equation (1.3) is
P(t 6 T < t + δt)
,
P(T > t)
which is equal to
F (t + δt) − F (t)
,
S(t)
SURVIVOR, HAZARD AND CUMULATIVE HAZARD FUNCTIONS 13
where F (t) is the distribution function of T . Then,
{ }
F (t + δt) − F (t) 1
h(t) = lim .
δt→0 δt S(t)
Now, { }
F (t + δt) − F (t)
lim
δt→0 δt
is the definition of the derivative of F (t) with respect to t, which is f (t), and
so
f (t)
h(t) = . (1.4)
S(t)
Taken together, Equations (1.1), (1.2) and (1.4) show that from any one of
the three functions, f (t), S(t), and h(t), the other two can be determined.
1.3.3 The cumulative hazard function

From Equation (1.4), it follows that
d
h(t) = − {log S(t)}, (1.5)
dt
and so
S(t) = exp {−H(t)}, (1.6)
where ∫ t
H(t) = h(u) du. (1.7)
0
The function H(t) features widely in survival analysis, and is called the in-
tegrated or cumulative hazard function. From Equation (1.6), the cumulative
hazard function can also be obtained from the survivor function, since
H(t) = − log S(t). (1.8)
The cumulative hazard function, H(t), is the cumulative risk of an event

occurring by time t. If the event is death, then H(t) summarises the risk of
death up to time t, given that death has not occurred before t. The cumulative
hazard function at time t can also be interpreted as the expected number of
events that occur in the interval from the time origin to t.
It is possible for the cumulative hazard function to exceed unity. Using
Equation (1.8), H(t) > 1, when − log S(t) > 1, that is when S(t) 6 e−1 =
0.37. The cumulative hazard is then greater than unity when the probability
of an event occurring after time t is less than 0.37, and means that more than
one event is expected in the time interval (0, t). The survivor function, S(t), is
then more correctly defined as the probability that one or more events occur
after time t. The interpretation of a cumulative hazard function in terms of
the expected number of events is only reasonable when repetitions of an event
are possible, such as when the event is the occurrence of an infection, migraine
or seizure. When the event of interest is death, this interpretation relies on
individuals being immediately resurrected after death has occurred! Methods
for analysing times to multiple occurrences of an event are considered later
in Chapter 13, and a more mathematical interpretation of the hazard and
cumulative hazard functions when multiple events are possible is included in
Section 13.1 of that chapter.
In the analysis of survival data, the survivor function, hazard function and
cumulative hazard function are estimated from the observed survival times.
Methods of estimation that do not require the form of the probability den-
sity function of T to be specified are described in Chapters 2 and 3, while
methods based on the assumption of a particular survival time distribution
are presented in Chapters 5 and 6.
1.4 Computer software for survival analysis
Most of the techniques for analysing survival data that will be presented in
this book require suitable computer software for their implementation. Many
computer packages for survival analysis are now available, but of the commer-
cially available software packages, SAS (SAS Institute Inc.), S-PLUS (TIBCO
Software Inc.) and Stata (StataCorp) have the most extensive range of fa-
cilities. In addition, the R statistical computing environment (R Core Team,
2013) is free software, distributed under the terms of the GNU General Public
License. Both S-PLUS and R are modern implementations of the S statisti-
cal programming language, and include a comprehensive range of modules for
survival analysis. Any of these four packages can be used to carry out the
analyses described in subsequent chapters of this book.
In this book, the data sets used to illustrate the different methods of sur-
vival analysis have been analysed using SAS 9.4 (SAS Institute, Cary NC),
mainly using the procedures lifetest, lifereg and phreg. Where published
SAS macros have been used for more specialised analyses, these are docu-
mented in the ‘Further reading’ section of each chapter.
In some circumstances, numerical results in the output produced by soft-
ware packages may differ. This is often due to different default methods of
calculation being used. A particularly important example of this occurs when
a data set includes two or more individuals with the same survival times. In
this case, the SAS phreg procedure and the R package survival (Therneau,
2014) default to different methods of handling these tied observations, leading
to differences in the output. The default settings can of course be changed,
and the treatment of tied survival times is described in Section 3.3.2 of Chap-
ter 3. Differences in numerical values may also result from different settings
being used for parameters that control the convergence of certain iterative
procedures, and different methods being used for numerical optimisation.
FURTHER READING 15
1.5 Further reading
An introduction to the techniques used in the analysis of survival data is

included in a number of general books on statistics in medical research, such
as those of Altman (1991) and Armitage, Berry and Matthews (2002). Machin,
Cheung and Parmar (2006) provide a practical guide to the analysis of survival
data from clinical trials, using non-technical language.
There a number of textbooks that provide an introduction to the methods
of survival analysis, illustrated with practical examples. Lee and Wang (2013)
provides a broad coverage of topics with illustrations drawn from biology and
medicine, and Marubini and Valsecchi (1995) describe the analysis of survival
data from clinical trials and observational studies. Hosmer, Lemeshow and
May (2008) give a balanced account of survival analysis, with excellent chap-
ters on model development and the interpretation of the parameter estimates
in a fitted model. Klein and Moeschberger (2005) include many example data
sets and exercises in their comprehensive textbook, and Kleinbaum and Klein
(2012) provide a self-learning text on survival analysis. Applications of sur-
vival analysis in the analysis of epidemiological data are described by Breslow
and Day (1987) and Woodward (2014). Introductory texts that describe the
application of survival analysis in other areas include those of Crowder et al.
(1991) who focus on the analysis of reliability data, and Box-Steffensmeier and
Jones (2004) who give a non-mathematical account of time to event analysis
in the social sciences.
Comprehensive accounts of the subject are given by Kalbfleisch and Pren-
tice (2002) and Lawless (2002). These books have been written for the post-
graduate statistician or research worker, and are usually regarded as reference
books rather than introductory texts. A concise review of survival analysis is
given in the research monograph of Cox and Oakes (1984), and in the chap-
ter devoted to this subject in Hinkley, Reid and Snell (1991). The book by
Hougaard (2000) on multivariate survival data incorporates more advanced
topics, after introductory chapters that cover the basic features of survival
analysis. Therneau and Grambsch (2000) base their presentation of survival
analysis on the counting process approach, leading to a more mathematical
development of the material. Harrell (2001) gives details on many issues that
arise in the development of a statistical model not found in other texts, and
includes an extensive discussion of two case studies.
There are many general books on the use of particular software packages
for data analysis, and some that give a detailed account of how they are used
in the analysis of survival data. Allison (2010) provides a comprehensive guide
to the SAS software for survival analysis. Der and Everitt (2013) also include
material on survival analysis in their text on the use of SAS for analysing
medical data. Therneau and Grambsch (2000) give a detailed account of how
SAS and S-PLUS are used to fit the Cox regression model, and extensions to
it. This book includes a description of a number of SAS macros and S-PLUS
functions that supplement the standard facilities available in these packages.
The use of S-PLUS in survival analysis is also described in Everitt and Rabe-
Hesketh (2001) and Tableman and Kim (2004), while Broström (2012) shows
how R is used in the analysis of survival data. Venables and Ripley (2002)
describe how graphical and numerical data analyses can be carried out in the
S environment that is implemented in both R and S-PLUS; note that S code
generally runs under R. A similarly comprehensive account of the R system
is given by Crawley (2013), while Dalgaard (2008) gives a more elementary
introduction to R. The short introduction to R of Venables and Smith (2009)
is also available from R Core Team (2013). The use of Stata in survival analysis
is presented by Cleves et al. (2010), and Rabe-Hesketh and Everitt (2007) give
a more general introduction to the use of Stata in data analysis.
Chapter 2
Some non-parametric procedures
An initial step in the analysis of a set of survival data is to present numerical

or graphical summaries of the survival times for individuals in a particular
group. Such summaries may be of interest in their own right, or as a precur-
sor to a more detailed analysis of the data. Survival data are conveniently
summarised through estimates of the survivor function and hazard function.
Methods for estimating these functions from a single sample of survival data
are described in Sections 2.1 and 2.3. These methods are said to be non-
parametric or distribution-free, since they do not require specific assumptions
to be made about the underlying distribution of the survival times.
Once the estimated survivor function has been found, the median and
other percentiles of the distribution of survival times can be estimated, as
shown in Section 2.4. Numerical summaries of the data, derived on the basis
of assumptions about the probability distribution from which the data have
been drawn, will be considered later in Chapters 5 and 6.
When the survival times of two groups of patients are being compared, an
informal comparison of the survival experience of each group of individuals
can be made using the estimated survivor functions. However, there are more
formal procedures that enable two groups of survival data to be compared.
Two non-parametric procedures for comparing two or more groups of sur-
vival times, namely the log-rank test and the Wilcoxon test, are described in
Section 2.6.
2.1 Estimating the survivor function

Suppose first that we have a single sample of survival times, where none of
the observations are censored. The survivor function S(t), defined in Equa-
tion (1.2), is the probability that an individual survives for a time greater
than or equal to t. This function can be estimated by the empirical survivor
function, given by
Number of individuals with survival times > t
Ŝ(t) = . (2.1)
Number of individuals in the data set
Equivalently, Ŝ(t) = 1−F̂ (t), where F̂ (t) is the empirical distribution function,
that is, the ratio of the total number of individuals alive at time t to the total
17
18 SOME NON-PARAMETRIC PROCEDURES
number of individuals in the study. Notice that the empirical survivor function
is equal to unity for values of t before the first death time, and zero after the
final death time.
The estimated survivor function Ŝ(t) is assumed to be constant between
two adjacent death times, and so a plot of Ŝ(t) against t is a step-function.
The function decreases immediately after each observed survival time.
Example 2.1 Pulmonary metastasis

One complication in the management of patients with a malignant bone tu-
mour, or osteosarcoma, is that the tumour often spreads to the lungs. This
pulmonary metastasis is life-threatening. In a study concerned with the treat-
ment of pulmonary metastasis arising from osteosarcoma, Burdette and Gehan
(1970) give the following survival times, in months, of eleven male patients.
11 13 13 13 13 13 14 14 15 15 17
Using Equation (2.1), the estimated values of the survivor function at times
11, 13, 14, 15 and 17 months are 1.000, 0.909, 0.455, 0.273 and 0.091. The
estimated value of the survivor function is unity from the time origin until 11
months, and zero after 17 months. A graph of the estimated survivor function
is given in Figure 2.1.
1.0
Estimated survivor function
0.8
0.6
0.4
0.2
0.0
0 3 6 9 12 15 18
Survival time
Figure 2.1 Estimated survivor function for the data from Example 2.1.
The method of estimating the survivor function illustrated in the above

example cannot be used when there are censored observations. The reason for
this is that the method does not allow information provided by an individual
ESTIMATING THE SURVIVOR FUNCTION 19
whose survival time is censored before time t to be used in computing the
estimated survivor function at t. Non-parametric methods for estimating S(t),
which can be used in the presence of censored survival times, are described in
the following sections.
2.1.1 Life-table estimate of the survivor function

The life-table estimate of the survivor function, also known as the actuarial
estimate of survivor function, is obtained by first dividing the period of ob-
servation into a series of time intervals. These intervals need not necessarily
be of equal length, although they usually are. The number of intervals used
will depend on the number of individuals in the study, but would usually be
somewhere between 5 and 15.
Suppose that the jth of m such intervals, j = 1, 2, . . . , m, extends from time
t′j−1 to immediately before time t′j , where we take t0 = 0 and tm = ∞. Also,
let dj and cj denote the number of deaths and the number of censored survival
times, respectively, in this interval, and let nj be the number of individuals
who are alive, and therefore at risk of death, at the start of the jth interval.
We now make the assumption that the censoring process is such that the
censored survival times occur uniformly throughout the jth interval, so that
the average number of individuals who are at risk during this interval is
n′j = nj − cj /2. (2.2)
This assumption is sometimes known as the actuarial assumption.

In the jth interval, the probability of death can be estimated by dj /n′j ,
so that the corresponding survival probability is (n′j − dj )/n′j . Now consider
the probability that an individual survives beyond time t′j−1 , j = 2, 3, . . . , m,
that is, until some time after the start of the jth interval. This will be the
product of the probabilities that an individual survives through each of the
j −1 preceding intervals, and so the life-table estimate of the survivor function
is given by
∏ ( n′ − d i )
j−1
∗ i
S (t) = , (2.3)
i=1
n′i
for t′j−1 6 t < t′j , j = 2, 3, . . . , m. The estimated probability of surviving

beyond the start of the first interval, t′0 , is of course unity, while the estimated
probability of surviving beyond t′m is zero. A graphical estimate of the survivor
function will then be a step-function with constant values of the function in
each time interval.

To illustrate the computation of the life-table estimate, consider the data on
the survival times of the 48 multiple myeloma patients given in Table 1.3. In
this illustration, the information collected on other explanatory variables for
each individual will be ignored.
The survival times are first grouped to give the number of patients who die,
dj , and the number who are censored, cj , in each of the first five years of the
study, and in the subsequent three-year period. The number at risk of death
at the start of each of these intervals, nj , is then computed, together with the
adjusted number at risk, n′j . Finally, the probability of survival through each
interval is estimated, from which the estimated survivor function is obtained
using Equation (2.3). The calculations are shown in Table 2.1, in which the
time period is given in months, and the jth interval that begins at time t′j−1
and ends just before time t′j , for j = 1, 2, . . . , m, is denoted t′j−1 –.
Table 2.1 Life-table estimate of the survivor function for the data from
Example 1.3.
Interval Time period dj cj nj n′j (n′j − dj )/n′j S ∗ (t)
1 0– 16 4 48 46.0 0.6522 1.0000
2 12– 10 4 28 26.0 0.6154 0.6522
3 24– 1 0 14 14.0 0.9286 0.4013
4 36– 3 1 13 12.5 0.7600 0.3727
5 48– 2 2 9 8.0 0.7500 0.2832
6 60– 4 1 5 4.5 0.1111 0.2124
A graph of the life-table estimate of the survivor function is shown in

Figure 2.2.
1.0
0.8
0.6
0.4
0.2
0.0
0 10 20 30 40 50 60 70
Survival time
Figure 2.2 Life-table estimate of the survivor function.
The form of the estimated survivor function obtained using this method
is sensitive to the choice of the intervals used in its construction, just as the
shape of a histogram depends on the choice of the class intervals. On the other
hand, the life-table estimate is particularly well suited to situations in which
the actual death times are unknown, and the only available information is the
number of deaths and the number of censored observations that occur in a
series of consecutive time intervals. In practice, such interval-censored survival
data occur quite frequently.
When the actual survival times are known, the life-table estimate can still
be used, as in Example 2.2, but the grouping of the survival times does result
in some loss of information. Alternative methods for estimating the survivor
function are then more appropriate, such as that leading to the Kaplan-Meier
estimate.
2.1.2 Kaplan-Meier estimate of the survivor function

The first step in the analysis of ungrouped censored survival data is normally
to obtain the Kaplan-Meier estimate of the survivor function. This estimate
is therefore considered in some detail. To obtain the Kaplan-Meier estimate, a
series of time intervals is constructed, as for the life-table estimate. However,
each of these intervals is designed to be such that one death time is contained
in the interval, and this death time is taken to occur at the start of the interval.
As an illustration, suppose that t(1) , t(2) and t(3) are three observed survival
times arranged in rank order, so that t(1) < t(2) < t(3) , and that c is a censored
survival time that falls between t(2) and t(3) . The constructed intervals then
begin at times t(1) , t(2) and t(3) , and each interval includes the one death time,
although there could be more than one individual who dies at any particular
death time. Notice that no interval begins at the censored time of c. Now
suppose that two individuals die at t(1) , one dies at t(2) and three die at
t(3) . The situation is illustrated diagrammatically in Figure 2.3, in which D
represents a death and C a censored survival time.
D
D D
D D C D
-
Time
t0 t(1) t(2) t(3)
Figure 2.3 Construction of intervals used in the derivation of the Kaplan-Meier

estimate.
The time origin is denoted by t0 , and so there is an initial period com-

mencing at t0 , which ends just before t(1) , the time of the first death. This
means that the interval from t0 to t(1) will not include a death time. The
first constructed interval extends from t(1) to just before t(2) , and since the
second death time is at t(2) , this interval includes the single death time at t(1) .
The second interval begins at time t(2) and ends just before t(3) , and includes
the death time at t(2) and the censored time c. There is also a third interval
beginning at t(3) , which contains the longest survival time, t(3) .
In general, suppose that there are n individuals with observed survival
times t1 , t2 , . . . , tn . Some of these observations may be right-censored, and
there may also be more than one individual with the same observed survival
time. We therefore suppose that there are r death times amongst the indi-
viduals, where r 6 n. After arranging these death times in ascending order,
the jth is denoted t(j) , for j = 1, 2, . . . , r, and so the r ordered death times
are t(1) < t(2) < · · · < t(r) . The number of individuals who are alive just
before time t(j) , including those who are about to die at this time, will be
denoted nj , for j = 1, 2, . . . , r, and dj will denote the number who die at this
time. The time interval from t(j) − δ to t(j) , where δ is an infinitesimal time
interval, then includes one death time. Since there are nj individuals who
are alive just before t(j) and dj deaths at t(j) , the probability that an indi-
vidual dies during the interval from t(j) − δ to t(j) is estimated by dj /nj .
The corresponding estimated probability of survival through that interval
is then (nj − dj )/nj .
It sometimes happens that there are censored survival times that occur at
the same time as one or more deaths, so that a death time and a censored
survival time appear to occur simultaneously. In this situation, the censored
survival time is taken to occur immediately after the death time when com-
puting the values of the nj .
From the manner in which the time intervals are constructed, the interval
from t(j) to t(j+1) − δ, the time immediately before the next death time,
contains no deaths. The probability of surviving from t(j) to t(j+1) − δ is
therefore unity, and the joint probability of surviving from t(j) − δ to t(j) and
from t(j) to t(j+1) −δ can be estimated by (nj −dj )/nj . In the limit, as δ tends
to zero, (nj − dj )/nj becomes an estimate of the probability of surviving the
interval from t(j) to t(j+1) .
We now make the assumption that the deaths of the individuals in the sam-
ple occur independently of one another. Then, the estimated survivor func-
tion at any time, t, in the kth constructed time interval from t(k) to t(k+1) ,
k = 1, 2, . . . , r, where t(r+1) is defined to be ∞, will be the estimated prob-
ability of surviving beyond t(k) . This is actually the probability of surviving
through the interval from t(k) to t(k+1) , and all preceding intervals, and leads
to the Kaplan-Meier estimate of the survivor function, which is given by
∏k ( )
nj − dj
Ŝ(t) = , (2.4)
j=1
nj
for t(k) 6 t < t(k+1) , k = 1, 2, . . . , r, with Ŝ(t) = 1 for t < t(1) , and where
t(r+1) is taken to be ∞. If the largest observation is a censored survival time,
t∗ , say, Ŝ(t) is undefined for t > t∗ . On the other hand, if the largest observed
survival time, t(r) , is an uncensored observation, nr = dr , and so Ŝ(t) is zero
for t > t(r) . A plot of the Kaplan-Meier estimate of the survivor function is
a step-function, in which the estimated survival probabilities are constant
between adjacent death times and decrease at each death time.
Equation (2.4) shows that, as for the life-table estimate of the survivor
function in Equation (2.3), the Kaplan-Meier estimate is formed as a product
of a series of estimated probabilities. In fact, the Kaplan-Meier estimate is
the limiting value of the life-table estimate in Equation (2.3) as the number
of intervals tends to infinity and their width tends to zero. For this reason,
the Kaplan-Meier estimate is also known as the product-limit estimate of the
survivor function.
Note that if there are no censored survival times in the data set, nj − dj =
nj+1 , j = 1, 2, . . . , k, in Equation (2.4), and on expanding the product we get
n2 n3 nk+1
Ŝ(t) = × × ··· × .
n1 n2 nk
This reduces to nk+1 /n1 , for k = 1, 2, . . . , r − 1, with Ŝ(t) = 1 for t < t(1) and
Ŝ(t) = 0 for t > t(r) . Now, n1 is the number of individuals at risk just before
the first death time, which is the number of individuals in the sample, and nk+1
is the number of individuals with survival times greater than or equal to t(k+1) .
Consequently, in the absence of censoring, Ŝ(t) is simply the empirical survivor
function defined in Equation (2.1). The Kaplan-Meier estimate is therefore a
generalisation of the empirical survivor function that accommodates censored
observations.

Data from 18 women on the time to discontinuation of the use of an intra-
uterine device (IUD) were given in Table 1.1. For these data, the survivor
function, S(t), represents the probability that a woman discontinues the use
of the contraceptive device after any time t. The Kaplan-Meier estimate of the
survivor function is readily obtained using Equation (2.4), and the required
calculations are set out in Table 2.2.
Table 2.2 Kaplan-Meier estimate of the survivor

function for the data from Example 1.1.
Time interval nj dj (nj − dj )/nj Ŝ(t)
0– 18 0 1.0000 1.0000
10– 18 1 0.9444 0.9444
19– 15 1 0.9333 0.8815
30– 13 1 0.9231 0.8137
36– 12 1 0.9167 0.7459
59– 8 1 0.8750 0.6526
75– 7 1 0.8571 0.5594
93– 6 1 0.8333 0.4662
97– 5 1 0.8000 0.3729
107 3 1 0.6667 0.2486
The estimated survivor function, Ŝ(t), is plotted in Figure 2.4. Note that
since the largest discontinuation time of 107 days is censored, Ŝ(t) is not
defined beyond t = 107.
1.0
0.8
0.6
0.4
0.2
0.0
0 20 40 60 80 100 120
Discontinuation time
Figure 2.4 Kaplan-Meier estimate of the survivor function for the data from Exam-
ple 1.1.
2.1.3 Nelson-Aalen estimate of the survivor function

An alternative estimate of the survivor function, which is based on the indi-
vidual event times, is the Nelson-Aalen estimate, given by
∏
k
S̃(t) = exp(−dj /nj ). (2.5)
j=1
This estimate can be obtained from an estimate of the cumulative hazard

function, as shown in Section 2.3.3. Moreover, the Kaplan-Meier estimate of
the survivor function can be regarded as an approximation to the Nelson-
Aalen estimate. To show this, we use the result that
x2 x3
e−x = 1 − x + − + ···,
2 6
which is approximately equal to 1 − x when x is small. It then follows that
exp(−dj /nj ) ≈ 1 − (dj /nj ) = (nj − dj )/nj , so long as dj is small relative
to nj , which it will be except at the longest survival times. Consequently,
the Kaplan-Meier estimate, Ŝ(t), in Equation (2.4), approximates the Nelson-
Aalen estimate, S̃(t), in Equation (2.5).
The Nelson-Aalen estimate of the survivor function, also known as Alt-
shuler’s estimate, will always be greater than the Kaplan-Meier estimate at
any given time, since e−x > 1 − x, for all values of x. Although the Nelson-
Aalen estimate has been shown to perform better than the Kaplan-Meier
STANDARD ERROR OF THE ESTIMATED SURVIVOR FUNCTION 25
estimate in small samples, in many circumstances, the estimates will be very
similar, particularly at the earlier survival times. Since the Kaplan-Meier esti-
mate is a generalisation of the empirical survivor function, the latter estimate
has much to commend it.

The values shown in Table 2.2, which gives the Kaplan-Meier estimate of the
survivor function for the data on the time to discontinuation of the use of an
intrauterine device, can be used to calculate the Nelson-Aalen estimate. This
estimate is shown in Table 2.3.
Table 2.3 Nelson-Aalen estimate of the sur-

vivor function for the data from Example 1.1.
Time interval exp(−dj /nj ) S̃(t)
0– 1.0000 1.0000
10– 0.9460 0.9460
19– 0.9355 0.8850
30– 0.9260 0.8194
36– 0.9200 0.7539
59– 0.8825 0.6653
75– 0.8669 0.5768
93– 0.8465 0.4882
97– 0.8187 0.3997
107 0.7165 0.2864
From this table we see that the Kaplan-Meier and Nelson-Aalen estimates
of the survivor function differ by less than 0.04. However, when we consider
the precision of these estimates, which we do in Section 2.2, we see that a
difference of 0.04 is of no practical importance.
2.2 Standard error of the estimated survivor function
An essential aid to the interpretation of an estimate of any quantity is the

precision of the estimate, which is reflected in the standard error of the esti-
mate. This is defined to be the square root of the estimated variance of the
estimate, and is used in the construction of an interval estimate for a quan-
tity of interest. In this section, the standard error of estimates of the survivor
function are given.
Because the Kaplan-Meier estimate is the most important and widely used
estimate of the survivor function, the derivation of the standard error of Ŝ(t)
will be presented in detail in this section. The details of this derivation can
be omitted on a first reading.
2.2.1 ∗ Standard error of the Kaplan-Meier estimate
The Kaplan-Meier estimate of the survivor function for any value of t in the
interval from t(k) to t(k+1) can be written as
∏
k
Ŝ(t) = p̂j ,
j=1
for k = 1, 2, . . . , r, where p̂j = (nj −dj )/nj is the estimated probability that an
individual survives through the time interval that begins at t(j) , j = 1, 2, . . . , r.
Taking logarithms,
∑k
log Ŝ(t) = log p̂j ,
j=1
and so the variance of log Ŝ(t) is given by

{ } ∑k
var log Ŝ(t) = var {log p̂j } . (2.6)
j=1
Now, the number of individuals who survive through the interval beginning
at t(j) can be assumed to have a binomial distribution with parameters nj
and pj , where pj is the true probability of survival through that interval. The
observed number who survive is nj −dj , and using the result that the variance
of a binomial random variable with parameters n, p is np(1 − p), the variance
of nj − dj is given by
var (nj − dj ) = nj pj (1 − pj ).
Since p̂j = (nj − dj )/nj , the variance of p̂j is var (nj − dj )/n2j , that is, pj (1 −
pj )/nj . The variance of p̂j may then be estimated by
p̂j (1 − p̂j )/nj . (2.7)
In order to obtain the variance of log p̂j , we make use of a general result
for the approximate variance of a function of a random variable. According to
this result, the variance of a function g(X) of the random variable X is given
by
{ }2
dg(X)
var {g(X)} ≈ var (X). (2.8)
dX
This is known as the Taylor series approximation to the variance of a function
of a random variable. Using Equation (2.8), the approximate variance of log p̂j
is var (p̂j )/p̂2j , and using Expression (2.7), the approximate estimated variance
of log p̂j is (1 − p̂j )/(nj p̂j ), which on substitution for p̂j , reduces to
dj
. (2.9)
nj (nj − dj )
* Sections marked with an asterisk may be omitted without loss of continuity.

Then, from Equation (2.6),
{ } ∑k
dj
var log Ŝ(t) ≈ , (2.10)
j=1
nj (nj − dj )
and a further application of the result in Equation (2.8) gives

{ } 1 { }
var log Ŝ(t) ≈ var Ŝ(t) ,
[Ŝ(t)]2
so that
{ } ∑
k
dj
var Ŝ(t) ≈ [Ŝ(t)]2 . (2.11)
j=1
nj (nj − dj )
Finally, the standard error of the Kaplan-Meier estimate of the survivor func-
tion, defined to be the square root of the estimated variance of the estimate,
is given by
  21
{ } ∑ k
dj 
se Ŝ(t) ≈ Ŝ(t) , (2.12)
 nj (nj − dj ) 
j=1
for t(k) 6 t < t(k+1) . This result is known as Greenwood’s formula.

If there are no censored survival times, nj −dj = nj+1 , and Expression (2.9)
becomes (nj − nj+1 )/nj nj+1 . Now,
∑k ∑k ( )
nj − nj+1 1 1 n1 − nk+1
= − = ,
j=1
n j n j+1 j=1
n j+1 n j n1 nk+1
which can be written as

1 − Ŝ(t)
,
n1 Ŝ(t)
since Ŝ(t) = nk+1 /n1 for t(k) 6 t < t(k+1) , k = 1, 2, . . . , r − 1, in the absence
of censoring. Hence, from Equation (2.11), the estimated variance of Ŝ(t) is
Ŝ(t)[1 − Ŝ(t)]/n1 . This is an estimate of the variance of the empirical sur-
vivor function, given in Equation (2.1), on the assumption that the number
of individuals at risk at time t has a binomial distribution with parameters
n1 , S(t).
2.2.2 ∗ Standard error of other estimates

The life-table estimate of the survivor function is similar in form to the
Kaplan-Meier estimate, and so the standard error of this estimator is ob-
tained in a similar manner. In the notation of Section 2.1.1, the standard
error of the life-table estimate is given by
  12
∑k
dj 
se {S ∗ (t)} ≈ S ∗ (t) .

j=1
n′j (n′j − dj ) 
The standard error of the Nelson-Aalen estimator is

  21
{ } ∑ k
dj 
se S̃(t) ≈ S̃(t) ,
 n2j  j=1
although other expressions have been proposed.
2.2.3 Confidence intervals for values of the survivor function

Once the standard error of an estimate of the survivor function has been
calculated, a confidence interval for the corresponding value of the survivor
function, at a given time t, can be found. A confidence interval is an interval
estimate of the survivor function, and is the interval which is such that there is
a prescribed probability that the value of the true survivor function is included
within it. The intervals constructed in this manner are sometimes referred to
as pointwise confidence intervals, since they apply to a specific survival time.
A confidence interval for the true value of the survivor function at a given
time t is obtained by assuming that the estimated value of the survivor func-
tion at t is normally distributed with mean S(t) and estimated variance
given by Equation (2.11). The interval is computed from percentage points
of the standard normal distribution. Thus, if Z is a random variable that
has a standard normal distribution, the upper (one-sided) α/2-point, or the
two-sided α-point, of this distribution is that value zα/2 which is such that
P(Z > zα/2 ) = α/2. This probability is the area under the standard normal
curve to the right of zα/2 , as illustrated in Figure 2.5. For example, the two-
sided 5% and 1% points of the standard normal distribution, z0.025 and z0.005 ,
are 1.96 and 2.58, respectively.
A 100(1 − α)% confidence interval for S(t), for a given value of t, is the
interval from Ŝ(t) − zα/2 se {Ŝ(t)} to Ŝ(t) + zα/2 se {Ŝ(t)}, where se {Ŝ(t)} is
found from Equation (2.12). These intervals for S(t) can be superimposed on
a graph of the estimated survivor function, as shown in Example 2.5.
One difficulty with this procedure arises from the fact that the confidence
intervals are symmetric. When the estimated survivor function is close to
zero or unity, symmetric intervals are inappropriate, since they can lead to
confidence limits for the survivor function that lie outside the interval (0,1).
A pragmatic solution to this problem is to replace any limit that is greater
than unity by 1.0, and any limit that is less than zero by 0.0.
An alternative procedure is to transform Ŝ(t) to a value in the range
(−∞, ∞), and obtain a confidence interval for the transformed value. The
-z 0 z
Value of z
Figure 2.5 Upper and lower α/2-points of the standard normal distribution.
resulting confidence limits are then back-transformed to give a confidence

interval for S(t) itself. Possible transformations are the logistic transforma-
tion, log[S(t)/{1 − S(t)}], and the complementary log-log transformation,
log{− log S(t)}. Note that from Equation (1.8), the latter quantity is the
logarithm of the cumulative hazard function. In either case, the standard
error of the transformed value of Ŝ(t) can be found using the approximation
in Equation (2.8).
For example, the variance of log{− log Ŝ(t)} is obtained from the expres-
sion for var {log Ŝ(t)} in Equation (2.10). Using the general result in Equa-
tion (2.8),
1
var {log(−X)} ≈ 2 var (X),
X
and setting X = log Ŝ(t) gives
[ ] 1 ∑
k
dj
var log{− log Ŝ(t)} ≈ .
{log Ŝ(t)}2 j=1
nj (nj − dj )
The standard error of log{− log Ŝ(t)} is the square root of this quantity. This
leads to 100(1 − α)% limits of the form
Ŝ(t)exp[±zα/2 se{log[− log Ŝ(t)]}] ,

where zα/2 is the upper α/2-point of the standard normal distribution.
A further problem is that in the tails of the distribution of the survival
times, that is, when Ŝ(t) is close to zero or unity, the variance of Ŝ(t) obtained
using Greenwood’s formula can underestimate the actual variance. In these
circumstances, an alternative expression for the standard error of Ŝ(t) may
be used. Peto et al. (1977) propose that the standard error of Ŝ(t) should be
obtained from the equation
√
Ŝ(t) {1 − Ŝ(t)}
se {Ŝ(t)} = √ ,
(nk )
for t(k) 6 t < t(k+1) , k = 1, 2, . . . , r, where Ŝ(t) is the Kaplan-Meier estimate
of S(t) and nk is the number of individuals at risk at t(k) , the start of the kth
constructed time interval.
This expression for the standard error of Ŝ(t) is conservative, in the sense
that the standard errors obtained will tend to be larger than they ought to
be. For this reason, the Greenwood estimate is recommended for general use.

The standard error of the estimated survivor function, and 95% confidence
limits for the corresponding true value of the function, for the data from
Example 1.1 on the times to discontinuation of use of an IUD, are given in
Table 2.4. In this table, confidence limits outside the range (0, 1) have been
replaced by zero or unity.
Table 2.4 Standard error of Ŝ(t) and confidence intervals for S(t)
for the data from Example 1.1.
Time interval Ŝ(t) se {Ŝ(t)} 95% confidence interval
0– 1.0000 0.0000
10– 0.9444 0.0540 (0.839, 1.000)
19– 0.8815 0.0790 (0.727, 1.000)
30– 0.8137 0.0978 (0.622, 1.000)
36– 0.7459 0.1107 (0.529, 0.963)
59– 0.6526 0.1303 (0.397, 0.908)
75– 0.5594 0.1412 (0.283, 0.836)
93– 0.4662 0.1452 (0.182, 0.751)
97– 0.3729 0.1430 (0.093, 0.653)
107 0.2486 0.1392 (0.000, 0.522)
From this table we see that, in general, the standard error of the esti-
mated survivor function increases with the discontinuation time. The reason
for this is that estimates of the survivor function at later times are based on
fewer individuals. A graph of the estimated survivor function, with the 95%
confidence limits shown as dashed lines, is given in Figure 2.6.
It is important to observe that the confidence limits for a survivor func-
tion, illustrated in Figure 2.6, are only valid for any given time. Different
methods are needed to produce confidence bands that are such that there is
a given probability, 0.95 for example, that the survivor function is contained
in the band for all values of t. These bands will tend to be wider than the
band formed from the pointwise confidence limits. Details will not be included,
but references to these methods are given in the final section of this chapter.
Notice also that the width of these intervals is very much greater than the
difference between the Kaplan-Meier and Nelson-Aalen estimates of the sur-
vivor function, shown in Tables 2.2 and 2.3. Similar calculations lead to con-
fidence limits based on life-table and Nelson-Aalen estimates of the survivor
function.
ESTIMATING THE HAZARD FUNCTION 31
Estimated survivor function 1.0
0.8
0.6
0.4
0.2
0.0
0 20 40 60 80 100 120
Figure 2.6 Estimated survivor function and 95% confidence limits for S(t).
2.3 Estimating the hazard function

A single sample of survival data may also be summarised through the hazard
function, which shows the dependence of the instantaneous risk of death on
time. There are a number of ways of estimating this function, two of which
are described in this section.
2.3.1 Life-table estimate of the hazard function

Suppose that the observed survival times have been grouped into a series of
m intervals, as in the construction of the life-table estimate of the survivor
function. An appropriate estimate of the average hazard of death per unit
time over each interval is the observed number of deaths in that interval,
divided by the average time survived in that interval. This latter quantity is
the average number of persons at risk in the interval, multiplied by the length
of the interval. Let the number of deaths in the jth time interval be dj , j =
1, 2, . . . , m, and suppose that n′j is the average number of individuals at risk
of death in that interval, where n′j is given by Equation (2.2). Assuming that
the death rate is constant during the jth interval, the average time survived
in that interval is (n′j − dj /2)τj , where τj is the length of the jth time interval.
The life-table estimate of the hazard function in the jth time interval is then
given by
dj
h∗ (t) = ′ ,
(nj − dj /2)τj
for t′j−1 6 t < t′j , j = 1, 2, . . . , m, so that h∗ (t) is a step-function.
The asymptotic standard error of this estimate has been shown by Gehan
(1969) to be given by
√
h∗ (t) {1 − [h∗ (t)τj /2]2 }
se {h∗ (t)} = √ ,
(dj )
and confidence intervals for the corresponding true hazard over each of the m
time intervals can be obtained in the manner described in Section 2.2.3.
The life-table estimate of the survivor function for the data from Example 1.3
on the survival times of 48 multiple myeloma patients was given in Table 2.1.
Using the same time intervals as were used in Example 2.2, calculations leading
to the life-table estimate of the hazard function are given in Table 2.5.
Table 2.5 Life-table estimate of the hazard function for

the data from Example 1.3.
Time period τj dj n′j h∗ (t)
0– 12 16 46.0 0.0351
12– 12 10 26.0 0.0397
24– 12 1 14.0 0.0062
36– 12 3 12.5 0.0227
48– 12 2 8.0 0.0238
60– 36 4 4.5 0.0444
The estimated hazard function is plotted as a step-function in Figure 2.7.

The general pattern is for the hazard to remain roughly constant over the
first two years from diagnosis, after which time it declines and then increases
gradually. However, some caution is needed in interpreting this estimate, as
there are few deaths two years after diagnosis.
2.3.2 Kaplan-Meier type estimate

A natural way of estimating the hazard function for ungrouped survival data
is to take the ratio of the number of deaths at a given death time to the
number of individuals at risk at that time. If the hazard function is assumed
to be constant between successive death times, the hazard per unit time can
be found by further dividing by the time interval. Thus, if there are dj deaths
at the jth death time, t(j) , j = 1, 2, . . . , r, and nj at risk at time t(j) , the
hazard function in the interval from t(j) to t(j+1) can be estimated by
dj
ĥ(t) = , (2.13)
nj τj
for t(j) 6 t < t(j+1) , where τj = t(j+1) − t(j) . Notice that it is not possible to
use Equation (2.13) to estimate the hazard in the interval that begins at the
final death time, since this interval is open-ended.
0.05
0.04
Estimated hazard function
0.03
0.02
0.01
0.00
0 10 20 30 40 50 60 70
Survival time
Figure 2.7 Life-table estimate of the hazard function for the data from Example 1.3.
The estimate in Equation (2.13) is referred to as a Kaplan-Meier type esti-

mate, because the estimated survivor function derived from it is the Kaplan-
Meier estimate. To show this, note that since ĥ(t), t(j) 6 t < t(j+1) , is an
estimate of the risk of death per unit time in the jth interval, the probabil-
ity of death in that interval is ĥ(t)τj , that is, dj /nj . Hence an estimate of
the corresponding survival probability in that interval is 1 − (dj /nj ), and the
estimated survivor function is as given by Equation (2.4).
The approximate standard error of ĥ(t) can be found from the variance
of dj , which, following Section 2.2.1, may be assumed to have a binomial
distribution with parameters nj and pj , where pj is the probability of death in
the interval of length τ . Consequently, var (dj ) = nj pj (1 − pj ), and estimating
pj by dj /nj gives
√
nj − d j
se {ĥ(t)} = ĥ(t) .
nj dj
However, when dj is small, confidence intervals constructed using this standard

error will be too wide to be of practical use.

Consider again the data on the time to discontinuation of the use of an IUD
for 18 women, given in Example 1.1. The Kaplan-Meier estimate of the sur-
vivor function for these data was given in Table 2.2, and Table 2.6 gives the
corresponding Kaplan-Meier type estimate of the hazard function, computed
from Equation (2.13). The approximate standard errors of ĥ(t) are also given.
Table 2.6 Kaplan-Meier type estimate of the hazard

function for the data from Example 1.1.
Time interval τj nj dj ĥ(t) se {ĥ(t)}
0– 10 18 0 0.0000 –
10– 9 18 1 0.0062 0.0060
19– 11 15 1 0.0061 0.0059
30– 6 13 1 0.0128 0.0123
36– 23 12 1 0.0036 0.0035
59– 16 8 1 0.0078 0.0073
75– 18 7 1 0.0079 0.0073
93– 4 6 1 0.0417 0.0380
97– 10 5 1 0.0200 0.0179
0.05
0.04
0.03
0.02
0.01
0.00
0 20 40 60 80 100 120
Figure 2.8 Kaplan-Meier type estimate of the hazard function for the data from
Example 1.1.
Figure 2.8 shows a plot of the estimated hazard function. From this figure,
there is some evidence that the longer the IUD is used, the greater is the risk of
discontinuation, but the picture is not very clear. The approximate standard
errors of the estimated hazard function at different times are of little help in
interpreting this plot.
In practice, estimates of the hazard function obtained in this way will often
tend to be rather irregular. For this reason, plots of the hazard function may be
‘smoothed’, so that any pattern can be seen more clearly. There are a number
of ways of smoothing the hazard function, that lead to a weighted average of
values of the estimated hazard ĥ(t) at death times in the neighbourhood of t.
For example, a kernel smoothed estimate of the hazard function, based on the
r ordered death times, t(1) , t(2) , . . . , t(r) , with dj deaths and nj at risk at time
t(j) , can be found from
{ ( )2 }
∑
r
t − t(j) dj
† −1
h (t) = b 0.75 1 − ,
j=1
b nj
where the value of b needs to be chosen. The function h† (t) is defined for
all values of t in the interval from b to t(r) − b, where t(r) is the greatest
death time. For any value of t in this interval, the death times in the interval
(t − b, t + b) will contribute to the weighted average. The parameter b is known
as the bandwidth and its value controls the shape of the plot; the larger the
value of b, the greater the degree of smoothing. There are formulae that lead
to ‘optimal’ values of b, but these tend to be rather cumbersome. Fuller details
can be found in the references provided in the final section of this chapter.
In this book, the use of a modelling approach to the analysis of survival data
is advocated, and so model-based estimates of the hazard function will be
considered in subsequent chapters.
2.3.3 Estimating the cumulative hazard function

The interpretation of the cumulative hazard function in terms of the expected
number of events that occur up to a given time, given in Section 1.3.3 of
Chapter 1, means that this function is important in the identification of models
for survival data, as will be seen later in Sections 4.4 and 5.2. In addition, since
the derivative of the cumulative hazard function is the hazard function itself,
the slope of the cumulative hazard function provides information about the
shape of the underlying hazard function. For example, a linear cumulative
hazard function over some time interval suggests that the hazard is constant
over this interval. Methods that can be used to estimate this function will now
be described.
The cumulative hazard at time t, H(t), was defined in Equation (1.7)
to be the integral of the hazard function, but is more conveniently found
using Equation (1.8). According to this result, H(t) = − log S(t), and
so if Ŝ(t) is the Kaplan-Meier estimate of the survivor function, Ĥ(t) =
− log Ŝ(t) is an appropriate estimate of the cumulative hazard function to
time t.
Now, using Equation (2.4),
∑
k ( )
nj − dj
Ĥ(t) = − log ,
j=1
nj
for t(k) 6 t < t(k+1) , k = 1, 2, . . . , r, and t(1) , t(2) , . . . , t(r) are the r ordered
death times, with t(r+1) = ∞.
If the Nelson-Aalen estimate of the survivor function is used, the estimated
cumulative hazard function, H̃(t) = − log S̃(t), is given by
∑k
dj
H̃(t) = .
n
j=1 j
This is the cumulative sum of the estimated probabilities of death from the
first to the kth time interval, k = 1, 2, . . . , r, and so this quantity has imme-
diate intuitive appeal as an estimate of the cumulative hazard.
An estimate of the cumulative hazard function also leads to an estimate
of the corresponding hazard function, since the differences between adjacent
values of the estimated cumulative hazard function provide estimates of the
underlying hazard, after dividing by the time interval. In particular, differences
in adjacent values of the Nelson-Aalen estimate of the cumulative hazard lead
directly to the hazard function estimate in Section 2.3.2.
2.4 Estimating the median and percentiles of survival times

Since the distribution of survival times tends to be positively skewed, the
median is the preferred summary measure of the location of the distribution.
Once the survivor function has been estimated, it is straightforward to obtain
an estimate of the median survival time. This is the time beyond which 50%
of the individuals in the population under study are expected to survive, and
is given by that value t(50) which is such that S{t(50)} = 0.5.
Because the non-parametric estimates of S(t) are step-functions, it will
not usually be possible to realise an estimated survival time that makes the
survivor function exactly equal to 0.5. Instead, the estimated median survival
time, t̂(50), is defined to be the smallest observed survival time for which the
value of the estimated survivor function is less than 0.5.
In mathematical terms,
t̂(50) = min{ti | Ŝ(ti ) < 0.5},
where ti is the observed survival time for the ith individual, i = 1, 2, . . . , n.

Since the estimated survivor function only changes at a death time, this is
equivalent to the definition
t̂(50) = min{t(j) | Ŝ(t(j) ) < 0.5},
where t(j) is the jth ordered death time, j = 1, 2, . . . , r.

In the particular case where the estimated survivor function is exactly
equal to 0.5 for values of t in the interval from t(j) to t(j+1) , the median
is taken to be the half-way point in this interval, that is (t(j) + t(j+1) )/2.
When there are no censored survival times, the estimated median survival
time will be the smallest time beyond which 50% of the individuals in the
sample survive.
ESTIMATING THE PERCENTILES OF SURVIVAL TIMES 37
The Kaplan-Meier estimate of the survivor function for the data from Ex-
ample 1.1 on the time to discontinuation of the use of an IUD was given in
Table 2.2. The estimated survivor function, Ŝ(t), for these data was shown in
Figure 2.4. From the estimated survivor function, the smallest discontinuation
time beyond which the estimated probability of discontinuation is less than
0.5 is 93 weeks. This is therefore the estimated median time to discontinuation
of the IUD for this group of women.
A similar procedure to that described above can be used to estimate other

percentiles of the distribution of survival times. The pth percentile of the
distribution of survival times is defined to be the value t(p) which is such that
F {t(p)} = p/100, for any value of p from 0 to 100. In terms of the survivor
function, t(p) is such that S{t(p)} = 1 − (p/100), so that for example the 10th
and 90th percentiles are given by
S{t(10)} = 0.9, S{t(90)} = 0.1,
respectively. Using the estimated survivor function, the estimated pth per-
centile is the smallest observed survival time, t̂(p), for which Ŝ{t̂(p)} <
1 − (p/100).
It sometimes happens that the estimated survivor function is greater than
0.5 for all values of t. In such cases, the median survival time cannot be
estimated. It would then be natural to summarise the data in terms of other
percentiles of the distribution of survival times, or the estimated survival
probabilities at particular time points.
Estimates of the dispersion of a sample of survival data are not widely
used, but should such an estimate be required, the semi-interquartile range
(SIQR) can be calculated. This is defined to be half the difference between
the 75th and 25th percentiles of the distribution of survival times. Hence,
1
SIQR = {t(75) − t(25)} ,
2
where t(25) and t(75) are the 25th and 75th percentiles of the survival time
distribution. These two percentiles are also known as the first and third quar-
tiles, respectively. The corresponding sample-based estimate of the SIQR is
{t̂(75) − t̂(25)}/2. Like the variance, the larger the value of the SIQR, the
more dispersed is the survival time distribution.

From the Kaplan-Meier estimate of the survivor function for the data from
Example 1.1, given in Table 2.2, the 25th and 75th percentiles of the distri-
bution of discontinuation times are 36 and 107 weeks, respectively. Hence, the
SIQR of the distribution is estimated to be 35.5 weeks.
2.5 ∗ Confidence intervals for the median and percentiles
Approximate confidence intervals for the median and other percentiles of a
distribution of survival times can be found once the variance of the estimated
percentile has been obtained. An expression for the approximate variance of
a percentile can be derived from a direct application of the general result for
the variance of a function of a random variable in Equation (2.8). Using this
result,
( )2
dŜ{t(p)}
var [Ŝ{t(p)}] = var {t(p)}, (2.14)
dt(p)
where t(p) is the pth percentile of the distribution and Ŝ{t(p)} is the Kaplan-
Meier estimate of the survivor function at t(p). Now,
dŜ{t(p)}
− = fˆ{t(p)},
dt(p)
an estimate of the probability density function of the survival times at t(p),
and on rearranging Equation (2.14), we get
( )2
1
var {t(p)} = var [Ŝ{t(p)}].
fˆ{t(p)}
The standard error of t̂(p), the estimated pth percentile, is therefore given by
1
se {t̂(p)} = se [Ŝ{t̂(p)}]. (2.15)
fˆ{t̂(p)}
The standard error of Ŝ{t̂(p)} is found using Greenwood’s formula for the
standard error of the Kaplan-Meier estimate of the survivor function, given in
Equation (2.12), while an estimate of the probability density function at t̂(p)
is
Ŝ{û(p)} − Ŝ{ˆl(p)}
fˆ{t̂(p)} = ,
ˆl(p) − û(p)
where { p }
û(p) = max t(j) | Ŝ(t(j) ) > 1 − +ϵ ,
100
and { }
ˆl(p) = min t(j) | Ŝ(t(j) ) 6 1 − p − ϵ ,
100
for j = 1, 2, . . . , r, and small values of ϵ. In many cases, taking ϵ = 0.05 will
be satisfactory, but a larger value of ϵ will be needed if û(p) and ˆl(p) turn
out to be equal. In particular, from Equation (2.15), the standard error of the
median survival time is given by
1
se {t̂(50)} = se [Ŝ{t̂(50)}], (2.16)
ˆ
f {t̂(50)}
CONFIDENCE INTERVALS FOR THE PERCENTILES 39
where fˆ{t̂(50)} can be found from
Ŝ{û(50)} − Ŝ{ˆl(50)}
fˆ{t̂(50)} = . (2.17)
ˆl(50) − û(50)
In this expression, û(50) is the largest survival time for which the Kaplan-
Meier estimate of the survivor function exceeds 0.55, and ˆl(50) is the smallest
survival time for which the survivor function is less than or equal to 0.45.
Once the standard error of the estimated pth percentile has been found, a
100(1 − α)% confidence interval for t(p) has limits of
t̂(p) ± zα/2 se {t̂(p)},
where zα/2 is the upper (one-sided) α/2-point of the standard normal distri-
bution.
This interval estimate is only approximate, in the sense that the probability
that the interval includes the true percentile will not be exactly 1 − α. A
number of methods have been proposed for constructing confidence intervals
for the median with superior properties, although these alternatives are more
difficult to compute than the interval estimate derived in this section.

The data on the discontinuation times for users of an IUD, given in Example
1.1, are now used to illustrate the calculation of a confidence interval for
the median discontinuation time. From Example 2.8, the estimated median
discontinuation time for this group of women is given by t̂(50) = 93 weeks.
Also, from Table 2.4, the standard error of the Kaplan-Meier estimate of the
survivor function at this time is given by se [Ŝ{t̂(50)}] = 0.1452.
To obtain the standard error of t̂(50) using Equation (2.16), we need an
estimate of the density function at the estimated median discontinuation time.
This is obtained from Equation (2.17). The quantities û(50) and ˆl(50) needed
in this equation are such that
û(50) = max{t(j) | Ŝ(t(j) ) > 0.55},
and
ˆl(50) = min{t(j) | Ŝ(t(j) ) 6 0.45},
where t(j) is the jth ordered discontinuation time, j = 1, 2, . . . , 9. Using Ta-

ble 2.4, û(50) = 75 and ˆl(50) = 97, and so
Ŝ(75) − Ŝ(97) 0.5594 − 0.3729

fˆ{t̂(50)} = = = 0.0085.
97 − 75 22
Then, the standard error of the median is given by
1
se {t̂(50)} = × 0.1452 = 17.13.
0.0085
A 95% confidence interval for the median discontinuation time has limits of
93 ± 1.96 × 17.13,
and so the required interval estimate for the median ranges from 59 to 127
days.
2.6 Comparison of two groups of survival data

The simplest way of comparing the survival times obtained from two groups of
individuals is to plot the corresponding estimates of the two survivor functions
on the same axes. The resulting plot can be quite informative, as the following
example illustrates.

Data on the survival times of women with breast cancer, grouped according to
whether or not sections of a tumour were positively stained with Helix pomatia
agglutinin (HPA), were given in Example 1.2. The Kaplan-Meier estimate of
the survivor function, for each of the two groups of survival times, is plotted
in Figure 2.9. Notice that in this figure, the Kaplan-Meier estimates extend
to the time of the largest censored observation in each group.
1.0
0.8
0.6
0.4
0.2
0.0
0 50 100 150 200 250
Survival time
Figure 2.9 Kaplan-Meier estimate of the survivor functions for women with tumours
that were positively stained (—) and negatively stained (·······).
This figure shows that the estimated survivor function for those women
with negatively stained tumours is always greater than that for women with
positively stained tumours. This means that at any time t, the estimated
probability of survival beyond t is greater for women with negative staining,
COMPARISON OF TWO GROUPS OF SURVIVAL DATA 41
suggesting that the result of the HPA staining procedure might be a useful
prognostic indicator. In particular, those women whose tumours are positively
stained appear to have a poorer prognosis than those with negatively stained
tumours.
There are two possible explanations for an observed difference between two
estimated survivor functions, such as those in Example 2.11. One explanation
is that there is a real difference between the survival times of the two groups
of individuals, so that those in one group have a different survival experience
from those in the other. An alternative explanation is that there are no real
differences between the survival times in each group, and that the difference
that has been observed is merely the result of chance variation. To help dis-
tinguish between these two possible explanations, we use a procedure known
as the hypothesis test. Because the concept of the hypothesis test has a central
role in the analysis of survival data, the underlying basis for this procedure is
described in detail in the following section.
2.6.1 Hypothesis testing
The hypothesis test is a procedure that enables us to assess the extent to which
an observed set of data are consistent with a particular hypothesis, known as
the working or null hypothesis. A null hypothesis generally represents a sim-
plified view of the data-generating process, and is typified by hypotheses that
specify that there is no difference between two groups of survival data, or that
there is no relationship between survival time and explanatory variables such
as age or serum cholesterol level. The null hypothesis is then the hypothesis
that will be adopted, and subsequently acted upon, unless the data indicate
that it is untenable.
The next step is to formulate a test statistic that measures the ex-
tent to which the observed data depart from the null hypothesis. In gen-
eral, the test statistic is so constructed that the larger the value of the
statistic, the greater the departure from the null hypothesis. Hence, if the
null hypothesis is that there is no difference between two groups, relatively
large values of the test statistic will be interpreted as evidence against this
null hypothesis.
Once the value of the test statistic has been obtained from the observed
data, we calculate the probability of obtaining a value as extreme or more
extreme than the observed value, when the null hypothesis is true. This quan-
tity summarises the strength of the evidence in the sample data against the
null hypothesis, and is known as the probability value, or P-value for short.
If the P -value is large, we would conclude that it is quite likely that the ob-
served data would have been obtained when the null hypothesis was true, and
that there is no evidence to reject the null hypothesis. On the other hand, if
the P -value is small, this would be interpreted as evidence against the null
hypothesis; the smaller the P -value, the stronger the evidence.
In order to obtain the P -value for a hypothesis test, the test statistic must
have a probability distribution that is known, or at least approximately known,
when the null hypothesis is true. This probability distribution is referred to
as the null distribution of the test statistic. More specifically, consider a test
statistic, W , which is such that the larger the observed value of the test
statistic, w, the greater the deviation of the observed data from that expected
under the null hypothesis. If W has a continuous probability distribution,
the P -value is then P(W > w) = 1 − F (w), where F (w) is the distribution
function of W , under the null hypothesis, evaluated at w.
In some applications, the most natural test statistic is one for which large
positive values correspond to departures from the null hypothesis in one di-
rection, while large negative values correspond to departures in the opposite
direction. For example, suppose that patients suffering from a particular ill-
ness have been randomised to receive either a standard treatment or a new
treatment, and their survival times are recorded. In this situation, a null hy-
pothesis of interest will be that there is no difference in the survival experience
of the patients in the two treatment groups. The extent to which the data are
consistent with this null hypothesis might then be summarised by a test statis-
tic for which positive values indicate that the new treatment is superior to the
standard, while negative values indicate that the standard treatment is supe-
rior. When departures from the null hypothesis in either direction are equally
important, the null hypothesis is said to have a two-sided alternative, and the
hypothesis test itself is referred to as a two-sided test.
If W is a test statistic for which large positive or large negative observed
values lead to rejection of the null hypothesis, a new test statistic, such as
|W | or W 2 , can be defined, so that only large positive values of the new
statistic indicate that there is evidence against the null hypothesis. For ex-
ample, suppose that W is a test statistic that under the null hypothesis has
a standard normal distribution. If w is the observed value of W , the appro-
priate P -value is P(W 6 − |w|) + P(W > |w|), which in view of the sym-
metry of the standard normal distribution, is 2 P(W > |w|). Alternatively,
we can make use of the result that if W has a standard normal distribu-
tion, W 2 has a chi-squared distribution on one degree of freedom, written χ21 .
Thus, a P -value for the two-sided hypothesis test based on the statistic W is
the probability that a χ21 random variable exceeds w2 . The required P -value
can therefore be found from the standard normal or chi-squared distribution
functions.
When interest centres on departures in a particular direction, the hypoth-
esis test is said to be one-sided. For example, in comparing the survival times
of two groups of patients where one group receives a standard treatment and
the other group a new treatment, it might be argued that the new treatment
cannot possibly be inferior to the standard. Then, the only relevant alternative
to the null hypothesis of no treatment difference is that the new treatment is
superior. If positive values of the test statistic W reflect the superiority of the
new treatment, the P -value is then P(W > w). If W has a standard normal
distribution, this P -value is half of that which would have been obtained for
the corresponding two-sided alternative hypothesis.
A one-sided hypothesis test can only be appropriate when there is no inter-
est whatsoever in departures from the null hypothesis in the opposite direction
to that specified in the one-sided alternative. For example, consider again the
comparison of a new treatment with a standard treatment, and suppose that
the observed value of the test statistic is either positive or negative, depending
on whether the new treatment is superior or inferior to the standard. If the
alternative to the null hypothesis of no treatment difference is that the new
treatment is superior, a large negative value of the test statistic would not be
regarded as evidence against the null hypothesis. Instead, it would be assumed
that this large negative value is simply the result of chance variation. Gen-
erally speaking, the use of one-sided tests can rarely be justified in medical
research, and so two-sided tests will be used throughout this book.
If a P -value is smaller than some value α, we say that the hypothesis is
rejected at the 100α% level of significance. The observed value of the test
statistic is then said to be significant at this level. But how do we decide on
the basis of the P -value whether or not a null hypothesis should actually be
rejected? Traditionally, P -values of 0.05 or 0.01 have been used in reaching a
decision about whether or not a null hypothesis should be rejected, so that if
P < 0.05, for example, the null hypothesis is rejected at the 5% significance
level. Guidelines such as these are not hard-and-fast rules and should not be
interpreted rigidly. For example, there is no practical difference between a
P -value of 0.046 and 0.056, even though only the former indicates that the
observed value of the test statistic is significant at the 5% level.
Instead of reporting that a null hypothesis is rejected or not rejected at
some specified significance level, a more satisfactory policy is to report the
actual P -value. This P -value can then be interpreted as a measure of the
strength of evidence against the null hypothesis, using a vocabulary that de-
pends on the range within which the P -value lies. Thus, if P > 0.1, there is
said to be no evidence to reject the null hypothesis; if 0.05 < P 6 0.1, there is
slight evidence against the null hypothesis; if 0.01 < P 6 0.05, there is mod-
erate evidence against the null hypothesis; if 0.001 < P 6 0.01, there is strong
evidence against the null hypothesis, and if P 6 0.001, the evidence against
the null hypothesis is overwhelming.
An alternative to quoting the exact P -value associated with a hypothesis
test is to compare the observed value of the test statistic with those values
that would correspond to particular P -values, when the null hypothesis is
true. Values of the test statistic that lead to rejection of the null hypothesis at
particular levels of significance can be found from percentage points of the null
distribution of that statistic. In particular, if W is a test statistic that has a
standard normal distribution, for a two-sided test, the upper α/2-point of the
distribution, depicted in Figure 2.5, is the value of the test statistic for which
the P -value is α. For example, values of the test statistic of 1.96, 2.58 and 3.29
correspond to P -values of 0.05, 0.01 and 0.001. Thus, if the observed value of
W were between 1.96 and 2.58, we would declare that 0.01 < P < 0.05. On
the other hand, if the null distribution of W is chi-squared on one degree of
freedom, the upper α-point of the distribution is the value of the test statistic
which would give a P -value of α. Then, values of the test statistic of 3.84,
6.64 and 10.83 correspond to P -values of 0.05, 0.01 and 0.001, respectively.
Notice that these values are simply the squares of those for the standard
normal distribution, which they must be in view of the fact that the square
of a standard normal random variable has a chi-squared distribution on one
degree of freedom.
For commonly encountered probability distributions, such as the normal
and chi-squared, percentage points are tabulated in many introductory text
books on statistics, or in statistical tables such as those of Lindley and Scott
(1984). Statistical software packages used in computer-based statistical analy-
ses of survival data usually provide the exact P -values associated with hypoth-
esis tests as a matter of course. Note that when these are rounded off to, say,
three decimal places, a P -value of 0.000 should be interpreted as P < 0.001.
In deciding on a course of action, such as whether or not to reject the
hypothesis that there is no difference between two treatments, the statistical
evidence summarised in the P -value for the hypothesis test will be just one
ingredient of the decision-making process. In addition to the statistical evi-
dence, there will also be scientific evidence to consider. This may, for example,
concern whether the size of the treatment effect is clinically important. In par-
ticular, in a large trial, a difference between two treatments that is significant
at, say, the 5% level may be found when the magnitude of the treatment effect
is so small that it does not indicate a major scientific breakthrough. On the
other hand, a new formulation of a treatment may prolong life by a factor of
two, and yet, because of small sample sizes used in the study, may not appear
to be significantly different from the standard.
Rather than report findings in terms of the results of a hypothesis testing
procedure, it is more informative to provide an estimate of the size of any
treatment difference, supported by a confidence interval for this difference.
Unfortunately, the non-parametric approaches to the analysis of survival data
being considered in this chapter do not lend themselves to this approach. We
will therefore return to this theme in subsequent chapters when we consider
models for survival data.
In the comparison of two groups of survival data, there are a number of
methods that can be used to quantify the extent of between-group differences.
Two non-parametric procedures will now be considered, namely the log-rank
test and the Wilcoxon test.
2.6.2 The log-rank test

In order to construct the log-rank test, we begin by considering separately
each death time in two groups of survival data. These groups will be labelled
Group I and Group II. Suppose that there are r distinct death times, denoted
t(1) < t(2) < · · · < t(r) , across the two groups, and that at time t(j) , d1j
individuals in Group I and d2j individuals in Group II die, for j = 1, 2, . . . , r.
Unless two or more individuals in a group have the same recorded death time,
the values of d1j and d2j will either be zero or unity. Suppose further that
there are n1j individuals at risk of death in the first group just before time
t(j) , and that there are n2j at risk in the second group. Consequently, at time
t(j) , there are dj = d1j + d2j deaths in total out of nj = n1j + n2j individuals
at risk. The situation is summarised in Table 2.7.
Table 2.7 Number of deaths at the jth death time in each of two
groups of individuals.
Group Number of Number surviving Number at risk
deaths at t(j) beyond t(j) just before t(j)
I d1j n1j − d1j n1j
II d2j n2j − d2j n2j
Total dj nj − dj nj
Now consider the null hypothesis that there is no difference in the survival
experience of the individuals in the two groups. One way of assessing the
validity of this hypothesis is to consider the extent of the difference between
the observed number of individuals in the two groups who die at each of the
death times, and the numbers expected under the null hypothesis. Information
about the extent of these differences can then be combined over each of the
death times.
If the marginal totals in Table 2.7 are regarded as fixed, and the null
hypothesis that survival is independent of group is true, the four entries in
this table are solely determined by the value of d1j , the number of deaths at
t(j) in Group I. We can therefore regard d1j as a random variable, which can
take any value in the range from 0 to the minimum of dj and n1j . In fact,
d1j has a distribution known as the hypergeometric distribution, according to
which the probability that the random variable associated with the number
of deaths in the first group takes the value d1j is
( )( )
dj nj − dj
d1j n1j − d1j
( ) . (2.18)
nj
n1j
In this formula, the expression
( )
dj
d1j
represents the number of different ways in which d1j times can be chosen from
dj times and is read as ‘dj C d1j ’. It is given by
( )
dj dj !
= ,
d1j d1j !(dj − d1j )!
where dj !, read as ‘dj factorial ’, is such that
dj ! = dj × (dj − 1) × · · · × 2 × 1.
The other two terms in Expression (2.18) are interpreted in a similar manner.
The mean of the hypergeometric random variable d1j is given by
e1j = n1j dj /nj , (2.19)
so that e1j is the expected number of individuals who die at time t(j) in
Group I. This value is intuitively appealing, since under the null hypothesis
that the probability of death at time t(j) does not depend on the group that
an individual is in, the probability of death at t(j) is dj /nj . Multiplying this
by n1j , gives e1j as the expected number of deaths in Group I at t(j) .
The next step is to combine the information from the individual 2 × 2
tables for each death time to give an overall measure of the deviation of the
observed values of d1j from their expected values. The most straightforward
way of doing this is to sum the differences d1j − e1j over the total number of
death times, r, in the two groups. The resulting statistic is given by
∑
r
UL = (d1j − e1j ). (2.20)
j=1
∑ ∑
Notice that this is d1j − e1j , which is the difference between the total
observed and expected numbers of deaths in Group I. This statistic will have
zero mean, since E (d1j ) = e1j . Moreover, since the death times are indepen-
dent of one another, the variance of UL is simply the sum of the variances of
the d1j . Now, since d1j has a hypergeometric distribution, the variance of d1j
is given by
n1j n2j dj (nj − dj )
v1j = , (2.21)
n2j (nj − 1)
so that the variance of UL is
∑
r
var (UL ) = v1j = VL , (2.22)
j=1
say. Furthermore, it can be shown that UL has an approximate normal dis-

tribution,√when the number of death times is not too small. It then follows
that UL / VL has a normal distribution with zero mean and unit variance,
denoted N (0, 1). We therefore write
UL
√ ∼ N (0, 1),
VL
where the symbol ‘∼’ is read as ‘is distributed as’. The square of a standard
normal random variable has a chi-squared distribution on one degree of free-
dom, denoted χ21 , and so we have that
UL2
∼ χ21 . (2.23)
VL
This method of combining information over a number of 2 × 2 tables was

proposed by Mantel and Haenszel (1959), and is known as the Mantel-Haenszel
procedure. In fact, the test based on this statistic has various names, including
Mantel-Cox and Peto-Mantel-Haenszel, but it is probably best known as the
log-rank test. The reason for this name is that the test statistic can be derived
from the ranks of the survival times in the two groups, and the resulting rank
test statistic is based on the logarithm of the Nelson-Aalen estimate of the
survivor function.
The statistic WL = UL2 /VL summarises the extent to which the observed
survival times in the two groups of data deviate from those expected under
the null hypothesis of no group differences. The larger the value of this statis-
tic, the greater the evidence against the null hypothesis. Because the null
distribution of W is approximately chi-squared with one degree of freedom,
the P -value associated with the test statistic can be obtained from the dis-
tribution function of a chi-squared random variable. Alternatively, percentage
points of the chi-squared distribution can be used to identify a range within
which the P -value lies. An illustration of the log-rank test is presented below
in Example 2.12.

In this example, we return to the data on the survival times of women with
breast cancer, grouped according to whether a section of the tumour was
positively or negatively stained. In particular, the null hypothesis that there
is no difference in the survival experience of the two groups will be examined
using the log-rank test. The required calculations are laid out in Table 2.8.
We begin by ordering the observed death times across the two groups
of women; these times are given in column 1 of Table 2.8. The numbers of
women in each group who die at each death time and the numbers who are
at risk at each time are then calculated. These values are d1j , n1j , d2j and
n2j given in columns 2 to 5 of the table. Columns 6 and 7 contain the total
numbers of deaths and the total numbers of women at risk over the two
groups, at each death time. The final two columns give the values of e1j and
v1j , computed from Equations (2.19)
∑ and (2.21),
∑ respectively. Summing the
entries in columns 2 and 8 gives d∑
1j and ∑e1j , from which the log-rank
∑
statistic can be calculated from UL = d1j − e1j . The value of VL = v1j
can be obtained by summing the entries in the final column. We find that
UL = 5 − 9.565 = −4.565 and VL = 5.929, and so the value of the log-rank
test statistic is WL = (−4.565)2 /5.929 = 3.515.
The corresponding P -value is calculated from the probability that a chi-
squared variate on one degree of freedom is greater than or equal to 3.515,
Table 2.8 Calculation of the log-rank statistic for the data from Example 1.2.
Death time d1j n1j d2j n2j dj nj e1j v1j
5 0 13 1 32 1 45 0.2889 0.2054
8 0 13 1 31 1 44 0.2955 0.2082
10 0 13 1 30 1 43 0.3023 0.2109
13 0 13 1 29 1 42 0.3095 0.2137
18 0 13 1 28 1 41 0.3171 0.2165
23 1 13 0 27 1 40 0.3250 0.2194
24 0 12 1 27 1 39 0.3077 0.2130
26 0 12 2 26 2 38 0.6316 0.4205
31 0 12 1 24 1 36 0.3333 0.2222
35 0 12 1 23 1 35 0.3429 0.2253
40 0 12 1 22 1 34 0.3529 0.2284
41 0 12 1 21 1 33 0.3636 0.2314
47 1 12 0 20 1 32 0.3750 0.2344
48 0 11 1 20 1 31 0.3548 0.2289
50 0 11 1 19 1 30 0.3667 0.2322
59 0 11 1 18 1 29 0.3793 0.2354
61 0 11 1 17 1 28 0.3929 0.2385
68 0 11 1 16 1 27 0.4074 0.2414
69 1 11 0 15 1 26 0.4231 0.2441
71 0 9 1 15 1 24 0.3750 0.2344
113 0 6 1 10 1 16 0.3750 0.2344
118 0 6 1 8 1 14 0.4286 0.2449
143 0 6 1 7 1 13 0.4615 0.2485
148 1 6 0 6 1 12 0.5000 0.2500
181 1 5 0 4 1 9 0.5556 0.2469
Total 5 9.5652 5.9289
and is 0.061, written P = 0.061. This P -value is sufficiently small to cast

doubt on the null hypothesis that there is no difference between the survivor
functions for the two groups of women. In fact, the evidence against the null
hypothesis is nearly significant at the 6% level. We therefore conclude that the
data do provide some evidence that the prognosis of a breast cancer patient
is dependent on the result of the staining procedure.
2.6.3 The Wilcoxon test

The Wilcoxon test, sometimes known as the Breslow test, is also used to test
the null hypothesis that there is no difference in the survivor functions for two
groups of survival data. The Wilcoxon test is based on the statistic
∑
r
UW = nj (d1j − e1j ),
j=1
where, as in the previous section, d1j is the number of deaths at time t(j)
in the first group and e1j is as defined in Equation (2.19). The difference
between UW and UL is that in the Wilcoxon test, each difference d1j − e1j is
weighted by nj , the total number of individuals at risk at time t(j) . The effect
of this is to give less weight to differences between d1j and e1j at those times
when the total number of individuals who are still alive is small, that is, at
the longest survival times. This statistic is therefore less sensitive than the
log-rank statistic to deviations of d1j from e1j in the tail of the distribution
of survival times.
The variance of the Wilcoxon statistic UW is given by
∑
r
VW = n2j v1j ,
j=1
where v1j is given in Equation (2.21), and so the Wilcoxon test statistic is
2
WW = UW /VW ,
which has a chi-squared distribution on one degree of freedom when the null
hypothesis is true. The Wilcoxon test is therefore conducted in the same man-
ner as the log-rank test.

For the data on the survival times of women with tumours that were positively
or negatively stained, the value of the Wilcoxon statistic is UW = −159, and
the variance of the statistic is VW = 6048.136. The value of the chi-squared
2
statistic, UW /VW , is 4.180, and the corresponding P -value is 0.041. This is
slightly smaller than the P -value for the log-rank test, and on the basis of
this result, we would declare that the difference between the two groups is
significant at the 5% level.
2.6.4 Comparison of the log-rank and Wilcoxon tests

Of the two tests, the log-rank test is the more suitable when the alternative
to the null hypothesis of no difference between two groups of survival times
is that the hazard of death at any given time for an individual in one group
is proportional to the hazard at that time for a similar individual in the
other group. This is the assumption of proportional hazards, which underlies a
number of methods for analysing survival data. For other types of departure
from the null hypothesis, the Wilcoxon test is more appropriate than the
log-rank test for comparing the two survivor functions.
In order to help decide which test is the more suitable in any given situa-
tion, we make use of the result that if the hazard functions are proportional,
the survivor functions for the two groups of survival data do not cross one
another. To show this, suppose that h1 (t) is the hazard of death at time t for
an individual in Group I, and h2 (t) is the hazard at that same time for an in-
dividual in Group II. If these two hazards are proportional, then we can write
h1 (t) = ψh2 (t), where ψ is a constant that does not depend on the time t. In-
tegrating both sides of this expression, multiplying by −1 and exponentiating
gives { ∫ } { ∫ }
t t
exp − h1 (u) du = exp − ψh2 (u) du . (2.24)
0 0
Now, from Equation (1.6),
{ ∫ t }
S(t) = exp − h(u) du ,
0
and so if S1 (t) and S2 (t) are the survivor functions for the two groups of
survival data, from Equation (2.24),
ψ
S1 (t) = {S2 (t)} .
Since the survivor function takes values between zero and unity, this result
shows that S1 (t) is greater than or less than S2 (t), according to whether ψ is
less than or greater than unity, at any time t. This means that if two hazard
functions are proportional, the true survivor functions do not cross. This is a
necessary, but not a sufficient condition for proportional hazards.
An informal assessment of the likely validity of the proportional hazards
assumption can be made from a plot of the estimated survivor functions for
two groups of survival data, such as that shown in Figure 2.9. If the two esti-
mated survivor functions do not cross, the assumption of proportional hazards
may be justified, and the log-rank test is appropriate. Of course, sample-based
estimates of survivor functions may cross even though the corresponding true
hazard functions are proportional, and so some care is needed in the interpre-
tation of such graphs. A more satisfactory graphical method for assessing the
validity of the proportional hazards assumption is described in Section 4.4.1
of Chapter 4.
In summary, unless a plot of the estimated survival functions, or previous
data, indicate that there is good reason to doubt the proportional hazards
assumption, the log-rank test should be used to test the hypothesis of equality
of two survivor functions.
From the graph of the two estimated survivor functions in Figure 2.9, we see
that the survivor function for the negatively stained women always lies above
that for the positively stained women. This suggests that the proportional
hazards assumption is appropriate, and that the log-rank test is more appro-
priate than the Wilcoxon test. However, in this example, there is very little
difference between the results of the two hypothesis tests.
2.7 ∗ Comparison of three or more groups of survival data

Both the log-rank and the Wilcoxon tests can be extended to enable three
or more groups of survival data to be compared. Suppose that the survival
COMPARISON OF THREE OR MORE GROUPS 51
distributions of g groups of survival data are to be compared, for g > 2. We
then define analogues of the U -statistics for comparing the observed numbers
of deaths in groups 1, 2, . . . , g − 1 with their expected values. In an obvious
extension of the notation used in Section 2.6, we obtain
∑r ( )
nkj dj
ULk = dkj − ,
j=1
nj
∑r ( )
nkj dj
UW k = nj dkj − ,
j=1
nj
for k = 1, 2, . . . , g − 1. These quantities are then expressed in the form of a

vector with (g − 1) components, which we denote by U L and U W .
We also need expressions for the variances of the ULk and UW k , and for
the covariance between pairs of values. In particular, the covariance between
ULk and ULk′ is given by
∑r ( )
nkj dj (nj − dj ) nk ′ j
VLkk′ = δkk′ − ,
j=1
nj (nj − 1) nj
for k, k ′ = 1, 2, . . . , g − 1, where δkk′ is such that

{
1 if k = k ′ ,
δkk′ =
0 otherwise.
These terms are then assembled in the form of a variance-covariance matrix,
V L , which is a symmetric matrix that has the variances of the ULk down
the diagonal, and covariance terms in the off-diagonals. For example, in the
comparison of three groups of survival data, this matrix would be given by
( )
VL11 VL12
VL= ,
VL12 VL22
where VL11 and VL22 are the variances of UL1 and UL2 , respectively, and VL12
is their covariance.
Similarly, the variance-covariance matrix for the Wilcoxon statistic is the
matrix V W , whose (k, k ′ )th element is
∑ r ( )
nkj dj (nj − dj ) nk ′ j
VW kk′ = n2j δkk′ − ,
j=1
nj (nj − 1) nj
for k, k ′ = 1, 2, . . . , g − 1.
Finally, in order to test the null hypothesis of no group differences, we
make use of the result that the test statistic U ′L V −1 ′ −1
L U L , or U W V W U W ,
has a chi-squared distribution on (g − 1) degrees of freedom, when the null
hypothesis is true.
Statistical software for the analysis of survival data usually incorporates
this methodology, and because the interpretation of the resulting chi-squared
statistic is straightforward, an example will not be given here.
2.8 Stratified tests
In many circumstances, there is a need to compare two or more sets of survival
data, after taking account of additional variables recorded on each individual.
As an illustration, consider a multicentre clinical trial in which two forms of
chemotherapy are to be compared in terms of their effect on the survival times
of lung cancer patients. Information on the survival times of patients in each
treatment group will be available from each centre. The resulting data are
then said to be stratified by centre.
Individual log-rank or Wilcoxon tests based on the data from each centre
will be informative, but a test that combines information about the treatment
difference in each centre provides a more precise summary of the treatment
effect. A similar situation would arise in attempting to test for treatment
differences when patients are stratified according to variables such as age
group, sex, performance status and other potential risk factors for the disease
under study.
In situations such as those described above, a stratified version of the log-
rank or Wilcoxon test may be employed. Essentially, this involves calculating
the values of the U - and V -statistics for each stratum, and then combining
these values over the strata. In this section, the stratified log-rank test will
be described, but a stratified version of the Wilcoxon test can be obtained in
a similar manner. An equivalent analysis, based on a model for the survival
times, is described in Section 11.2 of Chapter 11.
Let ULk be the value of the log-rank statistic for comparing two treatment
groups, computed from the kth of s strata using Equation (2.20). Also, denote
the variance of the statistic for the kth stratum by VLk , where VLk would be
computed for each stratum using Equation (2.21). The stratified log-rank test
is then based on the statistic
∑s 2
( ULk )
WS = ∑k=1 s , (2.25)
k=1 VLk
which has a chi-squared distribution on one degree of freedom (1 d.f.) under the
null hypothesis that there is no treatment difference. Comparing the observed
value of this statistic with percentage points of the chi-squared distribution
enables the hypothesis of no overall treatment difference to be tested.
Example 2.15 Survival times of patients with melanoma

The aim of a study carried out by the University of Oklahoma Health Sci-
ences Center was to compare two immunotherapy treatments for their ability
to prolong the life of patients suffering from melanoma, a highly malignant
tumour occurring in the skin. For each patient, the tumour was surgically
removed before allocation to Bacillus Calmette-Guérin (BCG) vaccine or to a
vaccine based on the bacterium Corynebacterium parvum (C. parvum).
The survival times of the patients in each treatment group were further
classified according to the age group of the patient. The data, which were
given in Lee and Wang (2013), are shown in Table 2.9.
STRATIFIED TESTS 53
Table 2.9 Survival times of melanoma patients in two treatment groups,

stratified by age group.
21–40 41–60 61–
BCG C. parvum BCG C. parvum BCG C. parvum
19 27* 34* 8 10 25*
24* 21* 4 11* 5 8
8 18* 17* 23* 11*
17* 16* 12*
17* 7 15*
34* 12* 8*
24 8*
8
8*
These data are analysed by first computing the log-rank statistics for com-
paring the survival times of patients in the two treatment groups, separately
for each age group. The resulting values of the U -, V - and W -statistics, found
using Equations (2.20), (2.22) and (2.23), are summarised in Table 2.10.
Table 2.10 Values of the log-rank statistic for

each age group.
Age group UL VL WL
21–40 −0.2571 1.1921 0.055
41–60 0.4778 0.3828 0.596
61– 1.0167 0.6497 1.591
Total 1.2374 2.2246
The values of the WL -statistic are quite similar for the three age groups,
suggesting that the treatment effect is consistent over these groups. Moreover,
none of them are significantly large at the 10% level.
To carry out a stratified log-rank test on these data, we calculate the
WS -statistic defined in Equation (2.25). Using the results in Table 2.10,
1.23742
WS = = 0.688.
2.2246
The observed value of WS is not significant when compared with percentage
points of the chi-squared distribution on 1 d.f. We therefore conclude that after
allowing for the different age groups, there is no significant difference between
the survival times of patients treated with the BCG vaccine and those treated
with C. parvum.
For comparison, when the division of the patients into the different age
groups is ignored, the log-rank test for comparing the two groups of patients
leads to WL = 0.756. The fact that this is so similar to the value that allows for
age group differences suggests that it is not necessary to stratify the patients
by age.
The stratified log-rank test can be extended to compare more than two treat-
ment groups. The resulting formulae render it unsuitable for hand calculation,
but the methodology can be implemented using computer software for sur-
vival analysis. However, this method of taking account of additional variables
is not as flexible as that based on a modelling approach, introduced in the
next chapter, and so further details are not given here.
2.9 Log-rank test for trend

In many applications where three or more groups of survival data are to be
compared, these groups are ordered in some way. For example, the groups
may correspond to increasing doses of a treatment, the stage of a disease, or
the age group of an individual. In comparing these groups using the log-rank
test described in previous sections, it can happen that the analysis does not
lead to a significant difference between the groups, even though the hazard
of death increases or decreases across the groups. A test procedure that uses
information about the ordering of the groups is more likely to lead to a trend
being identified as significant than a standard log-rank test.
The log-rank test for trend across g ordered groups is based on the statistic
∑
g
UT = wk (dk. − ek. ), (2.26)
k=1
where wk is a code assigned to the kth group, k = 1, 2, . . . , g, and

∑
rk ∑
rk
dk. = dkj , ek. = ekj ,
j=1 j=1
are the observed and expected numbers of deaths in the kth group, where
the summation is over the rk death times in that group. Note that the dot
subscript in the notation dk. and ek. stands for summation over the subscript
that the dot replaces. The codes are often taken to be equally spaced to
correspond to a linear trend across the groups. For example, if there are three
groups, the codes might be taken to be 1, 2 and 3, although the equivalent
choice of −1, 0 and 1 does simplify the calculations somewhat. The variance
of UT is given by
∑g
VT = (wk − w̄)2 ek. , (2.27)
k=1
where w̄ is a weighted sum of the quantities wk , in which the expected numbers
of deaths, ek. , are the weights, that is,
∑g
k=1 wk ek.
w̄ = ∑ g .
k=1 ek.
LOG-RANK TEST FOR TREND 55
The statistic WT = UT2 /VT then has a chi-squared distribution on 1 d.f. under
the hypothesis of no trend across the g groups.
Example 2.16 Survival times of patients with melanoma

The log-rank test for trend will be illustrated using the data from Example
2.15 on the survival times of patients suffering from melanoma. For the pur-
pose of this illustration, only the data from those patients allocated to the
BCG vaccine will be used. The log-rank statistic for comparing the survival
times of the patients in the three age groups turns out to be 3.739. When
compared to percentage points of the chi-squared distribution on 2 d.f., this
is not significant (P = 0.154).
We now use the log-rank test for trend to examine whether there is a linear
trend over age. For this, we will take the codes, wk , to be equally spaced, with
values −1, 0 and 1. Some of the calculations required for the log-rank test for
trend are summarised in Table 2.11.
Table 2.11 Values of wk and the observed and ex-

pected numbers of deaths in the three age groups.
Age group wk dk. ek.
21–40 −1 2 3.1871
41–60 0 1 1.1949
61– 1 2 0.6179
The log-rank test for trend is based on the statistic in Equation (2.26), the
value of which is
UT = (d3. − e3. ) − (d1. − e1. ) = 2.5692.
Using the values of the expected numbers of deaths in each group, given in
Table 2.11, the weighted mean of the wk ’s is given by
e3. − e1.
w̄ = = 0.5138.
e1. + e3.
The three values of (wk − w̄)2 are 0.2364, 0.2640 and 2.2917, and, from Equa-
tion (2.27), VT = 2.4849. Finally, the test statistic is
UT2
WT = = 2.656,
VT
which is just about significant at the 10% level (P = 0.103) when judged
against a chi-squared distribution on 1 d.f. We therefore conclude that there
is slight evidence of a linear trend across the age groups.
An alternative method of examining whether there is a trend across the
levels of an ordered categorical variable, based on a modelling approach to
the analysis of survival data, is described and illustrated in Section 3.8.1 of
the next chapter.
2.10 Further reading
The life-table, which underpins the calculation of the life-table estimate of
the survivor function, is widely used in the analysis of data from epidemio-
logical studies. Fuller details of this application can be found in Armitage,
Berry and Matthews (2002), and books on statistical methods in demography
and epidemiology, such as Pollard, Yusuf and Pollard (1990) and Woodward
(2014).
The product-limit estimate of the survivor function has been in use since
the early 1900s. Kaplan and Meier (1958) derived the estimate using the
method of maximum likelihood, which is why the estimate now bears their
name. The properties of the Kaplan-Meier estimate of the survivor function
have been further explored by Breslow and Crowley (1974) and Meier (1975).
The Nelson-Aalen estimate is due to Altshuler (1970), Nelson (1972) and
Aalen (1978b).
The expression for the standard error of the Kaplan-Meier estimate was
first given by Greenwood (1926), but an alternative result is given by Aalen
and Johansen (1978). Expressions for the variance of the Nelson-Aalen esti-
mate of the cumulative hazard function are compared by Klein (1991). Al-
though Section 2.2.3 shows how a confidence interval for the value of the sur-
vivor function at particular times can be found using Greenwood’s formula,
alternative procedures are needed for the construction of confidence bands
for the complete survivor function. Hall and Wellner (1980) and Efron (1981)
have shown how such bands can be computed, and these procedures are also
described by Harris and Albert (1991).
Methods for constructing confidence intervals for the median survival time
are described by Brookmeyer and Crowley (1982), Emerson (1982), Nair
(1984), Simon and Lee (1982) and Slud, Byar and Green (1984). Simon (1986)
emphasises the importance of confidence intervals in reporting the results of
clinical trials, and includes an illustration of a method described in Slud, Byar
and Green (1984). Klein and Moeschberger (2005) include a comprehensive
review of kernel-smoothed estimates of the hazard function.
The formulation of the hypothesis testing procedure in the frequentist
approach to inference is covered in many statistical texts. See, for example,
Altman (1991) and Armitage, Berry and Matthews (2002) for non-technical
presentations of the ideas in a medical context.
The log-rank test results from the work of Mantel and Haenszel (1959),
Mantel (1966) and Peto and Peto (1972). See Lawless (2002) for details of the
rank test formulation. A thorough review of the hypergeometric distribution,
used in the derivation of the log-rank test in Section 2.6.2, is included in
Johnson, Kemp and Kotz (2005). The log-rank test for trend is derived from
the test for trend in a 2 × k contingency table, given in Armitage, Berry and
Matthews (2002).
Chapter 3
The Cox regression model
The non-parametric methods described in Chapter 2 can be useful in the

analysis of a single sample of survival data, or in the comparison of two or
more groups of survival times. However, in most medical studies that give
rise to survival data, supplementary information will also be recorded on each
individual. A typical example would be a clinical trial to compare the survival
times of patients who receive one or other of two treatments. In such a study,
demographic variables such as the age and sex of the patient, the values of
physiological variables such as serum haemoglobin level and heart rate, and
factors that are associated with the lifestyle of the patient, such as smoking
history and dietary habits, may all have an impact on the time that the
patient survives. Accordingly, the values of these variables, which are referred
to as explanatory variables, would be recorded at the outset of the study.
The resulting data set would then be more complex than those considered
in Chapter 2, and the methods described in that chapter would generally be
unsuitable.
In order to explore the relationship between the survival experience of a
patient and explanatory variables, an approach based on statistical modelling
can be used. The particular model that is developed in this chapter, known
as the Cox regression model, both unifies and extends the non-parametric
procedures of Chapter 2.
3.1 Modelling the hazard function
Through a modelling approach to the analysis of survival data, we can explore

how the survival experience of a group of patients depends on the values of
one or more explanatory variables, whose values have been recorded for each
patient at the time origin. For example, in the study on multiple myeloma,
given as Example 1.3, the aim is to determine which of seven explanatory
variables have an impact on the survival time of the patients. In Example 1.4
on the survival times of patients in a clinical trial involving two treatments
for prostatic cancer, the primary aim is to identify whether patients in the
two treatment groups have a different survival experience. In this example,
variables such as the age of the patient and the size of their tumour are likely
57
58 THE COX REGRESSION MODEL
to influence survival time, and so it will be important to take account of these
variables when assessing the extent of any treatment difference.
In the analysis of survival data, interest centres on the risk or hazard of
death at any time after the time origin of the study. As a consequence, the
hazard function, defined in Section 1.3.2 of Chapter 1, is modelled directly in
survival analysis. The resulting models are somewhat different in form from
linear models encountered in regression analysis and in the analysis of data
from designed experiments, where the dependence of the mean response, or
some function of it, on certain explanatory variables is modelled. However,
many of the principles and procedures used in linear modelling carry over to
the modelling of survival data.
There are two broad reasons for modelling survival data. One objective of
the modelling process is to determine which combination of potential explana-
tory variables affect the form of the hazard function. In particular, the effect
that the treatment has on the hazard of death can be studied, as can the ex-
tent to which other explanatory variables affect the hazard function. Another
reason for modelling the hazard function is to obtain an estimate of the hazard
function itself for an individual. This may be of interest in its own right, but
in addition, from the relationship between the survivor function and hazard
function described by Equation (1.6), an estimate of the survivor function
can be found. This will in turn lead to an estimate of quantities such as the
median survival time, which will be a function of the explanatory variables in
the model. The median survival time could then be estimated for current or
future patients with particular values of these explanatory variables. The re-
sulting estimate could be particularly useful in devising a treatment regimen,
or in counselling the patient about their prognosis.
The model for survival data to be described in this chapter is based on the
assumption of proportional hazards, introduced in Section 2.6.4 of Chapter 2,
and is called a proportional hazards model. We first develop the model for the
comparison of the hazard functions for individuals in two groups.
3.1.1 A model for the comparison of two groups

Suppose that two groups of patients receive either a standard treatment or
a new treatment, and let hS (t) and hN (t) be the hazards of death at time t
for patients on the standard treatment and new treatment, respectively. Ac-
cording to a simple model for the survival times of the two groups of patients,
the hazard at time t for a patient on the new treatment is proportional to
the hazard at that same time for a patient on the standard treatment. This
proportional hazards model can be expressed in the form
hN (t) = ψhS (t), (3.1)
for any non-negative value of t, where ψ is a constant. An implication of this

assumption is that the corresponding true survivor functions for individuals
MODELLING THE HAZARD FUNCTION 59
on the new and standard treatments do not cross, as previously shown in
Section 2.6.4.
The value of ψ is the ratio of the hazard of death at any time for an
individual on the new treatment relative to an individual on the standard
treatment, and so ψ is known as the relative hazard or hazard ratio. If ψ < 1,
the hazard of death at t is smaller for an individual on the new drug, relative to
an individual on the standard. The new treatment is then an improvement on
the standard. On the other hand, if ψ > 1, the hazard of death at t is greater
for an individual on the new drug, and the standard treatment is superior.
An alternative way of expressing the model in Equation (3.1) leads to a
model that can more easily be generalised. Suppose that survival data are
available on n individuals and denote the hazard function for the ith of these
by hi (t), i = 1, 2, . . . , n. Also, write h0 (t) for the hazard function for an indi-
vidual on the standard treatment. The hazard function for an individual on
the new treatment is then ψh0 (t). The relative hazard ψ cannot be negative,
and so it is convenient to set ψ = exp(β). The parameter β is then the loga-
rithm of the hazard ratio, that is, β = log ψ, and any value of β in the range
(−∞, ∞) will lead to a positive value of ψ. Note that positive values of β are
obtained when the hazard ratio, ψ, is greater than unity, that is, when the
new treatment is inferior to the standard.
Now let X be an indicator variable, which takes the value zero if an indi-
vidual is on the standard drug, and unity if an individual is on the new drug.
If xi is the value of X for the ith individual in the study, i = 1, 2, . . . , n, the
hazard function for this individual can be written as
hi (t) = eβxi h0 (t), (3.2)
where xi = 1 if the ith individual is on the new treatment and xi = 0 oth-

erwise. This is the proportional hazards model for the comparison of two
treatment groups.
3.1.2 The general proportional hazards model

The model of the previous section is now generalised to the situation where
the hazard of death at a particular time depends on the values x1 , x2 , . . . , xp of
p explanatory variables, X1 , X2 , . . . , Xp . The values of these variables will be
assumed to have been recorded at the time origin of the study. An extension
of the model to cover the situation where the values of one or more of the
explanatory variables change over time will be considered in Chapter 8.
The set of values of the explanatory variables in the proportional hazards
model will be represented by the vector x, so that x = (x1 , x2 , . . . , xp )′ . Let
h0 (t) be the hazard function for an individual for whom the values of all the
explanatory variables that make up the vector x are zero. The function h0 (t) is
called the baseline hazard function. The hazard function for the ith individual
can then be written as
hi (t) = ψ(xi )h0 (t),
where ψ(xi ) is a function of xi , the vector of values of the explanatory variables
for the ith individual, whose components are x1i , x2i , . . . , xpi . The function
ψ(·) can be interpreted as the hazard at time t for an individual whose vector
of explanatory variables is xi , relative to the hazard for an individual for
whom x = 0.
Again, since the relative hazard, ψ(xi ), cannot be negative, it is convenient
to write this as exp(ηi ), where ηi is a linear combination of the values of the
p explanatory variables in xi . Therefore,
ηi = β1 x1i + β2 x2i + · · · + βp xpi ,

∑p ′
so that ηi = j=1 βj xji . In matrix notation, ηi = β xi , where β =
′
(β1 , β2 , . . . , βp ) is the vector of coefficients of the p explanatory variables
in the model. The quantity ηi is called the linear component of the model,
but it is also known as the risk score or prognostic index for the ith indi-
vidual. There are other possible forms for the function ψ(xi ), but the choice
ψ(xi ) = exp(β ′ xi ) leads to the most commonly used model for survival data.
The general proportional hazards model then becomes
hi (t) = exp(β1 x1i + β2 x2i + · · · + βp xpi )h0 (t). (3.3)
Notice that there is no constant term in the linear component of this propor-
tional hazards model. If a constant term β0 , say, were included, the baseline
hazard function could simply be rescaled by dividing h0 (t) by exp(β0 ), and
the constant term would cancel out. The model in Equation (3.3) can also be
re-expressed in the form
{ }
hi (t)
log = β1 x1i + β2 x2i + · · · + βp xpi ,
h0 (t)
to give a linear model for the logarithm of the hazard ratio.
The model in Equation (3.3), in which no assumptions are made about
the actual form of the baseline hazard function h0 (t), was introduced by Cox
(1972) and has come to be known as the Cox regression model or the Cox pro-
portional hazards model. Since no particular form of probability distribution is
assumed for the survival times, the Cox regression model is a semi-parametric
model, and Section 3.3 will show how the β-coefficients in this model can be
estimated. Of course, we will often need to estimate h0 (t) itself, and we will
see how this can be done in Section 3.10. Models in which specific assump-
tions are made about the form of the baseline hazard function, h0 (t), will be
described in Chapters 5 and 6.
3.2 The linear component of the model

There are two types of variable on which a hazard function may depend,
namely variates and factors. A variate is a variable that takes numerical val-
ues that are often on a continuous scale of measurement, such as age or systolic
THE LINEAR COMPONENT OF THE MODEL 61
blood pressure. A factor is a variable that takes a limited set of values, which
are known as the levels of the factor. For example, sex is a factor with two lev-
els, and type of tumour might be a factor whose levels correspond to different
histologies, such as squamous, adeno or small cell.
We now consider how variates, factors and terms that combine factors and
variates, can be incorporated in the linear component of a Cox regression
model.
3.2.1 Including a variate

Variates, either alone or in combination, are readily incorporated in a Cox
regression model. Each variate appears in the model with a corresponding β-
coefficient. As an illustration, consider a situation in which the hazard function
depends on two variates X1 and X2 . The value of these variates for the ith
individual will be x1i and x2i , respectively, and the Cox regression model for
the ith of n individuals is written as
hi (t) = exp(β1 x1i + β2 x2i )h0 (t).
In models such as this, the baseline hazard function, h0 (t), is the hazard
function for an individual for whom all the variates included in the model
take the value zero.
3.2.2 Including a factor

Suppose that the dependence of the hazard function on a single factor, A, is to
be modelled, where A has a levels. The model for an individual for whom the
level of A is j will then need to incorporate the term αj , which represents the
effect due to the jth level of the factor. The terms α1 , α2 , . . . , αa are known
as the main effects of the factor A. According to the Cox regression model,
the hazard function for an individual with factor A at level j is exp(αj )h0 (t).
Now, the baseline hazard function h0 (t) has been defined to be the hazard
for an individual with values of all explanatory variables equal to zero. To be
consistent with this definition, one of the αj must be taken to be zero. One
possibility is to adopt the constraint α1 = 0, which corresponds to taking the
baseline hazard to be the hazard for an individual for whom A is at the first
level. This is the constraint that will be used in the sequel.
Models that contain terms corresponding to factors can be expressed
as linear combinations of explanatory variables by defining indicator or
dummy variables for each factor. This procedure will be required when us-
ing computer software for survival analysis that does not allow factors to be
fitted directly.
If the first level of the factor A is set to zero, so that this is the base-
line level of the factor, the term αj can be included in the model by defin-
ing a − 1 indicator variables, X2 , X3 , . . . , Xa . These take the values shown
in Table 3.1.
Table 3.1 Indicator variables for a factor

with a levels.
Level of A X2 X3 . . . Xa
1 0 0 . . . 0
2 1 0 . . . 0
3 0 1 . . . 0
. . . . . . . . . . . . . . . .
a 0 0 . . . 1
The term αj can be incorporated in the linear part of the Cox regression
model by including the a − 1 explanatory variables X2 , X3 , . . . , Xa with coef-
ficients α2 , α3 , . . . , αa . In other words, the term αj in the model is replaced
by α2 x2 + α3 x3 + · · · + αa xa , where xj is the value of Xj for an individual
for whom A is at level j, j = 2, 3, . . . , a. There are then a − 1 parameters
associated with the main effect of the factor A, and A is said to have a − 1
degrees of freedom.
3.2.3 Including an interaction

When terms corresponding to more than one factor are to be included in the
model, sets of indicator variables can be defined for each factor in a manner
similar to that shown above. In this situation, it may also be appropriate to
include a term in the model that corresponds to individual effects for each
combination of levels of two or more factors. Such effects are known as inter-
actions.
For example, suppose that the two factors are the sex of a patient and
grade of tumour. If the effect of grade of tumour on the hazard of death is
different in patients of each sex, we would say that there is an interaction
between these two factors. The hazard function would then depend on the
combination of levels of these two factors.
In general, if A and B are two factors, and the hazard of death depends
on the combination of levels of A and B, then A and B are said to interact.
If A and B have a and b levels, respectively, the term that represents an
interaction between these two factors is denoted by (αβ)jk , for j = 1, 2, . . . , a
and k = 1, 2, . . . , b.
In statistical modelling, the effect of an interaction can only be investi-
gated by adding the interaction term to a model that already contains the
corresponding main effects. If either αj or βk are excluded from the model,
the term (αβ)jk represents the effect of one factor nested within the other.
For example, if αj is included in the model, but not βk , then (αβ)jk is the
effect of B nested within A. If both αj and βk are excluded, the term (αβ)jk
represents the effect of the combination of level i of A and level j of B on
the response variable. This means that (αβ)jk can only be interpreted as an
interaction effect when included in a model that contains both αj and βk ,
THE LINEAR COMPONENT OF THE MODEL 63
which correspond to the main effects of A and B. We will return to this point
when we consider model-building strategy in Section 3.6.
In order to include the term (αβ)jk in the model, products of indicator
variables associated with the main effects are calculated. For example, if A
and B have 2 and 3 levels respectively, indicator variables U2 and V2 , V3 are
defined as in Table 3.2.
Table 3.2 Indicator variables for two factors with

two and three levels, respectively.
Level of A U2 Level of B V2 V3
1 0 1 0 0
2 1 2 1 0
3 0 1
Let uj and vk be the values of Uj and Vk for a given individual, for j = 2,

k = 2, 3. The term (αβ)jk is then fitted by including variates formed from the
products of Uj and Vk in the model. The corresponding value of the product
for a given individual is uj vk . The coefficient of this product is denoted (αβ)jk ,
and so the term (αβ)jk is fitted as
(αβ)22 u2 v2 + (αβ)23 u2 v3 .
There are therefore two parameters associated with the interaction between A
and B. In general, if A and B have a and b levels, respectively, the two-factor
interaction AB has (a−1)(b−1) parameters associated with it, in other words
AB has (a − 1)(b − 1) degrees of freedom. Furthermore, the term (αβ)jk is
equal to zero whenever either A or B are at the first level, that is, when either
j = 1 or k = 1.
3.2.4 Including a mixed term

Another type of term that might be needed in a model is a mixed term formed
from a factor and a variate. Terms of this type would be used when the
coefficient of a variate in a model was likely to be different for each level
of a factor. For example, consider a contraceptive trial in which the time to
the onset of a period of amenorrhoea, the prolonged absence of menstrual
bleeding, is being modelled. The hazard of an amenorrhoea may be related to
the weight of a woman, but the coefficient of this variate may differ according
to the level of a factor associated with the number of previous pregnancies
that the woman has experienced.
The dependence of the coefficient of a variate, X, on the level of a factor,
A, would be depicted by including the term αj x in the linear component of
the Cox regression model, where x is the value of X for a given individual for
whom the factor A is at the jth level, j = 1, 2, . . . , a. To include such a term,
indicator variables Uj , say, are defined for the factor A, and each of these is
multiplied by the value of X for each individual. The resulting values of the
products Uj X are uj x, and the coefficient of uj x in the model is αj , where j
indexes the level of the factor A.
If the same definition of indicator variables in the previous discussion were
used, α1 , the coefficient of X for individuals at the first level of A, would be
zero. It is then essential to include the variate X in the model as well as the
products, for otherwise the dependence on X for individuals at the first level
of A would not be modelled. An illustration should make this clearer.
Suppose that there are nine individuals in a study, on each of whom the
value of a variate, X, and the level of a factor, A, have been recorded. We
will take A to have three levels, where A is at the first level for the first three
individuals, at the second level for the next three, and at the third level for
the final three. In order to model the dependence of the coefficient of the
variate X on the level of A, two indicator variables, U2 and U3 are defined
as in Table 3.3. Explanatory variables formed as the products U2 X and U3 X,
given in the last two columns of Table 3.3, would then be included in the
linear component of the model, together with the variate X.
Table 3.3 Indicator variables for the combination of a factor

with three levels and a variate.
Individual Level of A X U2 U3 U2 X U3 X
1 1 x1 0 0 0 0
2 1 x2 0 0 0 0
3 1 x3 0 0 0 0
4 2 x4 1 0 x4 0
5 2 x5 1 0 x5 0
6 2 x6 1 0 x6 0
7 3 x7 0 1 0 x7
8 3 x8 0 1 0 x8
9 3 x9 0 1 0 x9
Let the coefficients of the values of the products U2 X and U3 X be α2′ and
α3′ ,respectively, and let the coefficient of the value of the variate X in the
model be β. Then, the model contains the terms βx + α2′ (u2 x) + α3′ (u3 x).
From Table 3.3, u2 = 0 and u3 = 0 for individuals at level 1 of A, and so the
coefficient of x for these individuals is just β. For those at level 2 of A, u2 = 1
and u3 = 0, and the coefficient of x is β + α2′ . Similarly, at level 3 of A, u2 = 0
and u3 = 1, and the coefficient of x is β + α3′ .
Notice that if the term βx is omitted from the model, the coefficient of x
for individuals 1, 2 and 3 would be zero. There would then be no information
about the relationship between the hazard function and the variate X for
individuals at the first level of the factor A.
The manipulation described in the preceding paragraphs can be avoided
by defining the indicator variables in a different way. If a factor A has a levels,
and it is desired to include the term αj x in a model, without necessarily
FITTING THE COX REGRESSION MODEL 65
including the term βx, a indicator variables Z1 , Z2 , . . . , Za can be defined for
A, where Zj = 1 at level j of A and zero otherwise. The corresponding values
of these products for an individual, z1 x, z2 x, . . . , za x, are then included in the
model with coefficients α1 , α2 , . . . , αa . These are the coefficients of x for each
level of A.
Now, if the variate X is included in the model, along with the a products
of the form Zj X, there will be a + 1 terms corresponding to the a coeffi-
cients. It will not then be possible to obtain unique estimates of each of these
α-coefficients, and the model is said to be overparameterised. This overpa-
rameterisation can be dealt with by forcing one of the a + 1 coefficients to be
zero. In particular, taking α1 = 0 would be equivalent to a redefinition of the
indicator variables, in which Z1 is taken to be zero. This then leads to the
same formulation of the model that has already been discussed.
The application of these ideas in the analysis of actual data sets will be
illustrated in Section 3.4, after we have seen how the Cox regression model
can be fitted.
3.3 Fitting the Cox regression model

Fitting the Cox regression model given in Equation (3.3) to an observed set of
survival data entails estimating the unknown coefficients of the explanatory
variables, X1 , X2 , . . . , Xp , in the linear component of the model, β1 , β2 , . . . , βp .
The baseline hazard function, h0 (t), may also need to be estimated. It turns
out that these two components of the model can be estimated separately.
The βs are estimated first and these estimates are then used to construct an
estimate of the baseline hazard function. This is an important result, since
it means that in order to make inferences about the effects of p explanatory
variables, X1 , X2 , . . . , Xp , on the relative hazard, hi (t)/h0 (t), we do not need
an estimate of h0 (t). Methods for estimating h0 (t) will therefore be deferred
until Section 3.10.
The β-coefficients in the Cox regression model, which are the unknown
parameters in the model, can be estimated using the method of maximum
likelihood. To operate this method, we first obtain the likelihood of the sample
data. This is the joint probability of the observed data, regarded as a function
of the unknown parameters in the assumed model. For the Cox regression
model, this is a function of the observed survival times and the unknown β-
parameters in the linear component of the model. Estimates of the βs are then
those values that are the most likely on the basis of the observed data. These
maximum likelihood estimates are therefore the values that maximise the like-
lihood function. From a computational viewpoint, it is more convenient to
maximise the logarithm of the likelihood function. Furthermore, approxima-
tions to the variance of maximum likelihood estimates can be obtained from
the second derivatives of the log-likelihood function. Details will not be given
here, but Appendix A contains a summary of relevant results from the theory
of maximum likelihood estimation.
Suppose that data are available for n individuals, among whom there are
r distinct death times and n − r right-censored survival times. We will for the
moment assume that only one individual dies at each death time, so that there
are no ties in the data. The treatment of ties will be discussed in Section 3.3.2.
The r ordered death times will be denoted by t(1) < t(2) < · · · < t(r) , so that
t(j) is the jth ordered death time. The set of individuals who are at risk at
time t(j) will be denoted by R(t(j) ), so that R(t(j) ) is the group of individuals
who are alive and uncensored at a time just prior to t(j) . The quantity R(t(j) )
is called the risk set.
Cox (1972) showed that the relevant likelihood function for the model in
Equation (3.3) is given by
∏
r
exp(β ′ x(j) )
L(β) = ∑ ′ , (3.4)
j=1 l∈R(t(j) ) exp(β xl )
in which x(j) is the vector of covariates for the individual who dies at the jth
ordered death time, t(j) . The summation in the denominator of this likelihood
function is the sum of the values of exp(β ′ x) over all individuals who are at
risk at time t(j) . Notice that the product is taken over the individuals for whom
death times have been recorded. Individuals for whom the survival times are
censored do not contribute to the numerator of the log-likelihood function,
but they do enter into the summation over the risk sets at death times that
occur before a censored time.
The likelihood function that has been obtained is not a true likelihood,
since it does not make direct use of the actual censored and uncensored sur-
vival times. For this reason it is referred to as a partial likelihood function.
The likelihood function in Equation (3.4) depends only on the ranking of the
death times, since this determines the risk set at each death time. Conse-
quently, inferences about the effect of explanatory variables on the hazard
function depend only on the rank order of the survival times.
Now suppose that the data consist of n observed survival times, denoted
by t1 , t2 , . . . , tn , and that δi is an event indicator, which is zero if the ith
survival time ti , i = 1, 2, . . . , n, is right-censored, and unity otherwise. The
partial likelihood function in Equation (3.4) can then be expressed in the form
{ }δi
∏n
exp(β ′ xi )
∑ ′ , (3.5)
i=1 l∈R(ti ) exp(β xl )
where R(ti ) is the risk set at time ti . From Equation (3.5), the corresponding
partial log-likelihood function is given by
 
∑n  ∑ 
log L(β) = δi β ′ xi − log exp(β ′ xl ) . (3.6)
 
i=1 l∈R(ti )
The maximum likelihood estimates of the β-parameters in the Cox regres-

sion model can be found by maximising this log-likelihood function using
numerical methods. This maximisation is generally accomplished using the
Newton-Raphson procedure described below in Section 3.3.3.
Fortunately, most statistical software for survival analysis enables the Cox
regression model to be fitted. Such software also gives the standard errors of
the parameter estimates in the fitted model.
The justification for using Equation (3.4) as a likelihood function, and
further details on the structure of the likelihood function, are given in Sec-
tion 3.3.1. The treatment of tied survival times is then discussed in Sec-
tion 3.3.2 and the Newton-Raphson procedure is outlined in Section 3.3.3.
These three sections can be omitted without loss of continuity.
3.3.1 ∗ Likelihood function for the model

In the Cox regression model, the hazard of death at time t for the ith indi-
vidual, i = 1, 2, . . . , n is given by
hi (t) = exp(β ′ xi )h0 (t)
where β is the vector of coefficients of p explanatory variables whose values are

x1i , x2i , . . . , xpi for the ith individual, and h0 (t) is the baseline hazard function
of unspecified form. The basis of the argument used in the construction of a
likelihood function for this model is that intervals between successive death
times convey no information about the effect of explanatory variables on the
hazard of death. This is because the baseline hazard function has an arbitrary
form, and so it is conceivable that h0 (t), and hence hi (t), is zero in those time
intervals in which there are no deaths. This in turn means that these intervals
give no information about the values of the β-parameters. We therefore con-
sider the probability that the ith individual dies at some time t(j) , conditional
on t(j) being one of the observed set of r death times t(1) , t(2) , . . . , t(r) . If the
vector of values of the explanatory variables for the individual who dies at t(j)
is denoted by x(j) , this probability is
P(individual with variables x(j) dies at t(j) | one death at t(j) ). (3.7)
Next, from the result that the probability of an event A, given that an event
B has occurred, is given by
P(A | B) = P(A and B)/P(B),
the probability in Expression (3.7) becomes
P(individual with variables x(j) dies at t(j) )

. (3.8)
P(one death at t(j) )
Since the death times are assumed to be independent of one another, the
denominator of this expression is the sum of the probabilities of death at
time t(j) over all individuals who are at risk of death at that time. If these
individuals are indexed by l, with R(t(j) ) denoting the set of individuals who
are at risk at time t(j) , Expression (3.8) becomes
P(individual with variables x(j) dies at t(j) )

∑ . (3.9)
l∈R(t(j) ) P(individual l dies at t(j) )
The probabilities of death at time t(j) , in Expression (3.9), are now replaced
by probabilities of death in the interval (t(j) , t(j) + δt), and dividing both the
numerator and denominator of Expression (3.9) by δt, we get
P{individual with variables x(j) dies in (t(j) , t(j) + δt)}/δt

∑ .
l∈R(t(j) ) P{individual l dies in (t(j) , t(j) + δt)}/δt
The limiting value of this expression as δt → 0 is then the ratio of the prob-
abilities in Expression (3.9). But from Equation (1.3), this limit is also the
ratio of the corresponding hazards of death at time t(j) , that is,
Hazard of death at time t(j) for individual with variables x(j)

∑ .
l∈R(t(j) ) {Hazard of death at time t(j) for individual l}
If it is the ith individual who dies at t(j) , the hazard function in the numerator
of this expression can be written hi (t(j) ). Similarly, the denominator is the sum
of the hazards of death at time t(j) over all individuals who are at risk of death
at this time. This is the sum of the values hl (t(j) ) over those individuals in
the risk set at time t(j) , R(t(j) ). Consequently, the conditional probability in
Expression (3.7) becomes
hi (t(j) )
∑ .
l∈R(t(j) ) hl (t(j) )
On using Equation (3.3), the baseline hazard function in the numerator and
denominator cancels out, and we are left with
exp(β ′ x(j) )
∑ ′ .
l∈R(t(j) ) exp(β xl )
Finally, taking the product of these conditional probabilities over the r death
times gives the partial likelihood function in Equation (3.4).
In order to throw more light on the structure of the partial likelihood,
consider a sample of survival data from five individuals, numbered from 1
to 5. The survival data are illustrated in Figure 3.1. The observed survival
times of individuals 2 and 5 will be taken to be right-censored, and the three
ordered death times are denoted t(1) < t(2) < t(3) . Then, t(1) is the death time
of individual 3, t(2) is that of individual 1, and t(3) that of individual 4.
The risk set at each of the three ordered death times consists of the in-
dividuals who are alive and uncensored just prior to each death time. Hence,
1 D
2 C
Individual
3 D
4 D
5 C
0 t t t
Time
Figure 3.1 Survival times of five individuals.
the risk set R(t(1) ) consists of all five individuals, risk set R(t(2) ) consists of
individuals 1, 2 and 4, while risk set R(t(3) ) only includes individual 4. Now
write ψ(i) = exp(β ′ xi ), i = 1, 2, . . . , 5, where xi is the vector of explanatory
variables for the ith individual. The numerators of the partial likelihood func-
tion for times t(1) , t(2) and t(3) , respectively, are ψ(3), ψ(1) and ψ(4), since
individuals 3, 1 and 4, respectively, die at the three ordered death times. The
partial likelihood function over the three death times is then
ψ(3) ψ(1) ψ(4)
× × .
ψ(1) + ψ(2) + ψ(3) + ψ(4) + ψ(5) ψ(1) + ψ(2) + ψ(4) ψ(4)
It turns out that standard results used in maximum likelihood estimation
carry over without modification to maximum partial likelihood estimation. In
particular, the results given in Appendix A for the variance-covariance matrix
of the estimates of the βs can be used, as can distributional results associated
with likelihood ratio testing, to be discussed in Section 3.5.
3.3.2 ∗ Treatment of ties

The Cox regression model for survival data assumes that the hazard function
is continuous, and under this assumption, tied survival times are not possible.
Of course, survival times are usually recorded to the nearest day, month or
year, and so tied survival times can arise as a result of this rounding process.
Indeed, Examples 1.2, 1.3 and 1.4 in Chapter 1 all contain tied observations.
In addition to the possibility of more than one death at a given time, there
might also be one or more censored observations at a death time. When there
are both censored survival times and deaths at a given time, the censoring is
assumed to occur after all the deaths. Potential ambiguity concerning which
individuals should be included in the risk set at that death time is then re-
solved and tied censored observations present no further difficulties in the
computation of the likelihood function using Equation (3.4). Accordingly, we
only need consider how tied survival times can be handled in fitting the Cox
regression model.
In order to accommodate tied observations, the likelihood function in
Equation (3.4) has to be modified in some way. The appropriate likelihood
function in the presence of tied observations has been given by Kalbfleisch
and Prentice (2002). However, this likelihood has a very complicated form,
and will not be reproduced here. In addition, the computation of this like-
lihood function can be very time consuming, particularly when there are a
relatively large number of ties at one or more death times. Fortunately, there
are a number of approximations to the likelihood function that have compu-
tational advantages over the exact method. But before these are given, some
additional notation needs to be developed.
Let sj be the vector of sums of each of the p covariates for those individuals
who die at the jth death time, t(j) , j = 1, 2, . . . , r. If there are dj deaths at
∑dj
t(j) , the hth element of sj is shj = k=1 xhjk , where xhjk is the value of
the hth explanatory variable, h = 1, 2, . . . , p, for the kth of dj individuals,
k = 1, 2, . . . , dj , who die at the jth death time, j = 1, 2, . . . , r.
The simplest approximation to the likelihood function is that due to Bres-
low (1974), who proposed the approximate likelihood
∏
r
exp(β ′ sj )
{∑ }dj . (3.10)
′
j=1
l∈R(t(j) ) exp(β x l )
In this approximation, the dj deaths at time t(j) are considered to be distinct

and to occur sequentially. The probabilities of all possible sequences of deaths
are then summed to give the likelihood in Equation (3.10). Apart from a
constant of proportionality, this is also the approximation suggested by Peto
(1972). This likelihood is quite straightforward to compute, and is an adequate
approximation when the number of tied observations at any one death time is
not too large. For these reasons, this method is usually the default procedure
for handling ties in statistical software for survival analysis, and will be used
in the examples given in this book.
Efron (1977) proposed
∏
r
exp(β ′ sj )
∏d j [ ∑ ∑ ] (3.11)
j=1 k=1 l∈R(t(j) ) exp(β ′ xl ) − (k − 1)d−1
j l∈D(t(j) ) exp(β ′ xl )
as an approximate likelihood function for the Cox regression model, where

D(t(j) ) is the set of all individuals who die at time t(j) . This is a closer ap-
proximation to the appropriate likelihood function than that due to Breslow,
although in practice, both approximations often give similar results.
Cox (1972) suggested the approximation
∏
r
exp(β ′ sj )
∑ ′ , (3.12)
j=1 l∈R(t(j) ;dj ) exp(β sl )
where the notation R(t(j) ; dj ) denotes a set of dj individuals drawn from

R(t(j) ), the risk set at t(j) . The summation in the denominator is the sum
over all possible sets of dj individuals, sampled from the risk set without
replacement. The approximation in Expression (3.12) is based on a model for
the situation where the time-scale is discrete, so that under this model, tied
observations are permissible. Now, from Section 1.3.2 of Chapter 1, the hazard
function for an individual with vector of explanatory variables xi , hi (t), is the
probability of death in the unit time interval (t, t + 1), conditional on survival
to time t. A discrete version of the Cox regression model of Equation (3.3) is
the model
hi (t) h0 (t)
= exp(β ′ xi ) ,
1 − hi (t) 1 − h0 (t)
for which the likelihood function is that given in Equation (3.12). In fact, in
the limit as the width of the discrete time intervals becomes zero, this model
tends to the Cox regression model of Equation (3.3).
When there are no ties, that is, when dj = 1 for each death time, the ap-
proximations in Equations (3.10), (3.11) and (3.12) all reduce to the likelihood
function in Equation (3.4).
3.3.3 ∗ The Newton-Raphson procedure

Models for censored survival data are usually fitted by using the Newton-
Raphson procedure to maximise the partial likelihood function, and so the
procedure is outlined in this section.
Let u(β) be the p×1 vector of first derivatives of the log-likelihood function
in Equation (3.6) with respect to the β-parameters. This quantity is known
as the vector of efficient scores. Also, let I(β) be the p × p matrix of negative
second derivatives of the log-likelihood, so that the (j, k)th element of I(β) is
∂ 2 log L(β)
− .
∂βj ∂βk
The matrix I(β)is known as the observed information matrix.

According to the Newton-Raphson procedure, an estimate of the vector of
β-parameters at the (s + 1)th cycle of the iterative procedure, β̂ s+1 , is
β̂ s+1 = β̂ s + I −1 (β̂ s )u(β̂ s ),
for s = 0, 1, 2, . . ., where u(β̂ s ) is the vector of efficient scores and I −1 (β̂ s ) is

the inverse of the information matrix, both evaluated at β̂ s . The procedure
can be started by taking β̂ 0 = 0. The process is terminated when the change
in the log-likelihood function is sufficiently small, or when the largest of the
relative changes in the values of the parameter estimates is sufficiently small.
When the iterative procedure has converged, the variance-covariance ma-
trix of the parameter estimates can be approximated by the inverse of the
information matrix, evaluated at β̂, that is, I −1 (β̂). The square root of the
diagonal elements of this matrix are then the standard errors of the estimated
values of β1 , β2 , . . . , βp .
3.4 Confidence intervals and hypothesis tests
When statistical software is used to fit a Cox regression model, the parameter

estimates that are provided are usually accompanied by their standard errors.
These standard errors can be used to obtain approximate confidence inter-
vals for the unknown β-parameters. In particular, a 100(1 − α)% confidence
interval for a parameter β is the interval with limits β̂ ± zα/2 se (β̂), where β̂
is the estimate of β, and zα/2 is the upper α/2-point of the standard normal
distribution.
If a 100(1 − α)% confidence interval for β does not include zero, this is
evidence that the value of β is non-zero. More specifically, the null hypothesis
that β = 0 can be tested by calculating the value of the statistic β̂/ se (β̂). The
observed value of this statistic is then compared to percentage points of the
standard normal distribution in order to obtain the corresponding P -value.
Equivalently, the square of this statistic can be compared with percentage
points of a chi-squared distribution on one degree of freedom. This procedure
is sometimes called a Wald test, and the P -values for this test are often given
alongside parameter estimates and their standard errors in computer output.
When attempting to interpret the P -value for a given parameter, βj , say,
it is important to recognise that the hypothesis that is being tested is that
βj = 0 in the presence of all other terms that are in the model. For example,
suppose that a model contains the three explanatory variables X1 , X2 , X3 ,
and that their true coefficients are β1 , β2 , β3 . The test statistic β̂2 / se (β̂2 ) is
then used to test the null hypothesis that β2 = 0 in the presence of β1 and
β3 . If there was no evidence to reject this hypothesis, we would conclude that
X2 was not needed in the model in the presence of X1 and X3 .
In general, the individual estimates of the βs in a Cox regression model
are not all independent of one another. This means that the results of testing
separate hypotheses about the β-parameters in a model may not be easy to
interpret. For example, consider again the situation where there are three ex-
planatory variables, X1 , X2 , X3 . If β̂1 and β̂2 were not found to be significantly
different from zero, when compared with their standard errors, we could not
conclude that only X3 need be included in the model. This is because the
coefficient of X1 , for example, could well change when X2 is excluded from
CONFIDENCE INTERVALS AND HYPOTHESIS TESTS 73
the model, and vice versa. This would certainly happen if X1 and X2 were
correlated.
Because of the difficulty in interpreting the results of tests concerning the
coefficients of the explanatory variables in a model, alternative methods for
comparing different Cox regression models are required. It turns out that the
methods to be described in Section 3.5 are much more satisfactory than the
Wald tests. The results of these tests given in computer-based analyses of
survival data should therefore be treated with some caution.
3.4.1 Standard errors and confidence intervals for hazard ratios

We have seen that in situations where there are two groups of survival data,
the parameter β in a Cox regression model is the logarithm of the ratio of
the hazard of death at time t for individuals in one group relative to those
in the other. Hence the hazard ratio itself is ψ = eβ . The corresponding
estimate of the hazard ratio is ψ̂ = exp(β̂), and the standard error of ψ̂ can be
obtained from the standard error of β̂ using the result given as Equation (2.8)
in Chapter 2. From this result, the approximate variance of ψ̂, a function of
β̂, is
{ }2
exp(β̂) var (β̂),
that is, ψ̂ 2 var (β̂), and so the standard error of ψ̂ is given by
se (ψ̂) = ψ̂ se (β̂). (3.13)
Generally speaking, a confidence interval for the true hazard ratio will be
more informative than the standard error of the estimated hazard ratio. A
100(1 − α)% confidence interval for the true hazard ratio, ψ, can be found
simply by exponentiating the confidence limits for β. An interval estimate
obtained in this way is preferable to one found using ψ̂ ± zα/2 se (ψ̂). This is
because the distribution of the logarithm of the estimated hazard ratio will be
more closely approximated by a normal distribution than that of the hazard
ratio itself.
The construction of a confidence interval for a hazard ratio is illustrated
in Example 3.1 below. Fuller details on the interpretation of the parameters
in the linear component of a Cox regression model are given in Section 3.9.
3.4.2 Two examples

In this section, the results of fitting a Cox regression model to data from two
of the examples introduced in Chapter 1 are given.

Data on the survival times of breast cancer patients, classified according to
whether or not sections of their tumours were positively stained, were first
given in Example 1.2. The variable that indexes the result of the staining
process can be regarded as a factor with two levels. From the arguments given
in Section 3.2.2, this factor can be fitted by using an indicator variable X
to denote the staining result, where X = 0 corresponds to negative staining
and X = 1 to positive staining. Under the Cox regression model, the hazard
of death at time t for the ith woman, for whom the value of the indicator
variable is xi , is
hi (t) = eβxi h0 (t),
where xi is zero or unity. The baseline hazard function h0 (t) is then the hazard
function for a women with a negatively stained tumour. This is essentially the
model considered in Section 3.1.1, and given in Equation (3.2).
In the group of women whose tumours were positively stained, there are
two who die at 26 months. To cope with this tie, the Breslow approximation
to the likelihood function will be used. This model is fitted by finding that
value of β, β̂, which maximises the likelihood function in Equation (3.10).
The maximum likelihood estimate of β is β̂ = 0.908. The standard error of
this estimate is also obtained from statistical packages for fitting the Cox
regression model, and turns out to be given by se (β̂) = 0.501.
The quantity eβ is the ratio of the hazard function for a woman with
X = 1 to that for a woman with X = 0, so that β is the logarithm of the ratio
of the hazard of death at time t for positively stained relative to negatively
stained women. The estimated value of this hazard ratio is e0.908 = 2.48. Since
this is greater than unity, we conclude that a woman who has a positively
stained tumour will have a greater risk of death at any given time than a
comparable women whose tumour was negatively stained. Positive staining
therefore indicates a poorer prognosis for a breast cancer patient.
The standard error of the hazard ratio can be found from the standard
error of β̂, using the result in Equation (3.13). Since the estimated relative
hazard is ψ̂ = exp(β̂) = 2.480, and the standard error of β̂ is 0.501, the
standard error of ψ̂ is given by
se (ψ̂) = 2.480 × 0.501 = 1.242.
We can go further and construct a confidence interval for this hazard ratio.
The first step is to obtain a confidence interval for the logarithm of the hazard
ratio, β. For example, a 95% confidence interval for β is the interval from
β̂ − 1.96 se (β̂) to β̂ + 1.96 se (β̂), that is, the interval from −0.074 to 1.890.
Exponentiating these confidence limits gives (0.93, 6.62) as a 95% confidence
interval for the hazard ratio itself. Notice that this interval barely includes
unity, suggesting that there is evidence that the two groups of women have a
different survival experience.

Data on the survival times of 48 patients suffering from multiple myeloma were
given in Example 1.3. The database also contains the values of seven other
CONFIDENCE INTERVALS AND HYPOTHESIS TESTS 75
variables that were recorded for each patient. For convenience, the values of
the variable that describes the sex of a patient have been redefined to be
zero and unity for males and females, respectively. The sex of the patient and
the variable associated with the occurrence of Bence-Jones protein are factors
with two levels, and these terms are fitted using the indicator variables Sex
and Protein. The variables to be used in the modelling process are then as
follows:
Age: Age of the patient
Sex: Sex of the patient (0 = male, 1 = female)
Bun: Blood urea nitrogen
Ca: Serum calcium
Hb: Serum haemoglobin
Pcells: Percentage of plasma cells
Protein: Bence-Jones protein (0 = absent, 1 = present)
The Cox regression model for the ith individual is then

hi (t) = exp(β1 Age i + β2 Sex i + β3 Bun i + β4 Ca i + β5 Hb i
+ β6 Pcells i + β7 Protein i )h0 (t),
where the subscript i on an explanatory variable denotes the value of that
variable for the ith individual. The baseline hazard function is the hazard
function for an individual for whom the values of all seven of these variables
are zero. This function therefore corresponds to a male aged zero, who has
zero values of Bun, Ca, Hb and Pcells, and no Bence-Jones protein. In view of
the obvious difficulty in interpreting this function, it might be more sensible to
redefine the variables Age, Bun, Ca, Hb and Pcells by subtracting values for an
average patient. For example, if we took Age − 60 in place of Age, the baseline
hazard would correspond to a male aged 60 years. This procedure also avoids
the introduction of a function that describes the hazard of individuals whose
ages are rather different from the age range of patients in the study. Although
this leads to a baseline hazard function that has a more natural interpretation,
it will not affect inference about the influence of the explanatory variables on
the hazard of death. For this reason, the untransformed variables will be used
in this example. On fitting the model, the estimates of the coefficients of the
explanatory variables and their standard errors are found to be those shown
in Table 3.4.
We see from Table 3.4 that some of the estimates are close to zero. Indeed,
if individual 95% confidence intervals are calculated for the coefficients of the
seven variables, only those for Bun and Hb exclude zero. This suggests that
the hazard function does not depend on all seven explanatory variables.
We cannot deduce from this that Bun and Hb are the relevant variables,
since the estimates of the coefficients of the seven explanatory variables in the
fitted model are not independent of one another. This means that if one of
the seven explanatory variables were excluded from the model, the coefficients
Table 3.4 Estimated values of the

coefficients of the explanatory vari-
ables on fitting a Cox regression
model to the data from Example 1.3.
Variable β̂ se (β̂)
Age −0.019 0.028
Sex −0.251 0.402
Bun 0.021 0.006
Ca 0.013 0.132
Hb −0.135 0.069
Pcells −0.002 0.007
Protein −0.640 0.427
of the remaining six might be different from those in Table 3.4. For example,
if Bun is omitted, the estimated coefficients of the six remaining explanatory
variables, Age, Sex, Ca, Hb, Pcells and Protein, turn out to be −0.009, −0.301,
−0.036, −0.140, −0.001 and −0.420, respectively. Comparison with the values
shown in Table 3.4 shows that there are differences in the estimated coefficients
of each of these six variables, although in this case the differences are not very
great.
In general, to determine on which of the seven explanatory variables the
hazard function depends, a number of different models will need to be fitted,
and the results compared. Methods for comparing the fit of alternative models,
and strategies for model building are considered in subsequent sections of this
chapter.
3.5 Comparing alternative models
In a modelling approach to the analysis of survival data, a model is devel-

oped for the dependence of the hazard function on one or more explanatory
variables. In this development process, Cox regression models with linear com-
ponents that contain different sets of terms are fitted, and comparisons made
between them.
As a specific example, consider the situation where there are two groups
of survival times, corresponding to individuals who receive either a new treat-
ment or a standard. The common hazard function under the model for no
treatment difference can be taken to be h0 (t). This model is a special case of
the general proportional hazards model in Equation (3.3), in which there are
no explanatory variables in the linear component of the model. This model is
therefore referred to as the null model.
Now let X be an indicator variable that takes the value zero for individuals
receiving the standard treatment and unity otherwise. Under a proportional
hazards model, the hazard function for an individual for whom X takes the
value x is eβx h0 (t). The hazard functions for individuals on the standard and
COMPARING ALTERNATIVE MODELS 77
new treatments are then h0 (t) and eβ h0 (t), respectively. The difference be-
tween this model and the null model is that the linear component of the latter
contains the additional term βx. Since β = 0 corresponds to no treatment ef-
fect, the extent of any treatment difference can be investigated by comparing
these two Cox regression models for the observed survival data.
More generally, suppose that two models are contemplated for a partic-
ular data set, Model (1) and Model (2), say, where Model (1) contains a
subset of the terms in Model (2). Model (1) is then said to be parametrically
nested within Model (2). Specifically, suppose that the p explanatory variables,
X1 , X2 , . . . , Xp , are fitted in Model (1), so that the hazard function under this
model can be written as
exp{β1 x1 + β2 x2 + · · · + βp xp }h0 (t).
Also suppose that the p + q explanatory variables X1 , X2 , . . . , Xp , Xp+1 , . . . ,

Xp+q are fitted in Model (2), so that the hazard function under this model is
exp{β1 x1 + · · · + βp xp + βp+1 xp+1 + · · · + βp+q xp+q }h0 (t).
Model (2) then contains the q additional explanatory variables Xp+1 , Xp+2 ,
. . . , Xp+q . Because Model (2) has a larger number of terms than Model (1),
Model (2) must be a better fit to the observed data. The statistical problem
is then to determine whether the additional q terms in Model (2) significantly
improve the explanatory power of the model. If not, they might be omitted,
and Model (1) would be deemed to be adequate.
In the discussion of Example 3.2, we saw that when there are a number
of explanatory variables of possible relevance, the effect of each term cannot
be studied independently of the others. The effect of any given term therefore
depends on the other terms currently included in the model. For example,
in Model (1), the effect of any of the p explanatory variables on the hazard
function depends on the p − 1 variables that have already been fitted, and
the effect of Xp is said to be adjusted for the remaining p − 1 variables. In
particular, the effect of Xp is adjusted for X1 , X2 , . . . , Xp−1 , but we also speak
of the effect of Xp eliminating or allowing for X1 , X2 , . . . , Xp−1 . Similarly,
when the q variables Xp+1 , Xp+2 , . . . , Xp+q are added to Model (1), the effect
of these variables on the hazard function is said to be adjusted for the p
variables that have already been fitted, X1 , X2 , . . . , Xp .
3.5.1 The statistic −2 log L̂

In order to compare alternative models fitted to an observed set of survival
data, a statistic that measures the extent to which the data are fitted by
a particular model is required. Since the likelihood function summarises the
information that the data contain about the unknown parameters in a given
model, a suitable summary statistic is the value of the likelihood function
when the parameters are replaced by their maximum likelihood estimates. This
leads to the maximised likelihood, or in the case of the Cox regression model
the maximised partial likelihood, under an assumed model. This statistic can
be computed from Equation (3.4) by replacing the βs by their maximum
likelihood estimates under the model. For a given set of data, the larger the
value of the maximised likelihood, the better is the agreement between the
model and the observed data.
For reasons given in the sequel, it is more convenient to use minus twice
the logarithm of the maximised likelihood in comparing alternative models.
If the maximised likelihood for a given model is denoted by L̂, the summary
measure of agreement between the model and the data is −2 log L̂. From
Section 3.3.1, L̂ is in fact the product of a series of conditional probabilities,
and so this statistic will be less than unity. In consequence, −2 log L̂ will
always be positive, and for a given data set, the smaller the value of −2 log L̂,
the better the model.
The statistic −2 log L̂ cannot be used on its own as a measure of model
adequacy. The reason for this is that the value of L̂, and hence of −2 log L̂,
is dependent upon the number of observations in the data set. Thus if, after
fitting a model to a set of data, additional data became available to which the
fit of the model was the same as that to the original data, the value of −2 log L̂
for the enlarged data set would be different from that of the original data.
Accordingly the value of −2 log L̂ is only useful when making comparisons
between models fitted to the same data.
3.5.2 Comparing nested models

Consider again Model (1) and Model (2) defined earlier, where Model (1)
contains p explanatory variables and Model (2) contains an additional q ex-
planatory variables. Let the value of the maximised partial likelihood function
for each model be denoted by L̂(1) and L̂(2), respectively. The two models
can then be compared on the basis of the difference between the values of
−2 log L̂ for each model. In particular, a large difference between −2 log L̂(1)
and −2 log L̂(2) would lead to the conclusion that the q variables in Model
(2), that are additional to those in Model (1), do improve the adequacy of
the model. Naturally, the amount by which the value of −2 log L̂ changes
when terms are added to a model will depend on which terms have already
been included. In particular, the difference in the values of −2 log L̂(1) and
−2 log L̂(2), that is, −2 log L̂(1) + 2 log L̂(2), will reflect the combined effect of
adding the variables Xp+1 , Xp+2 , . . . , Xp+q to a model that already contains
X1 , X2 , . . . , Xp . This is said to be the change in the value of −2 log L̂ due to
fitting Xp+1 , Xp+2 , . . . , Xp+q , adjusted for X1 , X2 , . . . , Xp .
The statistic −2 log L̂(1) + 2 log L̂(2), can be written as
−2 log{L̂(1)/L̂(2)},
and this is the log-likelihood ratio statistic for testing the null hypothesis that
the q parameters βp+1 , βp+2 , . . . , βp+q in Model (2) are all zero. From results
associated with the theory of likelihood ratio testing (see Appendix A), this
statistic has an asymptotic chi-squared distribution, under the null hypoth-
esis that the coefficients of the additional variables are zero. The number of
degrees of freedom of this chi-squared distribution is equal to the difference
between the number of independent β-parameters being fitted under the two
models. Hence, in order to compare the value of −2 log L̂ for Model (1) and
Model (2), we use the fact that the statistic −2 log L̂(1) + 2 log L̂(2) has a
chi-squared distribution on q degrees of freedom, under the null hypothesis
that βp+1 , βp+2 , . . . , βp+q are all zero. If the observed value of the statistic is
not significantly large, the two models will be adjudged to be equally suitable.
Then, other things being equal, the more simple model, that is, the one with
fewer terms, would be preferred. On the other hand, if the values of −2 log L̂
for the two models are significantly different, we would argue that the addi-
tional terms are needed and the more complex model would be adopted.
Although the difference in the values of the −2 log L̂ for two nested models
has an associated number of degrees of freedom, the −2 log L̂ statistic itself
does not. This is because the value of −2 log L̂ for a particular model does
not have a chi-squared distribution. Sometimes, −2 log L̂ is referred to as a
deviance. However, this is inappropriate, since unlike the deviance used in the
context of generalised linear modelling, −2 log L̂ does not measure deviation
from a model that is a perfect fit to the data.

Consider again the data from Example 1.2 on the survival times of breast
cancer patients. On fitting a Cox regression model that contains no explana-
tory variables, that is, the null model, the value of −2 log L̂ is 173.968. As in
Example 3.1, the indicator variable X, will be used to represent the result
of the staining procedure, so that X is zero for women whose tumours are
negatively stained and unity otherwise. When the variable X is included in
the linear component of the model, the value of −2 log L̂ decreases to 170.096.
The values of −2 log L̂ for alternative models are conveniently summarised in
tabular form, as illustrated in Table 3.5.
Table 3.5 Values of −2 log L̂ on

fitting Cox regression models to
the data from Example 1.2.
Variables in model −2 log L̂
none 173.968
X 170.096
The difference between the values of −2 log L̂ for the null model and the
model that contains X can be used to assess the significance of the difference
between the hazard functions for the two groups of women. Since one model
contains one more β-parameter than the other, the difference in the values of
−2 log L̂ has a chi-squared distribution on one degree of freedom. The differ-
ence in the two values of −2 log L̂ is 173.968 − 170.096 = 3.872, which is just
significant at the 5% level (P = 0.049). We may therefore conclude that there
is evidence, significant at the 5% level, that the hazard functions for the two
groups of women are different.
In Example 2.12, the extent of the difference between the survival times
of the two groups of women was investigated using the log-rank test. The chi-
squared value for this test was found to be 3.515 (P = 0.061). This value is
not very different from the figure of 3.872 (P = 0.049) obtained above. The
similarity of these two P -values means that essentially the same conclusions
are drawn about the extent to which the data provide evidence against the
null hypothesis of no group difference. From the practical viewpoint, the fact
that one result is just significant at the 5% level, while the other is not quite
significant at that level, is immaterial.
Although the model-based approach used in this example is operationally
different from the log-rank test, the two procedures are in fact closely related.
This relationship will be explored in greater detail in Section 3.13.
Example 3.4 Treatment of hypernephroma

In a study carried out at the University of Oklahoma Health Sciences Center,
data were obtained on the survival times of 36 patients with a malignant
tumour in the kidney, or hypernephroma. The patients had all been treated
with a combination of chemotherapy and immunotherapy, but additionally
a nephrectomy, the surgical removal of the kidney, had been carried out on
some of the patients. Of particular interest is whether the survival time of
the patients depends on their age at the time of diagnosis and on whether or
not they had received a nephrectomy. The data obtained in the study were
given in Lee and Wang (2013), but in this example, the age of a patient has
been classified according to whether the patient is less than 60, between 60
and 70 or greater than 70. Table 3.6 gives the survival times of the patients
in months.
In this example, there is a factor, age group, with three levels (< 60,
60–70, > 70), and a factor associated with whether or not a nephrectomy was
performed. There are a number of possible models for these data depending on
whether the hazard function is related to neither, one or both of these factors,
labelled Model (1) to Model (5). Under Model (1), the hazard of death does
not depend on either of the two factors and is the same for all 36 individuals in
the study. In Models (2) and (3), the hazard depends on either the age group
or on whether a nephrectomy was performed, but not on both. In Model (4),
the hazard depends on both factors, where the impact of nephrectomy on the
hazard is independent of the age group of the patient. Model (5) includes an
interaction between age group and nephrectomy, so that under this model the
effect of a nephrectomy on the hazard of death depends on the age group of
the patient.
Suppose that the effect due to the jth age group is denoted by αj , j =
1, 2, 3, and that due to nephrectomy status is denoted by νk , k = 1, 2. The
Table 3.6 Survival times of 36 patients classified ac-

cording to age group and whether or not they have had
a nephrectomy.
No nephrectomy Nephrectomy
< 60 60–70 > 70 < 60 60–70 > 70
9 15 12 104* 108* 10
6 8 9 26 9
21 17 56 14 18
35 115 6
52 52
68 5*
77* 18
84 36
8 9
38
72
36
48
26
108
5
terms αj and νk may then be included in Cox regression models for hi (t), the
hazard function for the ith individual in the study. The five possible models
are then as follows:
Model (1): hi (t) = h0 (t)
Model (2): hi (t) = exp{αj }h0 (t)
Model (3): hi (t) = exp{νk }h0 (t)
Model (4): hi (t) = exp{αj + νk }h0 (t)
Model (5): hi (t) = exp{αj + νk + (αν)jk }h0 (t)
To fit the term αj , two indicator variables A2 and A3 are defined with

values shown in Table 3.7. The term νk is fitted by defining a variable N
which takes the value zero when no nephrectomy has been performed and
unity when it has. With this choice of indicator variables, the baseline hazard
function will correspond to an individual in the youngest age group who has
not had a nephrectomy.
Models that contain the term αj are then fitted by including the variables
A2 , A3 in the model, while the term νk is fitted by including N . The interaction
is fitted by including the products A2 N = A2 × N and A3 N = A3 × N in
Table 3.7 Indicator variables

for age group.
Age group A2 A3
< 60 0 0
60–70 1 0
> 70 0 1
the model. The explanatory variables fitted, and the values of −2 log L̂ for
each of the five models under consideration, are shown in Table 3.8. When
computer software for modelling survival data enables factors to be included
in a model without having to define appropriate indicator variables, the values
of −2 log L̂ in Table 3.8 can be obtained directly.
Table 3.8 Values of −2 log L̂ on fitting five models to the data in Table 3.6.
Model Terms in model Variables in model −2 log L̂
(1) null model none 177.667
(2) αj A2 , A3 172.172
(3) νk N 170.247
(4) αj + νk A2 , A3 , N 165.508
(5) αj + νk + (αν)jk A2 , A3 , N, A2 N, A3 N 162.479
The first step in comparing these different models is to determine if there is

an interaction between nephrectomy status and age group. To do this, Model
(4) is compared with Model (5). The reduction in the value of −2 log L̂ on
including the interaction term in the model that contains the main effects of
age group and nephrectomy status is 165.508 − 162.479 = 3.029 on 2 d.f. This
is not significant (P = 0.220) and so we conclude that there is no interaction
between age group and whether or not a nephrectomy has been performed.
We now determine whether the hazard function is related to neither, one
or both of the factors age group and nephrectomy status. The change in the
value of −2 log L̂ on including the term αj in the model that contains νk
is 170.247 − 165.508 = 4.739 on 2 d.f. This is significant at the 10% level
(P = 0.094) and so there is some evidence that αj is needed in a model
that contains νk . The change in −2 log L̂ when νk is added to the model that
contains αj is 172.172 − 165.508 = 6.664 on 1 d.f., which is significant at the
1% level (P = 0.010). Putting these two results together, the term αj may
add something to the model that includes νk , and νk is certainly needed in the
model that contains αj . This means that both terms are required, and that
the hazard function depends on both the patient’s age group and on whether
or not a nephrectomy has been carried out.
Before leaving this example, let us consider other possible results from
the comparison of the five models, and how they would affect the conclusion
as to which model is the most appropriate. If the term corresponding to age
group, αj , was needed in a model in addition to the term corresponding to
STRATEGY FOR MODEL SELECTION 83
nephrectomy status, νk , and yet νk was not needed in the presence of αj , the
model containing just αj , Model (2), is probably the most suitable. To make
sure that αj was needed at all, Model (2) would be further compared with
Model (1), the null model. Similarly, if the term corresponding to nephrectomy
status, νk , was needed in addition to the term corresponding to age group, αj ,
but αj was not required in the presence of νk , Model (3) would probably be
satisfactory. However, the significance of νk would be checked by comparing
Model (3) with Model (1). If neither of the terms corresponding to age group
and nephrectomy status were needed in the presence of the other, a maximum
of one variable would be required. To determine which of the two is necessary,
Model (2) would be compared with Model (1) and Model (3) with Model (1).
If both results were significant, on statistical grounds, the model that leads
to the biggest reduction in the value of −2 log L̂ from that for the null model
would be adopted. If neither Model (2) nor Model (3) was superior to Model
(1), we would conclude that neither age group nor nephrectomy status had
an effect on the hazard function.
There are two further steps in the modelling approach to the analysis of
survival data. First, we will need to critically examine the fit of a model to the
observed data in order to ensure that the fitted Cox regression model is indeed
appropriate. Second, we will need to interpret the parameter estimates in the
chosen model, in order to quantify the effect that the explanatory variables
have on the hazard function. Interpretation of parameters in a fitted model is
considered in Section 3.9, while methods for assessing the adequacy of a fitted
model will be considered in Chapter 4. But first, possible strategies for model
selection are discussed.
3.6 Strategy for model selection
An initial step in the model selection process is to identify a set of explanatory

variables that have the potential for being included in the linear component of
a Cox regression model. This set will contain those variates and factors that
have been recorded for each individual, but additional terms corresponding
to interactions between factors or between variates and factors may also be
required.
Once a set of potential explanatory variables has been isolated, the com-
bination of variables that are to be used in modelling the hazard function has
to be determined. In practice, a hazard function will not depend on a unique
combination of variables. Instead, there are likely to be a number of equally
good models, rather than a single ‘best’ model. For this reason, it is desirable
to consider a wide range of possible models.
The model selection strategy depends to some extent on the purpose of
the study. In some applications, information on a number of variables will
have been obtained, and the aim might be to determine which of them has
an effect on the hazard function, as in Example 1.3 on multiple myeloma.
In other situations, there may be one or more variables of primary interest,
such as terms corresponding to a treatment effect. The aim of the modelling
process is then to evaluate the effect of such variables on the hazard function,
as in Example 1.4 on prostatic cancer. Since the other variables that have been
recorded might also be expected to influence the magnitude of the treatment
effect, these variables will need to be taken account of in the modelling process.
An important principle in statistical modelling is that when a term corre-
sponding to the interaction between two factors is to be included in a model,
the corresponding lower-order terms should also be included. This rule is
known as the hierarchic principle, and means, for example, that interactions
between two factors should not be fitted unless the corresponding main effects
are present. Models that are not hierarchic are difficult to interpret.
3.6.1 Variable selection procedures

We first consider the situation where all explanatory variables are on an equal
footing, and the aim is to identify subsets of variables upon which the haz-
ard function depends. When the number of potential explanatory variables,
including interactions, non-linear terms and so on, is not too large, it might
be feasible to fit all possible combinations of terms, paying due regard to the
hierarchic principle. Alternative nested models can be compared by examining
the change in the value of −2 log L̂ on adding terms into a model or deleting
terms from a model.
Comparisons between a number of possible models, which need not nec-
essarily be nested, can also be made on the basis of Akaike’s information
criterion, given by
AIC = −2 log L̂ + 2q,
in which q is the number of unknown β-parameters in the model. The smaller

the value of this statistic, the better the model, but unlike the −2 log L̂ statis-
tic, the value of AIC will tend to increase when unnecessary terms are added
to the model.
An alternative to the AIC statistic is the Bayesian Information Criterion,
given by
BIC = −2 log L̂ + q log d,
where q is the number of unknown parameters in the fitted model and d is the
number of uncensored observations in the data set. The BIC statistic is also
known as the Schwarz Bayesian Criterion and denoted SBC. The Bayesian
Information Criterion is an adjusted value of the −2 log L̂ statistic that takes
account of both the number of unknown parameters in the fitted model and
the number of observations to which the model has been fitted. As for the
AIC statistic, smaller values of BIC are obtained for better models.
Of course, some terms may be identified as alternatives to those in a par-
ticular model, leading to subsets that are equally suitable. The decision on
which of these subsets is the most appropriate should not then rest on sta-
tistical grounds alone. When there are no subject matter grounds for model
choice, the model chosen for initial consideration from a set of alternatives
might be the one for which the value of −2 log L̂, AIC or BIC is a minimum.
It will then be important to confirm that the model does fit the data using
the methods for model checking described in Chapter 4.
In some applications, information might be recorded on a number of vari-
ables, all of which relate to the same general feature. For example, the variables
height, weight, body mass index (weight/height2 ), head circumference, arm
length and so on, are all concerned with the size of an individual. In view
of inter-relationships between these variables, a model for the survival times
of these individuals may not need to include each of them. It would then be
appropriate to determine which variables from this group should be included
in the model, although it may not matter exactly which variables are chosen.
When the number of variables is relatively large, it can be computation-
ally expensive to fit all possible models. In particular, if there is a pool of p
potential explanatory variables, there are 2p possible combinations of terms,
so that if p > 10, there are more than a thousand possible combinations of
explanatory variables. In this situation, automatic routines for variable selec-
tion that are available in many software packages might seem an attractive
prospect. These routines are based on forward selection, backward elimination
or a combination of the two known as the stepwise procedure.
In forward selection, variables are added to the model one at a time. At
each stage in the process, the variable added is the one that gives the largest
decrease in the value of −2 log L̂ on its inclusion. The process ends when the
next candidate for inclusion in the model does not reduce the value of −2 log L̂
by more than a prespecified amount. This is known as the stopping rule. This
rule is often couched in terms of the significance level of the difference in the
values of −2 log L̂ when a variable is added to a model, so that the selection
process ends when the next term for inclusion ceases to be significant at a
preassigned level.
In backward elimination, a model that contains the largest number of
variables under consideration is first fitted. Variables are then excluded one
at a time. At each stage, the variable omitted is the one that increases the
value of −2 log L̂ by the smallest amount on its exclusion. The process ends
when the next candidate for deletion increases the value of −2 log L̂ by more
than a prespecified amount.
The stepwise procedure operates in the same way as forward selection.
However, a variable that has been included in the model can be considered
for exclusion at a later stage. Thus, after adding a variable to the model, the
procedure then checks whether any previously included variable can now be
deleted. These decisions are again made on the basis of prespecified stopping
rules.
These automatic routines have a number of disadvantages. Typically, they
lead to the identification of one particular subset, rather than a set of equally
good ones. The subsets found by these routines often depend on the variable
selection process that has been used, that is, whether it is forward selection,
backward elimination or the stepwise procedure, and generally tend not to
take any account of the hierarchic principle. They also depend on the stop-
ping rule that is used to determine whether a term should be included in
or excluded from a model. For all these reasons, these automatic routines
have a limited role in model selection, and should certainly not be used
uncritically.
Instead of using automatic variable selection procedures, the following gen-
eral strategy for model selection is recommended.
1. The first step is to fit models that contain each of the variables one at a
time. The values of −2 log L̂ for these models are then compared with that
for the null model to determine which variables on their own significantly
reduce the value of this statistic.
2. The variables that appear to be important from Step 1 are then fitted
together. In the presence of certain variables, others may cease to be im-
portant. Consequently, those variables that do not significantly increase
the value of −2 log L̂ when they are omitted from the model can now be
discarded. We therefore compute the change in the value of −2 log L̂ when
each variable on its own is omitted from the set. Only those that lead to a
significant increase in the value of −2 log L̂ are retained in the model. Once
a variable has been dropped, the effect of omitting each of the remaining
variables in turn should be examined.
3. Variables that were not important on their own, and so were not under
consideration in Step 2, may become important in the presence of others.
These variables are therefore added to the model from Step 2, one at a
time, and any that reduce −2 log L̂ significantly are retained in the model.
This process may result in terms in the model determined at Step 2 ceasing
to be significant.
4. A final check is made to ensure that no term in the model can be omitted
without significantly increasing the value of −2 log L̂, and that no term not
included significantly reduces −2 log L̂.
When using this selection procedure, rigid application of a particular sig-

nificance level should be avoided. In order to guide decisions on whether to
include or omit a term, the significance level should not be too small, for
otherwise too few variables will be selected for inclusion; a level of 15% is
recommended for general use.
In some applications, a small number of interactions and other higher-
order terms, such as powers of certain variates, may need to be considered
for inclusion in a model. Such terms would be added to the model identified
in Step 3 above, after ensuring that any terms necessitated by the hierarchic
principle have already been included in the model. If any higher-order term
leads to a significant reduction in the value of −2 log L̂, that term would be
included in the model.
This model selection strategy is now illustrated in an example.

The analysis of the data on the survival times of multiple myeloma patients
in Example 3.2 suggested that not all of the seven explanatory variables, Age,
Sex, Bun, Ca, Hb, Pcells and Protein, are needed in a Cox regression model.
We now determine the most appropriate subsets of these variables. In this
example, transformations of the original variables and interactions between
them will not be considered. We will further assume that there are no medical
grounds for including particular variables in a model. A summary of the values
of −2 log L̂ for all models that are to be considered is given in Table 3.9.
Table 3.9 Values of −2 log L̂ for models

fitted to the data from Example 1.3.
none 215.940
Age 215.817
Sex 215.906
Bun 207.453
Ca 215.494
Hb 211.068
Pcells 215.875
Protein 213.890
Hb + Bun 202.938
Hb + Protein 209.829
Bun + Protein 203.641
Bun + Hb + Protein 200.503
Hb + Bun + Age 202.669
Hb + Bun + Sex 202.553
Hb + Bun + Ca 202.937
Hb + Bun + Pcells 202.773
The first step is to fit the null model and models that contain each of the
seven explanatory variables on their own. Of these variables, Bun leads to the
largest reduction in −2 log L̂, reducing the value of the statistic from 215.940
to 207.453. This reduction of 8.487 is significant at the 1% level (P = 0.004)
when compared with percentage points of the chi-squared distribution on 1 d.f.
The reduction in −2 log L̂ on adding Hb to the null model is 4.872, which is
also significant at the 5% level (P = 0.027). The only other variable that on
its own has some explanatory power is Protein, which leads to a reduction
in −2 log L̂ that is nearly significant at the 15% level (P = 0.152). Although
this P -value is relatively high, we will for the moment keep Protein under
consideration for inclusion in the model.
The next step is to fit the model that contains Bun, Hb and Protein,
which leads to a value of −2 log L̂ of 200.503. The effect of omitting each of
the three variables in turn from this model is shown in Table 3.9. In particular,
when Bun is omitted, the increase in −2 log L̂ is 9.326, when Hb is omitted
the increase is 3.138, and when Protein is omitted it is 2.435. Each of these
changes in the value of −2 log L̂ can be compared with percentage points of a
chi-squared distribution on 1 d.f. Since Protein does not appear to be needed
in the model, in the presence of Hb and Bun, this variable will not be further
considered for inclusion.
If either Hb or Bun is excluded from the model that contains both of these
variables, the increase in −2 log L̂ is 4.515 and 8.130, respectively. Both of
these increases are significant at the 5% level, and so neither Hb nor Bun can
be excluded from the model without significantly increasing the value of the
−2 log L̂ statistic.
Finally, we look to see if any of variables Age, Sex, Ca and Pcells should be
included in the model that contains Bun and Hb. Table 3.9 shows that when
any of these four variables is added, the reduction in −2 log L̂ is less than 0.5,
and so none of them need to be included in the model. We therefore conclude
that the most satisfactory model is that containing Bun and Hb.
We now turn to studies where there are variables of primary importance,

such as a treatment effect. Here, we proceed in the following manner.
1. The important prognostic variables are first selected, ignoring the treatment
effect. Models with all possible combinations of the variables can be fitted
when their number is not too large. Alternatively, the variable selection
process might follow similar lines to those described previously in Steps 1
to 4.
2. The treatment effect is then included in the model. In this way, any differ-
ences between the two groups that arise as a result of differences between
the distributions of the prognostic variables in each treatment group, are
not attributed to the treatment.
3. If the possibility of interactions between the treatment and other explana-
tory variables has not been discounted, these must be considered before the
treatment effect can be interpreted.
Additionally, it will often be of interest to fit a model that contains the treat-

ment effect alone. This enables the effect that the prognostic variables have
on the magnitude of the treatment effect to be evaluated.
In this discussion on strategies for model selection, the use of statistical
criteria to guide the selection process has been emphasised. In addition, due
account must be taken of the application area. In particular, on subject area
grounds, it may be inappropriate to include particular combinations of vari-
ables. On the other hand, there might be some variables that it is not sensible
to omit from the model, even if they appear not to be needed in modelling a
particular data set. Indeed, there is always a need for non-statistical consid-
erations in model building.

In the data from Example 1.4 on the survival times of 38 prostatic cancer
patients, there are four prognostic variables that might have an effect on the
survival times. These are the age of the patient (Age), serum haemoglobin level
(Shb), tumour size (Size) and Gleason index (Index). All possible combinations
of these variates are fitted in a Cox regression model and the values of −2 log L̂
computed. These are shown in Table 3.10, together with the values of Akaike’s
information criterion, obtained from AIC = −2 log L̂ + 2q and the Bayesian
Information Criterion obtained from BIC = −2 log L̂ + q log(6), where q is
the number of terms in a model fitted to a data set with 6 death times.
Table 3.10 Values of −2 log L̂, AIC and BIC for models fitted
to the data from Example 1.4.
Variables in model −2 log L̂ AIC BIC
none 36.349 36.349 36.349
Age 36.269 38.269 38.061
Shb 36.196 38.196 37.988
Size 29.042 31.042 30.834
Index 29.127 31.127 30.919
Age + Shb 36.151 40.151 39.735
Age + Size 28.854 32.854 32.438
Age + Index 28.760 32.760 32.344
Shb + Size 29.019 33.019 32.603
Shb + Index 27.981 31.981 31.565
Size + Index 23.533 27.533 27.117
Age + Shb + Size 28.852 34.852 34.227
Age + Shb + Index 27.893 33.893 33.268
Age + Size + Index 23.269 29.269 28.644
Shb + Size + Index 23.508 29.508 28.883
Age + Shb + Size + Index 23.231 31.231 30.398
The two most important explanatory variables when considered separately

are Size and Index. From the change in the value of −2 log L̂ on omitting either
of them from a model that contains both, we deduce that both variables are
needed in a Cox regression model. The value of −2 log L̂ is only reduced by a
very small amount when Age and Shb are added to the model that contains
Size and Index. We therefore conclude that only Size and Index are important
prognostic variables.
From the values of Akaike’s information criterion in Table 3.10, the model
with Size and Index leads to the smallest value of the statistic, confirming
that this is the most suitable model of those tried. Notice also that there are
no other combinations of explanatory variables that lead to similar values of
the AIC-statistic, which shows that there are no obvious alternatives to using
Size and Index in the model. The same conclusions follow from the values of
the Bayesian Information Criterion.
We now consider the treatment effect. Let Treat be a variable that takes
the value zero for individuals allocated to the placebo, and unity for those
allocated to diethylstilbestrol. When Treat is added to the model that con-
tains Size and Index, the value of −2 log L̂ is reduced to 22.572. This re-
duction of 0.961 on 1 d.f. is not significant (P = 0.327). This indicates
that there is no treatment effect, but first we ought to examine whether
the coefficients of the two explanatory variables in the model depend on
treatment. To do this, we form the products Tsize = Treat × Size and
Tindex = Treat × Index , and add these to the model that contains Size,
Index and Treat. When Tsize and Tindex are added to the model, −2 log L̂
is reduced to 20.829 and 20.792, respectively. On adding both of these mixed
terms, −2 log L̂ becomes 19.705. The reductions in −2 log L̂ on adding these
terms to the model are not significant, and so there is no evidence that the
treatment effect depends on Size and Index. This means that our original
interpretation of the size of the treatment effect is valid, and that on the ba-
sis of these data, treatment with DES does not appear to affect the hazard
of death. The estimated size of this treatment effect will be considered later
in Example 3.12.
Before leaving this example, we note that when either Tsize or Tindex is
added to the model, their estimated coefficient, and that of Treat, become
large. The standard errors of these estimates are also very large. In particular,
in the model that contains Size, Index, Treat and Tsize, the estimated coef-
ficient of Treat is −11.28 with a standard error of 18.50. For the model that
contains Size, Index, Treat and Tindex, the coefficients of Treat and Tindex are
−161.52 and 14.66, respectively, while the standard errors of these estimates
are 18476 and 1680, respectively! This is evidence of overfitting.
In an overfitted model, the estimated values of some of the β-coefficients
will be highly dependent on the actual data. A very slight change to the val-
ues of one of these variables could then have a large impact on the estimate
of the corresponding coefficient. This is the reason for such estimates hav-
ing large standard errors. In this example, overfitting occurs because of the
small number of events in the data set: there are only 6 of the 38 patients
who die.
An overfitted model is one that is more complicated than is justified by
the data, and does not provide a useful summary of the data. This is another
reason for not including the mixed terms in the model for the hazard of death
from prostatic cancer.
3.7 ∗ Variable selection using the lasso

A particularly useful aid to variable selection is the Least Absolute Shrinkage
and Selection Operator, referred to as the lasso. The effect of using the lasso is
to shrink the coefficients of explanatory variables in a model towards zero and
VARIABLE SELECTION USING THE LASSO 91
in so doing, some estimates are set automatically to exactly zero. Explanatory
variables whose coefficients become zero in the lasso procedure will be those
that have little or no explanatory power, or that are highly correlated with
others. The variables selected for inclusion in the model are then those with
non-zero coefficients. This shrinkage also improves the predictive ability of a
model, since the shrunken parameter estimates are less susceptible to changes
in the sample data, and so are more stable.
3.7.1 The lasso in Cox regression modelling

Consider a Cox regression model that contains p explanatory variables, where
the hazard of an event occurring at time t for the ith of n individuals is
hi (t) = exp(β ′ xi )h0 (t), where β ′ xi = β1 x1i + β2 x2i + · · · + βp xpi , and h0 (t)
is the baseline hazard function. In the lasso procedure, the β-parameters in
this model are estimated by maximising the partial likelihood function in
Equation (3.5), while constraining the sum of their absolute values to be less
than or equal to some value s. Denote the set of individuals at risk of an
event at time ti , the observed survival time of the ith individual, by R(ti ),
and let δi be the event indicator that is zero when ti is a censored survival
time and unity otherwise. The partial likelihood function for the model with
p explanatory variables is then
{ }δi
∏
n
exp(β ′ xi )
L(β) = ∑ ′ , (3.14)
i=1 l∈R(ti ) exp(β xl )
∑p
which is maximised subject to the constraint that j=1 |βj | 6 s. The quantity
∑p
j=1 |βj | is called the L1 -norm of the vector β, and is usually denoted ||β||1 .
The estimates β̂ that maximise the constrained partial likelihood function are
also the values that maximise
∑
p
Lλ (β) = L(β) − λ |βj |, (3.15)
j=1
where λ is called the lasso parameter. The likelihood function in Equa-

tion (3.15) is referred to as a penalised likelihood, since a penalty is assigned
to the more extreme values of the βs. The resulting estimates will be closer to
zero and have greater precision than the standard estimates on fitting a Cox
model, at the cost of introducing a degree of bias in the estimates.
To see the effect of this shrinkage, estimates of the coefficients of all the
explanatory variables that might potentially be included in the final model are
obtained for a range of values of the lasso parameter. A plot of these estimates
against λ, known as the lasso trace, provides an informative summary of the
dependence of the estimates on the value of λ. When λ is small, few variables in
the model will have zero coefficients, and the number of explanatory variables
will not be reduced by much. On the other hand, large values of λ result in
too many variables being excluded and an unsatisfactory model.
Although the lasso trace is a useful graphical aid to identifying a suitable
value of λ to use in the modelling process, this is usually supplemented by
techniques that enable the optimal value of the lasso parameter to be deter-
mined. This value could be taken to be that which maximises the penalised
likelihood, Lλ (β) in Equation (3.15), or equivalently the value of λ that min-
imises −2 log Lλ (β). However, other methods for determining the optimum
value of λ are generally preferred, such as the cross-validation method. In this
approach, the optimal value of λ is that which maximises the cross-validated
partial log-likelihood function. To define this, let β̂ (−i) (λ) be the estimated
vector of parameters on maximising the penalised partial likelihood function
in Equation (3.15) when the data for the ith individual are omitted from the
fitting process, i = 1, 2, . . . , n, and for a given value of the lasso parameter λ.
Writing log L(β) for the logarithm of the partial likelihood function defined in
Equation (3.14), let log L(−i) (β) be the partial log-likelihood when the data
for the ith individual are excluded. The cross-validated partial log-likelihood
function is then
∑ n { }
log L̂CV (λ) = log L[β̂ (−i) (λ)] − log L(−i) [β̂ (−i) (λ)] ,
i=1
and the value of λ that maximises this function is chosen as the optimal value
for the lasso parameter. The variables selected for inclusion in the model are
then those with non-zero coefficients at this value of λ.
Following use of the lasso procedure, standard errors of the parameter es-
timates, or functions of these estimates such as hazard ratios, are not usually
presented. One reason for this is that they are difficult to compute, but the
main reason is that the parameter estimates will be biased towards zero. As a
result, standard errors are not very meaningful and will tend to underestimate
the precision of the estimates. If standard errors are required, the lasso pro-
cedure could be used to determine the variables that should be included in a
model, that is those with non-zero coefficients at an optimal value of the lasso
parameter. A standard Cox regression model that contains these variables is
then fitted, but the advantages of shrinkage are then lost.
3.7.2 Data preparation

Before using the lasso procedure, the explanatory variables must be on similar
scales of measurement. When this is not the case, each of the explanatory vari-
ables is first standardised to have zero mean and unit variance, by subtracting
the sample mean and dividing by the standard deviation.
The lasso process can also be applied to categorical variables, after first
standardising the indicator variables associated with the factor. This approach
means that a factor is not regarded as a single entity, and the lasso procedure
may result in a subset of the associated indicator variables being set to zero.
VARIABLE SELECTION USING THE LASSO 93
The variables that are identified by the lasso procedure will then depend on
how the corresponding indicator variables have been coded. For example, if
the m − 1 indicator variables corresponding to an m-level factor are such that
they are all zero for the first level of the factor, as illustrated in Section 3.2,
the indicator variables will measure the effect of each factor level relative to
the first. This may be helpful as a means of identifying factor levels that do
not differ from a baseline level, but is inappropriate when, for example, the
factor levels are ordered. In this case, a coding that reflects differences in the
levels of the factor, known as forward difference coding, would be more useful.
Coefficients that are set to zero in the lasso procedure would then correspond
to levels that have similar effects to adjacent levels with non-zero coefficients.
To illustrate this coding, suppose that an ordered categorical variable A
has four levels. This factor might be represented by three indicator variables
A1 , A2 , A3 , defined in Table 3.11.

using forward difference coding.
Level of A A1 A2 A3
1 3/4 1/2 1/4
2 −1/4 1/2 1/4
3 −1/4 −1/2 1/4
4 −1/4 −1/2 −3/4
When these three variables are included in a model with coefficients

α1 , α2 , α3 , the effect of level 1 of A is (3α1 + 2α2 + α3 )/4, and that of
level 2 is (−α1 + 2α2 + α3 )/4. The coefficient of A1 , α1 , therefore rep-
resents the difference between level 1 and level 2 of A. Similarly, the co-
efficient of A2 represents the difference between levels 2 and 3 of A and
the coefficient of A3 represents the difference between levels 3 and 4. In
the general case of a factor A with m levels, the m − 1 indicator vari-
ables are such that A1 has values (m − 1)/m, −1/m, −1/m, . . . , −1/m, A2
has values (m − 2)/m, (m − 2)/m, −2/m, −2/m, . . . , −2/m, A3 has values
(m − 3)/m, (m − 3)/m, (m − 3)/m, −3/m, . . . , −3/m, and so on until Am−1
has values 1/m, 1/m, . . . , 1/m, −(m − 1)/m. Categorical variables can also be
accommodated using the group lasso, that enables a complete set of variables
to be included or excluded, but further details will not be given here.

The use of the lasso procedure in variable selection will be illustrated using
data on the survival times of 48 patients with multiple myeloma that were
first given in Example 1.3 of Chapter 1. The seven potential explanatory
variables include five continuous variables (Age, Bun, Ca, Hb, Pcells) and two
binary variables (Sex, Protein). All seven variables have very different scales
of measurement, and so they are first standardised to have a sample mean
of zero and unit standard deviation. For example, the 48 values of Age have
a mean of 62.90 and a standard deviation of 6.96, and so the standardised
values are (Age − 62.90)/6.96.
To apply the lasso procedure, estimates of the seven β-coefficients are ob-
tained by maximising the penalised likelihood function in Equation (3.15) for
a range of values of λ. The resulting trace of the coefficients of each variable,
plotted against λ, is shown in Figure 3.2. In this plot, the coefficients presented
are those of the original variables, obtained by dividing the estimated coeffi-
cient of the standardised explanatory variable that maximises the constrained
likelihood function, by the standard deviation of that variable.
0.2
Bun
0.0 Ca
Pcells
Age
Estimated coefficient
Hb
-0.2
Sex
-0.4
-0.6
Protein
-0.8
0 2 4 6 8 10
Value of
Figure 3.2 Trace of the estimated coefficients of the explanatory variables as a func-
tion of the lasso parameter, λ.
The estimates when λ = 0 are the parameter estimates on fitting a Cox

regression model that contains all seven explanatory variables. As λ increases,
these estimates get closer to zero, but at differing rates. The estimated co-
efficient of Ca has become zero when λ = 0.5 and that of Pcells is zero by
λ = 0.75. The estimated coefficients of Age and Sex are both zero by λ = 2.5.
This figure also illustrates another property of the lasso, which is that the
lasso trace is formed from a number of straight line segments, so that it is
piecewise linear.
To determine the optimal value of λ, the cross-validated partial log-
likelihood, log L̂CV (λ), is evaluated for a range of λ values, and the value
that maximises this likelihood function determined. The cross-validated par-
tial log-likelihood is shown as a function of the lasso parameter, λ, in Fig-
ure 3.3. This function is a maximum when λ = 3.90, and at this value of λ,
there are three variables with non-zero coefficients, Hb, Bun and Protein. The
lasso procedure therefore leads to a model that contains these three variables.
NON-LINEAR TERMS 95
Cross-validated partial log-likelihood -135
-140
-145
-150
0 2 4 6 8 10
Value of
Figure 3.3 Cross-validated partial log-likelihood as a function of the lasso parameter,

showing the optimum value of λ.
In this example, the cross-validated partial log-likelihood function shown in

Figure 3.3 is quite flat around its maximum, and has very similar values for λ
between 3 and 5. Also, from Figure 3.2, the same variables would be selected
for inclusion in the model for any value of λ between 3 and 5, although the
estimated coefficients will differ.
The coefficients of Hb, Bun and Protein in the model with λ = 3.90 are
−0.080, 0.015 and −0.330, respectively. The corresponding estimates from a
fitted Cox model with these three variables are −0.110, 0.020, and −0.617.
Notice that the estimated coefficients in the model from the lasso procedure
are all closer to zero than the corresponding estimates in the Cox model,
illustrating the shrinkage effect. In the Cox model, the coefficient of Pro-
tein is not significantly different from zero (P = 0.13), although the lasso
procedure suggests that this variable should be retained in the model. Ulti-
mately, one might wish to include Protein in the model so as not to miss
anything.
3.8 Non-linear terms
When the dependence of the hazard function on an explanatory variable that

takes a wide range of values is to be modelled, we should consider whether it
is appropriate to include that variable as a linear term in the Cox regression
model. If there are reasons for not assuming that a variable is linear, we then
need to consider how the non-linearity is modelled.
3.8.1 Testing for non-linearity
A straightforward way of examining whether non-linear terms are needed is
to add quadratic or cubic terms to the model, and examine the consequent
reduction in the value of the −2 log L̂ statistic. If the inclusion of such terms
significantly reduces the value of this statistic, we would conclude that there
is non-linearity. Polynomial terms might then be included in the model. How-
ever, in many situations, non-linearity in an explanatory variable cannot be
adequately represented by the inclusion of polynomial terms in a Cox regres-
sion model. The following procedure is therefore recommended for general
use.
To determine whether a variable exhibits non-linearity, the values of a pos-
sibly non-linear variate are first grouped into four or five categories containing
approximately equal numbers of observations. A factor is then defined whose
levels correspond to this grouping. For example, a variate reflecting the size
of a tumour could be fitted as a factor whose levels correspond to very small,
small, medium and large.
More specifically, let A be a factor with m levels formed from a continuous
variate, and let X be a variate that takes the value j when A is at level j,
for j = 1, 2, . . . , m. Linearity in the original variate will then correspond to
there being a linear trend across the levels of A. This linear trend can be
modelled by fitting X alone. Now, fitting the m − 1 terms X, X 2 , . . . , X m−1 is
equivalent to fitting A as a factor in the model, using indicator variables as in
Section 3.2.2. Accordingly, the difference between the value of −2 log L̂ for the
model that contains X, and that for the model that contains A, is a measure
of non-linearity across the levels of A. If this difference is not significant we
would conclude that there is no non-linearity and the original variate would
be fitted. On the other hand, if there is evidence of non-linearity the actual
form of this non-linearity can be further studied from the coefficients of the
indicator variables corresponding to A. A plot of these coefficients may help
in establishing the nature of any trend across the levels of the factor A.

In Example 3.5, we found that a Cox regression model that contained the
explanatory variables Bun and Hb appeared to be appropriate for the data
on the survival times of multiple myeloma patients. We now consider whether
there is any evidence of non-linearity in the values of serum haemoglobin level,
and examine whether a quadratic term is needed in the model that contains
Bun and Hb. When the term Hb 2 is added to this model, the value of −2 log L̂
is reduced from 202.938 to 202.917. This reduction of 0.021 on 1 d.f. is clearly
not significant, which suggests that a linear term in Hb is sufficient.
An alternative way of examining the extent of non-linearity is to use a
factor to model the effect of serum haemoglobin level on the hazard function.
Suppose that a factor with four levels is defined, where level 1 corresponds to
values of Hb less than or equal to 7, level 2 to values between 7 and 10, level 3
to values between 10 and 13 and level 4 to values greater than 13. This choice
NON-LINEAR TERMS 97
of levels corresponds roughly to the quartiles of the distribution of the values
of Hb. This factor can be fitted by defining three indicator variables, Hb2, Hb3
and Hb4, which take the values shown in Table 3.12.
Table 3.12 Indicator variables for a factor corresponding to

values of the variable Hb.
Level of factor (X) Value of Hb Hb2 Hb3 Hb4
1 Hb 6 7 0 0 0
2 7 < Hb 6 10 1 0 0
3 10 < Hb 6 13 0 1 0
4 Hb > 13 0 0 1
When a model containing Bun, Hb2, Hb3 and Hb4 is fitted, the value of
−2 log L̂ is 200.417. The change in the value of this statistic on adding the
indicator variables Hb2, Hb3 and Hb4 to the model that contains Bun alone
is 7.036 on 3 d.f., which is significant at the 10% level (P = 0.071). However,
it is difficult to identify any pattern across the factor levels.
A linear trend across the levels of the factor corresponding to haemoglo-
bin level can be modelled by fitting the variate X, which takes values 1, 2,
3, 4, according to the factor level. When the model containing Bun and X
is fitted, −2 log L̂ is 203.891, and the change in the value of −2 log L̂ due to
any non-linearity is 203.891 − 200.417 = 3.474 on 2 d.f. This is not significant
when compared with percentage points of the chi-squared distribution on 2 d.f.
(P = 0.176). We therefore conclude that the effect of haemoglobin level on
the hazard of death in this group of patients is adequately modelled by using
the linear term Hb.
3.8.2 Modelling non-linearity

If non-linearity is detected using the procedure described in Section 3.8.1,
it may be tempting to use the factor corresponding to the variable in the
modelling process. However, this means that a continuous variable is being
replaced by a step-function. Such a representation of an inherently continuous
variable is not usually plausible from a subject matter viewpoint. In addition,
this procedure requires category boundaries to be chosen, and the process of
categorisation leads to a loss in information. The use of polynomial terms to
represent non-linear behaviour in an explanatory variable is also not generally
recommended. This is because low order polynomials, such as a quadratic
or cubic expression, may not be a good fit to the data, and higher-order
polynomials do not usually fit well in the extremes of the range of values of
an explanatory variable. In addition, variables that have limiting values, or
asymptotes, cannot be adequately modelled using polynomial expressions. A
straightforward yet flexible solution is to use a model that contains different
powers of the same variable, which may be fractional, referred to as fractional
polynomials.
3.8.3 Fractional polynomials
A fractional polynomial in a variable X of order m contains m different pow-
ers of X. The expression β1 X p1 + β2 X p2 + · · · + βm X pm is then included
in the model, where each power pj , j = 1, 2, . . . , m, is taken to be one
of the values in the set {−2, −1, −0.5, 0, 0.5, 1, 2, 3}, with X 0 taken to be
log X. The representation of X to the power of 0 by log X is called the Box-
Tidwell transformation of X. Considerable flexibility in modelling the impact
of an explanatory variable on the hazard of death can be achieved by us-
ing just two different powers of the variable, and so m is generally taken to
be either 1 or 2. When m = 2, we can without any loss of generality take
p1 < p2 , since a model with powers p1 , p2 is the same as one with powers
p2 , p1 , and models with p1 = p2 are equivalent to the corresponding model
with m = 1.
With m = 1, models with the 8 possible powers would be fitted, and that
with the smallest value of −2 log L̂, in the presence of other variables under
consideration, is the best fitting model. When m = 2, there are 28 possi-
ble combinations of the 8 powers in the set, excluding the 8 cases where the
variable appears twice with the same power, and again the most appropriate
combination would be the one for which −2 log L̂ is minimised. When com-
paring non-nested models of different orders, for example a model with two
powers rather than one, where no power is common to both models, the AIC
or BIC statistics can be used.

In this example, we investigate whether there is evidence of non-linearity in
serum haemoglobin level in the data set on the survival times of multiple
myeloma patients. Fractional polynomials in Hb of order 1 and 2 are fitted,
in addition to the variable Bun. Thus, Hb is included in the model as a single
term with powers p1 and as two terms with powers p1 < p2 , where p1 and p2
are drawn from the set of values {−2, −1, −0.5, 0, 0.5, 1, 2, 3} and where Hb
raised to the power of zero is taken to mean log Hb. The values of −2 log L̂ for
Cox regression models with
hi (t) = exp(β1 Hb pi 1 + β2 Bun i )h0 (t),
and
hi (t) = exp(β1 Hb pi 1 + β2 Hb pi 2 + β3 Bun i )h0 (t),
are shown in Table 3.13.
From this table, the best models with just one power of Hb, that is for
m = 1, are those with a linear or a quadratic term, and of these, the model
with Hb alone is the simplest. When models with two powers of Hb are fitted,
that with p1 = −2 and p2 = −1 or −0.5 lead to the smallest values of −2 log L̂,
but neither leads to a significant improvement on the model with just one
power of Hb. If another power of Hb was to be added to the model that
includes Hb alone, we would add Hb−2 , but again there is no need to do this
INTERPRETATION OF PARAMETER ESTIMATES 99
Table 3.13 Values of −2 log L̂ on fitting fractional polynomials in Hb of order m =

1, 2 to the data from Example 1.3.
m=1 m=2
p1 −2 log L̂ p1 p2 −2 log L̂ p1 p2 −2 log L̂ p1 p2 −2 log L̂
−2 204.42 −2 −1 202.69 −1 1 202.81 0 2 202.94
−1 203.77 −2 −0.5 202.69 −1 2 202.94 0 3 203.04
−0.5 203.48 −2 0 202.71 −1 3 203.10 0.5 1 202.86
0 203.23 −2 0.5 202.74 −0.5 0 202.75 0.5 2 202.93
0.5 203.05 −2 1 202.79 −0.5 0.5 202.78 0.5 3 202.99
1 202.94 −2 2 202.94 −0.5 1 202.83 1 2 202.92
2 202.94 −2 3 203.14 −0.5 2 202.94 1 3 202.94
3 203.20 −1 −0.5 202.71 −0.5 3 203.07 2 3 202.80
−1 0 202.73 0 0.5 202.81
−1 0.5 202.76 0 1 202.84
as no model with two powers of Hb is a significant improvement on the model

with Hb alone. We conclude that a linear term in Hb suffices, confirming the
results of the analysis in Example 3.8.
3.9 Interpretation of parameter estimates

When a Cox regression model is used in the analysis of survival data, the
coefficients of the explanatory variables in the model can be interpreted as
logarithms of the ratio of the hazard of death to the baseline hazard. This
means that estimates of this hazard ratio, and corresponding confidence in-
tervals, can easily be found from the fitted model. The interpretation of pa-
rameters corresponding to different types of term in the Cox regression model
is described in the following sections.
3.9.1 Models with a variate

Suppose that a Cox regression model contains a single continuous variable X,
so that the hazard function for the ith of n individuals, for whom X takes the
value xi , is
hi (t) = eβxi h0 (t).
The coefficient of xi in this model can then be interpreted as the logarithm
of a hazard ratio. Specifically, consider the ratio of the hazard of death for
an individual for whom the value x + 1 is recorded on X, relative to one for
whom the value x is obtained. This is
exp{β(x + 1)}
= eβ ,
exp(βx)
and so β̂ in the fitted Cox regression model is the estimated change in the
logarithm of the hazard ratio when the value of X is increased by one unit.
Using a similar argument, the estimated change in the log-hazard ratio
when the value of the variable X is increased by r units is rβ̂, and the cor-
responding estimate of the hazard ratio is exp(rβ̂). The standard error of the
estimated log-hazard ratio will be r se (β̂), from which confidence intervals for
the true hazard ratio can be derived.
The above argument shows that when a continuous variable X is included
in a Cox regression model, the hazard ratio when the value of X is changed
by r units does not depend on the actual value of X. For example, if X refers
to the age of an individual, the hazard ratio for an individual aged 70, relative
to one aged 65, would be the same as that for an individual aged 20, relative
to one aged 15. This feature is a direct result of fitting X as a linear term in
the Cox regression model. If there is doubt about the assumption of linearity,
this can be checked using the procedure described in Section 3.8.1. Fractional
polynomials in X or a non-linear transformation of X might then be used in
the modelling process.
3.9.2 Models with a factor

When individuals fall into one of m groups, m > 2, which correspond to cat-
egories of an explanatory variable, the groups can be indexed by the levels of
a factor. Under a Cox regression model, the hazard function for an individual
in the jth group, j = 1, 2, . . . , m, is given by
hj (t) = exp(γj )h0 (t),
where γj is the effect due to the jth level of the factor, and h0 (t) is the baseline
hazard function. This model is overparameterised, and so, as in Section 3.2.2,
we take γ1 = 0. The baseline hazard function then corresponds to the hazard
of death at time t for an individual in the first group. The ratio of the hazards
at time t for an individual in the jth group, j > 2, relative to an individual
in the first group, is then exp(γj ). Consequently, the parameter γj is the
logarithm of this relative hazard, that is,
γj = log{hj (t)/h0 (t)}.
A model that contains the terms γj , j = 1, 2, . . . , m, with γ1 = 0, can

be fitted by defining m − 1 indicator variables, X2 , X3 , . . . , Xm , as shown in
Section 3.2.2. Fitting this model leads to estimates γ̂2 , γ̂3 , . . . , γ̂m , and their
standard errors. The estimated logarithm of the relative hazard for an indi-
vidual in group j, relative to an individual in group 1, is then γ̂j .
A 100(1 − α)% confidence interval for the true log-hazard ratio is the
interval from γ̂j − zα/2 se (γ̂j ) to γ̂j + zα/2 se (γ̂j ), where zα/2 is the upper α/2-
point of the standard normal distribution. A corresponding confidence inter-
val for the hazard ratio itself is obtained by exponentiating these confidence
limits.
Data on the survival times of patients with hypernephroma were given in
Table 3.6. In this example, we will only consider the data from those patients
on whom a nephrectomy has been performed, given in columns 4 to 6 of
Table 3.6. The survival times of this set of patients are classified according
to their age group. If the effect due to the jth age group is denoted by αj ,
j = 1, 2, 3, the Cox regression model for the hazard at time t for a patient in
the jth age group is such that
hj (t) = exp(αj )h0 (t).
This model can be fitted by defining two indicator variables, A2 and A3 , where
A2 is unity if the patient is aged between 60 and 70, and A3 is unity if the
patient is more than 70 years of age, as in Example 3.4. This corresponds to
taking α1 = 0.
The value of −2 log L̂ for the null model is 128.901, and when the term αj
is added, the value of this statistic reduces to 122.501. This reduction of 6.400
on 2 d.f. is significant at the 5% level (P = 0.041), and so we conclude that
the hazard function does depend on which age group the patient is in.
The coefficients of the indicator variables A2 and A3 are estimates of α2
and α3 , respectively, and are given in Table 3.14. Since the constraint α1 = 0
has been used, α̂1 = 0.
Table 3.14 Parameter estimates and their

standard errors on fitting a Cox regression
model to the data from Example 3.4.
Parameter Estimate se (Estimate)
α2 −0.065 0.498
α3 1.824 0.682
The hazard ratio for a patient aged 60–70, relative to one aged less than
60, is e−0.065 = 0.94, while that for a patient whose age is greater than 70,
relative to one aged less than 60, is e1.824 = 6.20. These results suggest that
the hazard of death at any given time is greatest for patients who are older
than 70, but that there is little difference in the hazard functions for patients
in the other two age groups.
The standard error of the parameter estimates in Table 3.14 can be used to
obtain a confidence interval for the true hazard ratios. A 95% confidence inter-
val for the log-hazard ratio for a patient whose age is between 60 and 70, rela-
tive to one aged less than 60, is the interval with limits −0.065±(1.96×0.498),
that is, the interval (−1.041, 0.912). The corresponding 95% confidence inter-
val for the hazard ratio itself is (0.35, 2.49). This confidence interval includes
unity, which suggests that the hazard function for an individual whose age is
between 60 and 70 is similar to that for a patient aged less than 60. Similarly,
a 95% confidence interval for the hazard for a patient aged greater than 70,
relative to one aged less than 60, is found to be (1.63, 23.59). This interval
does not include unity, and so an individual whose age is greater than 70 has
a significantly greater hazard of death, at any given time, than patients aged
less than 60.
In some applications, the hazard ratio relative to the level of a factor other
than the first may be required. In these circumstances, the levels of the factor,
and associated indicator variables, could be redefined so that some other level
of the factor corresponds to the required baseline level, and the model re-
fitted. The required estimates can also be found directly from the estimates
obtained when the first level of the original factor is taken as the baseline,
although this is more difficult.
The hazard functions for individuals at levels j and j ′ of the factor are
exp(αj )h0 (t) and exp(αj ′ )h0 (t), respectively, and so the hazard ratio for an
individual at level j, relative to one at level j ′ , is exp(αj −αj ′ ). The log-hazard
ratio is then αj − αj ′ , which is estimated by α̂j − α̂j ′ . To obtain the standard
error of this estimate, we use the result that the variance of the difference
α̂j − α̂j ′ is given by
var (α̂j − α̂j ′ ) = var (α̂j ) + var (α̂j ′ ) − 2 cov (α̂j , α̂j ′ ).
In view of this, an estimate of the covariance between α̂j and α̂j ′ , as well as
estimates of their variance, will be needed to compute the standard error of
(α̂j − α̂j ′ ). The calculations are illustrated in Example 3.11.
Consider again the subset of the data from Example 3.4, corresponding to
those patients who have had a nephrectomy. Suppose that an estimate of the
hazard ratio for an individual aged greater than 70, relative to one aged be-
tween 60 and 70, is required. Using the estimates in Table 3.14, the estimated
log-hazard ratio is α̂3 − α̂2 = 1.824 + 0.065 = 1.889, and so the estimated
hazard ratio is e1.889 = 6.61. This suggests that the hazard of death at any
given time for someone aged greater than 70 is more than six and a half times
that for someone aged between 60 and 70.
The variance of α̂3 − α̂2 is
var (α̂3 ) + var (α̂2 ) − 2 cov (α̂3 , α̂2 ),
and the variance-covariance matrix of the parameter estimates gives the re-
quired variances and covariance. This matrix can be obtained from statistical
software used to fit the Cox regression model, and is found to be
( )
A2 0.2484 0.0832
,
A3 0.0832 0.4649
A2 A3
from which var (α̂2 ) = 0.2484, var (α̂3 ) = 0.4649 and cov (α̂2 , α̂3 ) = 0.0832.
Of course, the variances are simply the squares of the standard errors in
Table 3.14. It then follows that
var (α̂3 − α̂2 ) = 0.4649 + 0.2484 − (2 × 0.0832) = 0.5469,
and so the standard error of α̂2 − α̂3 is 0.740. Consequently a 95% confidence
interval for the log-hazard ratio is (0.440, 3.338) and that for the hazard ratio
itself is (1.55, 8.18).
An easier way of obtaining the estimated value of the hazard ratio for an
individual who is aged greater than 70, relative to one aged between 60 and
70, and the standard error of the estimate, is to redefine the levels of the
factor associated with age group. Suppose that the data are now arranged so
that the first level of the factor corresponds to the age range 60–70, level 2
corresponds to patients aged greater than 70 and level 3 to those aged less
than 60. Choosing indicator variables to be such that the effect due to the
first level of the redefined factor is set equal to zero leads to the variables B2
and B3 defined in Table 3.15.
Table 3.15 Indicator variables

for age group.
Age group B2 B3
< 60 0 1
60–70 0 0
> 70 1 0
The estimated log-hazard ratio is now simply the estimated coefficient of B2 ,

and its standard error can be read directly from standard computer output.
The manner in which the coefficients of indicator variables are interpreted

is crucially dependent upon the coding that has been used for them. This
means that when a Cox regression model is fitted using a statistical package
that enables factors to be fitted directly, it is essential to know how indicator
variables used within the package have been defined.
As a further illustration of this point, suppose that individuals fall into
one of m groups and that the coding used for the m − 1 indicator ∑variables,
m
X2 , X3 , . . . , Xm , is such that the sum of the main effects of A, j=1 αj , is
equal to zero. The values of the indicator variables corresponding to an m-level
factor A, are then as shown in Table 3.16.
With this choice of indicator variables, a Cox regression model that con-
tains this factor can be expressed in the form
hj (t) = exp(α2 x2 + α3 x3 + · · · + αm xm )h0 (t),
where xj is the value of Xj for an individual for whom the factor A is at the
jth level, j = 2, 3, . . . , m. The hazard of death at a given time for an individual
at the first level of the factor is
exp{−(α2 + α3 + · · · + αm )}h0 (t),

where the main effects sum to zero.
Level of A X2 X3 . . . Xm
1 −1 −1 . . . −1
2 1 0 . . . 0
3 0 1 . . . 0
. . . . . . . . . . . . . . . . .
m 0 0 . . . 1
while that for an individual at the jth level of the factor is exp(αj )h0 (t), for
j > 2. The ratio of the hazards for an individual in group j, j > 2, relative to
that of an individual in the first group, is then
exp(αj + α2 + α3 + · · · + αm ).
For example, if m = 4 and j = 3, the hazard ratio is exp(α2 + 2α3 + α4 ), and

the variance of the corresponding estimated log-hazard ratio is
var (α̂2 ) + 4 var (α̂3 ) + var (α̂4 ) + 4 cov (α̂2 , α̂3 )

+ 4 cov (α̂3 , α̂4 ) + 2 cov (α̂2 , α̂4 ).
Each of the terms in this expression can be found from the variance-
covariance matrix of the parameter estimates after fitting a Cox regression
model, and a confidence interval for the hazard ratio obtained. However, this
particular coding of the indicator variables does make it much more compli-
cated to interpret the individual parameter estimates in a fitted model.
3.9.3 Models with combinations of terms

In previous sections, we have only considered the interpretation of parameter
estimates in Cox regression models that contain a single term. More generally,
a fitted model will contain terms corresponding to a number of variates, fac-
tors or combinations of the two. With suitable coding of indicator variables
corresponding to factors in the model, the parameter estimates can again be
interpreted as logarithms of hazard ratios.
When a model contains more than one variable, the parameter estimate
associated with a particular effect is said to be adjusted for the other variables
in the model, and so the estimates are log-hazard ratios, adjusted for the
other terms in the model. The Cox regression model can therefore be used
to estimate hazard ratios, taking account of other variables included in the
model.
When interactions between factors, or mixed terms involving factors and
variates, are fitted, the estimated log-hazard ratios for a particular factor will
differ according to the level of any factor, or the value of any variate with which
it interacts. In this situation, the value of any such factor level or variate will
need to be made clear when the estimated hazard ratios for the factor of
primary interest are presented.
Instead of giving algebraic details on how hazard ratios can be estimated
after fitting models with different combinations of terms, the general approach
will be illustrated in two examples. The first of these involves both factors and
variates, while the second includes an interaction between two factors.

In Example 3.6, the most important prognostic variables in the study on the
survival of prostatic cancer patients were found to be size of tumour (Size)
and the Gleason index of tumour stage (Index). The indicator variable Treat,
which represents the treatment effect, is also included in a Cox regression
model, since the aim of the study is to quantify the treatment effect. The
model for the ith individual can then be expressed in the form
hi (t) = exp{β1 Size i + β2 Index i + β3 Treat i }h0 (t),
for i = 1, 2, . . . , 38. Estimates of the β-coefficients and their standard errors

on fitting this model are given in Table 3.17.
Table 3.17 Estimated coefficients of the ex-

planatory variables on fitting a Cox regres-
sion model to the data from Example 1.4.
Variable β̂ se (β̂)
Size 0.083 0.048
Index 0.710 0.338
Treat −1.113 1.203
The estimated log-hazard ratio for an individual on the active treatment

DES (Treat = 1), relative to an individual on the placebo (Treat = 0), with
the same values of Size and Index as the individual on DES, is β̂3 = −1.113.
Consequently the estimated hazard ratio is e−1.113 = 0.329. The value of this
hazard ratio is unaffected by the actual values of Size and Index. However,
since these two explanatory variables were included in the model, the esti-
mated hazard ratio is adjusted for these variables.
For comparison, if a model that only contains Treat is fitted, the estimated
coefficient of Treat is −1.978. The estimated hazard ratio for an individual on
DES, relative to one on the placebo, unadjusted for Size and Index, is now
e−1.978 = 0.14. This shows that unless proper account is taken of the effect of
size of tumour and index of tumour grade, the extent of the treatment effect
is overestimated.
Now consider the hazard ratio for an individual on a particular treatment
with a given value of the variable Index and a tumour of a given size, relative to
an individual on the same treatment with the same value of Index, but whose
size of tumour is one unit less. This is e0.083 = 1.09. Since this is greater than
unity, we conclude that, other things being equal, the greater the size of the
tumour, the greater that hazard of death at any given time. Similarly, the
hazard ratio for an individual on a given treatment with a given value of Size,
relative to one on the same treatment with the same value of Size, whose value
of Index is one unit less, is e0.710 = 2.03. This again means that the greater
the value of the Gleason index, the greater is the hazard of death at any given
time. In particular, an increase of one unit in the value of Index doubles the
hazard of death.
Consider again the full set of data on survival times following treatment for
hypernephroma, given in Table 3.6. In Example 3.4, the most appropriate Cox
regression model was found to contain terms αj , j = 1, 2, 3, corresponding
to age group, and terms νk , k = 1, 2, corresponding to whether or not a
nephrectomy was performed. For illustrative purposes, in this example we
will consider the model that also contains the interaction between these two
factors, even though it was found not to be significant. Under this model, the
hazard function for an individual in the jth age group and the kth level of
nephrectomy status is
h(t) = exp{αj + νk + (αν)jk }h0 (t), (3.16)
where (αν)jk is the term corresponding to the interaction.

Consider the ratio of the hazard of death at time t for a patient in the
jth age group, j = 1, 2, 3, and the kth level of nephrectomy status, k = 1, 2,
relative to an individual in the first age group who has not had a nephrectomy,
which is
exp{αj + νk + (αν)jk }
.
exp{α1 + ν1 + (αν)11 }
As in Example 3.4, the model in Equation (3.16) is fitted by including the
indicator variables A2 , A3 , and N in the model, together with the products
A2 N and A3 N . The estimated coefficients of these variables are then α̂2 , α̂3 ,
d , and (αν)
ν̂2 , (αν) d , respectively. From the coding of the indicator variables
22 32
d and (αν)
that has been used, the estimates α̂1 , ν̂1 , (αν) d are all zero. The
11 12
estimated hazard ratio for an individual in the jth age group, j = 1, 2, 3, and
the kth level of nephrectomy status, k = 1, 2, relative to one in the first age
group who has not had a nephrectomy, is then just
d }.
exp{α̂j + ν̂k + (αν)jk
The non-zero parameter estimates are α̂2 = 0.005, α̂3 = 0.065, ν̂2 = −1.943,
d d
(αν) 22 = −0.051, and (αν)32 = 2.003, and the estimated hazard ratios are
summarised in Table 3.18.
Inclusion of the combination of factor levels for which the estimated hazard
ratio is 1.000 in tables such as Table 3.18, emphasises that the hazards are
relative to those for individuals in the first age group who have not had a
nephrectomy. This table shows that individuals aged less than or equal to 70,
ESTIMATING THE HAZARD AND SURVIVOR FUNCTIONS 107
Table 3.18 Estimated hazard ratios on fitting a model that

contains an interaction to the data from Example 3.4.
Age group No nephrectomy Nephrectomy
< 60 1.000 0.143
60–70 1.005 0.137
> 70 1.067 1.133
who have had a nephrectomy, have a much reduced hazard of death, compared
to those in the other age group and those who have not had a nephrectomy.
Confidence intervals for the corresponding true hazard ratios can be found
using the method described in Section 3.9.2. As a further illustration, a con-
fidence interval will be obtained for the hazard ratio for individuals who have
had a nephrectomy in the second age group relative to those in the first. The
d , and so the estimated hazard ratio is 0.955.
log-hazard ratio is now α̂2 + (αν) 22
The variance of this estimate is given by
d } + 2 cov {α̂2 , (αν)
var (α̂2 ) + var {(αν) d }.
22 22
From the variance-covariance matrix of the parameter estimates after fitting

d } = 0.942, and the
the model in Equation (3.16), var (α̂2 ) = 0.697, var {(αν) 22
d } = −0.695. Consequently, the variance of
covariance term is cov {α̂2 , (αν)22
the estimated log-hazard ratio is 0.248, and so a 95% confidence interval for
the true log-hazard ratio ranges from −0.532 to 0.441. The corresponding con-
fidence interval for the true hazard ratio is (0.59, 1.55). This interval includes
unity, and so the hazard ratio of 0.955 is not significantly different from unity
at the 5% level. Confidence intervals for the hazard ratios in Table 3.18 can
be found in a similar manner.
3.10 ∗ Estimating the hazard and survivor functions

So far in this chapter, we have only considered the estimation of the β-
parameters in the linear component of a Cox regression model. As we have
seen, this is all that is required in order to draw inferences about the effect
of explanatory variables in the model on the hazard function. Once a suitable
model for a set of survival data has been identified, the hazard function, and
the corresponding survivor function, can be estimated. These estimates can
then be used to summarise the survival experience of individuals in the study.
Suppose that the linear component of a Cox regression model contains p
explanatory variables, X1 , X2 , . . . , Xp , and that the estimated coefficients of
these variables are β̂1 , β̂2 , . . . , β̂p . The estimated hazard function for the ith
of n individuals in the study is then given by
ĥi (t) = exp(β̂ ′ xi )ĥ0 (t), (3.17)
where xi is the vector of values of the explanatory variables for the ith indi-
vidual, i = 1, 2, . . . , n, β̂ is the vector of estimated coefficients, and ĥ0 (t) is the
estimated baseline hazard function. Using this equation, the hazard function
for an individual can be estimated once an estimate of h0 (t) has been found.
The relationship between the hazard, cumulative hazard and survivor func-
tions can then be used to give estimates of the cumulative hazard function
and the survivor function.
An estimate of the baseline hazard function was derived by Kalbfleisch
and Prentice (1973) using an approach based on the method of maximum
likelihood. Suppose that there are r distinct death times which, when arranged
in increasing order, are t(1) < t(2) < · · · < t(r) , and that there are dj deaths
and nj individuals at risk at time t(j) . The estimated baseline hazard function
at time t(j) is then given by
ĥ0 (t(j) ) = 1 − ξˆj , (3.18)
where ξˆj is the solution of the equation

∑ exp(β̂ ′ xl ) ∑
′
= exp(β̂ ′ xl ), (3.19)
exp(β̂ xl )
l∈D(t(j) ) 1− ξˆj l∈R(t(j) )
for j = 1, 2, . . . , r. In Equation (3.19), D(t(j) ) is the set of all dj individuals

who die at the jth ordered death time, t(j) , and as in Section 3.3, R(t(j) )
is the set of all nj individuals at risk at time t(j) . The estimates of the βs,
which form the vector β̂, are those which maximise the likelihood function in
Equation (3.4). The derivation of this estimate of h0 (t) is quite complex, and
so it will not be reproduced here.
In the particular case where there are no tied death times, that is, where
dj = 1 for j = 1, 2, . . . , r, the left-hand side of Equation (3.19) will be a single
term. This equation can then be solved to give
( )exp(−β̂ ′ x(j) )
exp(β̂ ′ x(j) )
ξˆj = 1− ∑ ,
l∈R(t(j) ) exp(β̂ ′ xl )
where x(j) is the vector of explanatory variables for the individual who dies
at time t(j) .
When there are tied observations, that is, when one or more of the dj are
greater than unity, the summation on the left-hand side of Equation (3.19) is
the sum of a series of fractions in which ξˆj occurs in the denominators, raised
to different powers. Equation (3.19) cannot then be solved explicitly, and an
iterative scheme is required.
We now make the assumption that the hazard of death is constant be-
tween adjacent death times. An appropriate estimate of the baseline hazard
function in this interval is then obtained by dividing the estimated hazard in
Equation (3.18) by the time interval, to give the step-function
1 − ξˆj
ĥ0 (t) = , (3.20)
t(j+1) − t(j)
for t(j) 6 t < t(j+1) , j = 1, 2, . . . , r − 1, with ĥ0 (t) = 0 for t < t(1) .
The quantity ξˆj can be regarded as an estimate of the probability that
an individual survives through the interval from t(j) to t(j+1) . The baseline
survivor function can then be estimated by
∏
k
Ŝ0 (t) = ξˆj , (3.21)
j=1
for t(k) 6 t < t(k+1) , k = 1, 2, . . . , r − 1, and so this estimate is also a step-

function. The estimated value of the baseline survivor function is unity for
t < t(1) , and zero for t > t(r) , unless there are censored survival times greater
than t(r) . If this is the case, Ŝ0 (t) can be taken to be Ŝ0 (t(r) ) until the largest
censored time, but the estimated survivor function is undefined beyond that
time.
The baseline cumulative hazard function is, from Equation (1.8), given by
H0 (t) = − log S0 (t), and so an estimate of this function is
∑
k
Ĥ0 (t) = − log Ŝ0 (t) = − log ξˆj , (3.22)
j=1
for t(k) 6 t < t(k+1) , k = 1, 2, . . . , r − 1, with Ĥ0 (t) = 0 for t < t(1) .
The estimates of the baseline hazard, survivor and cumulative hazard func-
tions in Equations (3.20), (3.21) and (3.22) can be used to obtain the corre-
sponding estimates for an individual with a vector of explanatory variables
xi . In particular, from Equation (3.17), the hazard function is estimated by
exp(β̂ ′ xi )ĥ0 (t). Next, integrating both sides of Equation (3.17), we get
∫ t ∫ t
ĥi (u) du = exp(β̂ ′ xi ) ĥ0 (u) du, (3.23)
0 0
so that the estimated cumulative hazard function for the ith individual is
given by
Ĥi (t) = exp(β̂ ′ xi )Ĥ0 (t). (3.24)
On multiplying each side of Equation (3.23) by −1 and exponentiating, and

making use of Equation (1.6), we find that the estimated survivor function for
the ith individual is
{ }exp(β̂ ′ xi )
Ŝi (t) = Ŝ0 (t) , (3.25)
for t(k) 6 t < t(k+1) , k = 1, 2, . . . , r − 1. Note that once the estimated sur-
vivor function, Ŝi (t), has been obtained, an estimate of the cumulative hazard
function is simply − log Ŝi (t).
3.10.1 The special case of no covariates
When there are no covariates, so that we have just a single sample of survival
times, Equation (3.19) becomes
dj
= nj ,
1 − ξˆj
from which
nj − dj
ξˆj = .
nj
Then, the estimated baseline hazard function at time t(j) is 1 − ξˆj , which
is dj /nj . The corresponding estimate of the survivor function from Equa-
∏k
tion (3.21) is j=1 ξˆj , that is,
∏k ( )
nj − d j
,
j=1
nj
which is the Kaplan-Meier estimate of the survivor function given earlier in

Equation (2.4). This shows that the estimate of the survivor function given in
Equation (3.25) generalises the Kaplan-Meier estimate to the case where the
hazard function depends on explanatory variables.
Furthermore, the estimate of the hazard function in Equation (3.20) re-
duces to dj /{nj (t(j+1) − t(j) )}, which is the estimate of the hazard function
given in Equation (2.13) of Chapter 2.
3.10.2 Some approximations to estimates of baseline functions

When there are tied survival times, the estimated baseline hazard can only
be found by using an iterative method to solve Equation (3.19). This iterative
process can be avoided by using an approximation to the summation on the
left-hand side of Equation (3.19).
The term
′
exp(β̂ xl )
ξˆj ,
in the denominator of the left-hand side of Equation (3.19), can be written as
{ }
′
exp eβ̂ xl log ξˆj ,
and taking the first two terms in the expansion of the exponent gives
{ }
′ ′
exp eβ̂ xl log ξˆj ≈ 1 + eβ̂ xl log ξˆj .
Writing 1 − ξ˜j for the estimated baseline hazard at time t(j) , obtained using
′ ′
exp(β̂ xl )
this approximation, and substituting 1 + eβ̂ xl log ξ˜j for ξˆj in Equa-
˜
tion (3.19), we find that ξj is such that
∑ 1 ∑
− = exp(β̂ ′ xl ).
˜
log ξj
l∈D(t(j) ) l∈R(t (j) )
Therefore,
−dj ∑
= exp(β̂ ′ xl ),
log ξ˜j l∈R(t(j) )
since dj is the number of deaths at the jth ordered death time, t(j) , and so
( )
˜ −dj
ξj = exp ∑ ′
. (3.26)
l∈R(t(j) ) exp(β̂ xl )
From Equation (3.21), an estimate of the survivor function, based on the

values of ξ˜j , is given by
( )
∏k
−dj
S̃0 (t) = exp ∑ ′
, (3.27)
j=1 l∈R(t(j) ) exp(β̂ xl )
for t(k) 6 t < t(k+1) , k = 1, 2, . . . , r − 1. From this definition, the estimated

survivor function is not necessarily zero at the longest survival time, when
that time is uncensored, unlike the estimate in Equation (3.21). The estimate
of the baseline cumulative hazard function derived from S̃0 (t) is
∑
k
dj
H̃0 (t) = − log S̃0 (t) = ∑ , (3.28)
j=1 l∈R(t(j) ) exp(β̂ ′ xl )
for t(k) 6 t < t(k+1) , k = 1, 2, . . . , r − 1. This estimate is often referred to as

the Nelson-Aalen estimate or the Breslow estimate of the baseline cumulative
hazard function.
When there are no covariates, the estimated baseline survivor function in
Equation (3.27) becomes
∏ k
exp(−dj /nj ), (3.29)
j=1
since nj is the number of individuals at risk at time t(j) . This is the Nelson-
Aalen estimate of the survivor function given in Equation (2.5) of Chapter 2,
and
∑k the corresponding estimate of the baseline cumulative hazard function is
j=1 dj /nj , as in Section 2.3.3 of Chapter 2.
A further approximation is found from noting that the expression
−dj
∑ ,
l∈R(t(j) ) exp(β̂ ′ xl )
in the exponent of Equation (3.26), will tend to be small, unless there are
large numbers of ties at particular death times. Taking the first two terms of
the expansion of this exponent, and denoting this new approximation to ξj
by ξj∗ gives
dj
ξj∗ = 1 − ∑ ′
.
l∈R(t(j) ) exp(β̂ xl )
Adapting Equation (3.20), the estimated baseline hazard function in the in-
terval from t(j) to t(j+1) is then given by
dj
h∗0 (t) = ∑ , (3.30)
(t(j+1) − t(j) ) l∈R(t(j) ) exp(β̂ ′ xl )
for t(j) 6 t < t(j+1) , j = 1, 2, . . . , r − 1. Using ξj∗ in place of ξˆj in Equa-

tion (3.21), the corresponding estimated baseline survivor function is
( )
∏k
dj
∗
S0 (t) = 1− ∑ ′
,
j=1 l∈R(t(j) ) exp(β̂ xl )
and a further approximate estimate of the baseline cumulative hazard func-

tion is H0∗ (t) = − log S0∗ (t). Notice that the cumulative hazard function in
Equation (3.28) at time t can be expressed in the form
∑
k
H̃0 (t) = (t(j+1) − t(j) )h∗0 (t),
j=1
where h∗0 (t) is given in Equation (3.30). Consequently, differences in succes-

sive values of the estimated baseline cumulative hazard function in Equa-
tion (3.28) provide an approximation to the baseline hazard function, at times
t(1) , t(2) , . . . , t(r) , that can easily be computed.
In the particular case where there are no covariates, the estimates h∗0 (t),
S0 (t) and H0∗ (t) are the same as those given in Section 3.10.1. Equations
∗
similar to Equations (3.24) and (3.25) can be used to estimate the cumulative
hazard and survivor functions for an individual whose vector of explanatory
variables is xi .
In practice, it will often be computationally advantageous to use either
S̃0 (t) or S0∗ (t) in place of Ŝ0 (t). When the number of tied survival times is
small, all three estimates will tend to be very similar. Moreover, since the esti-
mates are generally used as descriptive summaries of the survival data, small
differences between the estimates are unlikely to be of practical importance.
Once an estimate of the survivor function has been obtained, the median
and other percentiles of the survival time distribution can be found from tab-
ular or graphical displays of the function for individuals with particular values
of explanatory variables. The method used is very similar to that described
in Section 2.4, and is illustrated in the following example.
In Example 3.4, a Cox regression model was fitted to the data on the survival
times of patients with hypernephroma. The hazard function was found to
depend on the age group of a patient, and whether or not a nephrectomy had
been performed. The estimated hazard function for the ith patient was found
to be
ĥi (t) = exp{0.013 A2i + 1.342 A3i − 1.412 Ni }ĥ0 (t),
where A2i is unity if the patient is aged between 60 and 70 and zero otherwise,
A3i is unity if the patient is aged over 70 and zero otherwise, and Ni is unity if
the patient has had a nephrectomy and zero otherwise. The estimated baseline
hazard function is therefore the estimated hazard of death at time t, for an
individual whose age is less than 60 and who has not had a nephrectomy.
In Table 3.19, the estimated baseline hazard function, ĥ0 (t), cumulative
hazard function, Ĥ0 (t), and survivor function, Ŝ0 (t), obtained using Equa-
tions (3.18), (3.22) and (3.21), respectively, are tabulated.
Table 3.19 Estimates of the baseline hazard, survivor and

cumulative hazard functions for the data from Example 3.4.
Time ĥ0 (t) Ŝ0 (t) Ĥ0 (t)
0 0.000 1.000 0.000
5 0.050 0.950 0.051
6 0.104 0.852 0.161
8 0.113 0.755 0.281
9 0.237 0.576 0.552
10 0.073 0.534 0.628
12 0.090 0.486 0.722
14 0.108 0.433 0.836
15 0.116 0.383 0.960
17 0.132 0.333 1.101
18 0.285 0.238 1.436
21 0.185 0.194 1.641
26 0.382 0.120 2.123
35 0.232 0.092 2.387
36 0.443 0.051 2.972
38 0.279 0.037 3.299
48 0.299 0.026 3.655
52 0.560 0.011 4.476
56 0.382 0.007 4.958
68 0.421 0.004 5.504
72 0.467 0.002 6.134
84 0.599 0.001 7.045
108 0.805 0.000 8.692
115 – 0.000 –
From this table, we see that the general trend is for the estimated baseline
hazard function to increase with time. From the manner in which the esti-
mated baseline hazard function has been computed, the estimates only apply
at the death times of the patients in the study. However, if the assumption
of a constant hazard in each time interval is made, by dividing the estimated
hazard by the corresponding time interval, the risk of death per unit time
can be found. This leads to the estimate in Equation (3.20). A graph of this
hazard function is shown in Figure 3.4.
0.20
0.15
0.10
0.05
0.00
0 20 40 60 80 100 120
Survival time
Figure 3.4 Estimated baseline hazard function, per unit time, assuming constant
hazard between adjacent death times.
This graph shows that the risk of death per unit time is roughly con-
stant over the duration of the study. Table 3.19 also shows that the values of
ĥ0 (t) are very similar to differences in the values of Ĥ0 (t) between successive
observations, as would be expected.
We now consider the estimation of the median survival time, defined as the
smallest observed survival time for which the estimated survivor function is
less than 0.5. From Table 3.19, the estimated median survival time for patients
aged less than 60 who have not had a nephrectomy is 12 months.
By raising the estimate of the baseline survivor function to a suitable
power, the estimated survivor functions for patients in other age groups,
and for patients who have had a nephrectomy, can be obtained using Equa-
tion (3.25). Thus, the estimated survivor function for the ith individual is
given by
{ }exp{0.013A2i +1.342A3i −1.412Ni }
Ŝi (t) = Ŝ0 (t) .
For an individual aged less than 60 who has had a nephrectomy, A2 = 0,

A3 = 0, and N = 1, so that the estimated survivor function for this individual
becomes { }exp{−1.412}
Ŝ0 (t) .
This function is plotted in Figure 3.5, together with the estimated baseline
survivor function, which is for an individual in the same age group but who
has not had a nephrectomy.
1.0
0.8
0.6
0.4
0.2
0.0
0 20 40 60 80 100 120
Survival time
Figure 3.5 Estimated survivor functions for patients aged less than 60, with (—) and
without (·······) a nephrectomy.
This figure shows that the probability of surviving beyond any given time is
greater for those who have had a nephrectomy, confirming that a nephrectomy
improves the prognosis for patients with hypernephroma.
Note that because of the assumption of proportional hazards, the two esti-
mated survivor functions in Figure 3.5 cannot cross. Moreover, the estimated
survivor function for those who have had a nephrectomy lies above that of
those on whom a nephrectomy has not been performed. This is a direct con-
sequence of the estimated hazard ratio for those who have had the operation,
relative to those who have not, being less than unity.
An estimate of the median survival time for this type of patient can be
obtained from the tabulated values of the estimated survivor function, or
from the graph in Figure 3.5. We find that the estimated median survival
time for a patient aged less than 60 who has had a nephrectomy is 36 months.
Other percentiles of the distribution of survival times can be estimated using
a similar approach.
In a similar manner, the survivor functions for patients in the different
age groups can be compared, either for those who have had or not had a
nephrectomy. For example, for patients who have had a nephrectomy, the
estimated survivor functions for patients in the three age groups are respec-
tively {Ŝ0 (t)}exp{−1.412} , {Ŝ0 (t)}exp{−1.412+0.013} and {Ŝ0 (t)}exp{−1.412+1.342} .
These estimated survivor functions are shown in Figure 3.6, which clearly
shows that patients aged over 70 have a poorer prognosis than those in the
other two age groups.
1.0
0.8
0.6
0.4
0.2
0.0
0 20 40 60 80 100 120
Survival time
Figure 3.6 Estimated survivor functions for patients aged less than 60 (—), between
60 and 70 (·······) and greater than 70 (- - -), who have had a nephrectomy.
3.11 Risk adjusted survivor function

Once a Cox regression model has been fitted, the estimated survivor function
can be obtained for each individual. It might then be of interest to see how the
fitted survivor functions compare with the unadjusted Kaplan-Meier estimate
of the survivor function. The impact of the risk adjustment can then be deter-
mined. To do this, a risk adjusted survivor function is obtained by averaging
the individual estimated values of the survivor function at each event time in
the data set.
Formally, suppose that a Cox regression model is fitted to survival data
from n individuals, in which the hazard of death for the ith individual, i =
1, 2, . . . , n, at time t is
hi (t) = exp(β ′ xi )h0 (t),
where β ′ xi = β1 x1i + β2 x2i + · · · + βp xpi , x1i , x2i , . . . , xpi are the values of p
explanatory variables measured on each individual, and h0 (t) is the baseline
hazard function. The corresponding survivor function for the ith individual is
′
exp(β xi )
Si (t) = {S0 (t)} ,
RISK ADJUSTED SURVIVOR FUNCTION 117
and the model fitting process leads to estimates of the p β-parameters,
β̂1 , β̂2 , . . . , β̂p , and the baseline survivor function, Ŝ0 (t). The average survivor
function at a given time t is then
1∑
n
Ŝ(t) = Ŝi (t), (3.31)
n i=1
where
{ }exp(β̂ ′ xi )
Ŝi (t) = Ŝ0 (t)
is the estimated survivor function for the ith individual. Risk adjusted esti-
mates of survival rates, the median and other percentiles of the survival time
distribution can then be obtained from Ŝ(t).

In Example 3.5, a Cox regression model that contained the explanatory vari-
ables Hb and Bun was found to be appropriate in modelling data on the
survival times of patients suffering from multiple myeloma, introduced in Ex-
ample 1.3. The estimated survivor function for the ith patient, i = 1, 2, . . . , 48,
is { } exp(η̂i )
Ŝi (t) = Ŝ0 (t) ,
where the risk score, η̂i , is given by η̂i = −0.134Hb i +0.019Bun i . The estimated
survivor function is then obtained at each of the event times in the data set,
for each of the 48 patients. Averaging the estimates across the 48 patients, for
each event time, leads to the risk adjusted survivor function. This is plotted
in Figure 3.7, together with the unadjusted Kaplan-Meier estimate of the
survivor function. The unadjusted and risk adjusted estimates of the survivor
function are very close, so that in this example, the risk adjustment process
makes very little difference to estimates of survival rates or the median survival
time.
3.11.1 Risk adjusted survivor function for groups of individuals

In many circumstances, it is of interest to estimate the survivor functions for
certain groups of individuals after adjustment has been made for differences
between these groups in terms of the values of measured explanatory variables.
For example, consider a study on disease-free survival following treatment with
one or other of two treatment regimens. If such a study were carried out as a
randomised controlled trial, it is likely that the values of measured explanatory
variables would be balanced between the treatment groups. It may then be
sufficient to summarise the data using the unadjusted Kaplan-Meier estimate
of the survivor function for each group, supplemented by unadjusted and
adjusted hazard ratios for the treatment effect. However, suppose that the
data were obtained in an observational study, where values of the explanatory
1.0
0.8
0.6
0.4
0.2
0.0
0 20 40 60 80 100
Survival time
Figure 3.7 Unadjusted (·······) and risk adjusted (—) estimated survivor functions for
the data on the survival of multiple myeloma patients.
variables were not evenly distributed across the two treatment groups. In
this situation, the unadjusted estimates of the two survivor functions may be
misleading. A risk adjustment process will then be needed to take account of
any imbalance between the characteristics of individuals in the two groups,
before conclusions can be drawn about the treatment effect.
If differences between the treatment groups can be assumed to be inde-
pendent of time, the group effect can be added to the survival model, and the
risk adjusted survivor function for the individuals in each treatment group
can be calculated using Equation (3.31). However, in many applications, such
as when comparing survival rates between institutions, to be considered in
Chapter 11, this assumption cannot be made. The Cox regression model is
then extended to have a different baseline hazard function for each group.
Suppose that in a study to compare the survival rates of individuals in g
groups, a Cox regression model containing relevant explanatory variables has
been fitted, excluding the group effect. The model for the hazard of death for
the ith individual, i = 1, 2, . . . , nj , in the jth group, j = 1, 2, . . . , g, at time t
is then
hij (t) = exp(β ′ xij )h0j (t),
where β ′ xij = β1 x1ij + β2 x2ij + · · · + βp xpij , x1ij , x2ij , . . . , xpij are the values
of p explanatory variables measured on each individual, and h0j (t) is the
baseline hazard function for the jth group. In this model, the coefficients of
the p explanatory variables, β1 , β2 , . . . , βp , are constant over the g groups, but
there is a different baseline hazard function for each group. This is a stratified
RISK ADJUSTED SURVIVOR FUNCTION 119
Cox regression model in which the g groups define the separate strata. These
models are considered in greater detail in Chapter 11.
On fitting the stratified model, the corresponding estimated survivor func-
tion for the ith patient in the jth group is
{ }exp(β̂ ′ xij )
Ŝij (t) = Ŝ0j (t) ,
where Ŝ0j (t) is the estimated baseline survivor function for individuals in
the jth group. If the groups can be assumed to act proportionately on the
hazard function, a common baseline hazard would be fitted and a group effect
included in the model. Then, h0j (t) is replaced by exp(gj )h0 (t), where gj is
the effect of the jth group. In general, it is less restrictive to allow the group
effects to vary over time by using a stratified model.
The risk adjusted survivor function for each group can be found from the
stratified model by averaging the values of the estimated survivor functions
at each of the event times, across the individuals in each group. The average
risk adjusted survivor function at time t is then
1 ∑
nj
Ŝj (t) = Ŝij (t),
nj i=1
for individuals in the jth group.

An alternative approach is to average the values of each explanatory vari-
able for individuals in each group, and to then use these values to estimate the
group-specific survivor functions. Writing x̄j for the vector of average values
of the p variables across the individuals in the jth group, the corresponding
estimated survivor function for an ‘average individual’ in group j is
{ }exp(β̂ ′ x̄j )
S̃j (t) = Ŝ0j (t) .
Although this approach is widely used, it can be criticised on a number of

grounds. Averaging the values of explanatory variables is not appropriate for
categorical variables. For example, suppose that the two levels of a variable
associated with the presence or absence of diabetes are coded as 1 and 0,
respectively, and that 10% of the individuals on whom data are available are
diabetic. Setting the average of the indicator variable to be 0.1 does not lead
to a meaningful interpretation of the survivor function at this value of diabetic
status. Also, even with continuous explanatory variables, the set of average
values of the explanatory variables within a group may not correspond to
a realistic combination of values, and the survivor function for an average
individual may be very different from the patterns of the individual survivor
functions.
Consider the data from Example 1.4 on the survival times of 38 prostatic
cancer patients in two treatment groups. In Example 3.6, the explanatory
variables Size and Index were found to affect the hazard of death, and so a
stratified Cox regression model that contains these two variables is fitted. The
estimated survivor function for the ith patient in the jth treatment group, for
j = 1 (placebo) and 2 (DES), is
{ }exp(η̂ij )
Ŝij (t) = Ŝ0j (t) ,
where
η̂ij = 0.0673 Size ij + 0.6532 Index ij ,
and Ŝ0j (t) is the estimated baseline survivor function for the jth treatment
group. Averaging the estimated survivor functions at each event time over
the patients in each group gives the risk adjusted survivor functions shown
in Figure 3.8. Also shown in this figure are the unadjusted Kaplan-Meier
estimates of the survivor functions for each group.
1.0 1.0
0.9 0.9
0.8 0.8
0.7 0.7
0.6 0.6
0.5 0.5
0 20 40 60 80 0 20 40 60 80
Survival time Survival time
Figure 3.8 (i) Unadjusted and (ii) risk adjusted survivor functions for prostatic can-
cer patients on DES (·······) and placebo (—).
This figure shows how the risk adjustment process has diminished the
treatment difference. Of course, this is also seen by comparing the unadjusted
hazard ratio for a patient on DES relative to placebo (0.14) with the corre-
sponding value adjusted for tumour size and Gleason index (0.33), although
this latter analysis assumes proportional hazards for the two treatments.
3.12 ∗ Explained variation in the Cox regression model

In linear regression analysis, the proportion of variation in the response vari-
able that is accounted for by the explanatory variables is widely used to sum-
marise the explanatory power of a model. This statistic is usually denoted
EXPLAINED VARIATION IN THE COX REGRESSION MODEL 121
by R2 . For a general linear model with p explanatory variables, where the
expectation of the ith value of a response variable Y , i = 1, 2, . . . , n, is given
by
E (Yi ) = β0 + β1 x1i + · · · + βp xpi ,
the R2 statistic is defined as
Model SS Model SS
R2 = = .
Total SS Model SS + Residual SS
In this equation, the variation in the response variable that∑is explained by
the model is summarised in the Model sum of squares (SS), i (ŷi − ȳ)2 , and
expressed as a∑proportion of the total variation
∑ in the data, represented by
the Total SS, i (yi − ȳ)2 , where ȳ = n−1 i yi and ŷi is the model-based
estimate of E (Yi ), the ith fitted value, given by
ŷi = β̂0 + β̂1 x1i + · · · + β̂p xpi .
The
∑ Total 2SS can be partitioned into the Model SS and the Residual SS,
i (yi − ŷi ) , which represents the unexplained variation. The larger the value
of R2 , the greater the proportion of variation in the response variable that is
accounted for by the model.
The R2 statistic for the general linear model can also be expressed in the
form
V̂M
R2 = , (3.32)
V̂M + σ̂ 2
where V̂M = Model SS/(n − 1) is an estimate of the variation in the data
explained by the model, and σ̂ 2 is an estimate of the residual variation. Note
that in this formulation, σ̂ 2 = Residual SS/(n − 1), rather than the usual
unbiased estimator with divisor n − p − 1. The quantity V̂M in Equation (3.32)
can be expressed in matrix form as β̂ ′ S β̂, where β̂ is the vector of estimated
coefficients for the p explanatory variables in the fitted regression model, and
S is the variance-covariance matrix of the explanatory variables. This matrix
is formed ∑ from the sample variance of each of the p explanatory variables,
(n − 1)−1∑ i (xij − x̄j )2 , on its diagonals, and the sample covariance terms,
(n − 1)−1 i (xij − x̄j )(xij ′ − x̄j ′ ), as the off-diagonal terms, for i = 1, 2, . . . , n
and j, j ′ = 1, 2, . . . , p for j ̸= j ′ , and where x̄j is the sample mean of the jth
variable.
In the analysis of survival data, a large number of different measures of the
proportion of variation in the data that is explained by a fitted Cox regression
model have been proposed. However, published reviews of these measures,
based on extensive simulation studies, suggest that three particular statistics
have desirable properties. All of them take values between 0 and 1, are largely
independent of the degree of censoring, are not affected by the scale of the
survival data, and increase in value as explanatory variables are added to the
model. These are described in the next section.
3.12.1 Measures of explained variation
Consider a Cox regression model for the hazard of death at time t in the
ith individual, hi (t) = exp(β ′ xi )h0 (t), where xi is the vector of values of
p explanatory variables, β is the vector of their unknown coefficients, and
h0 (t) is the baseline hazard function. One of the earliest suggestions for an
R2 type statistic, due to Kent and O’Quigley (1988), is similar in form to the
R2 statistic for linear regression analysis in Equation (3.32). This statistic is
defined by
2 V̂P
RP = ,
V̂P + π 2 /6
where V̂P = β̂ ′ S β̂ is an estimate of the variation in the risk score, β̂ ′ xi ,

between the n individuals, β̂ is the vector of parameter estimates in the fit-
ted Cox regression model, and S is the variance-covariance matrix of the
explanatory variables. The reason for including the term π 2 /6 in place of σ̂ 2
in Equation (3.32) will be explained later in Section 5.8 of Chapter 5.
2
The statistic RD , proposed by Royston and Sauerbrei (2004), is also based
on an estimate of the variation in the risk score between individuals. To obtain
this statistic, the values of the risk score for the fitted Cox regression model,
β̂ ′ xi , are first ordered from smallest value to largest, so that the ith of the n
ordered values is η̂(i) = β̂ ′ x(i) , where η̂(1) < η̂(2) < · · · < η̂(n) . The ith of these
ordered values is then associated with a quantity z(i) that is an approximation
to the expected value of the ith order statistic of a standard normal distri-
bution in a sample of n observations. This is usually referred to as a normal
score, and the most widely used score is z(i) = Φ−1 {(i − 3/8)/(n + 1/4)},
where Φ−1 (·) is the inverse standard normal distribution function. The values
η̂(i) are then regressed on the z(i) , and the resulting estimate of the coeffi-
cient of √ z(i) in the linear regression model, labelled D0 , is then scaled to give
2
D = D0 (π/8). The RD measure of explained variation for a Cox regression
model is then
2 D2
RD = 2 .
D + π 2 /6
2
This has a similar form to the RP statistic, and indeed, D2 can also be re-
garded as an estimate of the variation between the values of the risk score.
Another statistic from Kent and O’Quigley (1988) is based on the ‘distance’
between the model of interest and the model that has no explanatory variables.
This latter model is known as the null model, and is such that the hazard
function for each individual is simply the baseline hazard, h0 (t). This statistic
is
RW2
= 1 − exp(−W̃ ),
where W̃ , the distance measure, is an estimate of the expected value of the
likelihood ratio statistic, defined as W = −2 E {log L(0) − log L(β̂)}, for com-
paring the maximised log-likelihood under the fitted model, log L(β̂) with that
EXPLAINED VARIATION IN THE COX REGRESSION MODEL 123
of the null model, log L(0). This is given by
[ { }]
∑
n
−1
W̃ = 2 (1 − ω̃)Ψ(1) + log Γ(ω̃) + log n exp(−ω̃zi ) ,
i=1
where zi = β̂ ′ (xi − x̄), β̂ is the vector of parameter estimates in the fitted

Cox regression model, x̄ ∫ ∞is the vector of mean values of the p explanatory
variables, and Γ(ω̃) = 0 uω̃−1 e−u du is a gamma function. Also, ω̃ is the
value of ω that satisfies the non-linear equation
∑
n
exp(−ωzi )
Ψ(1) − Ψ(ω) + ∑n zi = 0,
i=1 l=1 exp(−ωzl )
where Ψ(ω) is the digamma function, defined by the series expansion

∞
∑ ω−1
Ψ(ω) = −λ + ,
j=0
(1 + j)(ω + j)
so that Ψ(1) = −λ, and λ = 0.577216 is Euler’s constant. This is a complicated

statistic to calculate, but it has been included here as it performs well, and
can now be obtained using several software packages for survival analysis.
3.12.2 Measures of predictive ability

In addition to measures of explained variation, statistics that summarise the
agreement or concordance between the ranks of observed and predicted sur-
vival times are useful in assessing the predictive ability of a model. These
statistics summarise the potential of a fitted model to discriminate between
individuals, by separating those with longer survival times from those with
shorter times. As for measures of explained variation, these statistics take
values between 0 and 1, corresponding to perfect discordance and perfect con-
cordance. Values around 0.5 are obtained when a model has no predictive
ability, and models with a reasonable degree of predictive ability would lead
to a value greater than 0.7.
A particular measure of concordance is the c-statistic described by Harrell,
Lee and Mark (1996). This statistic is an estimate of the probability that, for
any two individuals, the one with the shortest survival time is the one with
the greatest hazard of death. To calculate this statistic, consider all possible
pairs of survival times, where either both members of the pair have died, or
where one member of the pair dies before the censored survival time of the
other. Pairs in which both individuals have censored survival times, or where
the survival time of one exceeds the censored survival time of the other, are
not included. If in a pair where both individuals have died, the model-based
predicted survival time is greater for the individual who lived longer, the two
individuals are said to be concordant. In a proportional hazards model, an
individual in a pair who is predicted to have the greatest survival time will be
the one with the lower hazard of death at a given time, the higher estimated
survivor function at a given time, or the lower value of the risk score. For pairs
where just one individual dies, and one individual has a time that is censored
after the survival time of the other member of the pair, the individual with the
censored time has survived longer than the other, and so it can be determined
whether the two members of such a pair are concordant. The c-statistic is
obtained by dividing the number of concordant pairs by the number of all
possible pairs being considered.
Since pairs where both individuals are censored, or where the censored
survival time of one occurs before the death time of the other, the c-statistic
is affected by the pattern of censoring. This considerable disadvantage is over-
come in a statistic proposed by Gönen and Heller (2005). Their measure of
concordance is an estimate of the probability that, for any two individuals,
the survival time of one exceeds the survival time of the other, conditional on
the individual with the longer survival time having the lower risk score.
To obtain this statistic, let η̂i = β̂ ′ xi be the risk score for the ith individual.
A pair of survival times, (Ti , Tj ), say, is then concordant if the individual with
the lower risk score has the longer survival time. The probability of concor-
dance is then K = P(Ti > Tj | ηi 6 ηj ), and an estimate of this concordance
probability, K̂, is the Gönen and Heller measure of the predictive value of a
model. This estimate is given by
2 ∑ ∑ { I{(η̂j − η̂i ) < 0} I{(η̂i − η̂j ) < 0} }

K̂ = + ,
n(n − 1) i<j 1 + exp(η̂j − η̂i ) 1 + exp(η̂i − η̂j )
where the indicator function I(·) is such that I{X < 0} = 1 if X < 0 and zero
otherwise. In addition, the standard error of K̂ can be obtained, so that the
precision of K̂ can be assessed, although details will not be given here.
3.12.3 Model validation

The methods described in Sections 3.12.1 and 3.12.2 to determine the variation
in a set of survival data that is explained by a fitted Cox regression model, and
the predictive ability of the model, can lead to an over-optimistic assessment
of the strength of a model if the same data has been used both to fit the
model and to validate it. For this reason, measures of explained variation and
predictive ability are best determined from fitting the model to data that has
not been used in the modelling process. If an independent data set, such as
data from another region or country is available, then these data should be
used for the validation process. This is referred to as external validation. This
is often not possible and a data-splitting procedure, based on partitioning
the data at random into a modelling and a validation data set, may be used.
The model is then determined from the modelling data set and validated on
the remaining data, leading to an internal validation process. Depending on
PROPORTIONAL HAZARDS AND THE LOG-RANK TEST 125
the size of the data set, up to 50% of the data may be set aside for validation
purposes, although for smaller sets of data, this proportion may be somewhere
between 20% and 40%. In any case, this validation data set should contain a
sufficient number of events, and a general guideline is that at least 100 events
are needed.
In this example, the three different R2 measures will be compared for the data
on the survival times of multiple myeloma patients, presented in Example 1.3.
In Example 3.5, the most satisfactory Cox regression model for the hazard
of death contained the explanatory variables Hb (serum haemoglobin) and
Bun (blood urea nitrogen). The values of the different statistics are shown in
Table 3.20 for both the selected model that contains Hb and Bun, and the full
model that contains Age, Sex, Bun, Ca, Hb, Pcells and Protein.
Table 3.20 Measures of explained variation on fitting

the selected and full models.
Measure R2 (selected model) R2 (full model)
2
RP 0.27 0.31
2
RD 0.25 0.29
2
RW 0.30 0.34
The three different measures of explained variation have broadly similar

values for both models, and the difference between their values for the selected
2
and full models is the same for each statistic. On the basis of the RW statistic,
we conclude that 30% of the variation in the survival times is explained by
the levels of serum haemoglobin and blood urea nitrogen.
Turning to measures of predictive ability, the value of Harrell’s c-statistic
is 0.66 for the selected model and 0.70 for the full model. Gönen and Heller’s
measure of concordance is 0.67 (se = 0.044) for the selected model and 0.68
(se = 0.044) for the full model. In this example, the two measure of the dis-
criminatory power of either model are very similar because there is relatively
little censoring, and so most of the pairs of survival times can be used in the
computation of the c-statistic. There is also very little difference in the value of
either statistic for the selected and full model, confirming that the predictive
ability of the selected model is comparable to that of the full model. However
a value of 0.67 for the Gönen and Heller statistic indicates that the model
is not particularly good at discriminating between the survival times of the
individuals in this study.
3.13 ∗ Proportional hazards modelling and the log-rank test

The Cox regression model can be used to test the null hypothesis that there
is no difference between the hazard functions for two groups of survival times,
as illustrated in Example 3.3. This modelling approach therefore provides an
alternative to the log-rank test in this situation. However, there is a close
connection between the two procedures, which is explored in greater detail in
this section.
Following the notation used in Section 2.6.2, and summarised in Table 2.7,
the two groups will be labelled Group I and Group II, respectively. The num-
bers of individuals in the two groups who die at the jth ordered death time,
t(j) , j = 1, 2, . . . , r, will be denoted by d1j and d2j , respectively. Similarly,
the numbers of individuals at risk in the two groups at time t(j) , that is, the
numbers who are alive and uncensored just prior to this time, will be denoted
n1j and n2j , respectively.
Now let X be an indicator variable that is unity when an individual is in
Group I and zero when an individual is in Group II. The Cox regression model
for the ith individual can be written as
where xi is the value of X for the ith individual, i = 1, 2, . . . , n. When there

are no tied observations, that is, when dj = d1j + d2j = 1, this model can
be fitted by finding that value β̂ which maximises the likelihood function in
Equation (3.4). Denoting the value of X for the individual who dies at t(j) by
x(j) , the likelihood function is given by
∏
r
exp(βx(j) )
L(β) = ∑nj , (3.33)
j=1 l=1 exp(βxl )
since there are nj = n1j + n2j individuals in the risk set, R(t(j) ), at time t(j) ,
and the corresponding log-likelihood function is
{ nj }
∑r ∑r ∑
log L(β) = βx(j) − log exp(βxl ) .
j=1 j=1 l=1
Since x(j) is zero for individuals in Group II, the first summation in this
expression
∑r is over the death times in Group I, and so is simply d1 β, where
d1 = j=1 d1j is the total number of deaths in Group I. Also,
∑
nj
exp(βxl ) = n1j eβ + n2j ,
l=1
and so
∑
r
{ }
log L(β) = d1 β − log n1j eβ + n2j . (3.34)
j=1
The maximum likelihood estimate of β can be found by maximising this

expression with respect to β, using a non-linear optimisation routine. Then,
the null hypotheses that β = 0 can be tested by comparing ∑rthe value of
−2 log L(β̂) with −2 log L(0). This latter quantity is simply 2 j=1 log nj .
PROPORTIONAL HAZARDS AND THE LOG-RANK TEST 127
Computation of β̂ can be avoided by using a score test of the null hypothesis
that β = 0. This test procedure, which is outlined in Appendix A, is based on
the test statistic
u2 (0)
,
i(0)
where
∂ log L(β)
u(β) =
∂β
is the efficient score, and
∂ 2 log L(β)
i(β) = −
∂β 2
is Fisher’s (observed) information function. Under the null hypothesis that
β = 0, u2 (0)/i(0) has a chi-squared distribution on one degree of freedom.
Now, from Equation (3.34),
r ( )
∂ log L(β) ∑ n1j eβ
= d1j − ,
∂β j=1
n1j eβ + n2j
and
∂ 2 log L(β) ∑ (n1j eβ + n2j )n1j eβ − (n1j eβ )2

r
= −
∂β 2 j=1
(n1j eβ + n2j )2
∑
r
n1j n2j eβ
=− .
j=1
(n1j eβ + n2j )2
The efficient score and information function, evaluated at β = 0, are therefore

given by
∑r ( )
n1j
u(0) = d1j − ,
j=1
n1j + n2j
and
∑
r
n1j n2j
i(0) = .
j=1
(n1j + n2j )2
These are simply the expressions for UL and VL given in Equations (2.20) and
(2.22) of Chapter 2, for the special case where there are no ties, that is, where
dj = 1 for j = 1, 2, . . . , r.
When there are tied observations, the likelihood function in Equa-
tion (3.33) has to be replaced by one that allows for ties. In particular, if the
likelihood function in Equation (3.12) is used, the efficient score and informa-
tion function are exactly those given in Equations (2.20) and (2.22). Hence,
when there are tied survival times, the log-rank test corresponds to using the
score test for the discrete proportional hazards model due to Cox (1972). In
practice, the P -value that results from this score test will not usually differ
much from that obtained from comparing the values of the statistic −2 log L̂
for the models with and without a term corresponding to the treatment effect.
This was noted in the discussion of Example 3.3. Of course, one advantage of
using the Cox regression model in the analysis of such data is that it leads
directly to an estimate of the hazard ratio.
Comprehensive introductions to statistical modelling in the context of linear

regression analysis are given by Draper and Smith (1998) and Montgomery,
Peck and Vining (2012). McCullagh and Nelder (1989) include a chapter on
models for survival data in their encyclopaedic survey of generalised linear
modelling.
The proportional hazards model for survival data, in which the baseline
hazard function remains unspecified, was proposed by Cox (1972). This paper
introduced the notion of partial likelihood, which was subsequently considered
in greater detail by Cox (1975). See also the contributions to the discussion
of Cox (1972) by Kalbfleisch and Prentice (1972) and Breslow (1972). A de-
tailed review of the model, and extensions of it, is contained in Therneau and
Grambsch (2000).
Introductions to the Cox regression model intended for medical researchers
have been given by Christensen (1987), Elashoff (1983) and Tibshirani (1982).
More recent accounts are given in the textbooks referenced in Section 1.5
of Chapter 1. In particular, Hosmer, Lemeshow and May (2008) include a
careful discussion on model development and the interpretation of model-
based parameter estimates.
A detailed treatment of ties in survival data is given in Kalbfleisch and
Prentice (2002) and Lawless (2002); see also Breslow (1972) and Peto (1972).
The estimate of the baseline survivor function, denoted by Ŝ0 (t) in Sec-
tion 3.10, was introduced by Kalbfleisch and Prentice (1973) and is also de-
scribed in Kalbfleisch and Prentice (2002). The estimate S0∗ (t) was presented
by Breslow (1972, 1974), although it was derived using a different argument
from that used in Section 3.10.2.
Model formulation and strategies for model selection are discussed in books
on linear regression analysis, and also in Chapter 5 of Chatfield (1995), Chap-
ter 4 of Cox and Snell (1981), and Appendix 2 of Cox and Snell (1989). Miller
(2002) describes a wide range of procedures for identifying suitable subsets of
variables to use in linear regression modelling. What has come to be known as
Akaike’s information criterion was introduced by Akaike (1974). It is widely
used in times series analysis and described in books on this subject, such
as Chatfield (2004). The use of the BIC statistic in modelling survival data
was described by Volinsky and Raftery (2000). The hierarchic principle is
fully discussed by Nelder (1977), and in Chapter 3 of McCullagh and Nelder
FURTHER READING 129
(1989). Harrell (2001) addresses many practical issues in model building and
illustrates the process using two extensive case studies involving survival data.
The lasso was first proposed by Tibshirani (1996), and Tibshirani (1997)
showed how the method can be used for variable selection in the Cox
model. An efficient algorithm for computing the lasso estimates was described
by Goeman (2010). This algorithm is now implemented in the R package
penalized for fitting penalised Cox regression models; see Goeman, Meijer
and Chaturvedi (2013). The piecewise linearity property of the lasso was es-
tablished by Efron et al. (2004), and Yuan and Lin (2006) introduced the
group lasso to handle categorical variables.
The use of fractional polynomials in regression modelling was described by
Royston and Altman (1994); see also Sauerbrei and Royston (1999). The Box-
Tidwell transformation, referred to in the context of fractional polynomials,
was introduced by Box and Tidwell (1962).
Methods for the calculation of the adjusted survivor function have been
reviewed by Nieto and Coresh (1996) and in his comments on this paper,
Zahl (1997) outlines some related approaches. The method is also described
in Zhang et al. (2007), who present a SAS macro for its implementation.
Neuberger et al. (1986) used the method based on averaging the values of
explanatory variables to summarise observational data from a liver transplant
programme, which was subsequently referred to in Thomsen et al. (1991).
A considerable number of measures of explained variation have been pro-
posed, and there have been several reviews that compare and contrast their
properties. The most extensive reviews are those of Schemper and Stare
(1996), Hielscher et al. (2010) and the two consecutive papers of Choodari-
Oskooei, Royston and Parmar (2012). The R2 measures described in this
chapter were introduced by Kent and O’Quigley (1988) and Royston and
Sauerbrei (2004). The links between these and other measures of explained
variation are also described in Royston (2006). Heinzl (2000) describes SAS
2
macros for the calculation of Kent and O’Quigley’s RW statistic. Harrell’s
c-statistic is described in Harrell, Lee and Mark (1996), and a preferred mea-
sure of concordance was introduced by Gönen and Heller (2005). One further
measure of explained variation, due to Stare, Perme and Henderson (2011),
can be used in proportional hazards modelling, and has the advantage of be-
ing applicable to some of the other models presented in later chapters. This
statistic is based on explaining variability in the ranks of observations, and R
software for calculating the statistic is available.
One issue that has not been discussed in this chapter is how to handle
missing values of explanatory variables. A method of multiple imputation
based on fully conditional specification, also known as multiple imputation by
chained equations, is a widely used technique for dealing with this common
problem. The multiple imputation process involves generating a number of
sets of imputed values for each explanatory variable with missing values. The
parameter estimates on fitting a given model to each imputed data set are
then combined to give overall estimates, with standard errors that allow for
uncertainty in the imputed values. Advances in software development mean
that this technique can now be used routinely in conjunction with commonly
available software packages, including SAS, Stata and R. The tutorial paper
of White, Royston and Wood (2011) describes the method, its limitations and
possible dangers, and also includes a comprehensive reference list.
Chapter 4
Model checking in the Cox regression

model
After a model has been fitted to an observed set of survival data, the adequacy
of the model needs to be assessed. Indeed, the use of diagnostic procedures
for model checking is an essential part of the modelling process.
In some situations, careful inspection of an observed set of data may lead to
the identification of certain features, such as individuals with unusually large
or small survival times. However, unless there are only one or two explanatory
variables, a visual examination of the data may not be very revealing. The sit-
uation is further complicated by censoring, in that the occurrence of censored
survival times make it difficult to judge aspects of model adequacy, even in
the simplest of situations. Visual inspection of the data should therefore be
supplemented by diagnostic procedures for detecting inadequacies in a fitted
model.
Once a model has been fitted, there are a number of aspects of the fit
of a model that need to be studied. For example, the model must include an
appropriate set of explanatory variables from those measured in the study, and
we will need to check that the correct functional form of these variables has
been used. It might be important to identify observed survival times that are
greater than would have been anticipated, or individuals whose explanatory
variables have an undue impact on particular hazard ratios. Also, some means
of checking the assumption of proportional hazards might be required.
Many model-checking procedures are based on quantities known as resid-
uals. These are values that can be calculated for each individual in the study,
and have the feature that their behaviour is known, at least approximately,
when the fitted model is satisfactory. A number of residuals have been pro-
posed for use in connection with the Cox regression model, and this chapter
begins with a review of some of these. The use of residuals in assessing specific
aspects of model adequacy is then discussed in subsequent sections.
4.1 Residuals for the Cox regression model

Throughout this section, we will suppose that the survival times of n individ-
uals are available, where r of these are death times and the remaining n − r
131
132 MODEL CHECKING IN THE COX REGRESSION MODEL
are right-censored. We further suppose that a Cox regression model has been
fitted to the survival times, and that the linear component of the model con-
tains p explanatory variables, X1 , X2 , . . . , Xp . The fitted hazard function for
the ith individual, i = 1, 2, . . . , n, is therefore
ĥi (t) = exp(β̂ ′ xi )ĥ0 (t),
where β̂ ′ xi = β̂1 x1i + β̂2 x2i + · · · + β̂p xpi is the value of the risk score for that
individual and ĥ0 (t) is the estimated baseline hazard function.
4.1.1 Cox-Snell residuals

The residual that is most widely used in the analysis of survival data is the
Cox-Snell residual, so called because it is a particular example of the general
definition of residuals given by Cox and Snell (1968).
The Cox-Snell residual for the ith individual, i = 1, 2, . . . , n, is given by
rCi = exp(β̂ ′ xi )Ĥ0 (ti ), (4.1)
where Ĥ0 (ti ) is an estimate of the baseline cumulative hazard function at time
ti , the observed survival time of that individual. In practice, the Nelson-Aalen
estimate given in Equation (3.28) is generally used. Note that from Equa-
tion (3.24), the Cox-Snell residual, rCi , is the value of Ĥi (ti ) = − log Ŝi (ti ),
where Ĥi (ti ) and Ŝi (ti ) are the estimated values of the cumulative hazard and
survivor functions of the ith individual at ti .
This residual can be derived from a general result in mathematical statis-
tics on the distribution of a function of a random variable. According to this
result, if T is the random variable associated with the survival time of an
individual, and S(t) is the corresponding survivor function, then the random
variable Y = − log S(T ) has an exponential distribution with unit mean, irre-
spective of the form of S(t). The proof of this result is outlined in the following
paragraph, which can be omitted without loss of continuity.
From a general result, if fX (x) is the probability density function of the
random variable X, the density of the random variable Y = g(X) is given by

dy
fY (y) = fX {g (y)}/ ,
−1
dx
where fX {g −1 (y)} is the density of X expressed in terms of y. Using this

result, the probability density function of the random variable Y = − log S(T )
is given by
{ −1 −y } dy
fY (y) = fT S (e ) / , (4.2)
dt
where fT (t) is the probability density function of T . Now,
dy d{− log S(t)} fT (t)
= = ,
dt dt S(t)
RESIDUALS FOR THE COX REGRESSION MODEL 133
and when the absolute value of this function is expressed in terms of y, the
derivative becomes
{ } { }
fT S −1 (e−y ) fT S −1 (e−y )
= .
S {S −1 (e−y )} e−y
Finally, on substituting for the derivative in Equation (4.2), we find that
fY (y) = e−y ,
which, from Equation (5.3), is the probability density function of an exponen-

tial random variable, Y , with unit mean.
The next and crucial step in the argument is as follows. If the model fitted
to the observed data is satisfactory, a model-based estimate of the survivor
function for the ith individual at ti , the survival time of that individual,
will be close to the corresponding true value Si (ti ). This suggests that if the
correct model has been fitted, the values Ŝi (ti ) will have properties similar
to those of Si (ti ). Then, the negative logarithms of the estimated survivor
functions, − log Ŝi (ti ), i = 1, 2, . . . , n, will behave as n observations from a
unit exponential distribution. These estimates are the Cox-Snell residuals.
If the observed survival time for an individual is right-censored, then the
corresponding value of the residual is also right-censored. The residuals will
therefore be a censored sample from the unit exponential distribution, and a
test of this assumption provides a test of model adequacy, to which we return
in Section 4.2.1.
The Cox-Snell residuals, rCi , have properties that are quite dissimilar to
those of residuals used in linear regression analysis. In particular, they will
not be symmetrically distributed about zero and they cannot be negative.
Furthermore, since the Cox-Snell residuals are assumed to have an exponential
distribution when an appropriate model has been fitted, they have a highly
skew distribution and the mean and variance of the ith residual will both be
unity.
4.1.2 Modified Cox-Snell residuals

Censored observations lead to residuals that cannot be regarded on the same
footing as residuals derived from uncensored observations. We might therefore
seek to modify the Cox-Snell residuals so that explicit account can be taken
of censoring.
Suppose that the ith survival time is a censored observation, t∗i , and let
ti be the actual, but unknown, survival time, so that ti > t∗i . The Cox-Snell
residual for this individual, evaluated at the censored survival time, is then
given by
rCi = Ĥi (t∗i ) = − log Ŝi (t∗i ),
where Ĥi (t∗i ) and Ŝi (t∗i ) are the estimated cumulative hazard and survivor
functions, respectively, for the ith individual at the censored survival time.
If the fitted model is correct, then the values rCi can be taken to have a unit
exponential distribution. The cumulative hazard function of this distribution
increases linearly with time, and so the greater the value of the survival time
ti for the ith individual, the greater the value of the Cox-Snell residual for that
individual. It then follows that the residual for the ith individual at the actual
(unknown) failure time, Ĥi (ti ), will be greater than the residual evaluated at
the observed censored survival time.
To take account of this, Cox-Snell residuals can be modified by the addition
of a positive constant ∆, which can be called the excess residual. Modified
Cox-Snell residuals are therefore of the form
{
′ rCi for uncensored observations,
rCi =
rCi + ∆ for censored observations,
where rCi is the Cox-Snell residual for the ith observation, defined in Equa-
tion (4.1). It now remains to identify a suitable value for ∆. For this, we use
the lack of memory property of the exponential distribution.
To demonstrate this property, suppose that the random variable T has
an exponential distribution with mean λ−1 , and consider the probability that
T exceeds t0 + t1 , t1 > 0, conditional on T being at least equal to t0 . From
the standard result for conditional probability given in Section 3.3.1, this
probability is
P(T > t0 + t1 and T > t0 )

P(T > t0 + t1 | T > t0 ) = .
P(T > t0 )
The numerator of this expression is simply P(T > t0 + t1 ), and so the required
probability is the ratio of the probability of survival beyond t0 +t1 to the prob-
ability of survival beyond t0 , that is S(t0 + t1 )/S(t0 ). The survivor function
for the exponential distribution is given by S(t) = e−λt , as in Equation (5.2)
of Chapter 5, and so
exp{−λ(t0 + t1 )}
P(T > t0 + t1 | T > t0 ) = = e−λt1 ,
exp(−λt0 )
which is the survivor function of an exponential random variable at time t1 ,

that is P(T > t1 ). This result means that, conditional on survival to time t0 ,
the excess survival time beyond t0 also has an exponential distribution with
mean λ−1 . In other words, the probability of survival beyond time t0 is not
affected by the knowledge that the individual has already survived to time t0 .
From this result, since rCi has a unit exponential distribution, the excess
residual, ∆, will also have a unit exponential distribution. The expected value
of ∆ is therefore unity, suggesting that ∆ may be taken to be unity, and this
leads to modified Cox-Snell residuals, given by
{
′ rCi for uncensored observations,
rCi = (4.3)
rCi + 1 for censored observations.
The ith modified Cox-Snell residual can be expressed in an alternative form
by introducing an event indicator, δi , which takes the value zero if the observed
survival time of the ith individual is censored and unity if it is uncensored.
Then, Equation (4.3), the modified Cox-Snell residual is given by
′
rCi = 1 − δi + rCi . (4.4)
′
Note that from the definition of this type of residual, rCi must be greater
than unity for a censored observation. Also, as for the unmodified residuals,
′
the rCi can take any value between zero and infinity, and they will have a
skew distribution.
On the basis of empirical evidence, Crowley and Hu (1977) found that the
addition of unity to a Cox-Snell residual for a censored observation inflated the
residual to too great an extent. They therefore suggested that the median value
of the excess residual be used rather than the mean. For the unit exponential
distribution, the survivor function is S(t) = e−t , and so the median, t(50), is
such that e−t(50) = 0.5, whence t(50) = log 2 = 0.693. Thus, a second version
of the modified Cox-Snell residual has
{
′′ rCi for uncensored observations,
rCi = (4.5)
rCi + 0.693 for censored observations.
However, if the proportion of censored observations is not too great, the sets of
modified residuals from Equations (4.3) and (4.5) will not appear too different.
4.1.3 Martingale residuals

′
The modified residuals rCi defined in Equation (4.4) have a mean of unity for
uncensored observations. Accordingly, these residuals might be further refined
′
by relocating the rCi so that they have a mean of zero when an observation is
uncensored. If in addition the resulting values are multiplied by −1, we obtain
the residuals
rM i = δi − rCi . (4.6)
These residuals are known as martingale residuals, since they can also be
derived using what are known as martingale methods, referred to later in
Section 13.1 of Chapter 13. In this derivation, the rCi are based on the Nelson-
Aalen estimate of the cumulative hazard function.
Martingale residuals take values between −∞ and unity, with the residuals
for censored observations, where δi = 0, being negative. It can also be shown
that these residuals sum to zero and, in large samples, the martingale resid-
uals are uncorrelated with one another and have an expected value of zero.
In this respect, they have properties similar to those possessed by residuals
encountered in linear regression analysis.
Another way of looking at the martingale residuals is to note that the
quantity rM i in Equation (4.6) is the difference between the observed number
of deaths for the ith individual in the interval (0, ti ) and the corresponding
estimated expected number on the basis of the fitted model. To see this, note
that the observed number of deaths is unity if the survival time ti is uncen-
sored, and zero if censored, that is δi . The second term in Equation (4.6) is
an estimate of Hi (ti ), the cumulative hazard of death for the ith individual
over the interval (0, ti ). From Section 1.3.3 of Chapter 1, this can be inter-
preted as the expected number of deaths in that interval. This shows another
similarity between the martingale residuals and residuals from other areas of
data analysis.
4.1.4 Deviance residuals

Although martingale residuals share many of the properties possessed by resid-
uals encountered in other situations, such as in linear regression analysis, they
are not symmetrically distributed about zero, even when the fitted model is
correct. This skewness makes plots based on the residuals difficult to interpret.
The deviance residuals, which were introduced by Therneau et al. (1990), are
much more symmetrically distributed about zero. They are defined by
1
rDi = sgn(rM i ) [−2 {rM i + δi log(δi − rM i )}] 2 , (4.7)
where rM i is the martingale residual for the ith individual, and the function
sgn(·) is the sign function. This is the function that takes the value +1 if
its argument is positive and −1 if negative. Thus, sgn(rM i ) ensures that the
deviance residuals have the same sign as the martingale residuals.
The original motivation for these residuals is that they are components of
the deviance. The deviance is a statistic that is used to summarise the extent
to which the fit of a model of current interest deviates from that of a model
which is a perfect fit to the data. This latter model is called the saturated
or full model, and is a model in which the β-coefficients are allowed to be
different for each individual. The statistic is given by
{ }
D = −2 log L̂c − log L̂f ,
where L̂c is the maximised partial likelihood under the current model and
L̂f is the maximised partial likelihood under the full model. The smaller the
value of the deviance, the better the model. The deviance can be regarded as a
generalisation of the residual sum of squares used in modelling normal data to
the analysis of non-normal data, and features prominently in generalised linear
modelling. Note that differences in deviance between two alternative models
are the same as differences in the values of the statistic −2 log∑L̂ introduced
2
in Chapter 3. The deviance residuals are then such that D = rDi , so that
observations that correspond to relatively large deviance residuals are those
that are not well fitted by the model.
Another way of viewing the deviance residuals is that they are martingale
residuals that have been transformed to produce values that are symmetric
about zero when the fitted model is appropriate. To see this, first recall that
the martingale residuals rM i can take any value in the interval (−∞, 1). For
large negative values of rM i , the term in square brackets in Equation (4.7) is
dominated by rM i . Taking the square root of this quantity has the effect of
bringing the residual closer to zero. Thus, martingale residuals in the range
(−∞, 0) are shrunk toward zero. Now consider martingale residuals in the
interval (0, 1). The term δi log(δi − rM i ) in Equation (4.7) will only be non-
zero for uncensored observations, and will then have the value log(1 − rM i ).
As rM i gets closer to unity, 1 − rM i gets closer to zero and log(1 − rM i ) takes
large negative values. The quantity in square brackets in Equation (4.7) is
then dominated by this logarithmic term, and so the deviance residuals are
expanded toward +∞ as the martingale residual reaches its upper limit of
unity.
One final point to note is that although these residuals can be expected to
be symmetrically distributed about zero when an appropriate model has been
fitted, they do not necessarily sum to zero.
4.1.5 ∗ Schoenfeld residuals

Two disadvantages of the residuals described in Sections 4.1.1 to 4.1.4 are that
they depend heavily on the observed survival time and require an estimate
of the cumulative hazard function. Both of these disadvantages are overcome
in a residual proposed by Schoenfeld (1982). These residuals were originally
termed partial residuals, for reasons given in the sequel, but are now commonly
known as Schoenfeld residuals. This residual differs from those considered
previously in one important respect. This is that there is not a single value of
the residual for each individual, but a set of values, one for each explanatory
variable included in the fitted Cox regression model.
The ith Schoenfeld residual for Xj , the jth explanatory variable in the
model, is given by
rSji = δi {xji − âji }, (4.8)
where xji is the value of the jth explanatory variable, j = 1, 2, . . . , p, for the
ith individual in the study,
∑
l∈R(ti ) xjl exp(β̂ ′ xl )
âji = ∑ , (4.9)
l∈R(ti ) exp(β̂ ′ xl )
and R(ti ) is the set of all individuals at risk at time ti .

Note that non-zero values of these residuals only arise for uncensored ob-
servations. Moreover, if the largest observation in a sample of survival times
is uncensored, the value of âji for that observation, from Equation (4.9), will
be equal to xji and so rSji = 0. To distinguish residuals that are genuinely
zero from those obtained from censored observations, the latter are usually
expressed as missing values.
The ith Schoenfeld residual, for the explanatory variable Xj , is an estimate
of the ith component of the first derivative of the logarithm of the partial
likelihood function with respect to βj , which, from Equation (3.6), is given by
∂ log L(β) ∑
n
= δi {xji − aji } , (4.10)
∂βj i=1
where ∑ ′
∑l xjl exp(β xl )
aji = ′
. (4.11)
l exp(β xl )
The ith term in this summation, evaluated at β̂, is then the Schoenfeld residual
for Xj , given in Equation (4.8). Since the estimates of the β’s are such that
∂ log L(β)
βˆ = 0,
∂βj
the Schoenfeld residuals must sum to zero. These residuals also have the prop-
erty that, in large samples, the expected value of rSji is zero, and they are
uncorrelated with one another.
It turns out that a scaled version of the Schoenfeld residuals, proposed
by Grambsch and Therneau (1994), is more effective in detecting departures
from the assumed model. Let the vector of Schoenfeld residuals for the ith
individual be denoted r Si = (rS1i , rS2i , . . . , rSpi )′ . The scaled, or weighted,
∗
Schoenfeld residuals, rSji , are then the components of the vector
r ∗Si = d var (β̂)r Si ,
where d is the number of deaths among the n individuals, and var (β̂) is the
variance-covariance matrix of the parameter estimates in the fitted Cox re-
gression model. These scaled Schoenfeld residuals are therefore quite straight-
forward to compute.
4.1.6 ∗ Score residuals

There is one other type of residual that is useful in some aspects of model
checking, and which, like the Schoenfeld residual, is obtained from the first
derivative of the logarithm of the partial likelihood function with respect to
the parameter βj , j = 1, 2, . . . , p. However, the derivative in Equation (4.10)
is now expressed in a quite different form, namely
 
∑n  ∑ 
∂ log L(β) (ajr − xji )δr
= δi (xji − aji ) + exp(β ′ xi ) ∑ ′ ,
∂βj  exp(β xl ) 
i=1 tr 6 ti l∈R(t r )
(4.12)
where xji is the ith value of the jth explanatory variable, δi is the event indi-
cator which is zero for censored observations and unity otherwise, aji is given
in Equation (4.11), and R(tr ) is the risk set at time tr . In this formulation,
the contribution of the ith observation to the derivative only depends on in-
formation up to time ti . In other words, if the study was actually concluded
at time ti , the ith component of the derivative would be unaffected. Residuals
are then obtained as the estimated value of the n components of the deriva-
tive. From Appendix A, the first derivative of the logarithm of the partial
likelihood function, with respect to βj , is the efficient score for βj , written
u(βj ). These residuals are therefore known as score residuals, and are denoted
by rU ji .
From Equation (4.12), the ith score residual, i = 1, 2, . . . , n, for the jth
explanatory variable in the model, Xj , is given by
∑ (âjr − xji )δr
rU ji = δi (xji − âji ) + exp(β̂ ′ xi ) ∑ .
exp(β̂ ′ xl )
tr 6 ti l∈R(tr )
Using Equation (4.8), this may be written in the form

∑ (âjr − xji )δr
rU ji = rSji + exp(β̂ ′ xi ) ∑ , (4.13)
exp(β̂ ′ xl )
tr 6 ti l∈R(tr )
which shows that the score residuals are modifications of the Schoenfeld resid-
uals. As for the Schoenfeld residuals, the score residuals sum to zero, but will
not necessarily be zero when an observation is censored.
In this section, a number of residuals have been defined. We conclude with
an example that illustrates the calculation of these different types of residual
and that shows similarities and differences between them. This example will be
used in many illustrations in this chapter, mainly because the relatively small
number of observations allows the values of the residuals and other diagnostics
to be readily tabulated. However, the methods of this chapter are generally
more informative in larger data sets.
Example 4.1 Infection in patients on dialysis

In the treatment of certain disorders of the kidney, dialysis may be used to re-
move waste materials from the blood. One problem that can occur in patients
on dialysis is the occurrence of an infection at the site at which the catheter
is inserted. If any such infection occurs, the catheter must be removed, and
the infection cleared up. In a study to investigate the incidence of infection,
described by McGilchrist and Aisbett (1991), the time from insertion of the
catheter until infection was recorded for a group of kidney patients. Some-
times, the catheter has to be removed for reasons other than infection, giving
rise to right-censored observations. The data in this example relate to the 13
patients suffering from diseases of the kidney coded as type 3 in their paper.
Table 4.1 gives the number of days from insertion of the catheter until its
removal following the first occurrence of an infection, together with the value
of a variable that indicates the infection status of an individual. This variable
takes the value zero if the catheter was removed for a reason other than the
occurrence of an infection, and unity otherwise. The data set also includes the
age of each patient in years and a variable that denotes their sex (1 = male,
2 = female).
Table 4.1 Times to removal of a catheter

following a kidney infection.
Patient Time Status Age Sex
1 8 1 28 1
2 15 1 44 2
3 22 1 32 1
4 24 1 16 2
5 30 1 10 1
6 54 0 42 2
7 119 1 22 2
8 141 1 34 2
9 185 1 60 2
10 292 1 43 2
11 402 1 30 2
12 447 1 31 2
13 536 1 17 2
When a Cox regression model is fitted to these data, the estimated hazard
function for the ith patient, i = 1, 2, . . . , 13, is found to be
ĥi (t) = exp {0.030 Age i − 2.711 Sex i } ĥ0 (t), (4.14)
where Age i and Sex i refer to the age and sex of the ith patient.
The variable Sex is certainly important, since when Sex is added to the
model that contains Age alone, the decrease in the value of the −2 log L̂ statis-
tic is 6.445 on 1 d.f. This change is highly significant (P = 0.011). On the
other hand, there is no statistical evidence for including the variable Age in
the model, since the change in the value of the −2 log L̂ statistic on adding
Age to the model that contains Sex is 1.320 on 1 d.f. (P = 0.251). However, it
can be argued that from a clinical viewpoint, the hazard of infection may well
depend on age. Consequently, both variables will be retained in the model.
The values of different types of residual for the model in Equation (4.14)
are displayed in Table 4.2. In this table, rCi , rM i and rDi are the Cox-Snell
residuals, martingale residuals and deviance residuals, respectively. Also rS1i
and rS2i are the values of Schoenfeld residuals for the variables Age and Sex,
∗ ∗
respectively, rS1i and rS2i are the corresponding scaled Schoenfeld residuals,
and rU 1i , rU 2i are the score residuals.
The values in this table were computed using the Nelson-Aalen estimate
of the baseline cumulative hazard function given in Equation (3.28). Had the
estimate Ĥ0 (t), in Equation (3.22), been used, different values for all but the
Table 4.2 Different types of residual after fitting a Cox regression model.
∗ ∗
Patient rCi rM i rDi rS1i rS2i rS1i rS2i rU 1i rU 2i
1 0.280 0.720 1.052 −1.085 −0.242 0.033 −3.295 −0.781 −0.174
2 0.072 0.928 1.843 14.493 0.664 0.005 7.069 13.432 0.614
3 1.214 −0.214 −0.200 3.129 −0.306 0.079 −4.958 −0.322 0.058
4 0.084 0.916 1.765 −10.222 0.434 −0.159 8.023 −9.214 0.384
5 1.506 −0.506 −0.439 −16.588 −0.550 −0.042 −5.064 9.833 0.130
6 0.265 −0.265 −0.728 – – – – −3.826 −0.145
7 0.235 0.765 1.168 −17.829 0.000 −0.147 3.083 −15.401 −0.079
8 0.484 0.516 0.648 −7.620 0.000 −0.063 1.318 −7.091 −0.114
9 1.438 −0.438 −0.387 17.091 0.000 0.141 −2.955 −15.811 −0.251
10 1.212 −0.212 −0.199 10.239 0.000 0.085 −1.770 1.564 −0.150
11 1.187 −0.187 −0.176 2.857 0.000 0.024 −0.494 6.575 −0.101
12 1.828 −0.828 −0.670 5.534 0.000 0.046 −0.957 4.797 −0.104
13 2.195 −1.195 −0.904 0.000 0.000 0.000 0.000 16.246 −0.068
Schoenfeld residuals would be obtained. In addition, because the correspond-

ing estimate of the survivor function is zero at the longest removal time, which
is that for patient number 13, values of the Cox-Snell, martingale and deviance
residuals would not then be defined for this patient, and the martingale resid-
uals would no longer sum to zero.
In this data set, there is just one censored observation, which is for patient
number 6. The modified Cox-Snell residuals will then be the same as the Cox-
Snell residuals for all patients except number 6. For this patient, the values of
′ ′′
the two forms of modified residuals are rC6 = 1.265 and rC6 = 0.958. Also, the
Schoenfeld residuals are not defined for the patient with a censored removal
time, and are zero for the patient that has the longest period of time before
removal of the catheter.
The skewness of the Cox-Snell and martingale residuals is clearly shown
in Table 4.2, as is the fact that the Cox-Snell residuals are centred on unity
while the martingale and deviance residuals are centred on zero. Note also
that the martingale, Schoenfeld and score residuals sum to zero, as they
should do.
One unusual feature about the residuals in Table 4.2 is the large number
of zeros for the values of the Schoenfeld residual corresponding to Sex. The
reason for this is that for infection times greater than 30 days, the value of
the variable Sex is always equal to 2. This means that the value of the term
âji for this variable, given in Equation (4.9), is equal to 2 for a survival time
greater than 30 days, and so the corresponding Schoenfeld residual defined in
Equation (4.8) is zero.
We now consider how residuals obtained after fitting a Cox regression model
can be used to throw light on the extent to which the fitted model provides
an appropriate description of the observed data. We will then be in a position
to study the residuals obtained in Example 4.1 in greater detail.
4.2 Assessment of model fit
A number of plots based on residuals can be used in the graphical assessment
of the adequacy of a fitted model. Unfortunately, many graphical procedures
that are analogues of residual plots used in linear regression analysis have not
proved to be very helpful. This is because plots of residuals against quantities
such as the observed survival times, or the rank order of these times, often
exhibit a definite pattern, even when the correct model has been fitted. Tra-
ditionally, plots of residuals have been based on the Cox-Snell residuals, or
adjusted versions of them described in Section 4.1.2. The use of these residuals
is therefore reviewed in the next section, and this is followed by a description
of how some other types of residuals may be used in the graphical assessment
of the fit of a model.
4.2.1 Plots based on the Cox-Snell residuals

In Section 4.1.1, the Cox-Snell residuals were shown to have an exponential
distribution with unit mean, if the fitted model is correct. They therefore
have a mean and variance of unity, and are asymmetrically distributed about
the mean. This means that simple plots of the residuals, such as plots of the
residuals against the observation number, known as index plots, will not lead
to a symmetric display. The residuals are also correlated with the survival
times, and so plots of these residuals against quantities such as the observed
survival times, or the rank order of these times, are also unhelpful.
One particular plot of these residuals, which can be used to assess the
overall fit of the model, leads to an assessment of whether the residuals are
a plausible sample from a unit exponential distribution. This plot is based
on the fact that if a random variable T has an exponential distribution with
unit mean, then the survivor function of T is e−t ; see Section 5.1.1 of Chap-
ter 5. Accordingly, a plot of the cumulative hazard function H(t) = − log S(t)
against t, known as a cumulative hazard plot, will give a straight line through
the origin with unit slope.
This result can be used to examine whether the residuals have a unit
exponential distribution. After computing the Cox-Snell residuals, rCi , the
Kaplan-Meier estimate of the survivor function of these values is found. This
estimate is computed in a similar manner to the Kaplan-Meier estimate of
the survivor function for survival times, except that the data on which the
estimate is based are now the residuals rCi . Residuals obtained from censored
survival times are themselves taken to be censored. Denoting the estimate
by Ŝ(rCi ), the values of Ĥ(rCi ) = − log Ŝ(rCi ) are plotted against rCi . This
gives a cumulative hazard plot of the residuals. A straight line with unit
slope and zero intercept will then indicate that the fitted survival model is
satisfactory. On the other hand, a plot that displays a systematic departure
from a straight line, or yields a line that does not have approximately unit
slope or zero intercept, might suggest that the model needs to be modified in
some way. Equivalently, a log-cumulative hazard plot of the residuals, that is a
ASSESSMENT OF MODEL FIT 143
plot of log Ĥ(rCi ) against log rCi may be used. This plot is discussed in more
detail in Section 4.4.1.

Consider again the data on the time to the occurrence of an infection in kidney
patients, described in Example 4.1. In this example, we first examine whether
the Cox-Snell residuals are a plausible sample of observations from a unit
exponential distribution. For this, the Kaplan-Meier estimate of the survivor
function of the Cox-Snell residuals, Ŝ(rCi ), is obtained. The cumulative hazard
function of the residuals, Ĥ(rCi ), derived from − log Ŝ(rCi ), is then plotted
against the corresponding residual to give a cumulative hazard plot of the
residuals. The details of this calculation are summarised in Table 4.3, and the
cumulative hazard plot is shown in Figure 4.1. The residual for patient number
6 does not lead to values of Ŝ(rC6 ) and Ĥ(rC6 ) because this observation is
censored.
Table 4.3 Calculation of the cumulative

hazard function of the Cox-Snell residuals.
rCi Ŝ(rCi ) Ĥ(rCi )
0.072 0.9231 0.080
0.084 0.8462 0.167
0.235 0.7692 0.262
0.265 – –
0.280 0.6838 0.380
0.484 0.5983 0.514
1.187 0.5128 0.668
1.212 0.4274 0.850
1.214 0.3419 1.073
1.438 0.2564 1.361
1.506 0.1709 1.767
1.828 0.0855 2.459
2.195 0.0000 –
The relatively small number of observations in this data set makes it diffi-
cult to interpret plots of residuals. However, the plotted points in Figure 4.1
are fairly close to a straight line through the origin, which has approximately
unit slope. This could suggest that the model fitted to the data given in Ta-
ble 4.1 is satisfactory.
On the face of it, this procedure would appear to have some merit, but
cumulative hazard plots of the Cox-Snell residuals have not proved to be
very useful in practice. In an earlier section it was argued that since the val-
ues − log S(ti ) have a unit exponential distribution, the Cox-Snell residuals,
which are estimates of these quantities, should have an approximate unit ex-
ponential distribution when the fitted model is correct. This result is then
used when interpreting a cumulative hazard plot of the residuals. Unfortu-
Cumulative hazard of residual 2.5
2.0
1.5
1.0
0.5
0.0
0.0 0.5 1.0 1.5 2.0
Cox-Snell residual
Figure 4.1 Cumulative hazard plot of the Cox-Snell residuals.
nately this approximation is not very reliable, particularly in small samples.

This is because estimates of the βs, and also of the baseline cumulative hazard
function, H0 (t), are needed in the computation of the rCi . The substitution
of estimates means that the actual distribution of the residuals is not neces-
sarily unit exponential, but their exact distribution is not known. In fact, the
distribution of Cox-Snell residuals for n = 3 was shown by Lagakos (1981) to
be quite dissimilar to a unit exponential sample.
On other occasions, a straight line plot may be obtained when the model
fitted is known to be incorrect. Indeed, practical experience suggests that a fit-
ted model has to be seriously wrong before anything other than a straight line
of unit slope is seen in the cumulative hazard plot of the Cox-Snell residuals.
In the particular case of the null model, that is, the model that contains no
explanatory variates, the cumulative hazard plot will be a straight line with
unit slope and zero intercept, even if some explanatory variables should actu-
ally be included in the model. The reason for this is that when no covariates are
included, the Cox-Snell residual for the ith individual reduces to − log Ŝ0 (ti ).
From Equation (3.29) in Chapter 3, in the absence of ties this is approximately
∑k
j=1 1/nj at the kth uncensored survival time, k = 1, 2, . . . , r − 1, where nj
∑k
is the number at risk at time tj . This summation is simply j=1 1/(n − j + 1),
which is the expected value of the kth order statistic in a sample of size n
from a unit exponential distribution.
In view of the limitations of the Cox-Snell residuals in assessing model
adequacy, diagnostic procedures based on other types of residuals that are of
practical use, are described in the following section.
4.2.2 Plots based on the martingale and deviance residuals
The martingale residuals, introduced in Section 4.1.3, can be interpreted as

the difference between the observed and expected number of deaths in the
time interval (0, ti ), for the ith individual. Accordingly, these residuals high-
light individuals who, on the basis of the assumed model, have died too soon
or lived too long. Large negative residuals will correspond to individuals who
have a long survival time, but covariate values that suggest they should have
died earlier. On the other hand, a residual close to unity, the upper limit
of a martingale residual, will be obtained when an individual has an unex-
pectedly short survival time. An index plot of the martingale residuals will
highlight individuals whose survival time is not well fitted by the model. Such
observations may be termed outliers. The data from individuals for whom the
residual is unusually large in absolute value, will need to be the subject of
further scrutiny. Plots of these residuals against the survival time, the rank
order of the survival times, or explanatory variables, may indicate whether
there are particular times, or values of the explanatory variables, where the
model does not fit well.
Since the deviance residuals are more symmetrically distributed than the
martingale residuals, plots based on these residuals tend to be easier to inter-
pret. Consequently, an index plot of the deviance residuals may also be used
to identify individuals whose survival times are out of line.
In a fitted Cox regression model, the hazard of death for the ith individual
at any time depends on the values of explanatory variables for that individual,
xi , through the function exp(β̂ ′ xi ). This means that individuals for whom
β̂ ′ xi has a large negative value have a lower than average risk of death, and
individuals for whom β̂ ′ xi has a large positive value have a higher than average
risk. The quantity β̂ ′ xi is the risk score, introduced in Section 3.1 of Chapter
3, and provides information about whether an individual might be expected to
survive for a short or long time. By reconciling information about individuals
whose survival times are out of line, with the values of their risk score, useful
information can be obtained about the characteristics of observations that are
not well fitted by the model. In this context, a plot of the deviance residuals
against the risk score is a particularly helpful diagnostic.

Consider again the data on times to infection in kidney patients. From the
values of the martingale and deviance residuals given in Table 4.2, we see that
patient 2 has the largest positive residual, suggesting that the time to removal
of the catheter is shorter for this patient than might have been expected on the
basis of the fitted model. The table also shows that the two types of residual do
not rank the observations in the same order. For example, the second largest
negative martingale residual is found for patient 12, whereas patient 6 has
the second largest negative deviance residual. However, the observations that
have the most extreme values of the martingale and deviance residuals will
tend to be the same, as in this example. Index plots of the martingale and
deviance residuals are shown in Figure 4.2.
1.0 2.0
1.5
0.5
Martingale residual
Deviance residual
1.0
0.0
0.5
-0.5
0.0
-1.0
-0.5
-1.5 -1.0
1 3 5 7 9 11 13 1 3 5 7 9 11 13
Index Index
Figure 4.2 Index plots of the martingale and deviance residuals.
The plots are quite similar, but the distribution of the deviance residuals
is seen to be more symmetric. The plots also show that there are no patients
who have residuals that are unusually large in absolute value. Figure 4.3 gives
a plot of the deviance residuals against the risk scores, that are found from
the values of 0.030 Age i − 2.711 Sex i , for i = 1, 2, . . . , 13.
2.0
1.5
Deviance residual
1.0
0.5
0.0
-0.5
-1.0
-5 -4 -3 -2 -1
Risk score
Figure 4.3 Plot of the deviance residuals against the values of the risk score.
This figure shows that patients with the largest deviance residuals have
low risk scores. This indicates that these patients are at relatively low risk
of an early catheter removal, and yet their removal time is sooner than
expected.
4.2.3 Checking the functional form of covariates
Although the model-based approach to the analysis of survival data, described

in Chapter 3, identifies a particular set of covariates on which the hazard
function depends, it will be important to check that the correct functional form
has been adopted for these variables. An improvement in the fit of a model
may well be obtained by using some transformation of the values of a variable
instead of the original values. For example, it might be that a better fitting
model is obtained by using a non-linear function of the age of an individual
at baseline, or the logarithm of a biochemical variable such as serum bilirubin
level. Similarly, an explanatory variable such as serum cholesterol level may
only begin to exert an effect on survival when it exceeds some threshold value,
after which time the hazard of death might increase with increasing values of
that variable.
A straightforward means of assessing this aspect of model adequacy is
based on the martingale residuals obtained from fitting the null model, that is,
the model that contains no covariates. These residuals are then plotted against
the values of each covariate in the model. It has been shown by Therneau et
al. (1990) that this plot should display the functional form required for the
covariate. In particular, a straight line plot indicates that a linear term is
needed.
As an extension to this approach, if the functional form of certain co-
variates can be assumed to be known, martingale residuals may be calculated
from the fitted Cox regression model that contains these covariates alone. The
resulting martingale residuals are then plotted against the covariates whose
functional form needs to be determined.
The graphs obtained in this way are usually quite ‘noisy’ and their inter-
pretation is much helped by superimposing a smoothed curve that is fitted
to the scatterplot. There are a number of such smoothers in common use, in-
cluding smoothing splines. However, the most widely used smoothers are the
LOWESS (locally weighted scatterplot smoothing) and LOESS (locally esti-
mated scatterplot smoothing) methods, introduced by Cleveland (1979) and
implemented in many software packages.
Even with a smoother, it can be difficult to discern a specific functional
form when a non-linear pattern is seen in the plot. If a specific transforma-
tion is suggested, such as the logarithmic transformation, the covariate can
be so transformed, and the martingale residuals for the null model plotted
against the transformed variate. A straight line would then confirm that an
appropriate transformation has been applied.
In this example, we illustrate the use of martingale residuals in assessing
whether the age effect is linear in the Cox regression model fitted to the data
of Example 4.1. First, the martingale residuals for the null model are obtained
and these are plotted against the corresponding values of the age of a patient
in Figure 4.4.
1.0
0.5
Martingale residual for null model
0.0
-0.5
-1.0
-1.5
-2.0
-2.5
-3.0
10 20 30 40 50 60
Age
Figure 4.4 Plot of the martingale residuals for the null model against Age, with a
smoothed curve superimposed.
There is too little data to say much about this graph, but the smoothed
curve indicates that there is no need for anything other than a linear term in
Age. In fact, the age effect is not actually significant, and so it is not surprising
that the smoothed curve is roughly horizontal.
We end this section with a further illustrative example.

In this example we return to the data on the survival times of 48 patients with
multiple myeloma, described in Example 1.3. In Example 3.5, a Cox regression
model that contained the explanatory variables Hb (serum haemoglobin) and
Bun (blood urea nitrogen) was found to be a suitable model for the hazard
function. We now perform an analysis of the residuals in order to study the
adequacy of this fitted model.
First, a cumulative hazard plot of the Cox-Snell residuals is shown in
Figure 4.5. The line made by the plotted points in Figure 4.5 is reasonably
straight, and has a unit slope and zero intercept. On the basis of this plot,
there is no reason to doubt the adequacy of the fitted model. However, as
Cumulative hazard of residual 4
0
0 1 2 3
Cox-Snell residual
pointed out in Section 4.2.1, this plot is not very sensitive to departures from
the fitted model.
To further assess the fit of the model, the deviance residuals are plotted
against the corresponding risk scores in Figure 4.6.
2
Deviance residual
-1
-2
-2 -1 0 1 2
Risk score
Figure 4.6 Deviance residuals plotted against the risk score.

This plot shows that patients 41 and 38 have the largest values of the
deviance residuals, but these are not much separated from values of the resid-
uals for some of the other patients. Patients with the three largest risk scores
have residuals that are close to zero, suggesting that these observations are
well fitted by the model. Again, there is no reason to doubt the validity of the
fitted model.
In order to investigate whether the correct functional form for the variates
Hb and Bun has been used, martingale residuals are calculated for the null
model and plotted against the values of these variables. The resulting plots,
with a smoothed curve superimposed to aid in their interpretation, are shown
in Figures 4.7 and 4.8.
1
-1
-2
-3
4 6 8 10 12 14 16
Value of Hb
Figure 4.7 Plot of the martingale residuals for the null model against the values of
Hb, with a smoothed curve superimposed.
The plots for Hb and Bun confirm that linear terms in each variable are
required in the model. Note that the slope of the plot for Hb in Figure 4.7 is
negative, corresponding to the negative coefficient of Hb in the fitted model,
while the plot for Bun in Figure 4.8 has a positive slope.
In this data set, the values of Bun range from 6 to 172, and the distribution
of their values across the 48 subjects is positively skewed. In order to guard
against the extreme values of this variate having an undue impact on the co-
efficient of Bun, logarithms of this variable might be used in the modelling
process. Although there is no suggestion of this in Figure 4.8, for illustrative
purposes, we will use this type of plot to investigate whether a model contain-
ing log Bun rather than Bun is acceptable. Figure 4.9 shows the martingale
residuals for the null model plotted against the values of log Bun.
1
-1
-2
-3
0 25 50 75 100 125 150 175 200
Value of Bun
Bun, with a smoothed curve superimposed.
1
-1
-2
-3
1 2 3 4 5
Value of log Bun
log Bun, with a smoothed curve superimposed.
The smoothed curve in this figure does suggest that it is not appropriate
to use a linear term in log Bun. Indeed, if it were decided to use log Bun in the
model, Figure 4.9 indicates that a quadratic term in log Bun may be needed.
In fact, adding this quadratic term to a model that includes Hb and log Bun
leads to a significant reduction in the value of −2 log L̂, but the resulting value
of this statistic, 201.458, is then only slightly less than the corresponding value
for the model containing Hb and Bun, which is 202.938. This analysis confirms
that the model should contain linear terms in the variables Hb and Bun.
4.3 Identification of influential observations

In the assessment of model adequacy, it is important to determine whether
any particular observation has an undue impact on inferences made on the
basis of a model fitted to an observed set of survival data. Observations that
do have an effect on model-based inferences are said to be influential.
As an example, consider a survival study in which a new treatment is to
be compared with a standard. In such a comparison, it would be important
to determine if the hazard of death on the new treatment, relative to that on
the standard, was substantially affected by any one individual. In particular,
it might be that when the data record for one individual is removed from the
database, the relative hazard is increased or reduced by a substantial amount.
If this happens, the data from such an individual would need to be subject to
particular scrutiny.
Conclusions from a survival analysis are often framed in terms of estimates
of quantities such as the relative hazard and median survival time, which de-
pend on the estimated values of the β-parameters in the fitted Cox regression
model. It is therefore of particular interest to examine the influence of each
observation on these estimates. We can do this by examining the extent to
which the estimated parameters in the fitted model are affected by omitting in
turn the data record for each individual in the study. In some circumstances,
estimates of a subset of the parameters may be of special importance, such
as parameters associated with treatment effects. The study of influence may
then be limited to just these parameters. On many occasions, the influence
that each observation has on the estimated hazard function will be of interest,
and it would then be important to identify observations that influence the
complete set of parameter estimates under the model. These two aspects of
influence are discussed in the following sections.
In contrast to models encountered in the analysis of other types of data,
such as the general linear model, the effect of removing one observation from
a set of survival data is not easy to study. This is mainly because the log-
likelihood function for the Cox regression model cannot be expressed as the
sum of a number of terms, in which each term is the contribution to the log-
likelihood made by each observation. Instead, the removal of one observation
affects the risk sets over which quantities of the form exp(β ′ x) are summed.
This means that influence diagnostics are quite difficult to derive and so the
IDENTIFICATION OF INFLUENTIAL OBSERVATIONS 153
following sections of this chapter simply give the relevant results. References
to the articles that contain derivations of the quoted formulae are included in
the final section of this chapter.
4.3.1 ∗ Influence of observations on a parameter estimate

Suppose that we wish to determine whether any particular observation has
an untoward effect on β̂j , the jth parameter estimate, j = 1, 2, . . . , p, in a
fitted Cox regression model. One way of doing this would be to fit the model
to all n observations in the data set, and to then fit the same model to the
sets of n − 1 observations obtained by omitting each of the n observations
in turn. The actual effect that omitting each observation has on the param-
eter estimate could then be determined. This procedure is computationally
expensive, unless the number of observations is not too large, and so we use
instead an approximation to the amount by which β̂j changes when the ith
observation is omitted, for i = 1, 2, . . . , n. Suppose that the value of the jth
parameter estimate on omitting the ith observation is denoted by β̂j(i) . Cain
and Lange (1984) showed that an approximation to β̂j − β̂j(i) is based on the
score residuals, described in Section 4.1.6.
Let r U i denote the vector of values of the score residuals for the ith obser-
vation, so that r ′U i = (rU 1i , rU 2i , . . . , rU pi ), where rU ji , j = 1, 2, . . . , p, is the
ith score residual for the jth explanatory variable, given in Equation (4.13).
An approximation to β̂j − β̂j(i) , the change in β̂j on omitting the ith obser-
vation, is then the jth component of the vector
r ′U i var (β̂),
var (β̂) being the variance-covariance matrix of the vector of parameter esti-
mates in the fitted Cox regression model. The jth element of this vector, which
is called a delta-beta, will be denoted by ∆i β̂j , so that ∆i β̂j ≈ β̂j − β̂j(i) . Use
of this approximation means that the values of ∆i β̂j can be computed from
quantities available after fitting the model to the full data set.
Observations that influence a particular parameter estimate, the jth say,
will be such that the values of ∆i β̂j , the delta-betas for these observations, are
larger in absolute value than for other observations in the data set. Index plots
of the delta-betas for each explanatory variable in the model will then reveal
whether there are observations that have an undue impact on the parameter
estimate for any particular explanatory variable. In addition, a plot of the
values of ∆i β̂j against the rank order of the survival times yields information
about the relation between survival time and influence.
The delta-betas may be standardised by dividing ∆i β̂j by the standard
error of β̂j to give a standardised delta-beta. The standardised delta-beta can
be interpreted as the change in the value of the statistic β̂/ se (β̂), on omitting
the ith observation. Since this statistic can be used in assessing whether a par-
ticular parameter has a value significantly different from zero (see Section 3.4
of Chapter 3), the standardised delta-beta can be used to provide information
on how the significance of the parameter estimate is affected by the removal
of the ith observation from the database. Again, an index plot is the most
useful way of displaying the standardised delta-betas.
The statistic ∆i β̂j is an approximation to the actual change in the pa-
rameter estimate when the ith observation is omitted from the fit. The ap-
proximation is generally adequate in the sense that observations that have
an influence on a parameter estimate will be highlighted. However, the ac-
tual effect of omitting any particular observation on model-based inferences
will need to be studied. The agreement between the actual and approximate
delta-betas in a particular situation is illustrated in Example 4.6.

In this example, we return to the data on the times to infection following com-
mencement of dialysis. To investigate the influence that the data from each
of the 13 patients in the study has on the estimated value of the coefficients
of the variables Age and Sex in the linear component of the fitted Cox regres-
sion model, the approximate unstandardised delta-betas, ∆i β̂1 and ∆i β̂2 , are
obtained. These are given in Table 4.4.
Table 4.4 Approximate delta-betas

for Age (β̂1 ), and Sex (β̂2 ).
Observation ∆i β̂1 ∆i β̂2
1 0.0020 −0.1977
2 0.0004 0.5433
3 −0.0011 0.0741
4 −0.0119 0.5943
5 0.0049 0.0139
6 −0.0005 −0.1192
7 −0.0095 0.1270
8 −0.0032 −0.0346
9 −0.0073 −0.0734
10 0.0032 −0.2023
11 0.0060 −0.2158
12 0.0048 −0.1939
13 0.0122 −0.3157
The largest delta-beta for Age occurs for patient number 13, but there
are other delta-betas with similar values. The actual change in the parameter
estimate on omitting the data for this patient is 0.0195, and so omission of
this observation reduces the hazard of infection relative to the baseline hazard.
The standard error of the parameter estimate for Age in the full data set is
0.026, and so the maximum amount by which this estimate is changed when
one observation is deleted is about three-quarters of a standard error. When
the data from patient 13 is omitted, the age effect becomes less significant,
but the difference is unlikely to be of practical importance.
There are two large delta-betas for Sex that are quite close to one an-
other. These correspond to the observations from patients 2 and 4. The actual
change in the parameter estimate when each observation is omitted in turn is
0.820 and 0.818, and so the approximate delta-betas underestimate the actual
change. The standard error of the estimated coefficient of Sex in the full data
set is 1.096, and so again the change in the estimate on deleting an observa-
tion is less than one standard error. The effect of deleting either of these two
observations is to increase the hazard for males relative to females, so that
the sex effect is slightly more significant.
The approximate delta-betas can be compared with the actual values. In
this example, the agreement is generally quite good, although there is a ten-
dency for the actual changes in the parameter estimates to be underestimated
by the approximation. The largest difference between the actual and approxi-
mate value of the delta-beta for Age is 0.010, which occurs for patient number
8. That for Sex is 0.276, which occurs for patient number 2. These differ-
ences are about a quarter of the value of the standard error of each parameter
estimate.
4.3.2 ∗ Influence of observations on the set of parameter estimates

It may happen that the structure of the fitted model is particularly sensitive to
one or more observations in the data set. Such observations can be detected
using diagnostics that are designed to highlight observations that influence
the complete set of parameter estimates in the risk score. These diagnostics
therefore reflect the influence that individual observations have on the risk
score, and give information that is additional to that provided by the delta-
betas. In particular, excluding a given observation from the data set may not
have a great influence on any particular parameter estimate, and so will not
be revealed from a study of the delta-beta statistics. However, the change in
the set of parameter estimates might be such that the form of the estimated
hazard function, or values of summary statistics based on the fitted model,
change markedly when that observation is removed. Statistics for assessing
the influence of observations on the set of parameter estimates also have the
advantage that there is a single value of the diagnostic for each observation.
This makes them easier to use than diagnostics such as the delta-betas.
A number of diagnostics for assessing the influence of each observation on
the set of parameter estimates have been proposed. In this section, two will
be described, but references to others will be given in the concluding section
of this chapter.
One way of assessing the influence of each observation on the overall fit
of the model is to examine the amount by which the value of minus twice
the logarithm of the maximised partial likelihood, −2 log L̂, under a fitted
model, changes when each observation in turn is left out. Write log L(β̂) for
the value of the maximised log-likelihood when the model is fitted to all n
observations, and log L(β̂ (i) ) for the value of the maximised log-likelihood of
the n observations when the parameter estimates are computed after omitting
the ith observation from the fit. The diagnostic
{ }
2 log L(β̂) − log L(β̂ (i) )
can then be useful in the study of influence.

Pettitt and Bin Daud (1989) show that an approximation to this likelihood
displacement is
LDi = r ′U i var (β̂) r U i , (4.15)
where r U i is the p × 1 vector of score residuals, whose jth component is
given in Equation (4.13), and var (β̂) is the variance-covariance matrix of β̂,
the vector of parameter estimates. The values of this statistic may therefore
be straightforwardly obtained from terms used in computing the delta-betas
for each explanatory variable in the model. An index plot, or a plot of the
likelihood displacements against the rank order of the survival times, provides
an informative visual summary of the values of the diagnostic. Observations
that have relatively large values of the diagnostic are influential. Plots against
explanatory variables are not recommended, since, as demonstrated by Pettitt
and Bin Daud (1989), these plots can have a deterministic pattern, even when
the fitted model is correct.
Another diagnostic that can be used to assess the impact of each obser-
vation on the set of parameter estimates is based on the n × n symmetric
matrix
B = Θ′ var (β̂) Θ,
where Θ′ is the n×p matrix formed from the vectors r U i , for i = 1, 2, . . . , p. An
argument from linear algebra shows that the absolute values of the elements
of the n×1 eigenvector associated with the largest eigenvalue of the matrix B,
standardised to have unit length by dividing each component by the square
root of the sum of squares of all the components of the eigenvector, is a
measure of the sensitivity of the fit of the model to each of the n observations
in the data set. Denoting this eigenvector by lmax , the ith element of lmax
is a measure of the influence of the ith observation on the set of parameter
estimates. The sign of this diagnostic is immaterial, and so plots based on
the absolute values, |lmax | are commended for general use. Denoting the ith
element of |lmax | by |lmax |i , i = 1, 2, . . . , n, index plots of these values, plots
against the rank order of the survival times, and against explanatory variables
in the model, can all be useful in the assessment of influence.
The standardisation to unit length means that the squares of the values
|lmax |i must sum to 1.0. Observations for which the squares of the elements
of the eigenvector account for a substantial proportion of the total sum of
squares of unity will then be those that are most influential. Large elements
of this eigenvector will therefore correspond to observations that have most
effect on the value of the likelihood function. A final point to note is that
unlike other diagnostics, a plot of the values of |lmax |i against explanatory
variables will not have a deterministic pattern if the fitted model is correct.
This means that plots of |lmax |i against explanatory variables can be useful
in assessing whether there are particular ranges of values of the variates over
which the model does not fit well.

The data first given in Example 4.1 will again be used to illustrate the use of
diagnostics designed to reveal observations that influence the complete set of
parameter estimates. In Table 4.5, the approximate likelihood displacements
from Equation (4.15), and the values of |lmax |i are given.
Table 4.5 Values of the approximate likelihood

displacement, LDi , and the elements of |lmax |.
Observation LDi |lmax |i
1 0.033 0.161
2 0.339 0.309
3 0.005 0.068
4 0.338 0.621
5 0.050 0.104
6 0.019 0.058
7 0.136 0.291
8 0.027 0.054
9 0.133 0.124
10 0.035 0.193
11 0.061 0.264
12 0.043 0.224
13 0.219 0.464
The observations that most affect the value of the maximised log-likelihood
when they are omitted are those corresponding to patients 2 and 4. The value
of the likelihood displacement diagnostic is also quite large for patient number
13. This means that the set of parameter estimates are most affected by the
removal of either of these three patients from the database.
The fourth element of |lmax |, |lmax |4 , is the largest in absolute value, and
indicates that omitting the data from patient number 4 has the greatest effect
on the pair of parameter estimates. The elements corresponding to patients 2
and 13 are also large relative to the other values, suggesting that the data for
these patients are also influential. The sum of the squares of elements 2, 4 and
13 of |lmax | is 0.70. The total of the sums of squares of the elements is 1.00, and
so cases 2, 4 and 13 account for nearly three-quarters of the variability in the
vales |lmax |i . Note that the analysis of the delta-betas in Example 4.6 showed
that the observations from patients 2 and 4 most influence the parameter
estimate for Sex, while the observation for patient 13 has a greater effect on
the estimate for Age.
In summary, the observations from patients 2, 4 and 13 affect the form
of the hazard function to the greatest extent. Omitting each of these in turn
gives the following estimates of the linear component in the hazard functions
for the ith individual.
Omitting patient number 2: 0.031 Age i − 3.530 Sex i
For comparison, the linear component for the full data set is
0.030 Age i − 2.711 Sex i .
To illustrate the magnitude of the change in estimated hazard ratios, consider
the relative hazard of infection at time t for a patient aged 50 years relative
to one aged 40 years. For the full data set, this is e0.304 = 1.355. This value
increases to 1.365 and 1.564 when patients 2 and 4, respectively, are omitted,
and decreases to 1.114 when patient 13 is omitted. The effect on the haz-
ard function of removing these patients from the database is therefore not
particularly marked.
In the same way, the hazard of infection at time t for a male patient
(Sex = 1) relative to a female (Sex = 2) is e2.711 , that is, 5.041 for the full
data set. When observations 2, 4, and 13 are omitted in turn, the hazard ratio
for males relative to females is 4.138, 4.097 and 9.334, respectively. Omission
of the data from patient number 13 appears to have a great effect on the
estimated hazard ratio. However, some caution is needed in interpreting this
result. Since there are very few males in the data set, the estimated hazard
ratio is imprecisely estimated. In fact, a 95% confidence interval for the hazard
ratio, when the data from patient 13 are omitted, ranges from 0.012 to 82.96!
4.3.3 Treatment of influential observations

Once observations have been found to be unduly influential, it is difficult to
offer any firm advice on what should be done about them. So much depends
on the scientific background to the study.
When possible, the origin of influential observations should be checked. Er-
rors in transcribing and recording categorical and numerical data frequently
occur. If any mistakes are found, the data need to be corrected and the anal-
ysis repeated. If the observed value of a survival time, or other explanatory
variables, is unrealistic, and correction is not possible, the corresponding ob-
servation should be omitted from the database before repeating the analysis.
In many situations it will not be possible to confirm that the data corre-
sponding to an influential observation are valid. Certainly, influential observa-
tions should not then be rejected outright. In these circumstances, the most
appropriate course of action will be to establish the actual effect on the infer-
ences to be drawn from the analysis. For example, if a median survival time
is being used to summarise survival data, or a relative hazard is being used to
summarise a treatment effect, the values of these statistics with and without
the influential values can be contrasted. If the difference between the results
is so small as to not be of practical importance, the queried observations can
be retained. On the other hand, if the effect of removing the influential ob-
servations is large enough to be of practical importance, analyses based on
both the full and reduced data sets will need to be reported. The outcome
of consultations with the scientists involved in the study will then be a vital
ingredient in the process of deciding on the course of future action.

The effect of individual observations on the estimated values of the parameters
of a Cox regression model fitted to the data from Example 1.3 will now be
investigated. Plots of the approximate unstandardised delta-betas for Hb and
Bun against the rank order of the survival times are shown in Figures 4.10
and 4.11.
0.03
0.02
Delta-beta for Hb
0.01
0.00
-0.01
-0.02
0 10 20 30 40 50
Rank of survival time
Figure 4.10 Plot of the delta-betas for Hb against rank order of survival time.
From Figure 4.10, no one observation stands out as having a delta-beta for
Hb that is different from the rest. However, Figure 4.11 shows that the two
observations with the shortest survival times have relatively large positive or
large negative delta-betas for Bun. These correspond to patients 32 and 38 in
the data given in Table 1.3. Patient 32 has a survival time of just one month,
and the second largest value of Bun. Deletion of this observation from the
database decreases the parameter estimate for Bun. Patient number 38 also
survived for just one month after trial entry, but has a value of Bun that is
0.002
0.001
Delta-beta for Bun
0.000
-0.001
-0.002
0 10 20 30 40 50
Figure 4.11 Plot of the delta-betas for Bun against rank order of survival time.
rather low for someone surviving for such a short time. If the data from this
patient are omitted, the coefficient of Bun in the model is increased.
To identify observations that influence the set of parameter estimates, a
plot of the absolute values of the elements of the diagnostic lmax against the
rank order of the survival times is shown in Figure 4.12.
The observation with the largest value of |lmax | corresponds to patient 13.
This patient has an unusually small value of Hb, and a value of Bun that
is a little high, for someone who has survived as long as 65 months. If this
observation is omitted from the data set, the coefficient of Bun remains the
same, but that of Hb is reduced from −0.134 to −0.157. The effect of Hb on
the hazard of death is then a little more significant. In summary, the record
for patient 13 has little effect on the form of the estimated hazard function.
4.4 Testing the assumption of proportional hazards

So far in this chapter we have concentrated on how the adequacy of the linear
component of a survival model can be examined. A crucial assumption made
when using the Cox regression model is that of proportional hazards. Hazards
are said to be proportional if ratios of hazards are independent of time. If
there are one or more explanatory variables in the model whose coefficients
vary with time, or if there are explanatory variables that are time-dependent,
the proportional hazards assumption will be violated. We therefore require
techniques that can be used to detect whether there is some form of time de-
pendency in particular covariates, after allowing for the effects of explanatory
variables that are known, or expected to be, independent of time.
TESTING THE ASSUMPTION OF PROPORTIONAL HAZARDS 161
Absolute value of max 0.4
0.3
0.2
0.1
0.0
0 10 20 30 40 50
Figure 4.12 Plot of the absolute values of the elements of lmax against rank order of
survival time.
In this section, a straightforward plot that can be used in advance of model

fitting is first described, and this is followed by a description of how diagnostics
and test statistics derived from a fitted model can be used in examining the
proportional hazards assumption.
4.4.1 The log-cumulative hazard plot

In the Cox regression model, the hazard of death at any time t for the ith of
n individuals is given by
hi (t) = exp(β ′ xi )h0 (t), (4.16)
where xi is the vector of values of explanatory variables for that individual,

β is the corresponding vector of coefficients, and h0 (t) is the baseline hazard
function. Integrating both sides of this equation over t gives
∫ t ∫ t
hi (u) du = exp(β ′ xi ) h0 (u) du,
0 0
and so, using Equation (1.7),
Hi (t) = exp(β ′ xi )H0 (t),
where Hi (t) and H0 (t) are the cumulative hazard functions. Taking logarithms
of each side of this equation, we get
log Hi (t) = β ′ xi + log H0 (t),

from which it follows that differences in the log-cumulative hazard functions
do not depend on time. This means that if the log-cumulative hazard functions
for individuals with different values of their explanatory variables are plotted
against time, the curves so formed will be parallel if the proportional hazards
model in Equation (4.16) is valid. This provides the basis of a widely used
diagnostic for assessing the validity of the proportional hazards assumption.
It turns out that plotting the log-cumulative hazard functions against the log-
arithm of t, rather than t itself, is a useful diagnostic in parametric modelling,
and so this form of plot is generally used; see Section 5.2 of Chapter 5 for
further details on the use of this log-cumulative hazard plot.
To use this plot, the survival data are first grouped according to the levels
of one or more factors. If continuous variables are to feature in this analysis,
their values will first need to be grouped in some way to give a categorical vari-
able. The Kaplan-Meier estimate of the survivor function of the data in each
group is then obtained. A log-cumulative hazard plot, that is, a plot of the
logarithm of the estimated cumulative hazard function against the logarithm
of the survival time, will yield parallel curves if the hazards are proportional
across the different groups. This method is informative, and simple to operate
when there is a small number of factors, and a reasonable number of obser-
vations at each level. On the other hand, the plot will be based on very few
observations at the later survival times, and in more highly structured data
sets, a different approach needs to be taken.

We again use the data on the survival times of 48 patients with multiple
myeloma to illustrate the log-cumulative hazard plot. In particular we will in-
vestigate whether the assumption of proportional hazards is valid for the vari-
able Hb, which is associated with the serum haemoglobin level. Because this is
a continuous variable, we first need to categorise the values of Hb. This will be
done in the same manner as in Example 3.8 of Chapter 3, where four groups
were defined with values of Hb which are such that Hb 6 7, 7 < Hb 6 10,
10 < Hb 6 13 and Hb > 13. The patients are then grouped according to their
haemoglobin level, and the Kaplan-Meier estimate of the survivor function
is obtained for each of the four groups. From this estimate, the estimated
log-cumulative hazard is formed using the relation Ĥ(t) = − log Ŝ(t), from
Equation (1.8) of Chapter 1, and plotted against the values of log t. The re-
sulting log-cumulative hazard plot is shown in Figure 4.13.
This figure indicates that the plots for Hb 6 7, 7 < Hb 6 10 and Hb > 13
are roughly parallel. The plot for 10 < Hb 6 13 is not in line with the oth-
ers, although this impression results from relatively large cumulative hazard
estimates at the longest survival times experienced by patients in this group.
This plot takes no account of the values of the other variable, Bun, and it
could be that the survival times of the individuals in the third Hb group have
been affected by their Bun values. Overall, there is little reason to doubt the
proportional hazards assumption.
0
Log-cumulative hazard
-1
-2
-3
0 1 2 3 4 5
Log of survival time
Figure 4.13 Log-cumulative hazard plot for multiple myeloma patients in four groups
defined by Hb 6 7 ( •), 7 < Hb 6 10 ( ), 10 < Hb 6 13 (N) and Hb > 13 (∗).
4.4.2 ∗ Use of Schoenfeld residuals

The Schoenfeld residuals, defined in Section 4.1.5, are particularly useful in
evaluating the assumption of proportional hazards after fitting a Cox regres-
sion model. Grambsch and Therneau (1994) have shown that the expected
value of the ith scaled Schoenfeld residual, i = 1, 2, . . . , n, for the jth ex-
∗
planatory variable in the model, Xj , j = 1, 2, . . . , p, denoted rSji , is given by
∗
E (rSji ) ≈ βj (ti ) − β̂j , (4.17)
where βj (t) is taken to be a time-varying coefficient of Xj , βj (ti ) is the value
of this coefficient at the survival time of the ith individual, ti , and β̂j is the
estimated value of βj in the fitted Cox regression model. Note that these
residuals are only defined at death times.
∗
Equation (4.17) suggests that a plot of the values of rSji + β̂j , or equiva-
∗
lently just the scaled Schoenfeld residuals, rSji , against the observed survival
times should give information about the form of the time-dependent coefficient
of Xj , βj (t). In particular, a horizontal line will suggest that the coefficient
of Xj is constant, and the proportional hazards assumption is satisfied. A
smoothed curve can be superimposed on this plot to aid interpretation, as in
the plots of martingale residuals against the values of explanatory variables
in Section 4.2.3.

The data on catheter removal times for patients on dialysis, first given in Ex-
ample 4.1, are now used to illustrate the use of the scaled Schoenfeld residuals
in assessing non-proportional hazards. The scaled Schoenfeld residuals for the
variables Age and Sex were given in Table 4.2, and plotting these values against
the removal times gives the graphs shown in Figure 4.14. The smoothed curves
deviate little from horizontal lines, and in neither plot is there any suggestion
of non-proportional hazards.
0.2 10
Scaled Schoenfeld residual for Age
Scaled Schoenfeld residual for Sex

0.1 5
0.0 0
-0.1 -5
-0.2 -10
0 100 200 300 400 500 600 0 100 200 300 400 500 600
Survival time Survival time
Figure 4.14 Plot of scaled Schoenfeld residuals for Age and Sex.
4.4.3 ∗ Tests for non-proportional hazards

The graphical method for assessing the assumption of proportional hazards,
described in Section 4.4.2, leads to a formal test procedure. From the re-
sult given in Equation (4.17), the expected value of the ith scaled Schoenfeld
residual, i = 1, 2, . . . , n, for the jth explanatory variable, Xj , j = 1, 2, . . . , p,
depends on βj (ti ), the value of a time-varying coefficient of Xj at time ti .
A test of the proportional hazards assumption can then be based on testing
∗
whether there is a linear relationship between E (rSji ) and some function of
∗
time. If there is evidence that E (rSji ) is time-dependent, the hypothesis of
proportional hazards would be rejected.
For a particular explanatory variable Xj , linear dependence of the coeffi-
cient of Xj on time can be expressed by taking βj (ti ) = βj + νj (ti − t̄), where
νj is an unknown regression coefficient. This leads to a linear regression model
∗
with E (rSji ) = νj (ti − t̄), and a test of whether the slope νj is zero leads to
a test of whether the coefficient of Xj is time-dependent and hence of pro-
portional hazards with respect to Xj . Letting τi , τ2 , . . . , τd be the d observed
death times across all n individuals in the data set, Grambsch and Therneau
(1994) show that an appropriate test statistic is
∑d ∗
{ − τ̄ )rSji
i=1 (τi }2
∑d , (4.18)
d var (β̂j ) i=1 (τi − τ̄ )2
∑
where τ̄ = d−1 i=1 τi is the sample mean of the observed death times. Under
d
the null hypothesis that the slope is zero, this statistic has a χ2 distribution
on 1 d.f., and significantly large values of Expression (4.18) lead to rejection
of the proportional hazards assumption for the jth explanatory variable.
An overall or global test of the proportional hazards assumption across all
the p explanatory variables included in a Cox regression model is obtained by
aggregating the individual test statistics in Expression (4.18). This leads to
the statistic
(τ − τ̄ )′ S var (β̂)S ′ (τ − τ̄ )
∑d , (4.19)
i=1 (τi − τ̄ ) /d
2
where τ = (τ1 , τ2 , . . . , τd )′ is the vector formed from the d event times and
S is the d × p matrix whose columns are the (unscaled) Schoenfeld residuals
for the j explanatory variable, so that S = (rSj1 , rSj2 , . . . , rSjd )′ , and var (β̂)
is the variance-covariance matrix of the estimated coefficients of the explana-
tory variables in the fitted Cox regression model. The test statistic in Expres-
sion (4.19) has a χ2 distribution on p d.f. when the assumption of proportional
hazards across all p explanatory variables is true. This test is known as the
Grambsch and Therneau test of proportional hazards, and is sometimes more
enigmatically referred to as the zph test.
The test statistics in Expressions (4.18) and (4.19) can be adapted to
other time-scales by replacing the τi , i = 1, 2, . . . , d, by transformed values
of the death times. For example, using logarithms of the death times, rather
than the times themselves, would allow linearity in the coefficient of Xj to
be assessed on a logarithmic scale. The τi can also be replaced by the rank
order of the death times, or by the Kaplan-Meier estimate of the survivor
function at each event time. Plots of scaled Schoenfeld residuals against time,
discussed in Section 4.4.2, may indicate which of these possible options is the
most appropriate.

We now illustrate tests of proportional hazards using the data on catheter
removal times for patients on dialysis. The variances of the estimated coef-
ficients of the variables Age and Sex in the fitted Cox regression model are
0.000688 and 1.20099, respectively, the sum of squares of the 12 event times
is 393418.92, and the numerator of Expression (4.18) is calculated from the
scaled Schoenfeld residuals in Table 4.2. The values of the test statistic are
0.811 (P = 0.368) and 0.224 (P = 0.636) for Age and Sex, respectively. In
neither case is there evidence against the proportional hazards assumption, as
might have been expected from the graphical analysis in Example 4.10.
Matrix multiplication is required to obtain the numerator of the global
test for proportional hazards in Expression (4.19), and leads to 26578.805,
from which the Grambsch and Therneau test statistic is 0.811. This has a χ2
distribution on 2 d.f. leading to a P -value of 0.667. Again there is no reason
to doubt the validity of the proportional hazards assumption.
4.4.4 ∗ Adding a time-dependent variable
Specific forms of departure from proportional hazards can be investigated by
adding a time-dependent variable to the Cox regression model. Full details
on the use of time-dependent variables in modelling survival data are given
in Chapter 8, but in this section, the procedure is described in a particular
context.
Consider a survival study in which each patient has been allocated to one
of two groups, corresponding to a standard treatment and a new treatment.
Interest may then centre on whether the ratio of the hazard of death at time t
in one treatment group relative to the other, is independent of survival time.
A proportional hazards model for the hazard function of the ith individual in
the study is then
hi (t) = exp(β1 x1i )h0 (t), (4.20)
where x1i is the value of an indicator variable X1 that is zero for the standard
treatment and unity for the new treatment. The relative hazard of death at
any time for a patient on the new treatment, relative to one on the standard,
is then eβ1 , which is independent of the survival time.
Now define a time-dependent explanatory variable X2 , where X2 = X1 t.
If this variable is added to the model in Equation (4.20), the hazard of death
at time t for the ith individual becomes
hi (t) = exp(β1 x1i + β2 x2i )h0 (t), (4.21)
where x2i = x1i t is the value of X1 t for the ith individual. The relative hazard
at time t is now
exp(β1 + β2 t), (4.22)
since X2 = t under the new treatment, and zero otherwise. This hazard ratio
depends on t, and the model in Equation (4.21) is no longer a proportional
hazards model. In particular, if β2 < 0, the relative hazard decreases with
time. This means that the hazard of death on the new treatment, relative
to that on the standard, decreases with time. If β1 < 0, the superiority of
the new treatment becomes more apparent as time goes on. On the other
hand, if β2 > 0, the relative hazard of death on the new treatment increases
with time, reflecting an increasing risk of death on the new treatment relative
to the standard. In the particular case where β2 = 0, the relative hazard is
constant at eβ1 . This means that a test of the hypothesis that β2 = 0 is a
test of the assumption of proportional hazards. The situation is illustrated in
Figure 4.15.
In order to aid both the computation and interpretation of the parameters
in the model of Equation (4.21), the variable X2 can be defined in terms of
the deviation from some time, t0 . The estimated values of β1 and β2 will then
tend to be less highly correlated, and the numerical algorithm for maximising
the appropriate likelihood function will be more stable. If X2 is taken to be
Relative hazard 2>0
1.0
exp( 1) 2=0
2<0
0.0
0
Time
Figure 4.15 Plot of the relative hazard, exp(β1 + β2 t), against t, for different values
of β2 .
such that X2 = X1 (t − t0 ), the value of X2 is t − t0 for the new treatment and

zero for the standard. The relative hazard now becomes
exp{β1 + β2 (t − t0 )}.
In the model of Equation (4.21), with x2i = x1i (t − t0 ), the quantity eβ1 is the
hazard of death at time t0 for an individual on the new treatment relative to
one on the standard. In practical applications, t0 will generally be chosen to
provide a convenient interpretation for the time at which this relative hazard
is applicable. For example, taking t0 to be the mean or median survival time
means that exp(β̂1 ) is the estimated relative hazard of death at this time.
A similar model can be used to detect whether the coefficient of a con-
tinuous variate has a coefficient that depends on time. Suppose that X is
such a variate, and we wish to examine whether there is any evidence that
the coefficient of X is linearly dependent on time. To do this, the term Xt is
added to the model that includes X. The hazard of death at time t for the
ith individual is then
hi (t) = exp(β1 xi + β2 xi t)h0 (t),
where xi is the value of X for that individual. The hazard of death at time
t for an individual for whom X = xi + 1, relative to an individual for whom
X = xi , is then exp(β1 + β2 t), as in Equation (4.22).
The time-dependent variables considered in this section are such that their
coefficients are linearly dependent on time. A similar approach can be used
when a coefficient that is a non-linear function of time is anticipated. For ex-
ample, log t might be used in place of t in the definition of the time-dependent
variable X2 , used in Equation (4.21). In this version of the model, a test of
the hypothesis that β2 = 0 is a test of proportional hazards, where the al-
ternative hypothesis is that the hazard ratio is dependent on the logarithm
of time. Using log t in the definition of a time-dependent variable is also
helpful when the numerical values of the survival times are large, such as
when survival in a long-term study is measured in days. There may then be
computational problems associated with calculating the value of exp(β2 x2i )
in Equation (4.21), which are resolved by using log t in place of t in the
definition of X2 .
Models that include the time-dependent variable X2 cannot be fitted by
treating X2 in the same manner as other explanatory variables in the model.
The reason for this is that this variable will have different values at different
death times, complicating the calculation of the denominator of the partial
likelihood function in Equation (3.4). Full details on the fitting process will be
deferred to Chapter 8. However, inferences about the effect of time-dependent
variables on the hazard function can be evaluated as for other variables. In
particular, the change in the value of the −2 log L̂ statistic can be compared
to percentage points of the chi-squared distribution to test the significance of
the variable. This is therefore a formal test of proportional hazards.

The data on catheter removal times, used in Examples 4.10 and 4.11, will now
be used to illustrate how the proportional hazards assumption can be tested
using time-dependent variates.
We begin by fitting the Cox regression model containing just Age and Sex,
which leads to a value of −2 log L̂ of 34.468. We now define terms that are the
products of these variables with time, namely Tage = Age × t and T sex =
Sex × t. These variables are then added to the model. Note that we cannot
simply form these products from the observed survival times of the patients,
since the model-fitting process requires that these values be computed for
different values of t; see Chapter 8 for details on this.
When the variable Tage is added to the model that contains Age and Sex,
the value of −2 log L̂ reduces to 32.006, but this reduction is not significant at
the 5% level (P = 0.117). The reduction in −2 log L̂ when Tsex is added to the
model that has Age and Sex is only 0.364 (P = 0.546). This analysis confirms
that there is no reason to doubt the assumption of proportional hazards in
respect of the variables Age and Sex.
4.5 Recommendations
In this chapter, a variety of diagnostics have been presented. Which should

be used on a routine basis and which are needed when a more thorough as-
sessment of model adequacy is required?
FURTHER READING 169
In terms of assessing the overall fit of a model, a plot of the deviance
residuals against the risk score gives information on observations that are
not well fitted by the model, and their relation to the set of values of the
explanatory variables. This diagnostic is generally more informative than the
cumulative, or log-cumulative, hazard plot of the Cox-Snell residuals. Plots of
residuals against the survival times, the rank order of the survival times, or
explanatory variables may also be useful.
Plots of residuals might be supplemented by influence diagnostics. When
the inference to be drawn from a model centres on one or two particular
parameters, the delta-beta statistic for those parameters, will be the most
relevant. Plots of these values against the rank order of survival times will then
be useful. To investigate whether there are observations that have an influence
on the set of parameter estimates, or risk score, the diagnostic based on the
absolute values of the elements of lmax is probably the most suitable. Plots
of these values against the rank order of survival times will be informative,
but plots against particular explanatory variables might also be revealing.
An initial assessment of the validity of the proportional hazards assumption
can be made from log-cumulative hazard plots. However, plots based on the
scaled Schoenfeld residuals, and associated tests of non-proportionality are
more helpful. The Grambsch and Therneau test can be used for an overall
test of the proportional hazards assumption.
Once a suitable model has been determined, and satisfactory results have
been obtained from the model checking process, the final stage in the model
building process is to review the model critically, in conjunction with subject
matter experts. This may lead to further revisions to the model to ensure that
it will be accepted as an appropriate summary of the underlying survival data.
4.6 Further reading

General introductions to model checking in linear models are included in
Draper and Smith (1998) and Montgomery, Peck and Vining (2012). Cook
and Weisberg (1982) give a more detailed account of the theory underlying
residuals and influence diagnostics in a number of situations. Atkinson (1985)
describes model checking in linear models from a practical viewpoint, and
McCullagh and Nelder (1989) and Aitkin et al. (1989) discuss this topic in
the context of generalised linear models.
Many textbooks devoted to the analysis of survival data, and particularly
those of Cox and Oakes (1984), Hosmer, Lemeshow and May (2008), Lawless
(2002), and Kalbfleisch and Prentice (2002), include sections on the use of
residuals. Hinkley, Reid and Snell (1991) and Hastie and Tibshirani (1990)
also include brief discussions on methods for assessing the adequacy of models
fitted to survival data.
Early articles on the use of residuals in checking the adequacy of survival
models include Kay (1977) and Crowley and Hu (1977). These papers include a
discussion on the Cox-Snell residuals, which are based on the general definition
of residuals given by Cox and Snell (1968). Crowley and Storer (1983) showed
empirically that the cumulative hazard plot of the residuals is not particularly
good at identifying inadequacies in the fitted model. See also Crowley and
Storer (1983) for a practical application of the methods. Reviews of diagnostic
procedures in survival analysis were given in the mid-1980s by Kay (1984) and
Day (1985).
Martingale residuals were proposed by Barlow and Prentice (1988). Es-
sentially the same residuals were proposed by Lagakos (1981) and their use
is discussed by Therneau, Grambsch and Fleming (1990) and Henderson and
Milner (1991). Deviance residuals were also introduced in Therneau, Gramb-
sch and Fleming (1990). The Schoenfeld residuals for the Cox model were
proposed by Schoenfeld (1982). In accounts of survival analysis based on the
theory of counting processes, Fleming and Harrington (2005) and Therneau
and Grambsch (2000) show how different types of residual can be used, and
give detailed practical examples. Two other types of residual, introduced by
Nardi and Schemper (1999), are particularly suitable for the detection of out-
lying survival times.
Influence diagnostics for the Cox regression model have been considered by
many authors, but the major papers are those of Cain and Lange (1984), Reid
and Crépeau (1985), Storer and Crowley (1985), Pettitt and Bin Daud (1989)
and Weissfeld (1990). Pettitt and Bin Daud (1990) show how time-dependence
in the Cox proportional hazards model can be detected by smoothing the
Schoenfeld residuals. The LOWESS smoother was introduced by Cleveland
(1979), and the algorithm is also presented in Collett (2003).
Some other graphical methods for evaluating survival models, not men-
tioned in this chapter, have been proposed by Cox (1979) and Arjas (1988).
Gray (1990) describes the use of smoothed estimates of cumulative hazard
functions in evaluating the fit of a Cox model.
Most of the diagnostic procedures presented in this chapter rely on an in-
formal evaluation of tabular or graphical presentations of particular statistics.
In addition to these procedures, a variety of significance tests have been pro-
posed that can be used to assess the goodness of fit of the model. Examples
include the methods of Schoenfeld (1980), Andersen (1982), Nagelkerke, Oost-
ing and Hart (1984), Ciampi and Etezadi-Amoli (1985), Moreau, O’Quigley
and Mesbah (1985), Gill and Schumacher (1987), O’Quigley and Pessione
(1989), Quantin et al. (1996), Grønnesby and Borgan (1996), and Verweij,
van Houwelingen and Stijnen (1998). Reviews of some of these goodness of
fit tests for the Cox regression model are included in Lin and Wei (1991) and
Quantin et al. (1996). Many of these tests involve statistics that are quite
complicated, and the procedures are not widely available in computer soft-
ware for survival analysis. A more simple procedure for evaluating the overall
fit of a model has been proposed by May and Hosmer (1998).
Chapter 5
Parametric proportional hazards

models
When the Cox regression model is used in the analysis of survival data, there
is no need to assume a particular form of probability distribution for the sur-
vival times. As a result, the hazard function is not restricted to a specific
functional form, and the model has flexibility and widespread applicability.
On the other hand, if the assumption of a particular probability distribution
for the data is valid, inferences based on such an assumption will be more pre-
cise. In particular, estimates of quantities such as relative hazards and median
survival times will tend to have smaller standard errors than they would in
the absence of a distributional assumption. Models in which a specific proba-
bility distribution is assumed for the survival times are known as parametric
models, and parametric versions of the proportional hazards model, described
in Chapter 3, are the subject of this chapter.
A probability distribution that plays a central role in the analysis of sur-
vival data is the Weibull distribution, introduced by W. Weibull in 1951 in the
context of industrial reliability testing. Indeed, this distribution is as central
to the parametric analysis of survival data as the normal distribution is in lin-
ear modelling. Proportional hazards models based on the Weibull distribution
are therefore considered in some detail.
5.1 Models for the hazard function

Once a distributional model for survival times has been specified in terms of a
probability density function, the corresponding survivor and hazard functions
can be obtained from the relations
∫ t
S(t) = 1 − f (u) du,
0
and
f (t) d
h(t) = = − {log S(t)},
S(t) dt
where f (t) is the probability density function of the survival times. These
relationships were derived in Section 1.3. An alternative approach is to specify
171
172 PARAMETRIC PROPORTIONAL HAZARDS MODELS
a functional form for the hazard function, from which the survivor function
and probability density functions can be determined from the equations
S(t) = exp {−H(t)} , (5.1)
and
dS(t)
f (t) = h(t)S(t) = − ,
dt
where ∫ t
H(t) = h(u) du
0
is the integrated hazard function.
5.1.1 The exponential distribution

The simplest model for the hazard function is to assume that it is constant
over time. The hazard of death at any time after the time origin of the study
is then the same, irrespective of the time that has elapsed. Under this model,
the hazard function may be written as
h(t) = λ,
for 0 6 t < ∞. The parameter λ is a positive constant that can be estimated

by fitting the model to observed survival data. From Equation (5.1), the cor-
responding survivor function is
{ ∫ t }
S(t) = exp − λ du ,
0
= e−λt , (5.2)
and so the implied probability density function of the survival times is
f (t) = λe−λt , (5.3)
for 0 6 t < ∞. This is the probability density function of a random variable

T that has an exponential distribution with a mean of λ−1 . It is sometimes
convenient to write µ = λ−1 , so that the hazard function is µ−1 , and the
survival time distribution has a mean of µ. However, the former specification
of the hazard function will generally be used in this book.
The median of the exponential distribution, t(50), is such that S{t(50)} =
0.5, that is,
exp{−λt(50)} = 0.5,
so that
1
t(50) = log 2.
λ
MODELS FOR THE HAZARD FUNCTION 173
More generally, the pth percentile of the survival time distribution is the value
t(p) such that S{t(p)} = 1 − (p/100), and using Equation (5.2), this is
( )
1 100
t(p) = log .
λ 100 − p
A plot of the hazard function for three values of λ, namely 1.0, 0.1 and 0.01,
is given in Figure 5.1, and the corresponding probability density functions are
shown in Figure 5.2. For these values of λ, the means of the corresponding
exponential distributions are 1, 10 and 100, and the median survival times are
0.69, 6.93 and 69.31, respectively.
1.0 =1.0
0.8
Hazard function
0.6
0.4
0.2
=0.1
0.0 =0.01
0 20 40 60 80 100
Time
Figure 5.1 Hazard functions for exponential distributions with λ = 1.0, 0.1 and 0.01.
5.1.2 The Weibull distribution

In practice, the assumption of a constant hazard function, or equivalently
of exponentially distributed survival times, is rarely tenable. A more general
form of hazard function is such that
h(t) = λγtγ−1 , (5.4)
for 0 6 t < ∞, a function that depends on two parameters λ and γ, which

are both greater than zero. In the particular case where γ = 1, the hazard
function takes a constant value λ, and the survival times have an exponential
distribution. For other values of γ, the hazard function increases or decreases
monotonically, that is, it does not change direction. The shape of the hazard
function depends critically on the value of γ, and so γ is known as the shape
0.5
0.4 =1.0
Probability density function
0.3
0.2
0.1
=0.1
0.0 =0.01
0 1 2 3 4 5 6 7 8 9 10
Time
Figure 5.2 Probability density functions for exponential distributions with λ = 1.0,
0.1 and 0.01.
parameter, while the parameter λ is a scale parameter. The general form of

this hazard function for different values of γ is shown in Figure 5.3.
For this particular choice of hazard function, the survivor function is given
by { ∫ } t
S(t) = exp − λγuγ−1 du = exp(−λtγ ). (5.5)
0
The corresponding probability density function is then
f (t) = λγtγ−1 exp(−λtγ ),
for 0 6 t < ∞, which is the density of a random variable that has a Weibull
distribution with scale parameter λ and shape parameter γ. This distribution
will be denoted W (λ, γ). The right-hand tail of this distribution is longer than
the left-hand one, and so the distribution is positively skewed.
The mean, or expected value, of a random variable T that has a W (λ, γ)
distribution can be shown to be given by
E (T ) = λ−1/γ Γ(γ −1 + 1),
where Γ(x) is the gamma function defined by the integral

∫ ∞
Γ(x) = ux−1 e−u du.
0
The value of this integral is (x − 1)!, and so for integer values of x it can easily
MODELS FOR THE HAZARD FUNCTION 175
Hazard function >2
=2
=1
0< <1
0
0
Time
Figure 5.3 The form of the Weibull hazard function, h(t) = λγtγ−1 , for different
values of γ.
be calculated. The function is also defined for non-integer values of x, and can
then be evaluated as a standard function in many software packages.
Since the Weibull distribution is skewed, a more appropriate, and more
tractable, summary of the location of the distribution is the median survival
time. This is the value t(50) such that S{t(50)} = 0.5, so that
exp {−λ[t(50)]γ } = 0.5,
and { }1/γ
1
t(50) = log 2 .
λ
More generally, the pth percentile of the Weibull distribution, t(p), is such
that { ( )}1/γ
1 100
t(p) = log . (5.6)
λ 100 − p
The median and other percentiles of the Weibull distribution are therefore
much simpler to compute than the mean of the distribution.
The hazard function and corresponding probability density function for
Weibull distributions with a median of 20, and shape parameters γ = 0.5, 1.5
and 3.0, are shown in Figures 5.4 and 5.5, respectively. The corresponding
value of the scale parameter, λ, for these three Weibull distributions is 0.15,
0.0078 and 0.000087, respectively.
Since the Weibull hazard function can take a variety of forms, depending on
the value of the shape parameter, γ, and appropriate summary statistics can
0.20 =3.0
0.15
Hazard function
0.10
=1.5
0.05
=0.5
0.00
0 10 20 30 40
Time
Figure 5.4 Hazard functions for a Weibull distribution with a median of 20 and γ =
0.5, 1.5 and 3.0.
0.06
0.05
Probability density function
0.04
0.03
0.02
0.01 =1.5
=0.5
0.00 =3.0
0 10 20 30 40
Time
Figure 5.5 Probability density functions for a Weibull distribution with a median of
20 and γ = 0.5, 1.5 and 3.0.
ASSESSING THE SUITABILITY OF A PARAMETRIC MODEL 177
be easily obtained, this distribution is widely used in the parametric analysis
of survival data.
5.2 Assessing the suitability of a parametric model

Prior to fitting a model based on an assumed parametric form for the hazard
function, a preliminary study of the validity of this assumption should be
carried out. One approach would be to estimate the hazard function using
the methods outlined in Section 2.3. If the hazard function was reasonably
constant over time, this would indicate that the exponential distribution might
be a suitable model for the data. On the other hand, if the hazard function
increased or decreased monotonically with increasing survival time, a model
based on the Weibull distribution would be indicated.
A more informative way of assessing whether a particular distribution for
the survival times is plausible is to compare the survivor function for the
data with that of a chosen model. This is greatly helped by transforming
the survivor function to produce a plot that should give a straight line if the
assumed model is appropriate.
Suppose that a single sample of survival data is available, and that a
Weibull distribution for the survival times is contemplated. Since the sur-
vivor function for a Weibull distribution, with scale parameter λ and shape
parameter γ, is given by
S(t) = exp {−λtγ } ,
taking the logarithm of S(t), multiplying by −1, and taking logarithms a
second time, gives
log {− log S(t)} = log λ + γ log t. (5.7)
We now substitute the Kaplan-Meier estimate of the survivor function, Ŝ(t),
for S(t) in Equation (5.7). If the Weibull assumption is tenable, Ŝ(t) will be
‘close’ to S(t), and a plot of log{− log Ŝ(t)} against log t would then give an
approximately straight line. From Equation (1.8), the cumulative hazard func-
tion, H(t), is − log S(t) and so log{− log S(t)} is the log-cumulative hazard.
A plot of the values of log{− log Ŝ(t)} against log t is a log-cumulative hazard
plot, introduced in Section 4.4.1 of Chapter 4.
If the log-cumulative hazard plot gives a straight line, the plot can be used
to provide a rough estimate of the two parameters of the Weibull distribution.
Specifically, from Equation (5.7), the intercept and slope of the straight line
will be log λ and γ, respectively. Thus, the slope of the line in a log-cumulative
hazard plot gives an estimate of the shape parameter, and the exponent of the
intercept provides an estimate of the scale parameter. Note that if the slope
of the log-cumulative hazard plot is close to unity, the survival times could be
modelled using an exponential distribution.
In Example 2.3, the Kaplan-Meier estimate of the survivor function, Ŝ(t), for
the data on the time to discontinuation of an intrauterine device was obtained.
A log-cumulative hazard plot for these data, that is, a plot of log{− log Ŝ(t)}
against log t, is shown in Figure 5.6.
0
-1
-2
-3
2.0 2.5 3.0 3.5 4.0 4.5 5.0
Log of discontinuation time
Figure 5.6 Log-cumulative hazard plot for the data from Example 1.1.
The plot indicates that there is a straight line relationship between the
log-cumulative hazard and log t, confirming that the Weibull distribution is
an appropriate model for the discontinuation times. From the graph, the in-
tercept of the line is approximately −6.0 and the slope is approximately 1.25.
Approximate estimates of the parameters of the Weibull distribution are there-
fore λ∗ = exp(−6.0) = 0.002 and γ ∗ = 1.25. The estimated value of γ, the
shape parameter of the Weibull distribution, is quite close to unity, suggesting
that the discontinuation times might be adequately modelled by an exponen-
tial distribution.
These informal estimates of λ and γ can be used to estimate the parameters

of the distribution, and hence functions of these estimates, such as the me-
dian of the survival time distribution. However, this graphical approach does
not lead to a measure of the precision with which the quantities have been
estimated. In view of this limitation, a more formal way of fitting parametric
models to survival data is developed in the next section.
5.3 Fitting a parametric model to a single sample

Parametric models can be fitted to an observed set of survival data using
the method of maximum likelihood, outlined in Section 3.3. Consider first the
situation where actual survival times have been observed for n individuals, so
FITTING A PARAMETRIC MODEL TO A SINGLE SAMPLE 179
that there are no censored observations. If the probability density function of
the random variable associated with survival time is f (t), the likelihood of the
n observations t1 , t2 , . . . , tn is simply the product
∏
n
f (ti ).
i=1
This likelihood will be a function of the unknown parameters in the probability

density function, and the maximum likelihood estimates of these parameters
are those values for which the likelihood function is a maximum. In practice,
it is generally more convenient to work with the logarithm of the likelihood
function. Those values of the unknown parameters in the density function that
maximise the log-likelihood are of course the same values that maximise the
likelihood function itself.
We now consider the more usual situation where the survival data includes
one or more censored survival times. Specifically, suppose that r of the n indi-
viduals die at times t1 , t2 , . . . , tr , and that the survival times of the remaining
n − r individuals, t∗1 , t∗2 , . . . , t∗n−r , are right-censored. The r death times con-
tribute a term of the form
∏r
f (tj )
j=1
to the overall likelihood function. Naturally, we cannot ignore information

about the survival experience of the n − r individuals for whom a censored
survival time has been recorded. If a survival time is censored at time t∗ , say,
we know that the lifetime of the individual is at least t∗ , and the probability
of this event is P(T > t∗ ), which is S(t∗ ). Thus, each censored observation
contributes a term of this form to the likelihood of the n observations. The
total likelihood function is therefore
∏
r ∏
n−r
f (tj ) S(t∗l ), (5.8)
j=1 l=1
in which the first product is taken over the r death times and the second over
the n − r censored survival times.
More compactly, suppose that the data are regarded as n pairs of obser-
vations, where the pair for the ith individual is (ti , δi ), i = 1, 2, . . . , n. In this
notation, δi is an indicator variable that takes the value zero when the sur-
vival time ti is censored, and unity when ti is an uncensored survival time.
The likelihood function can then be written as
∏
n
δ 1−δi
{f (ti )} i {S(ti )} . (5.9)
i=1
This function, which is equivalent to that in Expression (5.8), can then be

maximised with respect to the unknown parameters in the density and sur-
vivor functions.
An alternative expression for this likelihood function can be obtained by
writing Expression (5.9) in the form
n {
∏ }δ
f (ti ) i
S(ti ),
i=1
S(ti )
so that, from Equation (1.4) of Chapter 1, this becomes
∏
n
δ
{h(ti )} i S(ti ). (5.10)
i=1
This version of the likelihood function is particularly useful when the probabil-
ity density function has a complicated form, as it often does. Estimates of the
unknown parameters in this likelihood function are then found by maximising
the logarithm of the likelihood function.
5.3.1 ∗ Likelihood function for randomly censored data

A more careful derivation of the likelihood function in Equation (5.9) is given
in this section, which shows the relevance of the assumption of independent
censoring, referred to in Section 1.1.2 of Chapter 1.
Suppose that survival data for a sample of n individuals is a mixture of
event times and right-censored observations. Denote the observed time for
the ith individual by ti , and let δi be the corresponding event indicator, i =
1, 2, . . . , n, so that δi = 1 if ti is an event time, and δi = 0 if the time is
censored.
The random variable associated with the event time of the ith individual
will be denoted by Ti . The censoring times will be assumed to be random,
and Ci will denote the random variable associated with the time to censoring.
The value ti is then an observation on the random variable τi = min(Ti , Ci ).
The density and survivor functions of Ti will be denoted by fTi (t) and STi (t),
respectively. Also, fCi (t) and SCi (t) will be used to denote the density and
survivor functions of the random variable associated with the censoring time,
Ci .
We now consider the probability distribution of the pair (τi , δi ) for censored
and uncensored observations, respectively. Consider first the case of a censored
observation, so that δi = 0. The joint distribution of τi and δi is described by
P(τi = t, δi = 0) = P(Ci = t, Ti > t).
This joint probability is a mixture of continuous and discrete components, but

to simplify the presentation, P(Ti = t), for example, will be understood to be
the probability density function of Ti . The distribution of the event time, Ti ,
FITTING EXPONENTIAL AND WEIBULL MODELS 181
is now assumed to be independent of that of the censoring time, Ci . Then,
P(Ci = t, Ti > t) = P(Ci = t) P(Ti > t),

= fCi (t)STi (t),
so that
P(τi = t, δi = 0) = fCi (t)STi (t).
Similarly, for an uncensored observation,
P(τi = t, δi = 1) = P(Ti = t, Ci > t),

= P(Ti = t) P(Ci > t),
= fTi (t)SCi (t),
again assuming that the distributions of Ci and Ti are independent. Putting

these two results together, the joint probability, or likelihood, of the n obser-
vations, t1 , t2 , . . . , tn , is therefore
∏
n
{fTi (ti )SCi (ti )}δi {fCi (ti )STi (ti )}1−δi ,
i=1

∏
n ∏
n
fCi (ti )1−δi SCi (ti )δi × fTi (ti )δi STi (ti )1−δi .
i=1 i=1
On the assumption of independent censoring, the first product in this expres-

sion will not involve any parameters that are relevant to the distribution of
the survival times, and so can be regarded as a constant. The likelihood of
the observed data is then proportional to
∏
n
fTi (ti )δi STi (ti )1−δi ,
i=1
which was given in Expression (5.9) of this chapter.

It can also be shown that when the study has a fixed duration, so that
individuals who have not experienced an event by the end of the study are
censored, the same likelihood function is obtained. Details are not given here,
but see Klein and Moeschberger (2005) or Lawless (2002), for example.
5.4 ∗ Fitting exponential and Weibull models

We now consider fitting exponential and Weibull distributions to a single
sample of survival data.
5.4.1 Fitting the exponential distribution
Suppose that the survival times of n individuals, t1 , t2 , . . . , tn , are assumed
to have an exponential distribution with mean λ−1 . Further suppose that the
data give the actual death times of r individuals, and that the remaining n − r
survival times are right-censored.
For the exponential distribution,
f (t) = λe−λt , S(t) = e−λt ,
and on substituting into Expression (5.9), the likelihood function for the n
observations is given by
∏
n
( )δi ( )1−δi
L(λ) = λe−λti e−λti ,
i=1
where δi is zero if the survival time of the ith individual is censored and unity
otherwise. After some simplification,
∏
n
L(λ) = λδi e−λti ,
i=1
and the corresponding log-likelihood function is

∑
n ∑
n
log L(λ) = δi log λ − λ ti .
i=1 i=1
∑n
Since the data contain r deaths, i=1 δi = r and the log-likelihood function
becomes
∑
n
log L(λ) = r log λ − λ ti .
i=1
We now need to identify the value λ̂, for which the log-likelihood function is
a maximum. Differentiation with respect to λ gives
r ∑
n
d log L(λ)
= − ti ,
dλ λ i=1
and equating the derivative to zero and evaluating it at λ̂ gives

∑
n
λ̂ = r/ ti (5.11)
i=1
for the maximum likelihood estimator of λ.

The mean of an exponential distribution is µ = λ−1 , and so the maximum
likelihood estimator of µ is
1∑
n
µ̂ = λ̂−1 = ti .
r i=1
This estimator of µ is the total time survived by the n individuals in the data
set divided by the number of deaths observed. The estimator therefore has
intuitive appeal as an estimate of the mean lifetime from censored survival
data.
The standard error of either λ̂ or µ̂ can be obtained from the second
derivative of the log-likelihood function, using a result from the theory of
maximum likelihood estimation given in Appendix A. Differentiating log L(λ)
a second time gives
d2 log L(λ) r
= − 2,
dλ2 λ
and so the asymptotic variance of λ̂ is
{ ( 2 )}−1
d log L(λ) λ2
var (λ̂) = −E = .
dλ2 r
Consequently, the standard error of λ̂ is given by

√
se (λ̂) = λ̂/ r. (5.12)
This result could be used to obtain a confidence interval for the mean survival
time. In particular, the limits of a 100(1 − α)% confidence interval for λ are
λ̂ ± zα/2 se (λ̂), where zα/2 is the upper α/2-point of the standard normal
distribution.
In presenting the results of a survival analysis, the estimated survivor and
hazard functions, and the median and other percentiles of the distribution of
survival times, are useful. Once an estimate of λ has been found, all these func-
tions can be estimated using the results given in Section 5.1.1. In particular,
under the assumed exponential distribution, the estimated hazard function is
ĥ(t) = λ̂ and the estimated survivor function is Ŝ(t) = exp(−λ̂t). In addition,
the estimated pth percentile is given by
( )
1 100
t̂(p) = log , (5.13)
λ̂ 100 − p
and the estimated median survival time is
t̂(50) = λ̂−1 log 2. (5.14)
The standard error of an estimate of the pth percentile of the distribution

of survival times can be found using the result for the approximate variance
of a function of a random variable given in Equation (2.8) of Chapter 2.
According to this result, an approximation to the variance of a function g(λ̂)
of λ̂ is such that { }2
dg(λ̂)
var {g(λ̂)} ≈ var (λ̂). (5.15)
dλ̂
Using this result, the approximate variance of the estimated pth percentile is
given by
{ ( )}2
1 100
var {t̂(p)} ≈ − log var (λ̂).
λ̂2 100 − p
Taking the square root, we get
( )
1 100
se {t̂(p)} = log se (λ̂),
λ̂2 100 − p
and on substituting for se (λ̂) from Equation (5.12) and t̂(p) from Equa-
tion (5.13), we find
√
se {t̂(p)} = t̂(p)/ r. (5.16)
In particular, the standard error of the estimated median survival time is
√
se {t̂(50)} = t̂(50)/ r. (5.17)
Confidence intervals for a true percentile are best obtained from exponenti-
ating the confidence limits for the logarithm of the percentile. This procedure
ensures that confidence limits for the percentile will be non-negative. Again
making use of the result in Equation (5.15), the standard error of log t̂(p) is
given by
se {log t̂(p)} = t̂(p)−1 se {t̂(p)},
and after substituting for se {t̂(p)} from Equation (5.16), this standard error
becomes
√
se {log t̂(p)} = 1/ r.
√ − α)% confidence limits √

Using this result, 100(1 for the 100pth percentile are
exp{log t̂(p) ± zα/2 / r}, that is, t̂(p) exp{±zα/2 / r}, where zα/2 is the upper
α/2-point of the standard normal distribution.

In this example, the data of Example 1.1 on the times to discontinuation
of an IUD for 18 women are analysed under the assumption of a constant
hazard of discontinuation. An exponential distribution is therefore fitted to
the discontinuation times. For these data, the total of the observed and right-
censored discontinuation times is 1046 days, and the number of uncensored
times is 9. Therefore, using Equation (5.11), λ̂ = 9/1046 = 0.0086, and the
√
standard error of λ̂ from Equation (5.12) is se (λ̂) = 0.0086/ 9 = 0.0029. The
estimated hazard function is therefore ĥ(t) = 0.0086, t > 0, and the estimated
survivor function is Ŝ(t) = exp(−0.0086 t). The estimated hazard and survivor
functions are shown in Figures 5.7 and 5.8, respectively.
Estimates of the median and other percentiles of the distribution of dis-
continuation times can be found from Figure 5.8, but more accurate estimates
are obtained from Equation (5.13). In particular, using Equation (5.14), the
0.010
0.008
0.006
0.004
0.002
0.000
0 20 40 60 80 100 120
Figure 5.7 Estimated hazard function on fitting the exponential distribution.
1.0
0.8
0.6
0.4
0.2
0.0
0 20 40 60 80 100 120
Figure 5.8 Estimated survivor function on fitting the exponential distribution.

median discontinuation time is 81 days, and an estimate of the 90th per-
centile of the distribution of discontinuation times is, from Equation (5.13),
t̂(90) = log 10/0.0086 = 267.61. This means that on the assumption that the
risk of discontinuing the use of an IUD is independent of time, 90% of women
will have a discontinuation time of less than 268 days.
From Equation (5.17),√ the standard error of the estimated median time to
discontinuation is 80.56/ 9, that is, 26.85 days. The limits of a 95% confidence
interval for the true median discontinuation time are
√
80.56 exp{±1.96/ 9},
and so the interval is from 42 days to 155 days. Confidence intervals for other
percentiles can be calculated in a similar manner.
5.4.2 Fitting the Weibull distribution

The survival times of n individuals are now taken to be a censored sample
from a Weibull distribution with scale parameter λ and shape parameter γ.
Suppose that there are r deaths among the n individuals and n − r right-
censored survival times. We can again use Expression (5.9) to obtain the
likelihood of the sample data. The probability density, survivor and hazard
function of a W (λ, γ) distribution are given by
f (t) = λγtγ−1 exp(−λtγ ), S(t) = exp(−λtγ ), h(t) = λγtγ−1 ,
and so, from Expression (5.9), the likelihood of the n survival times is
n {
∏ }δi
1−δ
λγtγ−1
i exp(−λt γ
i ) {exp(−λtγi )} i ,
i=1
where δi is zero if the ith survival time is censored and unity otherwise. Equiv-
alently, from Expression (5.10), the likelihood function is
n {
∏ }δi
λγtγ−1
i exp(−λtγi ).
i=1
This is regarded as a function of λ and γ, the unknown parameters in the

Weibull distribution, and so can be written L(λ, γ). The corresponding log-
likelihood function is given by
∑
n ∑
n ∑
n
log L(λ, γ) = δi log(λγ) + (γ − 1) δi log ti − λ tγi ,
i=1 i=1 i=1
∑n
and noting that i=1 δi = r, the log-likelihood becomes
∑
n ∑
n
log L(λ, γ) = r log(λγ) + (γ − 1) δi log ti − λ tγi .
i=1 i=1
The maximum likelihood estimates of λ and γ are found by differentiating
this function with respect to λ and γ, equating the derivatives to zero, and
evaluating them at λ̂ and γ̂. The resulting equations are
r ∑
n
− tγ̂i = 0, (5.18)
λ̂ i=1
and
r ∑ ∑
n n
+ δi log ti − λ̂ tγ̂i log ti = 0. (5.19)
γ̂ i=1 i=1
From Equation (5.18),

∑
n
λ̂ = r/ tγ̂i , (5.20)
i=1
and on substituting for λ̂ in Equation (5.19), we get the equation
r ∑ r ∑ γ̂
n n
+ δi log ti − ∑ γ̂ ti log ti = 0. (5.21)
γ̂ i=1 i ti i=1
This is a non-linear equation in γ̂, which can only be solved numerically using
an iterative procedure. Once the estimate, γ̂, which satisfies Equation (5.21),
has been found, Equation (5.20) can be used to obtain λ̂.
In practice, a numerical method, such as the Newton-Raphson procedure,
is used to find the values λ̂ and γ̂ which maximise the likelihood function
simultaneously. This procedure was described in Section 3.3.3 of Chapter 3,
in connection with fitting the Cox regression model. In that section it was
noted that an important by-product of the Newton-Raphson procedure is an
approximation to the variance-covariance matrix of the parameter estimates,
from which their standard errors can be obtained.
Once estimates of the parameters λ and γ have been found from fitting the
Weibull distribution to the observed data, percentiles of the survival time dis-
tribution can be estimated using Equation (5.6). The estimated pth percentile
of the distribution is
{ ( )}1/γ̂
1 100
t̂(p) = log , (5.22)
λ̂ 100 − p
and so the estimated median survival time is given by

{ }1/γ̂
1
t̂(50) = log 2 . (5.23)
λ̂
An expression for the standard error of a percentile of the Weibull distribution,
and a corresponding confidence interval, is derived in Section 5.4.3.
5.4.3 Standard error of a percentile of the Weibull distribution
The standard error of the estimated pth percentile of the Weibull distribution
with scale parameter λ and shape parameter γ, t̂(p), is most easily found from
the variance of log t̂(p). Now, from Equation (5.22),
{ ( )}
1 100
log t̂(p) = log λ̂−1 log ,
γ̂ 100 − p
and so
1{ }
log t̂(p) = cp − log λ̂ ,
γ̂
where ( )
100
cp = log log .
100 − p
This is a function of two parameter estimates, λ̂ and γ̂.
To obtain the variance of log t̂(p), we use the general result that the ap-
proximate variance of a function g(θ̂1 , θ̂2 ) of two parameter estimates, θ̂1 , θ̂2 ,
is
( )2 ( )2 ( )
∂g ∂g ∂g ∂g
var (θ̂1 ) + var (θ̂2 ) + 2 cov (θ̂1 , θ̂2 ). (5.24)
∂ θ̂1 ∂ θ̂2 ∂ θ̂1 ∂ θ̂2
This is an extension of the result given in Equation (2.8) of Chapter 2 for
the approximate variance of a function of a single random variable. Using
Equation (5.24),
( )2 ( )2
{ } ∂ log t̂(p) ∂ log t̂(p)
var log t̂(p) ≈ var (λ̂) + var (γ̂)
∂ λ̂ ∂γ̂
∂ log t̂(p) ∂ log t̂(p)
+2 cov (λ̂, γ̂).
∂ λ̂ ∂γ̂
Now, the derivatives of log t̂(p) with respect to λ̂ and γ̂ are given by
∂ log t̂(p) 1
= − ,
∂ λ̂ λ̂γ̂
∂ log t̂(p) cp − log λ̂
= − ,
∂γ̂ γ̂ 2
and so the approximate variance of log t̂(p) is
( )2 ( )
1 cp − log λ̂ 2 cp − log λ̂
var (λ̂) + var (γ̂) + cov (λ̂, γ̂). (5.25)
λ̂ γ̂ 2
2 γ̂ 4 λ̂γ̂ 3
The variance of t̂(p) itself is found from the result in Equation (2.8) of Chapter
2, from which
var {t̂(p)} ≈ t̂(p)2 var {log t̂(p)}, (5.26)
and using Expression (5.25),
{ ( )2
t̂(p)2
var {t̂(p)} ≈ γ̂ 2 var (λ̂) + λ̂2 cp − log λ̂ var (γ̂)
λ̂2 γ̂ 4
( ) }
+ 2λ̂γ̂ cp − log λ̂ cov (λ̂, γ̂) .
The standard error of t̂(p) is the square root of this expression, given by
{ ( )2
{ } t̂(p)
se t̂(p) = γ̂ 2 var (λ̂) + λ̂2 cp − log λ̂ var (γ̂)
λ̂γ̂ 2
( ) } 12
+ 2λ̂γ̂ cp − log λ̂ cov (λ̂, γ̂) . (5.27)
Note that for the special case of the exponential distribution, where the
shape parameter, γ, is equal to unity, the standard error of the estimated pth
percentile from Equation (5.27) is
t̂(p)
se (λ̂).
λ̂
Now, using Equation (5.12) of Chapter 5,
√
se (λ̂) = λ̂/ r,
where r is the number of death times in the data set, and so

√
se {t̂(p)} = t̂(p)/ r,
as in Equation (5.16).
A 100(1 − α)% confidence interval for the pth percentile of a Weibull dis-
tribution is found from the corresponding confidence limits for log t(p). These
limits are
log t̂(p) ± zα/2 se {log t̂(p)},
where se {log t̂(p)} is, from Equation (5.26), given by
1
se {log t̂(p)} = se {t̂(p)}, (5.28)
t̂(p)
and zα/2 is the upper α/2-point of the standard normal distribution. The
corresponding[ 100(1 − α)% confidence
] interval for the pth percentile, t(p), is
then t̂(p) exp ±zα/2 se {log t̂(p)} .
In Example 5.1, it was found that an exponential distribution provides a
satisfactory model for the data on the discontinuation times of 18 IUD users.
For comparison, a Weibull distribution will be fitted to the same data set. This
distribution is fitted using computer software, and from the resulting output,
the estimated scale parameter of the distribution is found to be λ̂ = 0.000454,
while the estimated shape parameter is γ̂ = 1.676. The standard errors of these
estimates are given by se (λ̂) = 0.000965 and se (γ̂) = 0.460, respectively.
Note that approximate confidence limits for the shape parameter, γ, found
using γ̂±1.96 se (γ̂), include unity, suggesting that the exponential distribution
would provide a satisfactory model for the discontinuation times.
The estimated hazard and survivor functions are obtained by substituting
these estimates into Equations (5.4) and (5.5), whence
ĥ(t) = λ̂γ̂tγ̂−1 ,
and ( )
Ŝ(t) = exp −λ̂tγ̂ .
These two functions are shown in Figures 5.9 and 5.10.
0.020
0.015
0.010
0.005
0.000
0 20 40 60 80 100 120
Figure 5.9 Estimated hazard function on fitting the Weibull distribution.
Although percentiles of the discontinuation time can be read from the

estimated survivor function in Figure 5.10, they are better estimated using
Equation (5.22). Hence, under the Weibull distribution, the median discon-
tinuation time can be estimated using Equation (5.23), and is given by
{ }1/1.676
1
t̂(50) = log 2 = 79.27.
0.000454
As a check, notice that this is perfectly consistent with the value of the dis-
continuation time corresponding to Ŝ(t) = 0.5 in Figure 5.10. The standard
1.0
0.8
0.6
0.4
0.2
0.0
0 20 40 60 80 100 120
Figure 5.10 Estimated survivor function on fitting the Weibull distribution.
error of this estimate, from Equation (5.27) is, after much arithmetic, found
to be
se {t̂(50)} = 15.795.
In order to obtain a 95% confidence interval for the median discontinuation
time, the standard error of log t̂(50) is required. From Equation (5.28),
15.795
se {log t̂(50)} = = 0.199,
79.272
and so the required confidence limits for the log median discontinuation
time are log 79.272 ± 1.96 × 0.199, that is, (3.982, 4.763). The correspond-
ing interval estimate for the true median discontinuation time, found from
exponentiating these limits, is (53.64, 117.15). This means that there is
a 95% chance that the interval from 54 days to 117 days includes the
true value of the median discontinuation time. This interval is rather
wide because of the small number of actual discontinuation times in the
data set.
It is interesting to compare these results with those found in Exam-
ple 5.2, where the discontinuation times were modelled using an exponen-
tial distribution. The estimated median survival times are very similar, at
80.6 days for the exponential and 79.3 days for the Weibull model. How-
ever, the standard error of the estimated median survival time is 26.8 days
when the times are assumed to have an exponential distribution, and only
15.8 days under the Weibull model. The median is therefore estimated more
precisely when the discontinuation times are assumed to have a Weibull
distribution.
Other percentiles of the discontinuation time distribution, and accompany-
ing standard errors and confidence intervals, can be found in a similar fashion.
For example, the 90th percentile, that is, the time beyond which 10% of those
in the study continue with the use of the IUD, is 162.23 days, and 95% confi-
dence limits for the true percentile are from 95.41 to 275.84 days. Notice that
the width of this confidence interval is larger than that for the median discon-
tinuation time, reflecting the fact that the median is more precisely estimated
than other percentiles.
5.5 A model for the comparison of two groups

We saw in Section 3.1 that a convenient general model for comparing two
groups of survival times is the proportional hazards model. Here, the two
groups will be labelled Group I and Group II, and X will be an indicator
variable that takes the value zero if an individual is in Group I and unity if an
individual is in Group II. Under the proportional hazards model, the hazard
of death at time t for the ith individual is given by
hi (t) = eβxi h0 (t), (5.29)
where xi is the value of X for the ith individual. Consequently, the hazard
at time t for an individual in Group I is h0 (t), and that for an individual in
Group II is ψh0 (t), where ψ = exp(β). The quantity β is then the logarithm of
the ratio of the hazard for an individual in Group II, to that of an individual
in Group I.
We will now make the additional assumption that the survival times for
the individuals in Group I have a Weibull distribution with scale parame-
ter λ and shape parameter γ. Using Equation (5.29), the hazard function for
the individuals in this group is h0 (t), where h0 (t) = λγtγ−1 . Now, also from
Equation (5.29), the hazard function for those in Group II is ψh0 (t), that is,
ψλγtγ−1 . This is the hazard function for a Weibull distribution with scale
parameter ψλ and shape parameter γ. We therefore have the result that if the
survival times of individuals in one group have a Weibull distribution with
shape parameter γ, and the hazard of death at time t for an individual in the
second group is proportional to that of an individual in the first, the survival
times of those in the second group will also have a Weibull distribution with
shape parameter γ. The Weibull distribution is then said to have the propor-
tional hazards property. This property is another reason for the importance of
the Weibull distribution in the analysis of survival data.
5.5.1 The log-cumulative hazard plot

When a single sample of survival times has a Weibull distribution W (λ, γ),
the log-cumulative hazard plot, described in Section 5.2, will give a straight
line with intercept log λ and slope γ. It then follows that if the survival times
A MODEL FOR THE COMPARISON OF TWO GROUPS 193
in a second group have a W (ψλ, γ) distribution, as they would under the pro-
portional hazards model in Equation (5.29), the log-cumulative hazard plot
will give a straight line, also of slope γ, but with intercept log ψ + log λ. If the
estimated log-cumulative hazard function is plotted against the logarithm of
the survival time for individuals in two groups, parallel straight lines would
mean that the assumptions of a proportional hazards model and Weibull sur-
vival times were tenable. The vertical separation of the two lines provides an
estimate of β = log ψ, the logarithm of the relative hazard.
If the two lines in a log-cumulative hazard plot are essentially straight,
but not parallel, this means that the shape parameter, γ, is different in the
two groups, and the hazards are no longer proportional. If the lines are not
particularly straight, the Weibull model may not be appropriate. However, if
the curves can be taken to be parallel, this would mean that the proportional
hazards model is valid, and the Cox regression model discussed in Chapter 3
might be more satisfactory.

In this example, we investigate whether the Weibull proportional hazards
model is likely to be appropriate for the data of Example 1.2 on the sur-
vival times of breast cancer patients. These data relate to women classified
according to whether their tumours were positively or negatively stained. The
Kaplan-Meier estimate of the survivor functions for the women in each group
were shown in Figure 2.9. From these estimates, the log-cumulative hazards
can be estimated and plotted against log t. The resulting log-cumulative haz-
ard plot is shown in Figure 5.11.
0
-1
-2
-3
-4
1 2 3 4 5 6
Figure 5.11 Log-cumulative hazard plot for women with tumours that were positively
stained (∗) and negatively stained (•).
In this figure, the lines corresponding to the two staining groups are rea-
sonably straight. This means that the assumption of Weibull distributions for
the survival times of the women in each group is quite plausible. Moreover,
the gradients of the two lines are very similar, which means that the propor-
tional hazards model is valid. The vertical separation of the two lines provides
an estimate of the log relative hazard. From Figure 5.11, the vertical distance
between the two straight lines is approximately 1.0, and so a rough estimate of
the hazard ratio is e1.0 = 2.72. Women in the positively stained group would
appear to have nearly three times the risk of death at any time compared to
those in the negatively stained group. More accurate estimates of the relative
hazard will be obtained from fitting exponential and Weibull models to the
data of this example, in Examples 5.5 and 5.6.
5.5.2 ∗ Fitting the model

The proportional hazards model in Equation (5.29) can be fitted using the
method of maximum likelihood. To illustrate the process, we consider the
situation where the survival times in each group have an exponential distri-
bution.
Suppose that the observations from n1 individuals in Group I can be ex-
pressed as (ti1 , δi1 ), i = 1, 2, . . . , n1 , where δi1 takes the value zero if the
survival time of the ith individual in that group is censored, and unity if that
time is a death time. Similarly, let (ti′ 2 , δi′ 2 ), i′ = 1, 2, . . . , n2 , be the obser-
vations from the n2 individuals in Group II. For individuals in Group I, the
hazard function will be taken to be λ, and the probability density function
and survivor function are given by
f (ti1 ) = λe−λti1 , S(ti1 ) = e−λti1 .
For those in Group II, the hazard function is ψλ, and the probability density
function and survivor function are
f (ti′ 2 ) = ψλe−ψλti′ 2 , S(ti′ 2 ) = e−ψλti′ 2 .
Using Equation (5.9), the likelihood of the n1 + n2 observations, L(ψ, λ),

is
∏
n1
{ −λti1 }δi1 { −λti1 }1−δi1 ∏
n2
{ }δ ′ { }1−δi′ 2
λe e ψλe−ψλti′ 2 i 2 e−ψλti′ 2 ,
i=1 i′ =1
which simplifies to
∏
n1 ∏
n2
λδi1 e−λti1 (ψλ)δi′ 2 e−ψλti′ 2 .
i=1 i′ =1
If the numbers of ∑
actual death times
∑ in the two groups are r1 and r2 , respec-
tively, then r1 = i δi1 and r2 = i′ δi′ 2 , and the log-likelihood function is
given by
∑
n1 ∑
n2
log L(ψ, λ) = r1 log λ − λ ti1 + r2 log(ψλ) − ψλ t i′ 2 .
i=1 i′ =1
Now write T1 and T2 for the total known time survived by the individuals in
Groups I and II, respectively. Then, T1 and T2 are the totals of uncensored
and censored survival times in each group, so that the log-likelihood function
becomes
log L(ψ, λ) = (r1 + r2 ) log λ + r2 log ψ − λ(T1 + ψT2 ).
In order to obtain the values ψ̂, λ̂ for which this function is a maximum,
we differentiate with respect to ψ and λ, and set the derivatives equal to zero.
The resulting equations that are satisfied by ψ̂, λ̂ are
r2
− λ̂T2 = 0, (5.30)
ψ̂
r1 + r2
− (T1 + ψ̂T2 ) = 0. (5.31)
λ̂
r2
λ̂ = ,
ψ̂T2
and on substituting for λ̂ in Equation (5.31) we get
r2 T1
ψ̂ = . (5.32)
r1 T2
Then, from Equation (5.30),
λ̂ = r1 /T1 .
Both of these estimates have an intuitive justification. The estimated value
of λ is the reciprocal of the average time survived by individuals in Group
I, while the estimated relative hazard, ψ̂, is the ratio of the average times
survived by the individuals in the two groups.
The asymptotic variance-covariance matrix of the parameter estimates is
the inverse of the information matrix, whose elements are found from the
second derivatives of the log-likelihood function; see Appendix A. We have
that
d2 log L(ψ, λ) r2 d2 log L(ψ, λ) r1 + r2 d2 log L(ψ, λ)
= − , = − , = −T2 ,
dψ 2 ψ2 dλ2 λ2 dλdψ
and the information matrix is the matrix of negative expected values of these
partial derivatives. The only second derivative for which expectations need
to be obtained is the derivative with respect to λ and ψ, for which E(T2 )
is required. This is straightforward when the survival times have an expo-
nential distribution, but as shown in Section 5.1.2, the expected value of a
survival time that has a Weibull distribution is much more difficult to calcu-
late. For this reason, the information matrix is usually approximated by using
the observed values of the negative second partial derivatives. The observed
information matrix is thus
( )
r2 /ψ 2 T2
I(ψ, λ) = ,
T2 (r1 + r2 )/λ2
and the inverse of this matrix is
( )
1 (r1 + r2 )ψ 2 −T2 ψ 2 λ2
.
(r1 + r2 )r2 − T22 ψ 2 λ2 −T2 ψ 2 λ2 r2 λ2
The standard errors of ψ̂ and λ̂ are found by substituting ψ̂ and λ̂ for ψ

and λ in this matrix, and taking square roots. Thus, the standard error of ψ̂
is given by √
(r1 + r2 )ψ̂ 2
se (ψ̂) = .
(r1 + r2 )r2 − T22 ψ̂ 2 λ̂2
On substituting for ψ̂ and λ̂ in the denominator of this expression, this stan-
dard error simplifies to √
r1 + r2
ψ̂ . (5.33)
r1 r2
Similarly, the standard error of λ̂ turns out to be given by
√
se (λ̂) = λ̂/ r1 .
The standard error of these estimates cannot be used directly in the construc-
tion of confidence intervals for ψ and λ. The reason for this is that the values
of both parameters must be positive and their estimated values will tend to
have skewed distributions. This means that the assumption of normality, used
in constructing a confidence interval, would not be justified. The distribution
of the logarithm of an estimate of either ψ or λ is much more likely to be sym-
metric, and so confidence limits for the logarithm of the parameter are found
using the standard error of the logarithm of the parameter estimate. The re-
sulting confidence limits are then exponentiated to give an interval estimate
for the parameter itself.
The standard error of the logarithm of a parameter estimate can be found
using the general result given in Equation (5.15). Thus, the approximate vari-
ance of log ψ̂ is
var (log ψ̂) ≈ ψ̂ −2 var (ψ̂),
and so the standard error of log ψ̂ is given by
√
r1 + r2
se (log ψ̂) ≈ ψ̂ −1 se (ψ̂) = . (5.34)
r1 r2
A 100(1 − α)% confidence interval for the logarithm of the relative hazard
has limits log ψ̂ ± zα/2 se (log ψ̂), and confidence limits for the hazard ratio ψ
are found by exponentiating these limits for log ψ. If required, a confidence
interval for λ can be found in a similar manner.

The theoretical results developed in this section will now be illustrated using
the data on the survival times of breast cancer patients. The survival times
for the women in each group are assumed to have exponential distributions, so
that the hazard of death, at any time, for a woman in the negatively stained
group is a constant value, λ, while that for a woman in the positively stained
group is ψλ, where ψ is the hazard ratio.
From the data given in Table 1.2 of Chapter 1, the numbers of death
times in the negatively and positively stained groups are, respectively, r1 = 5
and r2 = 21. Also, the total time survived in each group is T1 = 1652 and
T2 = 2679 months. Using Equation (5.32), the estimated hazard of death for
a woman in the positively stained group, relative to one in the negatively
stained group, is
21 × 1652
ψ̂ = = 2.59,
5 × 2679
so that a woman in the positively stained group has about two and a half times
the risk of death at any given time, compared to a woman whose tumour was
negatively stained. This is consistent with the estimated value of ψ of 2.72
from the graphical procedure used in Example 5.4.
Next, using Equation (5.33), the standard error of the estimated hazard
ratio is given by √
5 + 21
se (ψ̂) = 2.59 = 1.289.
5 × 21
In order to obtain a 95% confidence interval for the true relative hazard,
the standard error of log ψ̂ is required. Using Equation (5.34), this is found
to be given by se (log ψ̂) = 0.498, and so 95% confidence limits for log ψ are
log(2.59)±1.96 se (log ψ̂), that is, 0.952±(1.96×0.498). The confidence interval
for the log relative hazard is (−0.024, 1.927), and the corresponding interval
estimate for the relative hazard itself is (e−0.024 , e1.927 ), that is, (0.98, 6.87).
This interval only just includes unity, and suggests that women with positively
stained tumours have a poorer prognosis than those whose tumours were neg-
atively stained. This result is consistent with the result of the log-rank test in
Example 2.12, where a P -value of 0.061 was obtained on testing the hypothesis
of no group difference.
In practice, computer software is used to fit a parametric models to two

groups of survival data, assuming proportional hazards. When the model in
Equation (5.29) is fitted, estimates of β, λ and γ, and their standard errors,
can be obtained from the resulting output. Further calculation may then be
needed to obtain an estimate of the relative hazard, and the standard error
of this estimate. In particular, the estimated hazard ratio would be obtained
as ψ̂ = exp(β̂) and se (ψ̂) found from the equation
se (ψ̂) = exp(β̂) se (β̂),

a result that follows from Equation (5.15).
The median and other percentiles of the survival time distributions in the
two groups can be estimated from the values of λ̂ and ψ̂. For example, from
Equation (5.22), the estimated pth percentile for those in Group I is found
from
{ ( )}1/γ̂
1 100
t̂(p) = log ,
λ̂ 100 − p
and that for individuals in Group II is
{ ( )}1/γ̂
1 100
t̂(p) = log .
ψ̂ λ̂ 100 − p
An expression similar to that in (5.27) can be used to obtain the standard error
of an estimated percentile for individuals in each group, once the variances and
covariances of the parameter estimates in the model have been found. Specific
results for the standard error of percentiles of the survival time distributions
in each of the two groups will not be given. Instead, the general expression for
the standard error of the pth percentile after fitting a Weibull model, given in
Equation (5.27), may be used.
In Example 5.4, a Weibull proportional hazards model was found to be ap-
propriate for the data on the survival times of two groups of breast cancer
patients. Under this model, the hazard of death at time t is λγtγ−1 for a neg-
atively stained patient and ψλγtγ−1 for a patient who is positively stained.
The estimated value of the shape parameter of the fitted Weibull distri-
bution is γ̂ = 0.937. The estimated scale parameter for women in Group I is
λ̂ = 0.00414 and that for women in Group II is λ̂ψ̂ = 0.0105. The estimated
hazard ratio under this Weibull model is ψ̂ = 2.55, which is not very different
from the value obtained in Example 5.5, on the assumption of exponentially
distributed survival times.
Putting γ̂ = 0.937 and λ̂ = 0.00414 in Equation (5.23) gives 235.89 for
the median survival time of those in Group I. The estimated median survival
time for women in Group II is found by putting γ̂ = 0.937 and λ̂ = 0.0105
in that equation, and gives 87.07 for the estimated median survival time. The
median survival time of women whose tumour was positively stained is about
one-third that of those whose tumour was negatively stained.
Using the general result for the standard error of the median survival time,
given in Equation (5.27), the standard error of the two medians is found by
taking p = 50, γ̂ = 0.937 and λ̂ = 0.00414 and 0.0105 in turn. They turn out
to be 114.126 and 20.550, respectively.
THE WEIBULL PROPORTIONAL HAZARDS MODEL 199
As in Section 5.4.3, 95% confidence limits for the true median survival
times for each group of women are best obtained by working with the log-
arithm of the median. The standard error of log t̂(50) is found using Equa-
tion (5.28), from which
1
se {t̂(50)} = se {t̂(50)}.
t̂(50)
Confidence limits for log t(50) are then exponentiated to give the correspond-
ing confidence limits for t(50) itself.
In this example, 95% confidence intervals for the true median survival times
of the two groups of women are (91.4, 608.9) and (54.8, 138.3), respectively.
Notice that the confidence interval for the median survival time of patients
with positive staining is much narrower than that for women with negative
staining. This is due to there being a relatively small number of uncensored
survival times in the women whose tumours were negatively stained.
5.6 The Weibull proportional hazards model

The model in Equation (5.29) for the comparison of two groups of survival
data can easily be generalised to give a model that is similar in form to
the Cox regression model described in Section 3.1.2. Suppose that the val-
ues x1 , x2 , . . . , xp of p explanatory variables, X1 , X2 , . . . , Xp , are recorded for
each of n individuals. Under the proportional hazards model, the hazard of
death at time t for the ith individual is
hi (t) = exp(β1 x1i + β2 x2i + · · · + βp xpi )h0 (t), (5.35)
for i = 1, 2, . . . , n. Although this model has a similar appearance to that

given in Equation (3.3), there is one fundamental difference, which concerns
the specification of the baseline hazard function h0 (t). In the Cox regression
model, the form of h0 (t) is unspecified, and the shape of the function is es-
sentially determined by the actual data. In the model being considered in this
section, the survival times are assumed to have a Weibull distribution, and
this imposes a particular parametric form on h0 (t).
Consider an individual for whom the values of the p explanatory variables
in the model of Equation (5.35) are all equal to zero. The hazard function for
such an individual is h0 (t). If the survival time of this individual has a Weibull
distribution with scale parameter λ and shape parameter γ, then their hazard
function is such that
h0 (t) = λγtγ−1 .
Using Equation (5.35), the hazard function for the ith individual in the study
is then given by
hi (t) = exp(β ′ xi )λγtγ−1 , (5.36)
where β ′ xi stands for β1 x1i +β2 x2i +· · ·+βp xpi . From the form of this hazard
function, we can see that the survival time of the ith individual in the study
has a Weibull distribution with scale parameter λ exp(β ′ xi ) and shape param-
eter γ. This again is a manifestation of the proportional hazards property of
the Weibull distribution. This result shows that the effect of the explanatory
variates in the model is to alter the scale parameter of the distribution, while
the shape parameter remains constant.
The survivor function corresponding to the hazard function given in Equa-
tion (5.36) is found using Equation (1.6), and turns out to be
Si (t) = exp {− exp(β ′ xi )λtγ } . (5.37)
5.6.1 ∗ Fitting the model

The Weibull proportional hazards model is fitted by constructing the likeli-
hood function of the n observations, and maximising this function with respect
to the unknown parameters, β1 , β2 , . . . , βp , λ and γ. Since the hazard function
and survivor function differ for each individual, the likelihood function in Ex-
pression (5.10) is now written as
∏
n
δ
{hi (ti )} i Si (ti ). (5.38)
i=1
The logarithm of the likelihood function, rather than the likelihood itself,
is maximised with respect to the unknown parameters, and from Expres-
sion (5.38), this is
∑n
[δi log hi (ti ) + log Si (ti )].
i=1
On substituting for hi (ti ) and Si (ti ) from Equations (5.36) and (5.37), the
log-likelihood becomes
∑
n
[δi {β ′ xi + log(λγ) + (γ − 1) log ti } − λ exp(β ′ xi )tγ ] ,
i=1

∑
n ∑
n
[δi {β ′ xi + log(λγ) + γ log ti } − λ exp(β ′ xi )tγ ] − δi log ti . (5.39)
i=1 i=1
∑n
The final term in this expression, − i=1 δi log ti , does not involve any of the
unknown parameters, and can be omitted from the likelihood. The resulting
log-likelihood function is then
∑
n
[δi {β ′ xi + log(λγ) + γ log ti } − λ exp(β ′ xi )tγ ] , (5.40)
i=1
which differs from that obtained∑n from the full log-likelihood, given in Expres-
sion (5.39), by the value of i=1 δi log ti . When computer software is used
to fit the Weibull proportional hazards model, the log-likelihood is generally
computed from Expression (5.40). This expression will also be used in the
examples given in this book.
Computer software for fitting parametric proportional hazards models gen-
erally includes the standard errors of the parameter estimates, from which
confidence intervals for relative hazards and the median and other percentiles
of the survival time distribution can be found. Specifically, suppose that the es-
timates of the parameters in the model of Equation (5.36) are β̂1 , β̂2 , . . . , β̂p , λ̂
and γ̂. The estimated survivor function for the ith individual in the study, for
whom the values of the explanatory variables in the model are x1i , x2i , . . . , xpi ,
is then { }
Ŝi (t) = exp − exp(β̂1 x1i + β̂2 x2i + · · · + β̂p xpi )λ̂tγ̂ ,
and the corresponding estimated hazard function is
ĥi (t) = exp(β̂1 x1i + β̂2 x2i + · · · + β̂p xpi )λ̂γ̂tγ̂−1 .
Both of these functions can be estimated and plotted against t, for individuals
with particular values of the explanatory variables in the model.
Generalising the result in Equation (5.22) to the situation where the
Weibull scale parameter is λ exp(β ′ xi ), the estimated pth percentile of the
survival time distribution for an individual, whose vector of explanatory vari-
ables is xi , is
{ ( )}1/γ̂
1 100
t̂(p) = log . (5.41)
λ̂ exp(β̂ ′ xi ) 100 − p
The estimated median survival time for such an individual is therefore

{ }1/γ̂
log 2
t̂(50) = . (5.42)
λ̂ exp(β̂ ′ xi )
The standard error of t̂(p) and corresponding interval estimates for t(p), are
derived in Section 5.6.2.
5.6.2 ∗ Standard error of a percentile in the Weibull model

As in Section 5.4.3, the standard error of the estimated pth percentile of the
survival time distribution for the Weibull proportional hazards model, t̂(p),
given in Equation (5.41), is found from the variance of log t̂(p). Writing
( )
100
cp = log log ,
100 − p
we have from Equation (5.41),
1( )
log t̂(p) = cp − log λ̂ − β̂ ′ xi . (5.43)
γ̂
This is a function of the p + 2 parameter estimates λ̂, γ̂ and β̂1 , β̂2 , . . . , β̂p ,
and the approximate variance of this function can be found using a further
generalisation of the result in Equation (5.24). This generalisation is best
expressed in matrix form.
Suppose that θ̂ is a vector formed from k parameter estimates, θ̂1 ,
θ̂2 , . . . , θ̂k , and that g(θ̂) is a function of these k estimates. The approximate
variance of g(θ̂) is then given by
var {g(θ̂)} ≈ d(θ̂)′ var (θ̂) d(θ̂), (5.44)
where var (θ̂) is the k × k variance-covariance matrix of the estimates in θ̂,
and d(θ̂) is the k-component vector whose ith element is

∂g(θ̂)
,
∂ θ̂i θˆ
for i = 1, 2, . . . , k.
We now write V for the (p + 2) × (p + 2) variance-covariance matrix of
λ̂, γ̂ and β̂1 , β̂2 , . . . , β̂p . Next, from Equation (5.43), the derivatives of log t̂(p)
with respect to these parameter estimates are
∂ log t̂(p) 1
=− ,
∂ λ̂ λ̂γ̂
∂ log t̂(p) cp − log λ̂ − β̂ ′ xi
=− ,
∂γ̂ γ̂ 2
∂ log t̂(p) xji
=− ,
∂ β̂j γ̂
for j = 1, 2, . . . , p, where xji is the jth component of xi . The vector d(θ̂) in

Equation (5.44) can be expressed as −γ̂ −1 d0 , where d0 is a vector with com-
ponents λ̂−1 , γ̂ −1 {cp − log λ̂ − β̂ ′ xi }, and x1i , x2i , . . . , xpi . Then, the standard
error of log t̂(p) is given by
{ } √( ′ )
se log t̂(p) = γ̂ −1 d0 V d0 , (5.45)
from which the standard error of t̂(p) itself is obtained using
{ }
se {t̂(p)} = t̂(p) se log t̂(p) . (5.46)
Notice that for the null model that contains no explanatory variables, the stan-
dard error of t̂(p) in Equation (5.46) reduces to the result in Equation (5.27).
Confidence limits for log t(p), can be found from the standard error of
log t̂(p), shown in Equation (5.45), and a confidence interval for t(p) is obtained
by exponentiating these limits.
5.6.3 ∗ Log-linear form of the model
Most computer software for fitting the Weibull proportional hazards model
uses a different form of the model from that adopted in this chapter. The
reason for this will be given in the next chapter, but for the moment we note
that the model can be formulated as a log-linear model for Ti , the random
variable associated with the survival time of the ith individual. In this version
of the model,
log Ti = µ + α1 x1i + α2 x2i + · · · + αp xpi + σϵi , (5.47)
where α1 , α2 , . . . , αp are the coefficients of the p explanatory variables in the

model, whose values are x1i , x2i , . . . , xpi for the ith of n individuals, µ and σ
are unknown parameters, and ϵi is a random variable that has a probability
distribution that leads to Ti having a Weibull distribution. In this form of the
model, the survivor function of Ti , given in Equation (5.37), is expressed as
{ ( )}
log t − µ − α1 x1i − α2 x2i − · · · − αp xpi
Si (t) = exp − exp .
σ
The correspondence between these two representations of the model is such

that
λ = exp(−µ/σ), γ = σ −1 , βj = −αj /σ,
for j = 1, 2, . . . , p, and µ, σ are often termed the intercept and scale parameter,
respectively; see Sections 6.4 and 6.5 of Chapter 6 for full details.
Generally speaking, it will be more straightforward to use the log-linear
form of the model to estimate the hazard and survivor functions, percentiles
of the survival time distribution, and their standard errors. The relevant ex-
pressions are given as Equations (6.16), (6.17) and (6.19) in Section 6.5 of
Chapter 6.
However, the log-linear representation of the model makes it difficult to
obtain confidence intervals for a log-hazard ratio, β, in a proportional hazards
model, since only the standard error of the estimate of α is given in the output.
In particular, on fitting the Weibull proportional hazards model, the output
provides the estimated value of α = −σβ, α̂, and the standard error of α̂. The
corresponding estimate of β is easily found from β̂ = −α̂/σ̂, but the standard
error of β̂ is more complicated to calculate.
To obtain the standard error of β̂, we can use the result given in Equa-
tion (5.24) of Section 5.4.3. Using this result, the approximate variance of the
function
α̂
g(α̂, σ̂) = −
σ̂
is
( )2 ( )2 ( )
∂g ∂g ∂g ∂g
var (α̂) + var (σ̂) + 2 cov (α̂, σ̂). (5.48)
∂ α̂ ∂ σ̂ ∂ α̂ ∂ σ̂
The derivatives of g(α̂, σ̂), with respect to α̂ and σ̂, are given by
∂g 1 ∂g α̂
=− , = 2,
∂ α̂ σ̂ ∂ σ̂ σ̂
and so using Expression (5.48),
( ) ( )2 ( )2 ( )( )
α̂ 1 α̂ 1 α̂
var − ≈ − var (α̂) + var (σ̂) + 2 − cov (α̂, σ̂).
σ̂ σ̂ σ̂ 2 σ̂ σ̂ 2
After some algebra, the approximate variance becomes
1 { 2 }
4
σ̂ var (α̂) + α̂2 var (σ̂) − 2α̂σ̂ cov (α̂, σ̂) , (5.49)
σ̂
and the square root of this is the standard error of β̂.

The computation of a standard error of a log-hazard ratio is now illustrated
using the data on the survival times of two groups breast cancer patients. In
this example, computer output from fitting the log-linear form of the Weibull
proportional hazards model is used to illustrate the estimation of the hazard
ratio, and calculation of the standard error of the estimate.
On fitting the model that contains the treatment effect, represented by
a variable X, where X = 0 for a woman with negative staining and X =
1 for positive staining, we find that the estimated coefficient of X is α̂ =
−0.9967. Also, the estimates of µ and σ are given by µ̂ = 5.8544 and σ̂ =
1.0668, respectively. The estimated log-hazard ratio for a woman with positive
staining, (X = 1) relative to a woman with negative staining (X = 0) is
−0.9967
β̂ = − = 0.9343.
1.0668
The corresponding hazard ratio is 2.55, as in Example 5.6.
The standard errors of α̂ and σ̂ are generally included in standard computer
output, and are 0.5441 and 0.1786, respectively. The estimated variances of α̂
and σ̂ are therefore 0.2960 and 0.0319, respectively. The covariance between
α̂ and σ̂ can be found from computer software, although it is not usually part
of the default output. It is found to be −0.0213.
Substituting for α̂, σ̂ and their variances and covariance in Expres-
sion (5.49), we get
var (β̂) ≈ 0.2498,
and so the standard error of β̂ is given by se (β̂) = 0.4998. This can be used
in the construction of confidence intervals for the corresponding true hazard
ratio.
5.6.4 Exploratory analyses
In Sections 5.2 and 5.5.1, we saw how a log-cumulative hazard plot could be
used to assess whether survival data can be modelled by a Weibull distribution,
and whether the proportional hazards assumption is valid. These procedures
work perfectly well when we are faced with a single sample of survival data,
or data where the number of groups is small and there is a reasonably large
number of individuals in each group. But in situations where there are a small
number of death times distributed over a relatively large number of groups,
it may not be possible to estimate the survivor function, and hence the log-
cumulative hazard function, for each group.
As an example, consider the data on the survival times of patients with
hypernephroma, given in Table 3.6. Here, individuals are classified according
to age group and whether or not a nephrectomy has been performed, giving
six combinations of age group and nephrectomy status. To examine the as-
sumption of a Weibull distribution for the survival times in each group, and
the assumption of proportional hazards across the groups, a log-cumulative
hazard plot would be required for each group. The number of patients in each
age group who have not had a nephrectomy is so small that the survivor
function cannot be properly estimated in these groups. If there were more
individuals in the study who had died and not had a nephrectomy, it would
be possible to construct a log-cumulative hazard plot. If this plot featured six
parallel straight lines, the Weibull proportional hazards model is likely to be
satisfactory.
When a model contains continuous variables, their values will first need
to be grouped before a log-cumulative hazard plot can be obtained. This may
also result in there being insufficient numbers of individuals in some groups
to enable the log-cumulative hazard function to be estimated.
The only alternative to using each combination of factor levels in construct-
ing a log-cumulative hazard plot is to ignore some of the factors. However, the
resulting plot can be very misleading. For example, suppose that patients are
classified according to the levels of two factors, A and B. The log-cumulative
hazard plot obtained by grouping the individuals according to the levels of A
ignoring B, or according to the levels of B ignoring A, may not give cause
to doubt the Weibull or proportional hazards assumptions. However, if the
log-cumulative hazard plot is obtained for individuals at each combination
of levels of A and B, the plot may not feature a series of four parallel lines.
By the same token, the log-cumulative hazard plot obtained when either A
or B is ignored may not show sets of parallel straight lines, but when a plot
is obtained for all combinations of A and B, parallel lines may result. This
feature is illustrated in the following example.
Example 5.8 A numerical illustration

Suppose that a number of individuals are classified according to the levels of
two factors, A and B, each with two levels, and that their survival times are
as shown in Table 5.1.
Table 5.1 Artificial data on the survival times of 37 patients

classified according to the levels of two factors, A and B.
A=1 A=2
B=1 B=2 B=1 B=2
59 10 88 25*
20 4 70* 111
71 16 54 152
33 18 139 86
25 19 31 212
25 35 59 187*
15 11 111 54
53 149 357
47 30 301
44 195
25
The log-cumulative hazard plot shown in Figure 5.12 is derived from the
individuals classified according to the two levels of A, ignoring the level of
factor B. The plot in Figure 5.13 is from individuals classified according to
the two levels of B, ignoring the level of factor A.
1
-1
-2
-3
-4
1 2 3 4 5 6
Figure 5.12 Log-cumulative hazard plot for individuals for whom A = 1 (∗) and
A = 2 (•).
From Figure 5.12 there is no reason to doubt the assumption of a Weibull

distribution for the survival times at the two levels of A, and the assumption of
proportional hazards is clearly tenable. However, the crossed lines on the plot
1
-1
-2
-3
-4
1 2 3 4 5 6
Figure 5.13 Log-cumulative hazard plot for individuals for whom B = 1 (∗) and
B = 2 (•).
shown as Figure 5.13 strongly suggest that the hazards are not proportional
when individuals are classified according to the levels of B. A different picture
emerges when the 37 survival times are classified according to the levels of
both A and B. The log-cumulative hazard plot based on the four groups is
shown in Figure 5.14. The four parallel lines show that there is no doubt about
the validity of the proportional hazards assumption across the groups.
In this example, the reason why the log-cumulative hazard plot for B
ignoring A is misleading is that there is an interaction between A and B. An
examination of the data reveals that, on average, the difference in the survival
times of patients for whom B = 1 and B = 2 is greater when A = 2 than
when A = 1.
Even when a log-cumulative hazard plot gives no reason to doubt the as-
sumption of a Weibull proportional hazards model, the validity of the fitted
model will need to be examined using the methods to be described in Chap-
ter 7. When it is not possible to use a log-cumulative hazard plot to explore
whether a Weibull distribution provides a reasonable model for the survival
times, a procedure based on the Cox regression model, described in Chapter
3, might be helpful. Essentially, a Cox regression model that includes all the
relevant explanatory variables is fitted, and the baseline hazard function is
estimated, using the procedure described in Section 3.10. A plot of this func-
tion may suggest whether or not the assumption of a Weibull distribution is
tenable. In particular, if the estimated baseline hazard function in the Cox
model is increasing or decreasing, the Weibull model may provide a more con-
1
-1
-2
-3
-4
1 2 3 4 5 6
Figure 5.14 Log-cumulative hazard plot for individuals in the groups defined by the
four combinations of levels of A and B.
cise summary of the baseline hazard function than the Cox regression model.
Because the estimated baseline hazard function for a fitted Cox model can
be somewhat irregular, comparing the estimated baseline cumulative hazard
or the baseline survivor function, under the fitted Cox regression model, with
that of the Weibull model may be more fruitful.
5.7 Comparing alternative Weibull models
In order to ascertain which explanatory variables should be included in a

Weibull proportional hazards model, alternative models need to be compared.
Comparisons between different Weibull models can be made using methods
analogous to those for the Cox regression model described in Section 3.5.
Suppose that one model contains a subset of the explanatory variables
in another, so that the two models are nested. The two models can then be
compared on the basis of the statistic −2 log L̂, where L̂ is the maximised value
of the likelihood function under the fitted model. For a model that contains
p explanatory variables, the sample likelihood is a function of p + 2 unknown
parameters, β1 , β2 , . . . , βp , λ and γ. The maximised likelihood is then the value
of this function when these parameters take their estimates, β̂1 , β̂2 , . . . , β̂p , λ̂
and γ̂.
More specifically, suppose that one model, Model (1), say, contains p ex-
planatory variables, X1 , X2 , . . . , Xp , and another model, Model (2), contains
an additional q explanatory variables, Xp+1 , Xp+2 , . . . , Xp+q . The estimated
COMPARING ALTERNATIVE WEIBULL MODELS 209
hazard functions for the ith of n individuals under these two models are as
shown below:
Model (1): hi (t) = exp{β̂1 x1i + β̂2 x2i + · · · + β̂p xpi }λ̂γ̂tγ̂−1
Model (2): hi (t) = exp{β̂1 x1i + β̂2 x2i + · · · + β̂p+q xp+q,i }λ̂γ̂tγ̂−1
where x1i , x2i , . . . , xp+q,i are the values of the p + q explanatory variables for
the ith individual. The maximised likelihoods under Model (1) and Model
(2) will be denoted by L̂1 and L̂2 , respectively. The difference between the
values of −2 log L̂1 and −2 log L̂2 , that is, −2{log L̂1 − log L̂2 }, then has an
approximate chi-squared distribution with q degrees of freedom, under the
null hypothesis that the coefficients of the additional q variates in Model (2)
are all equal to zero. If the difference between the values of −2 log L̂ for these
two models is significantly large when compared with percentage points of the
chi-squared distribution, we would deduce that the extra q terms are needed
in the model, in addition to the p that are already included. Since differences
between values of −2 log L̂ are used in comparing models, it does not matter
whether the maximised log-likelihood, used in computing the value of −2 log L̂,
is based on Expression (5.39) or (5.40).
The description of the modelling process in Sections 3.5–3.8 applies equally
well to models based on the Weibull proportional hazards model, and so will
not be repeated here. However, the variable selection strategy will be illus-
trated using two examples.

Data on the survival times of 36 patients, classified according to their age
group and whether or not they have had a nephrectomy, were introduced in
Example 3.4 of Chapter 3. In that example, the data were analysed using
the Cox proportional hazards model. Here, the analysis is repeated using the
Weibull proportional hazards model. As in Example 3.4, the effect of the jth
age group will be denoted by αj , and that associated with whether or not a
nephrectomy was performed by νk . There are then five possible models for the
hazard function of the ith individual, hi (t), which are as follows:
Model (1): hi (t) = h0 (t)
Model (2): hi (t) = exp{αj }h0 (t)
Model (3): hi (t) = exp{νk }h0 (t)
Model (4): hi (t) = exp{αj + νk }h0 (t)
Model (5): hi (t) = exp{αj + νk + (αν)jk }h0 (t)

In these models, h0 (t) = λγtγ−1 is the baseline hazard function, and the pa-
rameters λ and γ have to be estimated along with those in the linear compo-
nent of the model. These five models have the interpretations given in Example
3.4. They can be fitted by constructing indicator variables corresponding to
the factors age group and nephrectomy status, as shown in Example 3.4, or
by using software that allows factors to be fitted directly.
Once a Weibull proportional hazards model has been fitted to the data,
values of −2 log L̂ can be found. These are given in Table 5.2 for the five
models of interest.
Table 5.2 Values of −2 log L̂ on fitting five

Weibull models to the data on hypernephroma.
Model Terms in model −2 log L̂
(1) null model 104.886
(2) αj 96.400
(3) νk 94.384
(4) αj + νk 87.758
(5) αj + νk + (αν)jk 83.064
The values of the −2 log L̂ statistic in Table 5.2, and other examples in
this book, have been computed using the log-likelihood in Expression (5.40).
Accordingly these values may differ from the∑values given by some computer
n
software packages by an amount equal to 2 i=1 δi log ti , which in this case
has the value 136.3733.
The reduction in the value of −2 log L̂ on adding the interaction term to
Model (4) is 4.69 on two degrees of freedom. This reduction is just about
significant at the 10% level (P = 0.096) and so there is some suggestion of an
interaction between age group and nephrectomy status. For comparison, note
that when the Cox regression model was fitted in Example 3.4, the interaction
was not significant (P = 0.220).
The interaction can be investigated in greater detail by examining the haz-
ard ratios under the model. Under Model (5), the estimated hazard function
for the ith individual is
d }ĥ0 (t),
ĥi (t) = exp{α̂j + ν̂k + (αν) jk
where
ĥ0 (t) = λ̂γ̂tγ̂−1
is the estimated baseline hazard function. The logarithm of the hazard ratio
for an individual in the jth age group, j = 1, 2, 3, and kth level of nephrectomy
status, k = 1, 2, relative to an individual in the youngest age group who has
not had a nephrectomy, is therefore
d − α̂1 − ν̂1 − (αν)
α̂j + ν̂k + (αν) d , (5.50)
jk 11
since the baseline hazard functions cancel out.

As in Example 3.4, models can be fitted to the data by defining indicator
variables A2 and A3 for age group and N for nephrectomy status. As in that
example, A2 is unity for an individual in the second age group and zero other-
wise, A3 is unity for an individual in the third age group and zero otherwise,
and N is unity if a nephrectomy has been performed and zero otherwise. Thus,
fitting the term αj corresponds to fitting the variables A2 and A3 , fitting νk
corresponds to fitting N , and fitting the interaction term (αν)jk corresponds
to fitting the products A2 N = A2 × N and A3 N = A3 × N . In particular,
to fit Model (5), the five variables A2 , A3 , N, A2 N, A3 N are included in the
d =0
model. With this choice of indicator variables, α̂1 = 0, ν̂1 = 0 and (αν) jk
d are the
when either i or j is unity. The remaining values of α̂j , ν̂k and (αν) jk
coefficients of A2 , A3 , N, A2 N, A3 N and are given in Table 5.3.
Table 5.3 Parameter estimates

on fitting a Weibull model to the
data on hypernephroma.
Parameter Estimate
α2 −0.085
α3 0.115
ν2 −2.436
(αν)22 0.121
(αν)32 2.538
Many computer packages set up indicator variables internally, and so es-

timates such as those in the above table can be obtained directly from the
output. However, to repeat an earlier warning, when packages are used to fit
factors, the coding used to define the indicator variables must be known if the
output is to be properly interpreted.
When the indicator variables specified above are used, the logarithm of
the hazard ratio given in Equation (5.50) reduces to
d ,
α̂j + ν̂k + (αν) jk
for j = 1, 2, 3, k = 1, 2. Table 5.4 gives the hazards for the individuals, relative
to the baseline hazard. The baseline hazard corresponds to an individual in
the youngest age group who has not had a nephrectomy, and so a hazard ratio
of unity for these individuals is recorded in Table 5.4.
Table 5.4 Hazard ratios for individuals classified by

age group and nephrectomy status.
< 60 1.00 0.09
60–70 0.92 0.09
> 70 1.12 1.24
This table helps to explain the interaction between age group and nephrec-
tomy status, in that the effect of a nephrectomy is not the same for individuals
in each of the three age groups. For patients in the two youngest age groups,
a nephrectomy substantially reduces the hazard of death at any given time.
Performing a nephrectomy on patients aged over 70 does not have much effect
on the risk of death. We also see that for those patients who have not had a
nephrectomy, age does not much affect the hazard of death.
Estimated median survival times can be found in a similar way. Using
Equation (5.42), the median survival time for a patient in the jth age group,
j = 1, 2, 3, and the kth level of nephrectomy status, k = 1, 2, becomes
{ }1/γ̂
log 2
t̂(50) = .
d }
λ̂ exp{α̂j + ν̂k + (αν)
jk
When the model containing the interaction term is fitted to the data,
the estimated values of the parameters in the baseline hazard function are
λ̂ = 0.0188 and γ̂ = 1.5538. Table 5.5 gives the estimated median survival
times, in months, for individuals with each combination of age group and
nephrectomy status.
Table 5.5 Median survival times for individuals

classified by age group and nephrectomy status.
< 60 10.21 48.94
60–70 10.78 47.81
> 70 9.48 8.87
This table shows that a nephrectomy leads to more than a fourfold increase
in the median survival time in patients aged up to 70 years. The median
survival time of patients aged over 70 is not much affected by the performance
of a nephrectomy.
We end this example with a note of caution. For some combinations of
age group and nephrectomy status, particularly the groups of individuals who
have not had a nephrectomy, the estimated hazard ratios and median survival
times are based on small numbers of survival times. As a result, the standard
errors of estimates of such quantities, which have not been given here, will be
large.
Example 5.10 Chemotherapy in ovarian cancer patients

Following surgical treatment of ovarian cancer, patients may undergo a course
of chemotherapy. In a study of two different forms of chemotherapy treat-
ment, Edmunson et al. (1979) compared the anti-tumour effects of cyclophos-
phamide alone and cyclophosphamide combined with adriamycin. The trial
involved 26 women with minimal residual disease and who had experienced
surgical excision of all tumour masses greater than 2 cm in diameter. Fol-
lowing surgery, the patients were further classified according to whether the
residual disease was completely or partially excised. The age of the patient
and their performance status were also recorded at the start of the trial. The
response variable was the survival time in days following randomisation to one
or other of the two chemotherapy treatments. The variables in the data set
are therefore as follows:
Time: Survival time in days

Status: Event indicator (0 = censored, 1 = uncensored)
Treat: Treatment (1 = single, 2 = combined)
Age: Age of patient in years
Rdisease: Extent of residual disease (1 = incomplete, 2 = complete)
Perf: Performance status (1 = good, 2 = poor)
The data, which were obtained from Therneau (1986), are given in Table 5.6.
Table 5.6 Survival times of ovarian cancer patients.

Patient Time Status Treat Age Rdisease Perf
1 156 1 1 66 2 2
2 1040 0 1 38 2 2
3 59 1 1 72 2 1
4 421 0 2 53 2 1
5 329 1 1 43 2 1
6 769 0 2 59 2 2
7 365 1 2 64 2 1
8 770 0 2 57 2 1
9 1227 0 2 59 1 2
10 268 1 1 74 2 2
11 475 1 2 59 2 2
12 1129 0 2 53 1 1
13 464 1 2 56 2 2
14 1206 0 2 44 2 1
15 638 1 1 56 1 2
16 563 1 2 55 1 2
17 1106 0 1 44 1 1
18 431 1 1 50 2 1
19 855 0 1 43 1 2
20 803 0 1 39 1 1
21 115 1 1 74 2 1
22 744 0 2 50 1 1
23 477 0 1 64 2 1
24 448 0 1 56 1 2
25 353 1 2 63 1 2
26 377 0 2 58 1 1
In modelling these data, the factors Treat, Rdisease and Perf each have two
levels, and will be fitted as variates that take the values given in Table 5.6.
This does of course mean that the baseline hazard function is not directly
interpretable, since there can be no individual for whom the values of all these
variates are zero. From both a computational and interpretive viewpoint, it is
more convenient to relocate the values of the variables Age, Rdisease, Perf and
Treat. If the variable Age − 50 is used in place of Age, and unity is subtracted
from Rdisease, Perf and Treat, the baseline hazard then corresponds to the
hazard for an individual of age 50 with incomplete residual disease, good
performance status, and who has been allocated to the cyclophosphamide
group. However, the original variables will be used in this example.
We begin by identifying which prognostic factors are associated with the
survival times of the patients. The values of the statistic −2 log L̂ on fitting a
range of models to these data are given in Table 5.7.
Table 5.7 Values of −2 log L̂ on fitting

models to the data in Table 5.6.
none 59.534
Age 43.566
Rdisease 55.382
Perf 58.849
Age, Rdisease 41.663
Age, Perf 43.518
Age, Treat 41.126
Age, Treat, Treat × Age 39.708
When Weibull models that contain just one of Age, Rdisease and Perf are
fitted, we find that both Age and Rdisease lead to reductions in the value of
−2 log L̂ that are significant at the 5% level. After fitting Age, the variables
Rdisease and Perf further reduce −2 log L̂ by 1.903 and 0.048, respectively,
neither of which is significant at the 10% level. Also, when Age is added to the
model that already includes Rdisease, the reduction in −2 log L̂ is 13.719 on
1 d.f., which is highly significant (P < 0.001). This leads us to the conclusion
that Age is the only prognostic variable that needs to be incorporated in the
model.
The term associated with the treatment effect is now added to the model.
The value of −2 log L̂ is then reduced by 2.440 on 1 d.f. This reduction of 2.440
is not quite large enough for it to be significant at the 10% level (P = 0.118).
There is therefore only very slight evidence of a difference in the effect of the
two chemotherapy treatments on the hazard of death.
For comparison, when Treat alone is added to the null model, the value
of −2 log L̂ is reduced from 59.534 to 58.355. This reduction of 1.179 is cer-
tainly not significant when compared to percentage points of the chi-squared
distribution on 1 d.f. Ignoring Age therefore leads to an underestimate of the
magnitude of the treatment effect.
To explore whether the treatment difference is consistent over age, the
interaction term formed as the product of Age and Treat is added to the model.
EXPLAINED VARIATION IN THE WEIBULL MODEL 215
On doing so, −2 log L̂ is only reduced by 1.419. This reduction is nowhere near
being significant and so there is no need to include an interaction term in the
model.
The variable Treat will be retained in the model, since interest centres
on the magnitude of the treatment effect. The fitted model for the hazard of
death at time t for the ith individual is then found to be
ĥi (t) = exp{0.144 Age i − 1.023 Treat i }λ̂γ̂tγ̂−1 ,
where λ̂ = 5.645×10−9 and γ̂ = 1.822. In this model, Treat = 1 for cyclophos-

phamide alone and Treat = 2 for the combination of cyclophosphamide with
adriamycin. The hazard for a patient on the single treatment, relative to one
on the combined treatment, is therefore estimated by
ψ̂ = exp{(−1.023 × 1) − (−1.023 × 2)} = 2.78.
This means that a patient receiving the single chemotherapy treatment is

nearly three times more likely to die at any given time than a patient on
the combined treatment. Expressed in this way, the benefits of the combined
chemotherapy treatment appear substantial. However, when account is taken
of the inherent variability of the data on which these results are based, this
relative hazard is only significantly greater than unity at the 12% level (P =
0.118).
The median survival time can be estimated for patients of a given age on
a given treatment from the equation
{ }1/γ̂
log 2
t̂(50) = .
λ̂ exp(0.144 Age − 1.023 Treat)
For example, a woman aged 60 (Age = 60) who is given cyclophosphamide

alone (Treat = 1) has an estimated median survival time of 423 days, whereas
someone of the same age on the combination of the two chemotherapy treat-
ments has an estimated median survival time of 741 days. Confidence intervals
for these estimates can be found using the method illustrated in Example 5.6.
5.8 ∗ Measures of explained variation in the Weibull model

Measures of explained variation in the Cox regression model were described in
Section 3.12 of Chapter 3. In Equation (3.32) of that section, the R2 measure
of explained variation for the general linear model was expressed as
V̂M
R2 = ,
V̂M + σ̂ 2
where V̂M is an estimate of variation in the data due to the fitted model and
σ̂ 2 is the residual variation. Now, VM can be expressed as β̂ ′ S β̂, where S is
the variance-covariance matrix of the explanatory variables; see Section 3.12
of Chapter 3. It then follows that R2 is a sample estimate of the quantity
β ′ Sβ
ρ2 = ′
β Sβ + σ 2
in which σ 2 is the variance of the response variable, Y.
We now adapt this measure for use in the analysis of survival data, which
requires the replacement of σ 2 by a suitable quantity. To do this, we take σ 2
to be the variance of the error term, ϵi , in the log-linear form of the Weibull
model given in Equation (5.47) of Section 5.6.3. In the particular case of
Weibull survival times, ϵi has a distribution which is such that the variance of
ϵi is π 2 /6; further details are given in Section 6.5.1 of Chapter 6. This leads
to the statistic
2 β̂ ′ S β̂
RP = ′ ,
β̂ S β̂ + π 2 /6
that was introduced in Section 3.12.1 of Chapter 3. This measure of explained
variation can be generally recommended for use with both the Cox and Weibull
proportional hazards models.
2
The RD statistic, also described in Section 3.12.1 of Chapter 3, can be
adapted for use with parametric survival models in a similar manner. This
leads to
2 D2
RD = 2 .
D + π 2 /6
where D is the scaled coefficient for the regression of the ordered values of the
risk score on normal scores as before.
To illustrate the use of measures of explained variation in a Weibull model
for survival data, consider the data on the survival times of patients with
multiple myeloma, for which the values of three R2 statistics on fitting Cox
regression models were given in Example 3.17. For a Weibull model containing
2 2
the variables Hb and Bun, RP = 0.25 and RD = 0.23. For the model that
2 2
contains all 7 variables, RP = 0.30 and RD = 0.28. These values are quite
similar to those obtained on fitting corresponding Cox regression models, so
that the Cox and Weibull models have similar explanatory power.
5.9 The Gompertz proportional hazards model

Although the Weibull model is the most widely used parametric proportional
hazards model, the Gompertz model has found application in demography and
the biological sciences. Indeed the distribution was introduced by Gompertz
in 1825, as a model for human mortality.
The hazard function of the Gompertz distribution is given by
h(t) = λeθt ,
THE GOMPERTZ PROPORTIONAL HAZARDS MODEL 217
for 0 6 t < ∞, and λ > 0. In the particular case where θ = 0, the hazard func-
tion has a constant value, λ, and the survival times then have an exponential
distribution. The parameter θ determines the shape of the hazard function,
positive values leading to a hazard function that increases with time. The haz-
ard function can also be expressed as h(t) = exp(α +θt), which shows that the
log-hazard function is linear in t. On the other hand, from Equation (5.4), the
Weibull log-hazard function is linear in log t. Like the Weibull hazard function,
the Gompertz hazard increases or decreases monotonically.
The survivor function of the Gompertz distribution is given by
{ }
λ
S(t) = exp (1 − eθt ) ,
θ
and the corresponding density function is
{ }
λ
f (t) = λeθt exp (1 − eθt ) .
θ
The pth percentile is such that
{ ( )}
1 θ 100 − p
t(p) = log 1 − log ,
θ λ 100
from which the median survival time is
{ }
1 θ
t(50) = log 1 + log 2 .
θ λ
A plot of the Gompertz hazard function for distributions with a median of
20 and θ = −0.2, 0.02 and 0.05 is shown in Figure 5.15. The corresponding
values of λ are 0.141, 0.028 and 0.020.
It is straightforward to see that the Gompertz distribution has the propor-
tional hazards property, described in Section 5.5.1, since if we take h0 (t) =
λeθt , then ψh0 (t) is also a Gompertz hazard function with parameters ψλ
and θ.
The general Gompertz proportional hazards model for the hazard of death
at time t for the ith of n individuals is expressed as
hi (t) = exp(β1 x1i + β2 x2i + · · · + βp xpi )λeθt ,
where x1i , x2i , . . . , xpi are the values of p explanatory variables X1 , X2 , . . . , Xp

for the ith individual, i = 1, 2, . . . , n, and the β’s, λ and θ are unknown pa-
rameters. The model can be fitted by maximising the likelihood function given
in Expression (5.9) or (5.10). The β-coefficients are interpreted as log-hazard
ratios, and alternative models are compared using the approach described in
Section 5.7. No new principles are involved.
In Example 5.10 on the survival times of ovarian cancer patients, a Weibull
0.15
0.10
Hazard function
0.05
0.00
0 10 20 30 40
Time
Figure 5.15 Hazard functions for a Gompertz distribution with a median of 20 and
θ = −0.2, 0.02 and 0.05.
proportional hazards model that contained the variables Age and Treat was
fitted. For comparison, a Gompertz proportional hazards model that contains
these two variables is now fitted. Under this model, the fitted hazard function
for the ith patient is
ĥi (t) = exp{0.122 Age i − 0.848 Treat i }λ̂ exp(θ̂t),
where λ̂ = 1.706 × 10−6 and θ̂ = 0.00138. The change in the value of −2 log L̂
on adding Treat to the Gompertz proportional hazards model that contains
Age alone is now 1.686 (P = 0.184). The hazard ratio for the treatment effect,
which is now exp(0.848) = 2.34, is therefore smaller and less significant under
this model than it was for the Weibull model.
5.10 Model choice

One attraction of the proportional hazards model for survival data is that it is
not necessary to adopt a specific probability distribution for the survival times.
However, when a Weibull distribution is appropriate for the observed survival
data, the parametric version of the proportional hazards model provides a
more suitable basis for modelling the data.
Diagnostic plots based on the log-cumulative hazard function, described in
Section 5.5.1, may throw light on whether the assumption of Weibull survival
times is plausible, but as has already been pointed out, this technique is often
not informative in the presence of explanatory variables that affect survival
times. In such circumstances, to help choose between the Cox and Weibull
FURTHER READING 219
proportional hazards models, it can be useful to fit the Cox regression model
and examine the shape of the baseline hazard function. The fitted Weibull
baseline cumulative hazard function, or the fitted baseline survivor function,
can also be compared with the corresponding estimates for the Cox regression
model, as described in Section 5.6.4.
A suitable analysis of residuals, to be discussed in Chapter 7, can be used
to investigate whether one model fits better than the other. However, it will
only be in exceptional circumstances that model-checking diagnostics provide
convincing evidence that one or other of the two models is more acceptable.
In general, discrimination between a Cox and a Weibull proportional haz-
ards model will be difficult unless the sample data contain a large number of
death times. In cases where there is little to choose between the two models
in terms of goodness of fit, the standard errors of the estimated β-parameters
in the linear component of the two models can be compared. If those for
the Weibull model are substantially smaller than those for the Cox model,
the Weibull model would be preferred on grounds of efficiency. On the other
hand, if these standard errors are similar, the Cox model is likely to be the
model of choice in view of its less restrictive assumptions.

The properties of the exponential, Weibull and Gompertz distributions are
presented in Johnson and Kotz (1994). A thorough discussion of the theory
of maximum likelihood estimation is included in Barnett (1999) and Cox and
Hinkley (1974), and a useful summary of the main results is contained in
Hinkley, Reid and Snell (1991). Numerical methods for obtaining maximum
likelihood estimates, and the Newton-Raphson procedure in particular, are
described by Everitt (1987) and Thisted (1988), for example; see also the
description in Section 3.3.3 of Chapter 3. Byar (1982) presents a comparison
of the Cox and Weibull proportional hazards models.
Papers that define and compare measures of explained variation for the
Weibull model include those of Kent and O’Quigley (1988) and Royston and
Sauerbrei (2004), cited in Chapter 3.
One other distribution with the proportional hazards property is the
Pareto distribution. This model is rarely used in practice, but see Davis and
Feldstein (1979) for further details.
Chapter 6
Accelerated failure time and other

parametric models
Although the proportional hazards model finds widespread applicability in the

analysis of survival data, there are relatively few probability distributions for
the survival times that can be used with this model. Moreover, the distribu-
tions that are available, principally the Weibull and Gompertz distributions,
lead to hazard functions that increase or decrease monotonically. A model
that encompasses a wider range of survival time distributions is the accel-
erated failure time model. In circumstances where the proportional hazards
assumption is not tenable, models based on this general family may prove to
be fruitful. Again, the Weibull distribution may be adopted for the distribu-
tion of survival times in the accelerated failure time model, but some other
probability distributions are also available. This chapter therefore begins with
a brief survey of alternative distributions for survival data, that may be used
in conjunction with an accelerated failure time model. The model itself is then
considered in detail in Sections 6.3 to 6.6. One other general family of sur-
vival models, known as the proportional odds model, may be useful in some
circumstances. This model is described in Section 6.7.
Parametric models based on standard probability distributions for survival
data may not adequately summarise the pattern of an underlying baseline
survivor or hazard function. An important extension to the basic proportional
hazards and proportional odds models that leads to more flexible models is
described and illustrated in Section 6.9.
6.1 Probability distributions for survival data

The Weibull distribution, described in Section 5.1.2, will not necessarily pro-
vide a satisfactory model for survival times in all circumstances, and so alter-
natives to this distribution need to be considered. Although any continuous
distribution for non-negative random variables might be used, the properties
of the log-logistic distribution make it a particularly attractive alternative to
the Weibull distribution. The lognormal, gamma and inverse Gaussian dis-
tributions are also used in accelerated failure time modelling, and so these
distributions are introduced in this section.
221
222 ACCELERATED FAILURE TIME AND OTHER MODELS
6.1.1 The log-logistic distribution
One limitation of the Weibull hazard function is that it is a monotonic function
of time. However, situations in which the hazard function changes direction
can arise. For example, following a heart transplant, a patient faces an increas-
ing hazard of death over the first ten days or so after the transplant, while the
body adapts to the new organ. The hazard then decreases with time as the
patient recovers. In situations such as this, a unimodal hazard function may
be appropriate.
A particular form of unimodal hazard is the function
eθ κtκ−1
h(t) = , (6.1)
1 + eθ t κ
for 0 6 t < ∞, κ > 0. This hazard function decreases monotonically if κ 6 1,
but if κ > 1, the hazard has a single mode. The survivor function correspond-
ing to the hazard function in Equation (6.1) is given by
{ }−1
S(t) = 1 + eθ tκ , (6.2)
and the probability density function is
eθ κtκ−1
f (t) = .
(1 + eθ tκ )2
This is the density of a random variable T that has a log-logistic distribution,

with parameters θ, κ. The distribution is so called because the variable log T
has a logistic distribution, a symmetric distribution whose probability density
function is very similar to that of the normal distribution.
The pth percentile of the log-logistic distribution is
( )1/κ
p e−θ
t(p) = ,
100 − p
and so the median of the distribution is
t(50) = e−θ/κ .
The hazard functions for log-logistic distributions with a median of 20 and

κ = 0.5, 2.0 and 5.0 are shown in Figure 6.1. The corresponding values of θ
for these distributions are −1.5, −6.0 and −15.0, respectively.
6.1.2 The lognormal distribution

The lognormal distribution is also defined for random variables that take pos-
itive values, and so may be used as a model for survival data. A random
variable, T , is said to have a lognormal distribution, with parameters µ and
PROBABILITY DISTRIBUTIONS FOR SURVIVAL DATA 223
0.20
0.15
Hazard function
=5.0
0.10
0.05
=2.0
=0.5
0.00
0 10 20 30 40
Time
Figure 6.1 Hazard functions for a log-logistic distribution with a median of 20 and
κ = 0.5, 2.0 and 5.0.
σ, if log T has a normal distribution with mean µ and variance σ 2 . The prob-
ability density function of T is given by
1 { }
f (t) = √ t−1 exp −(log t − µ)2 /2σ 2 ,
σ (2π)
for 0 6 t < ∞, σ > 0, from which the survivor and hazard functions can be
derived. The survivor function of the lognormal distribution is
( )
log t − µ
S(t) = 1 − Φ , (6.3)
σ
where Φ(·) is the standard normal distribution function, given by

∫ z
1 ( )
Φ(z) = √ exp −u2 /2 du.
(2π) −∞
The pth percentile of the distribution is then

{ }
t(p) = exp σΦ−1 (p/100) + µ ,
where Φ−1 (p/100), the pth percentile of the standard normal distribution, is
sometimes called the probit of p/100. In particular, the median survival time
under this distribution is simply t(50) = eµ .
The hazard function can be found from the relation h(t) = f (t)/S(t). This
function is zero when t = 0, increases to a maximum and then decreases to
zero as t tends to infinity. The fact that the survivor and hazard functions
can only be expressed in terms of integrals limits the usefulness of this model.
Moreover, in view of the similarity of the normal and logistic distributions,
the lognormal model will tend to be very similar to the log-logistic model.
6.1.3 ∗ The gamma distribution

The probability density function of a gamma distribution with mean ρ/λ and
variance ρ/λ2 is such that
λρ tρ−1 e−λt
f (t) = , (6.4)
Γ(ρ)
for 0 6 t < ∞, λ > 0, and ρ > 0. As for the lognormal distribution, the sur-
vivor function of the gamma distribution can only be expressed as an integral,
and we write
S(t) = 1 − Γλt (ρ),
where Γλt (ρ) is known as the incomplete gamma function, given by
∫ λt
1
Γλt (ρ) = uρ−1 e−u du.
Γ(ρ) 0
The hazard function for the gamma distribution is then h(t) = f (t)/S(t). This
hazard function increases monotonically if ρ > 1 and decreases if ρ < 1, and
tends to λ as t tends to ∞.
When ρ = 1, the gamma distribution reduces to the exponential distri-
bution described in Section 5.1.1, and so this distribution, like the Weibull
distribution, includes the exponential distribution as a special case. Indeed,
the gamma distribution is quite similar to the Weibull, and inferences based
on either model will often be very similar.
A generalisation of the gamma distribution is actually more useful than
the gamma distribution itself, since it includes the Weibull and lognormal
distributions as special cases. This model, known as the generalised gamma
distribution, may therefore be used to discriminate between alternative para-
metric models for survival data.
The probability density function of the generalised gamma distribution
is an extension of the gamma density in Equation (6.4), that includes an
additional parameter, θ, where θ > 0, and is defined by
θλρθ tρθ−1 exp{−(λt)θ }

f (t) = ,
Γ(ρ)
for 0 6 t < ∞. The survivor function for this distribution is again defined in
terms of the incomplete gamma function and is given by
S(t) = 1 − Γ(λt)θ (ρ),

EXPLORATORY ANALYSES 225
and the hazard function is again found from h(t) = f (t)/S(t). This distribu-
tion leads to a wide range of shapes for the hazard function, governed by the
parameter θ. This parameter is therefore termed the shape parameter of the
distribution. When ρ = 1, the distribution becomes the Weibull, when θ = 1,
the gamma, and as ρ → ∞, the lognormal.
6.1.4 ∗ The inverse Gaussian distribution

The inverse Gaussian distribution is a flexible model that has some important
theoretical properties. The probability density function of the distribution
which has mean µ and scale parameter λ is given by
( ) 21 { }
λ λ(t − µ2 )
f (t) = exp ,
2πt3 2µ2 t
for 0 6 t < ∞, and λ > 0. The corresponding survivor function is
{( ) √ ( −1 )} { ( ) √ ( −1 )}
S(t) = Φ 1 − tµ−1 λt − exp(2λ/µ) Φ − 1 + tµ−1 λt ,
and the hazard function is found from the ratio of the density and survivor
functions. However, the complicated form of the survivor function makes this
distribution difficult to work with.
6.2 Exploratory analyses

When the number of observations in a single sample is reasonably large, an
empirical estimate of the hazard function could be obtained using the method
described in Section 2.3.1. A plot of the estimated hazard function may then
suggest a suitable parametric form for the hazard function. For example, if
the hazard plot is found to be unimodal, a log-logistic distribution could be
used for the survival times. When the database includes a number of explana-
tory variables, the form of the estimated baseline hazard, cumulative hazard,
or survivor functions, for a fitted Cox regression model, may also indicate
whether a particular parametric model is suitable, in the manner described in
Section 5.6.4 of Chapter 5.
A method for exploring the adequacy of the Weibull model in describing
a single sample of survival times was described in Section 5.2. A similar pro-
cedure can be used to assess the suitability of the log-logistic distribution.
The basic idea is that a transformation of the survivor function is sought,
which leads to a straight line plot. From Equation (6.2), the odds of surviving
beyond time t are
S(t)
= e−θ t−κ ,
1 − S(t)
and so the log-odds of survival beyond t can be expressed as
{ }
S(t)
log = −θ − κ log t.
1 − S(t)
If the survivor function for the data is estimated using the Kaplan-Meier
estimate, and the estimated log-odds of survival beyond t are plotted against
log t, a straight line plot will be obtained if a log-logistic model for the survival
times is suitable. Estimates of the parameters of the log-logistic distribution,
θ and κ, can be obtained from the intercept and slope of the straight line plot.
The suitability of other parametric models can be investigated along simi-
lar lines. For example, from the survivor function of the lognormal distribution,
given in Equation (6.3),
log t − µ
Φ−1 {1 − S(t)} = ,
σ
and so a plot of Φ−1 {1 − Ŝ(t)} against log t should give a straight line, if the
lognormal model is appropriate. The slope and intercept of this line provide
estimates of σ −1 and −µ/σ, respectively.

In Example 5.1, a log-cumulative hazard plot was used to evaluate the fit of the
Weibull distribution to the data on times to discontinuation of an intrauterine
device (IUD), given in Example 1.1. We now consider whether the log-logistic
distribution is appropriate. A plot of log{Ŝ(t)/[1 − Ŝ(t)]} against log t for the
data on the times to discontinuation of an IUD is shown in Figure 6.2.
4
Log-odds of discontinuation
-2
2.0 2.5 3.0 3.5 4.0 4.5 5.0
Log of discontinuation time
Figure 6.2 A plot of the estimated log-odds of discontinuation after t against log t
for the data from Example 1.1.
From this plot, it appears that the relationship between the estimated
log-odds of discontinuing use of the contraceptive after time t, and log t, is
ACCELERATED FAILURE MODEL FOR TWO GROUPS 227
reasonably straight. This suggests that a log-logistic model could be used to
model the observed data.
Notice that there is very little difference in the extent of departures from
linearity in the plots of Figures 5.6 and 6.2. This means that either the Weibull
distribution or the log-logistic distribution is likely to be satisfactory, even
though the estimated hazard function under these two distributions may be
quite different. Indeed, when survival data are obtained for a relatively small
number of individuals, as in this example, there will often be little to choose
between alternative distributional models for the data. The model that is the
most convenient for the purpose in hand will then be adopted.
6.3 The accelerated failure time model for comparing two groups
The accelerated failure time model is a general model for survival data, in
which explanatory variables measured on an individual are assumed to act
multiplicatively on the time-scale, and so affect the rate at which an individual
proceeds along the time axis. This means that the models can be interpreted
in terms of the speed of progression of a disease, an interpretation that has
immediate intuitive appeal. Before the general form of the model is presented
in Section 6.4, the model for comparing the survival times of two groups of
patients is described in detail.
Suppose that patients are randomised to receive one of two treatments, a
standard treatment, S, or a new treatment, N . Under an accelerated failure
time model, the survival time of an individual on the new treatment is taken to
be a multiple of the survival time for an individual on the standard treatment.
Thus, the effect of the new treatment is to ‘speed up’ or ‘slow down’ the
passage of time. Under this assumption, the probability that an individual on
the new treatment survives beyond time t is the probability that an individual
on the standard treatment survives beyond time t/ϕ, where ϕ is an unknown
positive constant.
Now let SS (t) and SN (t) be the survivor functions for individuals in the
two treatment groups. Then, the accelerated failure time model specifies that
SN (t) = SS (t/ϕ),
for any value of the survival time t. One interpretation of this model is that
the lifetime of an individual on the new treatment is ϕ times the lifetime
that the individual would have experienced under the standard treatment.
The parameter ϕ therefore reflects the impact of the new treatment on the
baseline time-scale. When the end-point of concern is the death of a patient,
values of ϕ less than unity correspond to an acceleration in the time to death
of an individual assigned to the new treatment, relative to an individual on the
standard treatment. The standard treatment would then be the more suitable
in terms of promoting longevity. On the other hand, when the end-point is
the recovery from some disease state, values of ϕ less than unity would be
found when the effect of the new treatment is to speed up the recovery time.
In these circumstances, the new treatment would be superior to the standard.
The quantity ϕ−1 is therefore termed the acceleration factor.
The acceleration factor can also be interpreted in terms of the median
survival times of patients on the new and standard treatments, tN (50) and
tS (50), say. These values are such that SN {tN (50)} = SS {tS (50)} = 0.5.
Now, under the accelerated failure time model, SN {tN (50)} = SS {tN (50)/ϕ},
and so it follows that tN (50) = ϕtS (50). In other words, under the accelerated
failure time model, the median survival time of a patient on the new treatment
is ϕ times that of a patient on the standard treatment. In fact, the same
argument can be used for any percentile of the survival time distribution.
This means that the pth percentile of the survival time distribution for a
patient on the new treatment, tN (p), is such that tN (p) = ϕtS (p), where tS (p)
is the pth percentile for the standard treatment. This interpretation of the
acceleration factor is particularly appealing to clinicians.
From the relationship between the survivor function, probability density
function and hazard function given in Equation (1.4), the relationship between
the density and hazard functions for individuals in the two treatment groups
is
fN (t) = ϕ−1 fS (t/ϕ),
and
hN (t) = ϕ−1 hS (t/ϕ).
Now let X be an indicator variable that takes the value zero for an individual
in the group receiving the standard treatment, and unity for one who receives
the new treatment. The hazard function for the ith individual can then be
expressed as
hi (t) = ϕ−xi h0 (t/ϕxi ), (6.5)
where xi is the value of X for the ith individual in the study. Putting xi = 0
in this expression shows that the function h0 (t) is the hazard function for an
individual on the standard treatment. This is again referred to as the baseline
hazard function. The hazard function for an individual on the new treatment
is then ϕ−1 h0 (t/ϕ).
The parameter ϕ must be non-negative, and so it is convenient to set
ϕ = eα . The accelerated failure time model in Equation (6.5) then becomes
hi (t) = e−αxi h0 (t/eαxi ), (6.6)
so that the hazard function for an individual on the new treatment is now
e−α h0 (t/eα ).
6.3.1 ∗ Comparison with the proportional hazards model

To illustrate the difference between a proportional hazards model and the
accelerated failure time model, again suppose that the survival times of indi-
viduals in two groups, Group I and Group II, say, are to be modelled. Further
suppose that for the individuals in Group I, the hazard function is given by
{
0.5 if t 6 1,
h0 (t) =
1.0 if t > 1,
where the time-scale is measured in months. This type of hazard function

arises from a piecewise exponential model, since a constant hazard in each
time interval implies exponentially distributed survival times, with different
means, in each interval. This model provides a simple way of representing a
variable hazard function, and may be appropriate in situations where there is
a constant short-term risk that increases abruptly after a threshold time.
Now let hP (t) and hA (t) denote the hazard functions for individuals in
Group II under a proportional hazards model and an accelerated failure time
model, respectively. Consequently, we may write
hP (t) = ψh0 (t),
and
hA (t) = ϕ−1 h0 (t/ϕ),
∫t
for the two hazard functions. Using the result S(t) = exp{− 0
h(u) du}, the
baseline survivor function is
{ −0.5t
e if t 6 1,
S0 (t) =
e−0.5−(t−1) if t > 1.
Since S0 (t) > 0.61 if t < 1, the median occurs in the second part of the
survivor function and is when exp{−0.5 − (t − 1)} = 0.5. The median survival
time for those in Group I is therefore 1.19 months.
The survivor functions for the individuals in Group II under the two models
are
ψ
SP (t) = [S0 (t)] ,
and
SA (t) = S0 (t/ϕ),
respectively.
To illustrate the difference between the hazard functions under propor-
tional hazards and accelerated failure time models, consider the particular case
where ψ = ϕ−1 = 2.0. The median survival time for individuals in Group II
is 0.69 months under the proportional hazards model, and 0.60 months under
the accelerated failure time model. The hazard functions for the two groups
under both models are shown in Figure 6.3 and the corresponding survivor
functions are shown in Figure 6.4.
Under the accelerated failure time model, the increase in the hazard for
Group II from 1.0 to 2.0 occurs sooner than under the proportional hazards
model. The ‘kink’ in the survivor function also occurs earlier under the accel-
erated failure time model.
2.0
1.5
Hazard function
1.0 h0(t)
0.5
0.0
0.0 0.5 1.0 1.5 2.0 2.5
Time
Figure 6.3 The hazard functions for individuals in Group I, h0 (t), and in Group
II under (i) a proportional hazards model (—) and (ii) an accelerated failure time
model (· · ·).
1.0
0.8
Survivor function
0.6
0.4
0.2
S0(t)
0.0
0.0 0.5 1.0 1.5 2.0 2.5
Time
Figure 6.4 The survivor functions for individuals in Group I, S0 (t), and in Group
II, under (i) a proportional hazards model (—) and (ii) an accelerated failure time
model (· · ·).
6.3.2 The percentile-percentile plot
The percentile-percentile plot, also known as the quantile-quantile plot or the
Q-Q plot, provides an exploratory method for assessing the validity of an
accelerated failure time model for two groups of survival data. Recall that
the pth percentile of a distribution is the value t(p), which is such that the
estimated survivor function at time t(p) is 1 − (p/100), for any value of p in
the interval (0, 100). The pth percentile is therefore such that
( )
100 − p
t(p) = S −1 .
100
Now let t0 (p) and t1 (p) be the pth percentiles estimated from the survivor
functions of the two groups of survival data. The values of p might be taken
to be 10, 20, . . . , 90, so long as the number of observations in each of the two
groups is not too small. The percentiles of the two groups may therefore be
expressed as
( ) ( )
100 − p 100 − p
t0 (p) = S0−1 , t1 (p) = S1−1 ,
100 100
where S0 (t) and S1 (t) are the survivor functions for the two groups. It then
follows that
S1 {t1 (p)} = S0 {t0 (p)} , (6.7)
for any given value of p.
Under the accelerated failure time model, S1 (t) = S0 (t/ϕ), and so the pth
percentile for the second group, t1 (p), is such that
S1 {t1 (p)} = S0 {t1 (p)/ϕ} .
Using Equation (6.7),
S0 {t0 (p)} = S0 {t1 (p)/ϕ} ,
and hence
t0 (p) = ϕ−1 t1 (p).
Now let t̂0 (p), t̂1 (p) be the estimated percentiles in the two groups, so that
( ) ( )
100 − p 100 − p
t̂0 (p) = Ŝ0−1 , t̂1 (p) = Ŝ1−1 .
100 100
A plot of the quantity t̂0 (p) against t̂1 (p), for suitably chosen values of p, should
give a straight line through the origin if the accelerated failure time model is
appropriate. The slope of this line will be an estimate of the acceleration
factor, ϕ−1 . This plot may therefore be used in an exploratory assessment
of the adequacy of the accelerated failure time model. In this sense, it is an
analogue of the log-cumulative hazard plot, used in Section 4.4.1 to examine
the validity of the proportional hazards model.

In this example, the data on the survival times of women with breast tumours
that were negatively or positively stained, originally given as Example 1.2 in
Chapter 1, is used to illustrate the percentile-percentile plot. The percentiles
of the distribution of the survival times in each of the two groups can be
estimated from the Kaplan-Meier estimate of the respective survivor functions.
These are given in Table 6.1.
Table 6.1 Estimated percentiles of the distributions

of survival times for women with tumours that were
positively or negatively stained.
Percentile Negative staining Positive staining
10 47 13
20 69 26
30 148 35
40 181 48
50 – 61
60 – 113
70 – 143
80 – –
90 – –
The relatively small numbers of death times, and the censoring pattern in
the data from the two groups of women, mean that not all of the percentiles
can be estimated. The percentile-percentile plot will therefore have just four
pairs of points. For illustration, this is shown in Figure 6.5. The points fall on
a line that is reasonably straight, suggesting that the accelerated failure time
model would not be inappropriate. However, this conclusion must be regarded
with some caution in view of the limited number of points in the graph.
The slope of a straight line drawn through the points in Figure 6.5 is
approximately equal to 3, which is a rough estimate of the acceleration factor.
The interpretation of this is that for women whose tumours were positively
stained, the disease process is speeded up by a factor of three, relative to
those whose tumours were negatively stained. We can also say that the median
survival time for women with negatively stained tumours is estimated to be
three times that of women with positively stained tumours.
6.4 The general accelerated failure time model

The accelerated failure time model in Equation (6.6) can be generalised to the
situation where the values of p explanatory variables have been recorded for
each individual in a study. According to the general accelerated failure time
model, the hazard function of the ith individual at time t, hi (t), is then such
THE GENERAL ACCELERATED FAILURE TIME MODEL 233
Percentile for negative staining 200
150
100
50
0
0 20 40 60
Percentile for positive staining
Figure 6.5 Percentile-percentile plot for the data on the survival times of breast can-
cer patients.
that
hi (t) = e−ηi h0 (t/eηi ), (6.8)
where
ηi = α1 x1i + α2 x2i + · · · + αp xpi
is the linear component of the model, in which xji is the value of the jth
explanatory variable, Xj , j = 1, 2, . . . , p, for the ith individual, i = 1, 2, . . . , n.
As in the proportional hazards model, the baseline hazard function, h0 (t),
is the hazard of death at time t for an individual for whom the values of
the p explanatory variables are all equal to zero. The corresponding survivor
function for the ith individual is
Si (t) = S0 {t/ exp(ηi )},
where S0 (t) is the baseline survivor function.

Parametric accelerated failure time models are unified by the adoption of
a log-linear representation of the model, described in the sequel. This repre-
sentation shows that the accelerated failure time model for survival data is
closely related to the general linear model used in regression analysis. More-
over, this form of the model is adopted by most computer software packages
for accelerated failure time modelling.
6.4.1 ∗ Log-linear form of the accelerated failure time model
Consider a log-linear model for the random variable Ti , associated with the
lifetime of the ith individual in a survival study, according to which
log Ti = µ + α1 x1i + α2 x2i + · · · + αp xpi + σϵi . (6.9)
In this model, α1 , α2 , . . . , αp are the unknown coefficients of the values of

p explanatory variables, X1 , X2 , . . . , Xp , with values x1i , x2i , . . . , xpi for the
ith individual, and µ, σ are two further parameters, known as the intercept
and scale parameter, respectively. The quantity ϵi is a random variable used
to model the deviation of the values of log Ti from the linear part of the
model, and ϵi is assumed to have a particular probability distribution. In
this formulation of the model, the α-parameters reflect the effect that each
explanatory variable has on the survival times; positive values suggest that
the survival time increases with increasing values of the explanatory variable,
and vice versa.
To show the relationship between this representation of the model and that
in Equation (6.8), consider the survivor function of Ti , the random variable
associated with the survival time of the ith individual. Using Equation (6.9),
this is given by
Si (t) = P(Ti > t) = P {exp(µ + α′ xi + σϵi ) > t} ,
where α′ xi = α1 x1i + α2 x2i + · · · + αp xpi .

Now, Si (t) can be written in the form
Si (t) = P {exp(µ + σϵi ) > t/ exp(α′ xi )} ,
and the baseline survivor function, S0 (t), the survivor function of an individual
for whom x = 0, is
S0 (t) = P {exp(µ + σϵi ) > t} .
It then follows that
Si (t) = S0 {t/ exp(α′ xi )}, (6.10)
which is the general form of the survivor function for the ith individual in an
accelerated failure time model. In this version of the model, the acceleration
factor is exp(−α′ xi ) for the ith individual. The corresponding relationship
between the hazard functions is obtained using Equation (1.5) of Chapter 1.
Specifically, taking logarithms of both sides of Equation (6.10), multiplying
by −1, and differentiating with respect to t, leads to
hi (t) = exp(−α′ xi )h0 {t/ exp(α′ xi )},
which is the model in Equation (6.8) with ηi = α′ xi .

The log-linear formulation of the model can also be used to give a general
form of the survivor function for the ith individual, which is
Si (t) = P(Ti > t) = P(log Ti > log t).

THE GENERAL ACCELERATED FAILURE TIME MODEL 235
Si (t) = P(µ + α1 x1i + α2 x2i + · · · + αp xpi + σϵi > log t),

( )
= P ϵi > . (6.11)
σ
If we now write Sϵi (ϵ) for the survivor function of the random variable ϵi in the
log-linear model of Equation (6.9), the survivor function of the ith individual
can, from Equation (6.11), be expressed as
( )
Si (t) = Sϵi . (6.12)
σ
This result shows how the survivor function for Ti can be found from the
survivor function of the distribution of ϵi . The result also demonstrates that
an accelerated failure time model can be derived from many probability dis-
tributions for ϵi , although some are more tractable than others.
A general expression for the pth percentile of the distribution of survival
times also follows from the results in this section. The pth percentile for the
ith individual, ti (p), is given by
100 − p
Si {ti (p)} = ,
100
and using Equation (6.11),
( )
log ti (p) − µ − α1 x1i − α2 x2i − · · · − αp xpi 100 − p
P ϵi > = .
σ 100
If ϵi (p) is used to denote the pth percentile of the distribution of ϵi , then

100 − p
Sϵi {ϵi (p)} = P {ϵi > ϵi (p)} = .
100
Consequently,
log ti (p) − µ − α1 x1i − α2 x2i − · · · − αp xpi
ϵi (p) = ,
σ
and so
ti (p) = exp {σϵi (p) + µ + α1 x1i + α2 x2i + · · · + αp xpi } (6.13)
is the pth percentile of the distribution of survival times for the ith individual.
Note that the percentile in Equation (6.13) can be written in the form
ti (p) = exp(α1 x1i + α2 x2i + · · · + αp xpi )t0 (p),
where t0 (p) is the pth percentile for a baseline individual for whom all ex-
planatory variables take the value zero. This confirms that the α-coefficients
can be interpreted in terms of the effect of the explanatory variables on the
percentiles of the distribution of survival times.
The cumulative hazard function of the distribution of Ti is given by Hi (t) =
− log Si (t), and from Equation (6.12),
( )
Hi (t) = − log Sϵi ,
σ
( )
= Hϵi , (6.14)
σ
where Hϵi (ϵ) = − log Sϵi (ϵ) is the cumulative hazard function of ϵi . The cor-
responding hazard function, found by differentiating Hi (t) in Equation (6.14)
with respect to t, is
( )
1 log t − µ − α1 x1i − α2 x2i − · · · − αp xpi
hi (t) = hϵi , (6.15)
σt σ
where hϵi (ϵ) is the hazard function of the distribution of ϵi .

The distributions of ϵi that are most often used in accelerated failure time
modelling are such that their percentiles, ϵi (p), have a simple form. Models
based on such distributions are described in the following section.
6.5 Parametric accelerated failure time models

Particular choices for the distribution of ϵi in the log-linear formulation of the
accelerated failure time model, described in Section 6.4.1, lead to distributions
for the random variable associated with the survival time of the ith individual.
But the representation of the model in Equation (6.9) invariably leads to
different parameterisations of the models from those given in Sections 5.1 and
6.1. Parametric accelerated failure time models based on the Weibull, log-
logistic and lognormal distributions for the survival times are most commonly
used in practice, and so these models are described in detail, and summarised
in Section 6.5.4.
6.5.1 The Weibull accelerated failure time model

Suppose that survival times are assumed to have a Weibull distribution with
scale parameter λ and shape parameter γ, written W (λ, γ), so that the baseline
hazard function is
h0 (t) = λγtγ−1 .
The hazard function for the ith individual is then, from Equation (6.8), given
by
hi (t) = e−ηi λγ(e−ηi t)γ−1 = (e−ηi )γ λγtγ−1 ,
so that the survival time of this individual has a W (λe−γηi , γ) distribution.
The Weibull distribution is therefore said to possess the accelerated failure
PARAMETRIC ACCELERATED FAILURE TIME MODELS 237
time property. Indeed, this is the only probability distribution that has both
the proportional hazards and accelerated failure time properties.
Because the Weibull distribution has both the proportional hazards prop-
erty and the accelerated failure time property, there is a direct correspondence
between the parameters under the two models. If the baseline hazard function
is the hazard function of a W (λ, γ) distribution, the survival times under the
general proportional hazards model in Equation (5.35) of Chapter 5 have a
W (λ exp(β ′ xi ), γ) distribution, while those under the accelerated failure time
model have a W (λ exp(−γα′ xi ), γ) distribution. It then follows that when
the coefficients of the explanatory variables in the linear component of the
accelerated failure time model are multiplied by −γ, we get the correspond-
ing β-coefficients in the proportional hazards model. In the particular case of
comparing two groups, an acceleration factor of ϕ−1 = e−α under the accel-
erated failure time model corresponds to a hazard ratio of ϕ−γ = e−γα in a
proportional hazards model.
In terms of the log-linear representation of the model in Equation (6.9),
if Ti has a Weibull distribution, then ϵi does in fact have a type of extreme
value distribution known as the Gumbel distribution. This is an asymmetric
distribution with mean 0.5772, which is Euler’s constant, and variance π 2 /6.
The survivor function of the Gumbel distribution is given by
Sϵi (ϵ) = exp(−eϵ ),
for −∞ < ϵ < ∞, and the cumulative hazard and hazard functions of this
distribution are given by Hϵi (ϵ) = eϵ , and hϵi (ϵ) = eϵ , respectively.
To show that the random variable Ti = exp(µ + α′ xi + σϵi ) has a Weibull
distribution, from Equation (6.12), the survivor function of Ti is given by
{ ( )}
Si (t) = exp − exp . (6.16)
σ
This can be expressed in the form
( )
Si (t) = exp −λi t1/σ ,
where
λi = exp {−(µ + α1 x1i + α2 x2i + · · · + αp xpi )/σ} ,
which, from Equation (5.5) of Chapter 5, is the survivor function of a Weibull
distribution with scale parameter λi , and shape parameter σ −1 . Consequently,
Equation (6.16) is the accelerated failure time representation of the survivor
function of the Weibull model described in Section 5.6 of Chapter 5.
The cumulative hazard and hazard functions for the Weibull accelerated
failure time model can be found directly from the survivor function in Equa-
tion (6.16), or from Hϵi (ϵ) and hϵi (ϵ), using the general results in Equa-
tions (6.14) and (6.15). We find that the cumulative hazard function is
( )
Hi (t) = − log Si (t) = exp ,
σ
which can also be expressed as λi t1/σ , and the hazard function is given by
( )
hi (t) = exp , (6.17)
σt σ
−1
or hi (t) = λi σ −1 tσ −1 .
We now reconcile this form of the model with that for the Weibull pro-
portional hazards model. From Equation (5.37) of Chapter 5, the survivor
function for the ith individual is
Si (t) = exp {− exp(β1 x1i + β2 x2i + · · · + βp xpi )λtγ } , (6.18)
in which λ and γ are the parameters of the Weibull baseline hazard function.
There is a direct correspondence between Equation (6.16) and Equation (6.18),
in the sense that
λ = exp(−µ/σ), γ = σ −1 , βj = −αj /σ,
for j = 1, 2, . . . , p. We therefore deduce that the log-linear model where
1
log Ti = {− log λ − β1 x1i − β2 x2i − · · · − βp xpi + ϵi } ,
γ
and in which ϵi has a Gumbel distribution, provides an alternative represen-

tation of the Weibull proportional hazards model.
In this form of the model, the pth percentile of the survival time distribu-
tion for the ith individual is the value ti (p), which is such that Si {ti (p)} =
1−(p/100), where Si (t) is as given in Equation (6.16). Straightforward algebra
leads to the result that
[ { ( )} ]
100 − p ′
ti (p) = exp σ log − log + µ + α xi (6.19)
100
for that individual. Equivalently, the pth percentile of the distribution of ϵi ,

ϵi (p), is such that
{ } 100 − p
exp −eϵi (p) = ,
100
so that { ( )}
100 − p
ϵi (p) = log − log ,
100
and the general result in Equation (6.13) leads directly to Equation (6.19).
The survivor function and hazard function of the Weibull model follow
from Equations (6.16) and (6.17), and Equation (6.19) enables percentiles to
be estimated directly.
6.5.2 The log-logistic accelerated failure time model
Now suppose that the survival times have a log-logistic distribution. If the
baseline hazard function in the general accelerated failure time model in Equa-
tion (6.8) is derived from a log-logistic distribution with parameters θ, κ, this
function is given by
eθ κtκ−1
h0 (t) = .
1 + eθ t κ
Under the accelerated failure time model, the hazard of death at time t for
the ith individual is
hi (t) = e−ηi h0 (e−ηi t),
where ηi = α1 x1i + α2 x2i + · · · + αp xpi is a linear combination of the values
of p explanatory variables for the ith individual. Consequently,
e−ηi eθ κ(e−ηi t)κ−1

hi (t) = ,
1 + eθ (e−ηi t)κ
that is,
eθ−κηi κtκ−1
hi (t) = .
1 + eθ−κηi tκ
It then follows that the survival time for the ith individual also has a log-
logistic distribution with parameters θ−κηi and κ. The log-logistic distribution
therefore has the accelerated failure time property. However, this distribution
does not have the proportional hazards property.
The log-linear form of the accelerated failure time model in Equation (6.9)
also provides a representation of the log-logistic distribution. Suppose that
in this formulation, ϵi now has a logistic distribution with zero mean and
variance π 2 /3, so that the survivor function of ϵi is
1
Sϵi (ϵ) = .
1 + eϵ
Using Equation (6.12), the survivor function of Ti is then
{ ( )}−1
Si (t) = 1 + exp . (6.20)
σ
From Equation (6.2), the survivor function of Ti , when Ti has a log-logistic

distribution with parameters θ −κηi , κ, where ηi = α1 x1i +α2 x2i +· · ·+αp xpi ,
is
1
Si (t) = θ−κη i tκ
.
1+e
On comparing this expression with that for the survivor function in Equa-
tion (6.20), we see that the parameters θ and κ can be expressed in terms of
µ and σ. Specifically,
θ = −µ/σ, κ = σ −1 ,
and this shows that the accelerated failure time model with log-logistic survival
times can also be formulated in terms of a log-linear model. This is the form
of the model that is usually adopted by computer software, and so computer-
based parameter estimates are usually estimates of µ and σ, rather than θ
and κ.
The cumulative hazard and hazard functions of the distribution of ϵi are
such that
Hϵi (ϵ) = log (1 + eϵ ) ,
and ( )−1
hϵi (ϵ) = 1 + e−ϵ ,
respectively. Equations (6.14) and (6.15) may then be used to obtain the cu-
mulative hazard, and hazard function, of Ti . In particular, the hazard function
for the ith individual is
{ [ ( )]}−1
hi (t) = 1 + exp − .
σt σ
(6.21)
Estimates of quantities such as the acceleration factor, or the median
survival time, can be obtained directly from the estimates of µ, σ and
the αj ’s. For example, the acceleration factor for the ith individual is
exp{−(α1 x1i +α2 x2i +· · ·+αp xpi )} and the pth percentile of the survival time
distribution, from Equation (6.20), or the general result in Equation (6.13), is
{ ( ) }
p
ti (p) = exp σ log + µ + α1 x1i + α2 x2i + · · · + αp xpi .
100 − p
The median survival time is simply
ti (50) = exp {µ + α1 x1i + α2 x2i + · · · + αp xpi } , (6.22)
and so an estimate of the median can be straightforwardly obtained from the

estimated values of the parameters in the model.
6.5.3 The lognormal accelerated failure time model

If the survival times are assumed to have a lognormal distribution, the baseline
survivor function is given by
( )
log t − µ
S0 (t) = 1 − Φ ,
σ
where µ and σ are two unknown parameters. Under the accelerated failure
time model, the survivor function for the ith individual, is then
Si (t) = S0 (e−ηi t),

where ηi = α1 x1i + α2 x2i + · · · + αp xpi is a linear combination of the values
of p explanatory variables for the ith individual. Therefore,
( )
log t − ηi − µ
Si (t) = 1 − Φ , (6.23)
σ
which is the survivor function for an individual whose survival times have a
lognormal distribution with parameters µ + ηi and σ. The lognormal distri-
bution therefore has the accelerated failure time property.
In the log-linear formulation of the model, the random variable associated
with the survival time of the ith individual has a lognormal distribution if
log Ti is normally distributed. We therefore take ϵi in Equation (6.9) to have
a standard normal distribution, so that the survivor function of ϵi is
Sϵi (ϵ) = 1 − Φ(ϵ).
The cumulative hazard, and hazard function, of ϵi are
Hϵi (ϵ) = − log {1 − Φ(ϵ)} ,
and
fϵi (ϵ)
hϵi (ϵ) = ,
Sϵi (ϵ)
respectively, where fϵi (ϵ) is the density function of a standard normal random
variable, given by
1 ( )
fϵi (ϵ) = √ exp −ϵ2 /2 .
(2π)
The random variable Ti , in the general accelerated failure time model, then
has a lognormal distribution with parameters µ + α′ xi and σ. The survivor
function of Ti is as given in Equation (6.23), and the hazard function is found
from Equation (6.15).
The pth percentile of the distribution of Ti , from Equation (6.13), is
{ }
ti (p) = exp σΦ−1 (p/100) + µ + α1 x1i + α2 x2i + · · · + αp xpi ,
and, in particular, t(50) = exp(µ + α′ xi ) is the median survival time for the
ith individual.
6.5.4 Summary
It is convenient to summarise the models and results that have been described
in this section, so that the different parameterisations of the distributions used
in accelerated failure time models can clearly be seen.
The general accelerated failure time model for the survival time of the ith
of n individuals, for whom x1i , x2i , . . . , xpi are the values of p explanatory
variables, X1 , X2 , . . . , Xp , is such that the random variable associated with
the survival time, Ti , can be expressed in the form
log Ti = µ + α1 x1i + α2 x2i + · · · + αp xpi + σϵi .
Particular distributions for Ti are derived from assumptions about the dis-
tribution of ϵi in this model. The survivor function and hazard function of
the distributions of ϵi , that lead to commonly used accelerated failure time
models for the survival times, are summarised in Table 6.2.
Table 6.2 Summary of parametric accelerated failure time models.

Distribution of Ti Sϵi (ϵ) hϵi (ϵ) Percentile,
{ ( ϵi (p))}
Exponential exp(−eϵ ) eϵ log − log 100−p
{ 100
( 100−p )}
Weibull exp(−eϵ ) eϵ log − log
( )
100
Log-logistic (1 + eϵ )−1 (1 + e−ϵ )−1 log p

100−p
exp(−ϵ2 /2)
Lognormal 1 − Φ(ϵ) √
{1−Φ(ϵ)} (2π)
Φ−1 (p/100)
The cumulative hazard function of ϵi is found from Hϵi (ϵ) = − log Sϵi (ϵ),
and, if desired, the density function of ϵi is fϵi (ϵ) = hϵi (ϵ) Sϵi (ϵ). From the
survivor and hazard function of ϵi , the survivor and hazard function of Ti can
be found from
( )
Si (t) = Sϵi ,
σ
and ( )
hi (t) = hϵ ,
σt i σ
results that were first given in Equations (6.12) and (6.15), respectively.
The pth percentile of the distribution of ϵi is also given in Table 6.2, from
which ti (p), the pth percentile of the survival times for the ith individual, can
be found from
ti (p) = exp {σϵi (p) + µ + α1 x1i + α2 x2i + · · · + αp xpi } ,
given as Equation (6.13).

The log-linear representation of the Weibull and log-logistic models leads
to parameterisations of the survival time distributions that differ from those
used in Sections 5.1 and 6.1, when the distributions were first presented. The
link between the two sets of parameters is summarised in Table 6.3, which
includes the number of the equation that gives the survivor function of each
distribution in terms of the original parameters.
For the proportional hazards representation of the Weibull model, where
hi (t) = exp(β ′ xi )h0 (t),

FITTING AND COMPARING MODELS 243
Table 6.3 Summary of the parameterisation of accelerated failure time models.

Distribution of Ti Equation number Parameterisation of survivor function
Exponential (5.2) λ = e−µ , γ = 1 (σ = 1)
Weibull (5.5) λ = e−µ/σ , γ = 1/σ,
Log-logistic (6.2) θ = −µ/σ, κ = 1/σ
the correspondence between βj , and αj in the accelerated time model, is such

that βj = −αj /σ, j = 1, 2, . . . , p. This shows how the β-parameters in a
Weibull proportional hazards model, which represent log-hazard ratios, can
be obtained from the fitted α-parameters in an accelerated failure time model.
6.6 ∗ Fitting and comparing accelerated failure time models

Accelerated failure time models are fitted using the method of maximum likeli-
hood. The likelihood function is best derived from the log-linear representation
of the model, after which iterative methods are used to obtain the estimates.
The likelihood of the n observed survival times, t1 , t2 , . . . , tn , is, from Expres-
sion (5.9) in Chapter 5, given by
∏
n
δ 1−δi
L(α, µ, σ) = {fi (ti )} i {Si (ti )} ,
i=1
where fi (ti ) and Si (ti ) are the density and survivor functions for the ith
individual at ti , and δi is the event indicator for the ith observation, so that
δi is unity if the ith observation is an event and zero if it is censored. Now,
from Equation (6.12),
Si (ti ) = Sϵi (zi ) ,
where zi = (log ti − µ − α1 x1i − α2 x2i − · · · − αp xpi )/σ, and differentiation
with respect to t gives
1
fi (ti ) = fϵ (zi ) .
σti i
The likelihood function can then be expressed in terms of the survivor and
density functions of ϵi , giving
∏
n
(σti )−δi {fϵi (zi )} i {Sϵi (zi )}
δ 1−δi
L(α, µ, σ) = .
i=1
The log-likelihood function is then
∑
n
log L(α, µ, σ) = {− δi log(σti ) + δi log fϵi (zi ) + (1 − δi ) log Sϵi (zi )} ,
i=1
(6.24)
and the maximum likelihood estimates of the p + 2 unknown parameters, µ, σ
and α1 , α2 , . . . , αp , are found by maximising this function using the Newton-
Raphson procedure, described in Section 3.3.3.
Note that the expression∑n for the log-likelihood function in Equation (6.24)
includes the term − i=1 δi log ti , which does not involve any unknown pa-
rameters. This term may therefore be omitted from the log-likelihood function,
as noted in Section 5.6.1 of Chapter 5, in the context of the Weibull propor-
tional hazards model. Indeed, log-likelihood values given by most computer
software
∑n for accelerated failure time modelling do not include the value of
− i=1 δi log ti .
After fitting a model, the value of the statistic −2 log L̂ can be computed,
and used in making comparisons between nested models, just as for the pro-
portional hazards model. Specifically, to compare two nested models, the dif-
ference in the values of the statistic −2 log L̂ for the two models is calculated,
and compared with percentage points of the chi-squared distribution, with
degrees of freedom equal to the difference in the number of α-parameters
included in the linear component of the model.
Once a suitable model has been identified, estimates of the survivor and
hazard functions may be obtained and plotted. The fitted model can be inter-
preted in terms of the estimated value of the acceleration factor for particular
individuals, or in terms of the median and other percentiles of the distribution
of survival times. In particular, the estimated pth percentile of the distribution
of survival times, for an individual whose vector of values of the explanatory
variables is xi , is, from Equation (6.13), given by
t̂i (p) = exp{σ̂ϵi (p) + µ̂ + α̂1 x1i + α̂2 x2i + · · · + α̂p xpi }.
The estimated percentiles of the assumed distribution of survival times

are functions of the parameter estimates in the log-linear accelerated fail-
ure time model, and so the standard error of these estimates can be found
using the general result in Equation (5.44) of Chapter 5. Specifically, the vec-
tor θ̂ now has p + 2 components, namely µ̂, α̂1 , α̂2 , . . . , α̂p , σ̂, and var (θ̂) is
the variance-covariance matrix of these parameter estimates. Equation (5.44)
shows how the variance of a function of the parameter estimates can be ob-
tained from the vector of derivatives, d(θ̂), of the estimated percentiles. How-
ever, it is much more straightforward to first obtain the variance of log t̂i (p),
since the derivatives of log t̂i (p), with respect to µ̂, α̂1 , α̂2 , . . . , α̂p , σ̂, are then
1, x1i , x2i , . . . , xpi , ϵi (p), respectively. Equation (5.46) is then used to obtain
the standard error of the estimated percentile. Confidence intervals for a per-
centile will usually be formed from an interval estimate for log ti (p). Most
computer software for survival analysis provides the standard error of speci-
fied percentiles.

In this example, accelerated failure time models are fitted to the data on the
survival times of women with breast cancer. The Weibull accelerated failure
time model is first considered. A log-linear model for the random variable
associated with the survival time of the ith woman, Ti , is such that
log Ti = µ + αxi + σϵi ,
where ϵi has a Gumbel distribution, µ, σ and α are unknown parameters, and
xi is the value of an explanatory variable, X, associated with staining, such
that xi = 0 if the ith woman is negatively stained and xi = 1 if positively
stained. When this model is fitted, we find that µ̂ = 5.854, σ̂ = 1.067, and
α̂ = −0.997.
The acceleration factor, e−αxi , is estimated by e0.997 = 2.71 for a woman
with positive staining. The time to death of a woman with a positively stained
tumour is therefore accelerated by a factor of about 2.7 under this model. This
is in broad agreement with the estimated slope of the percentile-percentile plot
for this data set, found in Example 6.2.
The estimated survivor function for the ith woman is given by
{ ( )}
log t − µ̂ − α̂xi
Ŝi (t) = exp − exp ,
σ̂
from Equation (6.16), and may be plotted against t for the two possible values
of xi . The median survival time of the ith woman under the Weibull acceler-
ated failure time model, using Equation (6.19), is
ti (50) = exp {σ log(log 2) + µ + αxi } .
The estimated median survival time for a woman with negative staining
(xi = 0) is 236 days, while that for women with positive staining (xi = 1)
is 87 days, as in Example 5.6. Note that the ratio of the two medians is 2.71,
which is the acceleration factor. The median survival time for women with
positively stained tumours is therefore about one-third that of women whose
tumours were negatively stained.
The estimated hazard function for the ith woman is, from Equation (6.17),
given by ( )
−1 − µ̂ − α̂xi
ĥi (t) = σ̂ −1 tσ̂ −1 exp ,
σ̂
that is,
ĥi (t) = 0.937 t−0.063 exp(− 5.486 + 0.934 xi ).
A plot of this function for the two groups of women is shown in Figure 6.6.
In the proportional hazards representation of the Weibull model, given in
Section 5.6 of Chapter 5, the hazard of death at time t for the ith woman is
where xi takes the value zero if the ith woman had a negatively stained tu-
mour, and unity if the tumour was positively stained. For the Weibull distri-
bution, the baseline hazard function is
h0 (t) = λγtγ−1 ,
0.010
0.008
0.006
0.004
0.002
0.000
0 50 100 150 200 250
Survival time
Figure 6.6 Estimated hazard functions under the Weibull accelerated failure time
model for women with positively stained (—) and negatively stained (· · ·) tumours.
which is the hazard function for women with negatively stained tumours, and
hence
hi (t) = eβxi λγtγ−1 .
The corresponding estimated values of the parameters λ, γ and β are given

by λ̂ = exp(−µ̂/λ̂) = 0.00414, γ̂ = 1/σ̂ = 0.937 and β̂ = −α̂/σ̂ = 0.997. The
correspondence between the Weibull accelerated failure time model and the
Weibull proportional hazards model means that the hazard ratio under the
latter model is e−α/σ = eβ , which is estimated to be 2.55. This is in agreement
with the value found in Example 5.6.
We now fit the log-logistic accelerated failure time model to the same data
set. The log-linear form of the model now leads to µ̂ = 5.461, σ̂ = 0.805 and
α̂ = −1.149. The acceleration factor is e−α , which is estimated by 3.16. This
is slightly greater than that found under the Weibull accelerated failure time
model.
The median survival time for the ith woman under this model is, from
Equation (6.22), given by
exp(µ + αxi ),
from which the estimated median survival time for a women with negative
staining is 235 days, while that for women with positive staining is 75 days.
These values are very close to those obtained under the Weibull accelerated
failure time model.
The estimated hazard function for the ith woman is now
{ [ ( )]}−1
1 log t − µ̂ − α̂xi
ĥi (t) = 1 + exp − ,
σ̂t σ̂
from Equation (6.21), which is

{ }−1
ĥi (t) = 1.243 t−1 1 + t−1.243 exp (6.787 − 1.428 xi ) .
A graph of this function for the two groups of women is shown in Figure 6.7.
This can be compared with the graph in Figure 6.6.
0.010
0.008
0.006
0.004
0.002
0.000
0 50 100 150 200 250
Survival time
Figure 6.7 Estimated hazard functions under the log-logistic accelerated failure time
model for women with positively stained (—) and negatively stained (· · ·) tumours.
The hazard functions for those with negative staining are quite similar
under the two models. However, the hazard function for those with positive
staining under the log-logistic model is different from that under the Weibull
model. The values of the statistic −2 log L̂ for the fitted Weibull and log-
logistic models are 121.77 and 118.495. On this basis, the log-logistic model
is a slightly better fit. An analysis of residuals, to be discussed in Chapter
7, may help in choosing between these two models, although with this small
data set, such an analysis is unlikely to be very informative.
Finally, in terms of the parameterisation of the model given in Section
6.1.1, the baseline hazard function is
eθ κtκ−1
h0 (t) = ,
1 + eθ t κ
and so the hazard function for the ith woman in the study is
eθ−καxi κtκ−1
hi (t) = .
1 + eθ−καxi tκ
The corresponding estimated values of θ and κ are given by θ̂ = −µ̂/σ̂ =

−6.787, and κ̂ = 1/σ̂ = 1.243.

In a further illustration of modelling survival data using the log-logistic ac-
celerated failure time model, the data from a clinical trial to compare two
treatments for prostatic cancer are considered. These data were first given
in Example 1.4, and analysed using a Cox regression model in Examples 3.6
and 3.12.
To identify the terms that should be in the linear component of the log-
logistic accelerated failure time model, the procedure described in Example 3.6
can again be followed. The values of the statistic −2 log L̂ on fitting models
with all combinations of the four prognostic variables, Age, Shb, Size and
Index, are shown in Table 6.4.

fitted to the data from Example 1.4.
none 35.806
Age 35.752
Shb 35.700
Size 27.754
Index 27.965
Age + Shb 35.657
Age + Size 27.652
Age + Index 27.859
Shb + Size 27.722
Shb + Index 26.873
Size + Index 23.112
Age + Shb + Size 27.631
Age + Shb + Index 26.870
Age + Size + Index 23.002
Shb + Size + Index 22.895
Age + Shb + Size + Index 22.727
As in Example 3.6, the variables Size and Index are the ones that are
needed in the model. When either of these variables is omitted, the corre-
sponding increase in the value of −2 log L̂ is significant, and neither Age nor
Shb reduce −2 log L̂ by a significant amount when they are added to the model.
When the term corresponding to the treatment effect, Treat, is added to the
model that contains Size and Index, −2 log L̂ decreases to 21.245. When this
reduction of 1.867 is compared with percentage points of a chi-squared distri-
bution on 1 d.f., the reduction is not significant at the 10% level (P = 0.172).
There is no evidence of any interaction between Treat and the prognostic vari-
ables Size and Index, and so the conclusion is that there is no statistically
significant treatment effect.
The magnitude of the treatment effect can be assessed by calculating the
acceleration factor. According to the log-linear form of the model, the random
variable associated with the survival time of the ith patient, Ti , is such that
log Ti = µ + α1 Sizei + α2 Indexi + α3 Treati + σϵi ,
in which ϵi has a logistic distribution, Size i and Index i are the values of the
tumour size and Gleason index for the ith individual, and Treat i is zero if the
ith individual is in the placebo group and unity if in the treated group. The
maximum likelihood estimates of the unknown parameters in this model are
given by µ̂ = 7.661, σ̂ = 0.338, α̂1 = −0.029, α̂2 = −0.293 and α̂3 = 0.573.
The values of the α’s suggests that the survival time tends to be shorter for
larger values of the tumour size and tumour index, and longer for individuals
assigned to the active treatment.
Using Equation (6.20), the fitted survivor function for the ith patient is
[ { }]−1
log t − µ̂ − α̂1 Sizei − α̂2 Indexi − α̂3 Treati
Ŝi (t) = 1 + exp ,
σ̂
which can be written in the form

{ ( )}−1
Ŝi (t) = 1 + t1/σ̂ exp ζ̂i ,
where
1
ζ̂i = {−µ̂ − α̂1 Sizei − α̂2 Indexi − α̂3 Treati } ,
σ̂
that is,
ζ̂i = −22.645 + 0.085 Size i + 0.865 Index i − 1.693 Treat i .
The corresponding estimated hazard function can be found by differentiating

the estimated cumulative hazard function, Ĥi (t) = − log Ŝi (t), with respect
to t. This gives
1 { ( )}−1
ĥi (t) = 1 + t−1/σ̂ exp −ζ̂i ,
σ̂t
a result that can also be obtained directly from the hazard function given in
Equation (6.21).
The estimated acceleration factor for an individual in the treated group,
relative to an individual in the control group, is e−0.573 = 0.56. The inter-
pretation of this result is that after allowing for the size and index of the
tumour, the effect of the treatment with diethylstilbestrol is to slow down
the progression of the cancer by a factor of about 2. This effect might be of
clinical importance, even though it is not statistically significant. However,
before accepting this interpretation, the adequacy of the fitted model should
be checked using suitable diagnostics.
A confidence interval for the acceleration factor is found by exponentiat-
ing the confidence limits for the logarithm of the acceleration factor. In this
example, the logarithm of the acceleration factor for the treatment effect is
the estimated coefficient of Treat in the model for the hazard function, mul-
tiplied by −1, which is −0.573, and the standard error of this estimate is
0.473. Thus, a 95% confidence interval for the acceleration factor has limits
of exp{−0.573 ± 1.96 × 0.473}, and the required interval is from 0.70 to 4.48.
Notice that this interval estimate includes unity, which is consistent with the
earlier finding of a non-significant treatment difference.
Finally, in terms of the parameterisation of the model in Section 6.1.1, the
fitted hazard function for the ith patient, i = 1, 2, . . . , 38, is given by
ĥi (t) = e−η̂i ĥ0 (e−η̂i t),
where
η̂i = −0.029 Size i − 0.293 Index i + 0.573 Treat i ,
and from Equation (6.1),
eθ̂ κ̂tκ̂−1
ĥ0 (t) = .
1 + eθ̂ tκ̂
The estimated parameters in this form of the estimated baseline hazard func-
tion, ĥ0 (t), are given by θ̂ = −22.644 and κ̂ = 2.956. A graph of this function
is shown in Figure 6.8.
This figure indicates that the baseline hazard is increasing over time. Com-
parison with the baseline hazard function for a fitted Weibull model, also
shown in this figure, indicates that under the log-logistic model, the estimated
baseline hazard function does not increase quite so rapidly.
6.7 ∗ The proportional odds model

In the general proportional odds model, the odds of an individual surviving
beyond some time t are expressed as
Si (t) S0 (t)
= eηi , (6.25)
1 − Si (t) 1 − S0 (t)
where
ηi = β1 x1i + β2 x2i + · · · + βp xpi
is a linear combination of the values of p explanatory variables, X1 , X2 , . . .,
Xp , measured on the ith individual, and S0 (t), the baseline survivor function,
is the survivor function for an individual whose explanatory variables all take
the value zero.
THE PROPORTIONAL ODDS MODEL 251
0.0000025
0.0000020
0.0000015
0.0000010
0.0000005
0.0000000
0 20 40 60 80
Survival time
Figure 6.8 Estimated baseline hazard function for the fitted log-logistic model (—)
and a fitted Weibull model (· · ·).
In this model, the explanatory variables act multiplicatively on the odds

of survival beyond t. The logarithm of the ratio of the odds of survival beyond
t for the ith individual, relative to an individual for whom the explanatory
variables are all equal to zero, is therefore just ηi . The model is therefore a
linear model for the log-odds ratio.
Now consider the particular case of a two-group study, in which individuals
receive either a standard treatment or new treatment. Let the single indicator
variable X take the value zero if an individual is on the standard treatment
and unity if on the new. The odds of the ith individual surviving beyond time
t is then
Si (t) S0 (t)
= eβxi ,
1 − Si (t) 1 − S0 (t)
where xi is the value of X for the ith individual, i = 1, 2, . . . , n. Thus, if SN (t)
and SS (t) are the survivor functions for individuals on the new and standard
treatments, respectively,
SN (t) SS (t)
= eβ ,
1 − SN (t) 1 − SS (t)
and the log-odds ratio is simply β. The parameters in the linear component
of the model therefore have an immediate interpretation.
As for the proportional hazards model, a non-parametric estimate of the
baseline hazard function can be obtained. The model is then fitted by estimat-
ing the β-parameters in the linear component of the model, and the baseline
survivor function, from the data. A method for accomplishing this has been
described by Bennett (1983a), but details will not be included here. Fully
parametric versions of the proportional odds model can be derived by using
a specific probability distribution for the survival times. One such model is
described below in Section 6.7.1.
One particularly important property of the proportional odds model con-
cerns the ratio of the hazard function for the ith individual to the baseline
hazard, hi (t)/h0 (t). It can be shown that this ratio converges from the value
e−ηi at time t = 0, to unity at t = ∞. To show this, the model in Equa-
tion (6.25) can be rearranged to give
{ }−1
Si (t) = S0 (t) e−ηi + (1 − e−ηi )S0 (t) ,
and taking logarithms, we get

{ }
log Si (t) = log S0 (t) − log e−ηi + (1 − e−ηi )S0 (t) . (6.26)
Using the general result from Equation (1.5), the hazard function is
d
hi (t) = − log Si (t),
dt
and so
(1 − e−ηi )f0 (t)
hi (t) = h0 (t) − ,
e−ηi + (1 − e−ηi )S0 (t)
after differentiating both sides of Equation (6.26) with respect to t, where
f0 (t) is the baseline probability density function. After some rearrangement,
this equation becomes
f0 (t)
hi (t) = h0 (t) − . (6.27)
(eηi − 1)−1 + S0 (t)
From Equation (1.4), we also have that h0 (t) = f0 (t)/S0 (t) and substituting
for f0 (t) in Equation (6.27) gives
{ }
S0 (t)
hi (t) = h0 (t) 1 − ηi .
(e − 1)−1 + S0 (t)
Finally, after further rearrangement, the hazard ratio is given by

hi (t) −1
= {1 + (eηi − 1)S0 (t)} .
h0 (t)
As t increases from 0 to ∞, the baseline survivor function decreases monoton-
ically from 1 to 0. When S0 (t) = 1, the hazard ratio is e−ηi and as t increases
to ∞, the hazard ratio converges to unity.
In practical applications, it is common for the hazard functions obtained
for patients in two or more groups to converge with time. For example, in
a follow-up study of patients in a clinical trial, the effect on survival of the
THE PROPORTIONAL ODDS MODEL 253
treatment, or the initial stage of disease, may wear off. Similarly, in studies
where a group of patients with some disease are being compared with a control
group of disease-free individuals, an effective cure of the disease would lead
to the survival experience of each group becoming more similar over time.
This suggests that the proportional odds model, with its property of conver-
gent hazard functions, might be of considerable value. However, there are two
reasons why this general model has not been widely used in practice. The
first of these is that computer software is not generally available for fitting
a proportional odds model in which the baseline survivor function is of un-
specified form. The second is that the model is likely to give similar results
to a Cox regression model that includes a time-dependent variable to produce
non-proportional hazards. This particular approach to modelling survival data
with non-proportional hazards was outlined in Section 4.4.3, and is considered
more fully in Chapter 8.
6.7.1 The log-logistic proportional odds model

If survival times for individuals are assumed to have a log-logistic distribution,
the baseline survivor function is
{ }−1
S0 (t) = 1 + eθ tκ ,
where θ and κ are unknown parameters. The baseline odds of survival beyond
time t are then given by
S0 (t)
= e−θ t−κ .
1 − S0 (t)
The odds of the ith individual surviving beyond time t are therefore
Si (t)
= eηi −θ t−κ ,
1 − Si (t)
and so the survival time of the ith individual has a log-logistic distribution
with parameters θ − ηi and κ. The log-logistic distribution therefore has the
proportional odds property, and the distribution is the natural one to use in
conjunction with the proportional odds model. In fact, this is the only distribu-
tion to share both the accelerated failure time property and the proportional
odds property.
This result also means that the estimated β-coefficients in the linear com-
ponent of the proportional odds model in Equation (6.25) can be obtained by
multiplying the α-coefficients in the log-logistic accelerated failure time model
of Equation (6.20) by κ̂ = σ̂ −1 , where σ̂ is the estimated value of the param-
eter σ. The coefficients of the explanatory variables under the proportional
odds model can then be obtained from those under the accelerated failure
time model, and vice versa. The results of a survival analysis based on a pro-
portional odds model can therefore be interpreted in terms of an acceleration
factor or the ratio of the odds of survival beyond some time, whichever is the
more convenient.
As for other models for survival data, the proportional odds model can be
fitted using the method of maximum likelihood. Alternative models may then
be compared on the basis of the statistic −2 log L̂.
In a two-group study, a preliminary examination of the likely suitability
of the model can easily be undertaken. The log-odds of the ith individual
surviving beyond time t are
{ }
Si (t)
log = βxi − θ − κ log t,
1 − Si (t)
where xi is the value of an indicator variable that takes the value zero if an
individual is in one group and unity if in the other. The Kaplan-Meier estimate
of the survivor function is then obtained for the individuals
{ in each group} and
the estimated log-odds of survival beyond time t, log Ŝi (t)/[1 − Ŝi (t)] , are
plotted against log t. If the plot shows two parallel straight lines, this would
indicate that the log-logistic model was appropriate. If the lines were straight
but not parallel, this would suggest that the parameter κ in the model was
not the same for each treatment group. Parallel curves in this plot suggest
that although the proportional odds assumption is valid, the survival times
cannot be taken to have a log-logistic distribution.

In this example to illustrate the use of the proportional odds model, the model
is fitted to the data on the survival times of breast cancer patients. In order to
assess the likely suitability of the proportional odds model, the Kaplan-Meier
estimate of the survivor function for the negatively and positively stained
women is computed. For the two groups of women, the log-odds of survival
beyond time t are estimated and plotted against log t. The resulting graph is
shown in Figure 6.9. The lines are reasonably straight and parallel, and so
we go on to use the log-logistic proportional odds model to summarise these
data.
The model can be fitted using software for fitting the log-logistic accel-
erated failure time model. In Example 6.3, this latter model was fitted to
the data on the survival of breast cancer patients. The estimated values of
κ and θ in the proportional odds model are 1.243 and −6.787, respectively,
the same as those in the accelerated failure time model. However, the esti-
mated value of β in the linear component of the proportional odds model is
β̂ = −1.149×1.243 = −1.428. This is an estimate of the logarithm of the ratio
of the odds of a positively stained woman surviving beyond time t, relatively to
one who is negatively stained. The corresponding odds ratio is e−1.428 = 0.24,
so that the odds of a woman surviving beyond t are about four times greater
if that woman has a negatively stained tumour.
SOME OTHER DISTRIBUTIONS FOR SURVIVAL DATA 255
4
Log-odds of survival
-1
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
Figure 6.9 Estimated values of the log-odds of survival beyond t plotted against log t
for women with positively stained (∗) and negatively stained (•) tumours.
6.8 ∗ Some other distributions for survival data

Although a number of distributions for survival data have already been con-
sidered in some detail, there are others that can be useful in specific circum-
stances. Some of these are mentioned in this section.
When the hazard of death is expected to increase or decrease with time in
the short term, and to then become constant, a hazard function that follows a
general exponential curve or Mitscherlich curve may be appropriate. We would
then take the hazard function to be
h(t) = θ − βe−γt ,
where θ > 0, β > 0 and γ > 0. This is essentially a Gompertz hazard function,
defined in Section 5.9, with an additional constant. The general shape of this
function is depicted in Figure 6.10. This function has a value of θ − β when
t = 0 and increases to a horizontal asymptote at a hazard of θ. Similarly the
function
h(t) = θ + βe−γt ,
where θ > 0, β > 0 and γ > 0, could be used to model a hazard which
decreases from θ + β to a horizontal asymptote at θ.
Using Equation (1.6), the corresponding survivor function can be found,
from which the probability density function can be obtained. The probability
distribution corresponding to this specification of the hazard function is known
as the Gompertz-Makeham distribution.
Hazard function
0
0
Time
Figure 6.10 An asymptotic hazard function, where h(t) = θ − βe−γt .
To model a hazard function that decreases and then increases symmetri-

cally about the minimum value, a quadratic hazard function might be suitable.
Thus, if
h(t) = θ + βt + γt2 ,
for values of θ, β and γ which give the required shape of hazard and ensure
that h(t) > 0, explicit forms for the survivor function and probability density
function can be obtained.
Another form of hazard function that decreases to a single minimum and
increases thereafter is the ‘bathtub’ hazard. The model with
β
h(t) = αt +
1 + γt
provides a straightforward representation of this form of hazard, and corre-
sponding expressions for the survivor and density functions can be found.
Each of the models described in this section can be fitted by constructing a
log-likelihood function, using the result in Expression (5.38) of Chapter 5, and
maximising this with respect to the unknown model parameters. In principle,
the unknown parameters in the hazard function can also depend on the values
of explanatory variables. Non-linear optimisation routines can then be used
to maximise the log-likelihood.
6.9 ∗ Flexible parametric models

The main advantage of the Cox regression model for survival analysis is often
perceived to be the flexibility of the baseline hazard function, since it can
FLEXIBLE PARAMETRIC MODELS 257
accommodate the pattern needed for any particular data set. In contrast, the
parametric models described in this chapter lead to baseline hazard functions
that depend on a very small number of unknown parameters, and so have
a limited ability to capture the underlying form of a baseline hazard. This
advantage of the Cox model is often overplayed. Since the baseline hazard
function in a Cox model is estimated from the data to give a step-function
with jumps at each event time, it can behave very erratically, as illustrated
in Example 3.14 of Chapter 3. Also, the estimated survivor function for an
individual with given characteristics is constant between event times, so that it
may not be possible to estimate the survival rate at a precise time. Moreover,
survival rates at times that are beyond the longest event time in a data set
cannot be estimated.
Example 6.6 Recurrence-free survival in breast cancer patients

A cohort study of breast cancer in a large number of hospitals was carried
out by the German Breast Cancer Study Group to compare three cycles of
chemotherapy with six cycles, and also to investigate the effect of additional
hormonal treatment consisting of a daily dose of 30 mg of tamoxifen over
two years. The patients in the study had primary histologically proven non-
metastatic node-positive breast cancer who had been treated with mastec-
tomy. The response variable of interest is recurrence-free survival, which is
the time from entry to the study until a recurrence of the cancer or death.
Earlier analyses of the data had shown that recurrence-free survival was not
affected by the number of cycles of chemotherapy, and so only the factor as-
sociated with whether or not a patient received tamoxifen is included in this
example. In addition to this treatment factor, data were available on patient
age, menopausal status, size and grade of the tumour, number of positive
lymph nodes, progesterone and oestrogen receptor status. Further details on
the background to the study are given by Schumacher et al. (1994).
The data in this example relate to data from 41 centres and 686 patients
with complete data, and were included in Sauerbrei and Royston (1999). The
variables in this data set are as follows:
Id: Patient number

Treat: Hormonal treatment (0 = no tamoxifen, 1 = tamoxifen)
Age: Patient age (years)
Men: Menopausal status (1 = premenopausal, 2 = postmenopausal)
Size: Tumour size (mm)
Grade: Tumour grade (1 – 3)
Nodes: Number of positive lymph nodes
Prog: Progesterone receptor status (femtomoles)
Oest: Oestrogen receptor status, (femtomoles)
Time: Recurrence-free survival time (days)
Status: Event indicator (0 = censored, 1 = relapse or death)
Data for 20 of the 686 patients in this breast cancer study are shown in
Table 6.5.
Table 6.5 Recurrence-free survival times of 20 breast cancer patients.

Id Treat Age Men Size Grade Nodes Prog Oest Time Status
1 0 70 2 21 2 3 48 66 1814 1
2 1 56 2 12 2 7 61 77 2018 1
3 1 58 2 35 2 9 52 271 712 1
4 1 59 2 17 2 4 60 29 1807 1
5 0 73 2 35 2 1 26 65 772 1
6 0 32 1 57 3 24 0 13 448 1
7 1 59 2 8 2 2 181 0 2172 0
8 0 65 2 16 2 1 192 25 2161 0
9 0 80 2 39 2 30 0 59 471 1
10 0 66 2 18 2 7 0 3 2014 0
11 1 68 2 40 2 9 16 20 577 1
12 1 71 2 21 2 9 0 0 184 1
13 1 59 2 58 2 1 154 101 1840 0
14 0 50 2 27 3 1 16 12 1842 0
15 1 70 2 22 2 3 113 139 1821 0
16 0 54 2 30 2 1 135 6 1371 1
17 0 39 1 35 1 4 79 28 707 1
18 1 66 2 23 2 1 112 225 1743 0
19 1 69 2 25 1 1 131 196 1781 0
20 0 55 2 65 1 4 312 76 865 1
In analysing these data, we might be interested in the treatment effect

after adjusting for other explanatory variables, and in estimating the adjusted
hazard functions for the two treatment groups.
It is difficult to discern pattern in the estimated baseline hazard func-
tion in a fitted Cox regression model, as the value of the hazard at each
event time is determined from the number of events and the number at
risk at that time. A smoothed estimate of this function is therefore desir-
able. One approach to this is to apply a smoothing process to the estimated
cumulative baseline hazard, followed by numerical differentiation to give a
smooth estimate of the hazard function itself. Although this estimate is use-
ful in a descriptive analysis, it is not straightforward to use the fitted curve
to estimate survival rates from the fitted model, nor to validate the fit of
such a curve.
A much better approach is to model the underlying baseline hazard para-
metrical, but allowing this function to have a greater flexibility than that
allowed by the fully parametric models in Chapters 5 and 6. A particularly
appealing approach was described Royston and Parmar (2002), who showed
how the Weibull proportional hazards model and the log-logistic proportional
odds model can be extended to provide a flexible parametric modelling pro-
cedure.
6.9.1 The Royston and Parmar model
We begin with the Weibull model for the hazard of death at time t, where
hi (t) = exp(β ′ xi )h0 (t),
in which the baseline hazard function is h0 (t) = λγtγ−1 , λ and γ are unknown
parameters that determine the scale and shape of the underlying Weibull
distribution, and xi is the vector of values of p explanatory variables for the
ith of n individuals. The corresponding cumulative hazard function is
∫ t
Hi (t) = hi (u) du = exp(β ′ xi )λtγ ,
0
and so the log-cumulative hazard is
log Hi (t) = β ′ xi + log λ + γ log t.
Now set ηi = β ′ xi , and let γ0 = log λ, γ1 = γ and y = log t, so that the

log-cumulative hazard function for the Weibull model can be written
log Hi (t) = γ0 + γ1 y + ηi .
This formulation shows that the log-cumulative hazard function is linear in

y = log t.
The next step is to generalise the linear term in y to a natural cubic spline
in y. To define this, the range of values of y is divided into a number of
intervals, where the boundary between each interval is called a knot. In the
simplest case, we take the smallest and largest y-values and divide the range
into two halves. There would then be two boundary knots at the extreme
values of y and one internal knot between them. A cubic expression in y is
then fitted between adjacent knots. For example, suppose that the range of
values of y is from kmin to kmax , and that one knot is specified at the point
where y = k1 . A cubic expression in y is then defined for y ∈ (kmin , k1 ) and
for y ∈ (k1 , kmax ). These two cubic expressions are then constrained to have
a smooth join at the internal knot k1 , to give a cubic spline. Finally, a linear
term in y is assumed for y < kmin and for y > kmax , which leads to a restricted
cubic spline. This is illustrated in Figure 6.11, which shows a restricted cubic
spline with boundary knots at kmin and kmax and an internal knot at k1 .
The algebraic form of a restricted cubic spline in y with one knot is
γ0 + γ1 y + γ2 ν1 (y),
where
ν1 (y) = (y − k1 )3+ − λ1 (y − kmin )3+ − (1 − λ1 )(y − kmax )3+ ,
with
(y − a)3+ = max{0, (y − a)3 },
min 1 max
Figure 6.11 Restricted cubic spline with an internal knot at k1 and boundary knots
at kmin and kmax .
for any value a, and

kmax − k1
λ1 = .
kmax − kmin
The model for the log-cumulative hazard function then becomes


 γ0 + γ1 y + η i , y < kmin ,


 γ0 + γ1 y − γ2 {λ1 (y − kmin )3 } + ηi , y ∈ (kmin , k1 ),
log Hi (t) = γ0 + γ1 y + γ2 {(y − k1 )3 − λ1 (y − kmin )3 } + ηi , y ∈ (k1 , kmax ),


 γ0 + γ1 y + γ2 {(y − k1 )3 − λ1 (y − kmin )3


−(1 − λ1 )(y − kmax )3 } + ηi , y > kmax .
When y > kmax , the expression for log Hi (t) simplifies to
γ0 + γ1 y + γ2 (kmax − k1 )(kmin − k1 )(3y − kmin − k1 − kmax ) + ηi ,
which confirms that log Hi (t) is a linear function of y for y < kmin and y >
kmax , and cubic functions of y for values of y between kmin and k1 , and between
k1 and kmax .
The flexibility of the parametric model for log Hi (t) can be increased by
increasing the number of internal knots. The greater the number of knots,
the more complex the curve. In general, for a model with m internal knots,
non-linear terms ν1 (y), ν2 (y), . . . , νm (y) are defined, so that for a model with
m knots,
log Hi (t) = γ0 + γ1 y + γ2 ν1 (y) + · · · + γm+1 νm (y) + ηi , (6.28)

where for the jth knot at kj , j = 1, 2, . . . , m,
νj (y) = (y − kj )3+ − λj (y − kmin )3+ − (1 − λj )(y − kmax )3+ , (6.29)
and
kmax − kj
λj = .
kmax − kmin
The model defined by Equation (6.28) is the Royston and Parmar model.
The extended parametric form of the baseline hazard function means that
the survival times no longer have a Weibull distribution under this model,
nor any other recognisable distribution, although the model still assumes pro-
portional hazards amongst the explanatory variables. The model in Equa-
tion (6.28) can also be expressed in terms of a baseline cumulative hazard
function, H0 (t), by writing Hi (t) = exp(ηi )H0 (t), where
H0 (t) = exp{γ0 + γ1 y + γ2 ν1 (y) + · · · + γm+1 νm (y)}. (6.30)
The corresponding survivor function is Si (t) = exp{−Hi (t)}, for the ith indi-
vidual, so that
Si (t) = exp{− exp[γ0 + γ1 y + γ2 ν1 (y) + · · · + γm+1 νm (y) + ηi ]}, (6.31)
which can also be expressed as
Si (t) = {S0 (t)}exp(ηi ) ,
where S0 (t) = exp{−H0 (t)} is the baseline survivor function.

In terms of hazard functions, the model in Equation (6.28) can be expressed
as
dHi (t)
hi (t) = = exp(ηi )h0 (t). (6.32)
dt
In this equation, the baseline hazard function, h0 (t), is found by differentiating
H0 (t) in Equation (6.30) with respect to t, which gives
h0 (t) = t−1 {γ1 + γ2 ν1′ (y) + · · · + γm+1 νm

′
(y)}H0 (t),
where


 0, y 6 kmin ,

′ −3λj (y − kmin )2 , y ∈ (kmin , kj ),
νj (y) =

 3(y − kj )2 − 3λj (y − kmin )2 , y ∈ (kj , kmax ),

3(y − kj )2 − 3λj (y − kmin )2 − 3(1 − λj )(y − kmax )2 , y > kmax ,
is the first derivative of νj (y) with respect to y, j = 1, 2, . . . , m, and y = log t.

This baseline hazard involves m + 2 unknown parameters, γ0 , γ1 , . . . , γm+1 ,
and provides an approximate parametric representation of the non-parametric
baseline hazard function in a corresponding Cox regression model.
6.9.2 Number and position of the knots
The position chosen for the knots will affect the form of the parametric model
for the cumulative hazard function. Although some authors have proposed
using automatic or data driven procedures, Royston and Parmar (2002) cau-
tion against this, on the grounds that they can be unduly affected by local
features of the data set, and that it is difficult to take account of this when
constructing standard errors of parameter estimates. They suggest that the
boundary knots are taken at the smallest and largest values of the logarithms
of observed event times. Internal knots are then taken to give equal spacing
between percentiles of the distribution of the logarithms of the uncensored
survival times. Specifically, one internal knot would be placed at the median
of the uncensored values of log t, two internal knots would be placed at the
33% and 67% percentiles, three knots would be placed at the quartiles, and so
on. Usually, no more than 4 or 5 knots will be needed in practical applications.
To determine the number of knots to be used, we start with zero knots, and
fit the standard Weibull parametric model. A model with one internal knot
is then fitted, and compared to the model with zero knots, and then a model
with two internal knots is then fitted and compared to that with one knot,
and so on. As the position of the knots changes as their number is increased,
for example from one at the 50th percentile to two at the 33rd and 67th
percentiles of the distribution of the uncensored log survival times, models
with different numbers of knots are not necessarily nested. This means that
the −2 log L̂ statistic cannot generally be used to determine whether increasing
the number of knots gives a significant improvement in the fit. Instead, the
AIC statistic introduced in Section 3.6.1 of Chapter 3, can be used. We will
take the AIC statistic to be −2 log L̂ + 2q, where q is the number of unknown
parameters in the fitted model, so that for a model with p β-parameters and
m knots, q = p + m + 2. Models with the smallest values of this statistic
are generally the most suitable, and so the number of knots is increased until
the AIC statistic cannot be reduced any further. Models can be compared
in a similar manner using the BIC statistic, also defined in Section 3.6.1 of
Chapter 3.
Experience in using these parametric models suggests that the number of
knots is barely affected by covariate adjustment. This means that any variable
selection can be based on standard Weibull or Cox models, and once the
explanatory variables for inclusion in the Royston and Parmar model have
been determined, the form of the spline function can be found by increasing
the number of knots in the parametric model for the adjusted log-cumulative
baseline hazard function.
6.9.3 Fitting the model

The model for the log-cumulative hazard function in Equation (6.28), or equiv-
alently, the model for the hazard or survivor functions in Equations (6.31)
and (6.32), is fully parametric. This means that the model can be fitted using
the method of maximum likelihood. From Equation (5.38) in Section 5.6.1 of
Chapter 5, the likelihood function is
∏
n
δ
{hi (ti )} i Si (ti ),
i=1
where the survivor and hazard functions for the ith individual at time ti ,
Si (ti ) and hi (ti ), are given in Equations (6.31) and (6.32), and δi is the event
indicator. The logarithm of this likelihood function can be maximised using
standard optimisation routines, and this leads to fitted survivor and hazard
functions that are smooth functions of the survival time t. The fitting process
also leads to standard errors of the parameter estimates and functions of them,
such as estimates of the hazard and survivor functions at any given time.

We now return to the data on recurrence-free survival in breast cancer pa-
tients, introduced in Example 6.6. As an initial step in the analysis of these
data, the suitability of a Weibull model is investigated using a log-cumulative
hazard plot, stratified by treatment. Although the presence of additional ex-
planatory variables hinders the interpretation of such a plot for just one fac-
tor, the graph shown in Figure 6.12 does not exhibit straight lines for the two
treatment groups. The assumption of a Weibull distribution for the survival
times is therefore not appropriate. However, the vertical separation of two
curves in this plot appears constant suggesting that the proportional hazards
assumption for the treatment effect is valid.
-1
-2
-3
-4
-5
-6
-7
4 5 6 7 8
Log survival time
Figure 6.12 Log-cumulative hazard plot for the women not on tamoxifen (•) and
those in the tamoxifen group (∗).
At this stage, a variable selection process may be used to determine which
of the other explanatory factors, Age, Men, Size, Grade, Nodes, Prog, and
Oest, are needed in the model in addition to Treat, but in this example, all of
them will be included. Royston and Parmar models with increasing numbers
of knots are then fitted. Table 6.6 gives the value of the AIC statistic for the
models fitted, where the model with zero knots is the standard Weibull model.
Table 6.6 Values of the AIC statistic

for models with up to four knots.
Number of knots AIC
0 5182.4
1 5147.9
2 5147.1
3 5146.1
4 5148.2
There is a large reduction in the value of the AIC statistic on fitting a

restricted cubic spline with one internal knot, indicating that this model is a
substantial improvement on the standard Weibull model. On adding a second
and third knot, the AIC statistic is reduced further, but the decrease is much
less than that when one knot is added. On adding a fourth internal knot, the
AIC statistic increases. This analysis shows that the best fitting model has
three knots, although this is not a marked improvement on that with just one
knot. The fitted model with three knots is most simply expressed in terms of
the estimated log-cumulative hazard function for the ith patient, given by
log Ĥi (t) = exp(β̂1 Treati + β̂2 Agei + β̂3 Meni + β̂4 Sizei + β̂5 Gradei
+β̂6 Nodesi + β̂7 Progi + β̂8 Oesti )Ĥ0 (t),
where
Ĥ0 (t) = − exp{γ̂0 + γ̂1 y + γ̂2 ν1 (y) + γ̂3 ν2 (y) + γ̂4 ν3 (y)}
is the baseline cumulative hazard function for a model with 3 knots, y = log t
and the functions ν1 (y), ν2 (y), ν3 (y) are defined in Equation (6.29).
A visual summary of the fit of the Royston and Parmar models is shown in
Figure 6.13. This figure shows the adjusted baseline survivor function on fit-
ting a Cox regression model that contains the 8 explanatory variables, shown
as a step-function. In addition, the adjusted baseline survivor function for a
Weibull model, and models with one and three internal knots, is shown. This
figure confirms that the underlying risk adjusted baseline survivor function
from the Cox model is not well fitted by a Weibull model. A Royston and Par-
mar model with one knot tracks the estimated Cox baseline survivor function
much more closely, and that with three knots gives an improved performance
at the longer survival times.
The estimated values of the parameters and their standard errors in the
Royston and Parmar model with three knots, are given in Table 6.7. Also
1.0
0.9
0.8
0.7
0.6
0.5
0 500 1000 1500 2000 2500 3000
Survival time
Figure 6.13 Risk adjusted survivor functions for a fitted Cox regression model (—),
Weibull model ( ) and Royston-Parmar models with 1 (·······) and 3 (- - -) knots.
Table 6.7 Parameter estimates and their standard errors for a Royston and Parmar
model with 3 knots and a Cox regression model.
Parameter Royston and Parmar model Cox regression model
Estimate se (Estimate) Estimate se (Estimate)
β1 −0.3386 0.1290 −0.3372 0.1290
β2 −0.0096 0.0093 −0.0094 0.0093
β3 0.2753 0.1831 0.2670 0.1833
β4 0.0077 0.0040 0.0077 0.0039
β5 0.2824 0.1059 0.2801 0.1061
β6 0.0497 0.0074 0.0499 0.0074
β7 −0.0022 0.0006 −0.0022 0.0006
β8 0.0002 0.0004 0.0002 0.0004
γ0 −20.4691 3.2958
γ1 2.9762 0.5990
γ2 −0.4832 0.5873
γ3 1.4232 0.8528
γ4 −0.9450 0.4466
shown in this table are the estimated β-parameters in a fitted Cox model and
their standard errors. The estimates and their standard errors for the two
models are very similar. The adjusted hazard ratio for a patient on tamoxifen
relative to one who is not, is 0.71 under both models, so that the hazard of
recurrence of cancer or death is lower for patients on tamoxifen.
The Royston and Parmar model has the advantage of providing a paramet-
ric estimate of the baseline hazard. The fitted baseline hazard function, ad-
justed for the 8 explanatory variables, for the Weibull model and the Royston
and Parmar spline models with one and three knots, is shown in Figure 6.14.
The Royston and Parmar model indicates that the underlying hazard function
is unimodal, and so it is not surprising that the Weibull model is a poor fit.
0.0004
0.0003
0.0002
0.0001
0.0000
0 500 1000 1500 2000 2500 3000
Survival time
Figure 6.14 Adjusted baseline hazard functions for a fitted Weibull model (—) and
a Royston-Parmar model with 1 (·······) and 3 knots (- - -).
6.9.4 Proportional odds models

Parametric models based on cubic splines can also be used in conjunction with
proportional odds models described in Section 6.7. The general proportional
odds model for the odds of survival beyond time t, Si (t)/{1 − Si (t)}, is such
that
Si (t) ′ S0 (t)
= eβ x i ,
1 − Si (t) 1 − S0 (t)
where xi is the vector of values of the explanatory variables for the ith indi-
vidual and S0 (t) is the baseline survivor function, that is the survivor function
for an individual whose explanatory variables all take the value zero. If a log-
logistic model with parameters θ, κ is assumed for the survival times, as in
Section 6.7.1, the log odds of survival beyond time t is, from Equation (6.2),
{ }−1
S(t) = 1 + eθ tκ . Taking this to be the baseline survivor function, S0 (t)
is then such that
( )
S0 (t)
log = −θ − κ log t = γ0 + γ1 y,
1 − S0 (t)
where γ0 = −θ, γ1 = −κ, and the model is linear in y = log t. Extending this
model to incorporate non-linear terms to give a restricted cubic spline with
m internal knots, and using the same notation as in Equation (6.28), we get
( )
S0 (t)
log = γ0 + γ1 y + γ2 ν1 (y) + · · · + γm+1 νm (y),
1 − S0 (t)
which is analogous to the expression for the cumulative baseline hazard func-
tion in Equation (6.30). The proportional odds model that includes spline
terms can then be expressed as
( )
Si (t)
log = γ0 + γ1 y + γ2 ν1 (y) + · · · + γm+1 νm (y) + ηi ,
1 − Si (t)
where ηi = β ′ xi .
This flexible proportional odds model is used in just the same way as the
flexible proportional hazards model, and is particularly suited to situations
where the hazard function is unimodal. The model can also be expressed in
terms of the odds of an event occurring before time t, log[F (t)/{1 − F (t)}],
and as this is simply − log[S(t)/{1 − S(t)}], the resulting parameter estimates
will only differ in sign.
The model can be fitted using the method of maximum likelihood, as
described in Section 6.9.3, and leads to an estimate of the survivor function.
Estimates of the corresponding hazard and cumulative hazard functions can
then straightforwardly be obtained.

The unimodal hazard function identified in Example 6.7 suggests that a log-
logistic model may be a better fit to the recurrence-free survival times for
women with breast cancer. Table 6.8 gives the value of the AIC statistic for
the model for the log odds of surviving beyond time t that has a flexible
baseline survivor function, where the model with zero knots is the standard
proportional odds model. As in Example 6.7, the explanatory variables Age,
Men, Size, Grade, Nodes, Prog and Oest are included in the model, in addition
to the treatment factor, Treat. Notice that the values of the AIC statistic
behave somewhat erratically for models with 1, 2 and 3 knots, with the value of
the −2 log L̂ statistic increasing slightly on adding a fourth knot. This feature
can occur when the models being fitted are barely distinguishable, and since
the models being fitted have increasing numbers of knots at different locations,
not all the models are nested. In this case, a model for the log odds of surviving
Table 6.8 Values of the AIC statistic

for models with up to four knots.
Number of knots AIC
0 5154.0
1 5135.4
2 5135.8
3 5134.1
4 5136.2
beyond time t that contains three knots is the best fit, although the fit of this
model is barely distinguishable from the model with just one knot.
For the model with three knots, the estimated parameter associated with
the treatment effect is 0.5287. The ratio of the odds of surviving beyond time
t for a patient on tamoxifen, relative to one who is not, is exp(0.5287) = 1.70,
and so the odds of a patient on tamoxifen surviving beyond any given time are
1.7 times that for a patient who has not received that treatment. This result
is entirely consistent with the corresponding hazard ratio for the treatment
effect, given in Example 6.7.
6.10 ∗ Modelling cure rates
In survival analysis, it is generally assumed that all individuals will eventually

experience the end-point of interest, if the follow-up period is long enough.
This is certainly the case if the end-point is death from any cause. However, in
some studies, a substantial proportion of individuals may not have experienced
the end-point before the end of the study. This may be because the treatment
has effectively cured the patient. For example, in a cancer trial, interest may
centre on a comparison of two treatments, where the end-point is death from
a particular type of cancer. If the treatment cures the individual, there will
be a number of patients who do not die from the cancer during the course
of a relatively long follow-up period. This can lead to a larger proportion of
censored observations than is usually encountered, which is sometimes referred
to as heavy censoring. However, strictly speaking, an individual who does not
die during the follow-up period has a censored time, whereas those who are
cured cannot die from the disease under study.
Individuals who have been cured, or more generally, those who can no
longer experience the event under study, will eventually fail for some reason,
but may remain alive throughout a very long time period. In this situation, the
survivor function estimated from a group of individuals will tend to level off at
a value greater than zero. It may then be assumed that the population consists
of a mixture of individuals: those who are susceptible to the end-point, and
those who are not. The latter then correspond to the cured individuals, and
the proportion with prolonged survival is called the cured fraction. Standard
MODELLING CURE RATES 269
methods of survival analysis can then be adapted, so that the probability of
cure is modelled simultaneously with the time to the event.
Models for the time to an event can be extended to incorporate a cured
fraction, π, the probability of being cured. In a fully parametric model, the
survivor function becomes
S(t) = (1 − π)Sn (t) + π, (6.33)
where Sn (t) is the survivor function of the non-cured individuals. In Equa-

tion (6.33), S(t) is the overall survivor function for a group consisting of
cured and non-cured individuals, and the model is termed a parametric mix-
ture model. As t → ∞, the survivor function of non-cured individuals tends
to zero, and so, from Equation (6.33), S(t) tends to π, the probability of cure.
The corresponding hazard function for the whole group is
d log S(t) fn (t)
h(t) = − = , (6.34)
dt Sn (t) + π/(1 − π)
where
d
fn (t) = − Sn (t)
dt
is the density function of the non-cured individuals.
If a proportional hazards model can be assumed for the survival time of
the non-cured individuals, the hazard of death at time t in such an individual
is
hni (t) = exp(β ′ xi )hn0 (t),
where β ′ xi = β1 x1i + β2 x2i + · · · + βp xpi is a linear combination of the val-
ues of p explanatory variables, X1 , X2 , . . . , Xp , measured on this individual,
and hn0 (t) is the baseline hazard function of the non-cured individuals. The
survivor function for the non-cured individuals is then
′
Sni (t) = [Sn0 (t)]exp(β xi ) ,
∫t
where Sn0 (t) = exp{− 0 hn0 (u) du} is their baseline survivor function.
In addition, the probability of being cured may depend on a number of
explanatory variables, especially those relating to treatment group, denoted
Z1 , Z2 , . . . , Zp . The dependence of the probability that the ith of n individ-
uals is cured on these variables can then be modelled by taking the logistic
transformation of the cured fraction to be a linear combination of their values,
z1i , z2i , . . . , zpi . Then,
( )
πi
logit (πi ) = log = ϕ 0 + ϕ′ z i ,
1 − πi
for i = 1, 2, . . . , n, where ϕ′ z i = ϕ0 + ϕ1 z1i + ϕ2 z2i + · · · + ϕp zpi and
ϕ0 , ϕ1 , . . . , ϕp are unknown parameters. The cure probability is then
−1
πi = {1 + exp(−ϕ′ z i )} .
The same explanatory variables may feature in both the survival model for
the non-cured individuals and the model for the probability of cure. Also, in
contrast to a survival model, the model for the cure probability will generally
include a constant term ϕ0 , which is the logistic transformation of a common
cure probability, when there are no other explanatory variables in the model.
Now suppose that the survival times for non-cured individuals can be
modelled using a Weibull distribution, with baseline hazard function given
by hn0 (t) = λγtγ−1 . The hazard function for the non-cured individuals is
then
hni (t) = exp(β ′ xi )λγtγ−1 , (6.35)
and the corresponding survivor function is
′ ′
Sni (t) = [exp(−λtγ )]exp(β xi ) = exp{−eβ xi λtγ }. (6.36)
As for other models described in this chapter, the model that incorporates a
cured fraction can be fitted using the method of maximum likelihood. Suppose
that the data consist of n survival times t1 , t2 , . . . , tn , and that δi is the event
indicator for the ith individual so that δi = 1 if the ith individual dies and
zero otherwise. From Equation (5.38), the likelihood function is
∏
n
δ
L(β, ϕ, λ, γ) = {hi (ti )} i Si (ti ),
i=1
where hi (ti ) and Si (ti ) are found by substituting hni (ti ) and Sni (ti ) from
Equations (6.35) and (6.36) into Equations (6.33) and (6.34).
The corresponding log-likelihood function, log L(β, ϕ, λ, γ) can then be
maximised using computer software for numerical optimisation. This process
leads to estimates β̂, ϕ̂, λ̂, γ̂ of the unknown parameters and their standard
errors. In addition, models with different explanatory variables in either the
model for the probability of cure or the survival models for non-cured indi-
viduals, can be compared using the values of the statistic −2 log L(β̂, ϕ̂, λ̂, γ̂)
in the usual way.
A number of extensions to this model are possible. For example, an ac-
celerated failure time model can be used instead of a proportional hazards
model for the non-cured individuals. A Royston and Parmar model can also
be used to provide a more flexible model for the baseline hazard function in
the non-cured individuals.
6.11 ∗ Effect of covariate adjustment

We conclude this chapter with an illustration of an important feature that can
be encountered when developing parametric models for survival data. In lin-
ear regression analysis, one effect of adding covariates to a model is to reduce
the residual mean square, and hence increase the precision of estimates based
EFFECT OF COVARIATE ADJUSTMENT 271
on the model, such as a treatment effect. The estimated treatment effect, ad-
justed for explanatory variables, will then have a smaller standard error than
the unadjusted effect. In modelling survival data, the inclusion of relevant
explanatory variables often has a negligible effect on standard errors of pa-
rameter estimates. Indeed, the standard error of an estimated treatment effect,
for example, may even be larger after adjusting for covariates. Essentially, this
is because the treatment effect does not have the same interpretation in mod-
els with and without covariates, a point made by Ford, Norrie and Ahmadi
(1995). Nevertheless, it is important to include relevant explanatory variables
in the model, and to check that the fitted model is appropriate, in order to
ensure that a proper estimate of the treatment effect is obtained.
To illustrate this in a little more detail, suppose that ti is the observed
value of the random variable Ti , that is associated with the survival time
of the ith of n individuals, i = 1, 2, . . . , n. We will consider the situation
where there are two treatment groups, with n/2 individuals in each group,
and where there is a further explanatory variable whose values are available
for each individual. The two explanatory variables will be labelled X1 , X2 ,
where X1 refers to the treatment effect, and takes the value 0 or 1. The values
of X1 , X2 for the ith individual will be denoted x1i , x2i , respectively, and we
will write zji = xji − x̄j , for j = 1, 2, where x̄j is the sample mean of the values
of the explanatory variable Xj . A proportional hazards model will be adopted
for the dependence of the hazard of death at time t, for the ith individual,
on the values z1i , z2i , in which the baseline hazard is a constant value, λ.
Consequently, the hazard function can be expressed in the form
hi (t) = λ exp(β1 z1i + β2 z2i ), (6.37)
and under this model, the survival times are exponentially distributed, with
means {λ exp(β1 z1i +β2 z2i )}−1 . Using results given in Section 6.5.1, this model
may also be expressed in accelerated failure time form as
log Ti = µ − β1 z1i − β2 z2i + ϵi , (6.38)
where µ = − log λ and ϵi has a Gumbel distribution, that is log ϵi has a unit
exponential distribution. The model represented in Equations (6.37) or (6.38)
will be referred to as Model (1).
Using the results for maximum likelihood estimation given in Appendix A,
it can be shown that the approximate variance of the estimated treatment
effect, β̂1 , in Model (1), is
1
var (β̂1 ) = ∑n 2 ,
[1 − { corr (z1 , z2 )}2 ] i=1 z1i
where corr (z1 , z2 ) is the sample correlation between the values z1i and z2i .
∑n −0.5
Since z1i is either
2
or 0.5, and there are equal numbers of individuals in
each group, i=1 z1i = n/4, and so
4
var (β̂1 ) = . (6.39)
n[1 − { corr (z1 , z2 )}2 ]
Now consider the model that only includes the variable associated with the
treatment effect, X1 , so that
hi (t) = λ exp(β1 z1i ), (6.40)
or equivalently,
log Ti = µ − β1 z1i + ϵi , (6.41)
where again log ϵi has a unit exponential distribution. The model described
by Equations (6.40) or (6.41) will be referred to as Model (2). In this model,
the approximate variance of β̂1 is given by var (β̂1 ) = 4n−1 . Since the term
1 − { corr (z1 , z2 )}2 in Equation (6.39) is always less than or equal to unity,
the variance of β̂1 in Model (1) is at least equal to that of Model (2). The
addition of the explanatory variable X2 to Model (1) cannot therefore decrease
the variance of the estimated treatment effect.
The reason for this is that Model (1) and Model (2) cannot both be valid
for the same data set. If Model (1) is correct, and Model (2) is actually fit-
ted, the residual term in Equation (6.41) is not ϵi but ϵi − β2 z2i . Similarly,
if Model (2) is correct, but Model (1) is actually fitted, we cannot assume
that the logarithm of ϵi in Equation (6.38) has a unit exponential distribu-
tion. Moreover, the parameter β1 is now estimated less precisely because a
redundant parameter, β2 , is included in the model.
More detailed analytic and simulation studies are given in the paper by
Ford et al. (1995), which confirm the general point that the inclusion of ex-
planatory variables in models for survival data cannot be expected to increase
the precision of an estimated treatment effect.

The properties of random variables that have probability distributions such
as the logistic, lognormal and gamma, are presented in Johnson and Kotz
(1994). Chhikara and Folks (1989) give a detailed study of the inverse Gaussian
distribution.
A description of the log-linear model for survival data is contained in many
of the major textbooks on survival analysis; see in particular Cox and Oakes
(1984), Kalbfleisch and Prentice (2002), Klein and Moeschberger (2005) or
Lawless (2002).
Cox and Oakes (1984) show that the Weibull distribution is the only one to
have both the proportional hazards property and the accelerated failure time
property. They also demonstrate that the log-logistic distribution is the only
one that shares the accelerated failure time property and the proportional
odds property.
A non-parametric version of the accelerated failure time model, which
does not require the specification of a probability distribution for the survival
data, has been introduced by Wei (1992). This paper, and the published dis-
cussion, Fisher (1992), includes comments on whether the accelerated failure
time model should be used more widely in the analysis of survival data.
FURTHER READING 273
The application of the accelerated failure time and proportional odds mod-
els to the analysis of reliability data is described by Crowder et al. (1991).
The general proportional odds model for survival data was introduced by
Bennett (1983a), and Bennett (1983b) describes the log-logistic proportional
odds model. The model has been further developed by Yang and Prentice
(1999).
The piecewise exponential model, mentioned in Section 6.3.1, in which
hazards are constant over particular time intervals, was introduced by Breslow
(1974). Breslow also points out that the Cox regression model is equivalent
to a piecewise exponential model with constant hazards between each death
time. The piecewise exponential model and the use of the normal, lognormal,
logistic and log-logistic distributions for modelling survival times are described
in Aitkin et al. (1989).
Use of the quadratic hazard function was discussed by Gaver and Acar
(1979) and the bathtub hazard function was proposed by Hjorth (1980).
A more general way of modelling survival data is to use a general family of
distributions for survival times, which includes the Weibull and log-logistic as
special cases. The choice between alternative distributions can then be made
within a likelihood framework. In particular, the exponential, Weibull, log-
logistic, lognormal and gamma distributions are special cases of the generalised
F -distribution described by Kalbfleisch and Prentice (2002). However, this
methodology will only tend to be informative in the analysis of data sets in
which the number of death times is relatively large.
Estimators of the hazard function based on kernel smoothing are described
by Ramlau-Hansen (1983) and in the text of Klein and Moeschberger (2005).
The use of cubic splines in regression models was described by Durrleman
and Simon (1989). The flexible parametric model for survival analysis was
introduced by Royston and Parmar (2002), and a comprehensive account of
the model, and its implementation in Stata, is given by Royston and Lambert
(2011). Parametric mixture models that incorporate cured fractions, and their
extension to semi-parametric models, have been described by a number of
authors, including Farewell (1982), Kuk and Chen (1992), Taylor (1995), Sy
and Taylor (2000) and Peng and Dear (2000).
Chapter 7
Model checking in parametric models
Diagnostic procedures for the assessment of model adequacy are as important

in parametric modelling as they are when the Cox regression model is used in
the analysis of survival data. Procedures based on residuals are particularly
relevant, and so we begin this chapter by defining residuals for parametric
models, some of which stem from those developed for the Cox model, de-
scribed in Chapter 4. This is followed by a summary of graphical procedures
for assessing the suitability of models fitted to data that are assumed to have
a Weibull, log-logistic or lognormal distribution. Other ways of examining the
fit of a parametric regression model are then considered, along with methods
for the detection of influential observations. We conclude with a summary of
how the assumption of proportional hazards can be examined after fitting the
Weibull proportional hazards model.
7.1 Residuals for parametric models

Suppose that Ti is the random variable associated with the survival time of
the ith individual, i = 1, 2, . . . , n, and that x1i , x2i , . . . , xpi are the values
of p explanatory variables, X1 , X2 , . . . , Xp , for this individual. Assuming an
accelerated failure time model for Ti , we have that
log Ti = µ + α1 x1i + α2 x2i + · · · + αp xpi + σϵi ,
where ϵi is a random variable with a probability distribution that depends on

the distribution adopted for Ti , and µ, σ and αj , j = 1, 2, . . . , p, are unknown
parameters. If the observed survival time of the ith individual is censored, the
corresponding residual will also be censored, complicating the interpretation
of these quantities.
7.1.1 Standardised residuals

A natural form of residual to adopt in accelerated failure time modelling is
the standardised residual defined by
rSi = {log ti − µ̂ − α̂1 x1i − α̂2 x2i − · · · − α̂p xpi } /σ̂, (7.1)
275
276 MODEL CHECKING IN PARAMETRIC MODELS
where ti is the observed survival time of the ith individual, and µ̂, σ̂, α̂j ,
j = 1, 2, . . . , p, are the estimated parameters in the fitted accelerated fail-
ure time model. This residual has the appearance of a quantity of the form
‘observation − fitted value’, and would be expected to have the same distri-
bution as that of ϵi in the accelerated failure time model, if the model were
correct. For example, if a Weibull distribution is adopted for Ti , the rSi would
be expected to behave as if they were a possibly censored sample from a Gum-
bel distribution, if the fitted model is correct. The estimated survivor function
of the residuals would then be similar to the survivor function of ϵi , that is,
Sϵi (ϵ). Using the general result in Section 4.1.1 of Chapter 4, − log Sϵi (ϵ) has
a unit exponential distribution, and so it follows that − log Sϵi (rSi ) will have
an approximate unit exponential distribution, if the fitted model is appro-
priate. This provides the basis for a diagnostic plot that may be used in the
assessment of model adequacy, described in Section 7.2.4.
7.1.2 Cox-Snell residuals

The Cox-Snell residuals that were defined for the Cox regression model in Sec-
tion 4.1.1 of Chapter 4 are essentially the estimated values of the cumulative
hazard function for the ith observation, at the corresponding event time, ti .
Residuals that have a similar form may also be used in assessing the adequacy
of parametric models. The main difference is that now the survivor and hazard
functions are parametric functions that depend on the distribution adopted
for the survival times. In particular, the estimated survivor function for the ith
individual, on fitting an accelerated failure time model, from Equation (6.12),
is given by
( )
log t − µ̂ − α̂1 x1i − α̂2 x2i − · · · − α̂p xpi
Ŝi (t) = Sϵi , (7.2)
σ̂
where Sϵi (ϵ) is the survivor function of ϵi in the accelerated failure time model,
α̂j is the estimated coefficient of xji , j = 1, 2, . . . , p, and µ̂, σ̂ are the estimated
values of µ and σ. The form of Sϵi (ϵ) for some commonly used distributions
for Ti was summarised in Table 6.2 of Chapter 6.
The Cox-Snell residuals for a parametric model are defined by
rCi = Ĥi (ti ) = − log Ŝi (ti ), (7.3)
where Ĥi (ti ) is the estimated cumulative hazard function, and Ŝi (ti ) is the es-
timated survivor function in Equation (7.2), evaluated at ti . As in the context
of the Cox regression model, these residuals can be taken to have a unit ex-
ponential distribution when the correct model has been fitted, with censored
observations leading to censored residuals; see Section 4.1.1 for details.
The Cox-Snell residuals in Equation (7.3) are very closely related to the
standardised residuals in Equation (7.1), since from Equation (7.2), we see
that rCi = − log Sϵi (rSi ). Assessment of whether the standardised residuals
RESIDUALS FOR PARAMETRIC MODELS 277
have a particular distribution is therefore equivalent to assessing whether the
corresponding Cox-Snell residuals have a unit exponential distribution.
7.1.3 Martingale residuals

The martingale residuals provide a measure of the difference between the
observed number of deaths in the interval (0, ti ), which is either 0 or 1, and the
number predicted by the model. Observations with unusually large martingale
residuals are not well fitted by the model. The analogue of the martingale
residual, defined for the Cox regression model in Equation (4.6) of Chapter 4,
is such that
rM i = δi − rCi , (7.4)
where δi is the event indicator for the ith observation, so that δi is unity if
that observation is an event and zero if censored, and now rCi is the Cox-
Snell residual given in Equation (7.3). For reasons given in Section 7.1.5, the
martingale residuals for a parametric accelerated failure time model sum to
zero, but are not symmetrically distributed about zero. Strictly speaking, it is
no longer appropriate to refer to these residuals as martingale residuals since
the derivation of them, based on martingale methods, does not carry over to
the accelerated failure time model. However, for semantic convenience, we will
continue to refer to the quantities in Equation (7.4) as martingale residuals.
7.1.4 Deviance residuals

The deviance residuals, which were first presented in Equation (4.7) of Chapter
4, can be regarded as an attempt to make the martingale residuals symmet-
rically distributed about zero, and are defined by
1
rDi = sgn(rM i ) [−2 {rM i + δi log(δi − rM i )}] 2 . (7.5)
It is important to note that these quantities are not components of the de-
viance for the fitted parametric model, but nonetheless it will be convenient
to continue to refer to them as deviance residuals.
7.1.5 ∗ Score residuals

Score residuals, which parallel the score residuals, or Schoenfeld residuals,
used in connection with the Cox regression model, can be defined for any
parametric model. The score residuals are the components of the derivatives
of the log-likelihood function, with respect to the unknown parameters, µ, σ
and αj , j = 1, 2, . . . , p, and evaluated at the maximum likelihood estimates
of these parameters, µ̂, σ̂ and α̂j . From Equation (6.24) of Chapter 6, the
log-likelihood function for n observations is
∑
n
log L(α, µ, σ) = {−δi log(σti ) + δi log fϵi (zi ) + (1 − δi ) log Sϵi (zi )} ,
i=1
where zi = (log ti − µ − α1 x1i − α2 x2i − · · · − αp xpi )/σ, fϵi (ϵ) and Sϵi (ϵ) are
the density and survivor functions of ϵi , and δi is the event indicator.
Differentiating this log-likelihood function with respect to the parameters
µ, σ, and αj , for j = 1, 2, . . . , p, gives the following derivatives:
∂ log L ∑n
−1
=σ g(zi ),
∂µ i=1
∂ log L ∑n
= σ −1 {zi g(zi ) − δi } ,
∂σ i=1
∂ log L ∑n
= σ −1 xji g(zi ),
∂αj i=1
where the function g(zi ) is given by
(1 − δi )fϵi (zi ) δi fϵ′i (zi )

g(zi ) = − ,
Sϵi (zi ) fϵi (zi )
and fϵ′i (zi ) is the derivative of fϵi (zi ) with respect to zi .

The ith component of each derivative, evaluated at the maximum like-
lihood estimates of the unknown parameters, is then the score residual for
the corresponding term. Consequently, from the definition of the standardised
residual in Equation (7.1), the ith score residual for µ is
σ̂ −1 g(rSi ),
that for the scale parameter, σ, is
σ̂ −1 {rSi g(rSi ) − δi } ,
and that for the jth explanatory variable in the model, Xj , is
rU ji = σ̂ −1 xji g(rSi ).
Of these, the score residuals for Xj are the most important, and as in Section
4.1.6 are denoted rU ji . Specific expressions for these residuals are given in
the sequel for some particular parametric models. Because the sums of score
residuals are the derivatives of the log-likelihood function at its maximum,
these residuals must sum to zero.
7.2 Residuals for particular parametric models

In this section, the form of the residuals for parametric models based on
Weibull, log-logistic and lognormal distributions for the survival times are
described.
RESIDUALS FOR PARTICULAR PARAMETRIC MODELS 279
7.2.1 Weibull distribution
The residuals described in Section 7.1 may be used in conjunction with either
the proportional hazards or the accelerated failure time representations of
the Weibull model. We begin with the proportional hazards model described
in Chapter 5, according to which the hazard of death at time t for the ith
individual is
hi (t) = exp(β1 x1i + β2 x2i + · · · + βp xpi )h0 (t),
where h0 (t) = λγtγ−1 is the baseline hazard function. The corresponding

estimate of the cumulative hazard function is
Ĥi (t) = exp(β̂1 x1i + β̂2 x2i + · · · + β̂p xpi )λ̂tγ̂
which are the Cox-Snell residuals, as defined in Equation (7.3).

In the accelerated failure time form of the model, ϵi has a Gumbel distri-
bution, with survivor function
Sϵi (ϵ) = exp(−eϵ ). (7.6)
The standardised residuals are then as given in Equation (7.1), and if an ap-
propriate model has been fitted, these will be expected to behave as a possibly
censored sample from a Gumbel distribution. This is equivalent to assessing
whether the Cox-Snell residuals, defined below, have a unit exponential dis-
tribution.
The Cox-Snell residuals, rCi = − log Sϵi (rSi ), are, from Equation (7.6),
simply the exponentiated standardised residuals, that is rCi = exp(rSi ). These
residuals lead immediately to the martingale and deviance residuals for the
Weibull model, using Equations (7.4) and (7.5).
The score residuals for the Weibull model are found from the general results
in Section 7.1.5. In particular, the ith score residual for the jth explanatory
variable in the model, Xj , is
rU ji = σ̂ −1 xji (erSi − δi ) ,
where rSi is the ith standardised residual and δi the event indicator. We also
note that the ith score residual for µ is σ̂ −1 (erSi − δi ), which is σ̂ −1 (rCi −
δi ). Since these score residuals sum to zero, it follows that the sum of the
martingale residuals, defined in Equation (7.4), must be zero in the Weibull
model.
7.2.2 Log-logistic distribution

In the log-logistic accelerated failure time model, the random variable ϵi has
a logistic distribution with survivor function
−1
Sϵi (ϵ) = (1 + eϵ ) .
Accordingly, the standardised residuals, obtained from Equation (7.1), should
behave as a sample from a logistic distribution, if the fitted model is correct.
Equivalently, the Cox-Snell residuals for the log-logistic accelerated failure
time model are given by
rCi = − log Sϵi (rSi ) ,
that is,
rCi = log {1 + exp (rSi )} ,
where rSi is the ith standardised residual. The score residuals are found from
the general results in Section 7.1.5, and we find that the ith score residual for
the jth explanatory variable in the model is
{ }
exp(rSi ) − δi
rU ji = σ̂ −1 xji .
1 + exp(rSi )
7.2.3 Lognormal distribution

If the survival times are assumed to have a lognormal distribution, then ϵi
in the log-linear formulation of the accelerated failure time model is normally
distributed. The estimated survivor function for the ith individual, from Equa-
tion (6.23), is
( )
log t − µ̂ − α̂1 x1i − α̂2 x2i − · · · − α̂p xpi
Ŝi (t) = 1 − Φ ,
σ̂
and so the Cox-Snell residuals become
rCi = − log {1 − Φ (rSi )} ,
where, as usual, rSi is the ith standardised residual in Equation (7.1). Again
the martingale and deviance residuals are obtained from these, and the score
residuals are obtained from the results in Section 7.1.5. Specifically, the ith
score residual for Xj , is
{ }
−1 (1 − δi )fϵi (rSi )
rU ji = σ̂ + δi rSi ,
1 − Φ(rSi )
where fϵi (rSi ) is the standard normal density function at rSi , and Φ(rSi ) is
the corresponding distribution function.
7.2.4 Analysis of residuals

In the analysis of residuals after fitting a parametric model to survival data,
one of the most useful plots is based on comparing the distribution of the
Cox-Snell residuals with the unit exponential distribution. As noted in Sec-
tion 7.1.1, this is equivalent to comparing the distribution of the standardised
residuals to that of the random variable ϵi in the log-linear form of the acceler-
ated failure time model. This comparison is made using a cumulative hazard,
or log-cumulative hazard, plot of the residuals, as shown in Section 4.2.1 of
Chapter 4, where the use of this plot in connection with residuals after fit-
ting the Cox regression model was described. In summary, the Kaplan-Meier
estimate of the survivor function of the Cox-Snell residuals, denoted Ŝ(rCi ),
is obtained, and − log Ŝ(rCi ) is plotted against rCi . A straight line with unit
slope and zero intercept will suggest that the fitted model is appropriate. Al-
ternatively, a log-cumulative hazard plot of the residuals, obtained by plotting
log{− log Ŝ(rCi )} against log rCi , will also give a straight line with unit slope
and passing through the origin, if the fitted survival model is satisfactory.
In Section 4.2.1, substantial criticisms were levied against the use of this
plot. However, these criticisms do not have as much force for residuals derived
from parametric models. The reason for this is that the non-parametric esti-
mate of the baseline cumulative hazard function, used in the Cox regression
model, is now replaced by an estimate of a parametric function. This function
usually depends on just two parameters, µ and σ, and so fewer parameters are
being estimated when an accelerated failure time model is fitted to survival
data. The Cox-Snell residuals for a parametric model are therefore much more
likely to be approximated by a unit exponential distribution, when the correct
model has been fitted.
Other residual plots that are useful include index plots of martingale or
deviance residuals, which can be used to identify observations not well fitted
by the model. A plot of martingale or deviance residuals against the survival
times, the rank order of the times, or explanatory variables, shows whether
there are particular times, or particular values of explanatory variables, for
which the model is not a good fit. Plots of martingale or deviance residuals
against the estimated acceleration factor, exp(−α̂′ xi ), or simply the estimated
linear component of the accelerated failure time model, α̂′ xi , also provide in-
formation about the relationship between the residuals and the likely survival
time of an individual. Those with large values of the estimated acceleration
factor will tend to have shorter survival times. Index plots of score residuals,
or plots of these residuals against the survival times, or the rank order of the
survival times, might be examined in a more comprehensive assessment of
model adequacy.

In Example 5.10 of Chapter 5, data on the survival times of patients with
ovarian cancer were presented. The data were analysed using a Weibull pro-
portional hazards model, and the model chosen contained variables corre-
sponding to the age of the woman, Age, and the treatment group to which the
woman was assigned, Treat. In the accelerated failure time representation of
the model, the estimated survivor function for the ith woman is
( )
log t − µ̂ − α̂1 Age i − α̂2 Treat i
Ŝi (t) = Sϵi ,
σ̂
where Sϵi (ϵ) = exp(−eϵ ), so that
{ ( )}
log t − 10.4254 + 0.0790 Age i − 0.5615 Treat i
Ŝi (t) = exp − exp .
0.5489
The standardised residuals are the values of
rSi = (log ti − 10.4254 + 0.0790 Age i − 0.5615 Treat i )/0.5489,
for i = 1, 2, . . . , 26, and these are given in Table 7.1. Also given are the values
of the Cox Snell residuals, which for the Weibull model, are such that rCi =
exp(rSi ).
Table 7.1 Values of the standardised and Cox-Snell resid-

uals for 26 ovarian cancer patients.
Patient rSi rCi Patient rSi rCi
1 −1.320 0.267 14 −1.782 0.168
2 −1.892 0.151 15 −0.193 0.825
3 −2.228 0.108 16 −1.587 0.204
4 −2.404 0.090 17 −0.917 0.400
5 −3.270 0.038 18 −1.771 0.170
6 −0.444 0.642 19 −1.530 0.217
7 −1.082 0.339 20 −2.220 0.109
8 −0.729 0.482 21 −0.724 0.485
9 0.407 1.503 22 −1.799 0.165
10 0.817 2.264 23 0.429 1.535
11 −1.321 0.267 24 −0.837 0.433
12 −0.607 0.545 25 −1.287 0.276
13 −1.796 0.166 26 −1.886 0.152
A cumulative hazard plot of the Cox-Snell residuals is given in Figure 7.1.

In this figure, the plotted points lie on a line that has an intercept and slope
close to zero and unity, respectively. However, there is some evidence of a
systematic deviation from the straight line, giving some cause for concern
about the adequacy of the fitted model.
Plots of the martingale and deviance residuals against the rank order of the
survival times are shown in Figures 7.2 and 7.3, respectively. Both of these
plots show a slight tendency for observations with longer survival times to
have smaller residuals, but these are also the observations that are censored.
The graphs in Figure 7.4 show the score residuals for the two variables in
the model, Age and Treat, plotted against the rank order of the survival times.
The plot of the score residuals for Age shows that there are three observations
with relatively large residuals. These correspond to patients 14, 4 and 26 in
the original data set given in Table 5.6. However, there does not appear to be
anything unusual about these observations. The score residual for Treat for
patient 26 is also somewhat larger than the others. This points to the fact
that the model is not a good fit to the data from patients 14, 4 and 26.
1.5
Cumulative hazard of residual
1.0
0.5
0.0
0.0 0.2 0.4 0.6 0.8 1.0
Cox-Snell residual
1
Martingale residual
-1
-2
0 4 8 12 16 20 24 28
Figure 7.2 Plot of the martingale residuals against rank order of survival time.
2
Deviance residual
-1
-2
0 4 8 12 16 20 24 28
Figure 7.3 Plot of the deviance residuals against rank order of survival time.
200 6
4
Score residual for Treat
Score residual for Age
100
2
0
0
-100
-2
-200 -4
0 4 8 12 16 20 24 28 0 4 8 12 16 20 24 28
Rank of survival time Rank of survival time
Figure 7.4 Score residuals plotted against rank order of survival time for Age and
Treat.
7.3 Comparing observed and fitted survivor functions
In parametric modelling, the estimated survivor function is a continuous func-

tion of the survival time, t, and so this function can be plotted for particular
values of the explanatory variables included in the model. When there is just
a single sample of survival data, with no explanatory variables, the fitted sur-
vivor function can be compared directly with the Kaplan-Meier estimate of the
COMPARING OBSERVED AND FITTED SURVIVOR FUNCTIONS 285
survivor function, described in Section 2.1.2 of Chapter 2. If the fitted survivor
function is close to the Kaplan-Meier estimate, which is a step-function, the
fitted model is an appropriate summary of the data. Similarly, suppose that
the model incorporates one or two factors that classify individuals according
to treatment group, or provide a cross-classification of treatment group and
gender. For each group of individuals defined by the combinations of levels of
the factors in the model, the fitted survivor function can then be compared
with the corresponding Kaplan-Meier estimate of the survivor function.
In situations where the values of a number of explanatory variables are
recorded, groups of individuals are formed from the values of the estimated
linear component of the fitted model. For the ith individual, whose values of
the explanatory variables in the model are xi , this is just the risk score in
a proportional hazards model, β̂ ′ xi , or the value of α̂′ xi in an accelerated
failure time model. The following discussion is based on the risk score, where
large positive values correspond to a greater hazard, but it could equally well
be based on the linear component of an accelerated failure time model, or the
value of the acceleration factor.
The values of the risk score for each individual are arranged in increasing
order, and these values are used to divide the individuals into a number of
groups. For example, if three groups were used, there would be individuals
with low, medium and high values of the risk score. The actual number of
groups formed in this way will depend on the size of the database. For larger
databases, five or even seven groups might be constructed. In particular, with
five groups, there would be 20% of the individuals in each group; those with
the lowest and highest values of the risk score would be at low and high risk,
respectively, while the middle 20% would be of medium risk.
The next step is to compare the observed and fitted survivor functions in
each of the groups. Suppose that Ŝij (t) is the model-based estimate of the
survivor function for the ith individual in the jth group. The average fitted
survivor function is then obtained for each group, or just the groups with the
smallest, middle and highest risk scores, from
1 ∑
nj
S̄j (t) = Ŝij (t),
nj i=1
where nj is the number of observations in the jth group. The value of S̄j (t)
would be obtained for a range of t values, so that a plot of the values of
S̄j (t) against t, for each value of j, yields a smooth curve. The corresponding
observed survivor function for a particular group is the Kaplan-Meier estimate
of the survivor function for the individuals in that group. Superimposing these
two sets of estimates gives a visual representation of the agreement between
the observed and fitted survivor functions. This procedure is analogous to that
described in Section 3.11.1 for the Cox regression model.
Using this approach, it is often easier to detect departures from the fitted
model, than from plots based on residuals. However, the procedure can be
criticised for using the same fitted model to define the groups, and to obtain
the estimated survivor function for each group. If the database is sufficiently
large, the survivor function could be estimated from half of the data, and the
fit of the model evaluated on the remaining half. Also, since the method is
based on the values of the risk score, no account is taken of differences between
individuals who have different sets of values of the explanatory variables, but
just happen to have the same value of the risk score.
In this example, we examine the fit of a Weibull proportional hazards model
to the data on the survival times of 26 women, following treatment for ovarian
cancer. A Weibull model that contains the variables Age and Treat is fitted,
as in Example 5.10, so that the fitted survivor function for the ith individual
is { }
Ŝi (t) = exp −eη̂i λ̂tγ̂ , (7.7)
where η̂i = 0.144 Age i − 1.023 Treat i is the risk score, i = 1, 2, . . . , 26. This
is equivalent to the accelerated failure time representation of the model, used
in Example 7.1. The values of η̂i are then arranged in ascending order and
divided into three groups, as shown in Table 7.2.
Table 7.2 Values of the risk score, with the patient number in parentheses, for
three groups of ovarian cancer patients.
Group Risk score
1 (low risk) 4.29 (14) 4.45 (2) 4.59 (20) 5.15 (22) 5.17 (5)
5.17 (19) 5.31 (17) 5.59 (4) 5.59 (12)
2 (medium risk) 5.87 (16) 6.02 (13) 6.16 (8) 6.18 (18) 6.31 (26)
6.45 (6) 6.45 (9) 6.45 (11) 7.03 (25)
3 (high risk) 7.04 (15) 7.04 (24) 7.17 (7) 8.19 (23)
8.48 (1) 9.34 (3) 9.63 (10) 9.63 (21)
The next step is to obtain the average survivor function for each group by
averaging the values of the estimated survivor function, in Equation (7.7), for
the patients in the three groups. This is done for t = 0, 1, . . . , 1230, and the
three average survivor functions are shown in Figure 7.5. The Kaplan-Meier
estimate of the survivor function for the individuals in each of the three groups
shown in Table 7.2 is then calculated, and this is also shown in Figure 7.5.
From this plot, we see that the model is a good fit to the patients in the
high-risk group. For those in the middle group, the agreement between the
observed and fitted survivor functions is not that good, as the fitted model
leads to estimates of the survivor function that are a little too high. In fact, the
patients in this group have the largest values of the martingale residuals, which
also indicates that the death times of these individuals are not adequately
summarised by the fitted model. There is only one death among the individuals
in the low-risk group, and so little can be said about the fit of the model to
this set of patients.
1.0
0.8
0.6
0.4
0.2
0.0
0 250 500 750 1000 1250
Survival time
Figure 7.5 Plot of the observed and fitted survivor functions for patients of low (·······),
medium (- - -) and high (—) risk. The observed survivor function is the step-function.
7.4 Identification of influential observations

As when fitting the Cox regression model, it will be important to identify
observations that exert an undue influence on particular parameter estimates,
or on the complete set of parameter estimates. These two aspects of influence
are considered in turn in this section.
A number of influence diagnostics for the Weibull proportional hazards
model have been proposed by Hall, Rogers and Pregibon (1982), derived from
the accelerated failure time representation of the model. However, they may
also be used with other parametric models. These diagnostics are computed
from the estimates of all p + 2 parameters in the model, and their variance-
covariance matrix. For convenience, the vector of p + 2 parameters will be
denoted by θ, so that θ ′ = (µ, α1 , α2 , . . . , αp , σ). The vector θ̂ ′ will be used
to denote the corresponding vector of estimates of the parameters.
7.4.1 ∗ Influence of observations on a parameter estimate

An approximation to the change in the estimated value of θj , the jth com-
ponent of the vector θ, on omitting the ith observation, ∆i θ̂j , is the jth
component of the (p + 2) × 1 vector
var (θ̂)ui . (7.8)
In Expression (7.8), var (θ̂) is the estimated variance-covariance matrix of the

parameters in θ, and ui is the (p + 2) × 1 vector of values of the first partial
derivatives of the log-likelihood for the ith observation, with respect to the
p + 2 parameters in θ, evaluated at θ̂. The vector ui is therefore the vector of
values of the score residuals for the ith observation, defined in Section 7.1.5.
The quantities ∆i α̂j are components 2 to p − 1 of the vector in Expres-
sion (7.8), which we will continue to refer to as delta-betas rather than as
delta-alphas. These values may be standardised through division by the stan-
dard error of α̂j , leading to standardised delta-betas. Index plots or plots of
the standardised or unstandardised values of ∆i α̂j provide informative sum-
maries of this aspect of influence.
7.4.2 ∗ Influence of observations on the set of parameter estimates

Two summary measures of the influence of the ith observation on the set of
parameters that make up the vector θ have been proposed by Hall, Rogers
and Pregibon (1982). These are the statistics Fi and Ci . The quantity Fi is
given by
u′i R−1 ui
Fi = , (7.9)
(p + 2){1 − u′i R−1 ui }
where the (p + 2) × (p ∑ + 2) matrix R is the cross-product matrix of score
residuals, that is, R = i=1 ui u′i . Equivalently, R = U ′ U , where U is the
n
n×(p+2) matrix whose ith row is the transpose of the vector of score residuals,
u′i . An alternative measure of the influence of the ith observation on the set
of parameter estimates is the statistic
u′i var (θ̂)ui

Ci = . (7.10)
{1 − u′i var (θ̂)ui }2
The statistics Fi and Ci will typically have values that are quite different
from each other. However, in each case a relatively large value of the statistic
will indicate that the corresponding observation is influential. Exactly how
such observations influence the estimates would need to be investigated by
omitting that observation from the data set and refitting the model.

We now go on to investigate whether there are any influential observations in
the data on the survival times following chemotherapy treatment for ovarian
cancer. The unstandardised delta-betas for Age and Treat, plotted against the
rank order of the survival times, are shown in Figures 7.6 and 7.7.
In Figure 7.6, two observations have relatively large values of the delta-
beta for Age. These occur for patients 4 and 5 in the original data set. Both
women have short survival times, and in addition one is relatively old at 74
years and the other relatively young at 43 years. The delta-betas for Treat
displayed in Figure 7.7 show no unusual features.
We next investigate the influence of each observation on the set of param-
eter estimates. The values of Fi and Ci , defined in Equations (7.9) and (7.10),
0.012
0.008
Delta-beta for Age
0.004
0.000
-0.004
-0.008
0 4 8 12 16 20 24 28
Figure 7.6 Plot of the delta-betas for Age against rank order of survival time.
0.20
0.15
0.10
Delta-beta for Treat
0.05
0.00
-0.05
-0.10
-0.15
0 4 8 12 16 20 24 28
Figure 7.7 Plot of the delta-betas for Treat against rank order of survival time.
are plotted against the rank order of the survival times in Figures 7.8 and 7.9.
Figure 7.8 clearly shows that the observation corresponding to patient 5 is
influential, and that the influence of patients 1, 4, 14 and 26 should be in-
vestigated in greater detail. Figure 7.9 strongly suggests that the data from
patients 5 and 26 is influential.
0.4
0.3
Value of F-statistic
0.2
0.1
0.0
0 4 8 12 16 20 24 28
Figure 7.8 Plot of the F -statistic against rank order of survival time.
3
Value of C-statistic
0
0 4 8 12 16 20 24 28
Figure 7.9 Plot of the C-statistic against rank order of survival time.
TESTING PROPORTIONAL HAZARDS IN THE WEIBULL MODEL 291
The linear component of the fitted hazard function in the model fitted to
all 26 patients is
0.144 Age i − 1.023 Treat i ,
while that on omitting each of observations 1, 4, 5, 14 and 26 in turn is as
follows:
Omitting patient number 1: 0.142 Age i − 1.016 Treat i
These results show that the effect of omitting the data from patient 1 on
the parameter estimates is small. When the data from patient 4 are omitted,
the estimated coefficient of Age is most affected, whereas when the data from
patient 14 are omitted, the coefficient of Treat is changed the most. On leaving
out the data from patients 5 and 26, both estimates are considerably affected.
The hazard ratio for a patient on the combined treatment (Treat = 2),
relative to one on the single treatment (Treat = 1), is estimated by e−1.023 =
0.36, when the model is fitted to all 26 patients. When the observations from
patients 1, 4, 5, 14 and 26 are omitted in turn, the estimated age-adjusted
hazard ratios are 0.36, 0.30, 0.49, 0.27 and 0.50, respectively. The data from
patients 5 and 26 clearly have the greatest effect on the estimated hazard ratio;
in each case the estimate is increased, and the magnitude of the treatment
effect is diminished. Omission of the data from patients 4 or 14 decreases the
estimated hazard ratio, thereby increasing the estimated treatment difference.
7.5 Testing proportional hazards in the Weibull model

The Weibull model is most commonly used as a parametric proportional haz-
ards model, and so it will be important to test that the proportional hazards
assumption is tenable. In Section 4.4.3 of Chapter 4, it was shown how a
time-dependent variable can be used in testing proportionality of hazards in
the Cox regression model. Parametric models containing time-dependent vari-
ables are more complicated, and because software for fitting such models is
not widely available, further details on this approach will not be given here.
In the Weibull model, the assumption of proportional hazards across a
number of groups, g, say, corresponds to the assumption that the shape pa-
rameter γ in the baseline hazard function is the same in each group. One
way of testing this assumption is to fit a separate Weibull model to each of
the g groups, where the linear component of the model is the same in each
case. The models fitted to the data from each group will then have differ-
ent shape parameters as well as different scale parameters. The values of the
statistic −2 log L̂ for each of these g separate models are then summed to give
a value of −2 log L̂ for a model that has different shape parameter for each
group. Denote this by −2 log L̂1 . We then combine the g sets of data and
fit a Weibull proportional hazards model that includes the factor associated
with the group effects and interactions between this factor and other terms
in the model. This model then corresponds to there being a common shape
parameter for each group. The inclusion of group effects in the model leads to
there being different scale parameters for each group. The value of −2 log L̂
for this model, −2 log L̂0 , say, is then compared with −2 log L̂1 . The difference
between the values of these two statistics is the change in −2 log L̂ due to
constraining the Weibull shape parameters to be equal, and can be compared
with a chi-squared distribution on g − 1 degrees of freedom. If the difference
is not significant, the assumption of proportional hazards is justified.

Data from the study of survival following treatment for ovarian cancer, given
in Example 5.10 of Chapter 5, are now used to illustrate the procedure for
testing the assumption that the Weibull shape parameter is the same for the
patients in each of the two treatment groups. The first step is to fit a Weibull
proportional hazards model that contains Age alone to the data from the
women in each treatment group. When such a model is fitted to the data
from those on the single chemotherapy treatment, the value of the statistic
−2 log L̂ is 22.851, while that for the women on the combined treatment is
16.757. The sum of the two values of −2 log L̂ is 39.608, which is the value
of the statistic for a Weibull model with different shape parameters for the
two treatment groups, and different coefficients of Age for each group. For the
model with different age effects for each treatment group, a treatment effect,
and common shape parameter, the value of −2 log L̂ is 39.708. The change in
−2 log L̂ on constraining the shape parameters to be equal is therefore 0.10,
which is not significant when compared with a chi-squared distribution on one
degree of freedom. The two shape parameters may therefore be taken to be
equal.
Some alternatives to the proportional hazards model are described in

Chapter 6, and further comments on how to deal with situations in which
the hazards are not proportional are given in Chapter 11.
7.6 Further reading

There have been relatively few publications on model checking in parametric
survival models, compared to the literature on model checking in the Cox re-
gression model. Residuals and influence measures for the Weibull proportional
hazards model are described by Hall, Rogers and Pregibon (1982). Hollander
and Proschan (1979) show how to assess whether a sample of censored ob-
FURTHER READING 293
servations is drawn from a particular probability distribution. Weissfeld and
Schneider (1994) describe and illustrate a number of residuals that can be used
in conjunction with parametric models for survival data. Cohen and Barnett
(1995) describe how the interpretation of cumulative hazard plots of residuals
can be helped by the use of simulated envelopes for the plots. Influence di-
agnostics for use in parametric survival modelling are given by Weissfeld and
Schneider (1990) and Escobar and Meeker (1992). A SAS macro for evaluating
influence diagnostics for the Weibull proportional hazards model is described
by Escobar and Meeker (1988). These papers involve arguments based on lo-
cal influence, a topic that is explored in general terms by Cook (1986), and
reviewed in Rancel and Sierra (2001). A method for testing the assumptions
of proportional hazards and accelerated failure times against a general model
for the hazard function is presented by Ciampi and Etezadi-Amoli (1985).
An interesting application of parametric modelling, based on data on times
to reoffending by prisoners released on parole, which incorporates elements of
model checking, is given by Copas and Heydari (1997).
Chapter 8
Time-dependent variables
When explanatory variables are incorporated in a model for survival data,

the values taken by such variables are those recorded at the time origin of
the study. For example, consider the study to compare two treatments for
prostatic cancer first described in Example 1.4 of Chapter 1. Here, the age of
a patient, serum haemoglobin level, size of the tumour, value of the Gleason
index, and of course the treatment group, were all recorded at the time when
a patient was entered into the study. The impact of these variables on the
hazard of death is then evaluated.
In many studies that generate survival data, individuals are monitored for
the duration of the study. During this period, the values of certain explanatory
variables may be recorded on a regular basis. Thus, in the example on prostatic
cancer, the size of the tumour and other variables, may be recorded at frequent
intervals. If account can be taken of the values of explanatory variables as they
evolve, a more satisfactory model for the hazard of death at any given time
would be obtained. For example, in connection with the prostatic cancer study,
more recent values of the size of the tumour may provide a better indication
of future life expectancy than the value at the time origin.
Variables whose values change over time are known as time-dependent
variables, and in this chapter we see how such variables can be incorporated
in models used in the analysis of survival data.
8.1 Types of time-dependent variables

It is useful to consider two types of variables that change over time, which
may be referred to as internal variables and external variables.
Internal variables relate to a particular individual in a study, and can only
be measured while a patient is alive. Such data arises when repeated measure-
ments of certain characteristics are made on a patient over time, and examples
include measures of lung function such as vital capacity and peak flow rate,
white blood cell count, systolic blood pressure and serum cholesterol level.
Variables that describe changes in the status of a patient are also of this type.
For example, following a bone marrow transplant, a patient may be suscep-
tible to the development of graft versus host disease. A binary explanatory
variable, that reflects whether the patient is suffering from this life-threatening
295
296 TIME-DEPENDENT VARIABLES
side effect at any given time, is a further example of an internal variable. In
each case, such variables reflect the condition of the patient and their values
may well be associated with the survival time of the patient.
On the other hand, external variables are time-dependent variables that
do not necessarily require the survival of a patient for their existence. One
type of external variable is a variable that changes in such a way that its
value will be known in advance at any future time. The most obvious example
is the age of a patient, in that once the age at the time origin is known,
that patient’s age at any future time will be known exactly. However, there
are other examples, such as the dose of a drug that is to be varied in a
predetermined manner during the course of a study, or planned changes to
the type of immunosuppressant to be used following organ transplantation.
Another type of external variable is one that exists totally independently of
any particular individual, such as the level of atmospheric sulphur dioxide, or
air temperature. Changes in the values of such quantities may well have an
effect on the lifetime of individuals, as in studies concerning the management
of patients with certain types of respiratory disease.
Time-dependent variables also arise in situations where the coefficient of
a time-constant explanatory variable is a function of time. In Section 3.9 of
Chapter 3, it was explained that the coefficient of an explanatory variable
in the Cox proportional hazards model is a log-hazard ratio, and so under
this model, the hazard ratio is constant over time. If this ratio were in fact a
function of time, then the coefficient of the explanatory variable that varies
with time is referred to as a time-varying coefficient. In this case, the log-
hazard ratio is not constant and so we no longer have a proportional hazards
model. More formally, suppose that the coefficient of an explanatory variable,
X, is a linear function of time, t, so that we may write the term as βtX. This
means that the corresponding log-hazard ratio is a linear function of time.
This was precisely the sort of term introduced into the model in order to test
the assumption of proportional hazards in Section 4.4.3 of Chapter 4. This
term can also be written as βX(t), where X(t) = Xt is a time-dependent
variable. In general, suppose that a model includes the explanatory variable,
X, with a time-varying coefficient of the form β(t). The corresponding term in
the model would be β(t)X, which can be expressed as βX(t). In other words,
a term that involves a time-varying coefficient can be expressed as a time-
dependent variable with a constant coefficient. However, if β(t) is a non-linear
function of one or more unknown parameters, for example β0 exp(β1 t), the
term is not so easily fitted in a model.
All these different types of time-dependent variables can be introduced into
the Cox regression model, in the manner described in the following section.
8.2 A model with time-dependent variables

According to the Cox proportional hazards model described in Chapter 3, the
hazard of death at time t for the ith of n individuals in a study can be written
A MODEL WITH TIME-DEPENDENT VARIABLES 297
in the form  
∑p 
hi (t) = exp βj xji h0 (t),
 
j=1
where xji is the baseline value of the jth explanatory variable, Xj , j = 1, 2, . . . ,

p, for the ith individual, i = 1, 2, . . . , n, and h0 (t) is the baseline hazard
function. Generalising this model to the situation in which some or all of the
explanatory variables are time-dependent, we write xji (t) for the value of the
jth explanatory variable at time t, in the ith individual. The Cox regression
model then becomes
 
∑ p 
hi (t) = exp βj xji (t) h0 (t). (8.1)
 
j=1
In this model, the baseline hazard function, h0 (t), is interpreted as the hazard
function for an individual for whom all the variables are zero at the time
origin, and remain at this same value through time.
Since the values of the variables xji (t) in the model given in Equation (8.1)
depend on the time t, the relative hazard hi (t)/h0 (t) is also time-dependent.
This means that the hazard of death at time t is no longer proportional to the
baseline hazard, and the model is no longer a proportional hazards model.
To provide an interpretation of the β-parameters in this model, consider
the ratio of the hazard functions at time t for two individuals, the rth and
sth, say. This is given by
hr (t)
= exp [β1 {xr1 (t) − xs1 (t)} + · · · + βp {xrp (t) − xsp (t)}] .
hs (t)
The coefficient βj , j = 1, 2, . . . , p, can therefore be interpreted as the log-

hazard ratio for two individuals whose value of the jth explanatory variable
at a given time t differ by one unit, with the two individuals having the same
values of all the other p − 1 variables at that time.
8.2.1 ∗ Fitting the Cox model

When the Cox regression model is extended to incorporate time-dependent
variables, the partial log-likelihood function, from Equation (3.6) in Chapter 3,
can be generalised to
  
∑n ∑p ∑ ∑p 
δi βj xji (ti ) − log exp  βj xjl (ti ) , (8.2)
 
i=1 j=1 l∈R(ti ) j=1
in which R(ti ) is the risk set at time ti , the death time of the ith individual in
the study, i = 1, 2, . . . , n, and δi is an event indicator that is zero if the survival
time of the ith individual is censored and unity otherwise. This expression can
then be maximised to give estimates of the β-parameters.
In order to use Equation (8.1) in this maximisation process, the values of
each of the variables in the model must be known at each death time for all
individuals in the risk set at time ti . This is no problem for external variables
whose values are preordained, but it may be a problem for external variables
that exist independently of the individuals in a study, and certainly for internal
variables.
To illustrate the problem, consider a trial of two maintenance therapies for
patients who have suffered a myocardial infarct. The serum cholesterol level
of such patients may well be measured at the time when a patient is admitted
to the study, and at regular intervals of time thereafter. This variable is then
a time-dependent variable, and will be denoted X(t). It is then plausible that
the hazard of death for any particular patient, the ith, say, at time t, hi (t), is
more likely to be influenced by the value of the explanatory variable X(t) at
time t, than its value at the time origin, where t = 0.
Now suppose that the ith individual dies at time ti and that there are
two other individuals, labelled r and s, in the risk set at time ti . We further
suppose that individual r dies at time tr , where tr > ti , and that the survival
time of individual s, ts , is censored at some time after tr . The situation is
illustrated graphically in Figure 8.1. In this figure, the vertical dotted lines
refer to points in patient time when the value of X(t) is measured.
i D
r D
Individual
s C
0 ti tr ts
Time
Figure 8.1 Survival times of three patients in patient time.
If individuals r and s are the only two in the risk set at time ti , and X
is the only explanatory variable that is measured, the contribution of the ith
individual to the log-likelihood function in Expression (8.2) will be
∑
βxi (ti ) − log exp{βxl (ti )},
l
where xi (ti ) is the value of X(t) for the ith individual at their death time,
ti , and l in the summation takes the values i, r, and s. This expression is
therefore equal to
{ }
βxi (ti ) − log eβxi (ti ) + eβxr (ti ) + eβxs (ti ) .
This shows that the value of the time-dependent variable X(t) is needed at
the death time of the ith individual, and at time ti for individuals r and s. In
addition, the value of the variable X(t) will be needed for individuals r and s
at tr , the death time of individual r.
For terms in a model that are explicit functions of time, such as interac-
tions between time and a variable or factor measured at baseline, there is no
difficulty in obtaining the values of the time-dependent variables at any time
for any individual. Indeed, it is usually straightforward to incorporate such
variables in the Cox model when using statistical software that has facilities
for dealing with time-dependent variables. For other variables, such as serum
cholesterol level, the values of the time-dependent variable at times other than
that at which it was measured has to be approximated. There are then several
possibilities.
One option is to use the last recorded value of the variable before the
time at which the value of the variable is needed. When the variable has
been recorded for an individual before and after the time when the value is
required, the value closest to that time might be used. Another possibility is to
use linear interpolation between consecutive values of the variable. Figure 8.2
illustrates these approximations.
In this figure, the continuous curve depicts the actual value of a time-
dependent variable at any time, and the dotted vertical lines signify times
when the variable is actually measured. If the value of the variable is required
at time t in this figure, we could use either the value at P, the last recorded
value of the variable, the value at R, the value closest to t, or the value at Q,
the linearly interpolated value between P and R.
Linear interpolation is clearly not an option when a time-dependent vari-
able is a categorical variable. In addition, some categorical variables may be
such that individuals can only progress through the levels of the variable in a
particular direction. For example, the performance status of an individual may
only be expected to deteriorate, so that the value of this categorical variable
might only change from ‘good’ to ‘fair’ and from ‘fair’ to ‘poor’. As another
example, following a biopsy, a variable associated with the occurrence of a
tumour will take one of two values, corresponding to absence and presence. It
might then be very unlikely for the status to change from ‘present’ to ‘absent’
in consecutive biopsies.
Time-dependent variable
R
Q
t
Time
Figure 8.2 Computation of the value of a time-dependent variable at intermediate

times.
Anomalous changes in the values of time-dependent variables can be de-

tected by plotting the values of the variable against time for each patient.
This may then lead on to a certain amount of data editing. For example,
consider the plot in Figure 8.3, which shows the biopsy result, absent or
present, for a particular patient at a number of time points. In this diagram,
at least one of the observations made at times tA and tB must be incorrect.
The observation at tA might then be changed to ‘absent’ or that at tB to
‘present’.
If inferences of interest turn out to be sensitive to the method of interpo-
lation used, extreme caution must be exercised when interpreting the results.
Indeed, this feature could indicate that the value of the time-dependent vari-
able is subject to measurement error, substantial inherent variation or perhaps
the values have not been recorded sufficiently regularly.
8.2.2 ∗ Estimation of baseline hazard and survivor functions
After a Cox regression model that includes time-dependent variables has been
fitted, the baseline hazard function, h0 (t) and the corresponding baseline sur-
vivor function, S0 (t), can be estimated. This involves an adaptation of the
results given in Section 3.10 of Chapter 3 to cope with the additional com-
plication of time-dependent variables, in which the values of the explanatory
variables need to be updated with their time-specific values. In particular, the
Nelson-Aalen estimate of the baseline cumulative hazard function, given in
present A
Biopsy result
absent B
0 tA tB
Time of biopsy
Figure 8.3 Values of a time-dependent categorical variable.
Equation (3.28), becomes
∑
k
dj
H̃0 (t) = − log S̃0 (t) = ∑ , (8.3)
j=1 l∈R(t(j) ) exp{β̂ ′ xl (t)}
for t(k) 6 t < t(k+1) , k = 1, 2, . . . , r − 1, where xl (t) is the vector of values

of the explanatory variables for the lth individual at time t, and dj is the
number of events at the jth ordered event time, t(j) , j = 1, 2, . . . , r. Similar
modifications can be made to the other results in Section 3.10. With this
modification, computation of the summation over the risk set is much more
complicated, since for every event time, t(j) , j = 1, 2, . . . , r, the value of each
time-dependent variable, for all individuals in the risk set, is needed at that
event time.
Having obtained an estimate of the cumulative hazard function, the cor-
responding baseline hazard function can be estimated using Equation (3.30),
and an estimate of the baseline survivor function is S̃0 (t) = exp{−H̃0 (t)}.
The survivor function for a particular individual is much more difficult to
estimate. This is because the result that Si (t) can be expressed as a power of
the baseline survivor function, S0 (t), given in Equation (3.25) of Chapter 3, no
longer holds. Instead, the survivor function for the ith individual is obtained
from the integrated hazard function, which, from Equation (1.7) in Chapter 1,
is given by
{ ∫ t (∑p ) }
Si (t) = exp − exp βj xji (u) h0 (u) du . (8.4)
0 j=1
This survivor function therefore depends not only on the baseline hazard func-
tion h0 (t), but also on the values of the time-dependent variables over the
interval from 0 to t. The survivor function may therefore depend on future
values of the time-dependent variables in the model, which will generally be
unknown. However, approximate conditional probabilities of surviving a cer-
tain time interval can be found from the probability that an individual survives
over an interval of time, from t to t + h, say, conditional on being alive at time
t. This probability is P(Ti > t + h | Ti > t), where Ti is the random variable
associated with the survival time of the ith individual. Using the standard
result for conditional probability, given in Section 3.3.1 of Chapter 3, this
probability becomes P(Ti > t + h)/P(Ti > t), which is the ratio of the sur-
vivor functions at times t + h and t, that is, Si (t + h)/Si (t). We now assume
that any time-dependent variable remains constant through this interval, so
that from Equation (8.4), the approximate conditional probability is
{ (∑ )∫ }
p t+h
exp − exp j=1 βj xji (t) 0
h0 (u) du
Pi (t, t + h) = { (∑ )∫ } ,
p t
exp − exp j=1 βj xji (t) h (u) du
0 0
[ (∑p )]
= exp − {H0 (t + h) − H0 (t)} exp βj xji (t) ,
j=1
where H0 (t) is the baseline cumulative hazard function. An estimate of this

approximate conditional probability of surviving through the interval (t, t+h)
is then
[ { } (∑p )]
P̃i (t, t + h) = exp − H̃0 (t + h) − H̃0 (t) exp β̂j xji (t) , (8.5)
j=1
where H̃0 (t) is the estimated baseline cumulative hazard function obtained
on fitting the Cox regression model with p possibly time-dependent variables
with values xji (t), j = 1, 2, . . . , p, for the ith individual, i = 1, 2, . . . , n, and
β̂j is the estimated coefficient of the jth time-dependent variable. This result
was given by Altman and De Stavola (1994).
Corresponding estimates of the conditional probability of an event in the
interval (t, t + h) are 1 − P̃i (t, t + h), and these quantities can be used to
obtain an estimate of the expected number of events in each of a number
of successive intervals of width h. Comparing these values with the observed
number of events in these intervals leads to an informal assessment of model
adequacy.
8.3 Model comparison and validation

Models for survival data that include time-dependent variables can be com-
pared in the same manner as Cox proportional hazards models, using the
procedure described in Section 3.5 of Chapter 3. In particular, the model-
fitting process leads to a maximised partial likelihood function, from which
the value of the statistic −2 log L̂ can be obtained. Changes in the value of this
MODEL COMPARISON AND VALIDATION 303
statistic between alternative nested models may then be compared to percent-
age points of the chi-squared distribution, with degrees of freedom equal to
the difference in the number of β-parameters being fitted. For this reason, the
model-building strategies discussed in Chapter 3 apply equally in situations
where there are time-dependent variables.
8.3.1 Comparison of treatments

In order to examine the magnitude of a treatment effect after taking account of
variables that change over time, the value of −2 log L̂ for a model that contains
the time-dependent variables and any other prognostic factors is compared
with that for the model that contains the treatment term, in addition to
these other variables. But in this analysis, if no treatment effect is revealed,
one explanation could be that the time-dependent variable has masked the
treatment difference.
To fix ideas, consider the example of a study to compare two cytotoxic
drugs in the treatment of patients with leukaemia. Here, a patient’s survival
time may well depend on subsequent values of that patient’s white blood cell
count. If the effect of the treatment is to increase white blood cell count, no
treatment difference will be identified after including white blood cell count
as a time-dependent variable in the model. On the other hand, the treatment
effect may appear in the absence of this variable. An interpretation of this is
that the time-dependent variable has accounted for the treatment difference,
and so provides an explanation as to how the treatment has been effective.
In any event, much useful information will be gained from a comparison
of the results of an analysis that incorporates time-dependent variables with
an analysis that uses baseline values alone.
8.3.2 Assessing model adequacy

After fitting a model that includes time-dependent variables, a number of the
techniques for evaluating the adequacy of the model, described in Chapter
4, can be implemented. In particular, an overall martingale residual can be
computed for each subject, from an adaptation of the result in Equation (4.6).
The martingale residual for the ith subject is now
rM i = δi − exp{β̂ ′ xi (ti )}H̃0 (ti ),
where xi (ti ) is the vector of values of explanatory variables for the ith in-
dividual, which may be time-dependent, evaluated at ti , the event time of
that individual. Also, β̂ is the vector of coefficients, δi is the event indicator
that takes the value unity if ti is an event and zero otherwise, and H̃0 (ti ) is
the estimated baseline cumulative hazard function at ti , obtained from Equa-
tion (8.3). The deviance residuals may also be computed from the martingale
residuals, using Equation (4.7) of Chapter 4.
The plots described in Section 4.2.2 of Chapter 4 will often be helpful.
In particular, an index plot of the martingale residuals will enable outlying
observations to be identified. However, diagnostic plots for assessing the func-
tional form of covariates, described in Section 4.2.3, turn out to be not so
useful when a time-dependent variable is being studied. This is because there
will then be a number of values of the time-dependent covariate for any one
individual, and it is not clear what the martingale residuals for the null model
should be plotted against.
For detecting influential values, the delta-betas, introduced in Section 4.3.1
of Chapter 4, provide a helpful means of investigating the effect of each obser-
vation on individual parameter estimates. Changes in the value of the −2 log L̂
statistic, on omitting each observation in turn, can give valuable information
about the effect of each observation on the set of parameter estimates.
8.4 Some applications of time-dependent variables

One application of time-dependent variables is in connection with evaluat-
ing the assumption of proportional hazards. This was discussed in detail in
Section 4.4.3 of Chapter 4. In this application, a variable formed from the
product of an explanatory variable, X, and time, t, is added to the linear
part of the Cox model, and the null hypothesis that the coefficient of Xt is
zero is tested. If this estimated coefficient is found to be significantly different
from zero, there is evidence that the assumption of proportional hazards is not
valid.
In many circumstances, the waiting time from the occurrence of some cata-
strophic event until a patient receives treatment may be strongly associated
with the patient’s survival. For example, in a study of factors affecting the
survival of patients who have had a myocardial infarct, the time from the
infarct to when the patient arrives in hospital may be crucial. Some patients
may die before receiving treatment, while those who arrive at the hospital soon
after their infarct will tend to have a more favourable prognosis than those
for whom treatment is delayed. It will be important to take account of this
aspect of the data when assessing the effects of other explanatory variables
on the survival times of these patients.
In a similar example, Crowley and Hu (1977) show how a time-dependent
variable can be used in organ transplantation studies. Here, one feature of
interest is the effect of a transplant on the patient’s survival time. Suppose
that in a study on the effectiveness of a particular type of organ transplant, a
patient is judged to be suitable for a transplant at some time t0 . They then
wait some period of time until a suitable donor organ is found, and if the
patient survives this period, they receive a transplant at time t1 .
In studies of this type, the survival times of patients who have received
a transplant cannot be compared with those who have not had a transplant
in the usual way. The reason for this is that in order to receive a trans-
plant, a patient must survive the waiting time to transplant. Consequently,
SOME APPLICATIONS OF TIME-DEPENDENT VARIABLES 305
the group who survive to the time of transplant is not directly comparable
with the group who receive no such transplant. Similarly, it is not possible
to compare the times that the patients who receive a transplant survive after
the transplant with the survival times of the group not receiving a trans-
plant. Here, the time origin would be different for the two groups, and so they
are not comparable at the time origin. This means that it is not possible to
identify a time origin from which standard methods for survival analysis can
be used.
The solution to this problem is to introduce a time-dependent variable
X1 (t), which takes the value zero if a patient has not received a transplant at
time t, and unity otherwise. Adopting a Cox regression model, the hazard of
death for the ith individual at time t is then
hi (t) = exp{ηi + β1 x1i (t)}h0 (t),
where ηi is a linear combination of the explanatory variables that are not

time-dependent, whose values have been recorded at the time origin for the
ith individual, and x1i (t) is the value of X1 for that individual at time t.
Under this model, the hazard function is exp(ηi )h0 (t) for patients who
do not receive a transplant before time t, and exp{ηi + β1 }h0 (t) thereafter.
The effect of a transplant on the patient’s survival experience is then reflected
in β1 . In particular, for two patients who have the same values of other ex-
planatory variables in a model, eβ1 is the hazard of death at time t for the
patient who receives a transplant before time t, relative to the hazard at that
time for the patient who does not. Values of −2 log L̂ can be compared after
fitting the models with and without the time-dependent variable X1 . A sig-
nificant difference in these values means that the transplant has an effect on
survival.
In a refinement to this model, Cox and Oakes (1984) suggested that the
term β1 x1i (t) be replaced by β1 + β2 exp{−β3 (t − t1 )} for patients receiving a
transplant at time t1 . In this model, the effect of the transplant is to increase
the hazard to some value exp(ηi + β1 + β2 )h0 (t) immediately after the trans-
plant, when t = t1 , and to then decrease exponentially to exp(ηi + β1 )h0 (t),
which is less than the initial hazard exp(ηi )h0 (t) if β1 < 0. See Figure 8.4,
which shows graphically the behaviour of the hazard ratio, hi (t)/h0 (t), for a
transplant patient for whom ηi is the linear component of the model. Although
this is an attractive model, it does have the disadvantage that specialist soft-
ware is required to fit it.
In situations where a particular explanatory variable is changing rapidly,
new variables that reflect such changes may be defined. The dependence of
the hazard on the values of such variables can then be explored. For example,
in an oncological study, the percentage increase in the size of a tumour over
a period of time might be a more suitable prognostic variable than either
the size of the tumour at the time origin, or the time-dependent version of
that variable. If this route is followed, the computational burden of fitting
time-dependent variables can be avoided.
exp( i+ 1+ 2 )
Hazard ratio
exp( i)
exp( i+ 1)
0 t1
Time
Figure 8.4 The hazard ratio exp{ηi + β1 + β2 e−β3 (t−t1 ) }, t > t1 , for individual i who
receives a transplant at t1 .
8.5 Three examples

In this section, three examples of survival analyses that involve time-
dependent variables are given. In the first, data from a study concerning the
use of bone marrow transplantation in the treatment of leukaemia is used to
illustrate how a variable that is associated with the state of a patient, and
whose value changes over the follow-up period, can be included in a model. In
the second example, data from Example 5.10 of Chapter 5 on the comparison
of two chemotherapy treatments for ovarian cancer are analysed to explore
whether there is an interaction between age and survival time. The third ex-
ample is designed to illustrate how information on a time-varying explanatory
variate recorded during the follow-up period can be incorporated in a model
for survival times. Studies in which the values of certain explanatory variables
are recorded regularly throughout the follow-up period of each patient gener-
ate large sets of data. For this reason, artificial data from a small number of
individuals will be used in Example 8.3 to illustrate the methodology.
Example 8.1 Bone marrow transplantation in the treatment of leukaemia

Patients suffering from acute forms of leukaemia often receive a bone marrow
transplant. This provides the recipient with a new set of parent blood-forming
cells, known as stem cells, which in turn produce a supply of healthy red
and white blood cells. Klein and Moeschberger (2005) describe a multicentre
study of factors that affect the prognosis for leukaemia patients treated in
this manner. This study involved patients suffering from acute lymphoblastic
leukaemia (ALL) and acute myelocytic leukaemia (AML), with those suffering
THREE EXAMPLES 307
from AML being further divided into low-risk and high-risk, according to
their status at the time of transplantation. The survival time from the date
of the transfusion is available for each patient, together with the values of a
number of explanatory variables concerning the characteristics of the donor
and recipient, and adverse events that occurred during the recovery process.
Before the bone marrow transplant, patients were treated with a combination
of cyclophosphamide and busulfan, in order to destroy all abnormal blood
cells. The time taken for the blood platelets to return to a normal level is
then an important variable in terms of the prognosis for a patient, and so the
values of this variable were also recorded.
This example is based on the data from just one hospital, St. Vincent in
Sydney, Australia. The observed survival time of each patient was recorded in
days, together with the values of an event indicator which is unity if a patient
died, and zero if the patient was still alive at the end of the study period. The
prognostic variables to be used in this example concern the disease group, the
ages of the patient and the bone marrow donor, an indicator variable that
denotes whether the platelet count returned to a normal level of 40 × 109 per
litre, and the time taken to return to this value. Two patients, numbers 7 and
21, died before the platelet count had returned to normal, and so for these
patients, no value is given for the time to return to a normal platelet count.
The variables recorded are therefore as follows:

Status: Event indicator (0 = censored, 1 = event)
Group: Disease group (1 = ALL, 2 = low-risk AML,
3 = high-risk AML)
Page: Age of patient
Dage: Age of donor
Precovery: Platelet recovery indicator (0 = no, 1 = yes)
Ptime: Time in days to return of platelets to normal level (if P = 1)
The database used in this example is given in Table 8.1.

The aim of the analysis of these data is to examine whether there are any
differences between the survival times of patients in the three disease groups,
after adjusting for prognostic variables. In order to investigate the effect of
the time taken for platelets to return to their normal level on patient survival,
a time-dependent variable, P late(t), is defined. This variable takes the value
zero at times t when the platelets have not returned to normal levels, and
then switches to unity once a normal level has been achieved. Formally,
{
0 if t < time at which platelets returned to normal,
Plate(t) =
1 if t > time at which platelets returned to normal,
so that P late(t) = 0 for all t when a patient dies before platelet recovery.
We first fit a Cox proportional hazards model that contains the variables
associated with the age of the patient and donor, Page and Dage. When either
Table 8.1 Survival times of patients following bone marrow transplantation.

Patient Time Status Group Page Dage P recovery Ptime
1 1199 0 1 24 40 1 29
2 1111 0 1 19 28 1 22
3 530 0 1 17 28 1 34
4 1279 1 1 17 20 1 22
5 110 1 1 28 25 1 49
6 243 1 1 37 38 1 23
7 86 1 1 17 26 0
8 466 1 1 15 18 1 100
9 262 1 1 29 32 1 59
10 1850 0 2 37 36 1 9
11 1843 0 2 34 32 1 19
12 1535 0 2 35 32 1 21
13 1447 0 2 33 28 1 24
14 1384 0 2 21 18 1 19
15 222 1 2 28 30 1 52
16 1356 0 2 33 22 1 14
17 1136 0 3 47 27 1 15
18 845 0 3 40 39 1 20
19 392 1 3 43 50 1 24
20 63 1 3 44 37 1 16
21 97 1 3 48 56 0
22 153 1 3 31 25 1 59
23 363 1 3 52 48 1 19
of these variables is added on their own or in the presence of the other, there
is no significant reduction in the value of the −2 log L̂ statistic.
The time-dependent variable P late(t) is now added to the null model. The
value of −2 log L̂ is reduced from 67.13 to 62.21, a reduction of 4.92 on 1 d.f.,
which is significant at the 5% level (P = 0.026). This suggests that time to
platelet recovery does affect survival. After allowing for the effects of this
variable, there is still no evidence that the hazard of death is dependent on
the age of the patient or donor.
The estimated coefficient of P late(t) in the model that contains this vari-
able alone is −2.696, and the fact that this is negative indicates that there is a
greater hazard of death at any given time for a patient whose platelets are not
at a normal level. The hazard ratio at any given time is exp(−2.696) = 0.067,
and so a patient whose platelets have recovered to normal at a given time
has about one-fifteenth the risk of death at that time. However, a 95%
confidence interval for the corresponding true hazard ratio is (0.006, 0.751),
which shows that the point estimate of the relative risk is really quite
imprecise.
To quantify the effect of disease group on survival, the change in −2 log L̂
when the factor Group is added to the model that contains the time-
THREE EXAMPLES 309
dependent variable P late(t) is 6.49 on 2 d.f., which is significant at the 5%
level (P = 0.039). The parameter estimates associated with disease group
show that the hazard of death is much greater for those suffering from ALL
and those in the high-risk group of AML sufferers. The hazard ratios for an
ALL patient relative to a low-risk AML patient is 7.97 and that for a high-risk
AML patient relative to a low-risk one is 11.77.
For the model that contains the factor Group and the time-dependent vari-
able P late(t), the estimated baseline cumulative hazard and survivor functions
are given in Table 8.2. These have been obtained using the estimate of the
baseline cumulative hazard function given in Equation (8.3).
Table 8.2 Estimated baseline cumulative hazard,

H̃0 (t), baseline survivor function, S̃0 (t), and survivor
function for an ALL patient with P late(t) = 1, S̃1 (t).
Time, t H̃0 (t) S̃0 (t) S̃1 (t)
0 0.0000 1.0000 1.0000
63 0.1953 0.8226 0.9810
86 0.3962 0.6728 0.9618
97 0.6477 0.5232 0.9383
110 1.2733 0.2799 0.8823
153 1.9399 0.1437 0.8264
222 2.6779 0.0687 0.7685
243 3.4226 0.0326 0.7143
262 4.2262 0.0146 0.6600
363 5.0987 0.0061 0.6057
392 6.0978 0.0022 0.5491
466 7.2663 0.0007 0.4895
1279 13.0700 0.0000 0.2766
In this table, H̃0 (t) and S̃0 (t) are the estimated cumulative hazard and
survivor functions for an individual with AML and for whom the platelet re-
covery indicator, Plate(t), remains at zero throughout the study. Also given in
this table are the values of the estimated survivor function for an individual
with ALL, but for whom Plate(t) = 1 for all values of t, denoted S̃1 (t). Since
the value of Plate(t) is zero for each patient at the start of the study, and
for most patients this changes to unity at some later point in time, these two
estimated survivor functions illustrate the effect of platelet recovery at any
specific time. For example, the probability of an ALL patient surviving be-
yond 97 days is only 0.52 if their platelets have not recovered to a normal level
by this time. On the other hand, if such a patient has experienced platelet
recovery by this time, they would have an estimated survival probability of
0.94. The estimated survivor function for an ALL patient whose platelet re-
covery status changes at some time t0 from 0 to 1 can also be obtained from
Table 8.2, since this will be S̃0 (t) for t 6 t0 and S̃1 (t) for t > t0 . Estimates of
the survivor function may also be obtained for individuals in the other disease
groups.
In this illustration, the data from two patients who died before their
platelet count had reached a normal level have a substantial impact on in-
ferences about the effect of platelet recovery. If patients 7 and 21 are omitted
from the database, the time-dependent variable is no longer significant when
added to the null model (P = 0.755). The conclusion about the effect of
platelet recovery time on survival is therefore dramatically influenced by the
data for these two patients.

Consider again the data on patient survival following diagnosis of ovarian
cancer, given in Example 5.10 of Chapter 5. When a Cox proportional hazards
model that contains the variables Age, the age of a patient at the time origin,
and Treat, the treatment group, is fitted to the data on the survival times of
patients with ovarian cancer, the estimated hazard function for the ith of 26
patients in the study is
ĥi (t) = exp{0.147 Age i − 0.796 Treat i }h0 (t).
The value of the −2 log L̂ statistic for this model is 54.148.

We now fit a model that contains Age and Treat, and a term corresponding
to an interaction between age and survival time. This interaction will be mod-
elled by including the time-dependent variable Tage, whose values are formed
from the product of Age and the survival time t, that is, Tage = Age × t. Since
the values of Tage are dependent upon t, this time-dependent variable cannot
be fitted in the same manner as Age and Treat. When Tage is added to the
model, the fitted hazard function becomes
ĥi (t) = exp{0.216 Age i − 0.664 Treat i − 0.0002 Age i t}h0 (t).
Under this model, the hazard of death at t for a patient of a given age on the
combined treatment (Treat = 2), relative to one of the same age on the single
treatment (Treat = 1), is exp(−0.664) = 0.52, which is not very different from
the value of 0.45 found using the model that does not contain the variable
Tage. However, the log-hazard ratio for a patient aged a2 years, relative to
one aged a1 years, is
0.216(a2 − a1 ) − 0.0002(a2 − a1 )t
at time t. This model therefore allows the log-hazard ratio for Age to be
linearly dependent on survival time.
The value of −2 log L̂ for the model that contains Age, Treat and Tage is
53.613. The change in −2 log L̂ on adding the variable Tage to a model that
contains Age and Treat is therefore 0.53, which is not significant (P = 0.465).
We therefore conclude that the time-dependent variable Tage is not in fact
needed in the model.
THREE EXAMPLES 311
Example 8.3 Data from a cirrhosis study
Although the data to be used in this example are artificial, it is useful to
provide a background against which these data can be considered. Suppose
therefore that 12 patients have been recruited to a study on the treatment of
cirrhosis of the liver. The patients are randomised to receive either a placebo or
a new treatment that will be referred to as Liverol. Six patients are allocated
to Liverol and six to the placebo. At the time when the patient is entered
into the study, the age and baseline value of the patient’s bilirubin level are
recorded. The natural logarithm of the bilirubin value (in µmol/l) will be used
in this analysis, and the variables measured are summarised below:
Time: Survival time of the patient in days

Status: Event indicator (0 = censored, 1 = uncensored)
Treat: Treatment group (0 = placebo, 1 = Liverol)
Age: Age of the patient in years
Lbr: Logarithm of bilirubin level
The values of these variables are given in Table 8.3.
Table 8.3 Survival times of 12 patients in a study

on cirrhosis of the liver.
Patient Time Status Treat Age Lbr
1 281 1 0 46 3.2
2 604 0 0 57 3.1
3 457 1 0 56 2.2
4 384 1 0 65 3.9
5 341 0 0 73 2.8
6 842 1 0 64 2.4
7 1514 1 1 69 2.4
8 182 0 1 62 2.4
9 1121 1 1 71 2.5
10 1411 0 1 69 2.3
11 814 1 1 77 3.8
12 1071 1 1 58 3.1
Patients are supposed to return to the clinic three, six and twelve months
after the commencement of treatment, and yearly thereafter. On these occa-
sions, the bilirubin level is again measured and recorded. Data are therefore
available on how the bilirubin level changes in each patient throughout the du-
ration of the study. Table 8.4 gives the values of the logarithm of the bilirubin
value at each time in the follow-up period for each patient.
In taking log(bilirubin) to be a time-dependent variable, the value of the
variate is that recorded at the most recent follow-up visit, for each patient.
In this calculation, the change to a new value will be assumed to take place
immediately after the reading was taken, so that patient 1, for example, is
assumed to have a log(bilirubin) value of 3.2 for any time t when t 6 47,
Table 8.4 Follow-up times and log(bilirubin) values

for the 12 patients in the cirrhosis study.
Patient Follow-up time Log(bilirubin)
1 47 3.8
184 4.9
251 5.0
2 94 2.9
187 3.1
321 3.2
3 61 2.8
97 2.9
142 3.2
359 3.4
440 3.8
4 92 4.7
194 4.9
372 5.4
5 87 2.6
192 2.9
341 3.4
6 94 2.3
197 2.8
384 3.5
795 3.9
7 74 2.9
202 3.0
346 3.0
917 3.9
1411 5.1
8 90 2.5
182 2.9
9 101 2.5
410 2.7
774 2.8
1043 3.4
10 182 2.2
847 2.8
1051 3.3
1347 4.9
11 167 3.9
498 4.3
12 108 2.8
187 3.4
362 3.9
694 3.8
THREE EXAMPLES 313
3.8 for 47 < t 6 184, 4.9 for 184 < t 6 251, and 5.0 for 251 < t 6 281. The
values of Lbr for a given individual then follow a step-function in which the
values are assumed constant between any two adjacent time points.
The data are first analysed using the baseline log(bilirubin) value alone. A
Cox proportional hazards model is used, and the values of −2 log L̂ on fitting
particular models are as shown in Table 8.5.

without a time-dependent variable.
Terms in model −2 log L̂
null model 25.121
Age 22.135
Lbr 21.662
Age, Lbr 18.475
Both Age and Lbr appear to be needed in the model, although the evidence
for including Age as well as Lbr is not very strong. When Treat is added to
the model that contains Age and Lbr, the reduction in the value of −2 log L̂ is
5.182 on 1 d.f. This is significant at the 5% level (P = 0.023). The coefficient
of Treat is −3.052, indicating that the drug Liverol is effective in reducing the
hazard of death. Indeed, other things being equal, Liverol reduces the hazard
of death by a factor of 0.047.
We now analyse these data, taking the log(bilirubin) values to be time-
dependent. Let Lbrt be the time-dependent variate formed from the values of
log(bilirubin). The values of −2 log L̂ on fitting Cox regression models to the
data are then given in Table 8.6.

with a time-dependent variable.
Terms in model −2 log L̂
null model 25.121
Age 22.135
Lbrt 12.050
Age, Lbrt 11.145
It is clear from this table that the hazard function depends on the time-
dependent variable Lbrt, and that after allowing for this, the effect of Age is
slight. We therefore add the treatment effect Treat to the model that contains
Lbrt alone. The effect of this is that −2 log L̂ is reduced from 12.050 to 10.676,
a reduction of 1.374 on 1 d.f. This reduction is not significant (P = 0.241)
leading to the conclusion that after taking account of the dependence of the
hazard of death on the evolution of the log(bilirubin) values, no treatment
effect is discernible.
The estimated hazard function for the ith individual is given by
ĥi (t) = exp{3.605 Lbr i (t) − 1.479 Treat i }h0 (t),
where Lbr i (t) is the value of log(bilirubin) for this patient at time t. The
estimated ratio of the hazard of death at time t for two individuals on the same
treatment who have values of Lbr that differ by 0.1 units at t is e0.3605 = 1.43.
This means that the individual whose log(bilirubin) value is 0.1 units greater
has close to a 50% increase in the hazard of death at time t.
One possible explanation for the difference between the results of these
two analyses is that the effect of the treatment is to change the values of
the bilirubin level, so that after changes in these values over time have been
allowed for, no treatment effect is visible.
The baseline cumulative hazard function may now be estimated for the
model that contains the time-dependent variable Lbrt and Treat. The esti-
mated values of this function are tabulated in Table 8.7.
Table 8.7 Estimated baseline cumulative haz-

ard function, H̃0 (t), for the cirrhosis study.
Follow-up time (t) H̃0 (t)
0 0.000
281 0.009 × 10−6
384 0.012 × 10−6
457 0.541 × 10−6
814 0.908 × 10−6
842 1.577 × 10−6
1071 3.318 × 10−6
1121 6.007 × 10−6
1514 6.053 × 10−6
This table shows that the cumulative hazard function is increasing in a

non-linear fashion, which indicates that the baseline hazard is not constant,
but increases with time. The corresponding baseline survivor function could
be obtained from this estimate. However, a model with Lbrt = 0 for all t is
not at all easy to interpret, and so the estimated survivor function is obtained
for an individual for whom Lbrt = 3. This function is shown in Figure 8.5, for
a patient in either treatment group. This figure clearly shows that patients on
placebo have a much poorer prognosis than those on Liverol.
Finally, we illustrate how estimates of the conditional probabilities of sur-
vival from time t to time t + 360 days can be obtained, using the result given
in Equation (8.5). Using the values of the time-dependent variable given in
∑p
Tables 8.3 and 8.4, the values of j=1 β̂j xji (t), the prognostic index at time t
for the ith patient, i = 1, 2, . . . , 12, can be calculated for t = 0, 360, 720, 1080,
and 1440. For this calculation, the log(bilirubin) value at one of these times is
taken to be the value recorded at the immediately preceding follow-up time.
Table 8.7 is then used to obtain the values of H̃0 (t + 360) − H̃0 (t), and then
THREE EXAMPLES 315
1.0
0.9
0.8
0.7
0.6
0.5
0 200 400 600 800 1000 1200 1400 1600
Survival time
Figure 8.5 Estimated survivor function for a patient with Lbr = 3, for all t, who is
on placebo (—) or Liverol (·······).
P̃i (t, t+360) is obtained from Equation (8.5). The full set of results will not be
given here, but as an example, the estimated approximate conditional prob-
abilities of surviving through consecutive intervals of 360 days, for patients 1
and 7, are shown in Table 8.8.
Table 8.8 Approximate conditional survival

probabilities for patients 1 and 7.
Time interval P̃1 (t, t + h) P̃7 (t, t + h)
0– 0.999 1.000
360– 0.000 0.994
720– 0.000 0.969
1080– 0.000 0.457
1440– 0.045 0.364
Note that because these estimates are survival probabilities conditional

on being alive at the start of an interval, they are not necessarily decreasing
functions of time. These estimates again show that patients on Liverol have
a greater probability of surviving for a period of one year, if they are alive
at the start of that year, than patients on the placebo. Finally, the values of
1 − P̃i (t, t + h) are approximate estimates of the probability of death within
each interval, conditional on a patient being alive at the start. Summing these
estimates over all 12 patients leads to the values 0.02, 2.46, 5.64, 6.53, 3.16, for
each of the five intervals, respectively. These can be compared to the observed
numbers of deaths in each interval, which are 1, 2, 3, 1 and 1, respectively.
There is therefore a tendency for the model to overestimate the numbers of
deaths, but because of the small size of the data set, this does not provide a
very reliable assessment of the predictive power of the model.
8.6 Counting process format

Models that include time-dependent variables can be fitted by maximising the
partial log-likelihood function in Equation (8.2). This can be accomplished us-
ing software that enables the value of any time-dependent variable to be de-
termined at different event times in the data set, as explained in Section 8.2.1,
and this approach was used in the examples of Section 8.5.
Time-dependent variables may also be included in a model by expressing
the data in counting process format. This involves setting intervals of time over
which the values of all explanatory variables are constant. The event indicator
is then set to zero for all intervals except the final interval. The upper limit of
the final interval is the event time, or censoring time, of the individual, and
the value of the event indicator is set to unity if there is an event at the end
of that interval, and zero otherwise. Variables associated with the lower and
upper limits of each time interval, Start and Stop, say, are then used to specify
survival experience, and the variable Status denotes whether or not an event
has occurred at each stop time. The survival data are then represented by a
trio of values (Start, Stop, Status) for each interval, and this format of data
is called the counting process or (start, stop, status) format. Many software
packages for survival analysis can process data expressed in this general form.
Further details on the counting process format, and its application to other
areas of survival analysis, are given in Section 13.1.3 of Chapter 13.
Example 8.4 Data from a cirrhosis study

To illustrate the counting process form of survival data, consider again the
data from Example 8.3. For patient 1, the data for whom were given in Ta-
ble 8.4, the log(bilirubin) value was 3.2 when measured at baseline, 3.8 from
day 47, 4.9 from day 184 and 5.0 from day 251. This patient died on day
281, and so the value of the event indicator is 0 until the end of the final
interval. The values of the explanatory variables Status, Treat, Age and the
time-dependent variable Lbrt in the intervals (0, 47], (47, 184], (184, 251] and
(251, 281], are then as shown in Table 8.9. In this notation for the time inter-
vals, (a, b] denotes the interval for a time t, where a < t 6 b.
Table 8.9 Data for the first patient in the cirrhosis study in the
counting process format.
Time interval Start Stop Status Treat Age Lbrt
(0, 47] 0 47 0 0 46 3.2
(47, 184] 47 184 0 0 46 3.8
(184, 251] 184 251 0 0 46 4.9
(251, 281] 251 281 1 0 46 5.0
FURTHER READING 317
The database now has four lines of data for Patient 1, and we proceed in
a similar manner for the data from the remaining patients. A Cox regression
model with the variables Age, Treat and the time-dependent variable Lbrt is
then fitted to the extended data set, which leads to the same results as those
given in Example 8.3.
8.7 Further reading

The possibility of incorporating time-dependent variables in a proportional
hazards model was raised by Cox (1972). The appropriate partial likelihood
function was given in his paper, and discussed in greater detail in Cox (1975).
Kalbfleisch and Prentice (2002) include a detailed account of the construc-
tion of the partial likelihood function. The classification of time-dependent
variables outlined in Section 8.1 is due to Prentice and Kalbfleisch (1979),
who amplify on this in Kalbfleisch and Prentice (2002). Andersen (1992) re-
views the uses of time-dependent variables in survival analysis and includes
an example on which the hypothetical study of Example 8.3 is loosely based.
A number of practical problems encountered in the analysis of survival data
with time-dependent variables are discussed by Altman and De Stavola (1994),
who include a review of software available at that time. A comprehensive
analysis of data on primary biliary cirrhosis, which includes an assessment of
conditional survival probabilities is also provided. See Christensen et al. (1986)
for a further illustration. Klein and Moeschberger (2005) give the full data set
on survival following bone marrow transplantation, part of which was used in
Example 8.1, and use this in a number of detailed illustrative examples.
The model described in Section 8.4 in connection with organ transplanta-
tion was presented by Crowley and Hu (1977) and Cox and Oakes (1984) in
an analysis of the ‘Stanford heart transplant data’. This famous data set is
given in Crowley and Hu (1977) and an update is provided by Cox and Oakes
(1984). See also Aitkin, Laird and Francis (1983) and the ensuing discussion.
Relatively little work has been done on incorporating time-dependent vari-
ables in a fully parametric model for survival data. However, Petersen (1986)
shows how a parametric model with time-dependent variables can be fit-
ted. The Royston and Parmar model, described in Section 6.9 of Chapter
6, can be extended to include time-dependent variables by adding an inter-
action between potential time-dependent variables and the spline variables,
ν1 (y), ν2 (y), . . ., in the baseline cumulative hazard function of Equation (6.30)
of Chapter 6.
Joint models for longitudinal and survival data that describe underlying
relationships between measurements made on an individual at multiple times
and the time to an event, are also relevant in this context. Recent review
papers include Tsiatis and Davidian (2004), Tseng, Hsieh and Wang (2005),
McCrink, Marshall and Cairns (2013) and the text of Rizopoulis (2012).
Chapter 9
Interval-censored survival data
In many studies where the response variable is a survival time, the exact time
of the event of interest will not be known. Instead, the event will be known
to have occurred during a particular interval of time. Data in this form are
known as grouped or interval-censored survival data.
Interval-censored data commonly arise in studies where there is a non-
lethal end-point, such as the recurrence of a disease or condition. However,
most survival analyses are based on interval-censored data, in the sense that
the survival times are often taken as the nearest day, week or month. In this
chapter, some methods for analysing interval-censored data will be described
and illustrated. Models in which specific assumptions are made about the form
of the underlying hazard function are considered in Sections 9.1 to 9.4, and
fully parametric models are discussed in Section 9.5.
9.1 Modelling interval-censored survival data

In this chapter, a number of methods for the analysis of interval-censored sur-
vival data will be discussed in the context of a study on disease recurrence.
In the management of patients who have been cured of ulcers, carcinomas or
other recurrent conditions, the patients are usually provided with medication
to maintain their recovery. These patients are subsequently examined at reg-
ular intervals in order to detect whether a recurrence has occurred. Naturally,
some patients may experience symptoms of a recurrence and be subsequently
diagnosed as having had a recurrence at a time other than one of the scheduled
screening times.
Now suppose that the study is designed to compare two maintenance ther-
apies, a new and a standard treatment, say, and that a number of explanatory
variables are recorded for each individual when they are recruited to the study.
The vector xi will be used to denote the set of values of p explanatory vari-
ables, X1 , X2 , . . . , Xp , for the ith individual in the study. The first of these
variables, X1 , will be taken to be an indicator variable corresponding to the
treatment group, where X1 = 0 if an individual is on the standard treatment
and X1 = 1 if on the new treatment.
Clearly, one way of analysing such data is to ignore the interval censoring.
A survival analysis is then carried out on the times of a detected recurrence.
319
320 INTERVAL-CENSORED SURVIVAL DATA
However, the data set used in this analysis will be based on a mixture of
recurrences detected at scheduled screening times, known as screen-detected
recurrences and recurrences diagnosed following the occurrence of symptoms
or interval-detected recurrences. This leads to a difficulty in interpreting the
results of the analysis.
To illustrate the problem, consider a study to compare two treatments for
suppressing the recurrence of an ulcer, a new and a standard treatment, say.
Also suppose that both treatments have exactly the same effect on the recur-
rence time, but that the new treatment suppresses symptoms. The recurrence
of an ulcer in a patient on the new treatment will then tend to be detected
later than that in a patient on the standard treatment. Therefore, interval-
detected recurrences will be identified sooner in a patient on the standard
treatment. The interval-detected recurrence times will then be shorter for this
group of patients, indicating an apparent advantage of the new treatment over
the standard.
If the time interval between successive screenings is short, relative to the
average time to recurrence, there will be few interval-detected recurrences.
Standard methods for survival analysis may then be used.
Example 9.1 Recurrence of an ulcer

In a double-blind clinical trial to compare treatments for the inhibition of
relapse after primary therapy has healed an ulcer, patients are randomised
to receive one or other of two treatments, labelled A and B. Regular visits
to a clinic were arranged for the patients, and endoscopies were performed
6 months and 12 months after randomisation. A positive endoscopy result
indicates that an ulcer has recurred in the time since the last negative result.
Information is therefore obtained on whether or not an ulcer has recurred
in the interval from 0 to 6 months or in the interval from 6 to 12 months.
Additionally, some patients presented at the clinic between scheduled visits,
suffering from symptoms of recurrence. These patients had endoscopies at
these times in order to detect if a recurrence had in fact occurred.
At entry to the trial, the age of each person, in years, and the duration
of verified disease (1 = less than five years, 2 = greater than or equal to five
years) was recorded, in addition to the treatment group (A or B). There are
two variables associated with ulcer detection in the data set, namely the time
of the last visit, in months, and the result of the endoscopy (1 = no ulcer
detected, 2 = ulcer detected). Those with times other than 6 or 12 months
had presented with symptoms between scheduled visits.
The study itself was multinational and the full set of data is given in
Whitehead (1989). In this example, only the data from Belgium will be used,
and the relevant data are given in Table 9.1.
Once an ulcer is detected by endoscopy, a patient is treated for this and
is then no longer in the study. There were some patients who either did not
have an endoscopy six months after trial entry, or who dropped out after a
negative unscheduled endoscopy in the first six months. These patients have
MODELLING INTERVAL-CENSORED SURVIVAL DATA 321
Table 9.1 Data on the recurrence of an ulcer following treatment for the
primary disease.
Patient Age Duration Treatment Time of last visit Result
1 48 2 B 7 2
2 73 1 B 12 1
3 54 1 B 12 1
4 58 2 B 12 1
5 56 1 A 12 1
6 49 2 A 12 1
7 71 1 B 12 1
8 41 1 A 12 1
9 23 1 B 12 1
10 37 1 B 5 2
11 38 1 B 12 1
12 76 2 B 12 1
13 38 2 A 12 1
14 27 1 A 6 2
15 47 1 B 6 2
16 54 1 A 6 1
17 38 1 B 10 2
18 27 2 B 7 2
19 58 2 A 12 1
20 75 1 B 12 1
21 25 1 A 12 1
22 58 1 A 12 1
23 63 1 B 12 1
24 41 1 A 12 1
25 47 1 B 12 1
26 58 1 A 3 2
27 74 2 A 2 2
28 75 2 A 6 1
29 72 1 A 12 1
30 59 1 B 12 2
31 52 1 B 12 1
32 75 1 B 12 2
33 76 1 A 12 1
34 34 2 A 6 1
35 36 1 B 12 1
36 59 1 B 12 1
37 44 1 A 12 2
38 28 2 B 12 1
39 62 1 B 12 1
40 23 1 A 12 1
41 49 1 B 12 1
42 61 1 A 12 1
43 33 2 B 12 1
been omitted from the data set on the grounds that there is no information
about whether an ulcer has recurred in the first six months of the study. This
means that those patients in Table 9.1 whose last visit was greater than 6
months after randomisation would have had a negative endoscopy at 6 months.
In modelling the data from this study, duration of disease is denoted by
an indicator variable Dur, which is zero when the duration is less than 5 years
and unity otherwise. The treatment effect is denoted by a variable Treat,
which takes the value zero if an individual is on treatment A and unity if on
treatment B. The patient’s age is reflected in the continuous variate Age.
We first analyse the recurrence times in Table 9.1 ignoring the interval
censoring. The recurrence times of those patients who have not had a detected
recurrence by the time of their last visit are taken to be censored, and a Cox
regression model is fitted.
Table 9.2 Values of −2 log L̂ on fitting

a Cox regression model to data on the
time to recurrence of an ulcer.
None 79.189
Dur 79.157
Age 78.885
Age + Dur 78.872
Age + Dur + Treat 78.747
Treat 79.097
From the values of the −2 log L̂ statistic for different models, given in
Table 9.2, it is clear that neither age nor duration of disease are important
prognostic factors. Moreover, the reduction in −2 log L̂ on adding the treat-
ment effect to the model, adjusted or unadjusted for the prognostic factors, is
nowhere near significant.
The estimated coefficient of Treat in the model that contains Treat alone
is 0.189, and the standard error of this estimate is 0.627. The estimated haz-
ard of a recurrence under treatment B (Treat = 1), relative to that under
treatment A (Treat = 0), is therefore exp(0.189) = 1.21. The standard error
of the estimated hazard ratio is found using Equation (3.13) in Chapter 3,
and is 0.758. The fact that the estimated hazard ratio is greater than unity
gives a slight indication that treatment A is superior to treatment B, but not
significantly so.
9.2 Modelling the recurrence probability in the follow-up period

Suppose that patients are followed up to time ts , at which time the last sched-
uled screening test is carried out. Information on whether or not a recurrence
was detected at any time up to and including the last screen is then recorded.
Let pi (ts ) be the probability of a recurrence up to time ts for the ith patient,
i = 1, 2, . . . , n, with vector of explanatory variables xi . We now adopt a Cox
proportional hazards model, according to which the hazard of a recurrence at
ts , for the ith patient, is given by
hi (ts ) = exp(ηi )h0 (ts ),
where ηi = β ′ xi = β1 x1i + β2 x2i + · · · + βp xpi is the risk score, and h0 (ts ) is

the baseline hazard function at ts .
The probability that the ith individual experiences a recurrence after time
ts is the survivor function Si (ts ), so that Si (ts ) = 1 − pi (ts ). Now, from
Equation (3.25) in Section 3.10 of Chapter 3,
Si (ts ) = {S0 (ts )}exp(ηi ) , (9.1)
where S0 (ts ) is the value of the survivor function at ts for an individual on

the standard treatment for whom all the other explanatory variables are zero.
The probability of a recurrence up to time ts under this model is therefore
pi (ts ) = 1 − {S0 (ts )}exp(ηi ) ,
and so
log[− log{1 − pi (ts )}] = ηi + log{− log S0 (ts )}.
Writing β0 = log{− log S0 (ts )}, the model can be expressed as
log[− log{1 − pi (ts )}] = β0 + β1 x1i + β2 x2i + · · · + βp xpi . (9.2)
This is a linear model for the complementary log-log transformation of the

probability of a recurrence up to time ts . The model can be fitted to data on
a binary response variable that takes the value zero for those individuals in
the study who have not experienced a recurrence before ts , the time of the
last screen, and unity otherwise.
As in modelling survival data, models fitted to data on a binary response
variable can be compared on the basis of the statistic −2 log L̂. Here, L̂ is
the maximised likelihood of the binary data, and −2 log L̂ is generally known
as the deviance. Differences in the deviance for two nested models have an
asymptotic chi-squared distribution, and so models fitted to binary data can
be compared in the same manner as the models used in survival analysis.
When the model in Equation (9.2) is fitted to the observed data, the
estimate of the constant, β̂0 , is an estimate of log{− log S0 (ts )}, from which
an estimate of the baseline survivor function at ts can be obtained. Also, the
ratio of the hazard of a recurrence for an individual on the new treatment,
relative to one on the standard, is exp(β1 ). This can be estimated by exp(β̂1 ),
where β̂1 is the parameter estimate corresponding to X1 , the indicator variable
that corresponds to the treatment group. Values of the hazard ratio less than
unity suggest that the risk of a recurrence at any time is smaller under the
new treatment than under the standard. A confidence interval for the hazard
ratio may be obtained from the standard error of β̂1 in the usual manner.
This method of estimating the hazard ratio from interval-censored survival
data is not particularly efficient, since data on the times that a recurrence is
detected are not utilised. However, the method is appropriate when interest
simply centres on the risk of a recurrence in a specific time period. It is also the
method that would be adopted in modelling quantities such as the probability
of a relapse in the first year of treatment, or the probability of no recurrence
in a five-year period after trial entry.
We now model the probability of an ulcer recurring in the 12 months following
recruitment to the study described in Example 9.1. Of the 43 patients in the
data set, 11 of them had experienced a recurrence in this 12-month period,
namely patients 1, 10, 14, 15, 17, 18, 26, 27, 30, 32 and 37. A binary response
variable is now defined, which takes the value unity if a patient has experienced
a recurrence and zero otherwise. A model in which the complementary log-
log transformation of the recurrence probability is related to age, duration of
disease and treatment group is then fitted to the binary observations.
Table 9.3 gives the deviances on fitting complementary log-log models with
different terms to the binary response variable. All the models fitted include
a constant term.
Table 9.3 Deviances on fitting complementary log-log models

to data on the recurrence of an ulcer in 12 months.
Variables in model Deviance d.f.
Constant 48.902 42
Dur 48.899 41
Age 48.573 41
Treat 48.531 41
Dur + Age 48.565 40
Dur + Treat 48.531 40
Age + Treat 48.175 40
Dur + Age + Treat 48.172 39
Dur + Age + Treat + Treat × Age 47.944 38
Dur + Age + Treat + Treat × Dur 48.062 38
In this example, the effects of age, duration of disease and treatment group
have been modelled using the variates Age, Dur, and Treat, defined in Example
9.1. However, factors corresponding to duration and treatment could have been
used in conjunction with packages that allow factors to be included directly.
This would not make any difference to the deviances in Table 9.3, but it may
have an effect on the interpretation of the parameter estimates. See Sections
3.2 and 3.9 for fuller details.
It is clear from Table 9.3 that no variable reduces the deviance by a signif-
icant amount. For example, the change in the deviance on adding Treat to the
model that only contains a constant is 0.371, which is certainly not significant
when compared to percentage points of the chi-squared distribution on 1 d.f.
MODELLING RECURRENCE AT DIFFERENT TIMES 325
Approximately the same change in deviance is found when Treat is added to
the model that contains Age and Dur, showing that the treatment effect is of
a similar magnitude after allowing for these two variables. Moreover, there is
no evidence whatsoever of an interaction between treatment and the variables
Age and Dur.
On fitting a model that contains Treat alone, the estimated coefficient of
Treat is 0.378, with a standard error of 0.629. Thus, the ratio of the hazard
of a recurrence before 12 months in a patient on treatment B (Treat = 1),
relative to that for a patient on treatment A (Treat = 0), is exp(0.378) =
1.46. The risk of a recurrence in the year following randomisation is thus
greater under treatment B than it is under treatment A, but not significantly
so. This hazard ratio is not too different from the value of 1.21 obtained in
Example 9.1. The standard error of the estimated hazard ratio, again found
using Equation (3.13) in Chapter 3, is 0.918, which is also very similar to that
found in Example 9.1.
A 95% confidence interval for the log-hazard ratio has limits of 0.378 ±
1.96 × 0.629, and so the corresponding interval estimate for the hazard ratio
itself is (0.43, 5.01). Notice that this interval includes unity, a result which was
foreshadowed by the non-significant treatment effect.
The estimated constant term in this fitted model is −1.442. This is an
estimate of log{− log S0 (12)}, the survivor function at 12 months for a patient
on treatment A. The estimated probability of a recurrence after 12 months for
a patient on treatment A is therefore exp(−e−1.442 ) = 0.79. The corresponding
value for a patient on treatment B is 0.79exp(0.378) = 0.71. The probabilities of
a recurrence in the first 12 months are therefore 0.21 for a patient on treatment
A, and 0.29 for a patient on treatment B. This again shows that patients on
treatment B have a slightly higher probability of the recurrence of an ulcer in
the year following randomisation.
9.3 ∗ Modelling the recurrence probability at different times

In this section, a method for analysing interval-censored survival data is de-
scribed, which takes account of whether or not a recurrence is detected at
different examination times.
Suppose that patients enter a study at time 0 and are followed up to time
tk . During the course of this follow-up period, the individuals are screened
on a regular basis in order to detect a recurrence of the disease or condition
under study. Denote the examination times by t1 , t2 , . . . , tk , which are such
that t1 < t2 < · · · < tk . Further, let t0 denote the time origin, so that t0 = 0
and let tk+1 = ∞.
For each individual, information will be recorded on whether or not a recur-
rence has occurred at times t1 , t2 , . . . , tk . It can then be determined whether
a given individual has experienced a recurrence in the jth time interval from
tj−1 to tj . Thus, a patient who has a recurrence detected at time tj has an
actual recurrence time of t, where tj−1 6 t < tj , j = 1, 2, . . . , k. Note that
the study will not provide any information about whether a recurrence occurs
after the final screening time, tk .
Now let pij be the probability of a recurrence being detected in the ith
patient, i = 1, 2, . . . , n, at time tj , so that pij is the probability that patient i
experiences a recurrence in the jth time interval, j = 1, 2, . . . , k. Also let πij
be the probability that the ith of n patients is found to be free of the disease at
time tj−1 and has a recurrence in the jth time interval, j = 1, 2, . . . , k. This is
therefore the conditional probability of a recurrence in the jth interval, given
that the recurrence occurs after tj−1 . Using Ti to denote the random variable
associated with the recurrence time of the ith individual, we therefore have
pij = P(tj−1 6 Ti < tj ),
and
πij = P(tj−1 6 Ti < tj | Ti > tj−1 ),
for j = 1, 2, . . . , k.
We now consider individuals who have not had a detected recurrence by
the last examination time, tk . For these individuals, we define Ti to be the
random variable associated with the time to either a recurrence or death, and
the corresponding probability of a recurrence or death in the interval from
time tk is given by
∑
k
pi,k+1 = P(Ti > tk ) = 1 − pij .
j=1
Also, the corresponding conditional probability of a recurrence or death in the

interval (tk , ∞) is
πi,k+1 = P(Ti > tk | Ti > tk ) = 1.
It then follows that
pij = (1 − πi1 )(1 − πi2 ) · · · (1 − πi,j−1 )πij , (9.3)
for j = 2, 3, . . . , k + 1, with pi1 = πi1 .

Now let rij be unity if the ith patient has a recurrence detected in the
interval from tj−1 to tj , j = 1, 2, . . . , k+1, and zero if no recurrence is detected
in that interval, with ri,k+1 = 1. Also let sij be unity if a patient has a detected
recurrence after tj , and zero otherwise. Then,
sij = ri,j+1 + ri,j+2 + · · · + ri,k+1 ,
for j = 1, 2, . . . , k.
The sample likelihood of the n(k + 1) values rij is
∏ ∏
n k+1
r
pijij ,
i=1 j=1
and on substituting for pij from Equation (9.3), the likelihood function be-
comes
∏ ∏
n k+1
{(1 − πi1 ) · · · (1 − πi,j−1 )πij }rij .
i=1 j=1
This function can be written as

∏
n
ri1
πi1 {(1 − πi1 )πi2 }ri2 · · · {(1 − πi1 ) · · · (1 − πik )πi,k+1 }ri,k+1 ,
i=1
which reduces to
∏
n
r
∏
k
r
i,k+1
πi,k+1 πijij (1 − πij )sij . (9.4)
i=1 j=1
However, πi,k+1 = 1, and so the likelihood function in Equation (9.4)

becomes
∏
n ∏k
r
πijij (1 − πij )sij . (9.5)
i=1 j=1
This is the likelihood function for nk observations rij from a binomial dis-
tribution with response probability πij , and where the binomial denominator
is rij + sij . This denominator is equal to unity when a patient is at risk of
having a detected recurrence after time tj , and zero otherwise. In fact, the
denominator is zero when both rij and sij are equal to zero, and the like-
lihood function in Expression (9.5) is unaffected by observations for which
rij + sij = 0. Data records for which the binomial denominator is zero are
therefore uninformative, and so they can be omitted from the data set. If
there are m observations remaining after these deletions, so that m 6 nk, the
likelihood function in Expression (9.5) is that of m observations from binomial
distributions with parameters 1 and πij , in other words, m observations from
a Bernoulli distribution.
The next step is to note that for the ith patient,
1 − πij = P(Ti > tj | Ti > tj−1 ),
so that
Si (tj )
1 − πij = .
Si (tj−1 )
Adopting a proportional hazards model for the recurrence times, the hazard
of a recurrence being detected at time tj in the ith individual can be expressed
as
hi (tj ) = exp(ηi )h0 (tj ),
where h0 (tj ) is the baseline hazard at tj , and ηi is the risk score for the ith
individual. Notice that this assumption means that the hazards need only
be proportional at the scheduled screening times tj , and not at intermediate
times. This is less restrictive than the usual proportional hazards assumption,
which requires that hazards be proportional at every time.
Using the result in Equation (9.1),
{ }exp(ηi )
S0 (tj )
1 − πij = ,
S0 (tj−1 )
and on taking logarithms we find that
log(1 − πij ) = exp(ηi ) log {S0 (tj )/S0 (tj−1 )} .
Consequently,
log{− log(1 − πij )} = ηi + log [− log {S0 (tj )/S0 (tj−1 )}]
= ηi + γj ,
say. This is a linear model for the complementary log-log transformation of

πij , in which the parameters γj , j = 1, 2, . . . , k, are associated with the k
time intervals. The model can be fitted using standard methods for modelling
binary data.
In modelling the probability of a recurrence in the jth time interval for
the ith patient, πij , the data are the values rij . Data records for which both
rij and sij are equal to zero are omitted, and so the binomial denominator is
unity for each remaining observation. The parameters γj are incorporated in
the model by fitting terms corresponding to a k-level factor associated with the
period of observation, or by including suitable indicator variables as described
in Section 3.2. Note that a constant term is not included in the model. The
estimates of the β-coefficients in ηi , obtained on fitting this model, can again
be interpreted as log-hazard ratios. Also, estimates of the γj can be used to
obtain estimates of the πij . This process is illustrated in Example 9.3 below.

The data on the time to detection of an ulcer recurrence, given in Example
9.1, are now analysed using the method described above. To prepare the data
set for analysis using this approach, the two additional variables, Period and
R, are introduced. The first of these, Period, is used to signify the period,
and the variable is given the value unity for each observation. The second
variable, R, contains the values ri1 , i = 1, 2, . . . , 43, and so R is equal to unity
if an ulcer is detected in the first period and zero otherwise. For these data,
patients 10, 14, 15, 26 and 27 experienced a recurrence in the interval from 0
to 6 months, and so the value of R is unity for these five individuals and zero
for the remaining 38.
We then add a second block of data to this set. This block is a duplication
of the records for the patients who have not had a recurrence at the six-month
mark and for whom the last visit is made after 6 months. There are 38 patients
who have not had a recurrence at six months, but three of these, patients 16,
28 and 34, took no further part in the study. The second block of data therefore
contains 35 records. The variable Period now takes the value 2 for these 35
observations, since they correspond to the second time period. The variable
R contains the values ri2 for this second block of data. Therefore, R takes the
value unity for patients 1, 17, 18, 30, 32 and 37 and zero otherwise, since these
are the only six patients who have a detectable recurrence at 12 months.
The combined set of data has 43 + 35 = 78 rows, and includes the variable
Period, which defines the period in which an endoscopy is performed (1 = 0–6
months, 2 = 6–12 months), and the variable R, which defines the endoscopy
result (0 = negative, 1 = positive). The value of sij is unity for all records
except those for which rij = 1, when it is zero. The binomial denominators
rij +sij are therefore equal to unity for each patient, since every patient in the
extended data set is at risk of a detectable recurrence. Instead of giving a full
listing of the modified data set, the records for the combinations of patient
and period, for the first 18 patients, are shown in Table 9.4.
The dependence of the complementary log-log transformation of the proba-
bilities πij on certain explanatory variables can now be investigated by fitting
models to the binary response variable R. Each model includes a two-level
factor, Period, associated with the period, but no constant term. The term in
the model that corresponds to the period effect is γj , j = 1, 2. The deviances
for the models fitted are summarised in Table 9.5.
From this table we see that the effect of adding either Age or Dur to
the model that contains Period alone is to reduce the deviance by less than
0.3. There is therefore no evidence that the age of a person or the duration
of disease are associated with a recurrence. Adding Treat to the model that
contains Period alone, the reduction in deviance is 0.10 on 1 d.f. This leads us
to conclude that there is no significant difference between the two treatments.
The treatment effect, after adjusting for the variables Age and Dur, is of a
similar magnitude.
To check whether there are interactions between treatment and the two
prognostic factors, we look at the effect of adding the terms Treat × Age and
Treat × Dur to that model that contains Period, Age and Dur. From Table 9.5,
the resulting change in deviance is very small, and so there is no evidence of
any such interactions.
In summary, the modelling process shows that πij , the probability that
the ith patient has a recurrence in the jth period, does not depend on the
patient’s age or the duration of the disease, and, more importantly, does not
depend on the treatment group.
To further quantify the treatment effect, consider the model that includes
both Treat and Period. The equation of the fitted model can be written as
log{− log(1 − π̂ij )} = γ̂j + β̂ Treat i , (9.6)
where γj is the effect of the jth period, j = 1, 2, and Treati is the value of
the indicator variable Treat, for the ith individual. This variable is zero if that
patient is on treatment A and unity otherwise.
Table 9.4 Modified data on the recurrence of an ulcer in two periods, for
the first 18 patients.
Patient Age Duration Treat- Time of Result Period R
ment last visit
1 48 2 B 7 2 1 0
1 48 2 B 7 2 2 1
2 73 1 B 12 1 1 0
2 73 1 B 12 1 2 0
3 54 1 B 12 1 1 0
3 54 1 B 12 1 2 0
4 58 2 B 12 1 1 0
4 58 2 B 12 1 2 0
5 56 1 A 12 1 1 0
5 56 1 A 12 1 2 0
6 49 2 A 12 1 1 0
6 49 2 A 12 1 2 0
7 71 1 B 12 1 1 0
7 71 1 B 12 1 2 0
8 41 1 A 12 1 1 0
8 41 1 A 12 1 2 0
9 23 1 B 12 1 1 0
9 23 1 B 12 1 2 0
10 37 1 B 5 2 1 1
11 38 1 B 12 1 1 0
11 38 1 B 12 1 2 0
12 76 2 B 12 1 1 0
12 76 2 B 12 1 2 0
13 38 2 A 12 1 1 0
13 38 2 A 12 1 2 0
14 27 1 A 6 2 1 1
15 47 1 B 6 2 1 1
16 54 1 A 6 1 1 0
17 38 1 B 10 2 1 0
17 38 1 B 10 2 2 1
18 27 2 B 7 2 1 0
18 27 2 B 7 2 2 1
The estimated coefficient of Treat in this model is 0.195 and the standard
error of this estimate is 0.626. The hazard of a recurrence on treatment B at
any given time, relative to that on treatment A, is exp(0.195) = 1.21. Since
this exceeds unity, there is the suggestion that the risk of recurrence is less
on treatment A than on treatment B, but the evidence for this is not statis-
tically significant. The standard error of the estimated hazard ratio is 0.757.
For comparison, from Example 9.2, the estimated hazard ratio at 12 months
was found to be 1.46, with a standard error of 0.918. These values are very
similar to those obtained in this example. Moreover, the results of analyses
Table 9.5 Deviances on fitting complementary log-log models that

do not include a constant to the variable R.
Terms fitted in model Deviance d.f.
Period 62.982 76
Period + Age 62.685 75
Period + Dur 62.979 75
Period + Age + Dur 62.685 74
Period + Age + Dur + Treat 62.566 73
Period + Age + Dur + Treat + Treat × Age 62.278 72
Period + Age + Dur + Treat + Treat × Dur 62.552 72
Period + Treat 62.884 75
that accommodate interval censoring are comparable to those found in Exam-

ple 9.1, in which the Cox proportional hazards model was used without taking
account of the fact that the data are interval-censored.
The model in Equation (9.6) can be used to provide estimates of the πij .
The estimates of the period effects in this model are γ̂1 = −2.206, γ̂2 = −1.794,
and so the estimated probability of a recurrence in the first period, for a patient
on treatment A, denoted π̂A1 , is given by
log{− log(1 − π̂A1 )} = γ̂1 + β̂ × 0 = −2.206,
and from this, π̂A1 = 0.104. Other fitted probabilities can be calculated in a
similar manner, and the results of these calculations are shown in Table 9.6.
The corresponding observed proportions of individuals with a recurrence for
each combination of treatment and period are also displayed. The agreement
between the observed and fitted probabilities is good, which indicates that the
model is a good fit.
Table 9.6 Fitted and observed probabilities of an ulcer

recurring in the two time periods.
Period Treatment A Treatment B
Fitted Observed Fitted Observed
(0, 6) 0.104 0.158 0.125 0.083
(6, 12) 0.153 0.077 0.183 0.227
If desired, probabilities of a recurrence in either period 1 or period 2 could

also be estimated. The probability that a patient on treatment A has a recur-
rence in either period 1 or period 2 is
P{recurrence in (0, 6)} + P{recurrence in (6, 12) and no recurrence in (0, 6)}.
The joint probability of a recurrence in (6, 12) and no recurrence in (0, 6) can
be expressed as
P{recurrence in (6, 12) | no recurrence in (0, 6)} × P{no recurrence in (0, 6)},
and so the required probability is estimated by
π̂A1 + π̂A2 (1 − π̂A1 ) = 0.104 + 0.153 × 0.896 = 0.241.
Similarly, that for treatment B is
π̂B1 + π̂B2 (1 − π̂B1 ) = 0.125 + 0.183 × 0.875 = 0.285.
This again indicates the superiority of treatment A, but there is insufficient

data for this effect to be declared significant.
9.4 ∗ Arbitrarily interval-censored survival data

The methods of analysis described in the previous sections may be adopted
when different individuals have the same observation times. In this section, we
consider a more general form of interval censoring, where the observation times
differ between individuals. Then, each individual may have a different time
interval in which the event of interest has occurred, and data in this form are
referred to as arbitrarily interval-censored data. A method for analysing such
data, assuming proportional hazards, which is based on a non-linear model
for binary data, proposed by Farrington (1996), is now developed.
9.4.1 Modelling arbitrarily interval-censored data

Suppose that the event time for the ith of n individuals is observed to occur
in the interval (ai , bi ], where the use of different types of bracket indicates
that the actual event time is greater than ai , but less than or equal to bi . In
other words, the event has not occurred by time ai , but has occurred by time
bi , where the values of ai and bi may well be different for each individual in
the study. We will further suppose that the values of a number of explanatory
variables have also been recorded for each individual in the study.
When the values of both ai and bi are observed for an individual, the
interval-censored observation is said to be confined. If the event time for an
individual is left-censored at time bi , so that the event is only known to have
occurred some time before bi , then ai = 0. Similarly, if the event time is right-
censored at time ai , so that the event is only known to have occurred after
time ai , the upper limit of the interval, bi , is effectively infinite.
The survivor function for the ith individual will be denoted by Si (t), so that
the probability of an event occurring in the interval (ai , bi ] is Si (ai ) − Si (bi ).
The likelihood function for the n observations is then
∏
n
{Si (ai ) − Si (bi )} . (9.7)
i=1
Now suppose that the n observations consist of l left-censored observations,

r right-censored observations, and c observations that are confined, so that
ARBITRARILY INTERVAL-CENSORED SURVIVAL DATA 333
n = l + r + c. For the purpose of this exposition, we will assume that the
data have been arranged in such a way that the first l observations are left-
censored (ai = 0), the next r are right-censored (bi = ∞), and the remaining c
observations are confined (0 < ai < bi < ∞). Since Si (0) = 1 and Si (∞) = 0,
the contributions of a left- and right-censored observation to the likelihood
function will be 1 − Si (bi ) and Si (ai ), respectively. Consequently, from Equa-
tion (9.7), the overall likelihood function can be written as
∏
l ∏
l+r ∏
n
{1 − Si (bi )} Si (ai ) {Si (ai ) − Si (bi )},
i=1 i=l+1 i=l+r+1
and a re-expression of the final product in this function gives
∏
l ∏
l+r ∏
n
{1 − Si (bi )} Si (ai ) Si (ai ){1 − Si (bi )/Si (ai )}. (9.8)
i=1 i=l+1 i=l+r+1
We now show that this likelihood is equivalent to that for a corresponding

set of n + c independent binary observations, y1 , y2 , . . . , yn+c , where the ith
is assumed to be an observation from a Bernoulli distribution with response
probability pi , i = 1, 2, . . . , n + c. The likelihood function for this set of binary
data is then
∏ y
n+c
pi i (1 − pi )1−yi , (9.9)
i=1
where yi takes the value 0 or 1, for i = 1, 2, . . . , n + c.

To see the relationship between the response probabilities, pi , in Expres-
sion (9.9), and the values of the survivor function in Expression (9.8), consider
first the left-censored observations. Suppose that each of these l observations
contributes a binary observation with yi = 1 and pi = 1−Si (bi ), i = 1, 2, . . . , l.
The contribution of these l observations to Expression (9.9) is then
∏
l ∏
l
pi = {1 − Si (bi )},
i=1 i=1
which is the first term in Expression (9.8). For a right-censored observation,

we take yi = 0 and pi = 1 − Si (ai ) in Expression (9.9), and the contribution
to the likelihood function in Expression (9.9) from r such observations is
∏
l+r ∏
l+r
(1 − pi ) = Si (ai ),
i=l+1 i=l+1
and this is the second term in Expression (9.8). The situation is a little more
complicated for an observation that is confined to the interval (ai , bi ], since
two binary observations are needed to give the required component of Expres-
sion (9.8). One of these is taken to have yi = 0, pi = 1−Si (ai ), while the other
is such that yc+i = 1, pc+i = 1−{Si (bi )/Si (ai )}, for i = l+r+1, l+r+2, . . . , n.
Combining these two terms leads to a component of the likelihood in Expres-
sion (9.9) of the form
∏
n
(1 − pi )pc+i ,
i=l+r+1
which corresponds to
∏
n
Si (ai ){1 − Si (bi )/Si (ai )}
i=l+r+1
in Expression (9.8).
This shows that by suitably defining a set of n + c binary observations,
with response probabilities expressed in terms of the survivor functions for the
three possible forms of interval-censored observation, the likelihood function
in Expression (9.9) is equivalent to that in Expression (9.8). Accordingly,
maximisation of the log-likelihood function for n + c binary observations is
equivalent to maximising the log-likelihood for the interval-censored data.
9.4.2 Proportional hazards model for the survivor function

The next step in the development of a procedure for modelling arbitrarily
interval-censored survival data is to construct expressions for the survivor
functions that make up the likelihood function in Expression (9.8). A propor-
tional hazards model will be assumed, so that from Equation (9.1),
′
Si (t) = S0 (t)exp(β xi ) , (9.10)
where S0 (t) is the baseline survivor function and xi is the vector of values of
p explanatory variables for the ith individual, i = 1, 2, . . . , n, with coefficients
that make up the vector β.
The baseline survivor function will be modelled as a step-function, where
the steps occur at the k ordered censoring times, t(1) , t(2) , . . . , t(k) , where 0 <
t(1) < t(2) < · · · < t(k) , which are a subset of the times at which observations
are interval-censored. This means that the t(j) , j = 1, 2, . . . , k, are a subset
of the values of ai and bi , i = 1, 2, . . . , n. Exactly how these times are chosen
will be described later in Section 9.4.3.
We now define
S0 (t(j−1) )
θj = log ,
S0 (t(j) )
where t(0) = 0, so that θj > 0, and at times t(j) , we have
S0 (t(j) ) = e−θj S0 (t(j−1) ), (9.11)
for j = 1, 2, . . . , k.
Since the first step in the baseline survivor function occurs at t(1) , S0 (t) = 1
for 0 6 t < t(1) . From time t(1) , the baseline survivor function, using Equa-
tion (9.11), has the value S0 (t(1) ) = exp(−θ1 )S0 (t(0) ), which, since t(0) = 0,
means that S0 (t) = exp(−θ1 ), for t(1) 6 t < t(2) . Similarly, from time t(2) ,
the survivor function is exp(−θ2 )S0 (t(1) ), that is S0 (t) = exp{−(θ1 + θ2 )},
t(2) 6 t < t(3) , and so on, until S0 (t) = exp{−(θ1 + θ2 + · · · + θk )}, t > t(k) .
Consequently,
( j )
∑
S0 (t) = exp − θr , (9.12)
r=1
for t(j) 6 t < t(j+1) , and so the baseline survivor function, at any time ti , is
given by  
∑
k
S0 (ti ) = exp − θj dij  , (9.13)
j=1
where {
1 if t(j) 6 ti ,
dij =
0 if t(j) > ti ,
for j = 1, 2, . . . , k. The quantities dij will be taken to be the values of k
indicator variables, D1 , D2 , . . . , Dk , for the ith observation in the augmented
data set. Note that the values of the Dj , j = 1, 2, . . . , k, will differ at each
observation time, ti .
Combining the results in Equations (9.10) and (9.13), the survivor function
for the ith individual, at times ai , bi , can now be obtained. In particular,
{ ( ∑ )}exp(β ′ xi )
′ k
exp(β xi )
Si (ai ) = S0 (ai ) = exp − θj dij ,
j=1
which can be expressed in the form

{ ∑k }
Si (ai ) = exp − exp(β ′ xi ) θj dij ,
j=1
where dij = 1 if t(j) 6 ai , and dij = 0, otherwise. An expression for Si (bi ) can
be obtained in a similar way, leading to
{ ∑k }
Si (bi ) = exp − exp(β ′ xi ) θj dij ,
j=1
where dij = 1 if t(j) 6 bi , and dij = 0, otherwise.

From these expressions for Si (ai ) and Si (bi ), the response probabilities, pi ,
used in Expression (9.9), can be expressed in terms of the unknown parameters
θ1 , θ2 , . . . , θk , and the unknown coefficients of the p explanatory variables in
the model, β1 , β2 , . . . , βp . As in Section 9.4.1, for a left-censored observation,
pi = 1 − Si (bi ), and for a right-censored observation, pi = 1 − Si (ai ). In
the case of a confined observation, pi = 1 − Si (ai ) for one of the two binary
observations. For the other,
pc+i = 1 − Si (bi )/Si (ai ),

{ ∑k }
exp − exp(β ′ xi ) j=1 θj d1ij
= 1− { ∑k },
exp − exp(β ′ xi ) j=1 θj d2ij
where the values d1ij in the numerator are equal to unity if t(j) 6 bi , and
zero otherwise, and the values d2ij in the denominator are equal to unity if
t(j) 6 ai , and zero otherwise. Consequently, the θ-terms in the numerator for
which t(j) 6 ai cancel with those in the denominator, and this leaves
{ ∑k }
pc+i = 1 − exp − exp(β ′ xi ) θj dij ,
j=1
where here dij = 1 if ai < t(j) 6 bi .

It then follows that in each case, the response probability can be expressed
in the form
{ ∑k }
pi = 1 − exp − exp(β ′ xi ) θj dij , (9.14)
j=1
where {
1 if t(j) is in the interval Ai ,
dij =
0 otherwise,
for j = 1, 2, . . . , k, and the intervals Ai are as shown in Table 9.7.
Table 9.7 Definition of intervals, Ai , used for constructing indicator variables.

Type of observation Value of yi Interval, Ai
Left-censored 1 (0, bi ], i = 1, 2, . . . , l
Right-censored 0 (0, ai ], i = l + 1, l + 2, . . . , l + r
Confined 0 (0, ai ], i = l + r + 1, l + r + 2, . . . , n
1 (ai−c , bi−c ], i = n + 1, n + 2, . . . , n + c
This leads to a non-linear model for a set of binary response variables,

with values yi , and corresponding response probabilities pi , found from Equa-
tion (9.14), for i = 1, 2, . . . , n + c. The model contains k + p unknown pa-
rameters, namely θ1 , θ2 , . . . , θk and β1 , β2 , . . . , βp . This model is known as a
generalised non-linear model, since it is not possible to express a simple func-
tion of pi as a linear combination of the unknown parameters, except in the
case where there are no explanatory variables in the model. The model can
be fitted using computer software for generalised non-linear modelling. Note
that in the fitting process, the θ-parameters should be constrained to be non-
negative.
After fitting a model, the value of the statistic −2 log L̂ can be found, and
this may be used to compare alternative models in the usual manner. The
general procedure is to fit the k terms involving the θ-parameters, and to
then examine the effect of adding and subtracting the explanatory variables
in the data set.
Once an appropriate model has been found, the baseline survivor function
in Equation (9.12) is estimated using
( )
∑
j
Ŝ0 (t) = exp − θ̂r , (9.15)
r=1
for t(j) 6 t < t(j+1) , j = 1, 2, . . . , k, where t(k+1) = ∞, and θ̂j is the estimated
value of θj . The estimated survivor function for the ith individual follows from
′
Ŝi (t) = Ŝ0 (t)exp(β̂ xi )
{ ∑j }
= exp − exp(β̂ ′ xi ) θ̂r , (9.16)
r=1
for i = 1, 2, . . . , n, where β̂ is the vector of estimated coefficients of the ex-

planatory variables. Furthermore, the estimates of the β-parameters are inter-
pretable as log-hazard ratios, in the usual manner, and their standard errors,
produced during the fitting process, can be used to obtain confidence limits.
9.4.3 Choice of the step times

We have seen that the baseline survivor function is assumed to have steps at
times t(j) , j = 1, 2, . . . , k, which are a subset of the observed censoring times,
ai and bi , i = 1, 2, . . . , n. It might be considered to be desirable for the t(j) to
be formed from all distinct censoring times, that is all the unique values of ai
and bi . However, this will generally lead to the introduction of far too many
θ-parameters in the non-linear model. Instead, a subset of the available times
is chosen.
Each interval used in the binary data model, and denoted by Ai in the
preceding section, must include at least one of the times t(j) . If this is the
case, at least one of the values of dij in Equation (9.14) will be equal to unity
∑k
and hence the term j=1 θj dij will be greater than zero. Suppose that the
interval Ai is (ui , vi ]. This requirement is then achieved by taking t(1) to be the
smallest of the values of vi , t(2) to be the smallest vi such that ui > t(1) , t(3)
to be the smallest vi such that ui > t(2) , and so on, until t(k) is the smallest
value of vi such that ui > t(k−1) .
Once this subset of k times has been identified, the model can be fitted.
Models containing explanatory variables are fitted, and for each model the
estimates of the k θ-parameters and the relevant β-parameters are found. The
fitting process will lead to a value of the −2 log L̂ statistic, and these values can
be compared for models with the same number of θ-parameters, but different
explanatory variables, in the usual way.
Sometimes, it may be desirable to increase the number of steps in the es-
timated baseline hazard function, by the addition of some of the remaining
censoring times. This entails adding a new θ-parameter for each additional
time point. One way of doing this is to fit the minimal set of censoring times,
and to then add each additional time point in turn, for the full set of ex-
planatory variables. The time that leads to the biggest reduction in the value
of −2 log L̂ is then added to the minimal set. All remaining times are then
added one by one, and again that which reduces −2 log L̂ the most is added
to the set. This process may be continued until the reduction on adding an
additional θ-parameter ceases to be significant at some chosen level, and so
long as all the estimated θ-parameters remain positive.
It is important to note that this modelling procedure is only valid if the
set of possible censoring times is finite, that is, it does not increase with the
number of observations. Otherwise, the number of θ’s increases indefinitely
and asymptotic results used in comparing models will no longer be valid.
This procedure for modelling arbitrarily interval-censored data is now il-
lustrated.
Example 9.4 Occurrence of breast retraction

In the treatment of early breast cancer, a tumourectomy, followed by radiation
therapy, may be used as an alternative to mastectomy. Chemotherapy may
also be used in conjunction with the radiotherapy in order to enhance its effect,
but there is evidence that this adjuvant chemotherapy increases the effect of
the radiation on normal tissue. This in turn leads to breast retraction, which
has a negative impact on the appearance of the breast. In a retrospective study
to assess the effect of this type of treatment on breast cosmesis, 46 women
who had been treated with radiotherapy alone were compared with 48 who
had received a combination of radiotherapy and chemotherapy. Patients were
observed every 4 to 6 months, but less frequently as their recovery progressed.
On these occasions, the cosmetic effect of the treatment was monitored, with
the extent of breast retraction being measured on a four-point scale: none,
minimal, moderate, severe. The event of interest was the time in months to the
first appearance of moderate or severe retraction. The exact time of occurrence
of breast retraction will be unknown, and the only information available will
concern whether or not retraction is identified when a patient visits the clinic.
Moreover, since the visit times were not the same for each patient, and a
number of patients failed to keep appointments, the data are regarded as
arbitrarily interval-censored. The data obtained in this study were included
in Finkelstein and Wolfe (1985), and are given in Table 9.8.
In this data set, there are five patients for whom breast retraction had
occurred before the first visit. For each of these patients, the start of the
interval is set to zero, that is ai = 0, and the observed times are left-censored,
so that l = 5. There are 38 patients who had not experienced retraction by
Table 9.8 Data on the time in months to breast retraction in patients with
breast cancer.
Radiotherapy Radiotherapy and Chemotherapy
(45, ∗ ] (25, 37] (37, ∗ ] (8, 12] (0, 5] (30, 34]
(6, 10] (46, ∗ ] (0, 5] (0, 22] (5, 8] (13, ∗ ]
(0, 7] (26, 40] (18, ∗ ] (24, 31] (12, 20] (10, 17]
(46, ∗ ] (46, ∗ ] (24, ∗ ] (17, 27] (11, ∗ ] (8, 21]
(46, ∗ ] (27, 34] (36, ∗ ] (17, 23] (33, 40] (4, 9]
(7, 16] (36, 44] (5, 11] (24, 30] (31, ∗ ] (11, ∗ ]
(17, ∗ ] (46, ∗ ] (19, 35] (16, 24] (13, 39] (14, 19]
(7, 14] (36, 48] (17, 25] (13, ∗ ] (19, 32] (4, 8]
(37, 44] (37, ∗ ] (24, ∗ ] (11, 13] (34, ∗ ] (34, ∗ ]
(0, 8] (40, ∗ ] (32, ∗ ] (16, 20] (13, ∗ ] (30, 36]
(4, 11] (17, 25] (33, ∗ ] (18, 25] (16, 24] (18, 24]
(15, ∗ ] (46, ∗ ] (19, 26] (17, 26] (35, ∗ ] (16, 60]
(11, 15] (11, 18] (37, ∗ ] (32, ∗ ] (15, 22] (35, 39]
(22, ∗ ] (38, ∗ ] (34, ∗ ] (23, ∗ ] (11, 17] (21, ∗ ]
(46, ∗ ] (5, 12] (36, ∗ ] (44, 48] (22, 32] (11, 20]
(46, ∗ ] (14, 17] (10, 35] (48, ∗ ]
their final visit. For these patients, the upper limit of the time interval is shown
as an asterisk (∗) in Table 9.8. The observations for these patients are therefore
right-censored and so r = 38. The remaining c = 51 patients experience breast
retraction within confined time intervals, and the total number of observations
is n = l + c + r = 94.
The first step in fitting a model to these arbitrarily interval-censored data
is to expand the data set by adding a further 51 lines of data, repeating that
for the patients whose intervals are confined, so that the revised database has
n + c = 145 observations. The values, yi , of the binary response variable, Y ,
are then added. These are such that Y = 1 for a left-censored observation, and
Y = 0 for a right-censored observation. For confined observations, where the
data are duplicated, one of the pairs of observations has Y = 0 and the other
Y = 1. The treatment effect will be represented by the value of a variable
labelled Treat, which will be zero for a patient on radiotherapy and unity for
a patient on radiotherapy and chemotherapy. For illustration, the values of
the binary response variable, Y , are shown for the first three patient treated
with radiotherapy alone, in Table 9.9.
Table 9.9 Augmented data set for the first

three patients on radiotherapy.
Patient A B U V Treat Y
1 45 ∗ 0 45 0 0
2 6 10 0 6 0 0
2 6 10 6 10 0 1
3 0 7 0 7 0 1
In this table, the variables A and B refer to the times of the start and
end of each interval, and so their values are ai , bi and the variables U and V
contain the values of ui , vi that form the limits of the intervals, Ai , in the
binary data model.
We now determine the time points, t(j) , that are to be used in calculating
the baseline survivor function, using the procedure described in Section 9.4.3.
The first of these is the smallest of the values vi , which form the variable V in
the data set. This is found for the observation (0, 4], for which V = 4, so that
t(1) = 4. The smallest value of V for which U > 4 occurs for the observation
(4, 8], and so t(2) = 8. Next, the smallest value of V with U > 8 occurs for the
observation (8, 12], giving t(3) = 12. There are six other times found in this
way, namely 17, 23, 30, 34, 39 and 48, and so the minimal subset of times has
k = 9 in this example.
Variables with values dij , that correspond to each of the k censoring times,
t(j) , are now added to the database that has 145 records. Since k = 9 in this
case, nine variables, D1 , D2 , . . . , D9 , are introduced, where the values of Dj ,
j = 1, 2, . . . , 9, are dij , for i = 1, 2, . . . , 145. These values are such that dij = 1
if ui < t(j) 6 vi , and zero otherwise, and so they can straightforwardly be ob-
tained from the variables U and V in Table 9.9. The values of D1 , D2 , . . . , D9 ,
for the three patients included in Table 9.9, are shown in Table 9.10.
Table 9.10 Database for binary data analysis, for the first three patients on
radiotherapy.
Patient Treat Y D1 D2 D3 D4 D5 D6 D7 D8 D9
1 0 0 1 1 1 1 1 1 1 1 0
2 0 0 1 0 0 0 0 0 0 0 0
2 0 1 0 1 0 0 0 0 0 0 0
3 0 1 1 0 0 0 0 0 0 0 0
We now have the database to which a non-linear model for the binary
response data in Y is fitted. The model is such that we take Y to have a
Bernoulli distribution with response probability as given in Equation (9.14),
that is a binomial distribution with parameters 1, pi . Here, the SAS proce-
dure proc nlmixed has been used to fit the non-linear model to the binary
data.
On fitting the null model, that is the model that contains all 9 D-variables,
but not the treatment effect, the value of the statistic −2 log L̂ is 285.417. On
adding Treat to the model, the value of −2 log L̂ is reduced to 276.983. This
reduction of 8.43 on 1 d.f. is significant at the 1% level (P = 0.0037), and
so we conclude that the interval-censored data do provide strong evidence
of a treatment effect. The estimated coefficient of Treat is 0.8212, with a
standard error of 0.2881. The corresponding hazard ratio for a patient on
the combination of radiotherapy and chemotherapy, relative to a patient on
radiotherapy alone, is exp(0.8212) = 2.27. The interpretation of this is that
patients on the combined treatment have just over twice the risk of breast
retraction, compared to patients on radiotherapy. A 95% confidence interval
for the corresponding true hazard ratio has limits exp(0.8212 ± 1.96 × 0.2881),
which leads to the interval (1.29, 4.00).
The minimal subset of times at which the estimated baseline survivor
function is estimated can be enlarged by adding additional censoring times
from the data set. However, there are no additional times that lead to a
significant reduction in the value of the −2 log L̂ statistic, with the estimated
θ-parameters remaining positive.
The estimated values of the coefficients of the D-variables are the values θ̂j ,
for j = 1, 2, . . . , 9, and these can be used to provide an estimate of the survivor
function for the two treatment groups. Equation (9.15) gives the form of the
estimated baseline survivor function, which is the estimated survivor function
for the patients on radiotherapy alone. The corresponding estimate for the pa-
tients who receive adjuvant chemotherapy is obtained using Equation (9.16),
and is just {Ŝ0 (t(j) )}exp(β̂) , with β̂ = 0.8212. On fitting the model that con-
tains the treatment effect and the 9 D-variables, the estimated values of θj ,
for j = 1, 2, . . . , 9, are 0.0223, 0.0603, 0.0524, 0.0989, 0.1620, 0.0743, 0.1098,
0.2633 and 0.4713, respectively. From these values, the baseline survivor func-
tion, at the times t(j) , j = 1, 2, . . . , 9, can be estimated, and this estimate
is given as Ŝ0 (t(j) ) in Table 9.11. Also given in this table is the estimated
survivor function for patients on the combined treatment, denoted Ŝ1 (t(j) ).
Table 9.11 Estimated survivor functions

for a patient on radiotherapy alone, Ŝ0 (t),
and adjuvant chemotherapy, Ŝ1 (t).
Time interval Ŝ0 (t) Ŝ1 (t)
0– 1.000 1.000
4– 0.978 0.951
8– 0.921 0.829
12– 0.874 0.736
17– 0.791 0.588
23– 0.673 0.407
30– 0.625 0.343
34– 0.560 0.268
39– 0.430 0.147
48– 0.269 0.050
The survivor functions for the two groups of patients are shown in Figure 9.1.
From the estimated survivor functions, the median time to breast retraction
for patients on radiotherapy is estimated to be 39 months, while that for
patients who received adjuvant chemotherapy is 23 months. More precise esti-
mates of these median times could be obtained if a greater number of censoring
times were used in the analysis.
1.0
0.8
0.6
0.4
0.2
0.0
0 10 20 30 40 50
Time to retraction
Figure 9.1 Estimated survivor functions for a patient on radiotherapy (—) and the
combination of radiotherapy and chemotherapy (·······).
9.5 Parametric models for interval-censored data

The methods for modelling interval-censored data that have been described
in previous sections are based on the Cox proportional hazards model. In fact,
it is much more straightforward to model such data assuming a parametric
model for the survival times. In Section 9.4, the likelihood function for n
arbitrarily interval-censored observations consisting of l that are left-censored
at time ai , r that are right-censored at bi , and c that are confined to the
interval (ai , bi ), was given as
∏
l ∏
l+r ∏
n
{1 − Si (bi )} Si (ai ) {Si (ai ) − Si (bi )}. (9.17)
i=1 i=l+1 i=l+r+1
If a parametric model for the survival times is assumed, then Si (t) has a fully
parametric form. For example, if the survival times have a Weibull distribu-
tion with scale parameter λ and shape parameter γ, from Equation (5.37) in
Section 5.6 of Chapter 5, the survivor function for the ith individual is
{ }
Si (t) = exp − exp(β ′ xi )λtγ ,
where xi is the vector of values of explanatory variables for that individual,

with coefficients β. Alternatively, the accelerated failure time form of the
survivor function, given in Equation (6.16) of Section 6.5 of Chapter 6, may
be used. The Weibull survivor function leads to expressions for the survivor
functions in Expression (9.17), for any values of ai and bi . The corresponding
DISCUSSION 343
log-likelihood function can then be maximised with respect to the parameters
λ, γ and the β’s. No new principles are involved. The same procedure can be
adopted for any other parametric model described in Chapter 6.
In some situations, the database may be a combination of censored and un-
censored observations. The likelihood function in Expression (9.17) may then
be further extended to allow for the uncensored∏observations. This is achieved
by including an additional factor of the form i f (ti ), which is the product
of the density functions at the event times, over the uncensored observations.
A number of software packages include facilities for the parametric modelling
of interval-censored data.
Example 9.5 Occurrence of breast retraction

The interval-censored data on the times to breast retraction in women being
treated for breast cancer, given in Example 9.4, are now used to illustrate
parametric modelling for interval-censored data.
If a common Weibull distribution is assumed for the data in both treatment
groups, the value of −2 log L̂ is 297.585. If the treatment effect is added to
this model, and the assumption of proportional hazards is made, so that the
shape parameter is constant, the value of −2 log L̂ reduces to 286.642. This
reduction of 10.943 on 1 d.f. is highly significant (P = 0.001). Comparing this
with the results of the analysis in Example 9.4, we find that the significance
of the treatment effect is a little greater under the assumed Weibull model.
Furthermore, the estimated hazard ratio for a patient on radiotherapy with
adjuvant chemotherapy, relative to a patient on radiotherapy alone, is 2.50,
and a corresponding 95% confidence interval is (1.44, 4.35). These values do
not differ much from those found in Example 9.4.
9.6 Discussion
In many situations, the method for analysing interval-censored survival data

that has been presented in Section 9.4, or the fully parametric approach of
Section 9.5, will be the most appropriate. Even when studies are designed to
be such that the examination times are the same for each patient in the study,
missed, postponed or cancelled appointments may lead to observation times
that do differ across the patients. In this case, and for studies where this is a
natural feature, methods for handling arbitrarily interval-censored data will
be required.
When the observation times are the same for each patient, the method
for analysing interval-censored data that has been presented in Section 9.3
will generally be the most suitable. However, this approach is not optimal,
since recurrences detected between scheduled examinations, that is, interval-
detected recurrences, are only counted at the next examination time. If the
intervals between successive examination times are not too large, the difference
between the results of an analysis based on the model in Section 9.3, and one
that uses the actual times of interval-detected recurrences, will be negligible.
In fact, if the number of intervals is not too small, and the time between
successive examinations not too large, the results will not be too different
from an analysis that assumes the recurrence times to be continuous, outlined
in Section 9.1.
As mentioned earlier, the model described in Section 9.3 only requires
hazards to be proportional at scheduled screening times. This means that the
model is useful when the hazards are not necessarily proportional between
screening times. Furthermore, the model could be relevant in situations where
although actual survival times are available, the hazards can only be taken to
be proportional at specific times. On the other hand, the method for analysing
arbitrarily interval-censored data in Section 9.4, requires hazards to be propor-
tional at each of the times used in constructing the baseline survivor function,
which is more restrictive. Further comments on methods for analysing survival
data where hazards are non-proportional are included in Chapter 11.
9.7 Further reading

Much of Sections 9.1 to 9.3 of this chapter are based on the summary of
methods for processing interval-censored survival data given by Whitehead
(1989). The approach described in Section 9.3 is based on Prentice and Gloeck-
ler (1978). A method for fitting the proportional hazards model to interval-
censored data was proposed by Finkelstein (1986), but the method for mod-
elling arbitrarily interval-censored data in Section 9.4 is due to Farrington
(1996), who also develops additive and multiplicative models for such data.
A number of approaches to the analysis of interval-censored data involve
the analysis of binary observations. Collett (2003) describes a model-based
approach to the analysis of binary data. Other books that include material
on the analysis of binary data include Hosmer and Lemeshow (2000), Dobson
(2001) and Morgan (1992).
The use of the complementary log-log transformation in the analysis of
interval-censored data was described by Thompson (1981). Becker and Melbye
(1991) show how a log-linear model can be used to obtain an estimate of the
survivor function from interval-censored data, assuming a constant hazard in
each interval.
There are a number of other approaches to the analysis of interval-censored
data, some of which are described in the tutorial provided by Lindsey and
Ryan (1998). Lindsey (1998) reviews the use of parametric models for the
analysis of interval-censored data. Pan (2000) suggests using multiple impu-
tation, based on the approximate Bayesian bootstrap, to impute values for
censored observations. Farrington (2000) provides a comprehensive account
of diagnostic methods for use with proportional hazards models for interval-
censored data.
Chapter 10
Frailty models
In modelling survival data, we attempt to explain observed variation in times

to events such as death by differences in the values of certain explanatory
variables that have been recorded for each individual in a study. However,
even after allowing for the effects of such variables, there will still be variation
in their observed survival times. Some individuals may have a greater hazard
of death than others, and are therefore likely to die sooner. They may be
described as being more frail, and variation in survival times can be explained
by variability in this frailty effect across individuals.
Situations where the survival times amongst groups of individuals in a
study are not independent are frequently encountered. The survival times of
individuals within the same group will then tend to be more similar than
they would be for individuals from different groups. This effect can then be
modelled by assuming that the individuals within a group share the same
frailty. Since the frailty is common to all individuals within a group, a degree
of dependence between the survival times of individuals within the group is
introduced. Consequently, models with shared frailty can be used in modelling
survival data where some association between the event times is anticipated,
as in studies involving groups of individuals that share a common feature, or
when an individual experiences repeated times to an event. Such models are
described in this chapter.
10.1 Introduction to frailty

Variation in survival times is almost always found amongst individuals in a
study. Even when a number of individuals share the same values of demo-
graphic variables such as age and gender, or the same values of other explana-
tory variables that have been measured, their survival times will generally be
different. These differences may be explained by the fact that certain variables
that influence survival were not measured, or that an individual’s survival time
depends on variables that it was not possible to measure. There may be many
such variables and our knowledge of these may be limited to the extent that we
simply do not know what the variables are that might explain this variability.
To illustrate, consider the survival times of patients who have been diag-
nosed with a life-threatening cancer. Patient survival time following diagnosis
345
346 FRAILTY MODELS
will depend on many factors, including the type of cancer and stage of the tu-
mour, characteristics of the patient such as their age, weight and lifestyle, and
the manner in which the cancer is treated. A group of patients who have the
same values of each measured explanatory variable may nevertheless be ob-
served to have different survival times. This variation or heterogeneity between
individuals may arise because some individuals are not as strong as others in
ways that cannot be summarised in terms of a relatively small number of
known variables. Indeed, we can never aspire to know what all the factors are
that may have an effect on the survival of cancer patients, let alone be able
to measure them, but we can take account of them in a modelling process.
Variation between the survival times of a group of individuals can be de-
scribed in terms of some individuals being more frail than others. Those who
have higher values of a frailty term tend to die sooner than those who are
less frail. However, the extent of an individual’s frailty cannot be measured
directly; if it could, we might attempt to include it in a model for survival
times. Instead, we only observe the impact that the frailty effects have on the
observable survival times.
10.1.1 Random effects

In the models described in earlier chapters, the effects corresponding to factors
of interest have always been assumed to be fixed. The possible values of a
fixed effect do not vary, and they are assumed to be measured without error.
For example, the factor associated with gender has two distinct levels, male
and female, that will not change from study to study, and it will often be of
interest to summarise outcome differences between the two genders. On the
other hand, a random effect is assumed to have levels drawn from a population
of possible values, where the actual levels are representative of that population.
As an example, in a multicentre clinical trial, the centres adopted might be
considered to be drawn from a larger number of possible centres. We would
then model centre variation using a random effect, and there would be little
interest in comparing outcomes between particular centres.
Random effects are assumed to be observations on a random variable that
has some underlying probability distribution, with the variance of this dis-
tribution used to summarise the extent of differences in their values. Such
effects typically have many possible values and it is unrealistic to represent
differences between them using fixed effects, as this would introduce a large
number of unknown parameters into a model. By taking them to be random
effects, there is just one parameter to estimate, namely the variance of their
assumed underlying distribution.
10.1.2 Individual frailty

Since the unknown frailty values attributed to individuals in a study are es-
sentially drawn from a large number of possible values, we take frailty to be a
INTRODUCTION TO FRAILTY 347
random effect. In situations where the variance of the underlying distribution
of random effects is small, the frailty effects will not differ much and the effect
on inferences will be negligible. However, if the frailty variance is large, the
observed survival times of individuals may differ markedly, even after allowing
for the effects of other potential explanatory variables.
In the presence of frailty, individuals who are more frail will tend to die
earlier. This means that at any point in time when the survival times of a
sample of individuals are observed, those who are still alive will be less frail
than the corresponding population from which the sample has been drawn.
Since frailty is acting as a selection effect, there are consequences for what can
actually be observed. For example, if each individual has a constant hazard of
death, and frailty effects are present, the more frail individuals will die earlier,
while those who are less frail will survive longer. The observed hazard function
may therefore appear to decline over time. As an illustration of this, consider a
group of individuals who have experienced a non-fatal heart attack. For such
individuals, the hazard of death is generally observed to decline with time,
and there are two possible reasons for this. First, the individuals may simply
adjust to any damage to the heart that has been caused by the heart attack, so
that the underlying hazard of death does decline. Alternatively, the hazard of
death may be constant, but the observed decrease in the hazard may be due to
frailty; higher risk individuals die earlier, so that at any time, the individuals
who remain alive are those that are less frail. Data on the survival times
cannot be used to distinguish between these two possible explanations for the
apparent decline in the hazard of death. By the same token, when allowance is
made for frailty effects, estimated hazard ratios and their associated standard
errors may be quite different from the values obtained when frailty effects are
ignored. This feature will be illustrated in Section 10.3.
10.1.3 Shared frailty

Models that take account of individual levels of frailty can be used to assess
how the values of unmeasured variables may affect survival. However, the no-
tion of frailty is more widely used in situations where a number of individuals
share something in common. In a multicentre study, for example, the survival
experience of individuals from the same centre may be more similar than that
for individuals from different centres. This could be because of different clin-
ical teams in the different centres, or different nursing practices across the
centres. Similarly, in an animal experiment, animals from the same litter will
be more alike than animals from different litters, because of genetic and en-
vironmental influences. Their survival times will therefore be correlated, and
one way of modelling this association between survival times is to assume
that all the individuals within a group share the same frailty. In paired or-
gan studies involving eyes, ears or kidneys, where the time to some event is
recorded on each organ within the pair, frailty might be included in a model
to allow for any association between the pairs of observed event times. Shared
348 FRAILTY MODELS
frailty models therefore provide a method for modelling survival data when
the survival times are not independent.
Some studies lead to repeated or recurrent event times within an individ-
ual. Examples of this type of situation include studies on the times between
successive adverse events, such as migraine or nausea, the times to the failure
of tooth fillings in an individual, and times to the failure of transplanted kid-
neys in patients that receive more than one transplant. Here again, the event
times within an individual may not be independent, and to model this we can
assume that the frailty is the same for each of these event times. The concept
of a shared frailty can therefore be extended to situations where there are
repeated events within an individual.
10.2 Modelling individual frailty

Variation between the survival times of individuals can be modelled by sup-
posing that each person has his or her own value of a frailty. This frailty
term is assumed to act multiplicatively on the hazard of death of an individ-
ual. Those with a frailty greater than unity will have increased hazard, while
those who have a frailty less than unity will have a reduced hazard of the
event occurring.
Consider a proportional hazards model for the hazard of an event occurring
at some time t, in an individual with a hazard function that depends on
the values of p explanatory variables X1 , X2 , . . . , Xp , and on some unknown
baseline hazard function h0 (t). The hazard function for the ith of n individuals
is then
hi (t) = exp(β ′ xi )h0 (t), (10.1)
where xi is the vector of values of the explanatory variables for the ith individ-
ual, and β is the vector of their unknown coefficients β1 , β2 , . . . , βp . The base-
line hazard h0 (t) may be unspecified as in the Cox regression model (Chapter
3) or a parametric function of t, such as a Weibull model (Chapters 5, 6).
We now introduce a multiplicative frailty term, zi , and write
hi (t) = zi exp(β ′ xi )h0 (t). (10.2)
The frailty term zi cannot be negative, and the greater the frailty, the greater
is the hazard of the event occurring. A value of zi = 1 brings us back to the
standard model of Equation (10.1), corresponding to the situation where there
are no frailty effects. Values of zi between 0 and 1 correspond to situations
where the hazard is less than that in the underlying proportional hazards
model, and occur when individuals have increased fortitude. Differences in
the values of the zi across the sample of n individuals will then represent
variation in the hazard of death at any given time due to frailty effects.
It is generally more convenient to work with an alternative representation
of the frailty effect, obtained by setting zi = exp(ui ). The model in Equa-
tion (10.2) can then be written
hi (t) = exp(β ′ xi + ui )h0 (t). (10.3)
MODELLING INDIVIDUAL FRAILTY 349
In this model, ui = log zi is a random effect in the linear component of the
proportional hazards model. Note that whereas a frailty zi cannot be negative,
ui can take any value, positive or negative, and ui = 0 corresponds to the case
where zi = 1 and there is no frailty. In Equation (10.3), the linear component
of the proportional hazards model has been extended to include a random
effect, and this part of the model for survival times is now akin to mixed
models used in other areas of general and generalised linear modelling.
The ui in Equation (10.3) are regarded as the observed values, or real-
isations, of n independent and identically distributed random variables Ui ,
i = 1, 2, . . . , n, where each random variable, Ui , is assumed to have a common
probability distribution. So although the individual frailties across a group
of individuals may all be different, they are drawn from the same underlying
probability distribution.

The data in Example 1.2 of Chapter 1 on the survival times of two groups of
women, with tumours that were negatively or positively stained, shows that
there is variability in the survival times of the women within each group. In
the negatively stained group, actual survival times range from 23 to beyond
224 months, while in the positively stained group, they range from 5 to over
225 months. To summarise these data, a Weibull proportional hazards model
containing a factor associated with positive or negative staining is fitted, as in
Example 5.6. The estimated baseline hazard function is ĥ0 (t) = λ̂γ̂tγ̂−1 , where
λ̂ = 0.00414 and γ̂ = 0.937. The corresponding estimated survivor function
for the ith woman, i = 1, 2, . . . , 45, is given by
Ŝi (t) = exp{− exp(β̂xi )λ̂tγ̂ },
using Equation (5.37) of Chapter 5, where xi = 0 for a woman with negative

staining and xi = 1 for one with positive staining, and β̂ = 0.934. The corre-
sponding hazard ratio for a woman with a positively stained tumour, relative
to one whose tumour is negatively stained, is 2.545, with a 95% confidence
interval of (0.93, 6.98). The fitted Weibull survivor functions are shown in
Figure 10.1, superimposed on the Kaplan-Meier estimate of the survivor func-
tions. The Weibull model does not appear to be a very good fit to the survival
times of women with positively stained tumours.
In this figure, the data have been summarised in just two curves. However,
the introduction of frailty would lead to separate estimated survivor functions
for each woman, and we will return to this in Example 10.2.
10.2.1 ∗ Frailty distributions

In the model of Equation (10.3), it is common to assume that the random
variables Ui , i = 1, 2, . . . , n, that are associated with the random effects,
have a normal distribution with zero mean and common variance σu2 . If
350 FRAILTY MODELS
1.0
0.8
0.6
0.4
0.2
0.0
0 50 100 150 200 250
Survival time
Figure 10.1 Observed and fitted Weibull survivor functions for women with positive
(·······) and negative (—) staining.
Ui ∼ N (0, σu2 ), then Zi = exp(Ui ) has a lognormal distribution. This dis-

tribution was introduced in Section 6.1.2 of Chapter 6. Since zi = 1 in the
model of Equation (10.2) corresponds to the situation where there is no frailty,
it is desirable for the frailty distribution to be centred on unity. By taking the
random frailty effects, Ui , to be distributed about a mean of zero, the cor-
responding distribution of Zi has a median of unity. However, the expected
value of Zi , which is
E (Zi ) = exp(σu2 /2), (10.4)
will be different from 1.0, and the variance of Zi is given by
var (Zi ) = exp(σu2 ){exp(σu2 ) − 1}. (10.5)
The results in Equations (10.4) and (10.5) show how the mean and variance
of a lognormal frailty distribution can be found from the variance, σu2 , of the
corresponding normally distributed random effect. An alternative approach is
to take Ui to have a normal distribution with mean −σu2 /2 and variance σu2 .
Although Zi would then have a mean of unity, this formulation is not widely
used in practice.
A distribution that is very similar to the lognormal distribution is the
gamma distribution, described in Section 6.1.3, and so this is an alternative
distribution for the frailty random variable, Zi . This frailty distribution is
sometimes more convenient to work with than the lognormal model, and can
be used to investigate some of the consequences of introducing a frailty effect,
which we do in Section 10.3.
MODELLING INDIVIDUAL FRAILTY 351
If we take Zi to have a gamma distribution with both unknown parameters
equal to θ, and density
θθ ziθ−1 e−θzi
f (zi ) = , θ > 0,
Γ(θ)
for zi > 0, so that Zi ∼ Γ(θ, θ), this frailty distribution has a mean of unity
and a variance of 1/θ. The larger the value of θ, the smaller the frailty variance,
and as θ → ∞, the frailty variance tends to zero, corresponding to the case
where the zi are all equal to unity and there is no frailty. The corresponding
distribution of Ui = exp(Zi ) has density
θθ eθui exp(−θeui )
f (ui ) = , θ > 0, (10.6)
Γ(θ)
for −∞ < ui < ∞, and Ui is said to have an exp-gamma distribution. This
distribution is also referred to as the log-gamma distribution, but this nomen-
clature is inconsistent with the definition of a lognormal distribution. The
modal value of Ui is zero, but the distribution is asymmetric and the mean
and variance of Ui are expressed in terms of digamma and trigamma functions.
Specifically,
E (Ui ) = Ψ(θ) − log θ, (10.7)
where the digamma function, Ψ(θ), can be obtained from the series expansion
∞
∑ θ−1
Ψ(θ) = −λ + ,
j=0
(1 + j)(θ + j)
and here, λ = 0.577216 is Euler’s constant. Also, the variance of Ui is the

derivative of Ψ(θ), written Ψ′ (θ), and known as the trigamma function. This
function can also be obtained using a series expansion, and
∞
∑ 1
var (Ui ) = Ψ′ (θ) = . (10.8)
j=0
(θ + j)2
The digamma and trigamma functions are available as standard functions in

many statistical software packages.
The lognormal and gamma distributions for frailty, and the corresponding
normal and exp-gamma distributions for the random effects, will be compared
later in Example 10.3.
10.2.2 ∗ Observable survivor and hazard functions

The survivor function for the ith individual, from Equations (1.6) and (1.7)
in Chapter 1, is { ∫ t }
Si (t) = exp − hi (t) dt ,
0
352 FRAILTY MODELS
where hi (t) is given in Equation (10.2). Therefore
′
Si (t) = exp{−zi eβ xi H0 (t)}, (10.9)
where H0 (t) is the baseline cumulative hazard function. This is the survivor
function for the ith individual conditional on the frailty zi , and is termed a
conditional model for the survival times.
The individual frailties zi cannot be observed directly, and so what we
observe is the effect that they have on the overall survivor function for the
group of individuals. The observable survivor function is therefore the individ-
ual functions averaged over all possible values of zi , that is the expected value
of Si (t) with respect to the frailty distribution. If the corresponding random
variables Zi had a discrete distribution, where only a few specific values of
the frailty were possible, we would obtain this expectation by summing the
products of the possible values of Si (t) for the different zi values and the
probability that Zi was equal to zi . However, we generally assume a continu-
ous distribution for the Zi , and in this situation, the survivor function that is
observed is found by integrating Si (t) in Equation (10.9) with respect to the
distribution of Zi . The resulting survivor function is
∫ ∞ ∫ ∞
′
∗
Si (t) = Si (t)f (zi ) dzi = exp{−zi eβ xi H0 (t)}f (zi ) dzi , (10.10)
0 0
where f (zi ) is the density function of the frailty distribution. The quan-
tity Si∗ (t) is the unconditional or observable survivor function, and Equa-
tion (10.10) defines an unconditional model for the survival times.
Once Si∗ (t) has been obtained for a particular model, the corresponding
observable hazard function, h∗i (t), can be found using the relationship in Equa-
tion (1.5) of Chapter 1, so that
d
h∗ (t) = − {log S ∗ (t)}. (10.11)
dt
In general the integral in Equation (10.10) has to be evaluated numerically, but
as we shall see in the following section, this can be accomplished analytically
in the special case of gamma frailty effects.
10.3 The gamma frailty distribution

The gamma distribution is often used to model frailty effects as it leads to a
closed form representation of the observable survivor and hazard functions, as
shown in this section.
Suppose that each individual in a study has a distinct frailty value, zi , and
that these are observations on independent and identically distributed random
variables Zi , i = 1, 2, . . . , n, where Zi ∼ Γ(θ, θ). Now, from Equation (10.10),
the observable survivor function is
∫ ∞
Si∗ (t) = Si (t)f (zi ) dzi ,
0
THE GAMMA FRAILTY DISTRIBUTION 353
so that ∫ ∞ ′ θθ ziθ−1 e−θzi
Si∗ (t) = exp{−zi eβ xi H0 (t)} dzi ,
0 Γ(θ)
and on some rearrangement, this becomes
′ ∫ ∞
θθ [θ + eβ xi H0 (t)]−θ
yiθ−1 e−yi dyi ,
Γ(θ) 0
′
where yi = {θ+eβ xi H0 (t)}zi . Then, from the definition of a gamma function,
∫ ∞
yiθ−1 e−yi dyi = Γ(θ),
0
and so ′
Si∗ (t) = {1 + θ−1 eβ xi H0 (t)}−θ . (10.12)
Equation (10.11) leads to the observable hazard function, given by
′
eβ xi h0 (t)
h∗i (t) = ′ . (10.13)
1 + θ−1 eβ xi H0 (t)
In the next section, some of the consequences of introducing a frailty effect

are illustrated.
10.3.1 Impact of frailty on an observable hazard function

To illustrate the effect that a frailty term has on the hazard function, we will
take the frailty random variable Zi to have a Γ(θ, θ) distribution. Suppose that
the baseline hazard function h0 (t) is a constant, λ, and that we are fitting a
model to a single sample of survival times so that there are no covariates.
Using Equation (10.13), the observable hazard function is then
θλ
h∗ (t) = ,
θ + λt
which declines non-linearly from the value λ when t = 0 to zero. The presence
of frailty effects therefore provides an alternative explanation for a decreasing
hazard function. Specifically, the overall hazard of an event might actually be
constant, but heterogeneity between the survival times of the individuals in
the study will mean that a decline in hazard is observed.
Similar features are found if we assume that the underlying baseline hazard
is dependent upon time. In the case where the underlying baseline hazard is
Weibull, with h0 (t) = λγtγ−1 , and where there are no covariates,
θλγtγ−1
h∗ (t) = .
θ + λtγ
354 FRAILTY MODELS
6 =
5
Observable hazard function
2
=5
1
=2
0 =0.5
0 2 4 6 8 10
Time
Figure 10.2 Observable hazard function for Weibull baseline hazards with h0 (t) = 3t2
and gamma frailty distributions with θ = 0.5, 2, 5 and ∞.
A plot of h∗ (t) against t for a Weibull hazard with λ = 1, γ = 3 and various

values of θ is shown in Figure 10.2.
One limitation of a Weibull baseline hazard function, noted in Section 5.1.2
of Chapter 5, is that it is monotonic. However, as we can see in Figure 10.2,
when there is a degree of frailty present, we may observe unimodal hazard
functions. Consequently, an observed unimodal hazard may be the result of
frailty effects, rather than the intrinsic behaviour of the underlying baseline
hazard.
10.3.2 Impact of frailty on an observable hazard ratio

Suppose that the hazard function for the ith individual is
hi (t) = zi exp(βxi )h0 (t),
where xi is the value of a binary covariate X that takes values 0, 1 and zi is a

realisation of a Γ(θ, θ) random variable. The corresponding observable hazard
function is
eβxi h0 (t)
h∗i (t) = ,
1 + θ−1 eβxi H0 (t)
from Equation (10.13), and the observable hazard ratio for X = 1 relative to
X = 0 is
1 + θ−1 H0 (t) β
ψ ∗ (t) = e . (10.14)
1 + θ−1 eβ H0 (t)
THE GAMMA FRAILTY DISTRIBUTION 355
This hazard ratio is no longer constant but a function of time t, and so propor-
tional hazards no longer pertains. For example, in the particular case where
the underlying baseline hazard is assumed to be a constant λ, H0 (t) = λt and
the hazard ratio becomes
1 + θ−1 λt β
e ,
1 + θ−1 eβ λt
a non-linear function of t.
To illustrate the dependence of the observable hazard ratio on time, Fig-
ure 10.3 shows the hazard ratio ψ ∗ (t) for a Weibull baseline hazard with λ = 1,
γ = 3, ψ = eβ = 3 and various values of θ.
3 =
Observable hazard ratio
=5
=2
1 =0.5
0
0.0 0.5 1.0 1.5 2.0
Survival time
Figure 10.3 Observable hazard ratio for Weibull baseline hazard with h0 (t) = 3t2 ,
β = log 3, and gamma frailty distributions with θ = 0.5, 2, 5 and ∞.
The hazard ratio has a constant value of 3 when there is no frailty, that is
when θ = ∞, but when the frailty variance exceeds zero, the observable hazard
ratio declines over time. This dependence of the hazard ratio on time could
be used to account for any observed non-proportionality in the hazards.
These results show that the inclusion of a random effect in a Weibull pro-
portional hazards regression model can lead to hazards not being proportional,
and to a non-monotonic hazard function. Although this has been illustrated
using a Weibull model with gamma frailty, the conclusions drawn apply more
generally. This means that models that include random effects provide an al-
ternative way of modelling data where the hazard function is unimodal, or
where the hazards are not proportional.
356 FRAILTY MODELS
10.4 Fitting parametric frailty models
Fitting the model in Equation (10.3) entails estimating the values of the co-
efficients of the explanatory variables, the variance of frailty distribution and
the baseline hazard function. Models in which h0 (t) is fully specified, such as
the Weibull proportional hazards model or an accelerated failure time model,
can be fitted using the method of maximum likelihood. Denote the observed
survival data by the pairs (ti , δi ), i = 1, 2, . . . , n, where ti is the survival time
and δi is an event indicator, which takes the value zero for a censored obser-
vation and unity for an event. If the random effects ui in Equation (10.3) had
known values, the likelihood function would be
∏
n
{hi (ti )}δi Si (ti ), (10.15)
i=1
as in Equation (5.10) of Chapter 5. In this expression, the hazard function

for the ith of n individuals is given in Equation (10.3) and the corresponding
survivor function is
Si (ti ) = exp {− exp(β ′ xi + ui )H0 (ti )} , (10.16)

where H0 (ti ) is the cumulative hazard function evaluated at time ti .
However, the ui are not known, but are independent and identically dis-
tributed realisations of a random variable that has a probability distribution
with density f (ui ). In this situation, we integrate the likelihood contributions
over possible values of the random effects, so that from Equation (10.15), the
likelihood function becomes
n ∫
∏ ∞
{hi (ti )}δi Si (ti )f (ui ) dui , (10.17)
i=1 0
or equivalently, when working with the frailty terms zi ,
n ∫
∏ ∞
{hi (ti )}δi Si (ti )f (zi ) dzi . (10.18)
i=1 0
Numerical methods are generally needed to maximise this function, or its

logarithm.
Once the unknown parameters in a fully parametric model have been es-
timated, it is possible to obtain estimates of the random effects. To do this,
we use a version of Bayes’ theorem, according to which the probability of an
event A, conditional on an event B, is given by
P (B | A)P (A)
P(A | B) = . (10.19)
P (B)
FITTING PARAMETRIC FRAILTY MODELS 357
We now write L(ti | ui ) for the likelihood of the ith event time ti when the
random effect ui is regarded as fixed, which from combining Equation (10.16)
with Expression (10.15), is
L(ti | ui ) = {exp(β ′ xi + ui )h0 (ti )}δi exp{− exp(β ′ xi + ui )H0 (ti )}.
This can be interpreted as the probability of ti conditional on the random

effect ui . Next, the probability of a particular value of the random effect
is the density function f (ui ). Using Equation (10.19), the probability of ui
conditional on ti can be expressed as
L(ti | ui )f (ui )
π(ui | ti ) = , (10.20)
P (ti )
where P (ti ) is the marginal probability of the data, obtained by integrating

L(ti | ui )f (ui ) with respect to ui . This ensures that π(ui | ti ) integrates to
1, and so defines a proper probability density function. The mean or mode
of the distribution of the random variable associated with the ith random
effect, Ui , conditional on the data, π(ui | ti ), can then be obtained. Finally,
substituting estimates of the unknown β-coefficients and the variance of the
frailty distribution, we get an estimate of the mean or modal value of the
distribution of Ui , ûi , say, which can be regarded as an estimate of the random
effect. Because of the link to Bayes’ theorem in Equation (10.19), π(ui | ti )
is referred to as the posterior density of Ui , and the estimates ûi are termed
empirical Bayes estimates. In a similar way, the posterior variance of Ui can be
obtained from π(ui | ti ), which leads to the standard error of ûi . Corresponding
estimates of the frailty effects and their standard errors are obtained from
ẑi = exp(ûi ).
Once frailty effects have been estimated, Equations (10.9) or (10.16) can
be used to obtain fully parametric estimates of the survivor function. The
median survival time for a particular individual can then be found from the
expression
{ }1/γ̂
log 2
, (10.21)
λ̂ exp(β̂ ′ xi + ûi )
which is a straightforward adaptation of the result in Equation (5.42) of Chap-
ter 5.
Parametric survival models with frailty are most easily fitted when the
frailty effects have a gamma distribution, and so the fitting process for this
particular model is described in the following section.
10.4.1 ∗ Gamma frailty

When the frailty random variable, Zi , is assumed to have a gamma distri-
bution, a closed form can be obtained for the integrated likelihood in Equa-
tion (10.18). To show this, on substituting for the hazard function, survivor
358 FRAILTY MODELS
function and density of Zi into Equation (10.18), the likelihood function is
n ∫
∏ ∞ ′ ′ θθ ziθ−1 e−θzi
{zi eβ xi h0 (ti )}δi exp{−zi eβ xi H0 (ti )} dzi .
i=1 0 Γ(θ)
Collecting terms in zi , this becomes
∏n ∫ ∞
θθ ′ ′
{eβ xi h0 (ti )}δi ziθ+δi −1 exp{−[θ + eβ xi H0 (ti )]zi } dzi .
i=1
Γ(θ) 0
Next, the density of a gamma random variable Y with parameters r and θ is
θr y r−1 e−θy
f (y) = , (10.22)
Γ(r)
and so ∫ ∞
Γ(r)
y r−1 e−θy dy = ,
0 θr
from which
∫ ∞
′ Γ(θ + δi )
ziθ+δi −1 exp{−[θ + eβ xi H0 (ti )]zi } dzi = ′ .
0 {θ + e xi H0 (ti )}θ+δi
β
Consequently, the likelihood function becomes
∏n
θθ ′ Γ(θ + δi )
{eβ xi h0 (ti )}δi ′ .
i=1
Γ(θ) {θ + e xi H0 (ti )}θ+δi
β
The corresponding log-likelihood function is
∑
n
{θ log θ − log Γ(θ) + log Γ(θ + δi ) + δi [β ′ xi + log h0 (ti )]}
i=1
∑
n
′
− (θ + δi ) log{θ + eβ xi H0 (ti )}, (10.23)
i=1
and now that no integration is involved, standard numerica1 methods can be

used to maximise this function with respect to θ, the βs and the parameters
in the baseline hazard function, to give the maximum likelihood estimates.
Standard errors of these estimates are found from the corresponding informa-
tion matrix; see Appendix A. The estimated variance of the random effects
can then be found by using the maximum likelihood estimate of θ, θ̂, in Equa-
tion (10.8).
To obtain estimates of the frailty effects when they are assumed to have
a gamma distribution, it is more convenient to work with the zi rather than
the corresponding random effects ui . Writing L(ti | zi ) for the likelihood of ti
when the frailty zi is regarded as fixed, the numerator of Equation (10.20) is
then L(ti | zi )f (zi ), which is
′ ′ θθ ziθ−1 e−θzi
[zi eβ xi h0 (ti )]δi exp{−zi eβ xi H0 (ti )} .
Γ(θ)
Ignoring terms that do not involve zi , the posterior density of Zi , π(zi | ti ), is
proportional to
′
ziθ+δi −1 exp{−[θ + eβ xi H0 (ti )]zi }.
From the general form of a two-parameter gamma density function, shown in
Equation (10.22), it follows that the posterior distribution of Zi is a gamma
′
distribution with parameters θ + δi and θ + eβ xi H0 (ti ). Then since the ex-
pected value of a gamma random variable with parameters r, θ is r/θ, the
expectation of Zi given the data, is
θ + δi
E (Zi | ti ) = ′ .
θ + e xi H0 (ti )
β
From this, an estimate of the frailty effect for the ith individual is
θ̂ + δi
ẑi = ′
,
θ̂ + eβ̂ xi Ĥ0 (ti )
where Ĥ0 (t) is the estimated baseline cumulative hazard function. Similarly,
the variance of a gamma random variable is r/θ2 and so
θ + δi
var (Zi | ti ) = ′ .
{θ + e xi H0 (ti )}2
β
The estimated variance of Zi reduces to ẑi2 /(θ̂ + δi ) and so the standard error
of ẑi is given by
ẑi
se (ẑi ) = √ .
(θ̂ + δi )
Interval estimates for the frailty terms can then be found, and the ratio of
zî to its standard error can be compared to percentage points of a standard
normal distribution to give a P -value for a test of the hypothesis that the ith
frailty effect is zero. However, the results of a series of unplanned hypothesis
tests about frailty effects must be adjusted to allow for repeated significance
testing, for example by using the Bonferroni correction. With this correction,
the P -value for the frailty terms of n individuals are each divided by n before
interpreting the significance levels.
As this analysis shows, working with gamma frailties is mathematically
straightforward and leads to closed form estimates of many useful quantities.
In other situations, numerical methods are needed to evaluate the summary
statistics of the conditional distribution of the frailty given the data.
360 FRAILTY MODELS
When a frailty term with a gamma distribution is included in a Weibull model
for the survival times of women with negatively and positively stained tu-
mours, outlined in Example 10.1, the resulting hazard function is hi (t) =
zi exp(βxi )λγtγ−1 , and the corresponding fitted survivor function is
Ŝi (t) = exp{−ẑi exp(β̂xi )λ̂tγ̂ }.
The variance of the underlying gamma distribution for the frailty is θ̂−1 =
3.40, and estimates of the frailty effects follow from the method outlined in
this section.
In the presence of frailty, β̂ = 2.298, and the hazard ratio for a positively
stained woman relatively to one who is negatively stained is 9.95. However,
the 95% confidence interval for this estimate ranges from 0.86 to 115.54, re-
flecting the much greater uncertainty about the staining effect when account
is taken of frailty. It is also important to note that the estimated hazard ratio
is conditional on frailty, and so refers to a woman with a specific frailty value.
The revised estimates of λ and γ are 0.00008 and 1.9091, respectively, and
the individual frailty values are shown in Table 10.1, alongside the observed
survival times. Note that women with the shortest survival times have the
largest estimated frailty effects.
Table 10.1 Survival times of women with tumours that were negatively or positively
stained and corresponding gamma frailties.
Negative staining Positive staining
Survival time Frailty Survival time Frailty Survival time Frality
23 3.954 5 4.147 68 0.444
47 3.050 8 3.827 71 0.412
69 2.290 10 3.579 76* 0.083
70* 0.514 13 3.192 105* 0.047
71* 0.507 18 2.581 107* 0.045
100* 0.348 24 1.981 109* 0.044
101* 0.344 26 1.816 113 0.179
148 0.888 26 1.816 116* 0.039
181 0.646 31 1.472 118 0.166
198* 0.127 35 1.254 143 0.116
208* 0.117 40 1.038 154* 0.023
212* 0.113 41 1.001 162* 0.021
224* 0.103 48 0.788 188* 0.016
50 0.739 212* 0.013
59 0.564 217* 0.012
61 0.534 225* 0.011
The individual estimates of the survivor functions are shown in Figure 10.4.
This figure illustrates the extent of variation in the estimated survivor func-
tions that stems from the frailty effect, and shows that there is some separation
0.8
0.6
0.4
0.2
0.0
0 50 100 150 200 250
Survival time
Figure 10.4 Fitted Weibull survivor functions for women with positive (·······) and
negative (—) staining.
in the estimates for those women whose tumours were positively or negatively
stained.
The median survival times, found using Equation (10.21), vary from 55
to 372 months for women in the negatively stained group, and from 16 to
355 months for those in the positively stained group. Of the 32 women with
positively stained tumours, 18 have estimated median survival times that are
less than any of those in the negatively stained group. This confirms that once
allowance is made for frailty effects, differences between survival times in the
two groups of women are not as pronounced.
The observable survivor function under this model, from Equation (10.12),
can be estimated by {1 + θ̂−1 Ĥ0 (t)}−θ̂ for a woman with a negatively stained
tumour and {1 + θ̂−1 eβ̂ Ĥ0 (t)}−θ̂ for one with a positively stained tumour.
These functions are shown in Figure 10.5, superimposed on the corresponding
Kaplan-Meier estimate of the survivor functions. Comparing this with Fig-
ure 10.1, we see that once allowance is made for frailty, the Weibull model is
a much better fit to the observed survival times. This figure also provides a
visual confirmation of the suitability of the fitted model.
The observable, or unconditional, hazard ratio, for a woman with a pos-
itively stained tumour relative to one whose tumour was negatively stained,
derived from Equation (10.14), is given in Figure 10.6. This figure shows how
the observable hazard ratio varies over time; the hazard is much greater at
earlier times, but declines quite rapidly. This is due to the selection effect of
frailty, whereby women who are more frail die sooner. There are also many
362 FRAILTY MODELS
1.0
0.8
0.6
0.4
0.2
0.0
0 50 100 150 200 250
Survival time
Figure 10.5 Fitted survivor functions for the Weibull gamma frailty model, with the
corresponding observed survivor functions, for women with positive (·······) and negative
(—) staining.
10
8
Hazard ratio
0
0 50 100 150 200 250
Survival time
Figure 10.6 Ratio of the hazard of death for a women with a positively stained tumour
relative to one whose tumour is negatively stained.
FITTING SEMI-PARAMETRIC FRAILTY MODELS 363
more early deaths in the positively stained group, which is why the observed
hazard ratio varies in this way.
10.5 Fitting semi-parametric frailty models

When no underlying parametric form is assumed for the baseline hazard func-
tion, as in the Cox regression model, the procedure for fitting a frailty model
described in the previous section can no longer be used. This is because the
likelihood function in Expression (10.15) is no longer fully specified. The ap-
proach that is now most widely implemented in standard software packages
involves maximising the sum of the partial log-likelihood for the Cox model
that includes the random effects, log Lp (β, u), and the log-likelihood of the
random effects. This log-likelihood function is
∑
n
log Lp (β, u) + log f (ui ), (10.24)
i=1
where β is the vector of coefficients of p explanatory variables in a Cox regres-

sion model, and u is the vector of random effects for the n individuals. The
partial log-likelihood is found by replacing β ′ xi in Equation (3.6) of Chapter 3
by β ′ xi + ui , to give
 
∑n  ∑ 
log Lp (β, u) = δi β ′ xi + ui − log exp(β ′ xl + ul ) ,
 
i=1 l∈R(ti )
where δi is the event indicator and R(ti ) is the set of patients at risk of death
at time ti . The random effects, ui , are the observed values of frailty random
variables, Ui = exp(Zi ), where Zi is usually taken to have either a lognormal
or a gamma distribution.
10.5.1 ∗ Lognormal frailty effects

When frailty effects are assumed to be lognormally distributed, the random
effects, ui , in Expression (10.24), are realisations of an N (0, σu2 ) random vari-
able, with density ( )
1 u2i
f (ui ) = √ exp − 2 .
σu (2π) 2σu
The log-likelihood function in Expression (10.24) is then
1 ∑ 2
n
√
log Lp (β, u) − n log{σu (2π)} − 2 u . (10.25)
2σu i=1 i
Since this log-likelihood is only used to estimate the p components of β and
the n components of u, the term involving σu2 alone can be omitted to give
1 ∑ 2
n
log Lpen (β, u, σu2 ) = log Lp (β, u) − u . (10.26)
2σu2 i=1 i
364 FRAILTY MODELS
This is known as a penalised partial log-likelihood, since the effect of the second
term is to assign penalties to the partial log-likelihood function when the ui
have more extreme values, that is values that are further from their expected
value of zero. The σu2 term in Equation (10.26) essentially controls the relative
importance of the two components of this log-likelihood function.
The maximisation process proceeds iteratively by starting with a provi-
sional estimate of σu2 and finding the estimates of the βs and the us that
maximise log Lpen (β, u, σu2 ). Next, a marginal log-likelihood, log Lm (β, σu2 ),
is obtained by integrating the log-likelihood in Expression (10.25) over the
random effects ui . Ripatti and Palmgren (2000) show that, at estimates β̂ of
β, this is well-approximated by
1 1
log Lm (σu2 ) = log Lpen (β̂, û, σu2 ) − log(σu2n ) − log |I u |, (10.27)
2 2
where I u is the n × n observed information matrix for the random effects,
formed from the negative second partial derivatives of log Lpen (β, u, σu2 ) with
respect to the ui , i = 1, 2, . . . , n, evaluated at β̂, û, and |I u | is the determinant
of that matrix. Using the estimates of the βs, the marginal log-likelihood in
Equation (10.27) is then maximised with respect to σu2 to give a revised esti-
mate of σu2 . This process is repeated until the difference between two successive
estimates of σu2 is sufficiently small.
Estimates of the random effects, ûi , and hence the frailty terms ẑi =
exp(ûi ), are also obtained from this iterative process. Moreover, the maximum
likelihood estimate of the variance of the random effect is given by
{ n }
∑
σ̂u2 = n−1 û2i + trace (I −1
u ) ,
i=1
where trace (I −1
u )
is the sum of the diagonal elements of the inverse of I u .
The standard error of this estimate is
[ ]−1/2
√ 2 1 −1 −1 1 −1
se (σ̂u ) = (2σ̂u ) n + 4 trace (I u I u ) − 2 2 trace (I u )
2
.
σ̂u σ̂u
Maximum likelihood estimates of variances tend to be biased, and instead
estimates based on the method of restricted maximum likelihood (REML) es-
timation are preferred. For example, in the case of estimating the variance
of a single sample of observations, x1 , x2 , . . . , xn , from a normal distribu-
tion,∑
the maximum likelihood estimate of the variance is the biased estimate
n−1 i (xi − x̄)2 whereas the corresponding
∑ REML estimate turns out to be
the usual unbiased estimate (n − 1)−1 i (xi − x̄)2 .
REML estimates are obtained from a likelihood function that is indepen-
dent of β, and the REML estimate of the variance of the random frailty term
is { n }
∑
2 −1 2
σ̃u = n ũi + trace (Ṽ u ) ,
i=1
FITTING SEMI-PARAMETRIC FRAILTY MODELS 365
where Ṽ u is the estimated variance-covariance matrix of the REML estimates
of the ui , ũi . The trace of this matrix is just the sum of the estimated vari-
ances of the ũi . This is the preferred estimate of the variance of a normally
distributed random frailty effect. The standard error of σ̃u2 can be found from
[ ]−1/2
√ 2 1 1
se (σ̃u ) = (2σ̃u ) n + 4 trace (Ṽ u Ṽ u ) − 2 2 trace (Ṽ u )
2
.
σ̃u σ̃u
Both σ̃u2 and its standard error generally feature in the output of software
packages that have the facility for fitting Cox regression models with lognormal
frailty.
For fully parametric frailty models, the cumulative baseline hazard func-
tion and the corresponding survivor function can be estimated using the max-
imum likelihood estimates of the unknown parameters. However, in semi-
parametric frailty models, estimates of the baseline hazard and cumulative
hazard cannot be extended to take account of the frailty terms, and so esti-
mated survivor functions cannot easily be obtained.
10.5.2 ∗ Gamma frailty effects

When the random variable associated with frailty effects is assumed to have a
gamma distribution with unit mean and variance 1/θ, the corresponding ran-
dom effects, ui , have an exp-gamma distribution, introduced in Section 10.2.1.
Using the density function in Equation (10.6), the log-likelihood function in
Expression (10.24) is now
∑
n
log Lp (β, u) + nθ log θ − n log Γ(θ) − {θeui − θui }.
i=1
This log-likelihood function will be used to estimate the components of β and

u, and so terms involving θ alone can be omitted to give the penalised partial
log-likelihood function
∑
n
log Lpen (β, u, θ) = log Lp (β, u) − θ {eui − ui }, (10.28)
i=1
∑n
in which θ i=1 {eui − ui } is the penalty term.
To obtain estimates of β, u and θ, estimates of β and u are first taken
to be the values that maximise Equation (10.28) for a given value of θ. The
marginal log-likelihood for θ, at estimates β̂ and û of β and u, is then used
to obtain a revised estimate of θ. This has been shown by Therneau and
Grambsch (2000) to be given by
log Lm (θ) = log Lpen (β̂, û, θ) + nθ(1 + log θ)

∑n { [ ]}
Γ(θ + δi )
− (θ + δi ) log(θ + δi ) − log , (10.29)
i=1
Γ(θ)
366 FRAILTY MODELS
where δi is the event indicator for the ith individual. Maximising log Lm (θ)
in Equation (10.29) with respect to θ, leads to θ̂. This estimate is then used
with Equation (10.28) to obtain updated estimates of β and u, and so on until
the process converges. No closed form estimate of θ, nor its standard error, is
available.
As the frailty variance, 1/θ, tends to zero, the marginal log-likelihood in
Equation (10.29) becomes
∑
n
log Lp (β̂) − δi ,
i=1
where log Lp (β̂) is the maximised partial log-likelihood for the Cox regression
model when∑ there is no frailty. By taking the marginal log-likelihood to be
n
log Lm (θ) + i=1 δi , the maximised marginal log-likelihood in the presence of
frailty is then directly comparable to that for the model with no frailty. This
is helpful in comparing models, as we will see in Section 10.6.
10.6 Comparing models with frailty

In many cases, frailty acts as a nuisance term in the model. While it is impor-
tant to take account of its effects, the main interest is in determining which
explanatory variables are needed in a model, and obtaining estimates of their
effects, in the presence of frailty. The methods described in Section 5.7 of
Chapter 5, based on the change in the −2 log L̂ statistic, can be used to com-
pare parametric models that incorporate frailty. Model selection strategies for
determining which factors to include in a parametric model in the presence
of frailty can then be straightforwardly implemented. The fitted models can
subsequently be used to draw inferences about their effects, allowing for frailty.
The situation is more complicated for the Cox regression model. This is
because a penalised likelihood does not behave in the same way as the usual
likelihood or partial likelihood functions, since it contains unobservable frailty
effects. Instead, the maximised marginal log-likelihood statistic is used, which
is log Lm (σ̂u2 ) or log Lm (θ̂) for lognormal and gamma frailties, respectively.
This statistic is generally given in computer output from fitting Cox regres-
sion models that include frailty. Differences in the values of −2 log Lm (σ̂u2 ) or
−2 log Lm (θ̂), or −2 log L̂m for short, can be compared with percentage points
of a χ2 distribution in the usual manner. In the case of lognormal frailty ef-
fects, the marginal likelihood based on the maximum likelihood estimates of
the βs is used, rather than that based on the REML estimates.
10.6.1 Testing for the presence of frailty

In parametric models, the hypothesis of no frailty effects can be evaluated by
comparing −2 log L̂ values for models with and without frailty effects, for a
given set of covariates, with percentage points of a chi-squared distribution
COMPARING MODELS WITH FRAILTY 367
on one degree of freedom, χ21 . However, some care is needed as the standard
asymptotic theory that underpins the result that differences in −2 log L̂ values
have a chi-squared distribution, is no longer valid. Essentially this is because
the hypothesis of no frailty effect corresponds to testing the hypothesis that
the variance of the frailty term is zero, and this is the smallest possible value
that the variance can take. Tests based on changes in −2 log L̂ then tend to
be conservative, and the resulting P -value for a difference will be larger than
it should be. In this situation, the difference in −2 log L̂ values, when frailty is
added, can be compared informally to percentage points of a χ21 distribution.
If the observed difference is relatively large or small, conclusions about the
degree of frailty will be clear.
For a more precise determination of the significance of the frailty effect, we
can use the result that the asymptotic distribution of the change in −2 log L̂,
when a frailty term is added to a fully parametric model, is an equally weighted
mixture of a χ20 and χ21 distributions, written 0.5(χ20 + χ21 ). A χ20 distribution
has a point mass at zero, and so if the random variable W has a 0.5(χ20 + χ21 )
distribution, P(W = 0) = 0.5 and P(W > w) = 0.5P(χ21 > w) for w > 0.
Consequently, to apply this result, the P -value for testing the hypothesis that
the frailty variance is zero is just half of that obtained from using a χ21 distri-
bution.
In borderline cases, other methods may be needed, such as bootstrapping.
This procedure can be used to obtain an interval estimate of the variance
of a frailty term. Briefly, bootstrapping involves resampling the data with
replacement to give a new sample of n survival times. The survival model
that includes a frailty term is then fitted to this sample, and the variance of
the random effect is then estimated. This is repeated a large number of times.
The 2.5% and 97.5% percentiles of the distribution of the bootstrap variance
estimates then define a 95 per cent interval estimate for the variance.
In the case of the Cox regression model, fitted using the method of pe-
nalised partial log-likelihood outlined in Section 10.5, the significance of a
frailty term is assessed using the −2 log L̂m statistic formed from the max-
imised marginal log-likelihood. For lognormal frailty effects, the maximised
marginal log-likelihood for the model with frailty, obtained from maximum
likelihood estimation and not REML, can be compared directly with the max-
imised partial log-likelihood for the corresponding Cox regression model with-
out frailty. A similar process is used when
∑n gamma frailty effects are assumed,
but here the total number of events, i=1 δi , has to be added to the max-
imised marginal log-likelihood, as explained in Section 10.5.2. As with para-
metric modelling, this test procedure is only approximate, and bootstrapping
may be required for a more reliable assessment of the magnitude of the frailty
effect.
Example 10.3 Survival of patients registered for a lung transplant

The likely survival time of a patient from when they are assessed as needing an
organ transplant is of great interest to patients and clinicians alike. In a study
368 FRAILTY MODELS
to quantify this survival time and to identify factors associated with survival
from listing for a lung transplant, data were obtained on the UK patients
who were registered for a transplant in 2004. Survival times were measured
from the date of registration until death, regardless of all intervening treat-
ments, including transplantation. Survival was censored at the date of removal
from the list, the last known follow-up date, or at 30 April 2012 for patients
still alive. In addition, the age at registration, gender, body mass index and
primary disease were recorded, where primary disease was categorised as fi-
brosis, chronic obstructive pulmonary disease (COPD), suppurative disease
and other. There are 196 patients in the data set, of whom 123 died during
the study period. Data for the first 20 patients are shown in Table 10.2.
Table 10.2 Survival times of 20 patients from listing for a lung transplant.
Patient Survival time Status Age Gender BMI Disease
1 2324 1 59 1 29.6 COPD
2 108 1 28 1 22.6 Suppurative
3 2939 0 55 1 32.1 Fibrosis
4 1258 0 62 1 30.0 Fibrosis
5 2904 0 51 1 30.4 Fibrosis
6 444 1 59 1 26.9 Fibrosis
7 158 1 55 2 24.6 COPD
8 1686 1 53 2 26.8 COPD
9 142 1 47 1 32.2 Fibrosis
10 1624 1 53 2 15.7 COPD
11 16 1 62 1 26.4 Fibrosis
12 2929 0 50 1 29.0 COPD
13 1290 1 55 2 17.1 COPD
14 2854 0 47 1 20.0 Other
15 237 1 23 1 15.9 Suppurative
16 136 1 65 1 16.0 COPD
17 2212 1 24 2 19.5 Suppurative
18 371 1 54 1 28.9 Fibrosis
19 683 1 24 2 20.2 Suppurative
20 290 1 53 1 25.2 Fibrosis
We first fit a Weibull proportional hazards model that contains age, gender,

body mass index and primary disease. The value of the −2 log L̂ statistic for
this model is 2043.73. A lognormal frailty term is now introduced, by adding
a normally distributed random effect to the linear component of the model,
and maximising the likelihood function in Expression (10.17). The Weibull
model with lognormal frailty has a −2 log L̂ value of 2026.85, and the change
in −2 log L̂ on adding the random effect is 16.88. Comparing this reduction
with percentage points of a χ21 distribution, the frailty effect is highly signifi-
cant, with a P -value of less than 0.001. This means that there is substantial
heterogeneity between the survival times of the patients in this study, after
allowing for the effects of the four explanatory variables. A less conserva-
tive estimate of the significance of the frailty effect is found by referring the
change in −2 log L̂ to a 0.5(χ20 + χ21 ) distribution, equivalent to halving the
P -value from a comparison with a χ21 distribution, and it is then even more
significant. The variance of the normal random effect corresponding to the
lognormal frailty is σ̂u2 = 2.254. Using Equations (10.4) and (10.5), the mean
and variance of the lognormal frailty effect are 3.09 and 81.20, respectively.
We next fit a Weibull model that contains age, gender, body mass index
and primary disease, together with a gamma frailty effect. This model is fitted
by maximising the log-likelihood function in Expression (10.23), from which
the gamma frailty effect has variance θ̂−1 = 3.015. The corresponding variance
of the random effect is calculated from Equation (10.8), and is 10.19. The
Weibull model with gamma frailty has a −2 log L̂ value of 2023.69, and so
addition of the frailty term leads to a reduction in the value of the −2 log L̂
statistic of 20.04. This change is highly significant (P < 0.001), and greater
than that when a lognormal frailty is introduced.
In this example, the variance of the two distributions for the frailty term,
and the corresponding distributions of the random effect, are quite different.
To explore this in more detail, Figures 10.7 and 10.8 show the fitted normal
and exp-gamma probability density functions for the random variable U , and
the corresponding fitted lognormal and gamma probability density functions
for the frailty random variable, Z.
0.3
Probability density function of U
0.2
0.1
0.0
-6 -4 -2 0 2 4
Value of u
Figure 10.7 Fitted probability density functions for the normal (—) and exp-gamma
(·······) random effects.
Figure 10.7 shows the asymmetry in the exp-gamma distribution of the

random effect, and although the mode is zero, the mean, from Equation (10.7),
is −2.05. This explains the disparity in the estimated variances of the normal
370 FRAILTY MODELS
1.0
0.9
Probability density function of Z
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0 2 4 6
Value of z
Figure 10.8 Fitted probability density functions for the lognormal (—) and gamma
(·······) frailty effects.
and exp-gamma distributions for the random effect. Figure 10.8 shows that
the estimated density functions of the frailty effect are quite similar when
zi exceeds 0.2. However, the density of the fitted gamma distribution tends
to ∞ as zi tends to zero, whereas the fitted lognormal distribution is unimodal.
Also, the mean of the fitted gamma distribution is unity, whereas from Equa-
tion (10.4), the fitted lognormal distribution has a mean of 3.09, which is why
the lognormal frailty variance is so much greater than that for the gamma
distribution.
To compare the parameter estimates for the various terms and their stan-
dard errors, Table 10.3 shows the estimates for the different fitted Weibull
models, including the estimated variances of the random effects, ui , and the
frailty effects, zi , denoted d var (ui ) and d
var (zi ), respectively. Note that for
a normal random effect, the estimated variance is d var (ui ) = σ̂u2 , while for
gamma frailty, d var (zi ) = 1/θ̂. Equation (10.5) is then used to obtain the vari-
ance of a lognormal frailty term and Equation (10.8) gives the variance of the
random effect corresponding to gamma frailty.
This table shows that there is a general tendency for the parameter esti-
mates to be further from zero when a frailty term is added, and the standard
errors are also larger. However, inferences about the impact of the factors on
patient survival from listing are not much affected by the inclusion of a frailty
term, and there is little difference in the results obtained when either a lognor-
mal or gamma frailty effect is included. The only factor to have any effect on
patient survival is primary disease, with the hazard of death in patients with
COPD being less than that for patients with other diseases. Also, on the basis
Table 10.3 Parameter estimates (standard error) in fitted Weibull models.

Term No frailty Lognormal frailty Gamma frailty
Age 0.015 (0.012) 0.014 (0.018) 0.003 (0.020)
Gender: Male 0.080 (0.198) 0.226 (0.325) 0.433 (0.412)
BMI −0.024 (0.024) −0.029 (0.039) −0.014 (0.049)
Disease: COPD −0.675 (0.377) −0.960 (0.597) −1.008 (0.694)
Disease: Fibrosis 0.363 (0.344) 0.887 (0.575) 1.382 (0.724)
Disease: Other −0.254 (0.391) −0.231 (0.623) −0.076 (0.768)
Weibull shape (γ̂) 0.663 (0.051) 1.040 (0.141) 1.304 (0.231)
Weibull scale (λ̂) 0.0061 (0.0042) 0.0004 (0.0005) 0.0003 (0.0004)
d
var (ui ) – 2.254 10.186
d
var (zi ) – 81.197 3.015
−2 log L̂ 2043.73 2026.85 2023.69
of the −2 log L̂ statistic, the model with gamma frailty might be preferred to
that with lognormal frailty.
To avoid making specific assumptions about the form of the underlying
baseline hazard function, we fit a Cox regression model that contains age,
gender, body mass index and primary disease, and that also includes a log-
normal or gamma frailty term. These models are fitted by maximising the
penalised log-likelihood functions in Equations (10.26) and (10.28), respec-
tively. To test the hypothesis that all frailty effects are zero in the pres-
ence of the four explanatory variables, we compare the value of the −2 log L̂
statistic for the fitted Cox model, which is 1159.101, with values of the
maximised marginal log-likelihood statistic, −2 log L̂m , for Cox models with
frailty.
When a lognormal frailty term is added, the value of the maximised
marginal log-likelihood statistic, −2 log Lm (σ̂u2 ) from Equation (10.27), is
1157.477, and this leads to a change in the value of the −2 log L̂m statistic of
1.62, which is not significant (P = 0.203).
On fitting models with gamma frailty, the value of the log Lm (θ̂) statistic
from Equation (10.29) is −699.653, and adding the observed number of deaths,
123, to this and multiplying by −2, gives 1153.306. This can be compared
with the value of −2 log L̂ for the Cox regression model that contains the
same explanatory variables, but no frailty effects, 1159.101. The reduction in
the value of this test statistic is 5.80, which is significant when compared to
percentage points of a χ2 distribution on 1 d.f. (P = 0.016). This now shows
a significant frailty effect after allowing for survival differences due to primary
disease and the other factors.
The variances of the normal and exp-gamma random effects are smaller in
the Cox model than in the corresponding Weibull model, and the frailty effect
is less significant. This suggests that the fitted Cox regression model explains
more of the variation in survival times from listing for a transplant than a
Weibull model.
372 FRAILTY MODELS
To compare the parameter estimates for the various terms in a Cox re-
gression model and their standard errors, Table 10.4 shows the estimates for
the different models fitted, and the estimated variances of the random effects,
d
var (ui ) and frailty effects, d
var (zi ). For lognormal frailty, REML estimates
are given. Equations (10.5) and (10.8) have again been used to estimate the
variance of the frailty effects for lognormal frailty and the variance of the
random effects for gamma frailty.
Table 10.4 Parameter estimates (standard error) in fitted Cox models.

Term No frailty Lognormal frailty Gamma frailty
Age 0.013 (0.012) 0.012 (0.015) 0.006 (0.018)
Gender: Male 0.078 (0.198) 0.146 (0.267) 0.305 (0.352)
BMI −0.023 (0.024) −0.028 (0.032) −0.024 (0.042)
Disease: COPD −0.621 (0.371) −0.754 (0.482) −0.887 (0.598)
Disease: Fibrosis 0.377 (0.340) 0.704 (0.459) 1.225 (0.607)
Disease: Other −0.218 (0.386) −0.185 (0.512) −0.089 (0.659)
d
var (ui ) – 1.223 2.147
d
var (zi ) – 8.149 5.572
−2 log L̂m 1159.10 1157.48 1153.31
As with the Weibull model, some parameter estimates are larger when a
frailty term is added and the standard errors are larger, but the parameter
estimates in Tables 10.3 and 10.4 are broadly similar. As for the Weibull
model, the gamma frailty model leads to the smallest value of the −2 log L̂
statistic and is a better fit to the data.
Estimates of frailty effects can be obtained using the approach outlined in
Section 10.4. For example, from estimates of the frailty terms, ẑi , for the Cox
model with lognormal frailty, there are 7 patients with values of ẑi greater
than 3, namely patients 11, 36, 69, 70, 87, 113 and 163. The survival times
of these patients are 16, 21, 3, 4, 38, 22 and 35 days respectively, and so the
patients with largest frailty are those whose survival is shortest, as expected.
10.7 The shared frailty model

Models that seek to explain heterogeneity in the survival times of individuals
can provide improved estimates of the effects of covariates, or help to explain
features such as non-proportional hazards. However, frailty models are partic-
ularly useful in modelling situations where there is some characteristic whose
values are shared by groups of individuals. Some examples of potential areas
of application were given in Section 10.1.3.
To formulate a shared frailty model for survival data, suppose that there
are g groups of individuals with ni individuals in the ith group, i = 1, 2, . . . , g.
For the proportional hazards model, the hazard of death at time t for the jth
individual, j = 1, 2, . . . , ni , in the ith group, is then
hij (t) = zi exp(β ′ xij )h0 (t), (10.30)
THE SHARED FRAILTY MODEL 373
where xij is a vector of values of p explanatory variables for the jth individual
in the ith group, β is the vector of their coefficients, h0 (t) is the baseline hazard
function, and the zi are frailty effects that are common for all ni individuals
within the ith group. The hazard function in Equation (10.30) can also be
written in the form
hij (t) = exp(β ′ xij + ui )h0 (t),
where ui = log(zi ), and are assumed to be realisations of g random variables
U1 , U2 , . . . , Ug . The distribution assumed for Ui is taken to have zero mean,
and the normal distribution is a common choice.
The form of the baseline hazard may be fully specified as in a Weibull
model, or unspecified as in the Cox model. The general parametric acceler-
ated failure time model that incorporates a shared frailty component is, from
Equation (6.8) in Chapter 6, of the form
hij (t) = e−ηij h0 (t/eηij ),
where ηij = α′ xij + ui . Equivalently, extending Equation (6.9), this model
can be expressed in log-linear form as
log Tij = µ + α′ xij + ui + σϵij ,
where Tij is the random variable associated with the survival time of the jth
individual in the ith group, µ and σ are intercept and scale parameters, and
ϵij has some specified probability distribution.
10.7.1 ∗ Fitting the shared frailty model

Suppose that the survival times for the jth individual in the ith group are
denoted tij , for j = 1, 2, . . . , ni and i = 1, 2, . . . , g, and write δij for the
corresponding event indicator, which is unity if tij is an event time and zero
otherwise. Also, let Sij (t) and hij (t) be the survivor and hazard functions
for the jth individual in the ith group. For a fully parametric model, the
likelihood function for the observations in the ith group, when the frailty
terms are known, is
∏
ni
Li (β) = hij (tij )δij Sij (tij ).
j=1
As in Section 10.4, we next integrate this likelihood over the ui to give

∫ ∞
Li (β)f (ui ) dui ,
0
where f (ui ) is the probability density function of Ui . Over the g groups, the
likelihood function is
∏g ∫ ∞
L(β) = Li (β)f (ui ) dui .
i=1 0
374 FRAILTY MODELS
As in the case of individual frailty effects, the integration can only be carried
out analytically if the frailty effects have a Γ(θ, θ) distribution. In this case,
the likelihood function for the ith group is
ni {
∏ }δij
′ θθ Γ(θ + di )
eβ xij h0 (tij ) ∑ ′ θ+di
,
j=1
Γ(θ){θ + j exp(β xij )H0 (tij )}
∑
where di = i δij is the number of deaths in the ith group. The corresponding
log-likelihood over the g groups is
∑
g
log L(β) = {log Γ(θ + di ) − log Γ(θ) − di log θ}
i=1
 
∑
g ∑
ni
− (θ + di ) log 1 + θ−1 exp(β ′ xij )H0 (tij )
i=1 j=1
∑
g ∑
ni
+ δij [β ′ xij + log h0 (tij )] ,
i=1 j=1
which can be maximised to give estimates of θ, the parameters in the baseline

hazard function, and the βs.
Once this model has been fitted, estimates of the frailty effects, ẑi , can be
obtained in the same manner as described in Section 10.4, and we find that
θ̂ + di
ẑi = ∑ni ′
,
θ̂ + j=1 exp(β̂ xij )Ĥ0 (tij )
with standard error ẑi /(θ̂ + di ).

Cox regression models with frailty can again be fitted using penalised
partial log-likelihood methods, as in Section 10.5. To adapt the formulae given
in that section to the case of shared frailty models, the event indicator δi is
replaced by di , the number of deaths in the ith of g groups, and summations
over i = 1 to n are replaced by summations over i = 1 to g.
10.7.2 Comparing shared frailty models

Again, the results given in Section 10.6 apply equally to shared frailty mod-
els. In particular, fully parametric models that include a shared frailty term
can be compared using the −2 log L̂ statistic. To test for frailty, the change
in the value of the −2 log L̂ statistic on adding frailty effects can be formally
compared with percentage points of a 0.5(χ20 + χ21 ) distribution, as in Sec-
tion 10.6.1.
For the Cox model with shared frailty that is fitted using penalised partial
log-likelihood methods, the marginal log-likelihood statistics, −2 log Lm (σ̂u2 )
and −2 log Lm (θ̂), can be used to compare models with lognormal or gamma
THE SHARED FRAILTY MODEL 375
frailties, respectively. These statistics can also be compared to the value of
−2 log L̂ for the Cox regression model without frailty, to assess the extent of
frailty effects. However, the limitations of this approach, outlined in Section
10.6.1,
∑g also apply here. Note that in the case of gamma frailty, the quantity
i=1 di needs to be added to log Lm (θ̂) to ensure comparability with the
maximised partial log-likelihood in the absence of frailty, as for univariate
frailty.
Estimates of the random effects can be found using the method described
in Section 10.4. Such estimates are particularly useful when the frailty term
represents centres in a multicentre study, since the rank order of the estimated
centre effects provides information about the merits of the different centres
in terms of patient survival. But here, the estimated coefficients of the ex-
planatory variables in the model are interpreted as conditional on the shared
frailty. This means that hazard ratios relate to a comparison of effects in in-
dividuals within the same group. Estimates of the random effects also lead
to estimates of the survivor function for individuals with given characteris-
tics, and median survival times can easily be obtained. No new principles are
involved.
We conclude with an example that illustrates some of these features.
Example 10.4 Survival following kidney transplantation

Deceased organ donors generally donate both kidneys which are subsequently
transplanted into two different recipients, often by different transplant cen-
tres. In modelling short-term graft and patient survival, account needs to be
taken of a number of factors associated with the recipient, and factors as-
sociated with the transplant procedure itself. This illustration is based on
the outcomes of all kidney transplants carried out in a particular year, using
donors who are deceased following circulatory death. The outcome variable
is transplant survival, defined as the earlier of graft failure or death with a
functioning graft. Of the many explanatory factors that were recorded, patient
age (years), diabetes status (0 = absent, 1 = present) and the cold ischaemic
time (CIT), the time in hours between retrieval of the kidney from the donor
and transplantation into the recipient, are used in this example.
In addition to these recipient and transplant factors, donor factors may
also have an impact on the outcome of the transplant. Although many factors
related to the donor are routinely recorded, it is difficult to take account of
everything. Moreover, the survival times of the patients who receive the two
kidneys from the same donor may be expected to be more highly correlated
than the times for recipients of kidneys from different donors. One way of
taking account of donor factors, while allowing for association between the
outcomes when organs from the same donor are used, is to include the donor
as a shared frailty effect. The frailty term is then the same in models for the
survival times of two recipients of kidneys from the same donor.
In this study there were 434 transplants using organs from 270 deceased
donors. Of these, 106 gave one kidney and 164 donated both kidneys. Data
376 FRAILTY MODELS
for the 22 recipients of kidneys from the first 15 donors in the year are shown
in Table 10.5.
Table 10.5 Survival times of 20 patients following kidney transplantation.

Patient Donor Survival time Status Age Diabetes CIT
1 1 1436 0 61 0 14.5
2 2 1442 0 30 0 7.2
3 3 310 1 62 0 14.3
4 4 1059 1 62 0 10.6
5 5 236 1 50 0 18.7
6 6 1372 0 65 0 19.3
7 6 382 1 71 0 12.2
8 7 736 1 46 0 13.6
9 7 1453 0 66 0 13.9
10 8 1122 0 53 0 18.0
11 8 1010 0 42 0 13.8
12 9 1403 0 44 0 12.9
13 10 1449 0 67 0 13.9
14 11 843 1 34 0 16.6
15 12 1446 0 32 0 10.3
16 12 1397 0 51 0 19.0
17 13 1301 0 75 0 12.5
18 13 1442 0 47 0 11.3
19 14 1445 0 69 0 18.8
20 14 1445 0 53 0 23.2
21 15 1445 0 46 0 15.5
22 15 1445 0 49 0 17.2
On fitting a Weibull model containing cold ischaemic time, and the age and
diabetic status of the recipient, the value of the −2 log L̂ statistic is 1352.24.
To take account of donor effects, a lognormal shared frailty effect is added,
and the −2 log L̂ statistic decreases slightly to 1351.60. This reduction of 0.64
on 1 d.f. is not significant (P = 0.42) when compared to a χ21 distribution,
and remains non-significant (P = 0.21) compared to the percentage points of
the 0.5(χ20 + χ21 ) distribution.
When a lognormal frailty is added to a Cox regression model that con-
tains the same explanatory variables, the maximised marginal log-likelihood,
multiplied by −2, is 831.469, whereas for the corresponding Cox regression
model without frailty, −2 log L̂ = 831.868. Again, the reduction on adding a
frailty effect is not significant. Very similar results are obtained when donor
effects are modelled using a shared gamma frailty term, and we conclude that
there is no reason to include donor effects in the model. The estimated coef-
ficients of the three explanatory variables are very similar under a Cox and
Weibull model, and are hardly affected by including random donor effects.
We conclude that taking proper account of the donor effects has not materi-
SOME OTHER ASPECTS OF FRAILTY MODELLING 377
ally affected inferences about the effect of explanatory factors on short-term
survival of these transplant recipients.
10.8 ∗ Some other aspects of frailty modelling
A number of other issues that arise in connection with frailty modelling are
discussed briefly in this section.
10.8.1 Model checking
The techniques for model checking described in Chapters 4 and 7 may reveal
outlying observations, influential values and inadequacies in the functional
form of covariates. In addition, an informal way of assessing the adequacy
of a parametric model with gamma frailty is to compare the observed sur-
vivor function with the observable function derived from a frailty model, as
in Example 10.2. In more complex problems, the baseline survivor function
estimated from fitting a Cox model with the same explanatory variables as
the frailty model, can be compared to the observable survivor function for
an individual with all explanatory variables set to zero. For other choices of
frailty distribution the observable survivor function can only be determined
numerically, and so this approach is not so useful.
In frailty modelling, the validity of the chosen frailty distribution may
also be critically examined. Two particular distributional models have been
described in this chapter, the gamma and lognormal distributions, but there
are a number of other possibilities. Which of these frailty distributions is
chosen is often guided by mathematical convenience and the availability of
statistical software. One approach to discriminating between alternative frailty
distributions is to consider a general family that includes specific distributions
as special cases. The power variance function (PVF) distribution is widely
used in this context. This distribution has a density function that is a function
of three parameters, α, δ and θ, and whose mean and variance are given by
E (Z) = δθα−1 , var (Z) = δ(1 − α)θα−2 ,
for θ > 0, 0 < α 6 1, δ > 0. Setting δ = θ1−α gives a distribution with unit
mean and variance (1 − α)/θ. When α = 0 the distribution reduces to a
gamma density with variance θ−1 and when α = 0.5 to an inverse Gaussian
distribution. Tests of hypotheses about the value of α can then inform model
choice.
Another approach to model checking is to assess the sensitivity of key
model-based inferences to the specific choice of frailty model. Comparing haz-
ard ratios from a fully parametric models with those from a Cox regression
model that has the same frailty distribution can also be valuable.
378 FRAILTY MODELS
10.8.2 Correlated frailty models
An important extension of shared frailty models is to the situation where the
frailties of individuals within a group are not identical as in the shared frailty
model, but are merely correlated. Such models are particularly relevant when
interest centres on the association between event times, as might be the case
when studying event times of paired organs, or amongst twins for example. In
the bivariate frailty case, correlated frailty can be modelled by extending the
model in Equation (10.30) to have a separate frailty for each member of the
pair. Then, in the ith pair, the hazard function is
hij (t) = zij exp(β ′ xij )h0 (t),
for j = 1, 2, and a bivariate distribution is adopted for the corresponding

frailty random variables (Zi1 , Zi2 ). In the case of lognormal frailties, a bivariate
normal distribution could be assumed for the corresponding random effects
Uij = log(Zij ) in the linear component of the model, with corr (Ui1 , Ui2 ) = ρ.
This correlation need not be the same for each group. The extent of this
correlation can then be evaluated. Such models may be fitted using likelihood
or penalised likelihood methods, but can also be fitted using Monte Carlo
Markov Chain (MCMC) methods.
10.8.3 Dependence measures

When a shared frailty model is being used to account for association between
the event times of individuals within a group, a large frailty variance corre-
sponds to a high degree of dependence between the times. In this sense, the
frailty variance is a measure of dependence, and the extent of this depen-
dence can be compared in alternative models. For example, on adding certain
covariates, the frailty effect may diminish and so a reduction in the frailty vari-
ance would indicate that the revised model has been successful in reducing
unexplained variation.
Other measures of dependence in parametric models can be based on corre-
lation. In the special case of gamma frailty distributions, Kendall’s coefficient
of rank correlation, τ , is a useful measure of association between a pair of
survival times. Kendall’s τ is simply related to the frailty variance θ−1 , and is
found from τ = (1 + 2θ)−1 . The advantage of this is that it yields a summary
statistic of dependence in the range (0, 1).
10.8.4 Numerical problems in model fitting

Frailty models are generally fitted using a combination of numerical integra-
tion and optimisation. For this reason, it is important to specify suitable initial
values for the unknown parameters. These are best obtained from fitting the
corresponding model without frailty, and using a small value for the initial
frailty variance. Even then, numerical procedures can lead to computational
FURTHER READING 379
problems in the fitting process, and instability in the resulting estimates. This
is particularly the case when the parameters being estimated have very dif-
ferent scales. For instance, in Example 10.3, estimates of the Weibull shape
parameter, λ, are much smaller than estimates of the other parameters, leading
to convergence problems. Such difficulties can often be overcome by rescaling
certain parameters, in this case by working with λ′ = 10000λ, or by repa-
rameterising the model. Variance parameters such as σu2 in a normal random
effect may be recoded by estimating ω = log σu2 , and writing exp(ω) in place
of σu2 in the model. If changes such as these are not successful, some changes
may be needed to the settings used in the numerical integration process, or
the criteria that define convergence of the optimisation process.

Hougaard (2000) includes an extensive discussion of frailty models, and a num-
ber of illustrative examples. More recently, the books of Wienke (2011) and
Duchateau and Janssen (2008) both give thorough accounts of frailty mod-
elling. General review papers include those of Aalen (1994), Aalen (1998),
and Hougaard (1995). General texts on mixed models include Brown and
Prescott (2000), Demidenko (2013), West, Welch and Galecki (2007) and
Stroup (2013). Texts that include accounts of restricted maximum likelihood
estimation (REML) include those of McCulloch, Searle and Neuhaus (2008)
and Searle, Casella and McCulloch (2006)
McGilchrist and Aisbett (1991) describe how a Cox model with normal
random effects can be fitted and illustrate this with a much discussed exam-
ple on times to the recurrence of infection following insertion of a catheter in
patients on dialysis. Their approach is generalised in Ripatti and Palmgren
(2000) who show how random frailty effects that have a multivariate distri-
bution can be fitted using penalised partial log-likelihood methods. Therneau
and Grambsch (2000) include a chapter on frailty models and describe how
penalised partial log-likelihood methods can be used to fit Cox models with
frailty and to test hypotheses about parameters in the fitted models; see also
Therneau, Grambsch and Pankratz (2003). An alternative fitting method is
based on the expectation-maximization (EM) algorithm, described by Klein
(1992) and Klein and Moeschberger (2005).
Instead of comparing Cox regression models with frailty using the max-
imised marginal log-likelihood, Wald tests can be used. This approach has
been described by Therneau and Grambsch (2000), and uses the notion of
generalised degrees of freedom, described by Gray (1992). The result that
the change in −2 log L̂ on adding a frailty effect to a parametric model has
an asymptotic 0.5(χ20 + χ21 ) distribution was given in Claeskens, Nguti and
Janssen (2008).
Lambert et al. (2004) and Nelson et al. (2006) show how the probability
integral transformation can be used to fit parametric accelerated failure time
models with non-normal random frailty effects. This allows other distributions
380 FRAILTY MODELS
to be adopted when using computer software designed for fitting models with
normal random effects.
There has been relatively little published on methods for checking the
adequacy of frailty models, with most emphasis on testing the validity of a
gamma frailty. One of the first papers to develop a test for the assumption
of a gamma frailty distribution was Shih and Louis (1995), and a goodness of
fit test for use with bivariate survival data was given by Shih (1998). Some
extensions are described by Glidden (1999). More recently a formal goodness
of fit test is given in Geerdens, Claeskens and Janssen (2013), who also briefly
review other work in this area.
Chapter 11
Non-proportional hazards and

institutional comparisons
Proportional hazards models are widely used in modelling survival data in

medical research and other application areas. This assumption of propor-
tionality means that the effect of an explanatory variable or factor on the
hazard of an event occurring does not depend on time. This can be quite a
strong assumption, and situations frequently arise where this assumption is
untenable. Although models that do not assume proportional hazards have
been described in earlier chapters, some additional methods are described in
this chapter. A particularly important application of non-proportional haz-
ards modelling is the comparison of survival rates between institutions, and
so methods that can be used in this context are also described.
11.1 Non-proportional hazards

Models that do not require the assumption of proportional hazards include the
accelerated failure time model and the proportional odds model introduced
in Chapter 6, and the Cox regression model that includes a time-dependent
variable, described in Chapter 8. But often we are faced with a situation where
the assumption of proportional hazards cannot be made, and yet none of the
above models is satisfactory.
As an illustration, consider a study to compare a surgical procedure with
chemotherapy in the treatment of a particular form of cancer. Suppose that
the survivor functions under the two treatments are as shown in Figure 11.1,
where the time-scale is in years. Clearly the hazards are non-proportional.
Death at an early stage may be experienced by patients on the surgical treat-
ment, as a result of patients not being able to withstand the surgery or compli-
cations arising from it. In the longer term, patients who have recovered from
the surgery have a better prognosis. A similar situation arises when an aggres-
sive form of chemotherapy is compared to a standard. Here also, a long-term
advantage to the aggressive treatment may be at the expense of short-term
excess mortality.
One approach, which is useful in the analysis of data arising from situations
such as these, is to define the end-point of the study to be survival beyond
381
382 NON-PROPORTIONAL HAZARDS
1
Survivor function
0
0 1 2
Time
Figure 11.1 Long-term advantage of surgery (—) over chemotherapy (·······).
some particular time. For example, in the study leading to the survivor func-
tions illustrated in Figure 11.1, the treatment difference is roughly constant
after two years. The dependence of the probability of survival beyond two
years on prognostic variables and treatment might therefore be modelled. This
approach was discussed in connection with the analysis of interval-censored
survival data in Section 9.2. As shown in that section, there are advantages
in using a linear model for the complementary log-log transformation of the
survival probability. In particular, the coefficients of the explanatory variables
in the linear component of the model can be interpreted as logarithms of haz-
ard ratios. The disadvantages of this approach are that all patients must be
followed until the point in time when the survival rates are to be analysed,
and that the death data cannot be used until this time. Moreover, faith in
the long-term benefits of one or other of the two treatments will be needed to
ensure that the trial is not stopped early because of excess mortality in one
treatment group.
Strictly speaking, an analysis based on the survival probability at a partic-
ular time is only valid when that time is specified at the outset of the study,
which may be difficult to do. If the data are used to suggest end-points such
as the probability of survival beyond two years, some caution will be needed
in interpreting the results of a significance test.
In the study that leads to the survivor functions shown in Figure 11.1,
it is clear that an analysis of the two-year survival rate will be appropriate.
Now consider a study to compare the use of chemotherapy in addition to
surgery with surgery alone, in which the survivor functions are as shown in
Figure 11.2. Here, the short-term benefit of the chemotherapy may certainly be
STRATIFIED PROPORTIONAL HAZARDS MODELS 383
Survivor function 1
0
0 1 2
Time
Figure 11.2 Short-term advantage of chemotherapy and surgery (—) over surgery
alone (·······).
worthwhile, but an analysis of the two-year survival rates will fail to establish
a treatment difference. The fact that the difference between the two survival
rates is not constant makes it difficult to use an analysis based on survival
rates at a given time. However, it might be reasonable to assume that the
hazards are proportional over the first year of the study, and to carry out a
survival analysis at that time.
11.2 Stratified proportional hazards models

A situation that sometimes occurs is that hazards are not proportional on
an overall basis, but that they are proportional in different subgroups of the
data. For example, consider a situation in which a new drug is being compared
with a standard in the treatment of a particular disease. If the study involves
two participating centres, it is possible that in each centre the new treatment
halves the hazard of death, but that the hazard functions for the standard drug
differ between centres. Then, the hazards between centres for individuals on
a given drug are not proportional. This situation is illustrated in Figure 11.3.
In problems of this kind, it may be assumed that patients in each of the sub-
groups, or strata, have a different baseline hazard function, but that all other
explanatory variables satisfy the proportional hazards assumption within each
stratum. Suppose that the patients in the jth stratum have a baseline hazard
function h0j (t), for j = 1, 2, . . . , g, where g is the number of strata. The
effect of explanatory variables on the hazard function can then be represented
by a proportional hazards model for hij (t), the hazard function for the ith
Hazard function
Hazard function
Standard
Standard
New
New
Time Time
Centre A Centre B
Figure 11.3 Hazard functions for individuals on a new drug (—) and a standard
drug (·······) in two centres.
individual in the jth stratum, where i = 1, 2, . . . , nj , say, and nj is the number

of individuals in the jth stratum. We then have the stratified proportional
hazards model, according to which
hij (t) = exp(β ′ xij )h0j (t),
where xij is the vector of values of p explanatory variables, X1 , X2 , . . . , Xp ,

recorded on the ith individual in the jth stratum. This model was introduced
in connection with the risk adjusted survivor function, described in Section
3.11.1 of Chapter 3.
As an example of this model, consider the particular case where there are
two treatments being compared in each of g centres, and no other explanatory
variables. Let xij be the value of an indicator variable X, which is zero if the
ith patient in the jth centre is on the standard treatment and unity if on the
new treatment. The hazard function for this individual is then
hij (t) = eβxij h0j (t).
On fitting this model, the estimated value of β is the log-hazard ratio for a
patient on the new treatment, relative to one on the standard, in each centre.
This model for stratified proportional hazards is easily fitted using stan-
dard software packages for survival analysis, and nested models can be com-
pared using the −2 log L̂ statistic. Apart from the fact that the stratifying
variable cannot be included in the linear part of the model, no new principles
are involved. When two or more groups of survival data are being compared,
the stratified proportional hazards model is in fact equivalent to the stratified
log-rank test described in Section 2.8 of Chapter 2.
11.2.1 Non-proportional hazards between treatments
If there are non-proportional hazards between two treatments, misleading in-
ferences can result from ignoring this phenomenon. To illustrate this point,
suppose that the hazard function for two groups of individuals, on a new and
standard treatment, are as shown in Figure 11.4 (i). If a proportional hazards
model were fitted, the resulting fitted hazard functions are likely to be as
shown in Figure 11.4 (ii). Incorrect conclusions would then be drawn about
the relative merit of the two treatments.
Hazard function
Hazard function
0 1 2 0 1 2
Time Time
Figure 11.4 (i) Non-proportional hazards and (ii) the result of fitting a propor-
tional hazards model for individuals on a new treatment (—) and a standard treat-
ment (·······).
Non-proportional hazards between treatments can be modelled by assum-

ing proportional hazards in a series of consecutive time intervals. This is
achieved using a piecewise Cox model, which is analogous to the piecewise ex-
ponential model introduced in Chapter 6. To illustrate the use of the model,
suppose that the time period over which the hazard functions in Figure 11.4
are given is divided into three intervals, namely (0, t1 ], (t1 , t2 ] and (t2 , t3 ].
Within each of these intervals, hazards might be assumed to be proportional.
Now let X be an indicator variable associated with the two treatments,
where X = 0 if an individual is on the standard treatment and X = 1 if an
individual is on the new treatment. The piecewise Cox regression model can
then be fitted by defining two time-dependent variables, X2 (t) and X3 (t), say,
which are as follows:
{
1 if t ∈ (t1 , t2 ] and X = 1,
X2 (t) =
0 otherwise;
{
1 if t ∈ (t2 , t3 ] and X = 1,
X3 (t) =
0 otherwise.
In the absence of other explanatory variables, the model for the hazard func-
tion for the ith individual at time t can be written as
hi (t) = exp{β1 xi + β2 x2i (t) + β3 x3i (t)}h0 (t), (11.1)
where xi is the value of X for the ith individual, and x2i (t) and x3i (t) are
the values of the two time-dependent variables for the ith individual at t.
Under this model, the log-hazard ratio for an individual on the new treatment,
relative to one on the standard, is then β1 for t ∈ (0, t1 ], β1 + β2 for t ∈ (t1 , t2 ]
and β1 + β3 for t ∈ (t2 , t3 ]. This model can be fitted in the manner described
in Chapter 8.
The model in Equation (11.1) allows the assumption of proportional haz-
ards to be tested by adding the variables x2i (t) and x3i (t) to the model that
contains xi alone. A significant decrease in the value of the −2 log L̂ statistic
would indicate that the hazard ratio for the new treatment (X = 1) relative
to the standard (X = 0) was not constant. An equivalent formulation of the
model in Equation (11.1) is obtained by defining x1i (t) to be the value of
{
1 if t ∈ (0, t1 ] and X = 1,
X1 (t) =
0 otherwise,
for the ith individual, and fitting a model containing x1i (t), x2i (t) and x3i (t).
The coefficients of the three time-dependent variables in this model are then
the log-hazard ratios for the new treatment relative to the standard in each of
the three intervals. Confidence intervals for the hazard ratio can be obtained
directly from this version of the model.
Example 11.1 Survival times of patients with gastric cancer

In a randomised controlled trial carried out by the Gastrointestinal Tumor
Study Group, 90 patients with locally advanced gastric carcinoma were re-
cruited. They were randomised to receive chemotherapy alone consisting of a
combination of 5-fluorouracil and 1-(2-chloroethyl)-3-(4-methylcyclohexyl)-1-
nitrosoureamethyl (methyl CCNU), or a combination of the same chemother-
apy treatment with external radiotherapy of 5000 rad. Patients were followed
up for over eight years, and the outcome of interest is the number of days
from randomisation to death from gastric cancer. The data, as reported by
Stablein and Koutrouvelis (1985), are shown in Table 11.1.
Kaplan-Meier estimates of the survivor functions for the patients in each
treatment group are shown in Figure 11.5. This figure shows that over the first
two years, patients on the combined chemotherapy and radiotherapy treat-
ment have lower survival rates than those on chemotherapy alone. However,
in the longer term, the combined therapy is more advantageous. The two lines
in the log-cumulative hazard plot in Figure 11.6 are not parallel, again con-
firming that the treatment effect varies over time. If this aspect of the data
had been ignored, and a Cox regression model fitted, the estimated hazard ra-
tio for the combined treatment, relative to the chemotherapy treatment alone,
would have been 1.11, and the P -value for the change in the −2 log L̂ statistic
Table 11.1 Survival times of gastric cancer patients.

Chemotherapy alone Chemotherapy and Radiotherapy
1 383 748 17 185 542
63 383 778 42 193 567
105 388 786 44 195 577
129 394 797 48 197 580
182 408 955 60 208 795
216 460 968 72 234 855
250 489 1000 74 235 1366
262 499 1245 95 254 1577
301 523 1271 103 307 2060
301 524 1420 108 315 2412∗
342 535 1551 122 401 2486∗
354 562 1694 144 445 2796∗
356 569 2363 167 464 2802∗
358 675 2754∗ 170 484 2934∗
∗
380 676 2950 183 528 2988∗
1.0
0.8
0.6
0.4
0.2
0.0
0 500 1000 1500 2000 2500
Survival time
Figure 11.5 Estimates of the survivor functions for gastric cancer patients on
chemotherapy alone (—) or a combination of chemotherapy and radiotherapy (·······).
0
-1
-2
-3
-4
-5
0 1 2 3 4 5 6 7 8
Log survival time
Figure 11.6 Log-cumulative hazard plot for gastric cancer patients on chemotherapy
alone (—) or a combination of chemotherapy and radiotherapy (·······).
on adding the treatment effect to a null model is 0.64. We would then have
concluded there was no evidence of a treatment effect, but since the hazards
of death on the two treatments are not proportional, this analysis is incorrect.
A more appropriate summary of the treatment effect is obtained using a
piecewise Cox regression model, where the treatment effect is assumed con-
stant in each of a number of separate intervals, but differs across the intervals.
Four time intervals will be used in this analysis, namely 1–360, 361–720, 721–
1080 and 1081– days. A time-dependent treatment effect is then set up by
defining four variables, X1 (t), X2 (t), X3 (t), X4 (t), where Xj (t) = 1 when t is
within the jth time interval for a patient on the combined treatment, and
zero otherwise, for j = 1, 2, 3, 4. This is equivalent to fitting an interaction
between a time-dependent variable associated with the four intervals and the
treatments effect. If xji (t) is the value of the jth variable for the ith individual
at time t, the Cox regression model for the hazard of death at time t is
hi (t) = exp {β1 x1i (t) + β2 x2i (t) + β3 x3i (t) + β4 x4i (t)} h0 (t),
where h0 (t) is the baseline hazard function. In this model, the four β-
coefficients are the log-hazard ratios for the combined treatment relative to
chemotherapy alone in the four intervals, and h0 (t) is the hazard function for
a patient on the chemotherapy treatment.
On fitting all four time-dependent variables, the value of −2 log L̂ is
602.372. For the model in which a treatment effect alone is fitted, on the
assumption of proportional hazards, the hazard of death for the ith patient
at time t is hi (t) = exp(βxi )h0 (t), where xi is the value of a variable X
RESTRICTED MEAN SURVIVAL 389
for the ith patient, where X = 0 for a patient on the chemotherapy treat-
ment and X = 1 for a patient on the combined treatment. On fitting this
model, −2 log L̂ = 614.946. The two models are nested and the difference in
their −2 log L̂ values provides a test of proportional hazards. This difference
of 12.57 on 3 d.f. is highly significant, P = 0.005, confirming that there is
clear evidence of non-proportionality in the hazard ratio for the treatment ef-
fect. On fitting the four time-dependent variables, the hazard ratios and 95%
confidence intervals are as shown in Table 11.2.
Table 11.2 Hazard ratios and their corresponding 95% con-

fidence limits on fitting a piecewise Cox regression model.
Time interval Hazard ratio 95% confidence limits
1–360 2.40 (1.25, 4.64)
361–720 0.78 (0.34, 1.76)
721–1080 0.33 (0.07, 1.60)
1081– 0.31 (0.08, 1.24)
This table summarises how the treatment effect varies over the four time
intervals. In the first year, patients on the combined treatment have over
twice the risk of death at any time, compared to those on the chemotherapy
treatment alone. In subsequent years, patients on the combined treatment have
a reduced hazard of death, although the three interval estimates all include
unity, suggesting that these hazard ratios are not significantly different from 1.
11.3 ∗ Restricted mean survival

When hazards are not proportional, the hazard ratio is not an appropriate
summary of the effect of a covariate or a treatment effect. This is because the
hazard ratio is time-dependent, and so a single estimate of this ratio may be
misleading. An alternative summary measure, that has the advantage of being
straightforward to interpret, is the mean survival time to some predetermined
time point, called the restricted mean survival time.
To define the restricted mean survival time, suppose that T is a random
variable associated with a survival time. The restricted mean survival time
to t0 , µ(t0 ), is then the expected value of the minimum of T and t0 over the
follow-up period, so that µ(t0 ) = E (min{T, t0 }).
Now,
E (min{T, t0 }) = E (T ; T 6 t0 ) + t0 P(T > t0 ),
and ∫ t0
E (T ; T 6 t0 ) = uf (u) du,
0
when T has a parametric distribution with density f (t). Integrating by parts,
∫ t0 ∫ t0 ∫ t0
uf (u) du = uF (u) |t00 − F (u) du = t0 F (t0 ) − F (u) du,
0 0 0
and in terms of the survivor function, S(t) = 1 − F (t),
∫ t0 ∫ t0
uf (u) du = t0 {1 − S(t0 )} − {1 − S(u)} du,
0 0
so that ∫ ∫
t0 t0
uf (u) du = S(u) du − t0 S(t0 ).
0 0
Finally, since t0 P(T > t0 ) = t0 S(t0 ),

∫ t0
E (min{T, t0 }) = S(u) du.
0
The restricted mean survival time is therefore the area under the estimated
survivor function to t0 , and is an easily understood summary statistic. For
example, if time is measured in months, µ(24) is the average number of months
survived over a 24-month period, and so is the two-year life expectancy. This
statistic can also be used to summarise the effect of a treatment parameter or
other explanatory variables on life expectancy over a defined period of time.
The restricted mean survival can be determined from the Kaplan-Meier
estimate of the survivor function. For example, the estimated restricted mean
at the rth ordered event time, t(r) , is
∑
r
µ̂(t(r) ) = Ŝ(t(j) )(t(j) − t(j−1) ),
j=1
where Ŝ(t(j) ) is the Kaplan-Meier estimate of the survivor function at the

jth event time, t(j) , and t(0) is defined to be zero. The standard error of this
estimate is given by
  12
∑
r−1
 A2j /{nj Ŝ(t(j) )} ,
j=1
where
∑
r−1
Aj = Ŝ(t(i) )(t(i+1) − t(i) ).
i=j
A minor modification to these results is needed when an estimate of restricted

survival is required at a time that is not actually an event time.
When there are two groups of survival times, as in a clinical trial to com-
pare two treatments, and non-proportional hazards is anticipated or found,
the restricted mean survival time can be determined from the area under the
Kaplan-Meier estimates of the two survivor functions. The difference in the
estimated restricted mean survival to a given time can then be used as an
unadjusted summary measure of the overall treatment effect.
RESTRICTED MEAN SURVIVAL 391
More generally, consider the situation where there is non-proportional haz-
ards between two treatment groups and where information on the values of
other explanatory variables is available. Here, parametric models with differ-
ent underlying hazards could be used, such as a Weibull model with a different
shape parameter for each treatment group. The survivor functions under such
a model are fully parameterised, and so the restricted mean survival time
can be estimated by integrating the fitted survivor function, analytically or
numerically.
A greater degree of flexibility in modelling the underlying baseline hazard
function is achieved by using the Royston and Parmar model described in
Section 6.9 of Chapter 6. By adding interaction terms between the treatment
factor and the cubic spline functions that define the baseline hazard, different
underlying hazard functions for each treatment group can be fitted. Again,
the restricted mean survival time can subsequently be obtained by integrating
the fitted survivor functions.
11.3.1 Use of pseudo-values

A versatile, and straightforward way of modelling the dependence of the re-
stricted mean survival on explanatory variables is based on pseudo-values.
These values are formed from the differences between an estimate of some
quantity obtained from a complete set of n individuals, and that obtained on
omitting the ith, for i = 1, 2, . . . , n. Standard regression models are then used
to model the dependence of the pseudo-values on explanatory variables.
To obtain the pseudo-values for the restricted mean, the Kaplan-Meier es-
timate of the survivor function for the complete set of survival times is first
determined, from which the restricted mean survival at some value t0 is ob-
tained, µ̂(t0 ), say. Then, each observation is omitted in turn, and the restricted
mean survival is estimated from the reduced data set, to give µ̂(−i) (t0 ), for
i = 1, 2, . . . , n. The ith pseudo-value is then
zi = nµ̂(t0 ) − (n − 1)µ̂(−i) (t0 ), (11.2)
for i = 1, 2, . . . , n. This measures the contribution made by the ith individual

to the estimated restricted mean survival time at t0 , and is defined irrespective
of whether or not the observed survival time for that individual is censored.
The ith pseudo-value, zi , is then assumed to be the observed value of
a random variable Zi , and a generalised linear model may then be used to
model the dependence of the expected value of Zi on treatment factors and
explanatory variables. A natural choice would be to assume that Zi is normally
distributed, with a log-linear model for its mean, E (Zi ). Then, log E (Zi ) =
β0 + β ′ xi , where β0 is a constant and xi is the vector of values of explanatory
variables for the ith individual. A linear regression model for E (Zi ) will often
give similar results.
In practical applications, the value to use for t0 has to be determined in
advance of the analysis. When the occurrence of an event over a particular
time period is of interest, this time period will determine the value of t0 . For
example, if the average time survived over a one-year period following the
diagnosis of a particular form of cancer is of interest, t0 would be 365 days.
In other cases, t0 will be usually be taken to be close to the longest observed
event time in the data set.
Example 11.2 Survival times of patients with gastric cancer

An alternative analysis of the data on survival times of patients suffering from
gastric cancer, given in Example 11.1, is based on the restricted mean. The
restricted mean survival will be calculated at 5 years, obtained as the area
under the Kaplan-Meier estimate of the survivor function for patients in each
treatment group up to t = 1826 days. This is found to be 661.31 (se = 74.45)
for patients on the single treatment and 571.89 (se = 94.04) for those on the
combined treatment. This means that over a five-year period, patients are
expected to survive for 1.8 years on average when on chemotherapy alone,
and 1.6 years when on the combined treatment. The 95% confidence inter-
vals for the restricted mean survival time to 1826 days are (515.39, 807.23)
and (387.57, 756.21), respectively, for the two treatment groups. As there is
a substantial overlap between these intervals, there is no evidence that the
restricted mean survival time differs between the two groups.
We next illustrate the use of pseudo-values in analysing these data. First,
the Kaplan-Meier estimate of the survivor function for the complete data set is
obtained, and from this the estimated restricted mean to 1826 days is 616.60
days. This is an overall estimate of the average patient survival time over
this time period. Each of the 90 observations is then omitted in turn, and
the Kaplan-Meier estimate recalculated from each set of 89 observations. The
restricted mean to 1826 days is then estimated for each of these 90 sets of
data, and Equation (11.2) is then used to obtain the 90 pseudo-values. These
represent the contribution of each observation to the overall estimate of the
five-year restricted mean.
For the data in Table 11.1, the pseudo-values are equal to the observed
survival times for patients that die before 1826 days, while for the 10 patients
that survive beyond that time, the pseudo-values are all equal to 1826. This
pattern in the pseudo-values is due to the censored survival times being the
longest observed follow-up times in the data set.
The next step is to model the dependence of the pseudo-values on the
treatment effect. This can be done using a log-linear model for the expected
value of the ith pseudo-observation, E (Zi ), i = 1, 2, . . . , n, where log E (Zi ) =
β0 +β1 xi , and xi = 0 if the ith patient is on the chemotherapy treatment alone
and xi = 1 otherwise. On adding the treatment effect to the log-linear model,
the deviance is reduced from 28591297 on 89 d.f. to 28411380 on 88 d.f., and
so the F -ratio for testing whether there is a treatment effect is (28591297 −
28411380)/{28411380/88} = 0.56, which is not significant as an F -statistic on
1, 88 d.f. (P = 0.46). Much the same result is obtained using a linear model for
E(Zi ), which is equivalent to using a two-sample t-test to compare the means
INSTITUTIONAL COMPARISONS 393
of the pseudo-values in each treatment group. Patient survival over 1826 days
is therefore unaffected by the addition of radiotherapy to the chemotherapy
treatment.
In summary, the data in Table 11.1 indicate that there is a longer-term
benefit for patients on the combination of chemotherapy and radiotherapy, so
long as they survive the first 18 months or so where there is an increased risk
of death on this treatment. However, over the five-year follow-up period, on
average there is no survival benefit from being on the combined treatment.
11.4 Institutional comparisons

The need for accountability in the health service has led to the introduction
of a range of performance indicators that measure the quality of care pro-
vided by different institutions. One of the key measures is the risk adjusted
mortality rate, which summarises the mortality experienced by patients in
different healthcare institutions in a way that takes account of differences in
the characteristics of patients being treated. Statistics such as these provide a
means of comparing institutional performance on an equal footing. Analogous
measures can be defined for survival rates, recovery rates and infection rates,
and the methods that are described in this section apply equally to compar-
isons between other types of organisation, such as schools, universities and
providers of financial services. In this section, we describe and illustrate how
point and interval estimates for a risk adjusted failure rate, RAFR, can be
obtained from survival data.
The RAFR is an estimate of
Observed failure rate
× Overall failure rate, (11.3)
Expected failure rate
at a given time, where the observed failure rate is obtained from the Kaplan-
Meier estimate of the survivor function for a specific institution at a given
time, and the overall failure rate at that time is estimated from the survivor
function fitted to the data across all institutions, ignoring differences between
them. The expected failure rate for an institution can be estimated from the
risk adjusted survivor function, defined in Section 3.11 of Chapter 3, which is
the average of the estimated survivor functions for individuals within an insti-
tution, at a given time, based on a risk adjustment model. Estimates of each
failure rate at a specified time are obtained by subtracting the corresponding
value of the estimated survivor function from unity. Once the RAFR has been
obtained, the analogous risk adjusted survival rate, or RASR, can simply be
found from RASR = 1− RAFR.
Example 11.3 Comparisons between kidney transplant centres

The methods of this section will be illustrated using data on the transplant
survival rates experienced by recipients of organs from deceased donors in eight
kidney transplant centres in the UK, in the three-year period from January
2009 to December 2011. There are a number of factors that may affect the
centre specific survival rates, and in this illustration, account is taken of donor
age, an indicator of whether the donor is deceased following brain death (DBD
donor) or circulatory death (DCD donor), recipient age at transplant, diabetic
status of the recipient (absent, present) and the elapsed time between retrieval
of the kidney from the donor and transplantation into the recipient, known as
the cold ischaemic time. Also recorded is the transplant survival time, defined
to be the earlier of graft failure and patient death, and an event indicator that
is zero if the patient was alive with a functioning graft at the last known date
of follow-up, or December 2012 at the latest. The variables are summarised
below.
Patient: Patient identifier
Centre: Transplant centre (1, 2, . . . , 8)
Tsurv: Transplant survival time (days)
Tcens: Event indicator (0 = censored, 1 = transplant failure)
Dage: Donor age (years)
Dtype: Donor type (0 = DBD, 1 = DCD)
Rage: Recipient age (years)
Diab: Diabetic status (0 = absent, 1 = present)
CIT: Cold ischaemic time (hours)
There are 1439 patients in the data set, and data for the first 30 patients
transplanted in the time period are shown in Table 11.3. The transplanted
kidney did not function in patient 7, and so Tsurv = 0 for this patient.
Of particular interest is the transplant failure rate at one year, and so the
eight transplant centres will be compared using this metric. We first obtain
the unadjusted Kaplan-Meier estimate of the survivor function across all 1439
patients, from which the overall one-year failure rate is 1 − 0.904 = 0.096. The
centre-specific one-year failure rates are similarly obtained from the Kaplan-
Meier estimate of the survivor function for patients in each centre. Next,
the risk adjusted survival rate in each centre is calculated using the method
described in Section 3.11.1 of Chapter 3. A Cox regression model that contains
the variables Dage, Dtype, Rage, Diab, and CIT, is fitted to the data, from
which the estimated survivor function for the ith patient in the jth centre is
Ŝij (t) = {Ŝ0 (t)}exp(η̂ij ) , (11.4)
where
η̂ij = 0.023 Dage ij + 0.191 Dtype ij + 0.002 Rage ij − 0.133 Diab ij + 0.016 CIT ij ,
and Ŝ0 (t) is the estimated baseline survivor function. The average survivor
function at time t in the jth centre, j = 1, 2, . . . , 8, is
1 ∑
nj
Ŝj (t) = Ŝij (t),
nj i=1
Table 11.3 Transplant survival times of 30 kidney recipients.

Patient Centre Tsurv Tcens Dage Dtype Rage Diab CIT
1 6 1384 0 63 0 44 0 25.3
2 8 1446 0 17 0 33 0 24.0
3 8 1414 0 34 0 48 0 16.7
4 5 1459 0 36 0 31 0 13.6
5 4 1385 0 63 0 58 0 30.5
6 2 1455 0 40 0 35 0 15.8
7 3 0 1 59 0 46 0 14.9
8 6 1453 0 17 0 38 0 17.4
9 8 236 1 57 1 50 0 18.7
10 8 1445 0 59 0 34 0 15.0
11 8 1428 0 53 0 63 0 13.0
12 4 1454 0 59 0 50 0 13.7
13 4 1372 0 27 1 65 0 19.3
14 4 382 1 27 1 71 0 12.2
15 3 1452 0 47 0 68 0 17.8
16 3 1436 0 29 0 37 0 12.1
17 7 1452 0 28 0 49 0 12.9
18 4 310 1 51 1 62 0 14.3
19 1 1059 1 73 1 62 0 10.6
20 1 1446 0 17 1 32 0 10.3
21 1 1397 0 17 1 51 0 19.0
22 1 1369 0 37 1 46 0 10.4
23 1 1444 0 37 1 48 0 15.1
24 8 1443 0 63 0 53 0 14.8
25 7 1395 0 32 0 60 0 19.9
26 3 1396 0 36 0 36 0 13.4
27 4 1261 0 30 0 48 0 15.1
28 5 1441 0 44 1 50 1 12.4
29 2 1439 0 54 0 27 0 11.6
30 4 1411 0 38 0 26 0 15.9
where nj is the number of patients in the jth centre, and the expected trans-
plant failure rate at one year can then be estimated from this survivor function.
The unadjusted and adjusted one-year survival rates are shown in Table 11.4,
together with the RAFR.
To illustrate the calculation, for Centre 1, the unadjusted and adjusted
one-year survival rates are 0.9138 and 0.8920, respectively, and since overall
unadjusted survival rate is 0.096, the RAFR is
1 − 0.9138
× 0.096 = 0.077.
1 − 0.8920
The corresponding risk adjusted transplant survival rate for this centre is
0.923. Across the 8 centres, the risk adjusted one-year survival rates vary
between 6% and 17%.
Table 11.4 Values of RAFR found from risk adjusted failure rates.
Centre Number of Estimated one-year survival RAFR
patients Unadjusted Adjusted
1 267 0.9138 0.8920 0.0768
2 166 0.8887 0.9084 0.1170
3 148 0.8986 0.9150 0.1148
4 255 0.9254 0.9061 0.0764
5 164 0.9024 0.9002 0.0941
6 160 0.9438 0.9068 0.0581
7 102 0.8922 0.8947 0.0985
8 177 0.8475 0.9128 0.1682
An alternative and more convenient way of calculating the RAFR in Ex-

pression (11.3) is based on an estimate of
Observed number of failures
× Overall failure rate, (11.5)
Expected number of failures
in a given time period. To estimate the RAFR using this definition, we need
an estimate of the expected number of failures in the time period of interest.
From Section 1.3 of Chapter 1, the cumulative hazard of a failure at time
t, H(t), is the expected number of failures in the period to t, given that no
failure has occurred before then. Therefore, for the ith individual in the jth
institution who survives to time tij , Hij (tij ) is the expected number of failures
during their follow-up. Using the result in Equation (1.8) of Chapter 1, the
estimated number of failures in centre j is then
∑
nj
∑
nj
êj = Ĥij (tij ) = − log Ŝij (tij ), (11.6)
i=1 i=1
where Ŝij (tij ) is defined in Equation (11.4).

Transplant outcomes for eight kidney transplant centres, from the data de-
scribed in Example 11.3, are now used to estimate the RAFR using Expres-
sion (11.5). The observed number of transplant failures in the first year in each
centre is first determined. Next, a Cox regression model is fitted to the data
to give an estimate of the survivor function for the ith patient in the jth cen-
tre, Ŝij (t). For patients with Tsurv 6 365, their estimated survivor function
is obtained at their value of Tsurv, while for patients with Tsurv > 365, their
estimated survivor function at t = 365 is calculated. The corresponding esti-
mate of the cumulative hazard is then obtained at the survival time for each
patient, and from Equation (11.6), summing these estimates over the patients
in the jth centre gives an estimate of the expected number of failures in that
centre. These values are shown in Table 11.5. To illustrate the calculation, for
Centre 1, we observe 23 transplant failures in the first year and the estimated
Table 11.5 Values of RAFR found from the observed and expected numbers of
transplant failures.
Centre Number of Number of transplant failures RAFR
patients Observed Expected
1 267 23 28.96 0.0764
2 166 18 14.93 0.1160
3 148 15 12.48 0.1157
4 255 19 24.23 0.0755
5 164 16 16.68 0.0923
6 160 9 15.38 0.0563
7 102 11 10.67 0.0992
8 177 27 14.66 0.1772
expected number is 28.96. The overall failure rate is 0.096, and so the RAFR
for this centre is
23
RAFR = × 0.096 = 0.076.
28.96
The RAFR values in this table are very similar to those given in Table 11.4.
11.4.1 ∗ Interval estimate for the RAFR

To obtain an interval estimate for the risk adjusted failure rate, we use the
definition in Expression (11.5), and assume that the expected number of fail-
ures and the overall failure rate are estimated with negligible error. This will
often be the case, as these quantities will have been obtained from a large data
set covering all the institutions. Now let Yj be the random variable associated
with the number of failures out of nj in the jth institution in a given time
period, and let yj be the corresponding observed value. Also let pj be the
probability of failure in the jth institution, so that Yj has a binomial distri-
bution. Generally, nj will be sufficiently large for the Poisson approximation
to the binomial distribution to be valid, according to which a binomial distri-
bution with parameters nj , pj tends to a Poisson distribution with mean µj
as nj → ∞ and pj → 0, whilst µj = nj pj remains finite. Then, the random
variable associated with the number of failures in the jth institution, Yj , has
a Poisson distribution with mean µj . Note that this result is only valid when
the failure probability is small, as it will be when failures are not too frequent.
Also, this result means that we do not need to dwell on the value of nj to
use in the following calculations, which is particularly useful when individuals
may have censored survival times.
Approximate interval estimates for the number of deaths in the jth insti-
tution can be obtained from the result that log Yj is approximately normally
distributed with mean log µj and variance given by var (log Yj ) ≈ µ−2 j var (Yj ),
using the result in Equation (2.8) of Chapter 2. Now, the variance of a Poisson
random variable is equal to its mean, so that var (Yj ) = µj , and the variance
of log Yj is approximately 1/µj . Since the best estimate of µj is simply the
observed number of failures, yj , this variance can be estimated by 1/yj . It
then follows that a 95% interval
√ estimate for the number of failures is the
interval exp{log yj ± 1.96/ (yj )}.
It is also possible to calculate ‘exact’ Poisson limits for the number of
failures in a particular institution. Suppose that we observe a value y of a
random variable Y , which has a Poisson distribution with mean µ. The lower
limit of a 95% interval for µ is then yL , which is such that P(Y > y) = 0.025
when µ = yL . Similarly, the upper limit of the 95% interval for µ is the value
yU such that P(Y 6 y) = 0.025 when µ = yU . These limits can be shown to
have desirable optimality properties.
To calculate these exact limits, we can use the general result that if Y
has a Poisson distribution with mean µ, and X has a gamma distribution
with shape parameter y + 1 and unit scale parameter, that is Y ∼ P (µ) and
X ∼ Γ(y + 1, 1), then P (Y 6 y) = 1 − P (X 6 µ). This means that
∑
y ∫
e−µ µk µ
e−x xy
=1− dx, (11.7)
k! 0 Γ(y + 1)
k=0
where Γ(y + 1) = y! is a gamma function. The gamma distribution was in-

troduced in Section 6.1.3 of Chapter 6, and this result can be verified using
integration by parts.
The lower limit of a 95% interval for the number of events, yL , is the
expected value of a Poisson random variable, Y , which is such that P(Y > y) =
0.025. Adapting the result in Equation (11.7),
∑
y−1 ∫
e−yL yL k yL
e−x xy−1
P(Y > y) = 1 − P(Y 6 y − 1) = 1 − = dx,
k! 0 Γ(y)
k=0
and so yL is such that

∫ yL
e−x xy−1
dx = 0.025.
0 Γ(y)
This means that yL is the lower 2.5% point of a gamma random variable
with shape parameter y and unit scale parameter, and so can be obtained
from the inverse cumulative distribution function for the gamma distribution.
Similarly, the upper limit of the 95% confidence interval, yU , is the expected
value of a Poisson random variable for which P(Y 6 y) = 0.025. Again using
Equation (11.7),
∑
y ∫
e−yU yU k yU
e−x xy
P(Y 6 y) = =1− dx,
k! 0 Γ(y + 1)
k=0
so that ∫ yU
e−x xy
dx = 0.975,
0 Γ(y + 1)
and yU is the upper 2.5% point of a gamma distribution with shape parameter
y + 1 and unit scale parameter.
As an illustration, suppose that the observed number of events in a partic-
ular institution is y = 9, as it is for Centre 6 in the data on kidney transplant
failure rates in Example 11.3. The lower 2.5% point of a gamma distribu-
tion with shape parameter 9 and unit scale parameter is yL = 4.12, and so
P(Y > 9) = 0.025 when Y has a Poisson distribution with mean µ = 4.12.
Also, the upper 2.5% point of a gamma distribution with shape parameter
10 and unit scale parameter is yU = 17.08, and so P(Y 6 9) = 0.025 when
Y has a Poisson distribution with mean 17.08. These two distributions are
shown in Figure 11.7. The tail area to the right of y = 9 in the distribution
with mean 4.12, and the tail area to the left of y = 9 in the distribution with
mean 17.08, are both equal to 0.025. An exact 95% interval estimate for the
observed number of events is then (4.12, 17.08).
0.20
0.15
Probability
0.10
0.05
0.00
0 2 4 6 8 10 12 14 16 18 20 22 24
Value of y
Figure 11.7 Poisson distributions with means 4.12 and 17.09 (- - -) for deriving exact
confidence limits for an observed value of 9 (·······).
Once an interval estimate for the number of failures, (yL , yU ), has been
obtained using either the approximate or exact method, corresponding limits
for the RAFR are
yL yU
× overall failure rate, and × overall failure rate,
ej ej
where here ej is the estimated number of failures in the jth institution, ob-
tained using Equation (11.6), but taken to have negligible error.
For the data on transplant outcomes in eight kidney transplant centres, given
in Example 11.3, approximate and exact 95% confidence limits for the RAFR
Table 11.6 Approximate and exact 95% interval estimates for the RAFR.
Centre RAFR Approximate limits Exact limits
Lower Upper Lower Upper
1 0.0764 0.051 0.115 0.048 0.115
2 0.1160 0.073 0.184 0.069 0.183
3 0.1157 0.070 0.192 0.065 0.191
4 0.0755 0.048 0.118 0.045 0.118
5 0.0923 0.057 0.151 0.053 0.150
6 0.0563 0.029 0.108 0.026 0.107
7 0.0992 0.055 0.179 0.050 0.178
8 0.1772 0.122 0.258 0.117 0.258
To illustrate how these interval estimates are calculated, consider the data
for Centre 1, for which y1 = 23 and the corresponding expected number of
deaths is, from Example 11.4, estimated to be 28.96. The standard √ error of
the√logarithm of the number of transplant failures in this centre is 1/ (y1 ) =
1/ (23) = 0.209. A 95% confidence interval for y1 is then exp(log 23 ± 1.96 ×
0.209), that is (15.28, 34.61), and the corresponding interval for the RAFR is
( )
15.28 34.61
× 0.096, × 0.096 ,
28.96 28.96
that is (0.051, 0.115). Exact limits for the number of failures in this centre
can be found from the lower 2.5% point of a Γ(23, 1) random variable and
the upper 2.5% point of a Γ(24, 1) random variable. This leads to the interval
(14.58, 34.51), and corresponding limits for the RAFR are
( )
14.58 34.51
× 0.096, × 0.096 ,
28.96 28.96
that is (0.048, 0.115).

Table 11.6 shows that there is very good agreement between the approx-
imate and exact limits, as would be expected when the observed number of
failures is no smaller than 9.
11.4.2 ∗ Use of the Poisson regression model

Approximate interval estimates for the RAFR can conveniently be obtained
using a modelling approach. Let Fj be the risk adjusted failure rate and ej
the expected number of failures for centre j, so that
yj
Fj = × overall failure rate.
ej
Since we take the random variable associated with the number of failures in
the jth institution, Yj , to have a Poisson distribution with mean µj , it follows
that
µj
E (Fj ) = × overall failure rate. (11.8)
ej
To model the RAFR for centre j, Fj , we take log E (Fj ) = cj , where cj is the
effect due to the jth institution. Now, using Equation (11.8),
log E (Fj ) = log µj − log{ej /(overall failure rate)},
and this leads to a log-linear model for µj where
log µj = cj + log{ej /(overall failure rate)}. (11.9)
In this model, the term log{ej /(overall failure rate)} is a variate with a known
coefficient of unity, called an offset. When the log-linear model in Equa-
tion (11.9) is fitted to the observed number of failures in each institution,
yj , the model has the same number of unknown parameters as there are ob-
servations. It will therefore be a perfect fit to the data, and so the fitted values
µ̂j will be equal to the observed number of failures. The parameter estimates
ĉj will be the fitted values of log E (Fj ), so that the RAFR for the jth centre is
exp(ĉj ). A 95% confidence interval for the RAFR is then exp{ĉj ±1.96 se (ĉj )},
where ĉj and its standard error can be obtained from computer output from
fitting the log-linear model.

Interval estimates for the RAFR, obtained from fitting the log-linear model
in Equation (11.9) to the data on the number of transplant failures in each of
8 centres, are given in Table 11.7.
Table 11.7 The RAFR and interval estimates from fitting a log-linear model.
Centre ĉj se (ĉj ) RAFR (eĉj ) 95% limits for RAFR
Lower Upper
1 −2.571 0.209 0.0764 0.051 0.115
2 −2.154 0.236 0.1160 0.073 0.184
3 −2.157 0.258 0.1157 0.070 0.192
4 −2.584 0.229 0.0755 0.048 0.118
5 −2.382 0.250 0.0923 0.057 0.151
6 −2.877 0.333 0.0563 0.029 0.108
7 −2.310 0.302 0.0992 0.055 0.179
8 −1.730 0.192 0.1772 0.122 0.258
The RAFR values and their corresponding interval estimates, obtained

from the fitted log-linear model, are in exact agreement with the approximate
interval estimates given in Table 11.6, as they should be.
11.4.3 Random institution effects
The log-linear model for the expected number of events in Equation (11.9)
provides a framework for calculating interval estimates for the RAFR across
institutions. In this model, the parameters associated with institution effects
are fitted as fixed effects, which is entirely appropriate when the number of
institutions is small, as it is in Example 11.3. This methodology can also be
used when there is a much larger number of institutions, but in this situation,
it would be undesirable for institution effects to be incorporated as fixed ef-
fects. Instead, random effects would be used to model the variation between
institutions. Random effects were introduced in Section 10.1.1 of Chapter 10,
in the context of frailty models.
Suppose that the effect due to the jth institution, cj , is drawn from a
normal distribution with mean α and variance σc2 , denoted N (α, σc2 ). Since the
log-linear model for the RAFR in the jth centre is such that log E (Fj ) = cj ,
a lognormal distribution is implied for the variation in the expected RAFR
values, E (Fj ), across the centres. Alternatively, the model in Equation (11.9)
can be written as
log µj = α + cj + log{ej /(overall failure rate)}
where now the cj have a N (0, σc2 ) distribution, and α is the overall value of
the logarithm of the RAFR, so that eα is the overall failure rate.
Using random rather than fixed effects has two consequences. First, the
estimated institution effects, c̃j , are ‘shrunk’ towards the overall rate, α̃, and
the more extreme the RAFR, the greater the shrinkage. Second, interval esti-
mates for institution effects in the random effects model will be shorter than
when fixed effects are used, increasing the precision of prediction for future
patients. Both these features are desirable when using centre rates to guide
patient choice. The concept of shrinkage was referred to in Section 3.7 of
Chapter 3, when describing the lasso method for variable selection.

Random centre effects are now used in modelling the observed numbers of
transplant failures in the eight kidney transplant centres. The observed num-
ber of transplant failures in the jth centre, yj , is modelled by taking yj to be
the observed value of a Poisson random variable with mean µj , where
log µj = cj + log{ej /(overall failure rate)},
and cj ∼ N (α, σc2 ).

The estimate of α is α̃ = −2.341, so that exp(α̃) = 0.096, the same as the
failure rate obtained from the Kaplan-Meier estimate of the overall survivor
function. The estimated variance of the centre effects is σ̃c2 = 0.054 with a
standard error of 0.054, and so there is no evidence of significant between
centre variation. Estimates of the centre effects, c̃j , lead to RAFR estimates,
exp(c̃j ), obtained as estimates of the modal value of the posterior distribution
FURTHER READING 403
of the random effects (see Section 10.4 of Chapter 10), and are shown in
Table 11.8.
Table 11.8 The RAFR and interval estimates using random centre ef-
fects in a log-linear model.
Centre c̃j se (c̃j ) RAFR (ec̃j ) 95% limits for RAFR
Lower Upper
1 −2.471 0.170 0.084 0.061 0.118
2 −2.252 0.186 0.105 0.073 0.151
3 −2.261 0.194 0.104 0.071 0.152
4 −2.468 0.179 0.085 0.060 0.120
5 −2.360 0.181 0.094 0.066 0.135
6 −2.537 0.228 0.079 0.051 0.124
7 −2.330 0.199 0.097 0.066 0.144
8 −2.000 0.232 0.135 0.086 0.214
The estimates of the RAFR values on using a random effects model are
similar to those found with the fixed effects model shown in Table 11.7, but
closer to the overall rate of 0.096. This is particularly noticeable for Centre 8,
which has the largest RAFR. Also, the values of se (c̃j ) are generally smaller,
which in turn means that the corresponding interval estimates are narrower.
These two features illustrate the shrinkage effect caused by using random
centre effects.

Examples of survival analyses in situations where the proportional hazards
model is not applicable have been given by Stablein, Carter and Novak (1981)
and Gore, Pocock and Kerr (1984). Further details on the stratified propor-
tional hazards model can be found in Kalbfleisch and Prentice (2002) and
Lawless (2002), for example. A review of methods for dealing with non-
proportional hazards in the Cox regression model is included in Schemper
(1992). A discussion of strategies for dealing with non-proportional hazards is
also included in Chapter 6 of Therneau and Grambsch (2000).
Royston and Parmar (2011) describe and illustrate how the restricted mean
survival time can be used to summarise a treatment difference in the presence
of non-proportional hazards. The use of average hazard ratios is discussed by
Schemper, Wakounig and Heinze (2009). Andersen, Hansen and Klein (2004)
showed how pseudo-values could be used in modelling the restricted mean,
and Andersen and Perme (2010) review the applications of pseudo-values in
survival analysis. Klein et al. (2008) described a SAS macro and an R package
for the computation of the restricted mean.
Statistical methods for the comparison of institutional performance are de-
scribed and illustrated by Thomas, Longford and Rolph (1994) and Goldstein
and Spiegelhalter (1996). A detailed comparison of surgical outcomes between
a number of hospitals following paediatric cardiac surgery is given by Spiegel-
halter et al. (2002). Sun, Ono and Takeuchi (1996) showed how exact Poisson
limits for a standardised mortality ratio can be obtained using the relationship
between a Poisson distribution and a gamma or chi-squared distribution. Fun-
nel plots, described by Spiegelhalter (2005), provide a visual comparison of
institutional performance. Ohlssen, Sharples and Spiegelhalter (2007) describe
techniques that can be used to identify institutions that perform differently
from others, and Spiegelhalter et al. (2012) provide a comprehensive review
of statistical methods used in healthcare regulation.
Chapter 12
Competing risks
In studies where the outcome is death, individuals may die from one of a
number of different causes. For example, in a study to compare two or more
treatments for prostatic cancer, patients may succumb to a stroke, myocar-
dial infarction or the cancer itself. In some cases, an analysis of death from all
causes may be appropriate, and standard methods for survival analysis can
be used. More commonly, there will be interest in how the hazard of death
from different causes depends on treatment effects and other explanatory vari-
ables. Of course, death from any one of a number of possible causes precludes
its occurrence from any other cause, and this feature has implications for the
analysis of data of this kind. In this chapter, we review methods for summaris-
ing data on survival times for different causes of death, and describe models
for cause-specific survival data.
12.1 Introduction to competing risks

Individuals face death from a number of risks. These risks compete to become
the actual cause of death, which gives rise to the situation known as compet-
ing risks, in which a competing risk prevents the occurrence of an event of
particular interest. More generally, this term applies when an individual may
experience one of a number of different end-points, where the occurrence of
any one of these hinders or eliminates the potential for others to occur.
Data of this type occur in many application areas. For example, in a can-
cer clinical trial, we may be interested in deaths from a specific cancer, and
here events such as a myocardial infarct or stroke are competing risks. A
study involving the exposure of animals to a possible carcinogen may result
in exposed animals dying from different cancers, where each cause of death
is of interest, allowing for competing causes. In outcome studies following
bone marrow transplantation in patients suffering from leukaemia, possible
end-points might be discharged from hospital, relapse, the occurrence of graft
versus host disease or death, with interest centering on how various factors
affect the times to each event, in the presence of other risks.
There may be a number of aims when analysing data where there are
multiple end-points. For example, it may be important to determine which of
a number of factors are associated with a particular end-point, in the presence
405
406 COMPETING RISKS
of competing risks. Estimates of the effect of such factors on the hazard of
death from a particular cause, allowing for possible competing causes, may also
be needed. In other situations, it will be of interest to compare survival times
across different causes of death to identify those causes that lead to earlier
or later failure times. An assessment of the consistency of estimated hazard
ratios for certain factors across different end-points may also be needed. An
example of a data set with multiple end-points follows.
Example 12.1 Survival of liver transplant recipients

A number of conditions lead to failure of the liver, and the only possible treat-
ment is a liver transplant. Transplantation is generally very effective, and the
median survival time for adult transplant recipients is now well over 12 years.
However, following a liver transplant, the graft may fail as a result of acute or
chronic organ rejection, hepatic artery thrombosis, recurrent disease or other
reasons. This example is based on the times from a liver transplant to graft
failure for 1761 adult patients who had a first elective transplant with an organ
from a deceased donor between January 2000 and December 2010, who were
followed up until the end of 2012. These data concern patients transplanted
for three particular liver diseases, primary biliary cirrhosis (PBC), primary
sclerosing cholangitis (PSC) and alcoholic liver disease (ALD). In addition to
the graft survival time, information is given on the age and gender (1 = male,
2 = female) of the patient, their primary liver disease and the cause of graft
failure (0 = functioning graft, 1 = rejection, 2 = thrombosis, 3 = recurrent
disease, 4 = other). Because transplantation is so successful, these data ex-
hibit heavy censoring, and there are only 261 (15%) patients who suffer graft
failure, although an additional 211 die with a functioning graft. The failure
times of this latter group have been taken to be censored at their time of
death. The first 20 observations in the data set are given in Table 12.1.
12.2 Summarising competing risks data
In standard survival analysis, we have observations on a random variable T

associated with the survival time. In addition, there will be an event indica-
tor that denotes whether the end-point has actually occurred or whether the
observed time has been censored. When there are competing risks, the event
indicator is extended to cover the different possible end-points. The resulting
data are therefore a survival time T and a cause C. The data for a given
individual are then observed values of (T, C), and we will write (ti , ci ) for the
data from the ith individual, i = 1, 2, . . . , n, where the possible values of ci
are 0, 1, 2, . . . , m, and ci = 0 when no end-point has been observed.
From the data, we know that the jth of m possible causes, j = 1, 2, . . . , m,
has not occurred in the ith individual before time ti . Death from a particular
cause, the jth, say, may have occurred at time ti , and all other potential
causes could have occurred after ti if cause j had not occurred. As we do
not know which cause might have occurred when, difficulties in estimating a
SUMMARISING COMPETING RISKS DATA 407
Table 12.1 Graft survival times for 20 liver transplant recipients and the cause of
graft failure.
Patient Age Gender Primary disease Time Status Cause of failure
1 55 1 ALD 2906 0 0
2 63 1 PBC 4714 0 0
3 67 1 PBC 4673 0 0
4 58 1 ALD 27 0 0
5 59 1 PBC 4720 0 0
6 35 2 PBC 4624 0 0
7 51 2 PBC 18 1 1 Rejection
8 61 2 PSC 294 1 1 Rejection
9 51 2 ALD 4673 0 0
10 59 2 ALD 51 0 0
11 53 1 PSC 8 1 2 Thrombosis
12 56 1 ALD 4592 0 0
13 55 2 ALD 4679 0 0
14 44 1 ALD 1487 0 0
15 61 2 PBC 427 1 4 Other
16 59 2 PBC 4604 0 0
17 52 2 PSC 4083 1 3 Recurrent disease
18 61 1 PSC 559 1 3 Recurrent disease
19 57 1 PSC 4708 0 0
20 49 2 ALD 957 1 4 Other
cause-specific survivor function, that is the probability of a particular end-

point occurring after any time t, might be anticipated.
12.2.1 Kaplan-Meier estimate of survivor function

To summarise competing risks data, we might consider using a separate es-
timate of the survivor function for each cause. This would involve using the
event times for each cause of interest in turn, taking the event times for all
other causes as censored. This is justified on the grounds that if a specific
end-point has been observed, no other causes of death could have occurred
before this time. These other causes may have occurred later, had the specific
end-point not occurred. Data expressed in this form is known as cause-specific
survival data. There are, however, pitfalls to using the Kaplan-Meier estimate
of the survivor function to summarise such data, as the following example
shows.

Data on the times to graft failure, and the cause of failure, for eight pa-
tients from the data set used in Example 12.1 are given in Table 12.2. The
data sets used to construct the survivor functions for the patients suffering
408 COMPETING RISKS
Table 12.2 Graft survival times for eight liver trans-

plant recipients and the cause of graft failure.
Patient Survival time Cause of failure
1 18 Rejection
2 27 Thrombosis
3 63 Thrombosis
4 80 Rejection
5 143 Thrombosis
6 255 Thrombosis
7 294 Rejection
8 370 Rejection
from graft rejection and thrombosis are as follows, where an asterisk (∗) de-
notes a censored observation. For failure from rejection, the graft survival
times are:
18 27∗ 63∗ 80 143∗ 255∗ 294 370
and for failure from thrombosis, the graft survival times are:
18∗ 27 63 80∗ 143 255 294∗ 370∗
The Kaplan-Meier estimate of the survivor function can be calculated from

each set of cause-specific survival times, and this leads to the estimates shown
in Table 12.3.
Table 12.3 Kaplan-Meier estimates of survivor functions for two

causes of graft failure in eight liver transplant recipients.
Rejection Thrombosis
Time interval Ŝ(t) Time interval Ŝ(t)
0– 1.000 0– 1.000
18– 0.875 27– 0.857
80– 0.700 63– 0.714
294– 0.350 143– 0.536
370– 0.000 255– 0.357
From the estimated survivor functions in this table, the probability of

graft failure due to rejection in the period to 370 days is 1.000, and that from
thrombosis is 1−0.357 = 0.643. However, half of this group of 8 patients suffers
graft failure from rejection in 370 days, and the other half fail from thrombosis.
The survivor function should therefore be 0.5 at 370 days in each case. Also,
the probability that one or other causes of graft failure occur after 370 days
is zero, since all 8 patients have failed by then, rather than 1 − (1 + 0.643)
suggested by the Kaplan-Meier estimates.
HAZARD AND CUMULATIVE INCIDENCE FUNCTIONS 409
This simple example shows that the Kaplan-Meier estimate does not lead
to an appropriate summary of survival data when there are competing risks.
So what is the Kaplan-Meier estimate estimating? Consider the Kaplan-Meier
estimate for the jth cause, where the event times from other causes are taken
to be censored. The Kaplan-Meier estimate is the probability of death beyond
time t, if cause j is the only cause of death, that is if all other risks were
removed. Consequently, the estimate does not take proper account of the
competing risks from other causes. Also, the Kaplan-Meier estimates assume
that an individual will ultimately die from any given cause, and so do not
allow for the possibility that a particular cause of death may never occur. In
general, as in Example 12.2, the complement of the Kaplan-Meier estimator
overestimates the incidence of a particular event, and so this estimate rarely
has a meaningful interpretation in the context of competing risks.
12.3 Hazard and cumulative incidence functions

This section introduces two functions that are particularly useful in the con-
text of competing risks, namely the hazard function and the cumulative inci-
dence function for the different event types.
12.3.1 Cause-specific hazard function

The cause-specific hazard function for the jth cause, hj (t), for j = 1, 2, . . . , m,
is defined by
{ }
P(t 6 T 6 t + δt, C = j | T > t)
hj (t) = lim . (12.1)
δt→0 δt
This hazard function is the instantaneous failure rate from the jth cause at
time t, in the presence of all other risks.
From the definition of hj (t) in Equation (12.1) and following the method
used to derive Equation (1.4) in Section 1.3.2 of Chapter 1, we have that
{ }
P(t 6 T 6 t + δt, C = j)
hj (t) = lim ,
δt→0 δt P(T > t)
{ }
1 P(t 6 T 6 t + δt, C = j)
= lim ,
P(T > t) δt→0 δt
and so
fj (t)
hj (t) = , (12.2)
S(t)
where fj (t) is the cause-specific density function and S(t) = P(T > t) is the
overall survivor function.
Since only one of the m possible causes can lead to an event,
∑
m
P(t 6 T 6 t + δt | T > t) = P(t 6 T 6 t + δt, C = j | T > t).
j=1
410 COMPETING RISKS
∑m
The overall hazard function is then h(t) = j=1 hj (t), using the definition of
the cause-specific hazard in∑Equation (12.1). Similarly, the overall cumulative
m
hazard function is H(t) = j=1 Hj (t), where Hj (t) is the cumulative hazard
function for the jth cause. The overall survivor function is then
{ ∑ m }
S(t) = exp − Hj (t) . (12.3)
j=1
Although S(t) can also be expressed as

∏
m
S(t) = Sj† (t),
j=1
where Sj† (t) = exp{−Hj (t)}, Sj† (t) is not an observable survivor function. This
is because we can never know the cause of a death that may occur after time
t, when there is more than one possible cause.
Survival studies where there are competing risks can also be formulated
in terms of m random variables, T1 , T2 , . . . , Tm , that are associated with the
times to the m possible causes of failure. These random variables cannot be
observed directly, since only one event can occur, and so they are referred to
as latent random variables. In practice, we observe the earliest of the m events
to occur, and the random variable associated with the time to that event, T , is
such that T = min(T1 , T2 , . . . , Tm ). If the different causes are independent, the
hazard function in Equation (12.2) is the marginal hazard function associated
with Tj , the random variable for the jth event type, which is
{ }
P(t 6 Tj 6 t + δt | Tj > t)
lim .
δt→0 δt
Unfortunately, the joint distribution of the random variables associated with
times to the different causes, (T1 , T2 , . . . , Tm ), cannot be uniquely determined,
and it is not possible to use the competing risks data to test the assumption of
independence of the different causes. Consequently, this formulation of com-
peting risks will not be considered further.
12.3.2 Cause-specific cumulative incidence function

In competing risks data, no cause of death has occurred until the death of
an individual from a particular cause. The cause-specific cumulative incidence
function for a particular cause is therefore a more useful summary of the data
than a survivor function. This is the probability of surviving until t and death
is from cause j, in the presence of all other risks, and is given by
Fj (t) = P(T < t, C = j),
for j = 1, 2, . . . , m. The maximum value of the cumulative incidence function
is
P(T < ∞, C = j) = P(C = j) = πj ,
where πj is the ultimate probability of cause j occurring. Consequently,
Fj (t) → πj as t → ∞, and since Fj (t) does not tend to zero, Fj (t) is not
a ‘proper’ probability distribution function. Because of this, the cumulative
incidence function is also referred to as the subdistribution function.
From Equation (12.2), fj (t) = hj (t)S(t), and it then follows that Fj (t) can
be expressed in the form
∫ t
Fj (t) = hj (u)S(u) du. (12.4)
0
Equation (12.4) suggests that the cause-specific cumulative incidence function

can be estimated by
∑ δij
F̂j (t) = Ŝ(ti−1 ), (12.5)
ni
i:ti 6 t
where Ŝ(ti−1 ) is the Kaplan-Meier estimate of the overall survivor function

at ti−1 , ignoring different causes, ni is the number of individuals alive and
uncensored just prior to ti , and δij is the event indicator for the cause-specific
data. Thus, δij is unity if cause j is the cause of death for the ith individual
and zero otherwise, and the ratio δij /ni in Equation (12.5) is the Nelson-
Aalen estimate of the hazard function for the jth cause; see Section 2.3.3 of
Chapter 2. The summation in Equation (12.5) is over all event times up to
time t, and so the estimated cumulative incidence function for a given cause
uses information on the death times for all causes. This means that it is not
possible to estimate the cause-specific cumulative incidence function from the
corresponding cause-specific hazard function.
When there is just one event type, that is m = 1, we have δi1 ≡ δi , and
using Equation (2.4) of Chapter 2, we find that F̂1 (t) = 1 − Ŝ(t), where Ŝ(t)
is the usual Kaplan-Meier estimate of the survivor function.
The variance of the estimated cumulative incidence function at time t is
given by
∑ {[ ]2 }
δi 2 δij (ni − δij )
var {F̂j (t)} = F̂j (t) − F̂j (ti ) + Ŝ(ti−1 )
ni (ni − δi ) n3i
i:ti 6 t
∑ { } δij
−2 F̂j (t) − F̂j (ti ) Ŝ(ti−1 ) 2 ,
ni
i:ti 6 t
∑m
where δi = j=1 δij and ni is the number at risk of an event occurring at time
ti . Confidence intervals for the cumulative incidence function at any time, t,
can then be found in the manner described in Section 2.2.3 of Chapter 2.
When there is just one type of event, Ŝ(ti−1 ) = ni (ni − δi )−1 Ŝ(ti ), and the
square root of this variance reduces to Greenwood’s formula for the standard
error of the Kaplan-Meier estimate, given as Equation (2.12) of Chapter 2.
The overall cumulative incidence, F (t), ∑m can be also be estimated from the
m cause-specific functions, since F̂ (t) = j=1 F̂j (t). This in turn leads to an
412 COMPETING RISKS
estimate of the overall survivor function, given by Ŝ(t) = 1 − F̂ (t), where in
the absence of explanatory variables, Ŝ(t) is the usual Kaplan-Meier estimate,
ignoring differences in death from different causes.

Example 12.2 showed that the Kaplan-Meier estimate of the survivor function
cannot be used to estimate cumulative incidence. For the data shown in Ta-
ble 12.2, the calculations needed to estimate the cumulative incidence function
for the two causes of graft failure, rejection and thrombosis respectively, are
shown in Table 12.4. The values of Ŝ(ti−1 ) are found from the Kaplan-Meier
estimate of the survivor function across all event times.
Table 12.4 Estimates of the cumulative incidence of rejection,

F̂1 (t), and thrombosis, F̂2 (t), from data on the survival times of
eight liver transplant recipients.
Time interval ni δi1 δi2 Ŝ(ti−1 ) F̂1 (t) F̂2 (t)
0– 8 0 0 1.000 0.000 0.000
18– 8 1 0 1.000 0.125 0.000
27– 7 0 1 0.875 0.125 0.125
63– 6 0 1 0.750 0.125 0.250
80– 5 1 0 0.625 0.250 0.250
143– 4 0 1 0.500 0.250 0.375
255– 3 0 1 0.375 0.250 0.500
294– 2 1 0 0.250 0.375 0.500
370– 1 1 0 0.125 0.500 0.500
The estimate of cumulative incidence at 370 days, obtained using Equa-

tion (12.5), is now 0.5 for each cause, as required. Moreover,∑the overall sur-
m
vivor function estimate at 370 days is now Ŝ(370) = 1 − j=1 F̂j (370) =
1 − (0.5 + 0.5) = 0, as it should be.

For the data on the survival times of liver transplant recipients, the estimated
cumulative incidence functions for graft failure due to rejection, hepatic artery
thrombosis, recurrent disease or other reasons, are shown in Figure 12.1. This
figure shows that the incidence functions vary between the four causes of
graft failure, with failure due to other causes having the greatest incidence in
these liver recipients. Estimates of the ultimate probability of each cause of
graft failure occurring are obtained from estimates of the cumulative incidence
function at 12 years, from which approximate estimates are 0.025, 0.055, 0.060
and 0.102 for rejection, hepatic artery thrombosis, recurrent disease and other
reasons, respectively.
The estimated cumulative incidence function can also be used to sum-
marise the effect of certain explanatory variables on particular end-points. As
an illustration, the cumulative incidence of thrombosis for transplant recip-
ients with each of the three indications for transplantation, PBC, PSC and
Cumulative incidence function 0.15
0.10
0.05
0.00
0 2 4 6 8 10 12
Survival time (years)
Figure 12.1 Cumulative incidence of the four causes of graft failure, rejection (—),
thrombosis ( ), recurrent disease (·······) and other reasons (- - -).
ALD, are shown in Figure 12.2. The incidence of thrombosis is quite similar
in patients with PBC and PSC, and greater than that for patients with ALD.
However, a more formal analysis is needed to determine the significance of
these differences.
The cumulative incidence functions provide descriptive summaries of com-
peting risks data, but they can be supplemented by analogues of the log-rank
test for comparing two or more groups, in the presence of competing risks.
These tests include Gray’s test (Gray, 1988) and the method due to Pepe and
Mori (1993). Further details are not given here as an alternative procedure
is available through a modelling approach to the analysis of competing risks
data, and possible models are described in Sections 12.4 and 12.5.
12.3.3 ∗ Some other functions of interest

The cause-specific cumulative incidence function, Fj (t), leads to certain other
quantities that may be of interest in a competing risks situation. For example,
the probability of death before t when the cause is j is
P(T < t | C = j) = πj−1 Fj (t),
where πj is the probability of the jth cause occurring. Also, the probability
of death from cause j when death has occurred before time t is
Fj (t)
P(C = j | T < t) = ,
F (t)
414 COMPETING RISKS
Cumulative incidence function 0.100
0.075
0.050
0.025
0.000
0 2 4 6 8 10 12
Survival time (years)
Figure 12.2 Cumulative incidence of thrombosis for recipients with PBC (—), PSC
(·······) or ALD (- - -) at the time of transplant.
where F (t) = 1 − S(t), and S(t) is overall survivor function. Estimates of each
of these probabilities can be readily obtained from the estimated cumulative
incidence function.
12.4 Modelling cause-specific hazards

To model the dependence of the cause-specific hazard function on the values
of p explanatory variables, we take the hazard of death for the ith individual,
i = 1, 2, . . . , n, from the jth cause, j = 1, 2, . . . , m, to be
hij (t) = exp(βj′ xi )h0j (t),
where h0j (t) is the baseline hazard for the jth cause, xi is the vector of values
of the explanatory variables for the ith individual, and β j is the vector of
their coefficients for the jth cause.
From results given in the sequel, separate models can be developed for
each cause of death from the cause-specific survival data, illustrated in Ex-
ample 12.2. To do this, a set of survival data is produced for each cause in
turn, where death from that cause is an event and the death times for all
other causes are regarded as censored. Inferences about the impact of each
explanatory variable on the cause-specific hazard function can then be based
on hazard ratios in the usual way, using either a Cox regression model or a
parametric model.
In modelling cause-specific hazards, the event times of individuals who
experience a competing risk are censored, and so are treated as if there is
MODELLING CAUSE-SPECIFIC HAZARDS 415
the possibility of the event of interest occurring in the future. Consequently,
the estimated hazard ratios correspond to the situation where other causes of
death are removed or assumed not to occur. This can lead to the hazard of a
particular cause of failure being overestimated. Also, models for cause-specific
survival data are based on the usual assumption of independent censoring. If
the competing events do not occur independently of the event of interest, this
assumption is not valid. Unfortunately the assumption of independent compet-
ing risks cannot be tested using the observed data. Despite these drawbacks,
this approach may be warranted when interest centres on how the explanatory
variables directly influence the hazard associated with a particular cause of
death, ignoring deaths from other causes.
When there is just one event type, the survivor function, and hence the
cumulative incidence function, can be obtained from the hazard function using
Equations (1.7) and (1.6) of Chapter 1. The impact of a change in the value
of an explanatory variable on the hazard function can then be interpreted
in terms of the effect of this change on the cumulative incidence function.
However, in the presence of competing risks, the cumulative incidence function
for any cause depends on the hazard of occurrence of each potential cause
of death, as indicated in Equation (12.5). This means that we cannot infer
how explanatory variables affect the cumulative incidence of each cause from
analysing cause-specific survival data. For this, we need to model the Fj (t)
directly, which we return to in Section 12.5.
12.4.1 ∗ Likelihood functions for competing risks models
In this section, it is shown that either a Cox regression model or a fully

parametric model can be used to model the dependence of the cause-specific
hazard functions on the explanatory variables, by fitting separate models to
the m sets of cause-specific survival data.
Models for the cause-specific hazard function, where the baseline hazard
functions for each cause, h0j (t), j = 1, 2, . . . , m, are not specified, are fitted by
maximising a partial likelihood function, just as in the case of a single cause
of death. Consider the probability that the ith individual, i = 1, 2, . . . , n,
dies from cause j at the ith ordered death time ti , conditional on one of the
individuals at risk of death at time ti dying from cause j. Using the approach
described in Section 3.3.1 of Chapter 3, this probability is
exp(βj′ xi )
∑ ′ , (12.6)
l∈R(ti ) exp(βj xl )
where R(ti ) is the risk set at time ti , that is the set of individuals who are
alive and uncensored just prior to ti . Setting δij = 1 if the ith individual
dies from the jth cause, and zero otherwise, the partial likelihood function in
416 COMPETING RISKS
Expression (12.6) can be written as
{ }δij
∏m
exp(βj′ xi )
∑ ′ ,
j=1 l∈R(ti ) exp(βj xl )
and the partial likelihood function for all n individuals is then

{ }δij
∏n ∏ m
exp(βj′ xi )
∑ ′ .
i=1 j=1 l∈R(ti ) exp(βj xl )
This function factorises into the product of m terms of the form

{ }δij
∏n
exp(βj′ xi )
∑ ′ ,
i=1 l∈R(ti ) exp(βj xl )
which is the partial likelihood function for the cause-specific survival data
corresponding to the jth cause. This means that m separate Cox regression
models can be fitted to the cause-specific survival data to determine how the
explanatory variables affect the hazard of death from each cause.
When h0j (t) is fully specified, we have a parametric model that can be fit-
ted using standard maximum likelihood methods. Again, the likelihood func-
tion factorises into a product of likelihoods for the cause-specific survival data.
To show this, the contribution to the likelihood function for an individual who
dies from cause ci at ti is fci (ti ), where ci = 1, 2, . . . , m, for the ith individ-
ual, i = 1, 2, . . . , n. A censored survival time, for which the value of the cause
variable ci is zero, contains no information about the possible future cause
of death, and so the corresponding contribution to the likelihood function is
the overall survivor function, S(ti ). Ignoring covariates for the moment, the
likelihood function for data from n individuals is then
∏
n
L= fci (ti )δi S(ti )1−δi , (12.7)
i=1
where δi = 0 if the ith individual has a censored survival time and unity oth-
erwise. Using the result in Equation (12.2), and writing hci (t) for the hazard
function for the ith individual who experiences cause ci ,
∏
n
L= {hci (ti )}δi S(ti ). (12.8)
i=1
∏m
Now, from Equation (12.3), S(ti ) = j=1 exp{−Hj (ti )}, where the cumula-
tive hazard function for the jth cause at time ti is obtained from the corre-
sponding cause-specific hazard function, hj (ti ), and given by
∫ ti
Hj (ti ) = hj (u) du.
0
MODELLING CAUSE-SPECIFIC HAZARDS 417
Also, setting δij = 1 if ci = j, and zero otherwise, j = 1, 2, . . . , m, the likeli-
hood in Equation (12.8) can be expressed solely in terms of the hazard func-
tions for each of the m causes, where
 
∏ n ∏m ∏
m
L=  hj (ti )δij  exp{−Hj (ti )},
i=1 j=1 j=1
and this equation can be written as
∏
n ∏
m
L= hj (ti )δij exp{−Hj (ti )}. (12.9)
i=1 j=1
The likelihood function in Equation (12.9) is the product of m terms of

the form
∏
n
hj (ti )δij exp{−Hj (ti )},
i=1
and by comparing this expression with Equation (12.8), we see that this is the
likelihood function for the jth cause when the event times of all other causes
are taken to be censored. Consequently, the cause-specific hazard functions
can be estimated by fitting separate parametric models to the cause-specific
survival data. For example, if the baseline hazard function for the jth cause,
h0j (t), is taken to have a Weibull form, so that h0j (t) = λj γj tγj −1 , the param-
eters λj and γj in this baseline hazard function, together with the coefficients
of explanatory variables in the model, can be obtained by fitting separate
Weibull models to the cause-specific survival times. Standard methods can
then be used to draw inferences about the impact of the explanatory variables
on the hazard of death from each cause.

To illustrate modelling cause-specific hazards, separate Cox regression models
are fitted to the cause-specific data on graft failure times in liver transplant
recipients. The models fitted contain the variables associated with patient age
as a linear term, gender and primary disease. The corresponding hazard ratios
are shown in Table 12.5, together with their 95% confidence limits.
The hazard ratios in this table can be interpreted as the effect of each
variable on each of the four possible causes of graft failure, irrespective of the
occurrence of the other three causes. For example, the hazard ratios for rejec-
tion apply in the hypothetical situation where a patient can only suffer graft
failure from rejection. From this analysis, the risk of rejection and thrombosis
is less in older patients, but the risk of graft failure from other causes increases
as patients get older. There is no evidence that the hazard of graft failure is
affected by gender. Patients with PBC have less risk of graft failure from re-
current disease than patients with PSC and ALD, and there is a suggestion
that PBC patients have a greater incidence of graft failure from thrombosis.
418 COMPETING RISKS
Table 12.5 Cause-specific hazard ratios and their 95% confidence intervals for the
four causes of graft failure in liver transplant recipients.
Variable Cause of graft failure
Rejection Thrombosis Recurrent disease Other
Age (linear) 0.97 0.98 0.98 1.03
(0.93, 1.00) (0.96, 1.00) (0.95, 1.01) (1.01, 1.05)
Gender male 0.60 1.07 0.89 0.92

(0.26, 1.37) (0.61, 1.90) (0.44, 1.80) (0.59, 1.43)
female 1.00 1.00 1.00 1.00
Disease PBC 0.68 1.80 0.34 0.66

(0.24, 1.91) (0.94, 3.46) (0.12, 1.01) (0.38, 1.13)
PSC 1.14 1.61 1.76 1.34
(0.46, 2.84) (0.89, 2.92) (0.90, 3.44) (0.86, 2.08)
ALD 1.00 1.00 1.00 1.00
12.4.2 ∗ Parametric models for cumulative incidence functions

In Section 12.4.1, we saw how standard parametric models for the hazard
function for the jth cause, hj (t), for j = 1, 2, . . . , m, could be fitted by mod-
elling the cause-specific survival data. However, if the hazard function for the
jth cause has a Weibull form, in view of Equation (12.2), the correspond-
ing density function, fj (t), and cumulative incidence function, Fj (t), are not
those of a Weibull distribution for the survival times. Indeed, since the cu-
mulative incidence function is not a proper distribution function, we cannot
model this using a standard probability distribution, and we need to account
for the ultimate probability of a specific cause, Fj (∞), having a value of less
than unity.
To take the simplest case, suppose that survival times are assumed to be
exponential with mean θj−1 for cause j, j = 1, 2, . . . , m. Then, conditional on
cause j,
P(T < t | C = j) = 1 − e−θj t ,
so that the cumulative incidence of cause j is
Fj (t) = P(T < t | C = j)P (C = j) = πj (1 − e−θj t ),
where πj is the probability of death from cause j. As t → ∞, this cumulative

incidence function tends to πj , as required.
= πj θj e−θj t , and on using the
The corresponding density function is fj (t) ∑
m
result in Equation (12.2), and taking S(t) = 1− j=1 Fj (t), the corresponding
cause-specific hazard becomes
πj θj e−θj t
λj (t) = ∑ −θj t
.
j πj e
MODELLING CAUSE-SPECIFIC INCIDENCE 419
This cause-specific hazard function is not constant, even though the condi-
tional incidence function has an exponential form.
When a parametric model is adopted for the cumulative incidence of the
jth cause, Fj (t), models can be fitted by maximising the likelihood function
in Equation (12.7), from which the likelihood function for the n observations
(ti , ci ) is
∏n
fci (ti )δi S(ti )1−δi ,
i=1
where δi = 0 for a censored observation and unity otherwise. In this expres-

sion, ci = j if the ith individual experiences the jth event
∑mtype, and S(ti ) is
the survivor function at ti , obtained from S(ti ) = 1 − j=1 Fj (ti ). Even in
the case of exponential cause-specific survival times, the corresponding like-
lihood function has a complicated form, and numerical methods are required
to determine the estimates of the unknown parameters that maximise it.
12.5 ∗ Modelling cause-specific incidence

In standard survival analysis, where there is just one possible end-point, there
is a direct correspondence between the cumulative incidence, survivor and
hazard functions, and models for the survivor function are obtained directly
from those for the hazard function. As noted in Section 12.4, this is not the
case when there are competing risks. Although models for cause-specific haz-
ards can be used to determine the impact of explanatory variables on the
hazard of the competing risks, a different approach is needed to model how
they affect the cumulative incidence function. In this section, a model for the
dependence of cause-specific cumulative incidence functions on explanatory
variables is described. This model was introduced by Fine and Gray (1999),
and has become known as the Fine and Gray model.
12.5.1 The Fine and Gray competing risks model

The cause-specific cumulative incidence function, or subdistribution function,
for the jth cause is Fj (t) = P(T < t, C = j). Using the relationship first given
in Equation (1.5) of Chapter 1, the corresponding hazard function for the
subdistribution, known as the subdistribution hazard function or subhazard, is
d 1 dFj (t)
λj (t) = − log{1 − Fj (t)} = , (12.10)
dt 1 − Fj (t) dt
for the jth cause.
Now, 1 − Fj (t) is the probability that a person survives beyond time t or
who has previously died from a cause other than the jth, and as in Section 1.3
of Chapter 1, { }
dFj (t) Fj (t + δt) − Fj (t)
= lim .
dt δt→0 δt
420 COMPETING RISKS
It then follows that the subdistribution hazard function, λj (t), can be ex-
pressed as
{ }
P(t 6 T 6 t + δt, C = j | T > t or {T 6 t and C ̸= j})
λj (t) = lim .
δt→0 δt
This is the instantaneous death rate at time t from cause j, given that an
individual has not previously died from cause j. Since the definition of this
hazard function includes those who have died from a cause other than j before
time t, this subdistribution hazard function is different from the cause-specific
hazard in Equation (12.1) in both definition and interpretation.
To model the cause-specific cumulative incidence function, a Cox regression
model is assumed for the subhazard function for the jth cause. The hazard of
cause j at time t for the ith of n individuals is then
λij (t) = exp(βj′ xi )λ0j (t), (12.11)
where λ0j (t) is the baseline subdistribution hazard function for cause j, xi
is the vector of values of p explanatory variables for the ith individual, and
the vector β j contains their coefficients for the jth cause. In this model, the
subdistribution hazard functions are assumed to be proportional.
The model in Equation (12.11) is fitted by adapting the usual partial
likelihood in Equation (3.4) of Chapter 3 to include a weighted combination
of values in the risk set. The resulting partial likelihood function for the jth
of m causes is
∏
rj
exp(βj′ xh )
∑ ′ , (12.12)
h=1 l∈R(t(h) ) whl exp(βj xl )
where the product is over the rj individuals who die from cause j at event
times t(1) < t(2) < · · · < t(rj ) , and xh is the vector of values of the explanatory
variables for an individual who dies from cause j at time t(h) , h = 1, 2, . . . , rj .
The risk set R(t(h) ) is the set of all those who have not experienced an event
before the hth event time t(h) , for whom the survival time is greater than
or equal to t(h) , and those who have experienced a competing risk by t(h) ,
for whom the survival time is less than or equal to t(h) . This risk set is not
straightforward to interpret, since an individual who has died from a cause
other than the jth before time t is no longer at risk at t, but nevertheless they
do feature in the risk set for this model. The weights in Expression (12.12)
are defined as
Ŝc (t(h) )
whl = ,
Ŝc (min{t(h) , tl })
where Ŝc (t) is the Kaplan-Meier estimate of the survivor function for the
censoring times. This is obtained by regarding all event times, of any type,
in the data set as censored times, and likewise all censored times as event
times, and calculating the Kaplan-Meier estimate from the resulting data.
The weight whl will be 1.0 when tl > t(h) , that is for those in the risk set who
MODELLING CAUSE-SPECIFIC INCIDENCE 421
have not had an event before t(h) , and less than 1.0 otherwise. The effect of
this weighting function is that individuals that die from a cause other than
the jth remain in the risk set and are given a censoring time that exceeds all
event times. Also, the weights become smaller with increasing time between
the occurrence of a competing risk and the event time being considered, so
that earlier deaths from a competing risk have a diminishing impact on the
results.
The partial likelihood in Expression (12.12) is maximised to obtain esti-
mates of the β-parameters for a given cause. Since the weights used in this
partial likelihood may vary over the survival time of a particular individual,
the data must first be assembled in the counting process format. This was
outlined in Section 8.6 of Chapter 8. This also makes it straightforward to
include time-dependent variables in this model.
The subdistribution hazard function is difficult to interpret, and the fitted
model is best interpreted in terms of the effect of the explanatory variables
on the cause-specific cumulative incidence function for the jth cause, which
from Equation (12.10), can be estimated by
F̂ij (t) = 1 − exp{−Λ̂ij (t)},
for the ith individual, where Λ̂ij (t) is an estimate of the cumulative subdis-
tribution hazard function, Λij (t). This estimate is given by
Λ̂ij (t) = exp(β̂j′ xi )Λ̂0j (t),
where Λ̂0j (t) is the baseline cumulative subdistribution hazard function for
the jth event type. This function can be estimated using an adaptation of
the Nelson-Aalen estimate of the baseline cumulative hazard function, given
in Equation (3.28) of Chapter 3, where
∑ dh
Λ̂0j (t) = ∑ ,
whl exp(β̂j′ xl )
t(h) 6t l∈R(t(h) )
and dh is the number of deaths at time t(h) .

The Fine and Gray model can also be expressed in terms of a baseline
cumulative incidence function, Fij (t), for each cause, where
′
Fij (t) = 1 − {1 − F0j (t)}exp(βj xi ) ,
and F0j (t) = 1 − exp{−Λ0j (t)} is the baseline cumulative incidence function
for the jth cause. This emphasises that a quantity of direct interest is being
modelled.
In this example, the Fine and Gray model for the cumulative incidence func-
tion is used to model the data given in Example 12.1. Models for the cu-
mulative incidence of graft rejection, thrombosis, recurrent disease and other
422 COMPETING RISKS
Table 12.6 Subhazard ratios and their 95% confidence intervals in the Fine and Gray
model for the four causes of graft failure in liver transplant recipients.
Variable Cause of graft failure
Rejection Thrombosis Recurrent disease Other
Age (linear) 0.97 0.98 0.98 1.03
(0.93, 1.00) (0.96, 1.00) (0.95, 1.01) (1.01, 1.05)
Gender male 0.60 1.08 0.89 0.90

(0.27, 1.31) (0.67, 1.75) (0.45, 1.74) (0.58, 1.39)
female 1.00 1.00 1.00 1.00
Disease PBC 0.69 1.83 0.35 0.67

(0.25, 1.91) (1.06, 3.17) (0.13, 0.98) (0.39, 1.15)
PSC 1.10 1.59 1.68 1.34
(0.44, 2.76) (0.87, 2.91) (0.84, 3.39) (0.85, 2.10)
ALD 1.00 1.00 1.00 1.00
causes of failure are fitted in turn. Subhazard ratios and their corresponding
95% confidence intervals for these four failure types are shown in Table 12.6.
The subhazard ratios in this table summarise the direct effect of each
variable on the incidence of different causes of graft failure, in the presence of
competing risks. However, their values are very similar to the hazard ratios
shown in Table 12.5, which suggests that there is little association between
the competing causes of graft failure. In a later example, Example 12.7 in
Section 12.6, this is not the case.
12.6 Model checking

Many of the model checking procedures described in Chapter 4 can be applied
directly to examine the adequacy of a model for a specific cause. The methods
for checking the functional form of an explanatory variable (plots of martingale
and Schoenfeld residuals) and testing for proportional hazards (plot of scaled
Schoenfeld residuals, tests of proportional hazards) are particularly useful. In
addition, an adaptation of the method described in Section 7.3 of Chapter 7
for comparing observed survival with a model-based estimate can be used
to determine the validity of an assumed parametric model for the cumulative
incidence of each cause. Some of these methods are illustrated in Example 12.7.
Example 12.7 Survival of laboratory mice

In a laboratory study to compare the survival times of two groups of mice fol-
lowing exposure to radiation, one group was reared in a standard environment
and the other group was kept in a germ-free environment. The mice were RFM
strain males, a strain that is particularly susceptible to tumour development
following exposure to ionising radiation. The mice were exposed to a dose of
300 rads of radiation when they were five or six weeks old, and each mouse
MODEL CHECKING 423
was followed up until death. Following an autopsy, the cause of death of each
mouse was recorded as thymic lymphoma, reticulum cell sarcoma or other
causes, together with the corresponding survival times in days. The data were
first described by Hoel (1972), and are given in Table 12.7. These data are
unusual in that there are no censored survival times.
Estimated cumulative incidence functions for deaths from the three causes
are shown in Figure 12.3.
0.6 0.6
Cumulative incidence function

0.4 0.4
0.2 0.2
0.0 0.0
0 200 400 600 800 1000 1200 0 200 400 600 800 1000 1200
Survival time - thymic lymphoma Survival time - reticulum cell sarcoma
0.6
0.4
0.2
0.0
0 200 400 600 800 1000 1200
Survival time - other causes
Figure 12.3 Cumulative incidence functions for mice raised in a standard environ-
ment (—) and a germ-free environment (·······) for each cause of death.
These plots suggest that the incidence of thymic lymphoma is greater for
mice raised in the germ-free environment, but that reticulum cell sarcoma has
greater incidence for mice kept in the standard environment. The incidence of
death from other causes also differs between the two environments.
424 COMPETING RISKS
Table 12.7 Survival times and causes of death for two groups of irradiated mice.
Thymic lymphoma Reticulum cell sarcoma Other causes
Standard Germ-free Standard Germ-free Standard Germ-free
159 158 317 430 40 136
189 192 318 590 42 246
191 193 399 606 51 255
198 194 495 638 62 376
200 195 525 655 163 421
207 202 536 679 179 565
220 212 549 691 206 616
235 215 552 693 222 617
245 229 554 696 228 652
250 230 557 747 249 655
256 237 558 752 252 658
261 240 571 760 282 660
265 244 586 778 324 662
266 247 594 821 333 675
280 259 596 986 341 681
343 300 605 366 734
356 301 612 385 736
383 321 621 407 737
403 337 628 420 757
414 415 631 431 769
428 434 636 441 777
432 444 643 461 800
485 647 462 807
496 648 482 825
529 649 517 855
537 661 517 857
624 663 524 864
707 666 564 868
800 670 567 870
695 586 870
697 619 873
700 620 882
705 621 895
712 622 910
713 647 934
738 651 942
748 686 1015
753 761 1019
763
MODEL CHECKING 425
We next fit a Cox regression model to the cause-specific survival times.
For a mouse reared in the ith environment, i = 1, 2, the model for the jth
cause-specific hazard, j = 1, 2, 3, is
hij (t) = exp(βj xi )h0j (t),
where xi = 1 for a mouse reared in the germ-free environment and zero other-
wise, so that βj is the log-hazard ratio of death from cause j at any time for a
mouse raised in the germ-free environment, relative to one from the standard
environment. The estimated hazard ratio, together with the corresponding
95% confidence limits and the P -value for testing the hypothesis that the
hazard ratio is 1.0, for each cause of death, are shown in Table 12.8.
Table 12.8 Hazard ratios, confidence limits and P -value for the three causes
of death in a cause-specific hazards model.
Cause of death Hazard ratio 95% confidence interval P -value
Thymic lymphoma 1.36 (0.78, 2.38) 0.28
Reticulum cell sarcoma 0.13 (0.07, 0.26) < 0.001
Other causes 0.31 (0.17, 0.55) < 0.001
This table shows that the hazard of death from thymic lymphoma is not
significantly affected by the environment in which a mouse is reared, but that
mice raised in a germ-free environment have a significantly lower hazard of
death from reticulum cell sarcoma or other causes.
This analysis shows how the type of environment in which the mice were
raised influences the occurrence of each of the three causes of death in circum-
stances where the other two possible causes cannot occur. Since the cumulative
incidence of any particular cause of death depends on the hazard of all possible
causes, we cannot draw any conclusions about the effect of environment on
the cause-specific incidence functions from modelling cause-specific hazards.
For this, we fit a Fine and Gray model for the cumulative incidence of thymic
lymphoma, reticulum cell sarcoma and other causes. This enables the effect
of environment on each of the three causes of death to be modelled, in the
presence of competing risks.
From Equation (12.11), the model for the subhazard function for a mouse
reared in the ith environment that dies from the jth cause is
λij (t) = exp(βj xi )λ0j (t). (12.13)
The corresponding model for the cumulative incidence of death from cause j
is
Fij (t) = 1 − exp{−eβj xi Λ0j (t)},
where Λ0j (t) is the baseline cumulative subhazard function and xi = 1 for
the germ-free environment and zero otherwise. The estimated ratio of the
subhazard functions for mice raised in the germ-free environment relative to
426 COMPETING RISKS
Table 12.9 Subhazard ratios, confidence limits and P -value for the three
causes of death in the Fine and Gray model.
Cause of death Hazard ratio 95% confidence interval P -value
Thymic lymphoma 1.68 (0.97, 2.92) 0.066
Reticulum cell sarcoma 0.39 (0.22, 0.70) 0.002
Other causes 1.01 (0.64, 1.57) 0.98
those in the standard environment, P -values and 95% confidence limits, are
given in Table 12.9.
This table suggests that, in the presence of competing risks of death, the
environment has some effect on the subhazard of death from thymic lym-
phoma, a highly significant effect on death from reticulum cell carcinoma, but
no impact on death from other causes. The germ-free environment increases
the subhazard of thymic lymphoma and reduces that for reticulum cell sar-
coma.
At first sight the subhazard ratios in Table 12.9 appear quite surprising
when compared to the hazard ratios in Table 12.8. Although this feature
could result from the effect of competing risks on the incidence of death from
thymic lymphoma or other causes, an alternative explanation is suggested
by the estimates of the cumulative incidence functions, which were shown
in Figure 12.3. For death from other causes, the incidence of death in the
standard environment exceeds that in the germ-free environment at times
when there are relatively few events. At later times, where there are more
events, the incidence functions are closer together, and beyond 800 days, the
incidence of death in the germ-free environment is greater. The assumption
that the subhazard functions in the Fine and Gray model are proportional is
therefore doubtful.
To investigate this further, techniques described in Section 4.4 of Chapter 4
are used. First a plot of the scaled Schoenfeld residuals from the model for the
subhazard function for each cause is obtained, with a smoothed curve super-
imposed. The plot for deaths from other causes is shown in Figure 12.4. This
figure shows clearly that the smoothed curve is not horizontal, and strongly
suggests that the environment effect is time-dependent.
To further examine the assumption of proportional subhazards, the time-
dependent variable xi log t, with a cause-dependent coefficient βj1 , is added to
the model in Equation (12.13). The change in the value of −2 log L̂ on adding
this term to the subhazard model is not significant for death from thymic lym-
phoma (P = 0.16), but significant for reticulum cell sarcoma (P = 0.016), and
highly significant for death from other causes (P < 0.001). This demonstrates
that the effect of environment on the subhazard of death from reticulum cell
sarcoma and other causes is not independent of time. To illustrate this for
death from other causes, where j = 3, Figure 12.5 shows the time-dependent
subhazard ratio, exp{β3 + β31 log t}, plotted against log t, together with 95%
confidence bands. The hazard ratio is significantly less than 1.0 for survival
MODEL CHECKING 427
3
Scaled Schoenfeld residual
-1
-3
0 200 400 600 800 1000 1200
Survival time
Figure 12.4 Plot of the scaled Schoenfeld residuals against log survival time on fitting
a Fine and Gray model to data on mice that die from other causes.
6
Subhazard ratio
0
0 1 2 3 4 5 6 7
Figure 12.5 Time-dependent subhazard ratio (—), with 95% confidence bands (- - -),
for mice that die from other causes.
428 COMPETING RISKS
times less than 240 days and significantly greater than 1.0 for survival times
greater than 650 days.
Another feature of these data is that there are four unusually short times to
death from other causes in mice reared in the standard environment. However,
these observations have little influence on the results.

The earliest text to be published on competing risks was that of David and
Moeschberger (1978), who included a summary of the work in this area since
the seventeenth century. More up to date is the paper of Moeschberger and
Klein (1995), and the texts of Pintilie (2006), Crowder (2012), and Beyers-
mann, Allignol and Schumacher (2012). Each of these books shows how the
analyses can be carried out using the R software. A more mathematical ac-
count of the area is given by Crowder (2001). General texts on survival analy-
sis that contain chapters on competing risks include Kalbfleisch and Prentice
(2002), Lawless (2002), and Kleinbaum and Klein (2012). A number of tu-
torial papers on competing risks have also been published, such as those of
Putter, Fiocco and Geskus (2007) and Pintilie (2007a). Nomenclature is this
area is notoriously inconsistent; see Wolbers and Koller (2007), and Latouche,
Beyersmann and Fine (2007), for comments on that used in Pintilie (2007a),
and the response in Pintilie (2007b). An illustration of the use of competing
risks models in predictive modelling is given by Wolbers et al. (2009) and a
comparison of four different approaches is given by Tai et al. (2001).
Expressions for the variance of the estimated cumulative incidence function
are derived by Aalen (1978a), Marubini and Valsecchi (1995) and Lin (1997)
amongst others. Gray (1988) and Pepe and Mori (1993) describe various mod-
els and present tests for the comparison of cumulative incidence functions for
two or more groups of survival times in the presence of competing risks.
Parametric models for cause-specific hazards have been described by
Kalbfleish and Prentice (2002) and Maller and Zhao (2002), while Hinch-
cliffe and Lambert (2013) describe and illustrate flexible parametric models
for competing risks data. The Fine and Gray model for competing risks was
introduced by Fine and Gray (1999). Kohl and Heinze (2012) describe a SAS
macro for competing risks analysis using the Fine and Gray model, and Gray
(2013) describes the cmprsks package for use in the R system. A number of
papers have presented extensions to the Fine and Gray model, including the
additive model of Sun, Sun and Zhang (2006), and the parametric model of
Jeong and Fine (2007).
Chapter 13
Multiple events and event history

modelling
Individuals observed over time may experience multiple events of the same
type, or a number of events of different types. Studies in which there are
repeated occurrences of the same type of event, such as a headache or asth-
matic attack, are frequently encountered and techniques are needed to model
the dependence of the rate at which such events occur on characteristics of the
individuals or exposure factors. Multiple event data also arise when an indi-
vidual may experience a number of outcomes of different types, with interest
centering on factors affecting the time to each of these outcomes. Where the
occurrence of one event precludes the occurrence of all others, we have the
competing risks situation considered in Chapter 12, but in this chapter we
consider situations where this is not necessarily the case.
In modelling the course of a disease, individuals may pass through a num-
ber of phases or states that correspond to particular stages of the disease. For
example, a patient diagnosed with chronic kidney disease may be on dialy-
sis before receiving a kidney transplant, and this may be followed by phases
where chronic rejection occurs, or where the transplant fails. A patient may
experience some or all of these events during the follow-up period from the
time of diagnosis, and the sequence of events is termed an event history. Mul-
tistate models can be used to describe the movement of patients among the
various states, and to investigate the effect of explanatory variables on the
rate at which they transfer from one state to another.
Multiple event and event history data can be analysed using an extension
of the Cox regression model to the situation where more than one event can
occur. This extension involves the development of a probabilistic model for
events that occur over time, known as a counting process, and so this chapter
begins with an introduction to counting processes.
13.1 ∗ Introduction to counting processes

A counting process for the ith of n individuals is defined to be the sequence
of values of the random variable Ni (t), t > 0, that counts the number of oc-
currences of some event over the time period (0, t], for i = 1, 2, . . . , n. A
429
430 MULTIPLE EVENTS AND EVENT HISTORY MODELLING
realisation of the counting process Ni (t) is then a step-function that starts at
0 and increases in steps of one unit.
We next define Yi (t), t > 0, to be a process where Yi (t) = 1 when the ith
individual is at risk of an event occurring at time t, and zero otherwise, so
that Yi (t) is sometimes called the at-risk process. The value of Yi (t) must be
known at a time t−, the time immediately before t, and when this is the case,
Yi (t) is said to be a predictable process.
The counting process Ni (t) has an associated intensity, which is the rate
at which events occur. Formally, the intensity of a counting process is the
probability that Ni (t) increases by one step in unit time, conditional on the
history of the process up to time t. The history or filtration of the process
up to but not including time t is written H(t−), and is determined from the
set of values {Ni (s), Yi (s)} for all values of s up to time t. If we denote the
intensity process by λi (t), t > 0, then, in infinitesimal notation, where dt is an
infinitely small interval of time, we have
1
λi (t) = P{Ni (t) increases by one step in an interval of length dt | H(t−)},
dt
1
= P{Ni (t + dt) − Ni (t) = 1 | H(t−)}.
dt
If we set dNi (t) = Ni (t + dt) − Ni (t) to be the change in Ni (t) over an
infinitesimal time interval of length dt, λi (t) can be expressed as
1
λi (t) = P{dNi (t) = 1 | H(t−)}. (13.1)
dt
Since dNi (t) is either 0 or 1, it follows that
1
λi (t) = E {dNi (t) | H(t−)},
dt
and so the intensity of the counting process is the expected number of events
in unit time, conditional on the history of the process.
We can also define the cumulative intensity, or integrated intensity, Λi (t),
where ∫ t
Λi (t) = λi (u) du
0
is the cumulative expected number of events in the time interval (0, t], that is
Λi (t) = E {Ni (t)}.
13.1.1 Modelling the intensity function

A convenient general model for the intensity function for the ith individual is
that it depends on an observable, predictable stochastic process Yi (t), and an
unknown function of both time and known values of p possibly time-varying
INTRODUCTION TO COUNTING PROCESSES 431
explanatory variables, X1 (t), X2 (t), . . . , Xp (t). The intensity of the counting
process Ni (t) is then taken to be
λi (t) = Yi (t)g{t, xi (t)},
where xi (t) is the vector of values of the p explanatory variables at time t.

Setting g{t, xi (t)} = exp{β ′ xi (t)}λ0 (t), we have
λi (t) = Yi (t) exp{β ′ xi (t)}λ0 (t),
where λ0 (t) is a baseline intensity function that depends on t alone.

The argument used in Section 3.3.1 of Chapter 3 to derive the partial
likelihood of a sample of survival times can be adapted to obtain the likelihood
of data from a realisation of a counting process. Details are omitted but in
counting process notation, the partial likelihood has the form
n ∏ {
∏ }dNi (t)
Yi (t) exp{β ′ xi (t)}
∑n ′ , (13.2)
i=1 t >0 l=1 Yl (t) exp{β xl (t)}
where dNi (t) = 1 if Ni (t) increases by one unit at time t, and zero otherwise.
This function is then maximised with respect to unknown parameters in the
intensity function, leading to estimates of the βs.
A large body of theory has been developed for the study of counting pro-
cesses, and this enables the asymptotic properties of parameter estimates to
be determined. This theory is based on the properties of a type of stochastic
process with zero mean known as a martingale. In fact, the process defined by
Mi (t) = Ni (t) − Λi (t) (13.3)
is such that E {Mi (t)} = 0 for all t, and is a martingale. Theoretical details will
not be given here, but note that Equation (13.3) is the basis of the martingale
residuals defined in Equation (4.6) of Chapter 4.
13.1.2 Survival data as a counting process

Survival data with right-censored survival times, consisting of pairs formed
from a time, ti , and an event indicator δi , i = 1, 2, . . . , n, is a particular
example of a counting process. To show this, consider a counting process in
which Ni (t) = 0 when the ith individual is at risk of an event occurring, and
Ni (t) steps up to unity at the time when an event occurs. Since an individual
is at risk at the time origin, Ni (0) = 0, and so the process Ni (t) starts at 0 and
increases to 1 at the event time of the ith individual, or remains at 0 for those
individuals with censored survival times. The value of the at-risk process,
Yi (t), is unity until the occurrence of the event, or the time is censored, and
zero thereafter.
From Equation (13.1), the intensity of the counting process is such that
λi (t) dt = P{dNi (t) = 1 | H(t−)},
= P{t 6 Ti 6 t + dt | H(t−)},
where Ti is the random variable associated with the time to an event for the
ith individual. The intensity function can then be expressed as
λi (t) dt = Yi (t)hi (t) dt, (13.4)
where
hi (t) dt = P(t 6 Ti 6 t + dt | T > t)
is the hazard function, hi (t), first defined in Equation (1.3) of Chapter 1.
The result in Equation (13.4) follows from the fact that in survival data, the
hazard of an event occurring in the interval from t to t + dt depends only on
the individual being at risk at time t−, and so the relevant history at that
time, H(t−), is that Ti > t. Inclusion of Yi (t) in Equation (13.4) ensures that
λi (t) = 0 when the ith individual is not at risk.
From Equation (13.4), the intensity of the counting process is given by
λi (t) = Yi (t)hi (t), so that the intensity function is the hazard of an event
occurring at time t when the ith individual is at risk of an event. Also, the
intensity function of the standard Cox regression model can then be expressed
as
Yi (t) exp(β ′ xi )h0 (t),
where h0 (t) is the baseline hazard function. In the sequel, this will simply be
denoted hi (t), as usual.
To obtain the likelihood function for this model from the general result
in Expression (13.2), we note that dNi (t) is zero until the event time, when
dNi (t) = 1. The only contributions to the partial likelihood function in Equa-
tion (13.2) are from individuals who are at risk, and so Equation (13.2) be-
comes { }δi
∏n
exp{β ′ xi (t)}
∑ ′ ,
i=1 l∈R(ti ) exp{β xl (t)}
as given in Equation (3.5) of Chapter 8. When the explanatory variables are

not time-dependent, xj ≡ xj (t) for all values of t, j = 1, 2, . . . , p, and this is
Equation (3.5) in Section 3.3 of Chapter 3.
These results mean that the theory of counting processes can be used to
prove many of the standard results in survival analysis, and to derive asymp-
totic properties of test statistics. However, the great advantage of the counting
process formulation is that it can be used to extend the Cox model in many
directions. These include time-varying explanatory variables, dependent cen-
soring, recurrent events, multiple events per subject and models for correlated
data. It also allows problems with more complicated censoring patterns to be
managed. These include left truncation, where the follow-up process for an
individual only begins some known time after the time origin, and multistate
models. Some of these situations will be considered in this chapter.
INTRODUCTION TO COUNTING PROCESSES 433
13.1.3 Survival data in the counting process format
Counting process data consist of a sequence of time intervals (tj−1 , tj ], j =

1, 2, . . ., where t0 = 0, over which the values of the counting process Ni (t),
the at-risk process Yi (t), and any time-varying explanatory variables Xi (t),
are constant for the ith of n individuals. Associated with each interval is the
value of Ni (t) at the end of the interval, Ni (tj ), which marks the time when
Ni (t) increases by one unit, or when Yi (t) or an explanatory variable changes
in value. This leads to an alternative and more flexible way of formatting
survival data.
To express survival data in counting process format, suppose that t is the
event time and δ is the event indicator for a particular individual. The time
interval (0, t] is then divided into a series of r intervals with cut points at
times t1 , t2 , . . . , tr−1 , so that 0 < t1 < t2 < · · · < tr−1 < t. The correspond-
ing intervals are (0, t1 ], (t1 , t2 ], . . . , (tr−1 , t]. These intervals will be such that
the values of all time-varying explanatory variables are constant within each
interval, and where both the at-risk process and the counting process have
constant values. The event status at tj is N (tj ), so that N (tj ) = 0 if an in-
dividual remains at risk at tj , and N (tj ) = 1 if that individual experiences
an event at tj , the end of the jth interval, (tj−1 , tj ], for j = 1, 2, . . . , r, where
tr ≡ t. Then, for the ith individual, the observation on a survival time and an
event indicator, (ti , δi ), can be replaced by the sequence {ti,j−1 , tij , Ni (tij )},
for j = 1, 2, . . . , r, with ti0 ≡ 0 and tir ≡ ti . This form of the survival data
is often referred to as (start, stop, status) format, but more usually as the
counting process format. The partial likelihood function of data expressed in
this way is equivalent to that for the original data, and so the model fit is
identical. Values of the −2 log L̂ statistic for fitted models, as well as param-
eter estimates and their standard errors, are then identical to those obtained
when the data are analysed using the standard format.
Example 13.1 Illustration of the counting process format

Suppose that a patient in a study concerning cancer of the liver is observed
from the time when they enter the study until they die 226 days later. During
this time, the stage of the cancer is measured at particular time points, and
the data for this individual are given in Table 13.1.
Table 13.1 Tumour size at different times for an individual with

liver cancer.
Observation Time Tumour stage Start Stop Status
1 0 1 0 45 0
2 45 2 45 84 0
3 84 2 84 127 0
4 127 3 127 163 0
5 163 4 163 226 1
6 226 4
In the counting process format, a sequence of time intervals is defined,
leading to the intervals (0, 45], (45, 84], (84, 127], (127, 163], (163, 226]. These
are the start and stop times in Table 13.1. The status variable denotes whether
a patient is alive (status = 0) or dead (status = 1) at the end of each interval. It
is natural to use the stage of the tumour at the start of each of these intervals
in the analysis, which means that the tumour stage in the period leading to
a possible death at the stop time is used in the modelling process. Other
possibilities include using the stage at the end of the interval, or some form
of intermediate value. See Section 8.2.1 of Chapter 8 for a fuller discussion on
this. Since the tumour stage is the same at 45 and 84 days, and also at 163
and 226 days, the intervals (0, 84], (84, 127], (127, 226] could be used instead.
Also, if the individual were still alive at 226 days, and this marked the end of
follow-up for this patient, their status would be recorded as zero at that time.
13.1.4 Robust estimation of the variance-covariance matrix
When the counting process formulation of a Cox regression model is used to

analyse multiple event data, the event times within an individual will not be
independent. This in turn means that the standard errors of the parameter
estimates in fitted models, obtained in the usual manner, may be smaller
than they really should be. This overestimation of the precision of model-
based estimates can be avoided by using a robust estimate of the variance-
covariance matrix of the parameter estimates that is not so dependent on
the underlying model. The most widely used robust estimate is termed the
sandwich estimate, described in this section, and a number of studies have
shown that this estimate performs well in misspecified Cox regression models.
The standard model-based estimate of the variance-covariance matrix of
the parameter estimates in a Cox regression model, var (β̂), is the inverse
of the observed information matrix. This was introduced in Section 3.3.3 of
Chapter 3, and is a p × p matrix denoted I −1 (β̂), where p is the number of
unknown β-parameters. A robust estimate of the variance-covariance matrix
due to Lin and Wei (1989) is found from I −1 (β̂)A(β̂)I −1 (β̂), where the matrix
A(β̂) can be regarded as a correction term that allows for potential model
misspecification. This correction term is sandwiched between two copies of
the inverse of the information matrix, which is why it is called a sandwich
estimate. The correction matrix is A(β̂) = U ′ (β̂)U (β̂), where U (β̂) is the
n × p matrix of efficient scores, evaluated at the estimates β̂, for each of the
n observations.
In Section 4.3.1 of Chapter 4, the p × 1 vector of efficient scores for the
ith observation was written as r U i and termed the vector of score residu-
als for the ith observation, and so U (β̂) is an n × p matrix whose rows are
r ′U 1 , r ′U 2 , . . . , r ′U n . Now, the change in the vector of β-parameters, on omitting
the ith observation, was shown in Section 4.3.1 of Chapter 4 to be approx-
imately r ′U i var (β̂). The matrix of these values is D(β̂) = U (β̂)I −1 (β̂), in
MODELLING RECURRENT EVENT DATA 435
the notation of this section, and the components of the ith row of this matrix
are the delta-beta statistics for assessing the effect of the ith observation on
each of the p parameter estimates. It then follows that the sandwich estimate
can be written as D ′ (β̂)D(β̂), which shows how the sandwich estimate can
be obtained from the delta-beta’s.
In applications envisaged in this chapter, the data will consist of times to
a relatively small number of events within each individual. In this situation,
it is better to base the sandwich estimator on an aggregate of the event times
within an individual. This is achieved by summing the efficient scores, that
are the elements of the matrix U (β̂), over the event times within each of n
individuals, so that this matrix is still of dimension n × p.
When the robust estimate of the variance-covariance matrix is being used,
it can be helpful to compare the ratios of the robust standard errors to the
usual model-based standard errors. The effect of using the robust estimator on
quantities such as interval estimates for hazard ratios can then be determined.
13.2 Modelling recurrent event data

Situations where an individual experiences more than one event of the same
type, known as a recurrent event, are common in medical research, and exam-
ples include sequences of tumour recurrences, infection episodes, adverse drug
reactions, epileptic seizures, bleeding incidents, migraines and cardiothoracic
events. Techniques are then needed to estimate the rate or intensity at which
recurrences occur, and the dependence of the recurrence rate on characteristics
of the individuals or exposure factors.
In an initial analysis of recurrent event data, standard methods of survival
analysis can be used to model the time to the first event. Although straight-
forward, this approach ignores subsequent events and so is not making the
most of the data available. However, more detailed analyses are complicated
by two features. First, there will be variation between individuals in their sus-
ceptibility to recurrent events, and the recurrence times within an individual
may not be independent. Second, following the occurrence of an event such
as a heavy bleed, the treatment may mean that the individual is not at risk
of a subsequent bleed for some short period of time. It will then be necessary
to modify the set of individuals who are at risk of a recurrent event to take
account of this.
Variations on the Cox regression model provide a convenient framework
for a modelling approach to the analysis of recurrent event data. In particular,
the counting process approach, described in Section 13.1, enables features such
as variation in the at-risk status over time to be incorporated.
There are three models that are commonly used in recurrent event analysis.
The simplest of these is due to Anderson and Gill (1982), but a rather more
flexible model is due to Prentice, Williams and Peterson (1981). A third model,
due to Wei, Lin and Weissfeld (1989) will also be described, but this model is
more useful when there are events of different types, rather than recurrences
of events of the same type, and so is introduced later in Section 13.3.1. We
will follow others in referring to these as the AG, PWP and WLW models,
respectively.
13.2.1 The Anderson and Gill model

In this model, the baseline intensity function is common to all recurrences
and unaffected by the occurrence of previous events, and the individual recur-
rence times within an individual are independent. As a result, this model is
particularly useful when interest centres on modelling the overall recurrence
rate.
The AG model for the intensity, or hazard, of an event occurring at time
t in the ith of n individuals is given by
hi (t) = Yi (t) exp{β ′ xi (t)}h0 (t), (13.5)
where Yi (t) denotes whether or not the ith individual is at risk of an event
at time t, β ′ xi (t) is a linear combination of p possibly time-dependent ex-
planatory variables, and h0 (t) is a baseline hazard function. Patients who
may experience multiple events remain at risk, and so Yi (t) remains at unity
unless an individual temporarily ceases to be at risk in some time period, or
until the follow-up time is censored. The risk set at time t is then the set of all
individuals who are still being followed up at time t, just as in the standard
Cox model.
To fit this model, the recurrent event data are expressed in counting process
format, where the jth time interval, (tj−1 , tj ], is the time between the (j −1)th
and jth recurrences, j = 1, 2, . . . , where t0 = 0. The status variable denotes
whether or not an event occurred at the end-point of the interval, that is the
jth recurrence time, and is unity if tj was an event time and zero otherwise.
An individual who experiences no events contributes a single time interval
where the end-point is censored, so that the status variable is zero. Similarly,
the status variable will be zero at the time marking the end of the follow-up
period, unless an event is observed at that time. The format of the data that
is required to fit the AG model is illustrated later in Example 13.2.
The assumption of independent recurrence times can be relaxed to some
extent by including terms in the model that correspond to the number of
preceding events, or the time from the origin. Since any association between
recurrence times will usually result in the model-based standard errors of
the estimated β-parameters being too small, the robust form of the variance-
covariance matrix of the estimated β-parameters, described in Section 13.1.4,
can be used.
This approach may not properly account for within-subject dependence,
and so alternatively, and preferably, association between the recurrence times
can be accommodated by adding a random frailty effect for the ith subject,
as described in Chapter 10. The addition of a random subject effect, ui , to
the AG model leads to the model
hi (t) = Yi (t) exp{β ′ xi (t) + ui }h0 (t),
in which ui may be assumed to have a N (0, σu2 ) distribution. Fitting the AG
model with and without a frailty effect allows the extent of within-subject
correlation in the recurrence times to be assessed, using the methods described
in Section 10.6 of Chapter 10.
Once a model has been fitted, hazard ratios and their corresponding in-
terval estimates can be determined. In addition, the recurrence times can be
summarised using the unadjusted or adjusted cumulative intensity (or hazard)
function and the cumulative incidence function, estimated from the comple-
ment of the survivor function. This will be illustrated in Example 13.2.
13.2.2 The Prentice, Williams and Peterson model

A straightforward extension of the AG model in Equation (13.5) leads to a
more flexible model, in which separate strata are defined for each event. This
model allows the intensity to vary from one recurrence to another, so that
the within-subject recurrence times are no longer independent. In the PWP
model, the hazard of the jth occurrence, j = 1, 2, . . . , of an event at time t in
the ith of n individuals is
hij (t) = Yij (t) exp{βj′ xi (t)}h0j (t), (13.6)
where Yij (t) is unity until the (j − 1)th recurrent event and zero thereafter, β j
is the vector of coefficients of the p explanatory variables for the jth recurrence
time, xi (t) is the vector of values of the explanatory variables, and h0j (t) is
the baseline hazard for the jth recurrence. In this model, the risk set for the
jth recurrence is restricted to individuals who have experienced the previous
(j − 1) recurrences. As the intensity of the jth recurrent event is conditional
on the (j − 1)th having occurred, this is termed a conditional model, and is
widely regarded as the most satisfactory model for general use.
The PWP model, defined in Equation (13.6), can be fitted using a Cox
regression model, stratified by recurrence number. The data are expressed in
the same format as that used to fit the AG model, where the status indica-
tor is unity until the jth recurrence time, and zero thereafter. In addition, a
stratifying factor denotes the recurrence number. The coefficients of the ex-
planatory variables may take different values for each recurrence, but they can
be constrained to have a common value for a subset of the strata when large
numbers of recurrences do not occur very often, or across all the strata. As
usual, the resulting change in the value of the −2 log L̂ statistic can be used
to formally test the hypothesis of a constant hazard ratio across strata.
This model is particularly suited to situations where there is interest in
the times between different recurrences of an event, and in whether hazard
ratios vary with recurrence number. As for the AG model, a robust variance-
covariance matrix for the parameter estimates can be used to further allow
for correlated recurrence times, and the model can also be extended by using
a frailty term to account for any association between recurrence times.
Example 13.2 Recurrence of mammary tumours in female rats

In animal experiments that are designed to compare an active treatment for
the prevention of cancer with a control, the rate at which tumours occur is of
interest. Gail, Santner and Brown (1980) describe an experiment concerning
the development of mammary cancer in female rats that are allocated to one
of two treatment groups. A number of rats were injected with a carcinogen
and then all rats were treated with retinyl acetate, a retinoid that is a natural
form of vitamin A, with the potential to inhibit tumour development. After
60 days, the 48 rats that were tumour free were randomly allocated to receive
continuing prophylactic treatment with the retinoid, or to a control group
with no such treatment. The rats were assessed for evidence of tumours twice
weekly, and the observation period ended 122 days after randomisation to
treatment. Rats with no tumour at 122 days have censored recurrence times.
The data in Table 13.2 give the times from randomisation to the devel-
opment of mammary tumours for the 23 rats in the group receiving retinoid
prophylaxis and 25 controls. Since the rats were not assessed on a daily basis,
some rats were found to have developed more than one tumour at an assess-
ment time. For example, one rat in the control group was found to have three
tumours on day 42. To avoid the ambiguity of tied observations, these have
been replaced by times of 42, 42.1 and 42.2.
A simple analysis of these data is based on modelling the time to the first
event. Of the 48 rats in the experiment, just two do not develop a tumour and
so contribute censored survival times. Using standard methods, the hazard of a
first tumour occurring in a rat on the retinoid treatment, relative to one in the
control group, is 0.50. This is significantly different from unity (P = 0.027),
and the corresponding 95% confidence interval is (0.27, 0.93). There is strong
evidence that the hazard of a first tumour in treated rats is about half that
of control rats.
To illustrate how the data are organised for a recurrent event analysis,
consider the data from the fifth rat in the treated group, for whom the times
at which a tumour was detected are 70, 74, 85 and 92 days, and the rat was
subsequently followed up until day 122 with no further tumours detected.
The representation of this sequence in counting process format for the AG
and PWP models is shown in Table 13.3. The treatment group is represented
by an explanatory variable that is unity for a rat who receives the retinoid
treatment and zero otherwise.
In the AG model, the hazard of a tumour developing at time t is given by
hi (t) = Yi (t) exp(βxi )h0 (t),
where Yi (t) = 1 when a rat is at risk of tumour development and zero other-
wise, xi = 1 if the ith rat is in the treated group and zero otherwise, and β
is the log-hazard ratio for a rat on the retinoid treatment, relative to one in
Table 13.2 Recurrence times of tumours in rats on a retinoid treatment or control.

Treatment Times to tumour
Retinoid 122
122∗
3, 8, 122∗
92, 122∗
70, 74, 85, 92, 122∗
38, 92, 122
28, 35, 45, 70, 77, 107, 122∗
92, 122∗
21, 122∗
11, 24, 66, 74, 92, 122∗
56, 70, 122∗
31, 122∗
3, 8, 24, 35, 92, 122∗
45, 92, 122∗
3, 42, 92, 122∗
3, 17, 52, 80, 122∗
17, 59, 92, 101, 107, 122∗
45, 52, 85, 101, 122
92, 122∗
21, 35, 122∗
24, 31, 42, 48, 70, 74, 122∗
122∗
31, 122∗
Control 3, 42, 59, 101, 101.1, 112, 119, 122∗

28, 31, 35, 45, 52, 59, 59.1, 77, 85, 107, 112, 122∗
31, 38, 48, 52, 74, 77, 101, 101.1, 119, 122∗
11, 114, 122∗
35, 45, 74, 74.1, 77, 80, 85, 90, 90.1, 122∗
8, 8.1, 70, 77, 122∗
17, 35, 52, 77, 101, 114, 122∗
21, 24, 66, 74, 101, 101.1, 114, 122∗
8, 17, 38, 42, 42.1, 42.2, 122∗
52, 122∗
28, 28.1, 31, 38, 52, 74, 74.1, 77, 77.1, 80, 80.1, 92, 92.1, 122∗
17, 119, 122∗
52, 122∗
11, 11.1, 14, 17, 52, 56, 56.1, 80, 80.1, 107, 122∗
17, 35, 66, 90, 122∗
28, 66, 70, 70.1, 74, 122∗
3, 14, 24, 24.1, 28, 31, 35, 48, 74, 77, 119, 122∗
21, 28, 45, 56, 63, 80, 85, 92, 101, 101.1, 119, 122∗
28, 35, 52, 59, 66, 66.1, 90, 97, 119, 122∗
8, 8.1, 24, 42, 45, 59, 63, 63.1, 77, 101, 119, 122
80, 122∗
92, 122, 122.1
21, 122∗
3, 28, 74, 122∗
24, 74, 122
* Censored recurrence times.
Table 13.3 Representation of the recurrence times of one rat for the
AG and PWP models.
Model Event Interval Status Stratum Treatment
AG 1 (0, 70] 1 1 1
2 (70, 74] 1 1 1
3 (74, 85] 1 1 1
4 (85, 92] 1 1 1
5 (92, 122] 0 1 1
PWP 1 (0, 70] 1 1 1

2 (70, 74] 1 2 1
3 (74, 85] 1 3 1
4 (85, 92] 1 4 1
5 (92, 122] 0 5 1
the control group. A standard Cox regression model is then fitted to the stop
times, that is the end-point of each interval, in the data arranged as illustrated
in Table 13.3.
Using the robust standard error of the treatment effect, β̂ = −0.801 and
se (β̂) = 0.198. This estimate is significantly less than zero (P < 0.001), and
the overall hazard of a tumour occurring in a rat in the retinoid group, relative
to one in the control group, at any given time, is 0.45, with a 95% confidence
interval of (0.30, 0.66). The tumour occurrence rate for rats in the retinoid
treatment group is less than half that of rats in the control group.
The treatment difference can be summarised using the estimated cumu-
lative intensity, or hazard, of tumour recurrences for rats in each treatment
group. This is exp(β̂x)Ĥ0 (t), where Ĥ0 (t) is the estimated baseline cumula-
tive intensity function, and x = 1 for rats exposed to the retinoid treatment
and x = 0 for rats in the control group. This is also the cumulative expected
number of tumour recurrences over time. The cumulative intensity for each
treatment is shown in Figure 13.1. This shows that rats exposed to the retinoid
treatment have a lower intensity of tumour occurrence than rats in the control
group. At 45 days, it is estimated that one tumour would have occurred in a
rat in the treated group, but two tumours would have occurred by that time
in a rat in the control group.
The cumulative incidence of a recurrence can be obtained from 1 −
Ŝ0 (t)exp(β̂x) , for x = 0 or 1, where Ŝ0 (t) = exp{−Ĥ0 (t)} is the estimated
baseline survivor function in the fitted AG model. The two incidence functions
are shown in Figure 13.2. Again, this figure confirms that tumour incidence
is substantially greater for rats in the control group. The median tumour re-
currence time is 31 days for rats in the treated group and 17 days for rats in
the control group.
Instead of using the robust standard error of the treatment effect, associ-
ation between the recurrence times within a rat can be modelled by adding
a normally distributed random effect to the AG model. This random effect
5
Cumulative intensity
0
0 20 40 60 80 100 120
Recurrence time
Figure 13.1 Cumulative intensity of recurrent events for rats on the retinoid treat-
ment (—) or a control (·······).
1.0
0.8
Cumulative incidence
0.6
0.4
0.2
0.0
0 20 40 60 80 100 120
Recurrence time
Figure 13.2 Cumulative incidence of recurrent events for rats on the retinoid treat-
ment (—) or a control (·······).
differs for each rat but will be constant for the recurrence times within a rat.
When this is done, the rat effect is highly significant and the estimated vari-
ance of the random effect is 0.25. In this model, the estimated overall hazard
ratio is 0.46 with a 95% confidence interval of (0.31, 0.71). These estimates
are similar to those obtained using the robust standard error.
A limitation of the AG model is that the recurrence rate does not depend
on the number of preceding recurrences, and so we next fit the PWP model.
In this model, the hazard rate for the jth recurrence, j = 1, 2, . . . , in the ith
rat, i = 1, 2, . . . , 48, is
hij (t) = Yij (t) exp(βj xi )h0j (t),
where Yij (t) = 1 until the time of the (j − 1)th recurrence and zero there-
after, xi = 1 if the ith rat is in the treated group and zero otherwise, and βj
measures the treatment effect for the jth recurrence. On fitting this model,
estimates β̂j , the hazard ratios exp(β̂j ) and their standard errors, for the first
four recurrences and five or more recurrences, are as shown in Table 13.4. Both
the standard model-based standard errors and those obtained using the sand-
wich estimator are shown in this table. Unusually, those based on the robust
estimator are smaller, but the sandwich estimator will be used in subsequent
analyses.
Table 13.4 Estimates on fitting a PWP model to the recurrence times.

Recurrence β̂j exp(β̂j ) se (β̂j ) [model] se (β̂j ) [sandwich]
1 −0.686 0.503 0.312 0.283
2 −0.123 0.884 0.349 0.317
3 −0.831 0.436 0.425 0.381
4 0.060 1.062 0.482 0.296
>5 −0.850 0.427 0.419 0.274
The hazard ratio for the first recurrence is exp(−0.686) = 0.50, which is
identical to that from just modelling the time to the first tumour occurrence,
as it should be. The hazard ratio for a second and fourth recurrence are close
to unity, and not significant, P = 0.697, 0.839, respectively, while the hazard
ratios for three and five or more recurrences are significantly less than unity,
P = 0.029, 0.002, respectively. However, the value of −2 log L̂ for the fitted
PWP model is 812.82, and on constraining all the βs to be equal, the value of
this statistic increases to 816.95. This increase of 4.13 is not significant (P =
0.39) when compared to percentage points of the χ24 distribution, and so, on
an overall basis, there is no evidence of a difference between the hazard ratios
for different numbers of recurrences. The common estimate of β is −0.523, so
that the hazard ratio for any recurrence is 0.59, with a 95% confidence interval
of (0.46, 0.77). This is very similar to that found using the AG model, but
here the recurrence times have different baseline hazards.
The hazard functions, hij (t), can be constrained to be proportional by fit-
ting an unstratified model that includes a factor associated with the recurrence
MULTIPLE EVENTS 443
number, so that
hij (t) = Yij (t) exp(βj xi + ζj )h0 (t),
where ζj is the effect of the jth recurrence. If the ζj are all zero, the model re-
duces to the AG model. The stratified and unstratified models are not nested,
and so the two models cannot be directly compared. However, assuming pro-
portional hazards, the overall hazard rate is 0.59, with a 95% confidence in-
terval of (0.45, 0.77), which is very similar to that found using the stratified
model.
13.3 Multiple events

Individuals suffering from chronic diseases may experience multiple events,
each of which is of a different type. Any one individual may not experience
all of these events, and whereas recurrent events have a natural time order,
multiple events may occur in a different order in different individuals. For
example, a patient diagnosed with liver disease may experience events such
as histological progression of the disease, development of varices, development
of ascites, an increase in serum bilirubin level, a liver transplant and death.
Generally, each of these events will occur at most once in a sequence that may
vary from patient to patient. Such data could be analysed by fitting separate
models to the times to each event from the same time origin, but a more
efficient analysis is based on the Wei, Lin and Weissfeld (WLW) model.
13.3.1 The Wei, Lin and Weissfeld model

In this model, strata are defined for each event, in such a way that the total
number of strata is equal to the number of possible events. As for the PWP
model, the hazard of the occurrence of the jth event at time t in the ith
individual is given by
hij (t) = Yij (t) exp{βj′ xi (t)}h0j (t),
but now Yij (t) is unity until the jth event occurs and 0 thereafter. In this
model, the risk set at time t consists of all individuals who were being followed
up at time t and in whom the jth event has not occurred. This model has
the same form as a Cox regression model for the jth event, except that in the
WLW model, the βs are jointly estimated from the times to all events. The
WLW model is termed a marginal model since each event time is treated as a
separate event, and the time origin for each event is the start of the follow-up
time for each individual. Hazard ratios may vary across the different event
types, and the model also allows for differences in the underlying intensity of
each event to be accommodated.
To fit this model, the total number of possible events across all the individ-
uals in the study, r, is determined, and the time to each event is expressed as
a series of intervals from the time origin, (0, tj ], where tj is the time of the jth
event, j = 1, 2, . . . , r. A robust variance-covariance matrix may again be used,
or a frailty term can be incorporated to specifically allow for any correlation
between the event times within an individual. Also, the baseline hazards could
be taken to be proportional and constraints on the β-parameters may also be
introduced, as for the PWP model in Section 13.2.2.
The WLW model was originally proposed as a model for recurrent events,
but there is some controversy surrounding its use in this area. In particular,
when the WLW model is used to model recurrent events, the definition of the
risk set is such that an individual who has experienced just one recurrence is at
risk of not only a second recurrence, but also a third, fourth, fifth recurrence,
and so on. This makes it difficult to interpret the coefficients of the explanatory
variables. Because of this inconsistency, the model is only recommended for
use in situations where it is natural to consider times to separate events from
the time origin.
Example 13.3 Clinical trial of tamoxifen in breast cancer patients

There is a considerable body of evidence to suggest that following breast-
conserving surgery in women with breast cancer, local radiotherapy in addition
to tamoxifen treatment reduces the risk of recurrence in the same breast and
improves long-term survival. Breast screening is able to detect very small
tumours, and there is interest in determining whether patients at low risk
could avoid radiotherapy treatment. A multicentre clinical trial was begun
in 1992 to determine the benefit of adjuvant treatment with radiotherapy in
women aged 50 years or over, who had undergone breast-conserving surgery
for an invasive adenocarcinoma with a diameter less than or equal to 5 cm. The
patients were randomly assigned to receive tamoxifen treatment at the rate of
20 mg per day for five years in addition to breast irradiation, or to tamoxifen
alone. Patients were seen in clinic every three months for the first three years,
every six months for the next two and annually thereafter, and follow-up ended
in 2002. Full details of the trial and its results were given by Fyles et al. (2004).
This illustration is based on the data for one particular centre, used by Pintilie
(2006), and relates to 320 women who had been randomised to tamoxifen in
addition to radiotherapy, and 321 who received tamoxifen alone.
At randomisation, information was recorded on patient age, tumour size,
tumour histology, hormone receptor level, haemoglobin level and whether or
not axillary node dissection was performed. During the follow-up process, the
time from randomisation to the first occurrence of local relapse, axillary re-
lapse, distant relapse, second malignancy of any type and death, was recorded.
For each patient, the data set contains the number of days from randomisation
to the occurrence of any of these events, together with an event indicator. For
women who had no events and were alive at the end of follow-up, each of the
event times was censored at the last date that they were known to be alive.
The variables in the data set are listed below, and the explanatory variables
and survival data for 25 patients on the tamoxifen and radiotherapy treatment
Table 13.5 Data from 25 patients in a clinical trial on tamoxifen.
Id Treat Age Size Hist HR Hb ANdis Lsurv Ls Asurv As Dsurv Ds Msurv Ms Tsurv Ts
1 0 51 1.0 1 1 140 1 3019 0 3019 0 3019 0 3019 0 3019 0
2 0 74 0.5 1 1 138 1 2255 0 2255 0 493 1 2255 0 2255 0
3 0 71 1.1 1 1 157 1 2621 0 2621 0 2621 0 2621 0 2621 0
4 0 52 0.8 1 1 136 1 3471 0 3471 0 3471 0 3471 0 3471 0
5 0 62 1.5 1 1 123 1 3322 0 3322 0 3322 0 3322 0 3322 0
MULTIPLE EVENTS
6 0 75 0.7 1 1 122 1 2956 0 2956 0 2956 0 2956 0 2956 1

7 0 77 2.4 1 1 139 1 265 0 265 0 220 1 265 0 265 1
8 0 78 2.0 1 1 142 1 1813 0 1813 0 1813 0 1813 0 1813 0
9 0 65 2.0 1 1 121 1 3231 0 3231 0 3231 0 3231 0 3231 0
10 0 67 1.2 1 1 132 1 1885 0 1885 0 1885 0 1885 0 1885 0
11 0 64 1.5 1 1 133 1 3228 0 3228 0 3228 0 3228 0 3228 0
12 0 58 0.5 5 1 129 1 3284 0 3284 0 3284 0 3284 0 3284 0
13 0 71 2.2 1 1 143 1 2628 0 2628 0 2628 0 2628 0 2628 0
14 0 51 1.7 1 1 140 1 3428 0 3428 0 2888 1 344 1 3428 0
15 0 57 1.3 2 1 121 1 3288 0 3288 0 3288 0 3288 0 3288 0
16 0 70 3.8 1 1 152 1 377 0 377 0 269 1 377 0 377 1
17 0 67 1.5 1 1 132 1 125 0 125 0 125 0 113 1 125 1
18 0 54 3.0 1 1 153 1 3461 0 3461 0 3461 0 959 1 3461 0
19 0 53 0.5 1 1 121 1 3485 0 3485 0 3485 0 3485 0 3485 0
20 0 72 2.0 1 1 140 0 3399 0 3399 0 3399 0 3399 0 3399 0
21 0 57 2.5 1 1 129 1 2255 1 2835 0 2835 0 2835 0 2835 0
22 0 67 1.1 1 1 126 1 3322 0 3322 0 3322 0 3322 0 3322 0
23 0 56 1.0 1 1 134 1 2620 0 2620 0 2620 0 2620 0 2620 0
24 0 86 1.5 1 1 134 0 1117 0 1117 0 1117 0 299 1 1117 1
25 0 82 1.5 2 1 114 1 2115 0 2115 0 2115 0 2115 0 2115 0
445
Id: Patient identifier
Treat: Treatment group (0 = tamoxifen + radiotherapy, 1 = tamoxifen)
Age: Patient age (days)
Size: Tumour size (cm)
Hist: Tumour histology (1 = ductal, 2 = lobular, 3 = medullary,
4 = mixed, 5 = other)
HR: Hormone receptor level (0 = negative, 1 = positive)
Hb: Haemoglobin level (g/l)
ANdis: Axillary node dissection (0 = no, 1 = yes)
Lsurv: Time to local relapse or last follow-up
Ls: Local relapse (0 = no, 1 = yes)
Asurv: Time to axillary relapse or last follow-up
As: Axillary relapse (0 = no, 1 = yes)
Dsurv: Time to distant relapse or last follow-up
Ds: Distant relapse (0 = no, 1 = yes)
Msurv: Time to second malignancy or last follow-up
Ms: Second malignancy (0 = no, 1 = yes)
Tsurv: Time from randomisation to death or last follow-up
Ts: Status at last follow-up (0 = alive, 1 = dead)
These data will be analysed using the WLW model, and the first step is to
organise the data into the required format. Although no patient experienced
more than three of the five possible events, the revised database will have
five rows for each patient, corresponding respectively to time to local relapse,
axillary relapse, distant relapse, second malignancy and death. The values of
the explanatory variables are repeated in each row.
To illustrate this rearrangement, the event times and censoring indicators
for four patients, one with no events, one with one event, one with two events
and one with three events, are shown in Table 13.6. In this table, Nevents is
the number of events and the explanatory variables have been omitted.
Table 13.6 Data for four patients in the tamoxifen trial who experience 0, 1, 2 and
3 events.
Id Nevents Lsurv Ls Asurv As Dsurv Ds Msurv Ms Tsurv Ts
1 0 3019 0 3019 0 3019 0 3019 0 3019 0
2 1 2255 0 2255 0 493 1 2255 0 2255 0
14 2 3428 0 3428 0 2888 1 344 1 3428 0
349 3 949 1 2117 1 2775 1 3399 0 3399 0
The data for these four patients, in the required format, are given in Ta-
ble 13.7. Here, the variable Time is the time to the event concerned and Status
is the event status, where zero corresponds to a censored time and unity to
an event. In this table, the variable Event is coded as 1 for local relapse, 2
for axillary relapse, 3 for distant relapse, 4 for second malignancy and 5 for
death.
MULTIPLE EVENTS 447
Table 13.7 Rearranged data for four patients in the tamoxifen trial with 0, 1, 2 and
3 events.
Id Treat Age Size Hist HR Hb ANdis Event Time Status
1 0 51 1.0 1 1 140 1 1 3019 0
1 0 51 1.0 1 1 140 1 2 3019 0
1 0 51 1.0 1 1 140 1 3 3019 0
1 0 51 1.0 1 1 140 1 4 3019 0
1 0 51 1.0 1 1 140 1 5 3019 0
2 0 74 0.5 1 1 138 1 1 2255 0
2 0 74 0.5 1 1 138 1 2 2255 0
2 0 74 0.5 1 1 138 1 3 493 1
2 0 74 0.5 1 1 138 1 4 2255 0
2 0 74 0.5 1 1 138 1 5 2255 0
14 0 51 1.7 1 1 140 1 1 3428 0
14 0 51 1.7 1 1 140 1 2 3428 0
14 0 51 1.7 1 1 140 1 3 2888 1
14 0 51 1.7 1 1 140 1 4 344 1
14 0 51 1.7 1 1 140 1 5 3428 0
349 1 74 1.3 1 1 149 1 1 949 1
349 1 74 1.3 1 1 149 1 2 2117 1
349 1 74 1.3 1 1 149 1 3 2775 1
349 1 74 1.3 1 1 149 1 4 3399 0
349 1 74 1.3 1 1 149 1 5 3399 0
A Cox regression model, stratified by Event, in which the coefficients of

the explanatory variables are different for each event, is now fitted to the data
that have been organised as shown in Table 13.7. For each event type, the
hazard ratio for treatment with tamoxifen alone, relative to tamoxifen with
radiotherapy, the corresponding P -value based on the Wald test, and 95%
confidence limits for the hazard ratio, adjusted for the variables Age, Size,
Hist, HR, Hb and ANdis, are shown in Table 13.8. A robust estimate of the
variance-covariance matrix of the parameter estimates has been used.
Table 13.8 Adjusted hazard ratio for the treatment effect for each event and 95%
confidence intervals on fitting the WLW model.
Event Hazard ratio P -value 95% confidence interval
1: local relapse 10.30 < 0.001 (3.68, 28.82)
2: axillary relapse 3.14 0.092 (0.83, 11.91)
3: distant relapse 0.84 0.588 (0.45, 1.57)
4: second malignancy 1.19 0.502 (0.72, 1.98)
5: death 0.79 0.416 (0.45, 1.39)
From this table, it is estimated that there is more than 10 times the risk
of a local relapse if tamoxifen is used without radiotherapy, an effect that is
highly significant (P < 0.001). There is also some evidence, significant at the
10% level, that the absence of radiotherapy increases the risk of an axillary
relapse, but the estimated hazard ratios for all other events do not differ
significantly from unity. Further analysis of this data set would determine the
extent to which the hazards rates are proportional for different event types,
and which explanatory factors were relevant to each outcome.
13.4 Event history analysis

In studies where the primary outcome is survival from the time origin to death,
the occurrence of other non-fatal events during the follow-up period may help
to shed further light on the underlying mortality process. The sequence of
intermediate events defines an event history, and data of this kind can be
analysed using multistate models.
The experience of a patient in a survival study can be thought of as a
process that involves two states. At the point of entry to the study, the patient
is in a state that corresponds to their being alive. Patients then transfer from
this ‘live’ state to the ‘dead’ state at some transition rate, h(t), which is the
hazard of death at a given time t. The situation is expressed diagrammatically
in Figure 13.3. The dependence of the rate of transition from one state to the
other on explanatory variables is then modelled.
h(t) -
Alive Dead
Figure 13.3 A two-state model for survival analysis.
As an example in which there are a number of states, consider a survival

study concerning transplant patients. Following a transplant, there is the risk
that the transplanted organ will fail, as well as the risk of death, and the
transition rates for these two events may well be different. In addition, a
patient who suffers graft failure may be retransplanted, but may also die
without retransplant. The situation is shown diagrammatically in Figure 13.4.
This four-state model can be specified in terms of the transition rates
from transplant to death, hT D (t), from transplant to graft failure, hT F (t),
from graft failure to retransplant, hF R (t), from graft failure to death, hF D (t)
and from retransplant to death, hRD (t). Notice that although hT D (t), hF D (t)
and hRD (t) all denote the hazard of death at time t, the hazard rates depend
on whether the patient has had graft failure or has been retransplanted, and
may all be different.
It is straightforward to model the hazard of death following transplant
when graft failure has not occurred, hT D (t). Here, the survival times of those
patients who suffer graft failure are taken to be censored at the failure time.
Patients who are alive and who have not experienced graft failure also con-
tribute censored survival times. When modelling transition rates from graft
EVENT HISTORY ANALYSIS 449
hT F (t) hF R (t)
Transplant - Graft failure - Retransplant
Z
Z
Z
Z
Z
hT D (t)ZZ hF D (t)
hRD (t)
Z
Z
Z
Z
Z
Z~?
Z
=
Death
Figure 13.4 A four-state model for recipients of a transplant.
failure or retransplant to death, the set of patients at risk of death at any time
consists of those who have had graft failure or a retransplant, respectively, and
are still alive. Patients who have not yet had a graft failure or retransplant
cannot be in either risk set. Fortunately, by expressing the data in counting
process format, a Cox regression model can be used to model all four tran-
sitions between the states in this four-state model. Moreover, this approach
can be extended to more complex multistate models.
13.4.1 Models for event history analysis

Suppose that xi is the vector of values of p explanatory variables for the ith
individual and let β jk be the corresponding vector of their coefficients for the
transition from state j to state k. Also, denote the baseline intensity function
for this transition by h0jk (t). The transition rate of the ith individual from
state j to state k at time t in a multistate model is then
′
hijk (t) = Yijk (t) exp{βjk xi }h0jk (t),
where Yijk (t) = 1 if the ith of n individuals is in state j and at risk of entering
state k at time t, and Yijk (t) = 0, otherwise. As for other models described in
this chapter, the explanatory variables may also be time-dependent. Time is
generally measured from the point of entry into an initial state, which marks
the time origin.
When separate baseline hazards and separate regression coefficients are
assumed for each transition, stratified Cox regression models can be used to
model the different transition rates. Proportional hazard rates may also be
assumed, and modelled by including a factor in the model that corresponds
to the different states. In general, the counting process formulation of survival
data will be needed for this, as the risk set definition will depend on the
transition of interest. The data for a particular individual relating to the
transition from state j to state k will then consist of a record giving the time
of entry into state j, the time of exit from state j and a status indicator that
is unity if the transition is from state j to state k and zero otherwise. The
process is illustrated in Example 13.4.
Example 13.4 Patient outcome following bone marrow transplantation

The European Group for Blood and Marrow Transplantation (EBMT) was
established to promote stem cell transplantation and to allow clinicians to
develop collaborative studies. This is facilitated by the EBMT Registry,
which contains clinical and outcome data for patients who have undergone
haematopoietic stem cell transplantation. The Registry was established in the
early 1970s, and transplant centres and national Registries contribute data,
including annual follow-up data on their patients. As a result, the EBMT
Registry is a valuable source of data for retrospective studies on the outcomes
of stem cell transplantation in patients with leukaemia, lymphoma, myeloma,
and other blood disorders.
This example is based on data from 2204 patients with leukaemia who
received a bone marrow transplant reported to the EBMT Registry, and used
in Putter, Fiocco and Geskus (2007). The transplant is designed to allow the
platelet count to return to normal levels, but some individuals may relapse or
die before this state has been achieved. Others may relapse or die even though
their platelet count has returned to normal. A model with three states will
be used to represent this event history. The time following the transplant is
the initial state (state 1), and a patient may subsequently enter a state where
platelet recovery has occurred (state 2) or a state corresponding to relapse
or death (state 3). Transitions from state 2 to state 3 are also possible. This
multistate model is shown diagrammatically in Figure 13.5. In this figure, the
transition rates from state i to state j are denoted hij (t), so that the three
rates are h12 (t), h13 (t) and h23 (t).
For each patient, data are also available on the variable Leukaemia
that is associated with the disease type, categorised as acute myelogenous
leukaemia (AML), acute lymphoblastic leukaemia (ALL) and chronic myel-
ogenous leukaemia (CML). Their age group, Age ( 6 20, 21–40, > 40), was
also recorded, together with the patient’s donor recipient gender match, Match
(0 = no, 1 = yes), and whether or not there was T-cell depletion, Tcell (0 = no,
1 = yes). Data for the first 20 patients are shown in Table 13.9. This table
shows the times from transplantation to platelet recovery (Ptime) and the
time to relapse or death (RDtime). Also shown are the event indicators at the
time of platelet recovery (Pcens) and relapse or death (RDcens), which are
zero for a censored observation and unity for an event.
Of the 2204 patients who had a bone marrow transplant, 1169 experi-
enced platelet recovery and 458 relapsed or died before their platelet level
h12 (t)
- Platelet
Transplant
recovery
@
@
@
@
h13 (t) @ h23 (t)
@
@
@
@
@
R
Relapse
or death
Figure 13.5 A three-state model for recipients of a bone marrow transplant.
Table 13.9 Data on outcomes following bone marrow transplantation.

Id Leukaemia Age Match Tcell Ptime Pcens RDtime RDcens
1 CML > 40 1 0 23 1 744 0
2 CML > 40 0 0 35 1 360 1
3 CML > 40 0 0 26 1 135 1
4 AML 21–40 0 0 22 1 995 0
5 AML 21–40 0 0 29 1 422 1
6 ALL > 40 0 0 38 1 119 1
7 CML 21–40 1 0 30 1 590 0
8 ALL 6 20 0 0 35 1 1448 0
9 AML 21–40 0 0 1264 0 1264 0
10 CML > 40 0 0 1102 0 338 1
11 AML 21–40 1 0 50 1 84 1
12 AML > 40 1 0 22 1 114 1
13 AML 21–40 0 0 33 1 1427 0
14 AML > 40 0 0 29 1 775 1
15 CML > 40 0 0 24 1 1047 1
16 AML 21–40 0 0 31 1 1618 0
17 AML 21–40 1 0 87 1 1111 0
18 AML 21–40 0 0 469 0 255 1
19 ALL 6 20 0 0 59 0 59 1
20 CML 21–40 0 0 1727 0 1392 1
had returned to normal. A further 383 patients relapsed or died after platelet
recovery had occurred. There were also 152 patients that had experienced a
relapse but remained under surveillance for platelet recovery; this includes pa-
tients 10, 18 and 20 in Table 13.9. Since the model shown in Figure 13.5 does
not allow for transitions from relapse or death to platelet recovery, the times
to platelet recovery are censored at the time of relapse for these patients.
To fit multistate models, the data need to be formatted in such a way that
there is one row for each possible transition for each patient. There are three
possible transitions, shown as 1→2, 1→3, and 2→3. For the 2→3 transition,
only patients whose platelet level has returned to normal are at subsequent
risk of relapse or death, so that only two possible transitions are recorded for
these patients. Data for the first 10 patients in Table 13.9, expressed in this
format, are shown in Table 13.10.
Table 13.10 Reorganised data on outcomes following bone marrow transplantation.

Id Leukaemia Age Match Tcell Transition Start Stop Status
1 CML > 40 1 0 1→2 0 23 1
1 CML > 40 1 0 1→3 0 23 0
1 CML > 40 1 0 2→3 23 744 0
2 CML > 40 0 0 1→2 0 35 1
2 CML > 40 0 0 1→3 0 35 0
2 CML > 40 0 0 2→3 35 360 1
3 CML > 40 0 0 1→2 0 26 1
3 CML > 40 0 0 1→3 0 26 0
3 CML > 40 0 0 2→3 26 135 1
4 AML 21–40 0 0 1→2 0 22 1
4 AML 21–40 0 0 1→3 0 22 0
4 AML 21–40 0 0 2→3 22 995 0
5 AML 21–40 0 0 1→2 0 29 1
5 AML 21–40 0 0 1→3 0 29 0
5 AML 21–40 0 0 2→3 29 422 1
6 ALL > 40 0 0 1→2 0 38 1
6 ALL > 40 0 0 1→3 0 38 0
6 ALL > 40 0 0 2→3 38 119 1
7 CML 21–40 1 0 1→2 0 30 1
7 CML 21–40 1 0 1→3 0 30 0
7 CML 21–40 1 0 2→3 30 590 0
8 ALL 6 20 0 0 1→2 0 35 1
8 ALL 6 20 0 0 1→3 0 35 0
8 ALL 6 20 0 0 2→3 35 1448 0
9 AML 21–40 0 0 1→2 0 1264 0
9 AML 21–40 0 0 1→3 0 1264 0
10 CML > 40 0 0 1→2 0 338 0
10 CML > 40 0 0 1→3 0 338 1
The transitions between the three states are modelled by fitting a Cox
regression model to the data in the format shown in Table 13.10, stratified by
transition type. A stratified model that allows the parameters associated with
the four explanatory factors, Leukaemia, Age, Match and Tcell, to depend
on the transition type is first fitted, by including interactions between the
strata and these four factors. Comparing alternative models using the −2 log L̂
statistic shows that there is no evidence that the factors Tcell and Match vary
over the three transition types, and moreover, none of the three transitions
depend on Match. A reduced model therefore contains the main effects of
Leukaemia, Age and Tcell, together with interactions between Leukaemia and
Transition, and Age and Transition. For this model, the hazard ratios, and
their associated 95% confidence intervals, are shown in Table 13.11. In this
example, a robust estimate of variance is used, but the resulting standard
errors differ little from the model-based estimates.
Table 13.11 Hazard ratios and 95% confidence intervals for each transition.
Factor Transition 1→2 Transition 1→3 Transition 2→3
Leukaemia
AML 1.00 1.00 1.00
ALL 0.96 (0.83, 1.11) 1.29 (0.98, 1.69) 1.16 (0.86, 1.56)
CML 0.74 (0.65, 0.85) 1.02 (0.82, 1.26) 1.31 (1.04, 1.64)
Age
6 20 1.00 1.00 1.00
21–40 0.86 (0.74, 0.99) 1.29 (0.96, 1.72) 1.02 (0.76, 1.38)
> 40 0.92 (0.78, 1.08) 1.69 (1.25, 2.29) 1.68 (1.23, 2.30)
Tcell
No 1.00 1.00 1.00
Yes 1.42 (1.27, 1.59) 1.42 (1.27, 1.59) 1.42 (1.27, 1.59)
The effect of both type of leukaemia and age group differs for the three
transitions. Patients with CML progress to platelet recovery at a slower rate
than those with the other two types of disease, and are more likely to relapse
after platelet recovery has occurred. Patients with ALL have a greater hazard
of relapse or death before platelet recovery than those with AML or CML.
Patients aged 21–40 experience platelet recovery at a slower rate than others,
while those aged over 40 have an increased hazard of relapse or death, whether
or not platelet recovery has occurred. T-cell depletion leads to significantly
greater rates of transition to other states, with no evidence that these rates
differ between transition types.
The three baseline cumulative hazard and incidence rates, that is the rates
for patients in the 6 20 age group with AML and no T-cell depletion, are
shown in Figures 13.6 and 13.7. The baseline cumulative hazard plot shows
that the transition to platelet recovery occurs at a much faster rate than
transitions to relapse or death. The cumulative incidence functions have a
very similar pattern, and from Figure 13.7, the one-year cumulative incidence
of relapse or death from either of the two possible states is about one-third
that of platelet recovery, for patients in the baseline group.
1.0
Baseline cumulative hazard
0.8
0.6
0.4
0.2
0.0
0 1 2 3 4 5 6 7 8
Time since transplant (years)
Figure 13.6 Baseline cumulative hazard for the transitions 1 →2 (—), 1 →3 (- - -)

and 2→3 (·······).
1.0
Baseline cumulative incidence
0.8
0.6
0.4
0.2
0.0
0 1 2 3 4 5 6 7 8
Time since transplant (years)
Figure 13.7 Baseline incidence of transitions 1→2 (—), 1→3 (- - -) and 2→3 (·······).
FURTHER READING 455
Figure 13.6 suggests that the baseline hazard functions for the transitions
to relapse or death from either of the other two states may well be propor-
tional, although the hazard rate for the transition to a state of platelet recovery
has a different shape from the other two. Some further simplification of the
model is therefore possible, but this has little effect on the resulting inferences.

A number of texts describe counting process methods in survival analysis,
including Fleming and Harrington (2005), Aalen, Borgan and Gjessing (2008),
and Andersen et al. (1993). A mathematical treatment of models for survival
data, based on the counting process theory, is given by Andersen and Gill
(1982), while a more readable summary is contained in Gill (1984). See also
the comprehensive review paper of Andersen and Borgan (1985). The robust
sandwich estimate of the variance-covariance matrix of parameter estimates
in a Cox regression model was introduced by Lin and Wei (1989), and the link
with the delta-beta’s was described by Therneau and Grambsch (2000).
The three main models for recurrent and multiple events covered in this
chapter are due to Andersen and Gill (1982), Prentice, Williams and Petersen
(1981) and Wei, Lin and Weissfeld (1989). The text of Cook and Lawless
(2007) presents a comprehensive review of methodology for recurrent events,
and Therneau and Grambsch (2000) describe models for recurrent and mul-
tiple events in their text on extensions to the Cox regression model. Wei and
Glidden (1997) give an overview of statistical methods for the analysis of mul-
tiple events in clinical trials. Kelly and Lim (2000) give a careful discussion
of the different models for recurrent event data, concluding that the WLW
model is inappropriate. Arguments for and against the WLW model for the
analysis of recurrent event data are given by Metcalfe and Thompson (2007).
A thorough review of multistate models is given by Andersen and Keiding
(2002), and subsequent papers, in an issue of Statistical Methods in Medical
Research that is devoted to this topic. Other review articles include Com-
menges (1999), Hougaard (1999) and Meira-Machado et al. (2009). Hougaard
(2000) describes and illustrates event history models in his account of multi-
variate survival data. The tutorial article of Putter, Fiocco and Geskus (2007)
describe how the Cox regression model can be used in event history modelling,
and illustrates the approach with the EBMT data that was used in Example
13.4 of Section 13.4. A number of other articles have featured multistate mod-
els for patients who have a bone marrow transplant, including Keiding, Klein
and Horowitz (2001) and Klein and Shu (2002). Further examples of the use
of multistate models in medical research have been given by Hougaard and
Madsen (1985) and Andersen (1988).
The use of the R software for the analysis of competing risks and multistate
modelling is covered in the text of Beyersmann, Allignol and Schumacher
(2012), and by Putter, Fiocco and Geskus (2007). See also the special issue of
the Journal of Statistical Software, introduced by Putter (2011).
In order to use the methods presented in this section, the recurrence times
must be known. Multistate models that do not rely on the recurrence times
being known have been considered by many authors in connection with ani-
mal tumourigenicity experiments. In particular, see Dinse (1991), Kodell and
Nelson (1980), and McKnight and Crowley (1984). A useful review of this
literature is included in Lindsey and Ryan (1993).
Chapter 14
Dependent censoring
The methods described in this book for the analysis of censored survival data
are only valid if the censoring is independent or non-informative. Essentially,
this means that the censoring is not associated with the actual survival time,
so that individuals with censored survival times are representative of all oth-
ers who are at risk at that time, and who have the same values of measured
explanatory variables. For example, censoring would be considered to be in-
dependent if a censored time occurs because a patient has withdrawn from a
study on relocating to a different part of the country, or when survival data
are analysed at a prespecified time. On the other hand, if patients are with-
drawn because they experience life-threatening side effects from a particular
treatment, or if patients that do not respond to an active treatment are given
rescue medication, the censoring time is no longer independent of the survival
time. The censoring is then said to be dependent or informative. This chap-
ter shows how dependent censoring can be taken account of when modelling
survival data.
14.1 Identifying dependent censoring

Dependent censoring occurs when there is a dependence between the time to
an event such as death and the time to the occurrence of censoring. Since both
of these events cannot be observed in a given individual, it is not possible to use
the observed data to determine whether a data set has dependent censoring,
nor the extent of any such dependence. However, the context of a study can
often give some indication of whether or not there is likely to be dependent
censoring.
As an illustration, consider the double-blind clinical trial carried out by the
Acquired Immunodeficiency Syndrome (AIDS) Clinical Trials Group, known
as the ACTG 175 trial, and reported by Hammer et al. (1996). In this trial,
2467 patients were randomly assigned to one of four treatment regimens, and
the primary end-point was the time from randomisation to at least a 50%
reduction in the cluster of differentiation 4 (CD4) cell count, the development
of AIDS, or death. There were 1902 censored event times, with 933 of these
accounted for by the patient not having experienced any of the three events at
the end of the follow-up period, and so reflect independent administrative cen-
457
458 DEPENDENT CENSORING
soring. In addition, 283 discontinued treatment because of treatment toxicity,
and 380 left the study at the request of either the patient or the investigator.
It might be that patients in poor health would be more likely to experience
toxicity, and for these patients, their event times may be shorter. Similarly,
patients may have left the study to seek alternative treatments, and again
these patients may not be representative of the patient group as a whole. The
possibility of dependent censoring would therefore need to be considered when
analysing times to the primary end-point.
Statistical methods can be used to examine the assumption of independent
censoring in a number of ways. One approach is to plot observed survival times
against the values of explanatory variables, where the censored observations
are distinguished from the uncensored. If a pattern is exhibited in the censor-
ing, such as there being more censored observations at an earlier time on a
particular treatment, or if there is a greater proportion of censored survival
times in patients with a particular range of values of explanatory variables,
dependent censoring is suggested. More formally, a model could be used to
examine whether the probability of censoring is related to the explanatory
variables in the model. In particular, a linear logistic model could be used in
modelling a binary response variable that takes the value unity if an observed
survival time is censored and zero otherwise. If the probability of censoring is
found to depend on the values of certain explanatory variables, the assumption
of independent censoring may have been violated.
14.2 Sensitivity to dependent censoring
Even though dependent censoring is present, this feature may not necessarily
affect inferences of primary interest. It is therefore important to determine the
sensitivity of such inferences to the possibility of dependent censoring, and this
can be examined using two complementary analyses. In the first, we assume
that individuals who contribute censored observations are actually those at
high risk of an event. We therefore suppose that individuals for whom the
survival time is censored would have experienced the event immediately after
the censoring time, and designate all censored times as event times. In the
second analysis, we assume that the censored individuals are those at low risk
of an event. We then suppose that individuals with censored times experience
the event only after the longest observed survival time amongst all others in
the data set. The censored times are therefore replaced by the longest event
time. The impact of these two assumptions on the results of the analysis can
then be studied in detail. If essentially the same conclusions arise from the
original analysis and the two supplementary analyses, it will be safe to assume
that the results are not sensitive to the possibility of dependent censoring.
Another way to determine the potential impact of dependent censoring
is to obtain bounds on the survivor function in the presence of dependent
censoring. This has led to a number of proposals, but many of these lead to
SENSITIVITY TO DEPENDENT CENSORING 459
bounds that are too wide to be of practical value, and so this approach will
not be considered further.
Since the occurrence of dependent censoring cannot be identified without
additional information, it is useful to analyse the extent to which quantities
such as the risk score or median survival time are sensitive to the introduction
of some association between the time to failure and the time to censoring. A
sensitivity analysis for dependent censoring in parametric proportional haz-
ards models, described by Siannis, Copas and Lu (2005), will be outlined in
the next section.
14.2.1 ∗ A sensitivity analysis

Suppose that T is a random variable associated with the survival time and
C is a random variable associated with time to censoring. We will assume
Weibull proportional hazards models for both the hazard of death and the
hazard of censoring at time t. The two hazard functions for the ith individual,
i = 1, 2, . . . , n, for whom xi is a vector of values of p explanatory variables,
are respectively hi (t) and hci (t), where
hi (t) = exp(β ′ xi )h0 (t), hci (t) = exp(βc′ xi )hc0 (t).
The baseline hazard functions for the survival and censoring times are given
by
h0 (t) = λγtγ−1 , hc0 (t) = λc γc tγc −1 ,
respectively, in which λ, λc are scale parameters and γ, γc are shape parameters
in underlying Weibull distributions. The models for the survival time and
censoring time may contain different explanatory variables.
On fitting these models, the estimated values of β ′ xi and βc′ xi are the risk
score, β̂ ′ xi , and the censoring score, β̂c′ xi , respectively. A plot of the values
of the risk score against the corresponding censoring score will help determine
whether there is dependent censoring. In particular, if there is an association
between these two scores, dependent censoring is suggested.
We now introduce a parameter ϕ that measures the extent of dependence
between T and C. When ϕ = 0, censoring is independent of the survival times,
but as ϕ increases, censoring becomes more dependent on them. In fact, ϕ is
an upper bound to the correlation between T and C. Now let η̂(x0 ) denote
the risk score for the individual for whom x0 is the vector of values of the
explanatory variables in the model, and let η̂ϕ (x0 ) be the risk score for that
individual when ϕ measures the association between the survival time and
time to censoring. It can then be shown that an approximation to the change
in the risk score in a Weibull model that occurs when a small amount of
dependence between the time to failure and time to censoring variables, ϕ, is
assumed, is given by
∑n β̂c′ x0 tγ̂+γ̂c − (1 − δ )tγ̂ }
i=1 {e i i
B(x0 ) = η̂ϕ (x0 ) − η̂(x0 ) = ϕ i
∑n γ̂ , (14.1)
i=1 ti
where δi is an event indicator for the ith individual that is zero for a censored
observation and unity otherwise.
The sensitivity index, B(x0 ), only depends on the values of explanatory
variables through the censoring score, β̂c′ x0 , and Equation (14.1) shows that
the risk score is most sensitive to dependent censoring when the hazard of
censoring is greatest. A plot of B(x0 ) against a range of values of the censoring
score, β̂c′ x0 , for a given value of ϕ, provides information about the sensitivity
of the risk score to dependent censoring. Usually, ϕ is chosen to take values in
the range (−0.3, 0.3). This plot will indicate the values of the censoring score
that may result in dependent censoring having a non-negligible impact on the
risk score.
We can also examine the sensitivity of other summary statistics to depen-
dent censoring, such as a survival rate and the median survival time. For the
Weibull proportional hazards model, the estimated survivor function at time
t for the ith individual is
′
Ŝi (t) = {Ŝ0 (t)}exp(β̂ xi ) ,
where Ŝ0 (t) = λ̂tγ̂ is the estimated baseline survivor function. Using Equa-
tion (14.1), the estimated survivor function when there is dependent censoring
is approximately
′
{Ŝ (t)}exp{β̂ xi +B(xi )} ,
0
from which the impact of dependent censoring on survival rates can be deter-
mined, for a given value of ϕ. Similarly, the estimated median survival time
of an individual with vector of explanatory variables xi in the Weibull model
is { }1/γ̂
log 2
t̂(50) = ,
λ̂ exp(β̂ ′ xi )
and again using Equation (14.1), the estimated median when there is a degree
of dependence, ϕ, between the survival and censoring times is
{ }1/γ̂
log 2
t̂ϕ (50) = .
λ̂ exp[β̂ ′ xi + B(xi )]
An approximation to the relative reduction in the median survival time for
this individual is then
t̂(50) − t̂ϕ (50)
= 1 − exp{−γ̂ −1 B(xi )}, (14.2)
t̂(50)
for some value of ϕ. A plot of this quantity against a range of possible values
of the censoring score, β̂c′ xi , shows how the median survival time for individ-
uals with different censoring scores might be affected by dependent censoring.
The robustness of the estimated median to different amounts of dependent
censoring can also be explored by using a range of ϕ-values.
SENSITIVITY TO DEPENDENT CENSORING 461
14.2.2 Impact of dependent censoring
With independent censoring, individuals who are censored are representative
of the individuals at risk at the time of censoring, and estimated hazard ra-
tios and survivor functions obtained from a standard analysis of the time to
event data will be unbiased. On the other hand, if there is dependent cen-
soring, but it is assumed to be independent, model-based estimates may be
biased. The direction of this bias depends on whether there is positive or
negative association between the time to event and time to censoring vari-
ables. If there is a positive association between the two variables, those with
censored event times would be expected to experience a shorter event time
than those who remain at risk. Similarly, if there is a negative association,
individuals with censored event times may be those who would otherwise
have had a longer time before the occurrence of the event of interest. Stan-
dard methods for survival analysis would then lead to an overestimate or
underestimate of the survivor function, respectively, and the extent of the
bias will tend to increase as the number of dependently censored observations
increases.
Example 14.1 Time to death while waiting for a liver transplant

Once a patient has been judged to need a liver transplant they are added to the
registration list. However, the national shortage of livers means that a patient
may wait for some time for their operation, and a number die while waiting.
In a study to determine the mortality rate for patients registered for a liver
transplant, and the impact of certain factors on this outcome variable, data
were obtained from the UK Transplant Registry on the time from registration
to death on the list. The study cohort consisted of 281 adults with primary
biliary cirrhosis. This condition mainly affects females, and is characterised
by the progressive destruction of the small bile ducts of the liver, leading
to a build-up of bile in the liver, damage to liver tissue and cirrhosis. The
patients were first registered for a liver transplant in the five-year period from
1 January 2006, and the response variable of interest is the time from being
registered for a liver transplant until death on the list. For patients who receive
a transplant, the time from registration is censored at the time of transplant.
Times at which a patient are removed from the list because their condition
has deteriorated to the point where transplantation is no longer an option,
are regarded as death times.
In addition to the survival time from listing, information was available
on the age (years) and gender (male = 1, female = 0) of the patient, their
body mass index (BMI) (kg/m2 ) and the value of the UK end-stage liver
disease (UKELD) score, an indicator of disease severity where higher val-
ues correspond to a disease of greater severity. The first 20 observations in
this data set are given in Table 14.1, where the status variable is unity for
a patient who has died while waiting for a transplant and zero when their
time from listing has been censored. The data have been ordered by survival
time.
Table 14.1 Time from registration for a liver transplant until death while waiting.
Patient Time Status Age Gender BMI UKELD
1 1 0 60 0 24.24 60
2 2 1 66 0 30.53 67
3 3 0 71 0 26.56 61
4 3 0 65 1 23.15 63
5 3 0 62 0 22.55 64
6 4 1 56 0 36.39 73
7 5 0 52 0 24.77 57
8 5 0 65 0 33.87 49
9 5 1 58 0 27.55 75
10 5 1 57 0 22.10 64
11 6 0 62 0 21.60 55
12 7 0 56 0 25.69 66
13 7 0 52 0 32.39 59
14 8 1 45 0 28.98 66
15 9 0 50 0 31.67 60
16 9 0 65 0 24.67 57
17 9 0 44 0 24.34 64
18 10 0 67 0 22.65 61
19 12 0 67 0 26.18 57
20 13 0 57 0 22.23 53
Livers are generally allocated on the basis of need, and so tend to be offered
to those patients who are more seriously ill. As a result, patients who get a
transplant tend to be those who are nearer to death. The time to censoring
will then be associated with the time to death, and so the time from listing
until a transplant is dependent on the time from listing until death without a
transplant.
To illustrate the sensitivity analysis described in Section 14.2.1, consider a
Weibull model for the hazard of death at time t that contains the explanatory
variables that denote the age, gender, BMI and UKELD value of a patient
on the liver registration list. From a log-cumulative hazard plot, the Weibull
distribution fits well to the unadjusted survival times. The times to censoring
are also well fitted by a Weibull distribution, and the same four explanatory
variables will be included in this model. From these fitted models, the risk
score, β̂ ′ xi , and the censoring score, β̂c′ xi , can be obtained for each of the
281 patients, and a plot of the risk score against the censoring score for each
patient is given in Figure 14.1.
This figure shows that the risk score is positively correlated with the cen-
soring score, and so individuals that have a greater hazard of death, that is
those with a higher risk score, are more likely to be censored. This indicates
that there is dependent censoring in this data set.
To explore how this dependent censoring might affect the median survival
time, the relative change in the median is obtained for a moderate value of
MODELLING WITH DEPENDENT CENSORING 463
-2
-4
Risk score
-6
-8
-10
-12
-6.5 -6.0 -5.5 -5.0 -4.5
Censoring score
Figure 14.1 A plot of the values of the risk score, β̂ ′ xi against the censoring score,
β̂c′ xi , for each patient on the liver transplant waiting list.
ϕ, using Equation (14.2). A plot of the percentage relative reduction in the

median against the censoring score, β̂c′ xi , for values of β̂c′ xi in the range
(−6.5, −4.5) and ϕ = 0.3, corresponding to a moderate amount of dependent
censoring, is shown in Figure 14.2.
This figure shows that the relative reduction in the median survival time
is only near zero when the censoring score is around −6, corresponding to
individuals with the lowest hazard of censoring, that is the lowest chance of a
transplant. On the other hand, for individuals with a censoring score of −5.5,
which is close to the average censoring score for the 281 patients in the data
set, dependent censoring could decrease the median by about 30%. For those
at greatest risk of being censored, with a censoring score of at least −5, the
percentage decrease in median that might occur through dependent censoring
with ϕ = 0.3 is greater than 50%. An analysis based on the assumption of
independent censoring may therefore be seriously misleading. A technique that
allows account to be taken of dependent censoring in modelling survival data
is described in the next section.
14.3 Modelling with dependent censoring

In the presence of dependent censoring, standard methods for the analysis of
survival data must be modified. One possibility, useful when censoring does
not occur too early in patient time, is to analyse survival to a time before
any censoring has occurred. Alternatively, the probability of survival beyond
such a time could be modelled using logistic regression modelling, although
Percentage reduction in median 80
60
40
20
0
-6.5 -6.0 -5.5 -5.0 -4.5
Censoring score
Figure 14.2 Approximate percentage reduction in the median survival time from list-
ing for a liver transplant, as a function of the censoring score, when ϕ = 0.3.
this approach is unlikely to be useful when dependent censoring occurs early

in the study. A far better approach is to analyse the survival data using a Cox
regression model that directly accommodates dependent censoring.
14.3.1 Cox regression model with dependent censoring

Suppose that a Cox regression model is anticipated for data on the event
times of n individuals, some of which may be censored, so that the hazard of
an event at time t for the ith individual is
hi (t) = exp(β ′ xi )h0 (t),
where xi is the vector of values of p explanatory variables, β is the vector of

their coefficients, and h0 (t) is the baseline hazard function.
To allow for dependent censoring when fitting this model, we first develop
a model for the censored survival times. This is then used to weight the con-
tribution of the observed survival times to the partial likelihood function used
in the process of fitting the Cox regression model.
To summarise the basic idea of this technique, consider a particular in-
dividual, the ith say, who is at risk at an event time t. Suppose that the
probability that this individual’s survival time is censored at or after time t is
1/3. This means that on average, two other individuals who are identical to
the ith in terms of measured explanatory variables, will have survival times
that are censored before time t. If their survival times had not been censored,
these two individuals would have survived to at least t, as the ith individual
has done. Had all three of these individuals survived to t without being cen-
sored, each would have made the same contribution to the partial likelihood
function at time t. This can be modelled by weighting the contribution of the
ith individual to the partial likelihood at an event time t by 3, the reciprocal
of the censoring probability.
This process leads to Inverse Probability of Censoring Weighted (IPCW)
estimates of the unknown parameters in the model. The effect of this weight-
ing process is that greater weight is given to individuals that have a higher
probability of censoring before t, and so this accounts for censoring that is
associated with the survival time.
To calculate the weights used to adjust for dependent censoring, a model is
needed for the dependence of the probability of censoring at or after time t on
the values of measured explanatory variables. This is obtained from the data
by modelling the time to a censored observation by taking the time of censoring
as the end-point, and treating actual event times as censored observations. If
the data include an event indicator, δ, that is unity for an event and zero
for a censored observation, the corresponding indicator variable for censored
times is 1 − δ. Fitting a survival model to the censored survival times leads
to an estimated baseline survivor function, Ŝc0 (t). From this, the probability
of censoring occurring at or after time t in the ith individual, whose vector of
values of the explanatory variables is xi , is
′
Ŝci (t) = [Ŝc0 (t)]exp(β̂c xi ) , (14.3)
where β̂ c is the vector of estimated coefficients of the explanatory variables

in the censoring model. The weight for the ith individual at time t is wi (t) =
{Ŝci (t)}−1 , and these weights are then used in fitting a Cox regression model
for the time to an event.
The weighted partial likelihood function is found from the reformulation of
the Cox regression model in terms of a counting process. From an extension of
the formula given in Equation (13.2) of Chapter 13 to incorporate the weights
wi (t), this likelihood function is
n ∏ {
∏ }dNi (t)
wi (t)Yi (t) exp{β ′ xi }
∑n ′ , (14.4)
i=1 t >0 l=1 wl (t)Yl (t) exp{β xl }
where Yi (t) = 1 if an individual is at risk at time t and zero otherwise,

dNi (t) = 1 if an event occurs at time t and zero otherwise, and the inner
product is taken over all event times. Time-varying explanatory variables can
also be accommodated in this model.
The weights that are obtained using Equation (14.3) depend on the ex-
planatory variables in the censoring model, have different values at each event
time and change in value over the follow-up period for an individual. Because
of this, it is more convenient to model the censoring times using a parametric
model, as the weights are then a continuous function of t. The precise form of
the chosen model is not that important and the Weibull model is suggested
for general use. The hazard of censoring at time t for the ith individual is then
hci (t) = exp(βc′ xi )hc0 (t), where hc0 (t) = λc γc tγc −1 , and xi is the vector of
values of explanatory variables in the censoring model for the ith individual.
The corresponding survivor function is
Sci (t) = exp{−λc exp(βc′ xi )tγc }.
Note that the variables in the model for the censored survival times need not
be the same as those in the model for the survival times. In terms of the log-
linear formulation of Weibull model, described in Section 5.6.3 of Chapter 5,
{ [ ]}
log t − µc − αc′ xi
Sci (t) = exp − exp ,
σc
in which µc is the ‘intercept’, σc the ‘scale’, αcj is the coefficient of the jth
explanatory variable and λc = exp(−µc /σc ), γc = 1/σc , βcj = −αcj /σc , for
j = 1, 2, . . . , p.
To handle the time dependence of the weights, the data need to be ex-
pressed in counting process format, using the (start, stop, status) notation
that was described in Section 13.1.3 of Chapter 13. The stop times are taken
to be the event times in the data set.
Example 14.2 Data format for dependent censoring

Suppose that the first three ordered event times in a data set are 18, 55 and
73 days. The observed survival time of 73 days in counting process format is
taken to have the intervals (0, 18], (18, 55], (55, 73], where the status indicator
is 0 for the first two intervals and 1 for the third. If the survival time of 73 days
was censored, the status would be 0 for the third interval. Suppose further
that the model-based probability of censoring at or beyond times 18, 55 and
73, obtained using a model for the censored survival times, is 0.94, 0.82 and
0.75, respectively. The weights that are used in fitting Cox regression models
to the data based on the counting process format are then 1.064, 1.220 and
1.333, respectively.
The weights, wi (t), calculated from Equation (14.3), can get quite large
when the probability of censoring beyond t is small, and this can lead to com-
putational problems. It can then be more efficient to use stabilised weights,
wi∗ (t), calculated using wi∗ (t) = ŜKM (t)/Ŝci (t), where ŜKM (t) is the Kaplan-
Meier estimate of the probability of censoring after t. Using the term ŜKM (t)
in the numerator of the weights has no effect on parameter estimates, since
ŜKM (t) is independent of explanatory variables in the model and cancels out
in the numerator and denominator of the partial likelihood function in Equa-
tion (14.4). However, it does lead to greater stability in the model fitting
process.
Finally, to account for the additional uncertainty in the specification of
the model, a robust estimate of the variance-covariance matrix of the param-
eter estimates is recommended, such as the sandwich estimate introduced in
Section 13.1.4 of Chapter 13.
Example 14.3 Time to death while waiting for a liver transplant

The data from Example 14.1 on the survival times from registration for a
liver transplant are now analysed using a modelling approach that makes
allowance for any dependent censoring. To model the probability of censoring,
a Weibull model is fitted to the time from registration until censoring, where
the censoring indicator is zero when the status variable in Table 14.1 is unity,
and vice-versa. Using the log-linear formulation of the Weibull model, the
estimated probability of censoring at or beyond time t is
{ [ ]}
log t − µ̂c − α̂c1 x1i − α̂c2 x2i − α̂c3 x3i − α̂c4 x4i
Ŝci (t) = exp − exp ,
σ̂c
(14.5)
where x1i is the age, x2i the gender, x3i the BMI, and x4i the UKELD score
at registration for the ith patient. The estimated parameters and their stan-
dard errors in this model are shown in Table 14.2. At this stage, the model
could be modified by excluding variables that have no significant effect on the
probability of censoring, namely age, gender and BMI, but we will continue
to use a censoring model that contains all four variables.
Table 14.2 Parameter estimates and their standard errors in

a Weibull model for the censoring time.
Variable Parameter Estimate se (Estimate)
Age α̂c1 0.0070 0.0079
Sex α̂c2 −0.1839 0.1950
BMI α̂c3 0.0193 0.0141
UKELD α̂c4 −0.0535 0.0152
Intercept µ̂c 7.6261 1.0058
Scale σ̂c 0.9813 0.0507
Next, the data in Table 14.1 are expressed in the counting process format
and weights are calculated from the reciprocals of the censoring probabilities at
the ‘stop’ times, from the inverse of the estimated censoring probability, Ŝci (t)
in Equation (14.5). Data for the first 10 patients from Table 14.1 are shown
in the counting process format in Table 14.3, together with the censoring
probabilities and weights that are used in fitting a Cox regression model to
the survival times.
A weighted Cox regression model that contains the same four explanatory
variables is then fitted, and the estimated hazard of death at time t for the
ith patient is
ĥi (t) = exp{β̂1 x1i + β̂2 x2i + β̂3 x3i + β̂4 x4i }ĥ0 (t),
where ĥ0 (t) is the estimated baseline hazard function. The parameter esti-
mates and their standard errors in the weighted Cox regression model that
Table 14.3 Data from the first 10 patients in Table 14.1 in the counting process
format.
Patient Start time Stop time Status Censoring probability Weight
1 0 1 0 0.9955 1.0045
2 0 2 1 0.9888 1.0114
3 0 2 0 0.9915 1.0085
3 2 3 0 0.9872 1.0129
4 0 2 0 0.9873 1.0128
4 2 3 0 0.9809 1.0195
5 0 2 0 0.9885 1.0116
5 2 3 0 0.9827 1.0176
6 0 2 0 0.9851 1.0151
6 2 4 1 0.9701 1.0308
7 0 2 0 0.9919 1.0081
7 2 4 0 0.9837 1.0166
7 4 5 0 0.9796 1.0208
8 0 2 0 0.9960 1.0040
8 2 4 0 0.9919 1.0081
8 4 5 0 0.9899 1.0102
9 0 2 0 0.9806 1.0198
9 2 4 0 0.9610 1.0405
9 4 5 1 0.9513 1.0511
10 0 2 0 0.9880 1.0121
10 2 4 0 0.9758 1.0248
10 4 5 1 0.9698 1.0312
takes account of dependent censoring are shown in Table 14.4. Also shown
are the corresponding unweighted estimates when dependent censoring is not
taken into account. The sandwich estimate of the variance-covariance matrix
of the parameter estimates has been used in both cases, but this only makes
a very small difference to the standard errors.
Table 14.4 Parameter estimates and their standard errors in a weighted and un-
weighted Cox regression model.
Variable Parameter Weighted Unweighted
Estimate se (Estimate) Estimate se (Estimate)
Age β̂1 0.0118 0.0316 0.0774 0.0200
Sex β̂2 −0.9895 0.6549 −0.1471 0.4944
BMI β̂3 −0.0218 0.0492 0.0236 0.0210
UKELD β̂4 0.1559 0.0427 0.2162 0.0276
The two sets of estimates are somewhat different, which shows that
the adjustment for dependent censoring has affected the hazard ratios. In
the unweighted analysis, both age and UKELD score are highly significant
(P < 0.001), whereas age ceases to be significant in the weighted analysis.
From the hazard ratio for UKELD after adjustment for dependent censor-
ing, a unit increase in the UKELD leads to a 17% increase in the hazard
of death.
To illustrate the effect of the adjustment for dependent censoring, Fig-
ure 14.3 shows the estimated survivor functions for a female patient aged 50
with a UKELD score of 60 and a BMI of 25, from a weighted and unweighted
Cox regression model for the hazard of death.
1.0
0.8
0.6
0.4
0.2
0.0
0 200 400 600 800
Days from registration
Figure 14.3 Estimated survivor functions in a weighted (·······) and unweighted (—)
Cox regression model for a 50-year-old female with a UKELD score of 60 and a BMI
of 25.
This figure shows how survival rates are overestimated if account is not
taken of dependent censoring. In particular, if no allowance is made for de-
pendent censoring, the survival rate at six months is estimated to be 77%,
but after taking account of dependent censoring, the estimate is 65%. In ad-
dition, the 80% survival rate is overestimated by nearly two months. Fail-
ure to take account of the dependent censoring can therefore result in mis-
leading estimates of the waiting list mortality for patients awaiting a liver
transplant.
In an extension to this analysis, separate models could be entertained for
different causes of censoring, namely transplant and removal from the regis-
tration list because of a deteriorating condition, as well as for any independent
censoring. Additionally, variation in the UKELD score over time could also be
incorporated so that changes in disease severity over the registration period
can be accounted for.
Bounds for the survivor function in the presence of dependent censoring have
been described by a number of authors, including Peterson (1976), Slud and
Rubinstein (1983) and Klein and Moeschberger (1988). Tsiatis (1975) explains
why the extent of association between event times and censoring times cannot
be estimated from observed data, and Siannis (2004) and Siannis, Copas and
Lu (2005) have described methods for determining the sensitivity of infer-
ences to dependent censoring in parametric models. This approach has been
extended to the Cox proportional hazards model in Siannis (2011), although
the method is computationally intensive.
Models for data with dependent censoring have been described by Wu
and Carroll (1988) and Schlucter (1992). Inverse probability of censoring
weighted estimators were introduced by Robins and Rotnitzky (1992) and
Robins (1993). Robins and Finkelstein (2000) showed how these estimators
could be used to adjust Kaplan-Meier estimates to take account of dependent
censoring. Satten, Datta and Robins (2001) and Scharfstein and Robins (2002)
showed how to estimate the survivor function for a Cox regression model in
the presence of dependent censoring.
Chapter 15
Sample size requirements for a

survival study
There are many aspects of the design of a medical research programme that
need to be considered when the response variable of interest is a survival time.
These include factors such as the inclusion and exclusion criteria for study
participants, the unambiguous definition of the time origin and the end-point
of the study, and the duration of patient follow-up. In a clinical trial, the
specification of treatments, the method of randomisation to be employed in
allocating patients to treatment group, and the use of blinding must also be
specified. Consideration might also be given to whether the study should be
based on a fixed number of patients, or whether a sequential design should
be adopted, in which the study continues until there is a sufficient number
of events to be able to distinguish between treatments. The need for interim
analyses, or adaptive designs that allow planned modifications to be made to
the sample size or allocated treatment as data accumulate, also needs to be
discussed.
Many of these considerations are not unique to studies where survival is
the outcome of interest, and are discussed in a number of texts on the design
and analysis of clinical trials, such as Friedman, Furberg and DeMets (2010),
Matthews (2006) and Pocock (1983). However, there is one matter in the de-
sign of fixed sample size studies that will be discussed here. This is the crucial
issue of the number of patients that are required in a survival study. If too
few patients are recruited, there may be insufficient information available in
the data to enable a treatment difference to be pronounced significant. On the
other hand, it is unethical to waste resources in studies that are unnecessarily
large. Sample size calculations for survival data are presented in this chapter.
15.1 Distinguishing between two treatment groups

Many survival studies are concerned with distinguishing between two alter-
native treatments. For this reason, a study to compare the survival times of
patients who receive a new treatment with those who receive a standard will
be used as the focus for this chapter. The same formulae can of course be used
in other situations and for other end-points.
471
472 SAMPLE SIZE REQUIREMENTS FOR A SURVIVAL STUDY
Suppose that in this study, there are two groups of patients, and that the
standard treatment is allocated to the patients in Group I, while the new
treatment is allocated to those in Group II. Assuming a proportional hazards
model for the survival times, the hazard of death at time t for a patient on
the new treatment, hN (t), can be written as
hN (t) = ψhS (t),
where hS (t) is the hazard function at t for a patient on the standard treatment
and ψ is the unknown hazard ratio. We will also define θ = log ψ to be the log-
hazard ratio. If θ is zero, there is no treatment difference. On the other hand,
negative values of θ indicate that survival is longer under the new treatment,
while positive values of θ indicate that patients survive longer on the standard
treatment.
In order to test the null hypothesis that θ = 0, the log-rank test described
in Section 2.6 can be used. As was shown in Section 3.13, this is equivalent
to using the score test of the null hypothesis of equal hazards in the Cox
regression model. In this chapter, sample size requirements will be based on
the log-rank test statistic, but the formulae presented can also be used when
an analysis based on the Cox regression model is envisaged.
In a survival study, the occurrence of censoring means that it is not usu-
ally possible to measure the actual survival times of all patients in the study.
However, it is the number of actual deaths that is important in the analysis,
rather than the total number of patients. Accordingly, the first step in deter-
mining the number of patients in a study is to calculate the number of deaths
that must be observed. We then go on to determine the required number of
patients.
15.2 Calculating the required number of deaths

To determine the sample size requirement for a study, we calculate the num-
ber of patients needed for there to be a certain chance of declaring θ to be
significantly different from zero when the true, but unknown, log-hazard ratio
is θR . Here, θR is the reference value of θ. It will be a reflection of the mag-
nitude of the treatment difference that it is important to detect, using the
test of significance. In a study to compare a new treatment with a standard,
there is likely to be a minimum worthwhile improvement and a maximum
envisaged improvement. The actual choice of θR will then lie between these
two values. In practice, θR might be chosen on the basis of the desired hazard
ratio, the change in the median survival time that is to be detected, or the
difference in the probability of survival beyond a given time. This is discussed
and illustrated later in Example 15.1.
More formally, the required number of deaths is taken to be such that
there is a probability of 1 − β of declaring the observed log-hazard ratio to
be significantly different from zero, using a hypothesis test with a specified
significance level of α, when in fact θ = θR . The term 1 − β is the probability
CALCULATING THE REQUIRED NUMBER OF DEATHS 473
of rejecting the null hypothesis when it is in fact false, and is known as the
power of the test. The quantity β is the probability of not rejecting the null
hypothesis when it is false and is sometimes known as the type II error. Both
α and β are taken to be small. Typical values will be α = 0.05 and β = 0.1,
and with these values there would be a 90% chance of declaring the observed
difference between two treatments to be significant at the 5% level. The exact
specification of α and β will to some extent depend on the circumstances. If
it is important to detect a difference as being significant at a lower level of
significance, or if there needs to be a higher chance of declaring a result to be
significant, α and β will need to be modified accordingly.
The required number of deaths in a survival study, d, can be obtained from
the equation
4(zα/2 + zβ )2
d= 2 , (15.1)
θR
where zα/2 and zβ are the upper α/2- and upper β-points, respectively, of the
standard normal distribution. It is convenient to write c(α, β) = (zα/2 + zβ )2
in Equation (15.1), giving
2
d = 4c(α, β)/θR . (15.2)
The values of c(α, β) for commonly chosen values of the significance level α
and power 1 − β are given in Table 15.1.
Table 15.1 Values of the function c(α, β).

Value of α Value of 1 − β
0.80 0.90 0.95 0.99
0.10 6.18 8.56 10.82 15.77
0.05 7.85 10.51 13.00 18.37
0.01 11.68 14.88 17.81 24.03
0.001 17.08 20.90 24.36 31.55
Calculation of the required number of deaths then requires that a value

for θR be identified, and appropriate values of α and β chosen. Table 15.1 is
then used in conjunction with Equation (15.2) to give the number of deaths
required in a study.
The derivation of the result in Equation (15.2) assumes that the same
number of individuals is to be assigned to each treatment group. If this is not
the case, a modification has to be made. In particular, if the proportion of
individuals to be allocated to Group I is π, so that a proportion 1 − π will be
allocated to Group II, the required total number of deaths becomes
c(α, β)
d= 2 .
π(1 − π)θR
Notice that an imbalance in the number of individuals in the two treatment

groups leads to an increase in the total number of deaths required. The deriva-
tion also includes an approximation, which means that the calculated number
of deaths could be an underestimate. Some judicious rounding up of the cal-
culated value is therefore suggested to compensate for this.
The actual derivation of the formula for the required number of deaths is
important and so details are given below in Section 15.2.1. This section can be
omitted without loss of continuity. It is followed by an example that illustrates
the calculations.
15.2.1 ∗ Derivation of the required number of deaths

An expression for the required number of deaths is now derived on the basis of
a log-rank test to compare two treatment groups. As in Section 2.6, suppose
that there are r distinct death times, t(1) < t(2) < · · · < t(r) , among the
individuals in the study, and that in the ith group there are dij deaths at the
jth ordered death time t(j) , for i = 1, 2 and j = 1, 2, . . . , r. Also suppose that
the number at risk at t(j) in the ith group is nij , and write nj = n1j + n2j for
the total number at risk at t(j) and dj = d1j + d2j for the number who die at
t(j) . The log-rank statistic is then
∑
r
U= (d1j − e1j ),
j=1
where e1j is the expected number of deaths in Group I at t(j) , given by e1j =
n1j dj /nj , and the variance of the log-rank statistic is
∑r
n1j n2j dj (nj − dj )
V = . (15.3)
j=1
n2j (nj − 1)
When using the log-rank test, the null hypothesis that θ = 0 is rejected
if the absolute value of U is sufficiently large, that is, if |U | > k, say, where
k > 0 is a constant. We therefore require that
P(|U | > k; θ = 0) = α, (15.4)
and
P(|U | > k; θ = θR ) = 1 − β,
for a two-sided 100α% significance test to have power 1 − β.
We now quote without proof a result given in Sellke and Siegmund (1983),
according to which the log-rank statistic, U , has an approximate normal dis-
tribution with mean θV and variance V , for small values of θ. Indeed, the
result that U ∼ N (0, V ) under the null hypothesis θ = 0, is used as a basis
for the log-rank test. Then, since
P(|U | > k; θ = 0) = P(U > k; θ = 0) + P(U < −k; θ = 0),

and U has an N (0, V ) distribution when θ = 0, a distribution that is sym-
metric about zero,
P(U > k; θ = 0) = P(U < −k; θ = 0).
It then follows from Equation (15.4) that

α
P(U > k; θ = 0) = . (15.5)
2
Next, we note that
P(|U | > k; θ = θR ) = P(U > k; θ = θR ) + P(U < −k; θ = θR ).
For the sort of values of k that are likely to be used in the hypothesis test,
either P(U < −k; θ = θR ) or P(U > k; θ = θR ) will be negligible. For example,
if the new treatment is expected to increase survival so that θR is taken to be
less than zero, the probability of U having a value in excess of k, k > 0, will
be small. So without loss of generality we will take
P(|U | > k; θ = θR ) ≈ P(U < −k; θ = θR ).
We now denote the upper 100p% point of the standard normal distribu-
tion by zp . Then Φ(zp ) = 1 − p, where Φ(·) stands for the standard normal
distribution function. The quantity Φ(zp ) therefore represents the area under
a standard normal density function to the left of the value zp . Now, since
U ∼ N (0, V ) when θ = 0,
( )
k
P(U > k; θ = 0) = 1 − P(U 6 k; θ = 0) = 1 − Φ √ ,
(V )
and using Equation (15.5) we have that
( )
k
Φ √ = 1 − (α/2).
(V )
Therefore,
k
√ = zα/2 ,
(V )
where zα/2 is the upper α/2-point of the standard normal distribution, and
so k can be expressed as √
k = zα/2 (V ). (15.6)
In a similar manner, since U ∼ N (θR V, V ) when θ = θR ,
( )
−k − θR V
P(U < −k; θ = θR ) = Φ √ ≈ 1 − β,
(V )
and so we take
−k − θR V
√ = zβ ,
(V )
where zβ is the upper β-point of the standard normal distribution. If we now
substitute for k from Equation (15.6), we get
√ √
−zα/2 (V ) − θR V = zβ (V ),
and so V needs to be such that
V = (zα/2 + zβ )2 /θR
2
, (15.7)
to meet the specified requirements.

When the number of deaths is few relative to the number at risk, the
expression for V in Equation (15.3) is approximately
∑
r
n1j n2j dj
. (15.8)
j=1
n2j
Moreover, if θ is small, and recruitment to each treatment group proceeds at

a similar rate, then n1j ≈ n2j , for j = 1, 2, . . . , r, and so
n1j n2j n1j n2j n21j 1

2 = ≈ = .
nj (n1j + n2j )2 (2n1j )2 4
Then, V is given by
∑
r
V ≈ dj /4 = d/4,
j=1
∑r
where d = j=1 dj is the total number of deaths among the individuals in
the study.
Finally, using Equation (15.7), we now require d to be such that
d (zα/2 + zβ )2
= 2 ,
4 θR
which leads to the required number of deaths being that given in Equa-
tion (15.1).
At later death times, that is, when the values of j in Expression (15.8) are
close to r, the numbers of individuals at risk in the two groups will be small.
This is likely to mean that n1j and n2j will be quite different at the later
death times, and so n1j n2j /n2j will be less than 0.25. This in turn means that
V < d/4 and so the required number of deaths will tend to be underestimated.
Example 15.1 Survival from chronic active hepatitis

Patients suffering from chronic active hepatitis rapidly progress to an early
death from liver failure. A new treatment has become available and so a clinical
trial is planned to evaluate the effect of this new treatment on the survival
times of patients suffering from the disease. As a first step, information is
obtained on the survival times in years of patients in a similar age range
0.8
0.6
0.4
0.2
0.0
0 2 4 6 8 10
Survival time
Figure 15.1 Estimated survivor function for patients receiving a standard treatment
for hepatitis.
who have received the standard therapy. The Kaplan-Meier estimate of the
survivor function derived from such data is shown in Figure 15.1.
From this estimate of the survivor function, the median survival time is
3.3 years, and the survival rates at two, four and six years can be taken to be
given by S(2) = 0.70, S(4) = 0.45, and S(6) = 0.25.
The new treatment is expected to increase the survival rate at five years
from 0.41, the value under the standard treatment, to 0.60. This information
can be used to calculate a value for θR . To do this, we use the result that if
the hazard functions are assumed to be proportional, the survivor function
for an individual on the new treatment at time t is
SN (t) = [SS (t)]ψ , (15.9)
where SS (t) is the survivor function for an individual on the standard treat-
ment at t and ψ is the hazard ratio. Therefore,
log SN (t)
ψ= ,
log SS (t)
and so the value of ψ corresponding to an increase in S(t) from 0.41 to 0.60 is
log(0.60)
ψR = = 0.57.
log(0.41)
With this information, the survivor function for a patient on the new treat-
ment is [SS (t)]ψR , and so SN (2) = 0.82, SN (4) = 0.63, and SN (6) = 0.45. A
plot of the two survivor functions is shown in Figure 15.2.
1.0
0.8
0.6
0.4
0.2
0.0
0 2 4 6 8 10
Survival time
Figure 15.2 Estimated survivor functions for individuals on the standard treatment
(—) and the new treatment (·······).
The median survival time under the new treatment can be determined
from this estimate of the survivor function. Using Figure 15.2, the median
survival time under the new treatment is estimated to be about six years. A
hazard ratio of 0.57 therefore implies an increase in median survival time from
3.3 years on the standard treatment to 6 years on the new treatment.
To calculate the number of deaths that would be required in a study to
compare the two treatments, we will take α = 0.05 and 1 − β = 0.90. With
these values of α and β, the value of the function c(α, β) from Table 15.1
is 10.51. Substituting for c(0.05, 0.1) in Equation (15.2) and taking θR =
log ψR = log(0.57) = −0.562, the number of deaths required to have a 90%
chance of detecting a hazard ratio of 0.57 to be significant at the 5% level is
given by
4 × 10.51
d= = 133.
0.5622
Allowing for possible underestimation, this can be rounded up to 140 deaths
in total. This means that approximately 70 deaths would need to be observed
in each treatment group.
The treatment difference that it is required to detect may also be expressed
in terms of the desired absolute or relative change in the median survival time.
The corresponding log-hazard ratio, for use in Equation (15.1), can then be
found by reversing the preceding calculation. For example, suppose that an
increase in the median survival time from 3.3 years on the standard treatment
to just 5 years on the new treatment is anticipated. The survivor function
on the new treatment is then 0.5 when t = 5, and using Equation (15.9),
CALCULATING THE REQUIRED NUMBER OF PATIENTS 479
SN (5) = {SS (5)}ψR = 0.5. Consequently, ψR log{SS (5)} = log 0.5, and since
SS (5) = 0.41, ψR = 0.78. This reflects a less optimistic view of the treatment
effect than when ψR is taken to be 0.57. The corresponding number of deaths
that would need to be observed for this hazard ratio to be declared significantly
different from unity at the 5% level, with 90% power, is then around 680. This
is considerably greater than the number needed to identify a hazard ratio of
0.57 as significant.
The calculations described above are only going to be of direct use when a
study is to be continued until a given number of those entering the study have
died. Most trials will be designed on the basis of the number of patients to be
recruited and so we must now examine how this number can be obtained.
15.3 Calculating the required number of patients

In order to calculate the actual number of patients that are required in a
survival study, we need to consider the probability of death over the duration
of a study. Typically, patients are recruited over an accrual period of length a.
After recruitment is complete, there is an additional follow-up period of length
f . The total duration of a study will therefore be of length a + f . Notice that
if f is small, or even zero, there will need to be correspondingly more patients
recruited in order to achieve a specific number of deaths.
Once the probability of a patient dying in the study has been evaluated,
the required number of patients will be found from
d
n= , (15.10)
P(death)
where d is the required number of deaths found from Equation (15.2). Ac-
cording to a result derived in the next section, the probability of death can
be taken as
1
P(death) = 1 − {S̄(f ) + 4S̄(0.5a + f ) + S̄(a + f )}, (15.11)
6
where
SS (t) + SN (t)
S̄(t) = ,
2
and SS (t) and SN (t) are the estimated values of the survivor functions for
individuals on the standard and new treatments, respectively, at time t.
The above result shows how the required number of patients can be cal-
culated for a trial with an accrual period of a and a follow-up period of f . Of
course, the duration of the accrual period and follow-up period will depend
on the recruitment rate. So suppose that the recruitment rate is expected to
be m patients per month and that d deaths are required. If n patients are
to be entered into the study over a period of a months, this means that n/a
need to be recruited in each month. In practice, information is likely to be
available on the accrual rate, m, that can be expected. The number recruited
in an accrual period of length a is then ma and so the expected number of
deaths in the study is
ma × P(death).
Values of a and f which make this value close to the number of deaths required
can then be found numerically, for example, by trying out different values of
a and f . This algorithm could be computerised and an optimisation method
used to find the value of a that makes
d − {ma × P(death)} (15.12)
close to zero for a range of values of f. Alternatively, the value of f that yields
the result in Equation (15.12) for a range of values of a can be found. A two-
way table giving the required number of patients for different combinations of
values of a and f will be particularly useful in planning a study.
The following section gives details underlying the derivation of the result
in Equation (15.11), and can again be omitted without loss of continuity.
15.3.1 ∗ Derivation of the required number of patients

We begin with the general result from distribution theory that the marginal
probability of a patient dying during the course of a study can be obtained
from the joint probability of death and entry to the study at time t using
∫ a
P(death) = P(death and entry at time t) dt. (15.13)
0
The joint probability can in turn be found from the result
P(death and entry at time t) = P(death | entry at t) × P(entry at t),

(15.14)
which is simply a version of the result that P(A | B) = P(AB)/P(B).
We now assume a uniform recruitment rate over the accrual period. The
distribution of entry times to the study can then be taken to be uniform
over the time interval (0, a). Therefore, the probability of an individual being
recruited to the study at time t is a−1 , for any value of t in the interval (0, a).
From Equations (15.13) and (15.14), we have
∫ a
P(death) = P(death | entry at t)a−1 dt,
0
so that ∫ a
1
P(death) = 1 − P(survival | entry at t) dt.
a 0
A patient entering the study at time t who survives for the duration of the
study, that is, to time a + f , must have been alive for a period of length
a + f − t after entry. The conditional probability P(survival | entry at t) is
therefore the probability of survival beyond a + f − t. This probability is
the value of the survivor function for that individual at a + f − t, that is,
S(a + f − t). Consequently,
∫
1 a
P(death) = 1 − S(a + f − t) dt,
a 0
and on writing u = a + f − t, this result becomes

∫ a+f
1
P(death) = 1 − S(u) du. (15.15)
a f
The integral of the survivor function is now approximated using numerical

integration. According to Simpson’s rule,
∫ v { ( ) }
v−u u+v
f (x) dx ≈ f (u) + 4f + f (v) ,
u 6 2
so that
∫ a+f
a
S(u) du ≈ {S(f ) + 4S(0.5a + f ) + S(a + f )} ,
f 6
and hence, using Equation (15.15), the probability of death during the study
is given by
1
P(death) = 1 − {S(f ) + 4S(0.5a + f ) + S(a + f )} .
6
From this result, the approximate probability of death for an individual in
Group I, for whom the survivor function is SS (t), is
1
P(death; Group I) = 1 − {SS (f ) + 4SS (0.5a + f ) + SS (a + f )} ,
6
and similarly that for an individual in Group II is
1
P(death; Group II) = 1 − {SN (f ) + 4SN (0.5a + f ) + SN (a + f )} .
6
On the assumption that there is an equal probability of an individual being
assigned to either of the two treatment groups, the overall probability of death
is the average of these two probabilities, so that
P(death; Group I) + P(death; Group II)

P(death) = .
2
On substituting for the probabilities of death in the two treatment groups,
and setting S̄(t) = {SS (t) + SN (t)}/2, the overall probability of death can be
written
1
P(death) = 1 − {S̄(f ) + 4S̄(0.5a + f ) + S̄(a + f )},
6
as in Equation (15.11).
If the proportion of individuals to be allocated to Group I is π, the overall
probability of death becomes
π P(death; Group I) + (1 − π) P(death; Group II),
and the result for the overall probability of death given in Equation (15.11)
can be modified accordingly.

In Example 15.1, it was shown that 140 deaths needed to be observed for
the study on chronic hepatitis to have sufficient power to detect a hazard
ratio of 0.57 as significant, at the 5% level. Suppose that patients are to be
recruited to the study over an 18-month accrual period and that there is to
be a subsequent follow-up period of 24 months. From Equation (15.11), the
probability of death in the 42 months of the study will then be given by
1{ }
P(death) = 1 − S̄(24) + 4S̄(33) + S̄(42) .
6
Now, using the estimated survivor functions shown in Figure 15.2,
SS (24) + SN (24) 0.70 + 0.82
S̄(24) = = = 0.76,
2 2
SS (33) + SN (33) 0.57 + 0.73
S̄(33) = = = 0.65,
2 2
SS (42) + SN (42) 0.45 + 0.63
S̄(42) = = = 0.54,
2 2
and so the probability of death is
1
1 − {0.76 + (4 × 0.65) + 0.54} = 0.350.
6
From Equation (15.10), the required number of patients is
140
n= = 400,
0.350
and so 400 patients will need to be recruited to the study over the accrual
period of 18 months. This demands a recruitment rate of about 22 patients
per month.
If it is only expected that 18 patients can be found each month, the accrual
period will need to be extended to ensure that there is a sufficient number of
individuals to give the required number of deaths. The number of individuals
that could be recruited in a period of a months would be 18a. Various values
of a can then be tried in order to make this approximately equal to the value
obtained from Equation (15.10). For example, if we take a = 24 and continue
with f = 24, the probability of death over the four years of the study is
1
P(death) = 1 − {S̄(24) + 4S̄(36) + S̄(48)}.
6
From Figure 15.2, the survivor functions for patients on each treatment at
24, 36 and 48 months can be estimated, and we find that S̄(24) = 0.76,
S̄(36) = 0.65, and S̄(48) = 0.50. The probability of death then turns out to
be 0.357 and the required number of patients to give 140 deaths is now 393.
This is broadly consistent with an estimated recruitment rate of 18 per month.
Now suppose that it is decided that the study will not have a follow-up
period, so that the accrual period is equal to the duration of the study. If
the accrual period is taken to be 20 months, so that a = 20 and f = 0, the
probability of death is given by
1
P(death) = 1 − {S̄(0) + 4S̄(10) + S̄(20)}.
6
Now, S̄(0) = 1.00, S̄(10) = 0.82, and S̄(20) = 0.79, and the probability of
death is 0.155. The required number of patients is now 140/0.155 = 903, and
this would just about be met by a recruitment rate of 45 patients per month.
This shows that the absence of a follow-up period leads to an increase in the
number of patients that must be entered into the study.
15.3.2 An approximate procedure

A much simpler procedure for calculating the required number of patients in
a survival study to compare two treatments is outlined in this section. The
basis for this result is that {SS (τ ) + SN (τ )}/2 is the average probability that
a patient in the study survives beyond time τ , where SS (τ ) and SN (τ ) are the
survivor functions at time τ , for patients on the standard and new treatments,
respectively. The probability of death, in the period from the time origin to
τ , can then be approximated by
SS (τ ) + SN (τ )
1− .
2
Using Equation (15.10), the required number of patients becomes
2d
n= ,
2 − SS (τ ) − SN (τ )
where d is the required number of deaths.

A natural choice for the value of τ to use in this calculation is the average
length of the follow-up period for patients in the study, f + (a/2), where
a is the accrual period and f the follow-up period. This procedure is very
approximate because it does not take account of patient follow-up times that
extend beyond τ . As a consequence, this result will tend to overestimate the
required sample size.

The study on chronic active hepatitis is now used to illustrate the approxi-
mate procedure for determining the required number of patients. As in Ex-
ample 15.2, suppose that patients are recruited over an 18-month period and
that there is a further follow-up period of 24 months. The average length of
the follow-up period for a patient will then be τ = f + (a/2) = 33 months.
From Figure 15.2, the survivor functions for patients in the two groups are
given by SS (33) = 0.57 and SN (33) = 0.73, respectively. The approximate
probability of death in the period from 0 to 33 months is therefore 0.35 and
the number of patients required to give 140 deaths is 140/0.35 = 400. In this
illustration, the sample size produced by the approximate result is identical
to that found using the more complicated procedure illustrated in Example
15.2, but this will not usually be the case.
Full details on the issues to be considered when designing a clinical trial are
given in Friedman, Furberg and DeMets (2010), Matthews (2006) and Pocock
(1983) and the more general text of Altman (1991). Whitehead (1997) de-
scribes how a sequential clinical trial can be designed when the outcome vari-
able of interest is a survival time, and Jennison and Turnbull (2000) describe
group sequential methods.
Extensive tables of sample size requirements in studies involving different
types of response variable, including survival times, are provided by Machin
et al. (2009) and Julious (2010). A number of commercially available soft-
ware packages for sample size calculations, including PASS (Power Analysis
and Sample Size program) and nQuery Advisor, also implement methods for
calculating the required number of patients in a survival study.
The formula for the required number of deaths in Equation (15.1) appears
in many papers, including Bernstein and Lagakos (1978), Schoenfeld (1981),
Schoenfeld and Richter (1982) and Schoenfeld (1983), although the assump-
tions on which the result is based are different. Bernstein and Lagakos (1978)
obtain Equation (15.1) on the assumption that the survival times in each group
have exponential distributions. Lachin (1981), Rubinstein, Gail and Santner
(1981) and Lachin and Foulkes (1986) also discuss sample size requirements
in trials where the survival times are assumed to be exponentially distributed.
See also the earlier work of George and Desu (1974).
Schoenfeld (1981) obtains the same result as Bernstein and Lagakos (1978)
and others when the log-rank test is used to compare treatments, without
making the additional assumption of exponentiality. Schoenfeld (1983) shows
FURTHER READING 485
that Equation (15.1) holds when information on the values of explanatory
variables is allowed for.
The formulae for the required number of patients in Section 15.3 are based
on Schoenfeld (1983). When the assumption of exponential survival times is
made, these formulae simplify to the results of Schoenfeld and Richter (1982).
Although the resulting formulae are easier to use, it is dangerous to conduct
sample size calculations on the basis of restrictive assumptions about survival
time distributions.
A variant on the formula for the required number of deaths is given
by Freedman (1982). Freedman’s result has {(1 + ψ)/(1 − ψ)}2 in place of
4/(log ψ)2 in Equation (15.1). However, for small values of log ψ,
{(1 + ψ)/(1 − ψ)}2 ≈ 4/(log ψ)2 ,
and so the two expressions will tend to give similar results. The approximate
formula for the required number of patients, given in Section 15.3.2, is also
due to Freedman (1982).
Lakatos (1988) presented a method for estimating the required number of
patients to compare two treatments which can accommodate matters such as
staggered entry, non-compliance, loss to follow-up and non-proportional haz-
ards. Lakatos and Lan (1992) show that the Lakatos method procedure per-
forms well in a variety of circumstances. This approach is based on a Markov
model, and requires a computer program for its implementation; a SAS macro
has been given by Shih (1995).
Approximate sample size formulae for use in modelling the subdistribution
hazard function in the presence of competing risks, described in Section 12.5
of Chapter 12, have been given by Latouche, Porcher and Chevret (2004).
Appendix A
Maximum likelihood estimation
This appendix gives a summary of results on maximum likelihood estimation

that are relevant to survival analysis. The results presented apply equally
to inferences based on a partial likelihood function, and so can be used in
conjunction with the Cox regression model described in Chapter 3, and the
fully parametric models introduced in Chapters 5 and 6. A full treatment of
the theory of maximum likelihood estimation and likelihood ratio testing is
given by Cox and Hinkley (1974).
A.1 Inference about a single unknown parameter

Suppose that the likelihood of n observed survival times, t1 , t2 , . . . , tn , is a
function of a single unknown parameter β, and denoted L(β). The maximum
likelihood estimate of β is then the value β̂ for which this function is a maxi-
mum. In almost all applications, it is more convenient to work with the natural
logarithm of the likelihood function, log L(β). The value β̂, which maximises
the log-likelihood, is the same value that maximises the likelihood function
itself, and is generally found using differential calculus.
Specifically, β̂ is the value of β for which the derivative of log L(β), with
respect to β, is equal to zero. In other words, β̂ is such that
d log L(β)
β̂ = 0.
dβ
The first derivative of log L(β) with respect to β is known as the efficient score
for β, and is denoted u(β). Therefore,
d log L(β)
u(β) = ,
dβ
and so the maximum likelihood estimate of β, β̂, satisfies the equation
u(β̂) = 0.
The asymptotic variance of the maximum likelihood estimate of β can be
found from ( { 2 })−1
d log L(β)
−E , (A.1)
dβ 2
487
488 MAXIMUM LIKELIHOOD ESTIMATION
or from the equivalent formula,
( { }2 )−1
d log L(β)
E .
dβ
The variance calculated from either of these expressions can be regarded as

the approximate variance of β̂, although it is usually more straightforward to
use Expression (A.1). When the expected value of the derivative in Expres-
sion (A.1) is difficult to obtain, a further approximation to the variance of β̂
is found by evaluating the derivative at β̂. The approximate variance of β̂ is
then given by
( 2 )−1
d log L(β)
var (β̂) ≈ − . (A.2)
dβ 2
β̂
The second derivative of the log-likelihood function is sometimes known as

the Hessian, and the quantity
{ 2 }
d log L(β)
−E
dβ 2
is called the information function. Since the information function is formed

from the expected value of the second derivative of log L(β), it is sometimes
called the expected information function. In contrast, the negative second
derivative of the log-likelihood function itself is called the observed information
function. This latter quantity will be denoted i(β), so that
{ }
d2 log L(β)
i(β) = − .
dβ 2
The reciprocal of this function, evaluated at β̂, is then the approximate vari-
ance of β̂ given in Equation (A.2), that is,
1
var (β̂) ≈ .
i(β̂)
The standard error of β̂, that is, the square root of the estimated variance of
β̂, is found from
1
se (β̂) = √ .
{i(β̂)}
This standard error can be used to construct confidence intervals for β.
In order to test the null hypothesis that β = 0, three alternative test statis-
tics can be used. The likelihood ratio test statistic is the difference between the
values of −2 log L(β̂) and −2 log L(0). The Wald test is based on the statistic
β̂ 2 i(β̂),
INFERENCE ABOUT A VECTOR OF UNKNOWN PARAMETERS 489
and the score test statistic is
{u(0)}2
.
i(0)
Each of these statistics has an asymptotic chi-squared distribution on 1 d.f.,
under the null hypothesis that β = 0. Note that the Wald statistic is equivalent
to the statistic
β̂
,
se (β̂)
which has an asymptotic standard normal distribution.
A.2 Inference about a vector of unknown parameters

The results in Section A.1 can be extended to the situation where n obser-
vations are used to estimate the values of p unknown parameters, β1 , β2 , . . . ,
βp . These parameters can be assembled into a p-component vector, β, and the
corresponding likelihood function is L(β). The maximum likelihood estimates
of the p unknown parameters are the values β̂1 , β̂2 , . . . , β̂p , which maximise
L(β). They are therefore found by solving the p equations
d log L(β)
βˆ = 0,
dβj
for j = 1, 2, . . . , p, simultaneously.
The vector formed from β̂1 , β̂2 , . . . , β̂p is denoted β̂, and so the maximised
likelihood is L(β̂). The efficient score for βj , j = 1, 2, . . . , p, is
d log L(β)
u(βj ) = ,
dβj
and these quantities can be assembled to give a p-component vector of efficient

scores, denoted u(β). The vector of maximum likelihood estimates is therefore
such that
u(β̂) = 0,
where 0 is the p × 1 vector of zeroes.
Now let the matrix H(β) be the p × p matrix of second partial derivatives
of the log-likelihood function, log L(β̂). The (j, k)th element of H(β) is then
∂ 2 log L(β̂)
,
∂βj ∂βk
for j = 1, 2, . . . , p, k = 1, 2, . . . , p, and H(β) is called the Hessian matrix. The

matrix
I(β) = −H(β)
490 MAXIMUM LIKELIHOOD ESTIMATION
is called the observed information matrix. The (j, k)th element of the corre-
sponding expected information matrix is
( 2 )
∂ log L(β)
−E .
∂βj ∂βk
The variance-covariance matrix of the p maximum likelihood estimates, β̂1 , β̂2 ,

. . . , β̂p , written var (β̂), can then be approximated by the inverse of the ob-
served information matrix, evaluated at β̂, so that
var (β̂) ≈ I −1 (β̂).
The square root of the (j, j)th element of this matrix can be taken to be the
standard error of β̂j , for j = 1, 2, . . . , p.
The test statistics given in Section A.1 can be generalised to the multi-
parameter situation. Consider the test of the null hypothesis that all the
β-parameters in a fitted model are equal to zero. The likelihood ratio test
statistic is the value of
{ }
2 log L(β̂) − log L(0) ,
the Wald test is based on

β̂ ′ I(β̂) β̂,
and the score test statistic is
u′ (0)I −1 (0)u(0).
Each of these statistics has a chi-squared distribution on p d.f. under the null
hypothesis that β = 0.
In comparing alternative models, interest centres on the hypothesis that
some of the β-parameters in a model are equal to zero. To test this hypoth-
esis, the likelihood ratio test is the most suitable, and so we only consider
this procedure here. Suppose that a model that contains p + q parameters,
β1 , β2 , . . . , βp , βp+1 , . . . , βp+q , is to be compared with a model that only con-
tains the p parameters β1 , β2 , . . . , βp . This amounts to testing the null hy-
pothesis that the q parameters βp+1 , βp+2 , . . . , βp+q in the model with p + q
unknown parameters are all equal to zero. Let β̂ 1 denote the vector of es-
timates under the model with p + q parameters and β̂ 2 that for the model
with just p parameters. The likelihood ratio test of the null hypothesis that
βp+1 = βp+2 = · · · = βp+q = 0 in the model with p + q parameters is then
based on the statistic
{ }
2 log L(β̂ 1 ) − log L(β̂ 2 ) ,
which has a chi-squared distribution on q d.f., under the null hypothesis. This
test forms the basis for comparing alternative models, and was described in
greater detail in Section 3.5 of Chapter 3.
Appendix B
Additional data sets
This appendix contains a number of data sets, together with some suggestions
for analyses that could be carried out. These data sets may be downloaded
from the publishers’s web site, at the location given in the preface.
B.1 Chronic active hepatitis

In a clinical trial described by Kirk et al. (1980), 44 patients with chronic active
hepatitis were randomised to the drug prednisolone, or an untreated control
group. The survival time of the patients, in months, following admission to
the trial, was the response variable of interest. The data, which were given in
Pocock (1983), are shown in Table B.1.
Table B.1 Survival times of patients

suffering from chronic active hepatitis.
Prednisolone Control
2 131∗ 2 41
6 140∗ 3 54
12 141∗ 4 61
54 143 7 63
56∗ 145∗ 10 71
68 146 22 127∗
∗
89 148 28 140∗
∗
96 162 29 146∗
96 168 32 158∗
∗ ∗
125 173 37 167∗
∗ ∗
128 181 40 182∗
Summarise the data in terms of the estimated survivor function for each
treatment group. Compare the groups using the log-rank and Wilcoxon tests.
Fit Cox and Weibull proportional hazards model to determine the significance
of the treatment effect. Compare the results from these different analyses in
terms of the significance of the treatment effect, and, for the model-based
analyses, the estimated hazard ratio and corresponding 95% confidence limits.
491
492 ADDITIONAL DATA SETS
Obtain a log-cumulative hazard plot of the data, and comment on which
method of analysis is the most appropriate.
B.2 Recurrence of bladder cancer

In a placebo-controlled trial concerning bladder cancer, conducted by the Vet-
erans Administration Cooperative Urological Research Group, patients with
superficial bladder tumours first had their tumour removed transurethrally.
They were then randomised to a placebo, or to the chemotherapeutic agent,
thiotepa. The initial number of tumours in each patient, and the diameter of
the largest of these, was recorded at the time of randomisation. The original
data set, included in Andrews and Herzberg (1985), gave the times to up to
nine tumour recurrences, but the data presented in Table B.2 refer to the time
to the first recurrence, in months. Patients who had not experienced a recur-
rence by the end of the follow-up period contribute censored observations. The
variables in the data set are as follows:
Patient: Patient number (1–86)

Time: Survival time in months
Status: Status of patient (0 = censored, 1 = recurrence)
Treat: Treatment group (1 = placebo, 2 = thiotepa)
Init: Initial number of tumours
Size: Diameter of largest initial tumour in cm
Using a Cox proportional hazards model, determine whether the recur-

rence times depend on the explanatory variables Init and Size. After adjust-
ing for relevant variables, examine the significance of the treatment effect.
Check if there is any evidence of an interaction between treatment group and
the two explanatory variables. Using model-checking diagnostics, comment
on the suitability of the fitted model. In particular, by fitting a suitable time-
dependent variable, test the assumption of proportional hazards with respect
to treatment. Estimate the hazard ratio for the treatment effect and give a 95%
confidence interval for the ratio. Compare this estimate with that obtained
from fitting a Weibull proportional hazards model.
B.3 Survival of black ducks

In the first year of a study on the movements and overwintering survival of
black ducks, Anas rubripes, conducted by the U.S. Fish and Wildlife Service,
50 female black ducks from two locations in New Jersey were captured and
fitted with radios. The ducks were captured over a period of about four weeks
from 8 November 1983 to 14 December 1983, and included 31 hatch-year birds,
that is, birds born during the previous breeding season, and 19 birds of at least
one year of age. The body weight and wing length were measured for each
duck. The status of each duck, that is, whether it was alive, missing or dead,
SURVIVAL OF BLACK DUCKS 493
Table B.2 Time to first recurrence of a tumour in bladder cancer patients.

Patient Time Status Treat Init Size Patient Time Status Treat Init Size
1 0 0 1 1 1 44 3 1 1 3 1
2 1 0 1 1 3 45 59 0 1 1 1
3 4 0 1 2 1 46 2 1 1 3 2
4 7 0 1 1 1 47 5 1 1 1 3
5 10 0 1 5 1 48 2 1 1 2 3
6 6 1 1 4 1 49 1 0 2 1 3
7 14 0 1 1 1 50 1 0 2 1 1
8 18 0 1 1 1 51 5 1 2 8 1
9 5 1 1 1 3 52 9 0 2 1 2
10 12 1 1 1 1 53 10 0 2 1 1
11 23 0 1 3 3 54 13 0 2 1 1
12 10 1 1 1 3 55 3 1 2 2 6
13 3 1 1 1 1 56 1 1 2 5 3
14 3 1 1 3 1 57 18 0 2 5 1
15 7 1 1 2 3 58 17 1 2 1 3
16 3 1 1 1 1 59 2 1 2 5 1
17 26 0 1 1 2 60 17 1 2 1 1
18 1 1 1 8 1 61 22 0 2 1 1
19 2 1 1 1 4 62 25 0 2 1 3
20 25 1 1 1 2 63 25 0 2 1 5
21 29 0 1 1 4 64 25 0 2 1 1
22 29 0 1 1 2 65 6 1 2 1 1
23 29 0 1 4 1 66 6 1 2 1 1
24 28 1 1 1 6 67 2 1 2 2 1
25 2 1 1 1 5 68 26 1 2 8 3
26 3 1 1 2 1 69 38 0 2 1 1
27 12 1 1 1 3 70 22 1 2 1 1
28 32 0 1 1 2 71 4 1 2 6 1
29 34 0 1 2 1 72 24 1 2 3 1
30 36 0 1 2 1 73 41 0 2 3 2
31 29 1 1 3 1 74 41 0 2 1 1
32 37 0 1 1 2 75 1 1 2 1 1
33 9 1 1 4 1 76 44 0 2 1 1
34 16 1 1 5 1 77 2 1 2 6 1
35 41 0 1 1 2 78 45 0 2 1 2
36 3 1 1 1 1 79 2 1 2 1 4
37 6 1 1 2 6 80 46 0 2 1 4
38 3 1 1 2 1 81 49 0 2 3 3
39 9 1 1 1 1 82 50 0 2 1 1
40 18 1 1 1 1 83 4 1 2 4 1
41 49 0 1 1 3 84 54 0 2 3 4
42 35 1 1 3 1 85 38 1 2 2 1
43 17 1 1 1 7 86 59 0 2 1 3
was recorded daily from the date of release until 15 February 1984, when the
study was terminated. The data set, which is included in Hand et al. (1994),
contains the following variables:

Status: Status of bird (0 = alive or missing, 1 = dead)
Age: Age group (0 = hatch-year bird, 1 = bird aged > 1 year)
Weight: Weight of bird in g
Length: Length of wing in mm
The values of each of these variables for the ducks in the study are shown
in Table B.3.
Table B.3 Survival times of 50 black ducks.

Time Status Age Weight Length Time Status Age Weight Length
2 1 1 1160 277 44 1 0 1040 255
6 0 0 1140 266 49 0 0 1130 268
6 0 1 1260 280 54 0 1 1320 285
7 1 0 1160 264 56 0 0 1180 259
13 1 1 1080 267 56 0 0 1070 267
14 0 0 1120 262 57 0 1 1260 269
16 0 1 1140 277 57 0 0 1270 276
16 1 1 1200 283 58 0 0 1080 260
17 0 1 1100 264 63 0 1 1110 270
17 1 1 1420 270 63 0 0 1150 271
20 0 1 1120 272 63 0 0 1030 265
21 1 1 1110 271 63 0 0 1160 275
22 1 0 1070 268 63 0 0 1180 263
26 1 0 940 252 63 0 0 1050 271
26 1 0 1240 271 63 0 1 1280 281
27 1 0 1120 265 63 0 0 1050 275
28 0 1 1340 275 63 0 0 1160 266
29 1 0 1010 272 63 0 0 1150 263
32 1 0 1040 270 63 0 1 1270 270
32 0 1 1250 276 63 0 1 1370 275
34 1 0 1200 276 63 0 1 1220 265
34 1 0 1280 270 63 0 0 1220 268
37 1 0 1250 272 63 0 0 1140 262
40 1 0 1090 275 63 0 0 1140 270
41 1 1 1050 275 63 0 0 1120 274
Using a proportional hazards model, examine the effect of age, weight and
length on survival time. Investigate whether the coefficients of the explanatory
variables, Weight and Length, differ for the ducks in each age group. Is there
any evidence that non-linear functions of the variables Weight and Length are
needed in the model?
BONE MARROW TRANSPLANTATION 495
B.4 Bone marrow transplantation
Following the treatment of leukaemia, patients often undergo a bone marrow
transplant in order to help bring their blood cells back to a normal level. A
potentially fatal side effect of this is graft-versus-host disease, in which the
transplanted cells attack the host cells. In a study described by Bagot et
al. (1988), 37 patients who were in complete remission from acute myeloid
leukaemia (AML) or acute lymphocytic leukaemia (ALL), or in the chronic
phase of chronic myeloid leukaemia (CML), received a non-depleted allogeneic
bone marrow transplant. The age of the bone marrow donor, and whether or
not the donor had previously been pregnant, was recorded, together with
the age of the recipient, their type of leukaemia, an index of mixed epidermal
lymphocyte reactions, and whether or not the recipient developed graft-versus-
host disease. The variables in this data set are as follows:

Status: Status of patient (0 = alive, 1 = dead)
Rage: Age of patient in years
Dage: Age of donor in years
Type: Type of leukaemia (1 = AML, 2 = ALL, 3 = CML)
Preg: Donor pregnancy (0 = no, 1 = yes)
Index: Index of cell-lymphocyte reactions
Gvhd: Graft-versus-host disease (0 = no, 1 = yes)
The data, which were also given in Altman (1991), are presented in Ta-
ble B.4.
Using a Weibull accelerated failure time model, investigate the dependence
of the survival times on the prognostic variables. Estimate and plot the base-
line survivor function. Fit log-logistic and lognormal models that contain the
same explanatory variables, and estimate the baseline survivor function under
these models. Compare these estimates with the estimated baseline survivor
function obtained from fitting a Cox proportional hazards model. Hence com-
ment on which parametric model is the most appropriate, and further examine
the adequacy of this model using model-checking diagnostics.
B.5 Chronic granulomatous disease

Chronic granulomatous disease (CGD) is a group of inherited disorders of the
immune system that render patients vulnerable to recurrent infections and
chronic inflammatory conditions, such as gingivitis, enlarged lymph glands, or
non-malignant tumour-like masses, known as granulomas. These granulomas
can obstruct the passage of food through the digestive system, and may also
inhibit the flow of urine from the kidneys and bladder. The disease is generally
treated with antibiotics, but a multicentre trial carried out by Genentech, Inc.
compared the antiviral drug, interferon, with a placebo. Data were obtained
Table B.4 Survival times of leukaemia patients who received a bone marrow
transplant.
Patient Time Status Rage Dage Type Preg Index Gvhd
1 95 1 27 23 2 0 0.27 0
2 1385 0 13 18 2 0 0.31 0
3 465 1 19 19 1 0 0.39 0
4 810 1 21 22 2 0 0.48 0
5 1497 0 28 38 2 0 0.49 0
6 1181 1 22 20 2 0 0.50 0
7 993 0 19 19 2 0 0.81 0
8 138 1 20 23 2 0 0.82 0
9 266 1 33 36 1 0 0.86 0
10 579 0 18 19 1 0 0.92 0
11 600 0 17 20 2 0 1.10 0
12 1182 0 31 21 3 0 1.52 0
13 841 0 23 38 2 0 1.88 0
14 1364 0 27 15 2 0 2.01 0
15 695 0 26 16 2 0 2.40 0
16 1378 0 28 25 1 0 2.45 0
17 736 0 24 21 1 1 2.60 0
18 1504 0 18 20 2 0 2.64 0
19 849 0 24 25 1 1 3.78 0
20 1266 0 20 24 3 0 4.72 0
21 186 1 23 35 1 1 1.10 1
22 41 1 21 35 2 1 1.16 1
23 667 0 21 23 3 0 1.45 1
24 112 1 33 43 3 0 1.50 1
25 572 0 29 24 3 1 1.85 1
26 45 1 42 35 2 1 2.30 1
27 1019 0 27 31 3 0 2.34 1
28 479 1 43 29 2 1 2.44 1
29 190 1 22 20 1 0 3.70 1
30 100 1 35 39 1 1 3.73 1
31 177 1 16 14 1 0 4.13 1
32 80 1 39 35 2 1 4.52 1
33 142 1 28 25 3 1 4.52 1
34 1105 0 29 32 3 0 4.71 1
35 803 0 23 19 3 0 5.07 1
36 1126 0 33 34 3 0 9.00 1
37 114 1 19 20 1 0 10.11 1
CHRONIC GRANULOMATOUS DISEASE 497
on a number of prognostic variables, and the end-point of interest was the time
to the first serious infection following randomisation. The database, described
in Therneau and Grambsch (2000), contains the following variables:
Time: Time to first infection in days
Status: Status of patient (0 = censored, 1 = infection)
Centre: Centre (1 = Harvard Medical School
2 = Scripps Institute, California
3 = Copenhagen
4 = National Institutes of Health, Maryland
5 = Los Angeles Children’s Hospital
6 = Mott Children’s Hospital, Michigan
7 = University of Utah
8 = Children’s Hospital of Philadelphia, Pennsylvania
9 = University of Washington
10 = University of Minnesota
11 = University of Zurich
12 = Texas Children’s Hospital
13 = Amsterdam
14 = Mount Sinai Medical Center)
Treat: Treatment group (0 = placebo, 1 = interferon)
Age: Age in years
Sex: Sex (1 = male, 2 = female)
Height: Height in cm
Weight: Weight in kg
Pattern: Pattern of inheritance (1 = X-linked, 2 = autosomal recessive)
Cort: Use of corticosteroids at trial entry (1 = used, 2 = not used)
Anti: Use of antibiotics at trial entry (1 = used, 2 = not used)
Identify a suitable model for the times to first infection, and investigate the
adequacy of the model. Is there any evidence that the treatment effect is not
consistent across the different centres? Summarise the treatment difference in
terms of a relevant point estimate and corresponding confidence interval.
Now suppose that the actual infection times are interval-censored, and
that the infection status of any individual can only be recorded at 90, 180,
270 and 360 days. Construct a new observation from the survival times, that
gives the interval within which an infection, or censoring, occurs. The obser-
vations from individuals who do not experience an infection, or who develop
an infection after 360 days, are regarded as censored. For this constructed
database, analyse the interval-censored data using the methods described in
Chapter 9. Compare the estimate of the treatment effect with that found from
the original data.
Fit parametric and Cox regression models with shared frailty to these data,
where Centre is a random frailty effect. Compare and contrast the two sets of
results and comment on the extent of between centre variability.
Determine the number of patients who develop a first infection in the first
180 days at each centre. By fitting a Cox regression model to the first infection
times, obtain an estimate of the 180 day risk adjusted first infection rate at
each centre, using the methods described in Chapter 11. By treating centres
as either fixed or random, calculate corresponding interval estimates for this
rate, commenting on the extent of any differences.
Investigate the impact of dependent censoring in these data. First, fit a
Weibull model for the probability of censoring, and using this model, obtain
weights that are inversely proportional to censoring. After expressing the in-
fection time data in the counting process format, fit a weighted Cox regression
model to allow for any dependent censoring. Compare the results with those
obtained from an unweighted Cox regression analysis.
Bibliography
Aalen, O.O. (1978a) Nonparametric estimation of partial transition probabil-

ities in multiple decrement models. Annals of Statistics, 6, 534–545.
Aalen, O.O. (1978b) Nonparametric inference for a family of counting pro-
cesses. Annals of Statistics, 6, 701–726.
Aalen, O.O. (1994) Effects of frailty in survival analysis. Statistical Methods
in Medical Research, 3, 227–243.
Aalen, O.O. (1998) Frailty models. In: Statistical Analysis of Medical Data:
New Developments (eds. B.S. Everitt and G. Dunn), Arnold, London.
Aalen, O.O. and Johansen, S. (1978) An empirical transition matrix for non-
homogeneous Markov chains based on censored observations. Scandinavian
Journal of Statistics, 5, 141–150.
Aalen, O.O., Borgan, Ø. and Gjessing, H. (2008) Survival and Event History
Analysis: A Process Point of View, Springer, New York.
Aitkin, M., Anderson, D.A., Francis, B. and Hinde, J.P. (1989) Statistical
Modelling in GLIM, Clarendon Press, Oxford.
Aitkin, M., Laird, N. and Francis, B. (1983) A reanalysis of the Stanford
heart transplant data (with comments). Journal of the American Statistical
Association, 78, 264–292.
Akaike, H. (1974) A new look at the statistical model identification. IEEE
Transactions on Automatic Control, 19, 716–723.
Allison, P.D. (2010) Survival Analysis Using SAS⃝ R
: A Practical Guide, 2nd
ed., SAS Institute Inc., Cary, North Carolina.
Altman, D.G. (1991) Practical Statistics for Medical Research, Chapman &
Hall/CRC, London.
Altman, D.G. and De Stavola, B.L. (1994) Practical problems in fitting a
proportional hazards model to data with updated measurements of the
covariates. Statistics in Medicine, 13, 301–341.
Altshuler, B. (1970) Theory for the measurement of competing risks in animal
experiments. Mathematical Biosciences, 6, 1–11.
Andersen, P.K. (1982) Testing goodness of fit of Cox’s regression and life
model. Biometrics, 38, 67–77.
Andersen, P.K. (1988) Multistate models in survival analysis: A study of
nephropathy and mortality in diabetes. Statistics in Medicine, 7, 661–670.
499
500 BIBLIOGRAPHY
Andersen, P.K. (1992) Repeated assessment of risk factors in survival analysis.
Statistical Methods in Medical Research, 1, 297–315.
Andersen, P.K. and Borgan, Ø. (1985) Counting process models for life history
data: A review. Scandinavian Journal of Statistics, 12, 97–158.
Andersen, P.K. and Gill, R.D. (1982) Cox’s regression model for counting
processes: A large sample study. Annals of Statistics, 10, 1100–1120.
Andersen, P.K. and Keiding, N. (2002) Multi-state models for event history
analysis. Statistical Methods in Medical Research, 11, 91–115.
Andersen, P.K. and Perme, M.P. (2010) Pseudo-observations in survival anal-
ysis. Statistical Methods in Medical Research, 19, 71–99.
Andersen, P.K., Borgan, Ø., Gill, R.D. and Keiding, N. (1993) Statistical
Methods Based on Counting Processes, Springer, New York.
Andersen, P.K., Hansen, M.G. and Klein, J.P. (2004) Regression analysis of
restricted mean survival time based on pseudo-observations. Lifetime Data
Analysis, 10, 335–350.
Andrews, D.F. and Herzberg, A.M. (1985) Data, Springer, New York.
Arjas, E. (1988) A graphical method for assessing goodness of fit in Cox’s pro-
portional hazards model. Journal of the American Statistical Association,
83, 204–212.
Armitage, P., Berry, G. and Matthews, J.N.S. (2002) Statistical Methods in
Medical Research, 4th ed., Blackwells Science Ltd, Oxford.
Atkinson, A.C. (1985) Plots, Transformations and Regression, Clarendon
Press, Oxford.
Bagot, M., Mary, J.Y., Heslan, M., Kuentz, M., Cordonnier, C., Vernant, J.P.,
Dubertret, L. and Levy, J.P. (1988) The mixed epidermal cell lymphocyte-
reaction is the most predictive factor of acute graft-versus-host disease in
bone marrow graft recipients. British Journal of Haematology, 70, 403–409.
Barlow, W.E. and Prentice, R.L. (1988) Residuals for relative risk regression.
Biometrika, 75, 65–74.
Barnett, V. (1999) Comparative Statistical Inference, 3rd ed., Wiley, Chich-
ester.
Becker, N.G. and Melbye, M. (1991) Use of a log-linear model to compute
the empirical survival curve from interval-censored data, with application
to data on tests for HIV positivity. Australian Journal of Statistics, 33,
125–133.
Bennett, S. (1983a) Analysis of survival data by the proportional odds model.
Statistics in Medicine, 2, 273–277.
Bennett, S. (1983b) Log-logistic regression models for survival data. Applied
Statistics, 32, 165–171.
BIBLIOGRAPHY 501
Bernstein, D. and Lagakos, S.W. (1978) Sample size and power determination
for stratified clinical trials. Journal of Statistical Computation and Simu-
lation, 8, 65–73.
Beyersmann, J., Allignol, A. and Schumacher, M. (2012) Competing Risks and
Multistate Models with R, Springer, New York.
Box, G.E.P. and Tidwell, P.W. (1962) Transformation of the independent
variables. Technometrics, 4, 531–550.
Box-Steffensmeier, J.M. and Jones, B.S. (2004) Event History Modeling: A
Guide for Social Scientists, Cambridge University Press, Cambridge.
Breslow, N.E. (1972) Contribution to the discussion of a paper by D.R. Cox.
Journal of the Royal Statistical Society, B, 34, 216–217.
Breslow, N.E. (1974) Covariance analysis of censored survival data. Biomet-
rics, 30, 89–100.
Breslow, N.E. and Crowley, J. (1974) A large sample study of the life table
and product limit estimates under random censorship. Annals of Statistics,
2, 437–453.
Breslow, N.E. and Day, N.E. (1987) Statistical Methods in Cancer Research.
2: The Design and Analysis of Cohort Studies, I.A.R.C., Lyon, France.
Brookmeyer, R. and Crowley, J. (1982) A confidence interval for the median
survival time. Biometrics, 38, 29–41.
Broström, G. (2012) Event History Analysis with R, Chapman & Hall/CRC,
Boca Raton, Florida.
Brown, H. and Prescott, R.J. (2000) Applied Mixed Models in Medicine, Wiley,
Chichester.
Burdette, W.J. and Gehan, E.A. (1970) Planning and Analysis of Clinical
Studies, Charles C. Thomas, Springfield, Illinois.
Byar, D.P. (1982) Analysis of survival data: Cox and Weibull models with co-
variates. In: Statistics in Medical Research (eds. V. Mike and K.E. Stanley),
Wiley, New York.
Cain, K.C. and Lange, N.T. (1984) Approximate case influence for the pro-
portional hazards regression model with censored data. Biometrics, 40,
493–499.
Chatfield, C. (1995) Problem Solving: A Statisticians Guide, 2nd ed., Chap-
man & Hall/CRC, London.
Chatfield, C. (2004) The Analysis of Time Series, 6th ed., Chapman &
Hall/CRC, London.
Chhikara, S. and Folks, J.L. (1989) Inverse Gaussian Distribution, Marcel
Dekker, New York.
Choodari-Oskooei, B., Royston, P. and Parmar, M.K.B. (2012) A simulation
study of predictive ability measures in a survival model I: Explained vari-
ation measures. Statistics in Medicine, 31, 2627–2643.
502 BIBLIOGRAPHY
Choodari-Oskooei, B., Royston, P. and Parmar, M.K.B. (2012) A simulation
study of predictive ability measures in a survival model II: Explained ran-
domness and predictive accuracy. Statistics in Medicine, 31, 2644–2659.
Christensen, E. (1987) Multivariate survival analysis using Cox’s regression
model. Hepatology, 7, 1346–1358.
Christensen, E., Schlichting, P., Andersen, P.K., Fauerholdt, L., Schou, G.,
Pedersen, B.V., Juhl, E., Poulsen, H., Tygstrup, N. and Copenhagen Study
Group for Liver Diseases (1986) Updating prognosis and therapeutic ef-
fect evaluation in cirrhosis with Cox’s multiple regression model for time-
dependent variables. Scandinavian Journal of Gastroenterology, 21, 163–
174.
Ciampi, A. and Etezadi-Amoli, J. (1985) A general model for testing the pro-
portional hazards and the accelerated failure time hypotheses in the analy-
sis of censored survival data with covariates. Communications in Statistics,
A, 14, 651–667.
Claeskens, G., Nguti, R. and Janssen, P. (2008) One-sided tests in shared
frailty models. Test, 17, 69–82.
Cleveland, W.S. (1979) Robust locally weighted regression and smoothing
scatterplots. Journal of the Americal Statistical Association, 74, 829–836.
Cleves, M., Gould, W., Gutierrez, R.G. and Marchenko, Y.V. (2010) An In-
troduction to Survival Analysis Using Stata, 3rd ed., Stata Press.
Cohen, A. and Barnett, O. (1995) Assessing goodness of fit of parametric re-
gression models for lifetime data–graphical methods. Statistics in Medicine,
14, 1785–1795.
Collett, D. (2003) Modelling Binary Data, 2nd ed., Chapman & Hall/CRC,
Commenges, D. (1999) Multi-state models in epidemiology. Lifetime Data
Analysis, 5, 315–327.
Cook, R.D. (1986) Assessment of local influence (with discussion). Journal of
the Royal Statistical Society, B, 48, 133–169.
Cook, R.D. and Weisberg, S. (1982) Residuals and Influence in Regression,
Chapman & Hall/CRC, London.
Cook, R.J. and Lawless, J.F. (2007) The Statistical Analysis of Recurrent
Events, Springer, New York.
Copas, J.B. and Heydari, F. (1997) Estimating the risk of reoffending by
using exponential mixture models. Journal of the Royal Statistical Society,
A, 160, 237–252.
Cox, D.R. (1972) Regression models and life tables (with discussion). Journal
of the Royal Statistical Society, B, 74, 187–220.
Cox, D.R. (1975) Partial likelihood. Biometrika, 62, 269–276.
BIBLIOGRAPHY 503
Cox, D.R. (1979) A note on the graphical analysis of survival data. Biometrika,
66, 188–190.
Cox, D.R. and Hinkley, D. V. (1974) Theoretical Statistics, Chapman &
Hall/CRC, London.
Cox, D.R. and Oakes, D. (1984) Analysis of Survival Data, Chapman &
Hall/CRC, London.
Cox, D.R. and Snell, E.J. (1968) A general definition of residuals (with dis-
cussion). Journal of the Royal Statistical Society, A, 30, 248–275.
Cox, D.R. and Snell, E.J. (1981) Applied Statistics: Principles and Examples,
Cox, D.R. and Snell, E.J. (1989) Analysis of Binary Data, 2nd ed., Chapman
& Hall/CRC, London.
Crawley, M.J. (2013) The R Book, 2nd ed., John Wiley & Sons Ltd, Chichester.
Crowder, M.J. (2001) Classical Competing Risks, Chapman & Hall/CRC,
Crowder, M.J. (2012) Multivariate Survival Analysis and Competing Risks,
Chapman & Hall/CRC, Boca Raton, Florida.
Crowder, M.J., Kimber, A.C., Smith, R.L. and Sweeting, T.J. (1991) Statis-
tical Analysis of Reliability Data, Chapman & Hall/CRC, London.
Crowley, J. and Hu, M. (1977) Covariance analysis of heart transplant survival
data. Journal of the American Statistical Association, 72, 27–36.
Crowley, J. and Storer, B.E. (1983) Comment on a paper by M. Aitkin et al.
Journal of the American Statistical Association, 78, 277–281.
Dalgaard, P. (2008) Introductory Statistics with R, 2nd ed., Springer, New
York.
David, H.A. and Moeschberger, M.L. (1978) The Theory of Competing Risks,
Lubrecht & Cramer Ltd, New York.
Davis, H. and Feldstein, M. (1979) The generalized Pareto law as a model for
progressively censored survival data. Biometrika, 66, 299–306.
Day, L. (1985) Residual analysis for Cox’s proportional hazards model. In:
Proceedings of STATCOM-MEDSTAT ‘85’, MacQuarie University, Sydney.
Demidenko, E. (2013), Mixed Models: Theory and Applications with R, 2nd
ed., John Wiley & Sons, Hoboken, New Jersey.
Der, G. and Everitt, B.S. (2013) Applied Medical Statistics Using SAS, Chap-
man & Hall/CRC, Boca Raton, Florida.
Dinse, G.E. (1991) Constant risk differences in the analysis of animal tumouri-
genicity data. Biometrics, 47, 681–700.
Dobson, A.J. (2001) An Introduction to Generalized Linear Models, 2nd ed.,
Chapman & Hall/CRC, Boca Raton, Florida.
504 BIBLIOGRAPHY
Draper, N.R. and Smith, H. (1998) Applied Regression Analysis, 3rd ed., Wi-
ley, New York.
Duchateau, L. and Janssen, P. (2008) The Frailty Model, Springer, New York.
Durrleman, S. and Simon, R. (1989) Flexible regression models with cubic
splines. Statistics in Medicine, 8, 551–561.
Edmunson, J.H., Fleming, T.R., Decker, D.G., Malkasian, G.D., Jorgen-
son, E.O., Jeffries, J.A., Webb, M.J. and Kvols, L.K. (1979) Different
chemotherapeutic sensitivities and host factors affecting prognosis in ad-
vanced ovarian carcinoma versus minimal residual disease. Cancer Treat-
ment Reports, 63, 241–247.
Efron, B. (1977) The efficiency of Cox’s likelihood function for censored data.
Efron, B. (1981) Censored data and the bootstrap. Journal of the American
Statistical Association, 76, 312–319.
Efron, B., Hastie, T., Johnstone, I. and Tibshirani, R. (2004) Least angle
regression. Annals of Statistics, 32, 407–451.
Elashoff, J.D. (1983) Surviving proportional hazards. Hepatology, 3, 1031–
1035.
Emerson, J.D. (1982) Nonparametric confidence intervals for the median in
the presence of right censoring. Biometrics, 38, 17–27.
Escobar, L.A. and Meeker, W.Q. (1988) Using the SAS system to assess lo-
cal influence in regression analysis with censored data. Proceedings of the
Annual SAS User’s Group Conference, 1036–1041.
Escobar, L.A. and Meeker, W.Q. (1992) Assessing influence in regression anal-
ysis with censored data. Biometrics, 48, 507–528.
Everitt, B.S. (1987) Introduction to Optimisation Methods, Chapman &
Hall/CRC, London.
Everitt, B.S. and Rabe-Hesketh, S. (2001) Analyzing Medical Data Using S-
PLUS, Springer, New York.
Farewell V.T. (1982) The use of mixture models for the analysis of survival
data with long-term survivors. Biometrics, 38, 1041–1046.
Farrington, C.P. (1996) Interval censored survival data: A generalized linear
modelling approach. Statistics in Medicine, 15, 283–292.
Farrington, C.P. (2000) Residuals for proportional hazards models with
interval-censored data. Biometrics, 56, 473–482.
Fine, J.P. and Gray, R.J. (1999) A proportional hazards model for the sub-
distribution of a competing risk. Journal of the American Statistical Asso-
ciation, 94, 496–509.
Finkelstein, D.M. (1986) A proportional hazards model for interval-censored
failure time data. Biometrics, 42, 845–854.
BIBLIOGRAPHY 505
Finkelstein, D.M. and Wolfe, R.A. (1985) A semiparametric model for regres-
sion analysis of interval-censored failure time data. Biometrics, 41, 933–945.
Fisher, L.D. (1992) Discussion of a paper by L.J. Wei. Statistics in Medicine,
11, 1881–1885.
Fleming, T.R. and Harrington, D.P. (2005) Counting Processes and Survival
Analysis, John Wiley & Sons, Hoboken, New Jersey.
Ford, I., Norrie, J. and Ahmadi, S. (1995) Model inconsistency, illustrated by
the Cox proportional hazards model. Statistics in Medicine, 14, 735–746.
Freedman, L.S. (1982) Tables of the number of patients required in clinical
trials using the logrank test. Statistics in Medicine, 1, 121–129.
Friedman, L.M., Furberg, C.D. and DeMets, D.L. (2010) Fundamentals of
Clinical Trials, 4th ed., Springer, New York.
Fyles, A.W., McCready, D.R., Manchul, L.A., Trudeau, M.E., Merante, P.,
Pintilie, M., Weir, L. and Olivotto, I.A. (2004) Tamoxifen with or without
breast irradiation in women 50 years of age or older with early breast cancer.
New England Journal of Medicine, 351, 963–970.
Gail, M.H., Santner, T.J. and Brown, C.C. (1980) An analysis of comparative
carcinogenesis experiments based on multiple times to tumor. Biometrics,
36, 255–266.
Gaver, D.P. and Acar, M. (1979) Analytical hazard representations for use in
reliability, mortality and simulation studies. Communications in Statistics,
8, 91–111.
Geerdens, C., Claeskens, G. and Janssen, P. (2013) Goodness-of-fit tests for
the frailty distribution in proportional hazards models with shared frailty.
Biostatistics, 14, 433–446.
Gehan, E.A. (1969) Estimating survival functions from the life table. Journal
of Chronic Diseases, 21, 629–644.
George, S.L. and Desu, M.M. (1974) Planning the size and duration of a
clinical trial studying the time to some critical event. Journal of Chronic
Diseases, 27, 15–24.
Gill, R.D. (1984) Understanding Cox’s regression model: A martingale ap-
proach. Journal of the American Statistical Association, 79, 441–447.
Gill, R.D. and Schumacher, M. (1987) A simple test of the proportional haz-
ards assumption. Biometrika, 74, 289–300.
Glidden, D.V. (1999) Checking the adequacy of the gamma frailty model for
multivariate failure times. Biometrika, 86, 381–393.
Goeman, J., Meijer, R. and Chaturvedi, N. (2013) L1 (lasso and fused lasso)
and L2 (ridge) penalized estimation in GLMs and in the Cox model. R
package version 0.9-42, URL http://www.msbi.nl/goeman.
Goeman, J.J. (2010) L1 penalized estimation in the Cox proportional hazards
model. Biometrical Journal, 52, 70–84.
506 BIBLIOGRAPHY
Goldstein, H. and Spiegelhalter, D.J. (1996) League tables and their limita-
tions: statistical issues in comparisons of institutional performance, Journal
of the Royal Statistical Society, A, 159, 385–443.
Gönen, M. and Heller, G. (2005) Concordance probability and discriminatory
power in proportional hazards regression. Biometrika, 92, 965–970.
Gore, S.M., Pocock, S.J. and Kerr, G.R. (1984) Regression models and nonpro-
portional hazards in the analysis of breast cancer survival. Applied Statis-
tics, 33, 176–195.
Grambsch, P.M. and Therneau, T.M. (1994) Proportional hazards tests and
diagnostics based on weighted residuals. Biometrika, 81, 515–526.
Gray, R. (1990) Some diagnostic methods for Cox regression models through
hazard smoothing. Biometrics, 46, 93–102.
Gray, R. (2013) Subdistribution analysis of competing risks. R package version
2.2-6, URL http://www.r-project.org.
Gray, R.J. (1988) A class of k-sample tests for comparing the cumulative
incidence of a competing risk. Annals of Statistics, 16, 1141–1154.
Gray, R.J. (1992) Flexible methods for analyzing survival data using splines,
with applications to breast cancer prognosis. Journal of the American Sta-
tistical Association, 87, 942–951.
Greenwood, M. (1926) The errors of sampling of the survivorship tables. Re-
ports on Public Health and Statistical Subjects, number 33, Appendix 1,
HMSO, London.
Grønnesby, J.K. and Borgan, Ø. (1996) A method for checking regression
models in survival analysis based on the risk score. Lifetime Data Analysis,
2, 315–328.
Hall, W.J. and Wellner, J.A. (1980) Confidence bands for a survival curve
from censored data. Biometrika, 67, 133–143.
Hall, W.J., Rogers, W.H. and Pregibon, D. (1982) Outliers matter in sur-
vival analysis. Rand Corporation Technical Report P–6761, Santa Monica,
California.
Hammer, S.M., Katzenstein, D.A., Hughes, M.D., Gundacker, H., Schooley,
R.T., Haubrich, R.H., Henry, W.K., Lederman, M.M., Phair, J.P., Niu,
M., Hirsch, M.S. and Merigan, T.C. (1996) A trial comparing nucleoside
monotherapy with combination therapy in HIV-infected adults with CD4
cell counts from 200 to 500 per cubic millimeter. New England Journal of
Medicine, 335, 1081–1090.
Hand, D.J., Daly, F., Lunn, A.D., McConway, K.J. and Ostrowski, E. (1994)
A Handbook of Small Data Sets, Chapman & Hall/CRC, London.
Harrell, F.E. (2001) Regression Modelling Strategies, with Applications to Lin-
ear Models, Logistic Regression, and Survival Analysis, Springer-Verlag,
New York.
BIBLIOGRAPHY 507
Harrell, F.E., Lee, K.L. and Mark, D.B. (1996) Multivariable prognostic mod-
els: issues in developing models, evaluating assumptions and adequacy, and
measuring and reducing errors. Statistics in Medicine, 15, 361–387.
Harris, E.K. and Albert, A. (1991) Survivorship Analysis for Clinical Studies,
Marcel Dekker, New York.
Hastie, T. and Tibshirani, R. (1990) Generalized Additive Models, Chapman
& Hall/CRC, London.
Heinzl, H. (2000) Using SAS to calculate the Kent and O’Quigley measure
of dependence for Cox proportional hazards regression model. Computer
Methods and Programs in Biomedicine, 63, 71–76.
Henderson, R. and Milner, A. (1991) On residual plots for relative risk regres-
sion. Biometrika, 78, 631–636.
Hielscher, T., Zucknick, M., Werft, W. and Benner, A. (2010) On the prognos-
tic value of survival models with application to gene expression signatures.
Hinchcliffe, S.R. and Lambert, P.C. (2013) Flexible parametric modelling of
cause-specific hazards to estimate cumulative incidence functions. BioMed
Central Medical Research Methodology, 13, 1–14.
Hinkley, D.V., Reid, N. and Snell, E.J. (1991) Statistical Theory and Mod-
elling, Chapman & Hall/CRC, London.
Hjorth, U. (1980) A reliability distribution with increasing, decreasing and
bathtub-shaped failure rate. Technometrics, 22, 99–107.
Hoel, D.G. (1972) A representation of mortality data by competing risks.
Biometrics, 28, 475–488.
Hollander, M. and Proschan, F. (1979) Testing to determine the underlying
distribution using randomly censored data. Biometrics, 35, 393–401.
Hosmer, D.W. and Lemeshow, S. (2000) Applied Logistic Regression, 2nd ed.,
John Wiley & Sons, Hoboken, New Jersey.
Hosmer, D.W., Lemeshow, S. and May, S. (2008) Applied Survival Analysis:
Regression Modeling of Time to Event Data, 2nd ed., John Wiley & Sons,
Hoboken, New Jersey.
Hougaard, P. (1995) Frailty models for survival data. Lifetime Data Analysis,
1, 255–273.
Hougaard, P. (1999) Multi-state models: A review. Lifetime Data Analysis, 5,
239–264.
Hougaard, P. (2000) Analysis of Multivariate Survival Data, Springer, New
York.
Hougaard, P. and Madsen, E.B. (1985) Dynamic evaluation of short-term
prognosis of myocardial infarction. Statistics in Medicine, 4, 29–38.
Jennison, C. and Turnbull, B.W. (2000) Group Sequential Methods with Ap-
plications to Clinical Trials, Chapman & Hall/CRC, Boca Raton, Florida.
508 BIBLIOGRAPHY
Jeong, J.-H. and Fine, J.P. (2007) Parametric regression on cumulative inci-
dence function, Biostatistics, 8, 184–196.
Johnson, N.L. and Kotz, S. (1994) Distributions in Statistics: Continuous
Univariate Distributions, Volume 1, John Wiley & Sons, Hoboken, New
Jersey.
Johnson, N.L., Kemp, A.W. and Kotz, S. (2005) Distributions in Statistics:
Discrete Distributions, John Wiley & Sons, Hoboken, New Jersey.
Julious, S.A. (2010) Sample Sizes for Clinical Trials, Chapman & Hall/CRC,
Kalbfleisch, J.D. and Prentice, R.L. (1972) Contribution to the discussion of a
paper by D.R. Cox. Journal of the Royal Statistical Society, B, 34, 215–216.
Kalbfleisch, J.D. and Prentice, R.L. (1973) Marginal likelihoods based on
Cox’s regression and life model. Biometrika, 60, 267–278.
Kalbfleisch, J.D. and Prentice, R.L. (2002) The Statistical Analysis of Failure
Time Data, 2nd ed., Wiley, New York.
Kaplan, E.L. and Meier, P. (1958) Nonparametric estimation from incomplete
observations. Journal of the American Statistical Association, 53, 457–481.
Kay, R. (1977) Proportional hazard regression models and the analysis of
censored survival data. Applied Statistics, 26, 227–237.
Kay, R. (1984) Goodness of fit methods for the proportional hazards model.
Revue Epidemiologie et de Santé Publique, 32, 185–198.
Keiding, N., Klein, J.P. and Horowitz, M.M. (2001) Multistate models and
outcome prediction in bone marrow transplantation. Statistics in Medicine,
20, 1871–1885.
Kelly, P.J. and Lim, L.L.-Y. (2000) Survival analysis for recurrent event data:
An application to childhood infectious diseases. Statistics in Medicine, 19,
13–33.
Kent, J.T. and O’Quigley, J. (1988) Measures of dependence for censored
survival data. Biometrika, 75, 525–534.
Kirk, A.P., Jain, S., Pocock, S., Thomas, H.C. and Sherlock, S. (1980) Late re-
sults of the Royal Free Hospital prospective controlled trial of prednisolone
therapy in hepatitis B surface antigen negative chronic active hepatitis.
Gut, 21, 78–83.
Klein, J.P. (1991) Small-sample moments of some estimators of the variance
of the Kaplan-Meier and Nelson-Aalen estimators. Scandinavian Journal
of Statistics, 18, 333–340.
Klein, J.P. (1992) Semiparametric estimation of random effects using the Cox
model based on the EM algorithm. Biometrics, 48, 795–806.
Klein, J.P. and Moeschberger, M.L. (1988) Bounds on net survival probabili-
ties for dependent competing risks. Biometrics, 44, 529–538.
BIBLIOGRAPHY 509
Klein, J.P. and Moeschberger, M.L. (2005) Survival Analysis: Techniques for
Censored and Truncated Data, 2nd ed., Springer, New York.
Klein, J.P. and Shu, Y. (2002) Multi-state models for bone marrow transplan-
tation studies. Statistical Methods in Medical Research, 11, 117–139.
Klein, J.P., Gerster, M., Andersen, P.K., Tarima, S. and Perme, M.P. (2008)
SAS and R functions to compute pseudo-values for censored data regression.
Computer Methods and Programs in Biomedicine, 89, 289–300.
Kleinbaum, D.G. and Klein, J.P. (2012) Survival Analysis: A Self-Learning
Text, 3rd ed., Springer, New York.
Kodell, R.L. and Nelson, C.J. (1980) An illness-death model for the study
of the carcinogenic process using survival/sacrifice data. Biometrics, 36,
267–277.
Kohl, M. and Heinze, G. (2012) PSHREG: A SAS macro for proportional
and non proportional subdistribution hazards regression with competing risk
data, Technical Report 08/2012, Section for Clinical Biometrics, Medical
University of Vienna.
Krall, J.M., Uthoff, V.A. and Harley, J.B. (1975) A step-up procedure for
selecting variables associated with survival. Biometrics, 31, 49–57.
Kuk, A.Y.C. and Chen, C. (1992) A mixture model combining logistic regres-
sion with proportional hazards regression. Biometrika, 79, 531–541.
Lachin, J.M. (1981) Introduction to sample size determination and power
analysis for clinical trials. Controlled Clinical Trials, 2, 93–113.
Lachin, J.M. and Foulkes, M.A. (1986) Evaluation of sample size and power
for analyses of survival with allowance for nonuniform patient entry, losses
to follow-up, noncompliance, and stratification. Biometrics, 42, 507–519.
Lagakos, S.W. (1981) The graphical evaluation of explanatory variables in
proportional hazards models. Biometrika, 68, 93–98.
Lakatos, E. (1988) Sample sizes based on the log-rank statistic in complex
clinical trials. Biometrics, 44, 229–241.
Lakatos, E. and Lan, K.K.G. (1992) A comparison of sample size methods for
the log-rank statistic. Statistics in Medicine, 11, 179–191.
Lambert, P., Collett, D., Kimber, A. and Johnson, R. (2004) Parametric ac-
celerated failure time models with random effects and an application to
kidney transplant survival. Statistics in Medicine, 23, 3177–3192.
Latouche, A., Beyersmann, J. and Fine, J.P. (2007) Letter to the editor: Com-
ments on ‘Analysing and interpreting competing risk data’ by M. Pintilie.
Latouche, A., Porcher, R. and Chevret, S. (2004) Sample size formula for
proportional hazards modelling of competing risks. Statistics in Medicine,
23, 3263–3274.
510 BIBLIOGRAPHY
Lawless, J.F. (2002) Statistical Models and Methods for Lifetime Data, 2nd
ed., Wiley, New York.
Leathem, A.J. and Brooks, S.A. (1987) Predictive value of lectin binding on
breast-cancer recurrence and survival. The Lancet, 329, 1054–1056.
Lee, E.T. and Wang, J.W. (2013) Statistical Methods for Survival Data Anal-
ysis, 4th ed., Wiley, New York.
Lin, D.Y. (1997) Non-parametric inference for cumulative incidence functions
in competing risks studies. Statistics in Medicine, 16, 901–910.
Lin, D.Y. and Wei, L.J. (1989) The robust inference for the Cox proportional
hazards model. Journal of the American Statistical Association, 84, 1074–
1078.
Lin, D.Y. and Wei, L.J. (1991) Goodness-of-fit tests for the general Cox re-
gression model. Statistica Sinica, 1, 1–17.
Lindley, D.V. and Scott, W.F. (1984) New Cambridge Elementary Statistical
Tables, Cambridge University Press, Cambridge.
Lindsey, J.C. and Ryan, L.M. (1993) A three state multiplicative model for
rodent tumourigenicity experiments. Applied Statistics, 42, 283–300.
Lindsey, J.C. and Ryan, L.M. (1998) Methods for interval-censored data.
Lindsey, J.K. (1998) A study of interval censoring in parametric regression
models. Lifetime Data Analysis, 4, 329–354.
Machin, D., Campbell, M.J., Tan, S.B. and Tan, S.H. (2009) Sample Sizes for
Clinical Trials, 3rd ed., John Wiley & Sons Ltd, Chichester.
Machin, D., Cheung, Y.B. and Parmar, M.K.B. (2006) Survival Analysis: A
Practical Approach, Wiley, New York.
Maller, R.A. and Zhao, X. (2002) Analysis of parametric models for competing
risks. Statistica Sinica, 12, 725–750.
Mantel, N. (1966) Evaluation of survival data and two new rank order statis-
tics arising in its consideration. Cancer Chemotherapy Reports, 50, 163–
170.
Mantel, N. and Haenszel, W. (1959) Statistical aspects of the analysis of data
from retrospective studies of disease. Journal of the National Cancer Insti-
tute, 22, 719–748.
Marubini, E. and Valsecchi, M.G. (1995) Analysing Survival Data from Clin-
ical Trials and Observational Studies, Wiley, New York.
Matthews, J.N.S. (2006) Introduction to Randomized Controlled Clinical Tri-
als, 2nd ed., Chapman & Hall/CRC, Boca Raton, Florida.
May, S. and Hosmer, D.W. (1998) A simplified method of calculating an overall
goodness-of-fit test for the Cox proportional hazards model. Lifetime Data
Analysis, 4, 109–120.
BIBLIOGRAPHY 511
McCrink, L.M., Marshall, A.H. and Cairns, K.J. (2013) Advances in joint
modelling: a review of recent developments with application to the survival
of end stage renal disease patients. International Statistical Review, 81,
249–269.
McCullagh, P. and Nelder, J.A. (1989) Generalized Linear Models, 2nd ed.,
McCulloch, C.E., Searle, S.R. and Neuhaus, J.M. (2008) Generalized, Linear,
and Mixed Models, 2nd ed., John Wiley & Sons, Hoboken, New Jersey.
McGilchrist, C.A. and Aisbett, C.W. (1991) Regression with frailty in survival
analysis. Biometrics, 47, 461–466.
McKnight, B. and Crowley, J. (1984) Tests for differences in tumour incidence
based on animal carcinogenesis experiments. Journal of the American Sta-
tistical Association, 79, 639–648.
Meier, P. (1975) Estimation of a distribution function from incomplete obser-
vations. In: Perspectives in Probability and Statistics (ed. J. Gani), Aca-
demic Press, London, pp. 67–87.
Meira-Machado, L., de Uña-Alvarez, J., Cadarso-Suárez, C. and Andersen,
P.K. (2009) Multi-state models for the analysis of time-to-event data. Sta-
tistical Methods in Medical Research, 18, 195–222.
Metcalfe, C.R. and Thompson, S.G. (2007) Wei, Lin and Weissfeld’s marginal
analysis of multivariate failure time data: should it be applied to a recurrent
events outcome? Statistical Methods in Medical Research, 16, 103–122.
Miller, A.J. (2002) Subset Selection in Regression, 2nd ed., Chapman &
Hall/CRC, Boca Raton, Florida.
Moeschberger, M.L. and Klein, J.P. (1995) Statistical methods for dependent
competing risks. Lifetime Data Analysis, 1, 195–204.
Montgomery, D.C., Peck, E.A. and Vining, G. (2012) Introduction to Linear
Regression Analysis, 5th ed., Wiley, New York.
Moreau, T., O’Quigley, J. and Mesbah, M. (1985) A global goodness-of-fit
statistic for the proportional hazards model. Applied Statistics, 34, 212–
218.
Morgan, B.J.T. (1992) Analysis of Quantal Response Data, Chapman &
Hall/CRC, London.
Nagelkerke, N.J.D., Oosting, J. and Hart, A.A.M. (1984) A simple test for
goodness of fit of Cox’s proportional hazards model. Biometrics, 40, 483–
486.
Nair, V.N. (1984) Confidence bands for survival functions with censored data:
a comparative study. Technometrics, 26, 265–275.
Nardi, A. and Schemper, M. (1999) New residuals for Cox regression and their
application to outlier screening. Biometrics, 55, 523–529.
512 BIBLIOGRAPHY
Nelder, J.A. (1977) A reformulation of linear models (with discussion). Journal
of the Royal Statistical Society, A, 140, 48–77.
Nelson, K.P., Lipsitz, S.R., Fitzmaurice, G.M., Ibrahim, J., Parzen, M. and
Strawderman, R. (2006) Use of the probability integral transformation to
fit nonlinear mixed-effects models with nonnormal random effects. Journal
of Computational and Graphical Statistics, 15, 39–57.
Nelson, W. (1972) Theory and applications of hazard plotting for censored
failure data. Technometrics, 14, 945–965.
Neuberger, J., Altman, D.G., Christensen, E., Tygstrup, N. and Williams, R.
(1986) Use of a prognostic index in evaluation of liver transplantation for
primary biliary cirrhosis. Transplantation, 4, 713–716.
Nieto, F.J. and Coresh, J. (1996) Adjusting survival curves for confounders: A
review and a new method. American Journal of Epidemiology, 143, 1059–
1068.
O’Quigley, J. and Pessione, F. (1989) Score tests for homogeneity of regression
effect in the proportional hazards model. Biometrics, 45, 135–144.
Ohlssen, D., Sharples, L.D. and Spiegelhalter, D.J. (2007) A hierarchical
modelling framework for identifying unusual performance in health care
providers. Journal of the Royal Statistical Society, A, 170, 865–890.
Pan, W. (2000) A two-sample test with interval censored data via multiple
imputation. Statistics in Medicine, 19, 1–11.
Peng Y. and Dear K.B.G. (2000) A nonparametric mixture model for cure
rate estimation. Biometrics, 56, 237–243.
Pepe, M.S. and Mori, M. (1993) Kaplan-Meier, marginal or conditional prob-
ability curves in summarizing competing risks failure time data? Statistics
in Medicine, 12, 737–751.
Petersen, T. (1986) Fitting parametric survival models with time-dependent
covariates. Applied Statistics, 35, 281–288.
Peterson, A.V. (1976) Bounds for a joint distribution function with fixed sub-
distribution functions: application to competing risks. Proceedings of the
National Academy of Sciences, 73, 11–13.
Peto, R. (1972) Contribution to the discussion of a paper by D.R. Cox. Journal
Peto, R. and Peto, J. (1972) Asymptotically efficient rank invariant proce-
dures. Journal of the Royal Statistical Society, A, 135, 185–207.
Peto, R., Pike, M.C., Armitage, P., Breslow, N.E., Cox, D.R., Howard, S.V.,
Mantel, N., McPherson, K., Peto, J. and Smith, P.G. (1977) Design and
analysis of randomized clinical trials requiring prolonged observation of
each patient. II. Analysis and examples. British Journal of Cancer, 35,
1–39.
BIBLIOGRAPHY 513
Pettitt, A.N. and Bin Daud, I. (1989) Case-weighted measures of influence for
proportional hazards regression. Applied Statistics, 38, 51–67.
Pettitt, A.N. and Bin Daud, I. (1990) Investigating time dependence in Cox’s
proportional hazards model. Applied Statistics, 39, 313–329.
Pintilie, M. (2006) Competing Risks: A Practical Perspective, John Wiley &
Sons, Chichester.
Pintilie, M. (2007a) Analysing and interpreting competing risk data. Statistics
in Medicine, 26, 1360–1367.
Pintilie, M. (2007b) Authors reply to letter to the editor from A. Latouche,
J. Beyersmann and J.P. Fine. Statistics in Medicine, 26, 3679–3680.
Pocock, S.J. (1983) Clinical Trials: A Practical Approach, Wiley, Chichester.
Pollard, A.H., Yusuf, F. and Pollard, G.N. (1990) Demographic Techniques,
3rd ed., Pergamon Press, Sydney.
Prentice, R.L. and Gloeckler, L.A. (1978) Regression analysis of grouped sur-
vival data with application to breast cancer data. Biometrics, 34, 57–67.
Prentice, R.L. and Kalbfleisch, J.D. (1979) Hazard rate models with covari-
ates. Biometrics, 35, 25–39.
Prentice, R.L., Williams, B.J. and Peterson, A.V. (1981) On the regression
analysis of multivariate failure time data. Biometrika, 68, 373–379.
Putter, H. (2011) Special issue about competing risks and multi-state models.
Journal of Statistical Software, 38, 1–4.
Putter, H., Fiocco, M. and Geskus, R.B. (2007) Tutorial in biostatistics: Com-
peting risks and multi-state models. Statistics in Medicine, 26, 2389–2430.
Quantin, C., Moreau, T., Asselain, B., Maccario, J. and Lellouch, J. (1996) A
regression survival model for testing the proportional hazards hypothesis.
R Core Team. (2013) R: A language and environment for statistical
computing. R Foundation for Statistical Computing, Vienna, Austria.
URL http://www.R-project.org/.
Rabe-Hesketh, S. and Everitt, B.S. (2007) A Handbook of Statistical Analyses
Using Stata, 4th ed., Chapman & Hall/CRC, Boca Raton, Florida.
Ramlau-Hansen, H. (1983) Smoothing counting process intensities by means
of kernel functions. Annals of Statistics, 11, 453–466.
Rancel, M.M.S. and Sierra, M.A.G. (2001) Regression diagnostic using local
influence: A review. Communications in Statistics–Theory and Methods,
30, 799–813.
Reid, N. and Crépeau, H. (1985) Influence functions for proportional hazards
regression. Biometrika, 72, 1–9.
Ripatti, S. and Palmgren, J. (2000) Estimation of multivariate frailty models
using penalised partial likelihood. Biometrics, 56, 1016–1022.
514 BIBLIOGRAPHY
Rizopoulos, D. (2012) Joint Models for Longitudinal and Time-to-Event Data:
with Applications in R, Chapman & Hall/CRC, Boca Raton, Florida.
Robins, J.M. (1993) Information recovery and bias adjustment in proportional
hazards regression analysis of randomized trials using surrogate markers.
In Proceedings of the Biopharmaceutical Section, American Statistical As-
sociation, Virginia, 24–33.
Robins, J.M. and Finkelstein, D.M. (2000) Correcting for non-compliance and
dependent censoring in an AIDS clinical trial with inverse probability of
censoring weighted (IPCW) log rank tests. Biometrika, 56, 779–788.
Robins, J.M. and Rotnitzky, A. (1992) Recovery of information and ad-
justment for dependent censoring using surrogate markers. In AIDS
Epidemiology–Methodological Issues (eds. N. Jewell, K. Dietz and V.T.
Farewell), Birkhäuser, Boston, 297–331.
Royston, P. (2006) Explained variation for survival models. The Stata Journal,
6, 83–96.
Royston, P. and Altman, D.G. (1994) Regression using fractional polynomials
of continuous covariates: parsimonious parametric modelling (with discus-
sion). Applied Statistics, 43, 429–467.
Royston, P. and Lambert, P.C. (2011) Flexible Parametric Survival Analysis
Using Stata: Beyond the Cox Model, Stata Press, Texas.
Royston, P. and Parmar, M.K.B. (2002) Flexible proportional-hazards and
proportional-odds models for censored survival data, with application to
prognostic modelling and estimation of treatment effects. Statistics in
Medicine, 21, 2175–2197.
Royston, P. and Parmar, M.K.B. (2011) The use of restricted mean survival
time to estimate the treatment effect in randomized clinical trials when the
proportional hazards assumption is in doubt. Statistics in Medicine, 30,
2409–2421.
Royston, P. and Sauerbrei, W. (2004) A new measure of prognostic separation
in survival data. Statistics in Medicine, 23, 723–748.
Rubinstein, L.V., Gail, M.H. and Santner, T.J. (1981) Planning the dura-
tion of a comparative clinical trial with loss to follow-up and a period of
continued observation. Journal of Chronic Diseases, 34, 469–479.
Satten, G.A., Datta, S. and Robins, J.M. (2001) Estimating the marginal
survival function in the presence of time dependent covariates. Statistics
and Probability Letters, 54, 397–403.
Sauerbrei, W. and Royston, P. (1999) Building multivariate prognostic and
diagnostic models: transformations of the predictors by using fractional
polynomials. Journal of the Royal Statistical Society, A, 162, 71–94.
Scharfstein, D.O. and Robins, J.M. (2002). Estimation of the failure time
distribution in the presence of informative censoring. Biometrika, 89, 617–
634.
BIBLIOGRAPHY 515
Schemper, M. (1992) Cox analysis of survival data with non-proportional haz-
ard functions. The Statistician, 41, 455–465.
Schemper, M. and Stare, J. (1996) Explained variation in survival analysis.
Schemper, M., Wakounig, S. and Heinze, G. (2009) The estimation of average
hazard ratios by weighted Cox regression. Statistics in Medicine, 28, 2473–
2489.
Schlucter, M.D. (1992) Methods for the analysis of informatively censored
longitudinal data. Statistics in Medicine, 11, 1861–1870.
Schoenfeld, D.A. (1980) Chi-squared goodness of fit tests for the proportional
hazards regression model. Biometrika, 67, 145–153.
Schoenfeld, D.A. (1981) The asymptotic properties of comparative tests for
comparing survival distributions. Biometrika, 68, 316–319.
Schoenfeld, D.A. (1982) Partial residuals for the proportional hazards regres-
sion model. Biometrika, 69, 239–241.
Schoenfeld, D.A. (1983) Sample-size formula for the proportional-hazards re-
gression model. Biometrics, 39, 499–503.
Schoenfeld, D.A. and Richter, J.R. (1982) Nomograms for calculating the
number of patients needed for a clinical trial with survival as an endpoint.
Schumacher, M., Bastert, G., Bojar, H., Hübner, K., Olschewski, M., Sauer-
brei, W., Schmoor, C., Beyerle, C., Neumann, R.L.A. and Rauschecker,
H.F. (1994) Randomized 2 × 2 trial evaluating hormonal treatment and the
duration of chemotherapy in node-positive breast cancer patients. Journal
of Clinical Oncology, 12, 2086–2093.
Searle, S.R., Casella, G. and McCulloch, C.E. (2006) Variance Components,
John Wiley & Sons, Hoboken, New Jersey.
Sellke, T. and Siegmund, D. (1983) Sequential analysis of the proportional
hazards model. Biometrika, 70, 315–326.
Shih, J.H. (1995) Sample size calculation for complex clinical trials with sur-
vival endpoints. Controlled Clinical Trials, 16, 395–407.
Shih, J.H. (1998) A goodness-of-fit test for association in a bivariate survival
model. Biometrika, 85, 189–200.
Shih, J.H. and Louis, T.A. (1995) Inferences on the association parameter in
copula models for bivariate survival data. Biometrics, 51, 1384–1399.
Siannis, F. (2004) Applications of a parametric model for informative censor-
ing. Biometrics, 60, 704–714.
Siannis, F. (2011) Sensitivity analysis for multiple right censoring processes:
Investigating mortality in psoriatic arthritis. Statistics in Medicine, 30,
356–367.
516 BIBLIOGRAPHY
Siannis, F., Copas, J. and Lu, G. (2005) Sensitivity analysis for informative
censoring in parametric survival models. Biostatistics, 6, 77–91.
Simon, R. (1986) Confidence intervals for reporting results of clinical trials.
Annals of Internal Medicine, 105, 429–435.
Simon, R. and Lee, Y.J. (1982) Nonparametric confidence limits for survival
probabilities and median survival time. Cancer Treatment Reports, 66, 37–
42.
Slud, E.V. and Rubinstein, L.V. (1983) Dependent competing risks and sum-
mary survival curves. Biometrika, 70, 643–649.
Slud, E.V., Byar, D.P. and Green, S.B. (1984) A comparison of reflected ver-
sus test-based confidence intervals for the median survival time based on
censored data. Biometrics, 40, 587–600.
Spiegelhalter, D.J. (2005) Funnel plots for comparing institutional perfor-
mance. Statistics in Medicine, 24, 1185–1202.
Spiegelhalter, D.J., Aylin, P., Best, N.G., Evans, S.J.W. and Murray, G.D.
(2002) Commissioned analysis of surgical performance using routine data:
lessons from the Bristol inquiry (with discussion). Journal of the Royal
Statistical Society, A, 162, 191–231.
Spiegelhalter, D.J., Sherlaw-Johnson, C., Bardsley, M., Blunt, I., Wood, C.
and Grigg, O. (2012) Statistical methods for healthcare regulation: rating,
screening and surveillance (with discussion). Journal of the Royal Statistical
Society, A, 175, 1–47.
Stablein, D.M. and Koutrouvelis, I.A. (1985) A two-sample test sensitive to
crossing hazards in uncensored and singly censored data. Biometrics, 41,
643–652.
Stablein, D.M., Carter, W.H., Jr. and Novak, J.W. (1981) Analysis of survival
data with nonproportional hazard functions. Controlled Clinical Trials, 2,
148–159.
Stare, J., Perme, M.P. and Henderson, R. (2011) A measure of explained
variation for event history data. Biometrics, 67, 750–759.
Storer, B.E. and Crowley, J. (1985) A diagnostic for Cox regression and general
conditional likelihoods. Journal of the American Statistical Association, 80,
139–147.
Stroup, W. (2013) Generalized Linear Mixed Models: Modern Concepts, Meth-
ods and Applications, Chapman & Hall/CRC, Boca Raton, Florida.
Sun, J., Ono, Y. and Takeuchi, Y. (1996) A simple method for calculating the
exact confidence interval of the standardized mortality ratio with an SAS
function. Journal of Occupational Health, 38, 196–197.
Sun, L., Liu, J., Sun, J. and Zhang, M-J. (2006) Modeling the subdistribution
of a competing risk. Statistica Sinica, 16, 1367–1385.
BIBLIOGRAPHY 517
Sy, J.P. and Taylor, J.M.G. (2000) Estimation in a Cox proportional hazards
cure model. Biometrics, 56, 227–236.
Tableman, M. and Kim, J.S. (2004) Survival Analysis Using S: Analysis of
Time-to-Event Data, Chapman & Hall/CRC, Boca Raton, Florida.
Tai, B-C., Machin, D., White, I. and Gebski, V. (2001) Competing risks
analysis of patients with osteosarcoma: A comparison of four different ap-
proaches. Statistics in Medicine, 20, 661–684.
Taylor, J.M.G. (1995) Semi-parametric estimation in failure time mixture
models. Biometrics, 51, 899–907.
Therneau, T. (2014) A package for survival analysis in S. R package version
2.37-7, URL http://CRAN.R-project.org/package=survival.
Therneau, T.M. (1986) The COXREGR Procedure. In: SAS SUGI Supple-
mental Library User’s Guide, version 5 ed., SAS Institute Inc., Cary, North
Carolina.
Therneau, T.M. and Grambsch, P.M. (2000) Modelling Survival Data: Extend-
ing the Cox Model, Springer, New York.
Therneau, T.M., Grambsch, P.M. and Fleming, T.R. (1990) Martingale-based
residuals for survival models. Biometrika, 77, 147–160.
Therneau, T.M., Grambsch, P.M. and Pankratz, V.S. (2003) Penalized sur-
vival models and frailty. Journal of Computational and Graphical Statistics,
12, 156–175.
Thisted, R.A. (1988) Elements of Statistical Computing, Chapman &
Hall/CRC, London.
Thomas, N., Longford, N.T. and Rolph, J.E. (1994) Empirical Bayes methods
for estimating hospital-specific mortality rates. Statistics in Medicine, 13,
889–903.
Thompson, R. (1981) Survival data and GLIM. Letter to the editor of Applied
Statistics, 30, 310.
Thomsen, B.L., Keiding, N. and Altman, D.G. (1991) A note on the calcu-
lation of expected survival, illustrated by the survival of liver transplant
patients. Statistics in Medicine, 10, 733–738.
Tibshirani, R. (1982) A plain man’s guide to the proportional hazards model.
Clinical and Investigative Medicine, 5, 63–68.
Tibshirani, R. (1996) Regression shrinkage and selection via the lasso. Journal
Tibshirani, R. (1997) The lasso method for variable selection in the Cox model,
Tseng, Y.K., Hsieh, F. and Wang J.-L. (2005). Joint modeling of accelerated
failure time and longitudinal data. Biometrika, 92, 587–603.
Tsiatis, A.A. (1975) A nonidentifiability aspect of the problem of competing
risks. Proceedings of the National Academy of Sciences, 72, 20–22.
518 BIBLIOGRAPHY
Tsiatis, A.A. and Davidian, M. (2004) Joint modeling of longitudinal and
time-to-event data: An overview. Statistica Sinica, 14, 809–834.
Venables, W.N. and Ripley, B.D. (2002) Modern Applied Statistics with S, 4th
ed., Springer, New York.
Venables, W.N. and Smith, D.M. (2009) An Introduction to R, 2nd ed., Net-
work Theory Limited, Bristol.
Verweij, P.J.M., van Houwelingen, H.C. and Stijnen, T. (1998) A goodness-of-
fit test for Cox’s proportional hazards model based on martingale residuals.
Biometrics, 54, 1517–1526.
Volinsky, C.T. and Raftery, A.E. (2000) Bayesian information criterion for
censored survival models, Biometrics, 56, 256–262.
WHO Special Programme of Research, Development and Research Training
in Human Reproduction (1987) Vaginal bleeding patterns–The problem
and an example data set. Applied Stochastic Models and Data Analysis, 3,
27–35.
Wei, L.J. (1992) The accelerated failure time model: a useful alternative to
the Cox regression model in survival analysis. Statistics in Medicine, 11,
1871–1879.
Wei, L.J. and Glidden, D.V. (1997) An overview of statistical methods for
multiple failure time data in clinical trials. Statistics in Medicine, 16, 833–
839.
Wei, L.J., Lin, D.Y. and Weissfeld, L. (1989) Regression-analysis of multi-
variate incomplete failure time data by modeling marginal distributions.
Weissfeld, L.A. (1990) Influence diagnostics for the proportional hazards
model. Statistics and Probability Letters, 10, 411–417.
Weissfeld, L.A. and Schneider, H. (1990) Influence diagnostics for the Weibull
model fit to censored data. Statistics and Probability Letters, 9, 67–73.
Weissfeld, L.A. and Schneider, H. (1994) Residual analysis for parametric
models fit to censored data. Communications in Statistics–Theory and
Methods, 23, 2283–2297.
West, B.T., Welch, K.B. and Galecki, A.T. (2007) Linear Mixed Models: A
Practical Guide Using Statistical Software, Chapman & Hall/CRC, Boca
Raton, Florida.
White, I.R., Royston, P. and Wood, A.M. (2011) Multiple imputation using
chained equations: Issues and guidance for practice. Statistics in Medicine,
30, 377–399.
Whitehead, J.R. (1989) The analysis of relapse clinical trials, with application
to a comparison of two ulcer treatments. Statistics in Medicine, 8, 1439–
1454.
BIBLIOGRAPHY 519
Whitehead, J.R. (1997) The Design and Analysis of Sequential Clinical Trials,
2nd ed., Wiley, Chichester.
Wienke, A. (2011) Frailty Models in Survival Analysis, Chapman & Hall/CRC,
Wolbers, M. and Koller, M.T. (2007) Letter to the editor: Comments on
‘Analysing and interpreting competing risk data’ by M. Pintilie (original
article and author’s reply). Statistics in Medicine, 26, 3521–3523.
Wolbers, M., Koller, M.T., Witteman, J.C.M. and Steyerberg, E.W. (2009)
Prognostic models with competing risks: Methods and application to coro-
nary risk prediction. Epidemiology, 20, 555–561.
Woodward, M. (2014) Epidemiology: Study Design and Data Analysis, 3rd
ed., Chapman & Hall/CRC, Boca Raton, Florida.
Wu, M.C. and Carroll, R.J. (1988) Estimation and comparison of changes
in the presence of informative right censoring by modelling the censoring
process. Biometrics, 44, 175–188.
Yang, S. and Prentice, R.L. (1999) Semiparametric inference in the propor-
tional odds regression model. Journal of the American Statistical Associa-
tion, 94, 124–136.
Yuan, M. and Lin, Y. (2006) Model selection and estimation in regression with
grouped variables, Journal of the Royal Statistical Society, B, 68, 49–67.
Zahl, P.H. (1997) Letter to the editor: Comments on ‘Adjusting survival curves
for confounders: A review and a new method’ by F.J. Nieto and J. Coresh.
American Journal of Epidemiology, 146, 605.
Zhang, X., Loberiza, F.R., Klein, J.P. and Zhang, M.-J. (2007) A SAS macro
for estimation of direct adjusted survival curves based on a stratified Cox
regression model. Computer Methods and Programs in Biomedicine, 88,
95–101.
Statistics
Third
Edition Texts in Statistical Science
Modelling Survival Data in Medical Research describes the mod-
elling approach to the analysis of survival data using a wide range of
Modelling
Modelling Survival Data in Medical Research

examples from biomedical research.
Well known for its nontechnical style, this third edition contains new
Survival Data in
chapters on frailty models and their applications, competing risks,
non-proportional hazards, and dependent censoring. It also de-
scribes techniques for modelling the occurrence of multiple events
and event history analysis.
Earlier chapters are now expanded to include new material on a num-
ber of topics, including measures of predictive ability and flexible
parametric models. Many new data sets and examples are included
Medical Research
to illustrate how these techniques are used in modelling survival data.
Features
Third Edition
• Presents an accessible introduction to statistical methods for
handling survival data
• Includes modern statistical techniques for survival analysis and
key references
• Contains real data examples with many new data sets
• Provides additional data sets that can be used for coursework
Bibliographic notes and suggestions for further reading are provided
at the end of each chapter.
Additional data sets to obtain a fuller appreciation of the methodology,
or to be used as student exercises, are provided in the appendix.
All data sets used in this book are also available in electronic format
online.
This book is an invaluable resource for statisticians in the pharma-
ceutical industry, professionals in medical research institutes, scien-
tists and clinicians who are analysing their own data, and students
David Collett
Collett
taking undergraduate or postgraduate courses in survival analysis.
K12670
w w w. c rc p r e s s . c o m

Modelling Survival Data in Medical Research PDF

Uploaded by

Document Informationclick to expand document information

Document Informationclick to expand document information

Copyright:

Available Formats

Modelling Survival Data in Medical Research PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Modelling Survival Data in Medical Research PDF

Uploaded by

Copyright:

Available Formats

Statistics

Modelling Survival Data in Medical Research

taking undergraduate or postgraduate courses in survival analysis.

Statistical Theory: A Concise Introduction Modelling Binary Data, Second Edition

Stochastic Modeling and Mathematical Understanding Advanced Statistical Methods

Large Sample Methods in Statistics Epidemiology: Study Design and

Second edition published in 2003 by Chapman and Hall/CRC.

© 2015 by Taylor & Francis Group, LLC

No claim to original U.S. Government works

International Standard Book Number-13: 978-1-4987-3169-0 (eBook - PDF)

and the CRC Press Web site at

2 Some non-parametric procedures 17

3 The Cox regression model 57

4 Model checking in the Cox regression model 131

5 Parametric proportional hazards models 171

6 Accelerated failure time and other parametric models 221

7 Model checking in parametric models 275

8 Time-dependent variables 295

9 Interval-censored survival data 319

10 Frailty models 345

11 Non-proportional hazards and institutional comparisons 381

12 Competing risks 405

13 Multiple events and event history modelling 429

14 Dependent censoring 457

15 Sample size requirements for a survival study 471

A Maximum likelihood estimation 487

Index of Examples 521

1.1 Special features of survival data

1.1.2 Independent censoring

1.1.3 Study time and patient time

End of recruitment End of study

Figure 1.1 Study time for eight patients in a survival study.

Figure 1.2 Patient time for eight patients in a survival study.

1.2 Some examples

Example 1.1 Time to discontinuation of the use of an IUD

Table 1.1 Time in weeks to discontinuation of the use of an IUD.

Example 1.2 Prognosis for women with breast cancer

Table 1.2 Survival times of women with tumours

Example 1.3 Survival of multiple myeloma patients

Example 1.4 Comparison of two treatments for prostatic cancer

Table 1.3 Survival times of patients in a study on multiple myeloma.

1.3 Survivor, hazard and cumulative hazard functions

1.3.1 The survivor function

S(t) = P(T > t) = 1 − F (t). (1.2)

1.3.2 The hazard function

1.3.3 The cumulative hazard function

H(t) = − log S(t). (1.8)

The cumulative hazard function, H(t), is the cumulative risk of an event

1.4 Computer software for survival analysis

An introduction to the techniques used in the analysis of survival data is

Some non-parametric procedures

An initial step in the analysis of a set of survival data is to present numerical

2.1 Estimating the survivor function

Example 2.1 Pulmonary metastasis

The method of estimating the survivor function illustrated in the above

2.1.1 Life-table estimate of the survivor function

n′j = nj − cj /2. (2.2)