Sabatin Is

Download as pdf or txt
Download as pdf or txt
You are on page 1of 402

Awareness of age-related change (AARC) and its associations with cognitive,

mental, and physical health and well-being in the second half of life

Submitted by Serena Sabatini (670060780) to the University of Exeter as a thesis for


the degree of Doctor of Philosophy in Psychology (School of Psychology and
College of Medicine and Health), 24 January 2021

This thesis is available for Library use on the understanding that it is copyright material and that no
quotation from the thesis may be published without proper acknowledgement.

I certify that all material in this thesis which is not my own work has been identified and that any
material that has previously been submitted and approved for the award of a degree by this or any
other University has been acknowledged.
Acknowledgments
I would not have been able to complete this PhD without the support of many people
and I would like to thank all of them.
First, I am grateful to my supervisors. I am thankful to Linda for her constant
guidance, support, kindness, knowledge, inspiration, and enthusiasm for research. I
am grateful to Obi for patiently sharing his passion and knowledge about statistics
with me, but also for the supportive chats. I am thankful to Clive for his support which
has been valuable for data collection. I am thankful to Kaarin who provided me with
insightful feedback. You have all helped me to grow as a researcher and your
contributions have been invaluable.
I am thankful to numerous people who have extended their help to me at various
phases throughout my PhD work and growth into a researcher. I am grateful to
Rachel Collins and Sarang Kim for their valuable suggestions, to Anthony Martyr for
his expertise and guidance in meta-analyses and for always being available to
answer my questions, to Allyson Brothers, Roman Kaspar, Manfred Diehl, and Hans-
Werner Wahl for contributing to my work with their knowledge, to Helen Brooker for
her help with data collection, to Sharon Savage for allowing me to gain some
teaching experience, and to the REACH team for their kind words of encouragement,
career advice, team lunches, and virtual meetings. I am forever grateful to my family
who has helped me to grow into the person that I am today. I am grateful to Nicola
for being a patient listener and to all of my friends who always provide a welcome
escape from my studies. I thank you all for maintaining my happiness!

i
Abstract
Background: Promoting positive self-perceptions of ageing could foster health
maintenance in older age. Indeed, those individuals who perceive their ageing more
positively show better cognitive, mental, and physical health over time. Most studies
on this topic have relied on global and unidimensional assessments of self-
perceptions of ageing despite older individuals generally experiencing both positive
and negative changes. The awareness of age-related change (AARC) concept
instead captures the coexistence of the positive (gains) and negative (losses)
changes that individuals perceive across five life domains.
Aims: To validate in the UK population over 50s two questionnaires assessing
AARC and to examine the associations of AARC with cognitive, mental, and physical
health and well-being and sociodemographic variables.
Methods: A systematic review and meta-analysis synthesised and quantified
existing work on AARC and health outcomes. The psychometric properties for two
measures of AARC in the UK population over 50s were investigated. To further
explore the validity of the AARC measures and interpretation of AARC items,
participants’ comments made in relation to their responses on the AARC
questionnaires were categorised using content analysis.
Path analysis models and linear regressions estimated the association of AARC with
objective cognition. Latent profile analysis identified groups of individuals with
different combinations of levels of AARC gains and losses. Finally, differences in
cognitive, mental, and physical health and sociodemographic variables across the
identified profiles were examined.
Results: The AARC questionnaires have good psychometric properties among UK
residents over 50s. There is great variability in the combination of levels of AARC
gains and losses that individuals perceive. Higher losses are related to poorer
cognitive, mental, and physical health. Higher gains show negligible associations
with better mental and physical health and poorer cognition.
Conclusion: The AARC questionnaires may be useful to identify those UK
individuals experiencing poorer cognitive, mental, and/or physical health and who
therefore may benefit from interventions promoting adaptation to age-related
changes.

ii
Table of contents
Acknowledgments………………………………………………………………………....i
Abstract……………………………………………………………………………………. ii
Table of contents…………………………………………………………………........iii-v
List of tables……………………………………………………………………….......vi- ix
List of figures………………………………………………………………………….......x
List of appendices………………………………………………………………..……....xi
Chapter 1: Introduction……………………………………………………………..…...1
1.1 Summary…………………………………………………………………...2
1.2 Ageing population………………………………………………………..3
1.3 The ageing process……………………………………………………....4
1.4 Effects of ageing on cognitive health…………………………….......5
1.5 Effects of ageing on emotional well-being………………………......7
1.6 Ageing and depression………………………………………………….8
1.7 Ageing and anxiety…………………………………………………......10
1.8 The effects of ageing on physical health…………………………...11
1.9 Defining a possible solution……………………………………….....13
1.10 Subjective ageing……………………………………………………….14
1.11 Awareness of age-related change…………………………………....17
1.12 AARC heuristic framework and available evidence………………18
1.13 Measures of awareness of age-related change…………………....20
1.14 Psychometric properties of available measures of AARC………20
1.15 Possible uses of measures of AARC………………………………..22
1.16 AARC and developmental outcomes………………………………..23
1.16.1 AARC and cognitive health…………………………………………....23
1.16.2 AARC and emotional well-being and mental health……………...24
1.16.3 AARC and physical health………………………………………...…..24
1.17 Research questions…………………………………………………….25
1.18 Structure of the thesis………………………………………….……...26
1.19 Dissemination of findings…………………………………………......29
1.20 Conclusion……………………………………………………………….33
Chapter 2: Methodology………………………………………………………......…...34
2.1 Summary of the chapter……………………………………………….35
2.2 Platform for Research Online to investigate Genetics and

iii
Cognition in Ageing (PROTECT)…………………...……………......35
2.3 Design………………………………………………………………….....35
2.4 Participants………………………………………………………………36
2.5 Survey procedure……………………………………………………….37
2.6 Ethical approvals…………………………………………………….....40
2.7 Procedures for data access and adding questionnaires to the
PROTECT study………………………………………………………….40
2.8 Variables and measures………………………………………………..41
2.9 Data protection procedure…………………………………………….45
2.10 Statistical analyses……………….………………………………….....46
2.11 Conclusion………………………………………………………………..46
Chapter 3: Associations of awareness of age-related change with emotional
and physical well-being: A systematic review and meta-analysis………...…...48
3.1 Summary………………………………………………………………….49
3.2 Introduction………………………………………………………………51
3.3 Methods…….……………………………………………………………..53
3.4 Results……………………………………………………………………. 58
3.5 Discussion……………………………………………………………..... 71
3.6 Conclusion………………………………………………………………..77
Chapter 4: International relevance of two measures of awareness of age-
related change (AARC)…………………………………………………….…………...78
4.1 Summary………………………………………………………………….79
4.2 Introduction………………………………………………………………81
4.3 Methods…………………………………………………………………...84
4.4 Results…………………………………………………………………….93
4.5 Discussion………………………………………………………………120
5.6 Conclusion……………………………………………………………...127
Chapter 5: Exploration of awareness of age-related change among over 50s in
the UK: Qualitative and quantitative findings from the PROTECT study……129
5.1 Summary………………………………………………………………...130
5.2 Introduction………………………………………………………….....131
5.3 Methods………………………………………………………………....133
5.4 Results………………………………………………………………......136
5.5 Discussion……………………………………………………….……..149

iv
5.6 Conclusion………………………………………………………………153
Chapter 6: Cross-sectional association between objective cognitive
performance and perceived age-related gains and losses in cognition..……155
6.1 Summary…………………………………………………………….......156
6.2 Introduction……………………………………………………………..158
6.3 Methods……………………………………………………………….....161
6.4 Results………………………………………………………………......166
6.5 Discussion………………………………………………………………189
6.6 Conclusion………………………………………………………………195
Chapter 7: Differences in health among people with different profiles of
awareness of positive and negative age-related changes…………...………...196
7.1 Summary………………………………………………………………...197
7.2 Introduction……………………………………………………………..199
7.3 Methods………………………………………………………………….202
7.4 Results………………………………………………………………......208
7.5 Discussion………………………………………………………………215
7.6 Conclusion………………………………………………………………219
Chapter 8: Discussion………………………………………………………………...220
8.1 Introduction……………………………………………………………..221
8.2 Research question one……………………………………………….222
8.3 Research question two……………………………………………….224
8.4 Research question three……………………………………………..227
8.5 Research question four……..........................................................229
8.6 Research question five……………………………………………….232
8.7 Methodological considerations…………………………………......235
8.8 Future research directions…………………………………………..238
8.9 Implications of the research findings……………………………...242
8.10 Conclusion……………………………………………………………...247
References………………………………………………………………………………249
Appendices……………………………………………………………………………..322

v
List of tables
Table 3.1 Example of search terms used in Ovid Medline (R)………...…... 54
Table 3.2 Characteristics of studies included in the review………………... 61
Table 4.1 Means and standard deviations on the four objective cognitive
tasks for five levels of awareness of negative age-related
cognitive changes…………………………………………… 95
Table 4.2 Demographic characteristics of the study sample………………. 96
Table 4.3a Levels of AARC gains and losses stratified by age……………... 97
Table 4.3b Levels of AARC gains and losses stratified by sex and
educational level…………………………………………….. 98
Table 4.4 Proportions of gains and losses reported by participants on the
AARC-10 SF and the AARC-50 cognitive functioning
subscale………………………………………………………. 102
Table 4.5 Item characteristics and Cronbach’s αs for the two subscales of
the AARC-10 SF and the AARC-50 cognitive functioning
subscale………………………………………………………. 105
Table 4.6a Summary of the measurement invariance models for the
AARC-10 SF and the AARC-50 cognitive functioning
subscale-sex groups………………………………………… 108
Table 4.6b Summary of the measurement invariance models for the
AARC-10 SF and the AARC-50 cognitive functioning
subscale-educational level………………………………… 109
Table 4.6c Summary of the measurement invariance models for the
AARC-10 SF and the AARC-50 cognitive functioning
subscale - sensitivity analyses for educational level…… 110
Table 4.6d Summary of the measurement invariance models for the
AARC-10 SF and the AARC-50 cognitive functioning
subscale-age groups………………………………………... 111
Table 4.7a Correlations between AARC-10 SF and measures of subjective
ageing, mental and physical health……………………….. 113
Table 4.7b Correlations between AARC-50 cognitive functioning subscale
and measures of subjective ageing, subjective and
objective cognition………………………………………...… 115

vi
Table 4.8a Simple and multiple regressions with demographic variables as
predictors of AARC gains scores on the AARC-10 SF….. 117
Table 4.8b Simple and multiple regressions with demographic variables as
predictors of AARC losses scores on the AARC-10 SF… 117
Table 4.8c Simple and multiple regressions with demographic variables as
predictors of gains scores on the AARC-50 cognitive
functioning subscale………………………………………… 118
Table 4.8d Simple and multiple regressions with demographic variables as
predictors of losses scores on the AARC-50 cognitive
functioning subscale………………………………………… 118
Table 4.9 Simple and multiple regressions with education as predictor of
gains and losses scores on the AARC-10 SF and AARC-
50 cognitive functioning subscale – sensitivity analyses.. 119
Table 5.1 Demographic characteristics of PROTECT participants who
answered and who did not answer the AARC open-
ended question and differences in demographic
variables among the sample who did and did not answer
the AARC open-ended question….................................. 137
Table 5.2 Examples of participants’ comments for each of the identified
categories…………………………………………………….. 142
Table 5.3 Number of observations and frequencies of percentages for the
study sample who answered the open-ended question… 144
Table 5.4 For each of the ten categories, the difference in the AARC
gains mean between individuals whose comments were
assigned to a category and individuals whose comments
were not assigned to the category………………………… 146
Table 5.5 For each of the ten categories, the difference in the AARC
losses mean between individuals whose comments were
assigned to a category and individuals whose comments
were not assigned to the category………………………… 147
Table 6.1 Descriptive statistics of demographic variables and main study
variables for the study sample, study sub-samples, and
participants not included in the study……………………… 167

vii
Table 6.2 Path analysis model exploring AARC gains and losses in
cognition, AARC gains and losses across life domains,
subjective cognitive change, ATOA, and SA as
predictors of cognition in the overall study sample while
controlling for sex, education, employment status,
depressive and anxiety symptoms, and frequency of
cognitive training…………………………………………….. 170
Table 6.3 Multiple linear regressions with AARC gains and losses in
cognition and AARC gains and losses across life
domains as predictors of scores on the cognitive tasks in
the overall study sample………………………………….… 176
Table 6.4 Correlations for AARC gains in cognition, AARC losses in
cognition, and scores on the objective cognitive tasks….. 178
Table 6.5 Path analysis model exploring AARC gains and losses in
cognition as predictors of cognition in the three age sub-
groups while controlling for sex, education, employment
status, depressive and anxiety symptoms, and frequency
of cognitive training………………………........................... 182
Table 6.6 Multiple linear regressions exploring AARC gains and losses in
cognition as predictors of cognition while controlling for
sex, education, employment status, depression, anxiety
and frequency of cognitive training…………..................... 183
Table 6.7 Path analysis model exploring AARC gains and losses in
cognition as predictors of cognition while controlling for
sex, education, employment status……………………….. 184
Table 6.8 Multiple linear regressions exploring AARC gains and losses in
cognition as predictors of cognition while controlling for
sex, education, employment status……………………….. 185
Table 6.9 Associations of psychological variables and frequency of
cognitive training with AARC gains and losses in
cognition across three age sub-groups………………….. 187
Table 7.1 Descriptive statistics of demographic variables and main study
variables……………………………………………………… 203

viii
Table 7.2 Goodness of fit indices for the four latent profile models tests…. 209
Table 7.3 Estimated means for each AARC item by class membership….. 211
Table 7.4 Comparison of physical, mental and cognitive health outcomes
across the four profiles of AARC gains and losses……… 213
Table 7.5 Comparison of demographic variables and self-rated health
across the four profiles of AARC gains and losses……… 213

ix
List of figures
Figure 2.1 Diagram showing the number of participants that met
inclusion criteria in each of the four empirical
studies………………………………………………….. 38
Figure 3.1 Preferred reporting items for systematic reviews and
meta-analysis (PRISMA) 2009 flow diagram………. 59
Figure 3.2 Forest plot of the associations between AARC gains and
measures of emotional well-being…………………... 66
Figure 3.3 Funnel plots for the associations between (a) AARC gains
and emotional well-being, (b) AARC losses and
emotional well-being, (c) AARC gains and physical
well-being, (d) AARC losses and physical well-
being……………………………………………………. 67
Figure 3.4 Forest plot of the associations between AARC losses and
measures of emotional well-being…………………... 68
Figure 3.5 Forest plot of the associations between AARC gains and
measures of physical well-being…………………….. 69
Figure 3.6 Forest plot of the associations between AARC losses and
measures of physical well-being…………………….. 70
Figure 4.1a Distribution of scores on the digit span task……………….. 93
Figure 4.1b Distribution of scores on the paired associate learning task 94
Figure 4.1c Distribution of scores on the grammatical reasoning task… 94
Figure 4.1d Distribution of scores on the self-ordered search task……. 95
Figure 4.2a Distribution of scores on the AARC-10 SF gains-total
sample…………………………………………………. 99
Figure 4.2b Distribution of scores on the AARC-10 SF losses-total
sample…………………………………………………. 100
Figure 4.2c Distribution of scores on the AARC-50 cognitive
functioning subscale gains-total sample……………. 100
Figure 4.2c Distribution of scores on the AARC-50 cognitive
functioning subscale losses-total sample…………… 101
Figure 4.3a Two-factor model of the AARC-10 SF……………………… 103

x
Figure 4.3b Two-factor model of the AARC-50 cognitive functioning
subscale………………………………………………… 103
Figure 5.1 Thematic map illustrating predictors of AARC, factors
explaining lack of accuracy in AARC, and the
mental process of AARC……………………………… 139
Figure 6.1a Path analysis model exploring AARC gains and losses in
cognition as predictors of cognition in the overall
study sample while controlling for sex, education,
employment status, depressive and anxiety
symptoms, and frequency of cognitive training...... 171
Figure 6.1b Path analysis model exploring AARC gains and losses
across life domains as predictors of cognition in the
overall study sample while controlling for sex,
education, employment status, depressive and
anxiety symptoms, and frequency of cognitive
training………………………………………………… 172
Figure 6.1c Path analysis model exploring subjective cognitive decline
as predictor of cognition in the overall study sample
while controlling for sex, education, employment
status, depressive and anxiety symptoms, and
frequency of cognitive training……………………… 173
Figure 6.1d Path analysis model exploring attitudes toward own
ageing as predictor of cognition in the overall study
sample while controlling for sex, education,
employment status, depressive and anxiety
symptoms, and frequency of cognitive training…… 174
Figure 6.1e Path analysis model exploring subjective age as predictor
of cognition in the overall study sample while
controlling for sex, education, employment status,
depressive and anxiety symptoms, and frequency
of cognitive training………………………………….. 175
Figure 6.2a Path analysis model exploring AARC gains and losses as
predictors of cognition in the sub-sample of

xi
participants aged 51 to 65 years while controlling
for sex, education, employment status, depressive
and anxiety symptoms, and frequency of cognitive
training 179
Figure 6.2b Path analysis model exploring AARC gains and losses as
predictors of cognition in the sub-sample of
participants aged 66 to 75 years while controlling
for sex, education, employment status, depressive
and anxiety symptoms, and frequency of cognitive
training. 180
Figure 6.2c Path analysis model exploring AARC gains and losses as
predictors of cognition in the sub-sample of
participants aged 76 years and over while
controlling for sex, education, employment status,
depressive and anxiety symptoms, and frequency
of cognitive training. 181
Figure 7.1 Estimated average scores for each AARC item by class
membership……………………………………………. 210

xii
List of appendices
Appendix A: Questions added to the main PROTECT platform……………. 323
Appendix B: Questions used from the main PROTECT data set………….. 326
Appendix C: Application for access to PROTECT study data and /or
samples……………………………………………………. 331
Appendix D: Approval for access to PROTECT data………………………... 339
Appendix E: Ethical approval for the PROTECT study……………………… 340
Appendix F: Ethical approval for the secondary data analyses……………. 342
Appendix G: Summary of the PhD project outlined as part of the
application to request PROTECT data………………… 350
Appendix H: Memorandum of understanding set in place with the authors
of the AARC measures………………………………….. 358
Appendix I: Data agreement form…………………………………………….. 359
Appendix J: PROTECT protocol………………………………………………. 360
Appendix K: Participant information sheet for the PROTECT study………. 369
Appendix L: Application for access to PROTECT study data (Follow-up)... 370
Appendix M: Approval for access to PROTECT data (Follow-up)………….. 377
Appendix N: PROTECT data agreement form 2020………………………… 378
Appendix O: PRISMA 2009 Checklist…………………………………………. 385

xiii
Chapter 1: Introduction

Chapter 1: Introduction

1
Chapter 1: Introduction

1.1 Summary: Chapter one will introduce background information relevant for the
studies presented in this thesis. First, the challenges that an ageing population
presents for society will be introduced. Then, a brief definition of the effects that
ageing can have on cognitive, mental, and physical health will be provided. Individual
variability in ageing trajectories will be emphasised. Several subjective ageing
constructs will then be outlined and critically reviewed. Finally, this chapter will
introduce the concept of awareness of age-related change and the underlying
theoretical framework covering positive (AARC gains) and negative (AARC losses)
age-related change, available measures to assess AARC, and existing evidence on
the associations of AARC gains and losses with health outcomes. Presentation of
the research aims and questions will be followed by an outline of the overall structure
of the thesis.

2
Chapter 1: Introduction

1.2 Ageing population


The global proportion of older individuals is rapidly increasing. According to the
United Nations in 2019 people aged 65 years or over amounted to 9% of the global
population and this proportion is projected to reach 16% by 2050 (United Nations,
2019). In the United Kingdom 12 million people were aged 65 years or above in 2019
(Age UK, 2019). Ageing individuals are a resource for society as they are
experienced people who can contribute to the workforce directly, engage in volunteer
activity, and perform caregiving tasks for family members and friends. However,
such contributions are often disregarded, and the increasing proportion of older
people is frequently viewed as a problem at individual, societal, and political levels.
This negative vision is partly due to growing older being associated with increased
risk of chronic health conditions, complex co-morbidities, and disability. Indeed, even
though in the United Kingdom figures suggest that from age 65 onwards women and
men can expect to live an additional 21.1 years and 18.8 years respectively;
estimations also suggest that from age 65 onwards women and men can expect to
live 9.8 years and 8.2 years respectively in poor health (Public Health England,
2018).
As a consequence of the increasing proportion of older individuals, the total
number of people experiencing some type of cognitive decline (Kingston et al., 2018;
World Health Organization, 2020a) and/or poor mental health is also increasing
(World Health Organization, 2020b). Poor cognitive function and dementia place a
high economic and social burden on healthcare services (Prince et al., 2014; Prince
et al., 2015). In 2014 the estimated cost of dementia in the United Kingdom was
£26.3 billion (Prince et al., 2014). After dementia, depression and anxiety are the
most expensive mental health conditions in the United Kingdom, accounting for £7.5
and £8.9 billion respectively (McCrone et al., 2008). There is concern over the
projected costs for the increasing number of older people predicted to require
nursing and personal care (Brayne et al., 2001; Knapp & Wong, 2020; Morrison,
2008; Prince et al., 2014). Indeed because of functional difficulties on average men
spend 2.4 years and women 3.0 years with substantial care needs (Kingston et al.,
2017). As a result, emphasis is being placed on research into prevention of cognitive
decline, mental and physical illness and promotion of cognitive, emotional, and
physical health and well-being in later life (World Health Organization, 2008b, 2015).

3
Chapter 1: Introduction

1.3 The ageing process


The study of the ageing process is multidisciplinary; hence the concept of ageing
assumes different definitions depending from which perspective it is studied
(Balcombe & Sinclair, 2001). Chronological ageing captures the passage of time
from birth onwards. As chronological age increases, the incidence of physical and
mental problems and age-specific mortality rates also rises (Balcombe & Sinclair,
2001; Sheldon, 1947). Chronological age can be easily measured, but there is no
common agreement among researchers regarding the cut-off point at which
individuals are considered “old” (Forman et al., 1992; Kaspar et al., 2019; Wahl &
Ehni, 2020; Zizza et al., 2009).
From the biological perspective, ageing is the progressive, generalised
deterioration in function that starts in the post-maturity stage and continues until
death (Balcombe & Sinclair, 2001; Kuh et al., 2014; Rattan, 2013). As with
increasing age the biological system constantly suffers biochemical damage, the
incidence of chronic diseases increases (Balcombe & Sinclair, 2001). Survival rates
depend on how well the organism responds to such damage through mechanisms of
maintenance and repair (Rattan, 2013). Biological ageing can be measured by
assessing the presence of age-associated conditions and because trajectories of
biological ageing differ among individuals, it can be considered a better indicator of
health than chronological age.
According to Rowe and Kahn (Rowe & Kahn, 1987, 1997) individuals age
successfully when they live to a great age while having low risk of disease and
disease-related disability, good cognitive and physical abilities, are actively engaged
with life, and are able to adapt to environmental challenges. More recent work,
however, emphasises the additional importance of psychological features (e.g.
psychological well-being) as an essential aspect of successful ageing (Cosco et al.,
2015a; Cosco et al., 2013, 2014; Cosco et al., 2015b; Fries, 2002). When studying
ageing from a psychological perspective the focus of interest is on how individuals
experience their own ageing. The psychological concept of subjective ageing
(Kastenbaum et al., 1972) is an umbrella term including several constructs (e.g.
subjective age, attitudes toward own ageing, self-perceptions of ageing, and
awareness of age-related change) all referring to the ideas, beliefs, and experiences
that older adults themselves associate with their ageing. Subjective ageing is
associated with health-related behaviours and cognitive, mental, and physical health

4
Chapter 1: Introduction

(Beyer et al., 2015; Kavirajan et al., 2011; Robertson et al., 2016; Robertson et al.,
2015). Subjective ageing will be described in further detail in this chapter in Section
1.10.
Finally, sociological ageing encompasses the general beliefs about ageing
and older people that a society has, as well as the expectations regarding which
behaviours and social roles are considered appropriate for individuals in a specific
stage of life (Balcombe & Sinclair, 2001; Little, 2014). Hence, to some extent ageing
is a product of social norms and expectations that society applies to each stage of
life (Little, 2014). Indeed attitudes toward ageing (Bennett & Eckman, 1973) and age
stereotypes (Levy, 2009) can influence individuals’ behaviour, as well as the way in
which they experience their own ageing (Hess, 2006). Attitudes toward ageing
capture the images of ageing and older people that are held by societies as a whole.
Age-stereotypes cover generalised beliefs about older people as a whole (Levy,
2003, 2009). Age-stereotypes can be both positive (e.g. believing that older people
are wise) and negative (e.g. believing that older people are frail), whereas ageism
covers only those negative attitudes and beliefs about ageing that give rise to
prejudice and discrimination against older individuals (Levy, 2003; Levy et al.,
2020a).
The ageing process therefore entails change at chronological, biological,
social, and psychological levels. Changes in chronological or biological ageing do
not necessarily correspond to changes in sociological or subjective ageing: there is
generally a discrepancy between chronological ageing and subjective age (Kotter-
Grühn & Hess, 2012) and the progressive deterioration that characterises biological
ageing is not necessarily associated with negative changes in sociological ageing
such as decreased social participation (Bousquet et al., 2015). Since subjective age
differs from chronological age (Kotter-Grühn & Hess, 2012) and subjective age is
associated with cognitive, mental, and physical health (e.g. Kavirajan et al., 2011;
Robertson et al., 2015) and health-related behaviours (e.g. Beyer et al., 2015),
subjective perceptions should be considered when studying ageing.

1.4 Effects of ageing on cognitive health


Maintenance of cognitive health is important for successful ageing. Cognition is a
term used to refer to the group of mental abilities and processes related to
knowledge, attention, memory, judgement, decision making, language, and

5
Chapter 1: Introduction

comprehension (Hendrie et al., 2006). Even though with increasing age individuals
are likely to experience some sort of cognitive deterioration (Daviglus et al., 2010;
Wilson et al., 2002), mild cognitive impairment and dementia are not inevitable
(Clare et al., 2017; Gow et al., 2007). The cognitive decline experienced by ageing
individuals can be described as being on a continuum (Deary et al., 2009) with
normal cognitive ageing at one extreme and pathological cognitive ageing on the
other.
Normal cognitive ageing or “age-related cognitive decline” typically involves a
decrease in processing speed, reasoning, memory, and executive functions (Deary
et al., 2009; Liverman et al., 2015). Among individuals experiencing age-related
cognitive decline some aspects of cognitive functioning, such as verbal abilities and
general knowledge, are usually less affected and can even increase while ageing
(Christensen, 2001; Deary et al., 2009; Hedden & Gabrieli, 2004; Park & Reuter-
Lorenz, 2009).
Pathological cognitive ageing includes mild cognitive impairment and
dementia. Mild cognitive impairment refers to a cognitive decline that goes beyond
normal cognitive ageing but is less severe than dementia and does not impact on
individuals’ functional ability – which is the ability to conduct activities of daily living
and complex instrumental functions (Gauthier et al., 2006; Winblad et al., 2004). Mild
cognitive impairment is heterogeneous in its clinical presentation; the principal
cognitive impairment can be amnestic, non-amnestic, or involving multiple cognitive
domains (Matthews et al., 2007; Winblad et al., 2004). Dementia involves a
significant deterioration in one or more cognitive domains, such as memory and
language, and, contrary to mild cognitive impairment, the cognitive decline is severe
enough to interfere with individuals’ independence in everyday activities (American
Psychiatric Association, 2013). As dementia progress, impairments in functional
abilities become more severe to the point where the person is dependent on others
(Royall et al., 2007) and this often leads to institutionalisation (Gaugler et al., 2003).
In 2020, according to the World Health Organization 50 million people
worldwide are living with dementia and this number is predicted to rise to 152 million
by 2050 (Patterson, 2018; World Health Organization, 2020a). People who meet
criteria for mild cognitive impairment have an elevated risk of developing dementia,
but this progression is far from inevitable (Matthews et al., 2008; Winblad et al.,
2004). Indeed, many people who experience mild cognitive impairment do not

6
Chapter 1: Introduction

progress to dementia and some even return to previous levels of functioning


(Gauthier et al., 2006; Mitchell & Shiri-Feshki, 2009; Wilson et al., 2002; Winblad et
al., 2004). Secondary prevention of dementia is possible and this focuses on
preventing the progression from mild cognitive impairment to dementia through
strategies such as controlling for vascular risk factors, stress reduction, treatment of
depression, and cognitive training (Rakesh et al., 2017).
Among factors explaining variability in cognitive functioning in mid-to-late life
there are individual differences across the lifespan (Liverman et al., 2015) such as
cognitive ability in early life (Deary et al., 2000), genetic make-up (Gatz et al., 2006),
demographic factors, and history of chronic diseases (Brewster et al., 2014; Savva et
al., 2010; Stephan & Brayne, 2008; Yates et al., 2017). A common problem among
many of these factors is that they are not modifiable in adulthood and, as a
consequence, they cannot be targeted through midlife interventions. Health-related
behaviours (e.g. engagement in recreational activities) are also predictors of
cognitive decline (Clare et al., 2017; Williams & Kemper, 2010; Wilson et al., 2002;
Wilson et al., 2009) and, contrary to other variables, they are potentially modifiable.
Promoting health-related behaviours may therefore be a way of preventing cognitive
decline. Estimations suggest that by targeting lifestyle factors, 40% of incident
dementia and cognitive impairment could be avoided (Livingston et al., 2020).

1.5 Effects of ageing on emotional well-being


Another aspect of successful ageing is maintenance of emotional well-being.
Emotional well-being has been conceptualised in several ways and there is no
agreement on definition. Some authors define emotional well-being as a
multidimensional concept (e.g. Ryff & Keyes, 1995), whereas others understand it
through a more narrow definition (e.g. absence of depression; Antaramian et al.,
2010). Ryff’s multidimensional model of emotional well-being includes autonomy,
self-acceptance, environmental mastery, purpose in life, personal growth, and
positive interpersonal relations. All these components contribute to thriving in the
face of age-related challenges (Lara et al., 2013). Definitions of emotional well-being
also include Bradburn’s conceptualisation of happiness (Bradburn, 1969) as a
balance of positive and negative affect, and the concepts of life satisfaction and
quality of life (Andrews & McKennell, 1980; Bryant & Veroff, 1982).

7
Chapter 1: Introduction

In line with Lara and colleagues’ definition of emotional well-being (2013), in


Chapter 3 of this thesis emotional well-being will be considered as an overarching
term including concepts such as positive and negative affect, life satisfaction, quality
of life, and mental health. In this thesis emotional well-being and mental health will
be considered as two distinct concepts. This because the absence of mental illness
(e.g. depression) does not equate to good emotional well-being (Lara et al., 2013).
Indeed, many of the age-related losses (e.g. social isolation, loss of someone,
poverty) that people generally face while getting older can give rise to poor emotional
well-being even in the absence of a mental illness (Butcher & McGonigal-Kenney,
2005). Therefore, alongside emotional well-being in Chapters 4, 6, and 7 I will also
focus on mental health, and specifically on symptoms of depression and anxiety, as
these affect a significant number of people in later life, heavily impact on quality of
life, and are risk factors for dementia (Anstey, 2013; Bhalla et al., 2009; Rakesh et
al., 2017). A better understanding of emotional well-being and mental health in older
people is needed in order to find ways of promoting health maintenance and good
quality of life in older age.

1.6 Ageing and depression


Depression is a mental health disorder that is part of the category of mood disorders.
The criteria outlined in the Diagnostic and Statistical Manual of Mental Disorders
(DSM-5, American Psychiatric Association, 2013) state that in order to be diagnosed
with major depression a person must present either a depressed mood or a loss of
interest or pleasure and at least four of the following symptoms: significant weight
loss or weight gain, insomnia or hypersomnia, psychomotor agitation or retardation,
fatigue or loss of energy, feelings of worthlessness or excessive or inappropriate
guilt, diminished ability to think or concentrate or indecisiveness, and/or recurrent
thoughts of death or suicide attempt. For a diagnosis of depression these symptoms
must have been concomitantly present for two weeks, cause significant distress to
the individual and/or impairment in important areas of functioning and not be the
result of substance abuse or medical conditions.
Depression assumes specific characteristics in later life including less
frequent depressive thoughts, irritability, and hypersomnia but higher prevalence of
specific somatic symptoms (e.g. weight loss, aches, and psychomotor retardation),
poor motivation, higher anxiety, more severe cognitive dysfunction, greater insomnia,

8
Chapter 1: Introduction

and more frequent suicide attempts (Conwell et al., 2002; Minino et al., 2002;
Morrison, 2008). Late-life depressive symptoms are associated with adverse effects
on cognitive and physical health and social functioning (Bunce et al., 2014; Gallo et
al., 2005; Ho et al., 2014; Szanto et al., 2012). Some negative consequences of
depression such as cognitive impairment may persist even after the remission of
depressive symptoms (Bhalla et al., 2009).
Depression affects approximately 7% of people aged 60 and over worldwide
(World Health Organization, 2020b) and 8% of individuals aged between 65 and 85
and residing in Europe (Andreas et al., 2017). It has been estimated that at least half
of cases of major depression in older age represent late-onset depression (Bruce et
al., 2002; Fiske et al., 2009), meaning that these individuals have not had a previous
episode of depression in their life (Brodaty et al., 2001). Research comparing the
incidence and prevalence of depression in middle-age and older age is inconclusive.
Indeed, some studies document an increase in depression with ageing whereas
others report a decrease (Buchtemann et al., 2012). These contrasting results may
be explained by several factors. First, depression in older age may be harder to
detect due to older people underreporting the presence and severity of depressive
symptoms (Morrison, 2008) and/or depression being unrecognised or misdiagnosed
by specialists (Butcher & McGonigal-Kenney, 2005; Chapman, 2005; D' Mello, 2003;
Dooley & Kunik, 2017). Second, whereas major depression may decrease in early
old age, less severe forms of depression may become more frequent (Buchtemann
et al., 2012; Kemp et al., 2006; Kemp et al., 2005; Meeks et al., 2011). Third,
whereas the incidence and prevalence of major depression are lower in early old
age, they may increase in advanced old age (Alexopoulos, 2005) due to this age
group being more likely to experience age-related risk factors for depression such as
increased disability and poor socioeconomic status (Alexopoulos, 2005; Blazer,
2003). Indeed, the incidence of depression among individuals aged 79-85 years (44
per 1,000) was estimated to be more than the double than that among those aged
70-79 years (17 per 1,000) (Palsson et al., 2001; Weyerer et al., 2013).
Many variables contribute to the aetiology of late-life depression including
genetic factors (Heun et al., 2001), low socio-economic status (Fiske et al., 2009),
poor physical health (Fauth et al., 2012), increased dependence on others (Blazer,
2010), having little or no social support, being unmarried or widowed, being a
caregiver, and fearing death (Butcher & McGonigal-Kenney, 2005; Sable et al.,

9
Chapter 1: Introduction

2002). As many of these factors (e.g. poor physical health) are normative aspects of
older age, in older individuals identifying when everyday unhappiness and/or loss of
interest in life ends and clinical depression begins is challenging. To detect some
criteria that help distinguish depression from other expected negative emotional
states, Bugental (1984) conducted some interviews and subsequently proposed the
term “dispiritedness”. Dispiritedness refers to a psychological state for some aspects
similar to depression as it involves a recurrent sense of lack of meaning in life, loss
of vigour and animation, and being lost. Differently from depressed individuals,
people reporting dispiritedness do not experience insomnia, weight change,
psychomotor agitation or retardation, diminished ability to think or concentrate,
somatic complaints, recurrent thoughts of death, and impaired functioning in daily
activities. From the interviews, Bugental noted that this state of dispiritedness often
arises during transitional periods such as menopause and retirement.
Bugental concluded that in the face of age-related losses it is reasonable for
older individuals to experience some symptoms commonly attributed to depression
(such as a sense of lack of meaning in life and a general slowing down) without
having depression. However, depression is not an inevitable consequence of ageing
and when a state of being in low spirits becomes severe to the point in which it
impairs individuals daily functioning, perhaps it should attract clinical attention. Since
psychological variables, such as beliefs and expectations that individuals hold about
ageing, are associated with depressive symptoms (Kavirajan et al., 2011), they may
be useful in understanding why some older individuals experience symptoms of
depression when others do not.

1.7 Ageing and anxiety


Anxiety disorders encompass several mental health disorders that involve
cognitive, somatic, and emotional components. In this thesis I will focus on
generalised anxiety disorder (GAD). The DSM-5 criteria (American Psychiatric
Association, 2013) specify that for a diagnosis of GAD, the person must have had
excessive anxiety or worry for at least six months, difficulty controlling the worry, and
at least three of the following symptoms: restlessness, being easily fatigued, difficulty
concentrating, irritability, muscle tension, or sleep disturbance. For a diagnosis,
symptoms must lead to impairment in social activities, family relations, or ability to
work. In older age anxiety is associated with negative prognostic factors including

10
Chapter 1: Introduction

low quality of life and poor cognitive, physical and functional health (Nordhus, 2008;
Pietrzak et al., 2012; Porensky et al., 2009). Moreover, pure anxiety often
progresses to depression (Schoevers et al., 2005) and anxiety and depression
frequently co-occur in older age at both clinical and subclinical levels (Beekman et
al., 2000; Braam et al., 2014; Kasckow et al., 2013; Kvaal et al., 2008).
Anxiety affects approximately 3.8% of the global population aged 60 and over
(World Health Organization, 2020b) and 11.4% of individuals residing in Europe and
aged between 65 and 85 (Andreas et al., 2017). Similar to depressive symptoms,
many anxiety symptoms in older age can be misconceived as normal aspects of
ageing. As a consequence, anxiety disorders are often misdiagnosed and
underdiagnosed by health practitioners or underreported by older individuals
(Morrison, 2008; Sheikh et al., 2004). Between 30% and 40% of older individuals
with GAD experience an onset later in life (Le Roux et al., 2005). Contributing to the
aetiology of anxiety symptoms in old age are losses in health, mobility, financial
status, lifelong partners, and support systems, as well as fear of losses and low self-
efficacy (Banazak, 1997; Paukert et al., 2010). Among older adults with anxiety
disorders worries mainly revolve around specific topics such as health, fear of falling,
and becoming housebound (Porensky et al., 2009). As the majority of older people
experience the risk factors for anxiety (e.g. poverty and a decrease in health), less
severe worries about health or money may be reasonable in older age. A better
understanding of the relation between the experience of age-related changes and
anxiety may help to distinguish between normal preoccupations from those levels of
anxiety that require clinical attention.

1.8 The effects of ageing on physical health


Along with cognitive and mental health, the maintenance of reasonable physical
health is a precondition for older people to live independently, to engage in social
activities, and to contribute to society through volunteering or other activities. A
multidimensional definition of health has been provided by Huber et al. (2016) after
conducting interviews and focus groups with healthcare providers, patients with a
chronic condition and policymakers. They conceptualised health as including six
dimensions: bodily functioning; mental functions and perception; existential
dimension; quality of life; social and societal participation; and functional ability. In
line with the multidimensionality of Huber’s definition of health, in Chapter 3 I will

11
Chapter 1: Introduction

refer to physical well-being as an over-arching term including several indicators of


physical health such as functional ability, number of chronic health conditions, and
self-rated health. In the empirical studies presented in Chapters 4, 5, and 7 I will
focus on functional ability and self-rated health as indicators of physical health.
Generally physical health starts to decline in middle-age as at this stage of life
individuals are more likely to experience acute illnesses and to develop chronic
conditions (Aarts et al., 2012; Barnett et al., 2012; Brayne et al., 2001). In the United
Kingdom 44%, 64%, and 90% of people aged 40, 60, and 80 have one or more long-
term condition respectively (Public Health England, 2018). Generally functional
ability declines as a consequence of certain health conditions such as dementia
(Royall et al., 2007). Even though age-related changes in physical and functional
health result from biological factors, they can be influenced by potentially controllable
factors such as lack of engagement in health-related behaviours (e.g. physical
exercise) and engagement in health-damaging behaviours (e.g. smoking) (Morrison,
2008). The positive effects of health-related behaviour and the negative effects of
health-damaging behaviours are cumulative; hence the longer we engage in noxious
behaviours, the greater their negative effects on health (Morrison, 2008). Both
engagement in health-related behaviours and levels of health in older age can be
influenced by psychosocial factors such as optimism (Robinson-Whelen et al., 1997),
coping (Greer et al., 1990), social support (Jang et al., 2002) and levels of perceived
control over age-related changes (Bandura, 1977; Kornadt et al., 2019a; Kornadt &
Rothermund, 2012; Leventhal & Prohaska, 1986; Rothermund, 2005; Rothermund et
al., 2020; Rothermund & Kornadt, 2015).
Finally, to obtain a comprehensive assessment of health it is also important to
consider how individuals rate their own health status. Interestingly, even though
physical health and functional ability decline with ageing (e.g. Jylha et al., 2001), on
average self-rated health stays rather stable or at least it decreases at a slower rate.
To provide an example, a study found that 56% of women and 58% of men aged 65
or over rated their own health as good or fairly good despite having a limiting long-
term illness (Evandrou & ESRC SAGE Research Group, 2005). However, there are
also some individuals that rate their physical health as being poor even in the
absence of poor objective health and this may be due to them having poor emotional
well-being (French et al., 2012; Jylha, 2009). Nonetheless, self-rated health can be a

12
Chapter 1: Introduction

very important source of information as it predicts future levels of physical health


(Idler & Benyamini, 1997).
A review of studies on self-rated health (Idler & Benyamini, 1997) provided
several possible explanations for the discrepancy in scores obtained with measures
of self-rated and objective health. First, when rating their own health individuals may
capture their full array of illnesses including symptoms of undiagnosed diseases that
are present in preclinical or prodromal stages. Hence self-rated health may be a
more inclusive and accurate measure of health status compared to objective
measures. Second, unlike objective measures, self-rated health captures a dynamic
evaluation of health status. Indeed, a trajectory of improvement after a period of
illness may be evident to the respondent but not to an observer or through the use of
objective measures. Third, self-perceptions of health influence behaviours that
subsequently affect health status. Fourth, self-rated health may reflect the presence
or absence of environmental (e.g. living arrangements) and personal resources (e.g.
absence of depression) that can attenuate or increase decline in health.
In sum, psychological variables influence individuals’ physical health as well
as the way in which they perceive their own health. Indeed, even though older
individuals are more likely to experience chronic health conditions and a decrease in
functional ability, the beliefs they hold about ageing and their health seem to play an
important role in predicting both their engagement in healthy behaviours and future
levels of physical health. However, individuals’ evaluations of their health also
appear to shape self-perceptions of ageing (Barrett, 2003; Rubin & Berntsen, 2006).
A better understanding of the associations of objective and subjective health with
subjective ageing is needed and will be addressed in the research presented in this
thesis.

1.9 Defining a possible solution


To sum up, health-related behaviours account for some variability in cognitive and
physical health while ageing (Clare et al., 2017; Morrison, 2008; Williams & Kemper,
2010; Wilson et al., 2002; Wilson et al., 2009) and changes in physical health and
psychosocial factors account for some variability in mental health (Fauth et al., 2012;
Fiske et al., 2009). It is estimated that in the United Kingdom illness related to poor
diet costs the National Health Service (NHS) a total of £5.8 billion, with physical
inactivity costing £0.9 billion, smoking costing £3.3 billion, alcohol costing £3.3 billion

13
Chapter 1: Introduction

and overweight and obesity costing £35.1 billion (Scarborough et al., 2011). As the
same behavioural risk factors contribute to the aetiology of different chronic
diseases, elimination of shared risk factors could prevent several health conditions
(World Health Organization, 2008a). However, a critical challenge is how to
decrease engagement in damaging health-related behaviours and how to promote
engagement in health-related behaviours. Psychological variables including more
positive subjective ageing are related to better maintenance of cognitive, mental, and
physical health (Westerhof et al., 2014). In the United States it has recently been
estimated that negative subjective ageing is responsible for a yearly health care cost
of $33.7 billion (Levy et al., 2020a). Hence, subjective ageing could be targeted as a
way of promoting health maintenance in older age.
An explanation for better health that generally characterises those with
positive subjective ageing is provided by Levy’s Stereotype Embodiment Theory
(Levy, 2009), which has been corroborated by a wide amount of empirical evidence.
Stereotype Embodiment Theory posits that the behavioural, psychological, and/or
physiological features of those individuals with more positive subjective ageing
explain why they are more likely to experience better cognitive, mental, and physical
health. First, on average individuals with more positive subjective ageing engage
more frequently in health-related behaviours (Miche et al., 2015) such as healthy
eating (Klusmann et al., 2017), smoking abstinence, moderate alcohol intake (Kim,
2009; Levy & Myers, 2004; Sarkisian et al., 2005), and physical activity (Beyer et al.,
2015; Wurm et al., 2010; Wurm et al., 2013). Second, individuals with positive
subjective ageing have some psychological features such as better self-efficacy
(Dutt & Wahl, 2018; Levy et al., 2002a; Zhang & Neupert, 2020) that are related to
engagement in healthy behaviours and health maintenance. Third, individuals with
positive subjective ageing are less likely to experience stress and the health-related
changes associated with it (Chida & Steptoe, 2010; Levy et al., 2000; Panaite et al.,
2015; Stephan et al., 2015).

1.10 Subjective ageing


The psychological concept of subjective ageing (Kastenbaum et al., 1972) is an
overarching term encompassing a range of constructs attempting to capture the
beliefs and experiences that older adults associate with their own ageing. Examples
of such concepts are subjective age (Barrett, 2003; Kotter-Grühn & Hess, 2012),

14
Chapter 1: Introduction

attitudes toward own ageing (ATOA; Lawton, 1975), and self-perception of ageing
(Kleinspehn-Ammerlahn et al., 2008; Kotter-Grühn et al., 2009). Subjective age
captures how old individuals feel they are and can differ from chronological age
(Barrett, 2003; Kotter-Grühn & Hess, 2012) as a consequence of adjusting one’s age
in relation to distal (e.g. subjective age norms) and proximal reference points (e.g.
physical markers of one’s age) of age (Barrett, 2003; Kastenbaum et al., 1972;
Kotter-Grühn & Hess, 2012; Montepare, 2009). Conceptually, subjective age treats
positive (younger subjective age) and negative (older subjective age) evaluations of
ageing as two ends of the same spectrum. The majority of middle-aged and older
individuals feel younger than their chronological age (Bordone et al., 2019; Choi &
DiNitto, 2014; Opdebeeck et al., 2019; Rubin & Berntsen, 2006; Westerhof et al.,
2014) and this discrepancy between one’s chronological age and subjective age
increases with ageing (Kornadt et al., 2018). However, there is also a small
proportion of people feeling either their chronological age or older than their
chronological age (Choi & DiNitto, 2014; Kotter-Grühn et al., 2016; Rubin &
Berntsen, 2006; Westerhof et al., 2003). These individuals generally experience poor
health (Kwak et al., 2018; Montepare & Lachman, 1989; Stephan et al., 2014;
Stephan et al., 2016a; Stephan et al., 2016b; Uotinen et al., 2005). Even though
several dimensions of subjective age such as look age, do age, and interest age
have been proposed (Kastenbaum et al., 1972), at statistical level these dimensions
load onto a single factor supporting a unidimensional construct (Barak, 1987; Hubley
& Russell, 2009; Teuscher, 2009).
ATOA capture individuals’ perceptions and evaluations of the changes taking
place in their lives as they age (Bennett & Eckman, 1973; Kotter-Grühn & Hess,
2012; Lawton, 1975). More positive ATOA are associated with indicators of
successful ageing including lower risk of pathological cognitive decline (Levy et al.,
2018; Robertson & Kenny, 2016b; Robertson et al., 2016; Seidler & Wolff, 2017;
Siebert et al., 2016), better physical health and functional ability (Kavirajan et al.,
2011; Kotter-Grühn et al., 2009; Levy et al., 2002b; Maier & Smith, 1999; Moser et
al., 2011; Sargent-Cox et al., 2014), and higher satisfaction with life (Wurm et al.,
2008). ATOA start to develop in childhood, reflecting a person’s past experience
within their social, cultural, and historical context. ATOA involve little awareness and
impact on people’s behaviour at an unconscious level (Diehl et al., 2014; Hess,
2006). Most of the available evidence on the topic assessed ATOA with the ATOA

15
Chapter 1: Introduction

scale (Lawton, 1975) which captures affective and cognitive components of self-
related ageing attitudes. Similar to subjective age, the ATOA scale measures global
ATOA and treats positive and negative ATOA as two ends of the same spectrum.
Items included in the ATOA scale are reported in Appendix A.
Finally, “self-perceptions of ageing”, “subjective experiences of ageing” or
“self-views of ageing” (Connidis, 1989; Giles et al., 2010; Kleinspehn-Ammerlahn et
al., 2008; Kotter-Grühn et al., 2009; Sherman, 1994) are all interchangeable terms
capturing how individuals subjectively experience the effects of advancing
chronological age on their cognitive, physical, emotional, and social functioning.
Research on self-perceptions of ageing is mainly based on face-to-face interviews
and diaries asking participants to report events or changes that made them aware of
their increasing age (Furstenberg, 2002; Giles et al., 2010; Heckhausen et al., 1989;
Karp, 1988; Lin et al., 2004; Nilsson et al., 2000; Sherman, 1994; Steverink et al.,
2001). Overall, research on this topic shows that self-perceptions of ageing capture
both positive (e.g. increased self-efficacy) and negative (e.g. physical decline) events
across several domains of one’s life (Furstenberg, 2002; Giles et al., 2010;
Heckhausen et al., 1989; Karp, 1988; Keller et al., 1989; Lin et al., 2004; Nilsson et
al., 2000; Sherman, 1994; Steverink et al., 2001). Self-perceptions of ageing have
been hypothesised to be one of the antecedents of another subjective ageing
concept which captures a more conscious and explicit understanding of one’s own
ageing process: awareness of age-related change (AARC; Diehl & Wahl, 2010). The
AARC concept will be introduced in Section 1.11 of this chapter.
To summarise, a great deal of subjectivity characterises the ageing process
and variability in subjective ageing experiences is related to levels of cognitive,
mental, and physical health. Targeting subjective ageing may be useful to promote
maintenance of cognitive, physical, and emotional health in older age. However,
existing evidence on the topic relies on subjective ageing concepts and measures
that have several limitations. These include lack of recognition of the variations in
individuals’ subjective ageing experience across different domains (e.g. physical,
cognitive, and social domains), and failure to acknowledge that age-related gains
and losses may occur simultaneously (Diehl & Wahl, 2010). The AARC concept
attempts to overcome these limitations.

16
Chapter 1: Introduction

1.11 Awareness of age-related change


AARC is a multidimensional construct capturing subjective evaluations of ageing
across five behavioural and life domains (health and physical functioning, cognition,
interpersonal relationships, socio-cognitive and socio-emotional functioning, and
lifestyle and life engagement). AARC refers to “a person’s state of awareness that
his or her behaviour, level of performance, or way of experiencing life has changed
as a consequence of having grown older” (Diehl & Wahl, 2010; p. 342). AARC
acknowledges that while ageing individuals can experience both positive (gains) and
negative (losses) changes which can coexist, even within the same domain.
The concept of AARC is grounded in specific developmental theories such as
the theoretical work of Baltes (1987) and Brandtstädter and Rothermund (2002).
Life-span developmental theories have been outlined over the last 30 years in North
America and Europe with the aim to explaining life-long development by focusing on
constancy and change in human behaviour, interindividual differences and
similarities, and plasticity of development (Baltes et al., 1977; Lerner, 1984; Thomae,
1979). Baltes’ work emphasises that during a given developmental period the
functioning of some behavioural systems may be enhanced whereas the functioning
of others may decrease, which highlighted the importance of considering both
developmental gains and losses and their interaction throughout the entire lifespan
(Baltes, 1987; Baltes et al., 1977).
Brandtstädter and Rothermund (2002) in their Two-Process Framework model
theorise that with ageing individuals face losses and life constraints that leave them
unable to achieve goals as they used to. However, individuals can use two ways of
coping to reach their desired goals, termed assimilative and accommodative modes.
Individuals use the assimilative mode when they increase their efforts or engage in
new activities in order to reach the same personal goals that in the past would have
been achieved with less effort. Individuals adopt the accommodative mode when
they downsize their goals according to current available resources. Self-reflection
about personal changes is a precondition to the use of assimilative or
accommodative coping strategies. As AARC implies that individuals monitor the way
they grow older and consciously reflect on their own experiences, AARC may be a
necessary prerequisite for taking active control over one’s ageing process. However,
as AARC captures only those changes that individuals attribute to ageing, perceiving
changes as unavoidable aspects of ageing could also lead to psychological distress

17
Chapter 1: Introduction

and inactivity (Dunbar-Jacob et al., 1995; Kornadt et al., 2019a; Kornadt &
Rothermund, 2012; Lachman et al., 2006; Leventhal & Prohaska, 1986;
Rothermund, 2005; Rothermund et al., 2020; Rothermund & Kornadt, 2015;
Sarkisian et al., 2001; Stewart et al., 2011; Wolff et al., 2017).
Overall, AARC is conceptually different from subjective age and ATOA in
several ways. First, measures of subjective age and ATOA are unidimensional
(individuals can report either positive or negative subjective ageing) whereas AARC
assesses the coexistence of positive and negative subjective ageing perceptions.
The assessment of AARC gains is very important as it makes it possible to explore
for the first time whether perceptions of positive age-related change can be
promoted to enhance levels of health and well-being in older age, as well as to
attenuate the negative effects that age-related losses can have on the psychological
and physical well-being of older individuals. Second, most measures of subjective
age capture global evaluations of ageing, whereas AARC captures self-evaluations
of ageing across five domains. Third, ATOA are mostly unconscious whereas AARC
is conceptualised as a component of self-knowledge. Nonetheless, AARC is linked to
other subjective ageing concepts (Diehl & Wahl, 2010). Indeed, levels of AARC
gains and losses are rooted in concrete personal events and the subjective
evaluation of such experiences may be filtered by the age stereotypes and attitudes
towards ageing that individuals have internalised throughout their lives (Brothers et
al., 2020; Miche et al., 2014). In sum, the unique conceptual features of AARC may
make it possible to better capture the complexity of subjective ageing and to
advance the understanding of why some individuals age less well than others.

1.12 AARC heuristic framework and available evidence


Based on subjective ageing literature, Diehl & Wahl (2010) proposed an heuristic
framework placing AARC into a broader conceptual framework including potential
antecedents (distal and proximal), variables involved in the mental process, and
developmental outcomes of AARC. Distal antecedents of AARC comprise
sociodemographic, health-related, and psychological factors, whereas proximal
antecedents include personal goals, life limitations, age stereotypes, and life events.
Some distal antecedents of AARC have been empirically explored;
sociodemographic variables including age, sex, education level, and socioeconomic
status all explain significant variability in levels of AARC gains and AARC losses

18
Chapter 1: Introduction

(Brothers et al., 2016; English et al., 2019; Miche et al., 2014). People who are older,
female, and better educated perceive both higher gains and losses compared to
those who do not have these characteristics (Brothers et al., 2016; English et al.,
2019; Miche et al., 2014). Individuals who identify themselves as having high socio-
economic status report more AARC gains and fewer AARC losses (English et al.,
2019). However, evidence is limited to US and German participants; in Chapters 5
and 7 I will therefore explore whether sociodemographic variables are related to
levels of AARC gains and losses among UK individuals. Regarding psychological
antecedents of AARC, individuals with lower levels of neuroticism, higher levels of
openness, and/or higher conscientiousness report more AARC gains, whereas
individuals with lower levels of neuroticism report fewer AARC losses (Brothers et
al., 2019; Dutt & Wahl, 2018; Kaspar et al., 2019; Rupprecht et al., 2019). Finally,
those individuals with more negative age stereotypes experience more AARC losses
(Brothers et al., 2020; Brothers et al., 2017).
Hypothesised factors involved in the mental process of AARC include
personal meaning-making and self-regulation of behaviour. So far evidence suggests
that the less frequent use of self-regulatory strategies is related to fewer AARC gains
but more AARC losses (Dutt et al., 2016b). This suggests that experiencing high
levels of losses and interpreting them as a consequence of ageing may lead to
inactivity. In this thesis, I aim to better understand the mental process of AARC as
this is an area relatively unexplored. Finally, hypothesised outcomes of AARC are
psychological health, physical health, and life engagement. Empirical work
corroborates the hypothesised outcomes by showing that those individuals with
higher AARC losses and lower AARC gains are more likely to experience depressive
symptoms, negative affect, low life satisfaction, poorer physical well-being and
functional ability (Brothers et al., 2017; Dutt et al., 2016a; Dutt & Wahl, 2018; Miche
et al., 2014; Neupert & Bellingtier, 2017). The relation of AARC with developmental
outcomes will be further discussed in section 1.16 of this chapter.
To summarise, there is increasing interest in the concept of AARC and
evidence indicates that it may be associated with maintenance of well-being and
health in older age. This is important as AARC appears to be amenable to change
(Brothers & Diehl, 2017). Due to existing evidence being limited to US and German
participants, this thesis will explore whether the coexistence of AARC gains and
losses is related to cognitive, mental, and physical health among UK individuals.

19
Chapter 1: Introduction

1.13 Measures of awareness of age-related change


A questionnaire assessing AARC is available in three published versions of differing
lengths: a 50-item version (Brothers et al., 2019), and a further ultra-short version
(AARC-10 SF; Kaspar et al., 2019) are available in English whereas a 32-item
version is available in German (Wahl et al., 2013). The 50-item questionnaire
resulted from the refinement of two original forms of the questionnaire consisting of
189 and 100 items respectively (Brothers, 2016). The original 189 items were put
together based on data from three preliminary studies. In the first study Miche et al.
(2014) identified 49 potential items through open-ended diaries in which participants
were asked to record experiences related to AARC. In the second study, Wahl et al.
(2013) examined 60 potential new items as part of the development of the AARC-32
item German-language questionnaire. Fifty items were retained and included in the
189-item version. Finally the last 90 items were identified by Brothers and colleagues
through six focus groups (Brothers et al., 2019) in which groups of four to six middle-
aged participants were invited to think about recent positive and negative
experiences that made them realise they were growing old. A modified version of the
AARC questionnaire (Neupert & Bellingtier, 2017) adapted to capture daily
perceptions of AARC also exists. This includes 20 items that have been selected
from the 50-item version of the AARC questionnaire. Each item stem, instead of
asking participants to reflect on their increasing age (“With my increasing age…”),
invites participants to reflect on their awareness of ageing in that specific day (“With
my awareness of ageing today…”).

1.14 Psychometric properties of available measures of AARC


Psychometric properties of the AARC questionnaires have only been explored in the
United States or Germany. The 32-item German-language version of the AARC
questionnaire (Wahl et al., 2013) is a reliable measure showing Cronbach’s a
coefficients of .82 for AARC gains and of .87 for AARC losses. Regarding the
English-language versions of the AARC questionnaire, so far, no information has
been published on the psychometric properties of the 20-item version, whereas there
is evidence that the AARC-50 and the AARC-10 SF are both valid and reliable
measures in the United States and Germany (Brothers et al., 2019; Kaspar et al.,
2019). For the AARC-50 (Brothers et al., 2019) Cronbach’s a coefficients ranged

20
Chapter 1: Introduction

from .73 to .89 for the two higher order subscales (AARC gains and AARC losses)
as well as for the ten subscales assessing AARC behavioural domains (five domains
for AARC gains and five for AARC losses). Item-level reliabilities obtained
Cronbach’s a coefficients ranging from .26 to .85.
Convergent and divergent validity of the AARC-50 was calculated only for the
two subscales assessing AARC gains and AARC losses. Convergent validity of the
AARC-50 was examined with other subjective ageing constructs, namely subjective
age (assessed with a single-item question; Kastenbaum et al., 1972), ATOA
(assessed with Lawton’s ATOA scale; Lawton, 1975) and self-perceptions of ageing
(assessed with the AgeCog scales; Steverink et al., 2001). Higher levels of AARC
losses were associated with a more negative ATOA, an older subjective age, and
more negative self-perceptions of ageing. AARC gains were only significantly
associated with more positive self-perceptions of ageing. Divergent validity,
assessed by comparing the AARC-50 with age stereotypes (assessed with the
Views on Ageing Scale; Kornadt & Rothermund, 2011) showed that AARC and age
stereotypes are distinct constructs.
For the AARC-10 SF (Kaspar et al., 2019) psychometric properties have been
explored in a sample of individuals aged 40 and over. Regarding items reliability,
AARC gains and losses subscales obtained Cronbach’s a coefficients ranging from
.49 to .75 and from .58 to .75 respectively. Convergent validity of the AARC-10 SF
has been explored with the AARC-50 (Brothers et al., 2019), subjective age (Barrett,
2003) and ATOA (assessed with Lawton’s ATOA scale; Lawton, 1975). Cronbach
correlation coefficients between the AARC-10 SF and the AARC-50 reached as of
.89 for AARC losses and .88 for AARC gains. Small but significant associations were
found for more positive subjective age with more AARC gains and fewer losses. The
AARC-10 SF showed a moderate negative association between AARC losses and
ATOA but not between AARC gains and ATOA. External criterion validity of the
AARC questionnaire was explored using measures of emotional and physical well-
being. Higher levels of AARC losses were significantly related to more severe
depressive symptoms but to lower levels of overall life satisfaction and reduced
emotional well-being. On the contrary, higher levels of AARC gains were associated
with better well-being even though the associations were smaller compared to AARC
losses. Instead AARC gains showed stronger associations, compared to AARC

21
Chapter 1: Introduction

losses, with personality-related developmental outcomes such as personal growth


and personal relations. Personal growth and personal relations were assessed with
the AgeCog scales which capture age-related cognitions (Steverink et al., 2001).
AARC losses showed moderate associations with physical health status and
subjective health, whereas AARC gains did not show significant associations with
physical health.
Analyses of convergent and divergent validity for the AARC-50 have been
conducted on a sample of 424 individuals aged between 42 and 98 years old and
convergent, divergent, and external criterion validity of the AARC-10 SF were tested
in two subsamples of old (N=145) and advanced old (N= 118) individuals, aged from
70 to 79 years and 80 years and over respectively. Construct validity of the AARC-10
SF has never been explored in individuals younger than 70 years and construct
validity of the cognitive functioning subscale taken from the AARC-50 (Brothers et
al., 2019) has never been explored. Convergent and divergent validity for the AARC-
10 SF and AARC-50 cognitive functioning subscale with objective cognition and self-
reported cognitive variables (such as subjective memory complaints or subjective
cognitive impairment) have not been explored so far. Moreover, there is evidence
supporting cultural and age differences in levels of AARC (Brothers et al., 2016;
Miche et al., 2014). Hence the first step to enable research on AARC in the United
Kingdom is to validate the AARC measures in the UK population and to further
examine construct validity of the AARC-10 SF and the AARC-50 cognitive
functioning subscale. This is one of the aims of this research (Chapters 4 and 5).
The full list of items included in the AARC-10 SF and AARC-50 cognitive functioning
subscale is reported in Appendix A.

1.15 Possible uses of measures of AARC


Valid and reliable measures of AARC within the United Kingdom could serve several
purposes. First, the AARC-10 SF may be useful in clinical and counselling settings to
identify those individuals who, because of higher levels of AARC losses, may benefit
from interventions helping them to adapt to or counteract age-related changes
(Kaspar et al., 2019). Second, the AARC-10 SF could be used to assess the effects
of interventions on specific aspects of adults’ self-perceptions of ageing (Kaspar et
al., 2019). Third, among the five behavioural domains assessed with the AARC-50,
the cognitive functioning domain may be important in the area of prevention of

22
Chapter 1: Introduction

cognitive decline. Considering previous evidence showing that subjective cognitive


decline can predict future cognitive decline (Jessen et al., 2014), it may be that
subclinical cognitive changes, that are unlikely to be otherwise diagnosed, are
perceived and incorporated into perceptions of AARC. Hence scores on the AARC-
50 cognitive functioning subscale may serve as an indicator of subtle cognitive
change and a predictor of cognitive decline. The AARC-50 cognitive functioning
subscale could therefore potentially be used to identify those segments of the
population at greater risk of cognitive decline and who would benefit from closer
cognitive monitoring, cognitive training programs (Kaspar et al., 2019; pg. 3), or
interventions that help to minimise the negative psychological and functional impact
of age-related cognitive changes (Hahn & Lachman, 2015).

1.16 AARC and developmental outcomes


The empirical work conducted for this thesis will focus on the associations of AARC
with the developmental outcomes of cognitive functioning and physical health as
these generally decrease with ageing (Barnett et al., 2012; Liverman et al., 2015).
This research will also focus on depressive and anxiety symptoms as, after
dementia, they are the most common mental health disorders in old age (World
Health Organization, 2020b).

1.16.1 AARC and cognitive health


More positive scores on subjective ageing measures predict better cognitive
functioning (Hicks, 2017; Levy et al., 2002b; Robertson et al., 2016; Robertson et al.,
2015; Siebert et al., 2020; Siebert et al., 2016; Stephan et al., 2016a) and lower risk
of Alzheimer’s-like neuropathology (Levy et al., 2016), even among those individuals
with higher genetic risk of dementia (Levy et al., 2020b; Siebert et al., 2018).
Subjective ageing can even be a better predictor of cognition than well-established
risk factors for cognitive decline such as body mass index (Reijmer et al., 2011;
Stephan et al., 2015). Moreover, more positive subjective ageing is associated with
better psychological and physical health (Levy et al., 2015; Levy et al., 2009b; Moser
et al., 2011; Westerhof et al., 2014) which in turn predict better cognitive functioning
in later life (Stephan et al., 2016a; Vassilaki et al., 2015). However, the association
between cognitive functions and AARC has never been investigated.

23
Chapter 1: Introduction

1.16.2 AARC and emotional well-being and mental health


Better scores on measures of subjective ageing are associated with greater
emotional well-being and mental health (Kavirajan et al., 2011) including fewer
depressive (Kwak et al., 2014; Stephan et al., 2016a) and anxiety (Bryant et al.,
2012) symptoms. Similarly, those individuals with lower AARC losses and/or higher
AARC gains report better general emotional well-being (Brothers et al., 2017),
greater life satisfaction (Brothers et al., 2019), lower depressive symptoms (Dutt et
al., 2016a; Dutt et al., 2018), more positive and less negative affect (Miche et al.,
2014; Neupert & Bellingtier, 2017). However, the association of AARC with anxiety
has never been explored.
Existing evidence on other subjective ageing constructs such as age
stereotypes (Levy et al., 2014b) and subjective age (Stephan et al., 2015) suggests
that those with more positive subjective ageing are less likely to experience anxiety
symptoms. Moreover, on average individuals reporting more AARC losses score
higher on measures of neuroticism and negative affect and worry is a key
component of neuroticism, negative affect, and anxiety. (American Psychiatric
Association, 2013; Costa & McCrae, 1992; Harris & Dollinger, 2003; Miche et al.,
2014; Neupert & Bellingtier, 2017; Rupprecht et al., 2019). Finally, AARC is
associated with depressive symptoms and these often co-occur with anxiety in old
age (Schuurmans & Van Balkom, 2011; Van der Weele et al., 2009). It may therefore
be that AARC is also associated with anxiety symptoms. In the current thesis I will
investigate the associations of AARC with depressive and anxiety symptoms in the
UK and explore whether different profiles of individuals having varying degrees of
gains and losses differ in the experienced levels of depressive and anxiety
symptoms.

1.16.3 AARC and physical health


Individuals with positive subjective ageing generally have better physical health
including fewer chronic health conditions and better functional ability (Levy, 2003,
2009; Levy et al., 2015; Levy et al., 2002b; Moser et al., 2011; Robertson et al.,
2015). More AARC gains and fewer AARC losses are generally related to more
positive scores on indicators of physical health and well-being including self-rated
physical functioning, ability to complete daily tasks, bodily pain, and general health
(Brothers et al., 2017; Kaspar et al., 2019). However, existing evidence on the

24
Chapter 1: Introduction

association of AARC and physical health has several limitations. First, studies
reporting the association between AARC and physical health used self-report
measures of physical health (Brothers et al., 2016; Brothers et al., 2019; Kaspar et
al., 2019; Wahl et al., 2013). Second, the majority of studies (e.g. Dutt et al., 2016b)
treated self-rated health as a covariate while exploring associations between AARC
and other variables. Third, studies explored the associations of gains and losses with
physical health separately, without considering the coexistence of different
combinations of perceived gains and losses. Fourth, most studies are based on
cross-sectional analyses.
It would be important to explore the direction of the association of AARC with
indicators of physical health with longitudinal studies as there is evidence suggesting
that AARC and physical health may influence each other over time (Baali et al.,
2012; Brothers et al., 2020; Diehl & Wahl, 2010; Giles et al., 2010; Karp, 1988;
Kleinspehn-Ammerlahn et al., 2008; Kotter-Grühn et al., 2009; Lin et al., 2004;
Nilsson et al., 2000; Sherman, 1994; Spuling et al., 2013; Steverink et al., 2001;
Westerhof et al., 2014). Even though in this thesis I aimed to address all these
limitations, due to the disruption to data collection caused by COVID-19, this was not
possible. However, this thesis advances existing knowledge by exploring for the first
time whether individuals having varying degrees of AARC gains and losses differ in
their self-evaluations of health and in levels of functional ability.

1.17 Research questions


To sum up, self-perceptions of ageing are particularly important as they are related
to a wide range of health outcomes in the second half of life (Burmester et al., 2016;
Idler & Benyamini, 1997; Westerhof et al., 2014). Moreover, recent evidence
suggests that self-perceptions of ageing can be modified to enhance maintenance of
cognitive, mental, and physical health while ageing (Brothers & Diehl, 2017; Burnes
et al., 2019; Wolff et al., 2014). However, existing literature is mainly based on global
and unidimensional (either positive or negative) indicators of self-perceptions of
ageing. The multi-dimensional concept of awareness of positive (AARC gains) and
negative (AARC losses) age-related change (Diehl & Wahl, 2010) instead captures
the coexistence of AARC gains and losses across five life domains. Among
subjective ageing constructs, AARC appears to be most strongly associated with
developmental outcomes (Brothers & Diehl, 2017; Kaspar et al., 2019). However,

25
Chapter 1: Introduction

there is little evidence on the AARC concept in comparison to that available for other
subjective ageing concepts. This thesis aims to advance understanding of the AARC
concept by examining for the first time the psychometric properties of two measures
of AARC for the UK population aged 50 and over and by exploring the associations
of AARC gains and losses with a range of indicators of cognitive, mental, and
physical health. Some of these indicators such as subjective and objective cognitive
functioning and anxiety have never been explored in association with AARC. Finally,
this thesis will explore for the first time whether cognitive, mental, and physical health
and sociodemographic variables vary across individuals with different combinations
of degrees of AARC gains and losses.

The thesis has five research questions:


1. Are the AARC-10 SF and the AARC-50 cognitive functioning subscale valid,
reliable, and useful measures to assess perceived age-related gains and
losses among UK individuals aged 50 and over? (Chapters 4, 5, and 6)
2. Are AARC gains and losses associated with cognitive functioning? (Chapters
4, 6, and 7)
3. Are AARC gains and losses associated with mental health and emotional
well-being? (Chapters 3, 4, 6, and 7)
4. Are AARC gains and losses associated with indicators of physical health and
well-being? (Chapters 3, 4, and 7)
5. Are AARC gains and losses associated with sociodemographic variables?
(Chapters 4 and 7).

1.18 Structure of the thesis


This thesis consists of this general introduction followed by seven chapters: a
chapter outlining methodological details, a systematic review and meta-analysis, four
empirical cross-sectional papers, and the general discussion of the thesis. Chapters
3-7 are provided in the format of journal articles that have been published or have
been submitted for publication. These have been adapted where necessary for
inclusion in the thesis but are very similar to the published or submitted versions. A
summary of the content of each chapter follows below.

26
Chapter 1: Introduction

Chapter 2: Methodology
Chapter 2 provides an overview of the methodology of PROTECT (The Platform for
Research Online to investigate Genetics and Cognition in Ageing;
https://www.protectstudy.org.uk), which is the cohort study from which data used in
the empirical research presented in the thesis have been collected. It includes details
of the design, participants, study procedure, data access, and procedure for
inclusion of additional questionnaires, measures used, and data protection
procedure used in the thesis.

Chapter 3: Associations of awareness of age-related change with


emotional and physical well-being: A systematic review and meta-analysis
Chapter 3 is a systematic review and meta-analysis synthesising and quantifying
existing work on the associations of AARC with emotional and physical well-being in
adulthood. The review describes in which populations the associations of AARC with
emotional and physical well-being have been investigated, as well as the strengths
and limitations of the identified studies.

Chapter 4: International relevance of two measures of awareness of age-


related change (AARC)
The systematic search of the literature identified only studies based in the United
States and/or Germany. As subjective ageing can vary across cultures (Voss et al.,
2018), Chapter 4 investigates the psychometric properties for two measures of
AARC in the UK population aged 50 and over. One measure assesses general
evaluations of AARC (AARC-10 SF) whereas the other assesses AARC in the
cognitive domain (AARC-50 cognitive functioning subscale). This chapter also
explores whether age, sex, marital status, employment status, and education level
are predictors of AARC gains and losses.

Chapter 5: Exploring awareness of age-related change among over 50s


in the United Kingdom: Qualitative and quantitative findings from the
PROTECT study
The study presented in Chapter 5 uses qualitative analyses to examine and
categorise the written comments made by UK individuals aged 50 and over when
answering the AARC questionnaire. This study also uses quantitative analyses to

27
Chapter 1: Introduction

relate participants’ comments to levels of AARC gains and losses. Overall, this
chapter increases knowledge about what contributes to making UK individuals aware
of age-related changes as these develop in the second half of life. It also provides
additional information on the use of the AARC questionnaire in the United Kingdom.

Chapter 6: Cross-sectional association between objective cognitive


performance and perceived age-related gains and losses in cognition
The systematic search of existing work on AARC and health-related outcomes
(Chapter 3) identify no study on AARC and cognition. The study presented in
Chapter 6 therefore fills this gap in the literature. As AARC is the first concept that
assesses the coexistence of perceived positive and negative age-related changes,
this study investigates whether the association of AARC with objective cognition is
stronger compared to those found between objective cognition and well-established
unidimensional measures of subjective cognitive decline and subjective ageing.
Finally, to understand whether AARC in the cognitive domain is influenced by
psychological variables, in addition to objective cognitive functioning, Chapter 6
explores whether variability in AARC cognitive gains and losses is explained by
anxiety, depression, ATOA, subjective age, self-rated health.

Chapter 7: Differences in health among people with different profiles of


awareness of positive and negative age-related changes
Examination of previous literature on AARC (Chapter 3) showed that, despite the
AARC concept making it possible to assess the coexistence of perceived gains and
losses, this has never been explored in relation to health outcomes. The study
presented in Chapter 7 identifies in a sample of UK individuals aged 50 or over the
number and types of profiles characterised by different combinations of levels of
AARC gains and losses. This study also explores whether the identified profiles differ
in their levels of cognitive, mental, and physical health, as well as in their
sociodemographic characteristics.

Chapter 8: Discussion
The final chapter synthesises the findings from the five studies presented in this
thesis. Results are discussed in the context of the five research questions. This
chapter considers the validity of the AARC measures in capturing perceived positive

28
Chapter 1: Introduction

and negative age-related changes in individuals’ cognitive, mental, and physical


health. It also considers the utility of AARC gains and losses in providing information
about individuals’ levels of cognitive, mental, and physical health and well-being and
whether levels of AARC vary in relation to sociodemographic variables. Finally, this
chapter critically considers methodological issues relating to the research findings,
suggests directions for future work, and discusses the implications of this thesis.

1.19 Dissemination of findings


Dissemination through publication
To date, the following chapters have been submitted or accepted for publication:
Sabatini, S., Silarova, B., Martyr, A., Collins, R., Ballard, C., Anstey, K. J., Kim, S., &
Clare, L. (2020). Associations of awareness of age-related change with
emotional and physical well-being: A systematic review and meta-analysis.
The Gerontologist, 60(6), e477-e490. https://doi.org/10.1093/geront/gnz101
Sabatini, S., Ukoumunne, O. C., Ballard, C., Brothers, A. F., Kaspar, R., Collins, R.,
Kim, S., Corbett, A., Aarsland, D., Hampshire, A., Brooker, H., & Clare, L.
(2020). International relevance of two measures of awareness of age-related
change (AARC). BMC Geriatrics, 20(1), 359. https://doi.org/10.1186/s12877-
020-01767-6
Sabatini, S., Ukoumunne, O. C., Ballard, C., Brothers, A., Diehl, M., Wahl, H-W.,
Collins, R., Corbett, A., Brooker, H., Clare, L. (2020). Differences in health
among people with different profiles of awareness of positive and negative
age-related changes. Under review
Sabatini, S., Ukoumunne, O. C., Ballard, C., Collins, R., Kim, S., Corbett, A.,
Aarsland, D., Brooker, H., Clare, L. (2020). Exploration of Awareness of Age-
Related Changes among over 50s in the UK: Qualitative and Quantitative
Findings from the PROTECT study. Unider review
Sabatini, S., Ukoumunne, O. C., Ballard, C., Collins, R., Anstey, J., Diehl, M.,
Brothers, A., Wahl, H-W., Corbett, A., Hampshire, A., Brooker, H., Clare, L.
(2020). Cross-sectional association between objective cognitive performance
and awareness of age-related cognitive change. Under review

29
Chapter 1: Introduction

Dissemination through published abstracts


Sabatini, S., Silarova, B., Martyr, A., Collins, R., Ballard, C., Anstey, K. J., Kim, S., &
Clare, L. (2019). Associations of Awareness of Age-Related Change With
Emotional and Physical Well-being: A Systematic Review and Meta-analysis.
Oral presentation at Thinking Ageing and Older Age MICRA PhD and Early
Career Conference, British Society of Gerontology, Manchester
Sabatini, S., Silarova, B., Martyr, A., Collins, R., Ballard, C., Anstey, K. J., Kim, S., &
Clare, L. (2019). Associazioni tra consapevolezza di cambiamenti correlati
all'invecchiamento e benessere emotivo e fisico: review sistematica e meta-
analisi. Poster presentation at the XII Convegno nazionale Società Italiana di
Psicologia dell’Invecchiamento, Pescara, Italy
Sabatini, S., Silarova, B., Martyr, A., Collins, R., Ballard, C., Anstey, K. J., Kim, S. &
Clare, L. (2019). Meta-analysis of AARC and emotional and physical well-
being. Innovation in Aging, 3 (Suppl 1), S384
Sabatini, S., Ukoumunne, O. C., Ballard, C., Collins, R, Anstey, K. J., Diehl, M.,
Brothers, Kim, S., Corbett, A., Aarsland, D., Hampshire, A., Brooker, H.,
Clare, L. (2020). Higher perceived age-related gains and losses relate to
lower objective cognitive scores. Alzheimer’s Association International
Conference
Sabatini, S., Ukoumunne, O. C., Ballard, C., Brothers, A. F., Kaspar, R., Collins, R., Kim,
S., Corbett, A., Aarsland, D., Hampshire, A., Brooker, H., & Clare, L. (2020). UK
validation of two measures of awareness of age-related changes (AARC)”, Poster
presentation. 49th Annual General Meeting of the British Society of Gerontology,
online event
Sabatini, S., Ukoumunne, O. C., Ballard, C., Collins, R, Anstey, K. J., Diehl, M., Wahl, H.
-W., Brothers, Kim, S., Corbett, A., Aarsland, D., Hampshire, A., Brooker, H., Clare,
L. (2020). Cross-sectional relationship between objective cognitive performance
and awareness of age-related cognitive change. Emerging Researchers in Ageing
(ERA) virtual preconference of the British Society of Gerontology
Sabatini, S., Ukoumunne, O. C., Ballard, C., Brothers, A. F., Kaspar, R., Collins, R., Kim,
S., Corbett, A., Aarsland, D., Hampshire, A., Brooker, H., & Clare, L. (2020). UK
validation of two measures of awareness of age-related changes (AARC)”, Abstract
published for the online XII Convegno nazionale Società Italiana di Psicologia dell’
Invecchiamento

30
Chapter 1: Introduction

Sabatini, S., Ukoumunne, O. C., Ballard, C., Collins, R, Anstey, K. J., Diehl, M.,
Wahl, H. W., Brothers, Kim, S., Corbett, A., Aarsland, D., Hampshire, A.,
Brooker, H., Clare, L. (2020). Higher awareness of positive and negative age-
related changes relate to lower objective cognitive scores. GSA 2020 Annual
scientific meeting online
Sabatini, S., Ukoumunne, O. C., Ballard, C., Collins, R, Anstey, K. J., Diehl, M.,
Wahl, H. W., Brothers, A., Corbett, A., Hampshire, A., Brooker, H., Clare, L.
(2020) Health differences among people with varying profiles of awareness of
positive and negative age-related changes. GSA 2020 Annual scientific
meeting online.

Dissemination through academic presentations


Several scientific presentations have been made based on the findings from this
thesis:
Sabatini, S. (2018). Awareness of age-related change: a useful concept for
prevention of dementia in UK?. Oral presentation at the University of Exeter
School of Medicine and Health Annual Research Event, Torquay
Sabatini, S., Silarova, B., Martyr, A., Collins, R., Ballard, C., Anstey, K. J., Kim, S. &
Clare, L. (2019). Associations of Awareness of Age-Related Change With
Emotional and Physical Well-being: A Systematic Review and Meta-analysis.
Oral presentation at the University of Exeter School of Medicine and Health
Annual Research Event, Penryn
Sabatini, S. (May 2019). Awareness of age-related change (AARC) and its associations
with cognitive, mental and physical health and well-being in later life. Oral
presentation (upgrade talk) at the University of Exeter School of Medicine and
Health and School of Psychology, Exeter
Sabatini, S., Silarova, B., Martyr, A., Collins, R., Ballard, C., Anstey, K. J., Kim, S., &
Clare, L. (2019). Associations of awareness of age-related change with emotional
and physical well-being: A systematic review and meta-analysis. Poster
presentation during CLEScon 2019, University of Exeter, UK
Sabatini, S., (2019). Guest lecture on “Cognition in older age” for the module
“Neuropsychology of Ageing and Dementia”, University of Exeter, UK

31
Chapter 1: Introduction

Sabatini, S., (2020). Guest lecture on “Promotion of cognitive, mental and physical health
and Awareness of age-related change” for the module “Neuropsychology of Ageing
and Dementia”, University of Exeter
Sabatini, S., Ukoumunne, O. C., Ballard, C., Brothers, A., Diehl, M., Wahl, H-W., Collins,
R., Corbett, A., Brooker, H., Clare, L. (2020). Differences in health among people
with different profiles of awareness of positive and negative age-related changes,
Annual Research Event of the University of Exeter Medical School, online event

32
Chapter 1: Introduction

1.20 Conclusion
The way in which individuals perceive their own ageing is particularly important as it
is associated with maintenance of cognitive, physical and mental health in the
second half of life. Individuals’ perceptions of ageing appear to be modifiable and
hence a better understanding of their associations of these perceptions with health
outcomes may suggest ways of promoting health maintenance in later life. This
thesis focuses on the relatively new concept of awareness of positive and negative
age-related change as it captures in a more comprehensive way the complexity of
individuals’ experiences of ageing. As evidence on AARC is limited and restricted to
US and German samples, this thesis aims to validate two measures of AARC in the
United Kingdom and to better understand the associations of AARC with several
cognitive, mental, and physical health and wellbeing and sociodemographic
variables. The next chapter outlines the methodology used for the analyses in this
thesis.

33
Chapter 2: Methodology

Chapter 2: Methodology

34
Chapter 2: Methodology

2.1 Summary of the chapter


In this chapter I will provide a detailed account of the methodology used for the
systematic review and meta-analysis and for the four empirical studies included in
this thesis, which were all based on the same dataset. Three studies will involve
quantitative data analyses whereas one will involve both qualitative and quantitative
data analyses. I will provide an overview of the methodology of the PROTECT study
and how PROTECT has been used in order to conduct my PhD research. I will then
describe in detail the research design, participants, study procedure, measures
used, and data protection procedure used in the research reported in the thesis.

2.2 Platform for research online to investigate genetics and cognition in ageing
(PROTECT)
The UK version of the PROTECT study (https://www.protectstudy.org.uk) is a 25-
year longitudinal cohort study that started in November 2015. The PROTECT study
aims to explore the role of genetic, lifestyle, and medical factors on cognition in
individuals aged 50 years or over and living in the UK. The PROTECT study is
coordinated by researchers from the University of Exeter Medical School. University
of Exeter is the Sponsor of the UK version of the PROTECT project and both King’s
College London and University of Exeter Medical School are PROTECT study
research sites. PROTECT administrative teams are present both at King’s College
London and at the University of Exeter Medical School.

2.3 Design
Even though this thesis aimed to use longitudinal data, longitudinal data collection
was delayed due to COVID-19. Hence, the four empirical studies included in this
thesis are based on cross-sectional data collected in 2019 through the ongoing
PROTECT study. I decided to use PROTECT as a platform to implement the
research presented in this thesis for several reasons. First, it gave me the possibility
to include additional questionnaires (Appendix A) that were relevant to my thesis.
Second, the PROTECT study includes a variety of measures which make it possible
to explore the associations of AARC with several developmental outcomes. Third, I
was interested in a population of middle-aged and older individuals and the
PROTECT study includes participants aged 50 years and over. I was interested in
individuals aged 50 years and over as I aimed to examine the psychometric

35
Chapter 2: Methodology

properties of the AARC questionnaire among individuals aged 50 years and over as
they have been published only for individuals aged 70 year and older. Moreover, fifty
years is an appropriate cut-off that should make it possible to include in the research
individuals who are old enough to be likely to experience age-related changes (e.g.
Raz & Rodrigue, 2006; Shin et al., 2003; Siedlecki et al., 2005). In addition,
individuals in their 50s report being concerned with their physical health (Centre for
Ageing Better, 2015) and hence they may be willing to participate in studies that can
inform future interventions aiming to promote healthy ageing. Fourth, the wide age
range of participants and the large sample size of the PROTECT study made it
possible to explore the AARC concept across age sub-groups. Fifth, PROTECT
excludes individuals with dementia, which is an important exclusion criterion in the
present research.

2.4 Participants
For all the empirical studies conducted as part of my PhD I aimed for a minimum
sample size of 2,500 participants. Sample size calculation was based on the
analyses conducted in Chapter 4 as these analyses required the largest sample size.
To estimate the minimum sample size needed to obtain sufficient power in order to
correctly reject the null hypothesis when it should be rejected, I used an a-priori
sample size calculator for structural equation models
(https://www.danielsoper.com/statcalc/calculator.aspx?id=89 ). In this calculation I
considered several factors such as the number of latent (N = 2) and observed
variables (N = 10), anticipated effect size (correlation coefficient r = 0,1), alpha (α =
0,01) and power (0,95). The estimation indicated that a minimum sample size of
2,148 participants was needed. A minimum sample size of 2,500 participants was
chosen in order to take into account for possible drop-outs at follow-up.
What follows is an overview of PROTECT participants’ characteristics and an
explanation of how these participants were recruited at the beginning of the
PROTECT study. Individuals were potentially eligible to participate in the PROTECT
study if they were UK residents, English speakers, aged 50 years or older, had
access to a computer and internet and did not have a clinical diagnosis of dementia
at the time of recruitment. In PROTECT participants were recruited through several
channels including advertisement at King’s College London, invitation of participants
enrolled in existing UK-based cohort studies (https://exetercrfnihr.org/about/exeter-

36
Chapter 2: Methodology

10000/; https://www.joindementiaresearch.nihr.ac.uk/;
https://bdr.alzheimersresearchuk.org), and information leaflets placed in general
practitioners’ surgeries and memory clinics throughout the United Kingdom.
Recruitment for the PROTECT study began in 2013 and baseline assessment
started in November 2015. In 2015 (baseline assessment) 23,470 participants were
part of the PROTECT study. The majority of participants were white (92.2%) and
female (73.6%). Among the 23,470 PROTECT participants, 16.4% completed
secondary education, 11.8% completed post-secondary education, 20% obtained a
vocational qualification, 32% undertook an undergraduate degree, 16.5% undertook
a post-graduate degree, and 3.3% had a doctorate. Among the 23,470 PROTECT
participants, 43.1% were aged between 50 and 59 years, 43,2% were aged between
60 and 69, 12,4% were aged between 70 and 79 years old, 1,2% were aged
between 80 and 89 years, and 0.1% were aged 90 years or over. From 2017
additional PROTECT participants have been recruited through the University of
Exeter and the Royal Devon and Exeter NHS Trust.

2.5 Survey Procedure


In PROTECT data are collected using an online platform in which participants can
log in to the site and complete questionnaires and assessments. PROTECT
participants initially provided consent online through the PROTECT platform to
complete a baseline assessment and are invited to take part in a follow-up
assessment each year. Since when the PROTECT study was launched in November
2015 participants have been able to join the study at any point. Hence, the month in
which the annual assessment takes place varies across the cohort based on the
month in which participants undertook their baseline assessment. As part of the
PROTECT annual assessments in January 2019 (started on 10/01/2019) PROTECT
participants were asked to fill in additional questionnaires specific to this study. As
the intention was to sample from the PROTECT cohort until the desired sample size
was achieved, data collection for these additional questionnaires ended on
31/03/2019 as by that date a sufficient sample size was obtained. Only participants
that were sent a reminder about their PROTECT annual assessments between
January 2019 and the end of March 2019 could complete the additional measures
relevant to the current study. Between January 2019 and end of March 2019, 14,882
PROTECT participants completed the PROTECT annual assessment. Out of these

37
Chapter 2: Methodology

14,882 participants, 5,472 did not complete the additional measures added to the
PROTECT annual assessment for the purpose of the current study.

Figure 2.1. Diagram showing the number of participants that met inclusion
criteria in each of the four empirical studies

PROTECT participants that


undertook the PROTECT
annual assessment
between 1st January 2019
and 31st March 2019
N = 14,882

1st PROTECT participants that


empirical completed the AARC
study questionnaire in addition to
(Chapter 4) the PROTECT annual
assessment between
January and March 2019
N = 9,410

2nd PROTECT participants that 4th PROTECT participants that


empirical completed the AARC open- empirical completed the PROTECT
study ended question in addition study cognitive battery in addition
(Chapter 5) to the AARC questionnaire (Chapter 7) to the AARC questionnaire
N = 609. N = 6202

3rd PROTECT participants that


empirical completed the PROTECT
study cognitive battery in addition
(Chapter 6) to the AARC questionnaire
and that did not have
probable MCI or dementia
N = 6056

4rd PROTECT participants that

The following measures were added to the PROTECT study for the purpose of the
current research:
1. AARC-10 SF (Kaspar et al., 2019).
2. AARC-50 cognitive functioning subscale (Brothers et al., 2019).
3. Open-ended question asking participants to add their thoughts regarding the
AARC questionnaires.
4. ATOA five-item subscale taken from the Philadelphia Geriatric Center Morale
Scale (Lawton, 1975).

38
Chapter 2: Methodology

5. A single-item question assessing subjective age (adapted from the National


Survey of Midlife development in the Unites States; Barrett, 2003).
6. A single-item question assessing self-rated health (taken from the 36-ietm
Short Form Health Survey; Ware & Sherbourne, 1992).
As an additional measure, PROTECT participants were also asked to complete the
22-item Ruminative Response Scale (Nolen-Hoeksema et al., 2008). However, this
measure was not used in the current thesis as the intent was to explore rumination
as a mediating variable in the longitudinal association of AARC with mental health.

Moreover, to conduct this research I also used data collected as part of the main
PROTECT study during the annual assessment in January 2019. The following data
were requested:
1. Demographic information.
2. Lawton’s Instrumental Activities of Daily Living Scale (Lawton & Brody, 1969).
3. The Patient Health Questionnaire-9 (Kroenke et al., 2001).
4. The Composite International Diagnostic Interview-Short Form (Kessler et al.,
1998).
5. The Generalised Anxiety Disorder Assessment-7 (Spitzer et al., 2006).
6. The Informant Questionnaire on Cognitive Decline in the Elderly- Self Version
and Informant Version (Jorm & Jacomb, 1989).
7. PROTECT Cognitive Test Battery (Corbett et al., 2015).
8. Frequency of engagement in computerised cognitive training from 2015 to
2019.

The full questionnaires added to the PROTECT study for the purpose of this
research and the questionnaires that were requested from the main PROTECT
annual assessment can be found in Appendices A and B respectively. A detailed
description of these measures is provided in Section 2.9 of this chapter. The form for
the application for access to PROTECT study data and the approval can be found in
Appendices C and D.

39
Chapter 2: Methodology

2.6 Ethical approvals


Ethical approval for the present study was requested as an amendment to the ethical
approval for the PROTECT main study by the PROTECT team. The PROTECT
study was approved by NHS REC: London - London Bridge Research Ethics
Committee (13/LO/1578) on 15/11/2018. The HRA approval for the amendment was
granted on 15th November 2018 (details in Appendix E).
Ethical approval for the analyses proposed in this thesis was sought through the
ethics committee of the University of Exeter School of Psychology. (Application ID:
eCLESPsy000603 v1.0). Approval was granted on 06/02/2019 (details in Appendix
F).

2.7 Procedures for data access and adding questionnaires to the PROTECT
study
Permission to conduct this study was granted by the PROTECT Steering Committee
and PROTECT Academic Lead, Dr Anne Corbett, Senior lecturer, UEMS (Appendix
D). This covers:
1. Adding the AARC-10 SF and the AARC-50 cognitive functioning subscale to the
PROTECT platform.
2. Collecting data through the PROTECT online platform.
3. Creation by PROTECT team members of a bespoke data set for the analyses of
the present study.
In order to obtain permission from PROTECT Steering Committee, the PROTECT
Academic Lead was provided with a data access form (Appendix C); an outline of
the study (Appendix G); a Memorandum of Understanding that was set in place with
the authors of the AARC measures (Appendix H); and a data agreement form
(Appendix I). Full details of the PROTECT protocol and Participant Information Sheet
can be found in the appendices J and K. A similar procedure was undertaken to
request access to one-year follow-up data collected in 2020 (Appendix L). Even
though I obtained approval for access to PROTECT data collected in 2020
(Appendices M and N), due to delays in data collection related to COVID-19 the data
had not arrived 10 months after the expected date and by this time I needed to write
up the thesis.

40
Chapter 2: Methodology

2.8 Variables and measures


Sociodemographic information
Sociodemographic information collected included age, sex, ethnicity, marital status,
employment status, and education level.

Subjective ageing concepts


Awareness of age-related change (AARC)
The AARC-10 SF (Kaspar et al., 2019) is a valid and reliable tool for capturing
perceived age-related gains (AARC gains) and losses (AARC losses). It contains ten
items, five assessing AARC gains and five assessing AARC losses. Each of these
five items assesses a different AARC behavioural domain (health and physical
functioning, cognitive functioning, interpersonal relationships, socio-cognitive and
socio-emotional functioning, and lifestyle and life engagement). All ten items start
with the same stem “With my increasing age, I realise that…”. An example of an item
capturing AARC gains is “…I appreciate relationships and people much more”,
whereas an example of an item capturing AARC losses is “…I have less energy”.
Respondents have to rate how much each item applies to them on a five-point Likert
scale (1= not at all, 2= a little bit, 3= moderately, 4= quite a bit, and 5= very much).
Scores can be obtained for the AARC gains and AARC losses subscales by
summing items that fall into the respective subscales. Subscales scores range from
a minimum of five to a maximum of twenty-five with higher scores indicating higher
levels of awareness of age-related change. Cronbach’s a coefficients ranged
between .49 to .75 for items assessing AARC gains and between .58 to .75 for items
assessing AARC losses. Even though most studies on AARC used the 50-item
version of the AARC questionnaire, in my research I used the AARC-10 SF for
several reasons. First, research has shown that the AARC-10 SF has good
psychometric properties and can be used as a reliable alternative to the AARC 50-
item questionnaire. Second, as the sample used in my research includes individuals
in advanced old age and participants were asked to complete several
questionnaires, using the 50-item version of the AARC questionnaire would have
excessively burdened participants. Finally, for practical reasons I had to be restrictive
in the number of items I could add to the PROTECT annual assessment.

41
Chapter 2: Methodology

The cognitive functioning subscale of the AARC-50 questionnaire (Brothers et


al., 2019) includes ten items, five assessing AARC gains and five assessing AARC
losses. Two of the items included in the AARC-50 questionnaire overlap with items
included in the AARC-10 SF. An example of item capturing AARC gains in the
cognitive domain is “With my increasing age, I realise that I have become wiser”,
whereas an item capturing losses is “With my increasing age, I realise that I am
more forgetful”. The items response options and scoring procedure of the AARC-10
SF are applied. Cronbach’s a coefficients are .86 and .92 for AARC gains and AARC
losses respectively.
The following open-ended question was used to ask participants to add their
thoughts in relation to the AARC questionnaire: “If you wish, please add any
additional information or comments for this section in the box below”. This question
was asked only once at the end of both the AARC-10 SF and AARC-50 cognitive
functioning subscale.

Attitudes toward own ageing (ATOA)


To assess attitudes toward own ageing I used the ATOA five-item subscale (from the
Philadelphia Geriatric Center Morale Scale; Lawton, 1975). For each statement
respondents are asked to make temporal comparisons about changes in energy
level, perceived usefulness, happiness, and quality of life and to respond on a binary
response set (better versus worse, yes versus no). An example item is “things keep
getting worse as I get older”. A proportion-based score can be obtained by summing
participant’s item scores and by dividing it by the number of responses. A value
between zero and one can be obtained with one representing a positive response on
all answers and a score of zero representing a negative response in all answers.
Regarding validity of the ATOA scale, Cronbach’s a coefficient reached .85.

Subjective age
I assessed subjective age with the single-item question “Many people feel older or
younger than they actually are. Fill in the age (in years) that you feel most of the
time: ___” (adapted from the National Survey of Midlife development in the United
States; Barrett, 2003). A proportional discrepancy score was calculated by
subtracting the participants’ subjective age from their chronological age, and by

42
Chapter 2: Methodology

dividing this difference score by participants’ chronological age. A positive value


indicates a younger subjective age, while a negative value indicates an older
subjective age.

Indicators of physical health


Instrumental activities of daily living
To measure individuals’ overall functional ability at the present time I used a modified
version of the Lawton’s Instrumental Activities of Daily Living Scale (Lawton & Brody,
1969). It assesses seven activities including preparing meals, managing
medications, and using the telephone. Two of the original eight items (assessing
laundry and housekeeping) were combined. For each activity respondents rate how
difficult they find doing the activity. Answers vary from zero to two (0= no difficulty,
1= some difficulty, and 2= great difficulty) and the total score ranges from a possible
zero to fourteen. Little research on the psychometric properties of the IADL Scale
has been reported. Inter-rater reliability testing resulted in a correlation of .85
(Lawton & Brody, 1969).

Self-rated health
I assessed self-rated health with a single-item question “Would you say that for
someone of your age, your own health in general is:___?” (taken from the 36-ietm
Short Form Health Survey; Ware & Sherbourne, 1992). Participants had to rate their
own health on a four-point scale ranging from excellent to poor (1= excellent, 2=
good, 3= fair, and 4= poor).

Indicators of mental health


Depression
I assessed the presence of depressive symptoms over the past two weeks with the
Patient Health Questionnaire-9 (Kroenke et al., 2001) which is based on the
diagnostic criteria for major depressive disorder described in the Diagnostic and
Statistical Manual Fourth Edition (DSM IV; American Psychiatric Association, 2000).
Respondents are asked to indicate how frequently they experience each symptom
on a four-point Likert scale (1= not at all, 2= several days, 3= more than half the
days, and 4= nearly every day). The total score is the sum of the item scores and
can range from nine to thirty-six. The PHQ-9 has excellent internal reliability, with

43
Chapter 2: Methodology

Cronbach’s α coefficients ranging from .86 to .89, and excellent reliability with
Cronbach’s α coefficient of .84 (Kroenke et al., 2001).

Lifetime depression and anxiety


To assess lifetime depressive and anxiety symptoms I used the Composite
International Diagnostic Interview-Short Form (CIDI-SF; Kessler et al., 1998)
including nine items assessing depressive symptoms and eight assessing anxiety
symptoms. An example of a depressive symptom question is “did you lose interest in
most things?”. For each item, participants can answer “yes” if they have the
symptoms, “no” if they did not have the symptoms. For both depression and anxiety
a total score can be calculated by summing the items where the participants answer
yes. For depression and anxiety the total score can range from zero to nine and from
zero to eight, respectively.

Generalised anxiety
To assess symptoms of generalised anxiety I used the Generalised Anxiety
Disorder-7 (Spitzer et al., 2006). Respondents are asked to indicate the frequency of
occurrence of a list of symptoms over the past two weeks on a four-point scale (1=
not at all, 2= several days, 3= more than half the days, and 4= nearly every day).
The overall score is the sum of the item scores and ranges from seven to twenty-
eight. The Generalised Anxiety Disorder-7 is a valid measure with Cronbach’s as
between .69 and .81 (Spitzer et al., 2006).

Indicators of cognitive functioning


Cognitive function - informant rating
The Informant Questionnaire on Cognitive Decline in the Elderly short form (Jorm,
1994; Jorm & Jacomb, 1989) was administered to obtain information about
participant’s cognitive change over the last ten years from someone close to them.
Items describe both cognitive improvement and cognitive decline and can be
answered on a five-point scale (1= much improved, 2= a bit improved, 3= not much
change, 4= a bit worse, and 5= much worse). The final score is the mean of the item
scores. A parallel version of the questionnaire (Informant Questionnaire on Cognitive
Decline in the Elderly – Self) was administered to the participant (Jorm & Jacomb,

44
Chapter 2: Methodology

1989). The questionnaire has high internal reliability with Cronbach’s a of .95 (Jorm
& Jacomb, 1989).

Cognitive functioning - objective assessment


I measured cognitive functioning with the PROTECT Cognitive Test Battery (Corbett
et al., 2015; Huntley et al., 2018) which includes four tests: (1) the Grammatical
Reasoning task assesses verbal reasoning (Baddeley, 1968) by asking participants
to determine the accuracy of a series of grammatical statements about a presented
picture; (2) the Digit Span task (Huntley et al., 2017) assesses verbal working
memory by asking participants to repeat sequences of numbers; (3) the Self-Ordered
Search task measures spatial working memory (Owen et al., 1990) by asking
participants to recall the position of a hidden object behind a series of panels; and (4)
the Paired Associate Learning task (Owen et al., 1993) assesses visual episodic
memory. In this task participants are presented with a series of objects in cells and
instructed to remember the locations of the objects. Participants are subsequently
asked to select the location where the object was initially presented. For each task a
summary score can be obtained by subtracting the number of errors from the
number of correct answers. For digit span the summary score can range from zero to
twenty. For paired associates learning the summary score can range from zero to
sixteen. For grammatical reasoning the summary score has no upper or lower limit
as this depends on speed, accuracy, total correct answers, and total errors. Finally,
the self-ordered search summary score can range from zero to twenty. The same
scoring was applied to all individuals, irrespective of their age or education level.

2.9 Data protection procedures


Data were collected via the merged King’s College and Exeter PROTECT platforms.
Personal data were collected on a separate database from all other data. This was
matched up with a unique ID number for each participant. Personal data were held
separately by King’s College London and Exeter University, depending on the
recruitment route taken by the participant. Databases were password-protected and
located on University shared drives which only the study team at that University had
access to. The data were also stored on the Universities’ servers which were not
accessible to anyone outside the organisations. All non-identifiable data (i.e. data

45
Chapter 2: Methodology

that were not personal) were stored in independent databases at the two sites in
Exeter University and King’s College London. These data were accessible for import
to either site to enable creation of a full database for analyses purposes.
Raw data (not including personal details) were stored in encrypted form, subject to
ISO/IEC 27001:2005 Certification. Only core study team members had access to
personal data through the electronic database. This included the Chief Investigator,
study coordinators and IT specialist. Once data had been collected, colleagues from
the PROTECT team were able to extract relevant data requested for this study and
prepare this to be sent anonymously.
In order to facilitate analyses for the present study, PROTECT team members
created a bespoke dataset that did not include any personal data. Quantitative data
were managed according to the data protection and governance requirements set
out in the standard procedures for the ethical approval of research by the University
of Exeter. Access to data was limited to the minimum number of individuals
necessary for analyses. Access to the databases was password-protected and
limited to me (Serena Sabatini), Professor Linda Clare, Professor Obioha C
Ukoumunne, and Dr Rachel Collins. The study database was held on University
servers in restricted-access buildings. All university computers were password-
protected and access to electronic drives and folders containing study data was
restricted to the research team. I (Serena Sabatini) was the custodian of the
generated data, including the implementation of the data management plan, storage
and initial archive. University of Exeter was responsible for the storage of the data,
storage backup and recovery data during the project.

2.10 Statistical analyses


Description of statistical analyses used for each of the five studies presented in this
thesis will be discussed in the respective chapter.
To estimate the minimum sample size needed to obtain sufficient power in
order to correctly reject the null hypothesis in the four empirical studies, I used a-
priori sample size online calculator (Soper) and G* Power (Faul et al., 2009).

2.11 Conclusion
The chapter has provided an overview of the PROTECT study, including details
regarding participants, procedures, data collection, and measures used. It has also

46
Chapter 2: Methodology

provided an overview of the variables that are used in this thesis and how they are
scored for analyses.

47
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

Chapter 3: Associations of awareness of age-related change with emotional


and physical well-being: A systematic review and meta-analysis

48
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

3.1 Summary
Numerous studies have linked the way in which individuals experience ageing to
cognitive, mental, and physical health outcomes. However, a variety of concepts
capturing subjective ageing exist. Existing systematic reviews and meta-analyses
have so far synthesised and quantified the associations of health outcomes for some
of these concepts but not for awareness of age-related change. The systematic
review and meta-analysis presented in this chapter aims therefore to explore and
quantify the associations of AARC with cognitive, emotional, and physical well-being
in existing empirical work.
This paper has been published: Sabatini, S., Silarova, B., Martyr, A., Collins, R.,
Ballard, C., Anstey, K. J., Kim, S., & Clare, L. (2020). Associations of awareness of
age-related change with emotional and physical well-being: A systematic review and
meta-analysis. The Gerontologist, 60(6), e477-e490.
https://doi.org/10.1093/geront/gnz101

Abstract
Background: This systematic review aimed to synthesise and quantify the
associations of awareness of age-related change (AARC) with emotional well-being,
physical well-being, and cognitive functioning.
Method: We conducted a systematic review and a random effects meta-analysis to
pool correlation coefficients. We included quantitative studies, published from 1st
January 2009 to 3rd October 2018, exploring associations between AARC and one or
more of the following outcomes: emotional well-being, physical well-being, and
cognitive functioning. We assessed heterogeneity (I2) and publication bias.
Results: We included twelve studies in the review, nine exploring the association
between AARC and emotional well-being and eleven exploring the association
between AARC and physical well-being. No study explored the association between
AARC and cognitive functioning. Six articles were included in the meta-analysis. We
found a moderate association between a higher level of AARC gains and better
emotional well-being (r= .33; 95% CI: .18, .47; p< .001; I2= 76.0%) and between a
higher level of AARC losses and poorer emotional (r= -.31; 95% CI: -.38, -.24; p<
.001; I2= 0%) and physical well-being (r= -.38; 95% CI: -.51, -.24; p< .001; I2=

49
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

83.5%). We found a negligible association between AARC gains and physical well-
being (r= .08; 95% CI: .02, .14, p= .122; I2= 0%). Studies were of medium-to-high
methodological quality.
Discussion: There is some indication that AARC gains and losses can play a role in
emotional well-being and that AARC losses are associated with physical well-being.
However, the number of included studies is limited and there was some indication of
heterogeneity.
PROSPERO registration: CRD42018111472.

50
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

3.2 Introduction
The proportion of older individuals in the world is increasing (United Nations, 2019)
and as a consequence the number of people that experience poor emotional and
physical well-being and poor cognitive health, including dementia, is also increasing
(United Nations, 2019). Poor emotional and physical well-being and dementia place
a high economic burden on healthcare services (Morrison, 2008; Prince et al., 2015).
As a result, increasing emphasis is being placed on research into prevention of poor
emotional and physical well-being, cognitive impairment, and dementia (World
Health Organization, 2015). There is great individual variability in levels of emotional
and physical well-being, and of cognitive functioning, among older people (e.g.
Deary et al., 2009; Gow et al., 2007; Jagger et al., 2009). A better understanding of
those factors that account for individual variability may be useful in suggesting ways
of preventing poor emotional well-being, physical well-being, and cognitive decline
with ageing.
Empirical evidence has shown that positive subjective ageing, the way in
which individuals experience ageing, is associated with better health and well-being
(Westerhof et al., 2014). Previous studies, for example, have found associations
between positive subjective ageing and better physical and cognitive functioning
(Levy et al., 2002b; Robertson et al., 2016; Robertson et al., 2015), lower risk of
medical problems such as cardiovascular events, falls, and hospitalisations (Levy et
al., 2015; Levy et al., 2009b; Moser et al., 2011), higher longevity (Levy et al., 2002b;
Moser et al., 2011), and lower levels of Alzheimer’s-like neuropathology (Levy et al.,
2016). Subjective ageing covers a range of concepts attempting to understand the
ways in which individuals experience the ageing process (Westerhof et al., 2014).
Examples of such concepts are self-perception of ageing (Kleinspehn-Ammerlahn et
al., 2008; Kotter-Grühn et al., 2009), subjective age, age identity (Kotter-Grühn &
Hess, 2012), attitudes toward ageing (Bennett & Eckman, 1973), and age
stereotypes (Levy, 2003, 2009).
A more recently-introduced subjective ageing concept is awareness of age-
related change (AARC; Diehl & Wahl, 2010). AARC refers to “a person’s state of
awareness that his or her behaviour, level of performance, or way of experiencing
life has changed as a consequence of having grown older”(Diehl & Wahl, 2010; pg.

51
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

342). The term was introduced in 2010 by an international working group to


overcome some of the limitations of previous subjective ageing concepts. In contrast
to other subjective ageing concepts, AARC acknowledges that individuals’ subjective
ageing experience may vary across different life and behavioural domains (for
example health and physical functioning, cognitive functioning, interpersonal
relationships, socio-cognitive and socio-emotional functioning, and lifestyle and life
engagement). It is also the first construct recognising that people can experience
both positive and negative age-related changes, called AARC gains and AARC
losses respectively. An example of AARC gains in the cognitive functioning domain
is “I have become wiser”, whereas an example of AARC losses in the same domain
is “I am slower in my thinking”. An example of AARC gains in the health and physical
functioning domain is “I pay more attention to my health” whereas an example of
AARC losses in the health and physical functioning domain is “I have less energy”.
Interestingly, AARC gains and AARC losses may occur simultaneously even in the
same behavioural or life domain.
Different to other subjective ageing concepts (e.g. attitudes toward ageing and
age stereotypes) that act at a pre-conscious level, AARC has been conceptualised
as a form of conscious self-knowledge. This makes the AARC concept potentially
useful to identify individuals who, because they experience high awareness of
negative age-related change, may be highly motivated to engage in health-promoting
interventions.
Assessing AARC in a more standardised way has been made possible with
the introduction of a questionnaire specifically designed to investigate AARC (Wahl
et al., 2013). There are now four published versions of the questionnaire with
differing lengths across two languages. In English there is a 50-item questionnaire
(Brothers et al., 2019), a shorter 20-item version (Neupert & Bellingtier, 2017), and a
further ultra-short version (AARC-10 SF; Kaspar et al., 2019). The 50-item
questionnaire resulted from the refinement of two original forms of the questionnaire
consisting of 189 and 100 items respectively (Brothers, 2016). The 20-item and 10-
item versions are made up of selected items from the AARC 50-item version. In
German, an initial version consisted of 50 items, which were translated and included

52
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

in the 189-item version, and this 50-item version was subsequently refined to provide
a 32-item version (Wahl et al., 2013).
There has been some research investigating AARC gains and AARC losses
and their associations with emotional well-being (Dutt et al., 2016a; Miche et al.,
2014; Neupert & Bellingtier, 2017) and functional health (Brothers et al., 2017;
Kaspar et al., 2019). It is therefore an appropriate time to conduct a meta-analysis to
summarise findings on the associations of AARC gains and AARC losses with
emotional and physical well-being and cognitive functioning, to provide a foundation
for future AARC research. Since AARC gains and AARC losses have been
conceptualised as two distinct sub-components of the same AARC construct,
contrary to other subjective ageing concepts, it is possible to separately study AARC
gains and AARC losses and to investigate their unique associations with health
outcomes. This increased knowledge may be useful in understanding whether the
way in which people experience ageing explains some of the individual differences in
health and well-being over time.
The aim of this systematic review is to synthesise and quantify the
associations of AARC with emotional and physical well-being and cognitive
functioning reported in studies using quantitative study designs and including only
cognitively healthy adult participants (aged 18 years and over). In addition, we aimed
to understand study characteristics, the populations in which associations of AARC
with emotional and physical well-being and cognitive functioning have been
investigated, and the strengths and limitations of the available studies exploring such
associations.

3.3 Methods
We conducted a comprehensive systematic review in accordance with the Preferred
Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA; see Appendix
O) Statement (Moher et al., 2009) and following an a priori protocol. The protocol is
registered with PROSPERO (registration number: CRD42018111472).

53
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

Systematic search strategy


The search strategy was developed in consultation with expert librarians and used
free text terms (Table 3.1). The following search terms were used: “awareness of
age-related change” or “awareness of age-related change”; AARC*; “views on aging”
or “views on ageing”. We searched the following healthcare databases and trial
registers: Medline, EMBASE, PsycINFO, CINAHL, AgeLine, AMED, and Web of
Science (Core collection) from 1st January 2009 to 3rd October 2018. We searched
for grey literature through PROSPERO, Scopus, and OpenGrey from 1st January
2009 to 3rd October 2018. The start for the searches was set as 2009 because the
AARC concept was introduced in 2010. We also searched OpenThesis, British
Library EThOS, and PROQUEST with no time limits.

Table 3.1 Example of search terms used in Medline (R)


Step Terms
1 All fields:
“awareness of age related change” or “awareness of age-related change”
2 All fields:
AARC*
3 All fields:
“views on aging” or “views on ageing”
4 1 or 2 or 3
5 Additional limits: Humans, adults and adolescents, 2009 to current

Inclusion criteria for the review


We included articles if they: a) reported quantitative data, including clinical trials,
quasi-experimental studies with comparison groups, and cross-sectional and
longitudinal observational studies; b) explored the concept of AARC as defined by
Diehl and Wahl (2010) and assessed AARC with one of the following measures:
AARC-189, AARC-100, AARC-50 (Brothers et al., 2019), AARC-20 (Brothers et al.,
2017; Neupert & Bellingtier, 2017), AARC-10 SF (Kaspar et al., 2019), AARC-32
(Wahl et al., 2013) or diaries (Miche et al., 2014) asking individuals to report their
subjective ageing experiences in relation to the five behavioural domains described
by Diehl and Wahl (2010); c) investigated associations between AARC and one of

54
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

the following outcomes: emotional well-being, physical well-being, or cognitive


functioning; d) used any self-rated, informant-rated and/or objective assessments of
emotional and/or physical well-being or cognitive functioning; e) were published from
2009 onwards; f) were written in English or German; and g) included only cognitively-
healthy adult participants (aged 18 years and over) or reported data for cognitively-
healthy participants separately.

Procedure
All citations identified in the searches were independently assessed for eligibility by
two reviewers (SS, BS), and any obviously ineligible titles were removed. At title
level all articles on subjective ageing were retained, because it was not possible to
determine from titles whether studies focused on AARC or on another subjective
ageing concept, given that in the subjective ageing literature authors often use
different terms to refer to the same concept. The same reviewers (SS, BS) then
independently assessed the abstracts of the remaining articles. Any that clearly did
not meet inclusion criteria were excluded. At abstract level reviewers retained all
articles that included a measure of AARC irrespective of whether outcome measures
assessing emotional and/or physical well-being and/or cognitive functioning were
listed in the abstract. This was to avoid excluding articles that assessed the
association of AARC with emotional and/or physical well-being and/or cognitive
functioning but did not report exploration of these associations in the abstract. The
full-text version of all potentially eligible studies was assessed for suitability by two
reviewers (SS, BS). At every step, discrepancies between the first and second
reviewers were resolved through discussion and the involvement of a third reviewer
(RC). Once full texts were identified, potential additional articles meeting inclusion
criteria were sought through forward (studies that cited the article) and backward
(studies cited in the article) searching of references of included articles.
We used a data extraction proforma to extract data from the articles included
in the review. We extracted data covering study information (author, title, journal,
year of publication, country of origin, funding sources, and language), study
characteristics (study aims, design, inclusion/exclusion criteria, sample size, study
period, recruitment, response rate, and follow-up times), participants’ characteristics

55
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

(age, gender, education level, ethnicity, and health status), measure of AARC used,
outcome measures, and results.
Where relevant data were unavailable, the authors of the study were
contacted to request missing information. Additional information was sought for four
articles. All authors replied and provided baseline bivariate correlations between
AARC gains and AARC losses and physical well-being for these four articles
included in the systematic review of which two were included in the meta-analysis.
One author provided data for the association between AARC gains and AARC
losses and emotional well-being, and the article was also included in the meta-
analysis.
All data were extracted by the first reviewer (SS). The second reviewer (BS)
checked 30% of the extracted information. Disagreements between the reviewers
were resolved through discussion. We calculated the percent agreement as a
measure of inter-rater reliability. The methodological quality of included articles was
assessed by the first author (SS) using the Critical Appraisal Skills Programme
(CASP) checklist for cohort studies (Critical Appraisal Skills Programme, 2018).
Each item was scored as low, medium, or high and each article received an overall
quality score ranging from low to high. Some articles received an overall quality
score of medium-to-high as they included an equal number of items scored as
medium and high.

Statistical analyses
We conducted a random effects meta-analysis of cross-sectional findings following
the procedure outlined by Borenstein et al. (2005), as the included studies employed
different methods of assessing AARC and emotional and physical well-being, and
included heterogeneous samples of middle-aged and older individuals. The random
effects model estimates the magnitude of heterogeneity and incorporates this into
the overall estimated effect (DerSimonian & Laird, 1986). Where multiple studies
used data from the same cohort and reported findings based on the same emotional
and/or physical well-being and AARC measures, we planned to include in the meta-
analysis the articles reporting the largest sample size and longest follow-up duration.
We used a standardised correlational direction and where necessary the direction

56
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

was changed to facilitate cross-study comparisons. Estimated effect sizes ≤ .09 were
considered negligible, .10 to .29 small, .30 to .49 moderate, and ≥ .50 large (Cohen,
1988). Between-study heterogeneity for each measure of emotional and physical
well-being was quantified using an index of inconsistency (I2, Higgins, Thompson,
Deeks, & Altman, 2003). This statistic calculates a percentage of heterogeneity
resulting from study differences that is not due to chance. Larger values indicate
larger heterogeneity.
All computations were based on Fisher’s z transformation of the correlation
coefficient and were conducted using the Comprehensive Meta-Analysis 2
(Borenstein et al., 2005) software package which calculated average z scores and p-
values, weighted effect for Pearson’s r correlation coefficients, and 95% confidence
intervals (CI) for the pooled effect sizes. One of the included studies (Kaspar et al.,
2019) presented correlations between one measure of AARC and multiple measures
of emotional and physical well-being. The software package was therefore instructed
to average the multiple within-study correlations to correct for violations of
independence. For all the remaining studies only one correlation was reported per
analyses. To visualise the amount of variation between the studies and an estimate
of the overall effect size (Lewis & Clarke, 2001), we created forest plots using Forest
Plot Viewer (Boyles et al., 2011).
To address the risk of possible publication bias, where studies with non-significant
findings are less likely to be published than those with significant findings, funnel
plots and Egger’s regression intercepts were calculated using Comprehensive Meta-
Analysis 2. Funnel plots were used to visualise possible publication bias (Borenstein
et al., 2009) while Egger’s test provides information about the degree of possible
publication bias.
In total, we conducted four analyses. The first and second analyses
investigated the associations of emotional well-being with AARC gains and AARC
losses respectively. The third and fourth analyses investigated the associations of
physical well-being with AARC gains and AARC losses respectively.

57
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

3.4 Results
After removing duplicates, we screened 814 titles, 125 abstracts, and 27 full-text
articles. We excluded seventeen articles at full-text screening as they either
examined other subjective ageing constructs (N= 15), did not explore relevant
outcomes (N= 1), or were theses where data had been published in an included
journal article (N= 1). Ten full-text articles met the inclusion criteria for the systematic
review. Inter-rater agreement at title, abstract, and full-text level was 94%, 90%, and
96% respectively. Through forward and backward chasing of references two more
articles were identified, resulting in 12 articles being included in the review (Figure
3.1). Five studies (Brothers et al., 2019; Brothers et al., 2017; Dutt et al., 2016a,
2016b; Dutt & Wahl, 2018) were excluded from the meta-analysis as they reported
data from the same cohort of included studies and reported findings based on the
same outcome measures. One study (Miche et al., 2014) was excluded from the
meta-analysis as it reported associations between emotional and physical well-being
and overall AARC, instead of AARC gains and AARC losses. We included in the
meta-analysis articles that explored emotional and/or physical well-being as main
outcomes and/or as covariates. Six articles were included in the meta-analysis.
Authors of the studies provided correlational data on the association between AARC
and emotional well-being for one article and correlational data on the association
between AARC and physical well-being for two articles.

58
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

Records identified through


database searching
Identification

(N= 1138)

Records after removing duplicates


(N= 814)

Papers excluded, with reasons (N= 689)


Records screened at title level Before 2009 (N= 235)
(N= 814) Other languages (N= 2)
Title irrelevant (N= 452)
Screening

Papers excluded, with reasons (N= 98)


Records screened at abstract level
Conference abstracts (N= 16)
(N= 125)
Studies not on subjective ageing (N= 9)
Other subjective ageing constructs than
AARC and views on ageing (N= 22)
Theoretical/qualitative (N= 28)
Book (N= 6)
Meta-analysis (N= 1)
Not relevant outcome (N= 16)
Eligibility

Full-text articles assessed for Full-text articles excluded from systematic


eligibility review, with reasons (N= 17)
(N= 27) Other subjective ageing constructs (N= 15)
Not relevant outcome (N= 1)
Thesis of a paper (N= 1)
Articles included through
forward and backward
chasing (N= 2)

Studies included in narrative Full-text articles excluded from meta-


synthesis analysis, with reasons (N= 6)
(N= 12) Same study populations and same
measures of emotional and physical well-
Included

being (N= 5)
Assessment of overall AARC (not gains
and losses) (N= 1)

Studies included meta-analysis


(N= 6)

Figure 3.1 Preferred reporting items for systematic reviews and meta-analysis
(PRISMA) 2009 flow diagram.

59
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

Characteristics and populations of studies included in the review


A summary of the study characteristics, participants, AARC measures, and quality
assessment is shown in Table 3.2. Among the twelve articles included in the
systematic review, eight articles were cross-sectional and four were longitudinal.
Two studies (Miche et al., 2014; Neupert & Bellingtier, 2017) took diary-based
approaches, and for the present review these were conceptualised as cross-
sectional studies. One of these (Neupert & Bellingtier, 2017) included micro-
longitudinal work (a 9-day study) exploring variability in the association between
AARC and emotional well-being. We did not identify any clinical trials or quasi-
experimental studies with comparison groups. Most participants were white women
with above average levels of education and self-reported health. Six articles included
German participants only, three included US residents only, and three included both
US and German participants. Sample sizes ranged between 71 and 819. AARC,
emotional well-being, and physical health were assessed using different measures
across the included studies (Table 3.2).
Among the twelve articles included in the systematic review, three were
judged as being of medium methodological quality, five were judged as being of
medium-to-high methodological quality, and four were judged as being of high
methodological quality. No study was of low methodological quality (details of the
CASP are available in Table 3.2). Where articles were downgraded to medium and
medium-high ratings, this was due to limited generalisability of study findings
resulting from the specific composition of the sample.

60
Chapter 3: Associations of awareness of age-related change with emotional and physical well-being: a systematic review and meta-

analysis

Table 3.2. Characteristics of studies included in the review.


Study characteristics Population characteristics Measures
Author Study cohort Study design N in Age, M Women (%) Exposure: Outcome Study
analysis (SD), range AARC quality
in years measure (CASP)
(Brothers Cohort of adults Longitudinal Baseline 64.13, 63.3 AARC-50 1) Emotional well-being: SPWB- Medium
2
et al., from the United (2.5 years) N= 819; (12.85), 40- SF (Ryff & Keyes, 1995).
2016) States and 2.5 years 98 2) Physical well-being: Self-rated
Germany aged 40- follow-up health through a single-item
98 (Diehl and Wahl N = 537 question (Baali et al., 2012).
as principal
investigators)
(Brothers Same cohort as Cross- N= 819 64.13, 63.3 AARC-189 1) Physical well-being: SF-36v23 Medium-
et al., above sectional (12.85), 40- (Ware et al., 2007). High
2017) analysis 98 2) Emotional well-being: SWLS4
(Glaesmer et al., 2011).
(Kaspar Same cohort as Cross- N= 819 64.13, 60 AARC-50; 1) Physical well-being: a)SF-36v2 High
et al., above sectional (12.85), 40- AARC-10 SF (Ware et al., 2007). b) Subjective
2019) analysis 98 health through a single-item
question.
2) Emotional well-being: a) SWLS
(Diener et al., 1985). b) SPWB-SF
(Ryff & Keyes, 1995). c) CES-D-
R105 (Baron et al., 2017).

61
Chapter 3: Associations of awareness of age-related change with emotional and physical well-being: a systematic review and meta-

analysis

(Brothers Cohort of adults Cross- N= 424 69.53, 52.4 AARC-50 1) Physical well-being: self-rated High
et al., from the United sectional (12.52), 42- health through a single-item
2019) States and analysis 98 question.
Germany aged 40-
98 (Diehl and Wahl
as principal
investigators)
(Dutt et Cohort of adults Longitudinal Baseline Baseline Baseline 64 AARC-50 1) Emotional well-being: CES-D Medium-
al., from the United (2.5 years) N= 423; 62.94, (Radloff, 1977). High
2016a) States and 2.5 years (11.84), 40- 2) Physical well-being: SF-36-
Germany aged 40- follow-up 98 German version9 (Bullinger,
98 (Diehl and Wahl N= 356 1995).
as principal
investigators)
(Dutt et Same cohort as Cross- N= 356 64.04, 65 AARC-50 1) Physical well-being: SF-36- High
al., above sectional (11.38), 42- German version (Bullinger, 1995).
2016b) 100
Neupert Mindfulness and Micro- N= 116 64.71, 61 AARC 20- 1) Emotional well-being: PNAS6 Medium
and Anticipatory Coping longitudinal (4.98), 60- item version (Watson et al., 1988).
Bellingtier Everyday study (9 days) 90
(2017) (United States)

62
Chapter 3: Associations of awareness of age-related change with emotional and physical well-being: a systematic review and meta-

analysis

(Wahl et Cohort from the Cross- N= 265 64, (11), 69 AARC-32 1) Emotional well-being: SWLS- High
al., 2013) project “awareness sectional 40-87 item German version (Diener et al.,
of age-related 1985).
change: a cross- 2) Physical well-being: subjective
cultural health through a single-item
collaboration” question.

(Dutt & Cohort of adults Longitudinal Baseline Baseline 64 AARC-50 1) Emotional well-being: CES-D Medium-
Wahl, from the United (4.61 years) N= 423; 62.94 (Radloff, 1977). high
2018) States and 4.61 (11.84), 40- 2) Physical well-being: Sf-36-
Germany aged 40- years 98 German version (Bullinger, 1995).
98 (Diehl and Wahl follow-up
as principal N= 299
investigators)
(Dutt et Cohort of adults Longitudinal Baseline Baseline 64 AARC-50 1) Emotional well-being: CES-D Medium-
al., 2018) from the United (4.61 years) N= 423; 62.94 losses (Radloff, 1977). High
States and 2.5 years (11.84), 40- subscale 2) Physical well-being: SF-36-
Germany aged 40- follow-up 98 German version (Bullinger, 1995).
98 (Diehl and Wahl N= 356;
as principal 4.61
investigators) years
follow-up
N= 299

63
Chapter 3: Associations of awareness of age-related change with emotional and physical well-being: a systematic review and meta-

analysis

(English Mindfulness and Cross- N= 296 64.67 50 AARC-20 1) Physical well-being: Chronic Medium
et al., Anticipatory Coping sectional (4.36), condition checklist (Ryff & Keyes,
2019) Everyday study 1995).
(United States)
(Miche et BEWOHNT study Cross- N= 225 70-89 Not stated Diary based 1) Physical well-being: SF-LLFDI7 Medium
al., 2014) (Germany) sectional reports of (Denkinger et al., 2008).
analysis (14 SAEs in the 2) PNAS (Watson et al., 1988).
days) five domains
of behaviour
and
functioning
suggested by
Diehl and
Wahl (2010).
Note: AARC= Awareness of age-related change; CASP= Critical Appraisal Skills Programme; SPWB-SF= The Scales of
Psychological Well-being; SF-36V2= The Short Form 36 Health and Well-being Questionnaire, Version 2; SWLS= The Satisfaction
with Life Scale; CES-D= The Center for Epidemiologic Studies Depression Scale; CES-D-R10= The Center for Epidemiologic
Studies Depression Scale 10-item Short form; PNAS= Positive and Negative Affect Scale; SF-LLFDI= The Function Component of
the abbreviated Late Life Function and Disability Instrument.

64
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

The association between AARC and emotional well-being


In this review we considered the term emotional well-being as an overarching
concept (Lara et al., 2013) including concepts such as positive and negative affect
(Bradburn, 1969), psychological well-being (Ryff & Keyes, 1995), life satisfaction,
quality of life (Andrews & McKennell, 1980; Bryant & Veroff, 1982) and mental health
(Antaramian et al., 2010).
Nine articles (Brothers et al., 2016; Brothers et al., 2017; Dutt et al., 2016a;
Dutt & Wahl, 2018; Dutt et al., 2018; Kaspar et al., 2019; Miche et al., 2014; Neupert
& Bellingtier, 2017; Wahl et al., 2013) explored the association between AARC and
emotional well-being; five of these were included in the meta-analysis (Brothers et
al., 2016; Dutt et al., 2018; Kaspar et al., 2019; Neupert & Bellingtier, 2017; Wahl et
al., 2013). Among the five articles included in the meta-analysis, three (Brothers et
al., 2016; Dutt et al., 2018; Kaspar et al., 2019) were combined as they were
conducted with the same cohort but explored associations of AARC with different
measures of emotional well-being.
Below we describe first the association between AARC gains and emotional well-
being and then the association between AARC losses and emotional well-being.
Overall, AARC gains were associated with better emotional well-being (r= .33;
95% CI: .18, .47; p< .001; I2= 76.0%, see Figure 3.2). However, the strength of the
association varied depending on the measure of emotional well-being and the cohort
used in the study. For example, AARC gains were significantly associated with some
indicators of emotional well-being such as satisfaction with life (Brothers et al., 2017;
Wahl et al., 2013) and psychological well-being (Brothers et al., 2016) whereas the
results for other indicators such as depressive symptoms (Dutt et al., 2016a; Dutt &
Wahl, 2018; Dutt et al., 2018; Kaspar et al., 2019) were mixed. The study that found
a significant association between AARC gains and depressive symptoms involved
analyses of data from a sample of participants aged 80 and over (Kaspar et al.,
2019) whereas the studies (Dutt et al., 2016a; Dutt & Wahl, 2018; Dutt et al., 2018)
that did not find a significant association between AARC gains and depressive
symptoms were conducted with a sample of participants aged between 40 and 96
years. In addition, a study (Miche et al., 2014) that explored the concept of AARC

65
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

through the use of daily diaries found that positive ageing experiences in the
cognitive domain were associated with positive affect.

Figure 3.2 Forest plot of the associations between AARC gains and measures of
emotional well-being. Combined= Pearson’s r correlation coefficients of Brothers et
al. (2016); Dutt et al. (2018); Kaspar et al. (2019) were combined. Overall= Overall
effect size of the five studies included in the meta-analysis.

The funnel plot (Figure 3.3) suggests some indication of publication bias and
therefore the effect sizes of included studies may be slightly inflated. However,
Egger’s test indicates that the publication bias was not statistically significant (p=
.138).

66
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

Figure 3.3. Funnel plots for the associations between (a) AARC gains and emotional
well-being, (b) AARC losses and emotional well-being, (c) AARC gains and physical
well-being, (d) AARC losses and physical well-being.

All nine articles (Brothers et al., 2016; Brothers et al., 2017; Dutt et al., 2016a;
Dutt & Wahl, 2018; Dutt et al., 2018; Kaspar et al., 2019; Miche et al., 2014; Neupert
& Bellingtier, 2017; Wahl et al., 2013) that explored the association between AARC
losses and emotional well-being found an association between higher levels of
AARC losses and lower levels of emotional well-being. Of the nine articles exploring
the association between AARC losses and emotional well-being, five were included
in the meta-analysis (Brothers et al., 2016; Dutt et al., 2018; Kaspar et al., 2019;
Neupert & Bellingtier, 2017; Wahl et al., 2013). AARC losses were found to be
associated with lower emotional well-being (r= -.31; 95% CI: -.38, -.24; p= .001; I2=
0%, see Figure 3.4). Findings from a micro-longitudinal study exploring variability in
the association between AARC and emotional well-being indicated that individuals
who reported daily increases in AARC losses also experienced concomitant
increases in negative affect (Neupert & Bellingtier, 2017). There was no evidence of
heterogeneity, suggesting that the estimated effect size between AARC losses and

67
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

emotional well-being was reliable; there was also no indication of publication bias
(Figure 3.4) and Egger’s test was not statistically significant (p= .093).

Figure 3.4. Forest plot of the associations between AARC losses and measures of
emotional well-being. Combined= Pearson’s r correlation coefficients of Brothers et
al. (2016); Dutt et al. (2018); Kaspar et al. (2019) were combined. Overall= Overall
effect size of the five studies included in the meta-analysis.

The association between AARC and physical well-being


In this review we considered the term physical well-being as an overarching concept
including several constructs that provide information about the physical well-being of
participants such as self-rated health, presence of chronic illnesses, and ability to
carry out instrumental activities of daily living (Huber et al., 2016).
Eleven articles (Brothers et al., 2016; Brothers et al., 2019; Brothers et al.,
2017; Dutt et al., 2016a, 2016b; Dutt & Wahl, 2018; Dutt et al., 2018; English et al.,
2019; Kaspar et al., 2019; Miche et al., 2014; Wahl et al., 2013) explored the
association between AARC and physical well-being. Out of these eleven articles, five
were included in the meta-analysis (Brothers et al., 2016; Dutt et al., 2018; English et
al., 2019; Kaspar et al., 2019; Wahl et al., 2013). Among the five articles included in

68
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

the meta-analysis, three (Brothers et al., 2016; Dutt et al., 2018; Kaspar et al., 2019)
were combined as they were conducted with the same cohort but explored
associations of AARC with different measures of physical well-being. Below we
describe first the association between AARC gains and physical well-being, and then
the association between AARC losses and physical well-being. AARC gains were
not significantly associated with physical well-being (r= .08; 95% CI: .02, .14, p=
.122; I2= 0%, see Figure 3.5). Whereas the majority of the included studies did not
find an association between AARC gains and physical well-being (Brothers et al.,
2016; Brothers et al., 2017; Dutt et al., 2016a, 2016b; Dutt & Wahl, 2018; Dutt et al.,
2018; English et al., 2019; Kaspar et al., 2019; Wahl et al., 2013), one study did find
an association (Brothers et al., 2019).

Figure 3.5. Forest plot of the associations between AARC gains and measures of
physical well-being. Combined= Pearson’s r correlation coefficients of Brothers et al.
(2016); Dutt et al. (2018); Kaspar et al. (2019) were combined. Overall= Overall
effect size of the five studies included in the meta-analysis.

There was no evidence of heterogeneity, suggesting that the estimated effect


size between AARC gains and physical well-being was reliable. Funnel plots (Figure

69
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

3.3) suggested some evidence of publication bias; however, Egger’s test indicated
that this was not statistically significant (p= .280). All eleven articles found an
association between higher levels of AARC losses and poorer physical health. The
meta-analysis indicated that AARC losses were associated with lower physical well-
being (r= -.38; 95% CI: -.51, -.24; p< .001; I2= 83.5%, see Figure 3.6). There was a
large degree of heterogeneity, suggesting that the estimated effect size may not be
reliable. Funnel plots (Figure 3.3) indicated no publication bias and Egger’s test was
not statistically significant (p= .123).

Figure 3.6. Forest plot of the associations between AARC losses and measures of
physical well-being. Combined= Pearson’s r correlation coefficients of Brothers et al.
(2016); Dutt et al. (2018); Kaspar et al. (2019) were combined. Overall= Overall
effect size of the five studies included in the meta-analysis.

Association between AARC and cognitive functioning

70
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

We found no studies that explored the association between AARC and cognitive
functioning.

3.5 Discussion
This is the first systematic review and meta-analysis to explore the evidence on
AARC gains and losses, and their associations with emotional and physical well-
being and cognitive functioning, among cognitively healthy adults. Findings suggest
that overall there is a moderate association between higher levels of AARC gains
and better emotional well-being, but different indicators of emotional well-being may
have different associations with AARC gains. There is some evidence that higher
levels of AARC losses are associated with lower emotional well-being. Results for
AARC and physical well-being are equivocal. Whereas there is a negligible
association between AARC gains and physical well-being, there is a moderate
association between higher levels of AARC losses and poorer physical well-being.
No study explored the association between AARC and cognitive functioning.
The moderate associations between AARC gains and some indicators of
emotional well-being, such as satisfaction with life and psychological well-being,
suggest that higher levels of AARC gains may be important for a general positive
emotional state. The association between AARC gains and emotional well-being is in
line with the broader literature on subjective ageing (Bryant et al., 2012). Similarly,
the empirical literature on other concepts of subjective ageing indicates that there is
an association between more positive subjective ageing and better emotional well-
being, including lower depressive symptoms (Kavirajan et al., 2011; Stephan et al.,
2016a). However, our systematic review indicates that the results for AARC gains
and depressive symptoms are mixed, and this may be explained by the differing age
profiles of participants in the relevant studies. Our results suggest that AARC gains
may be a protective factor against depressive symptoms only for individuals aged 80
and over (Kaspar et al., 2019), and may not play a preventive role against
depressive symptoms among younger individuals (Dutt et al., 2016a; Dutt & Wahl,
2018; Dutt et al., 2018). As the experience of AARC losses increases with age (e.g.
Dutt et al., 2018) and AARC losses are associated with depressive symptoms (Dutt
et al., 2016a), it may be that the presence of AARC gains in those age groups that

71
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

are likely to experience high levels of AARC losses buffers the negative effect of
AARC losses, resulting in lower levels of depressive symptoms. In line with this
reasoning, a recent study found that the presence of AARC gains attenuates the
negative effect that AARC losses have on older adults’ ability to adjust to age-related
challenges (Wilton-Harding & Windsor, 2021). Hence, the use of measures, such as
the AARC questionnaire, that divide scores for gains and losses, are very important
when the aim is to understand how the presence of gains and losses interact to
influence health-related outcomes in middle and older age. However, it may also be
possible that while AARC plays a role in overall emotional well-being, other aspects
of subjective ageing such as subjective age and attitudes toward own ageing may
play a more important role in specific mental health conditions, including depressive
symptoms (Kwak et al., 2014; Stephan et al., 2016a).
The moderate association between AARC losses and emotional well-being
suggests that higher levels of AARC losses are associated with lower levels of
emotional well-being. These findings are also in line with the literature on subjective
ageing which indicates that negative subjective ageing predicts the onset and
persistence of depression (Freeman et al., 2016). High levels of AARC losses may
increase reflection on mortality (Greenberg et al., 1986; Wurm & Benyamini, 2014)
and awareness of limited time to live (Brothers et al., 2016) which may facilitate
depressive ideation. Empirical literature that explores the way in which individuals
perceive their future (Carstensen, 1993) suggests that the emotional well-being of
individuals with high levels of AARC losses may be improved by interventions
promoting ego-integrity and engagement in fulfilling relations (Westerhof et al.,
2015). Moreover, as the empirical literature shows how several constructs (e.g.
mindful attitude) impact on the association between AARC losses and mental health
outcomes (Dutt & Wahl, 2018; Dutt et al., 2018), positive emotional well-being could
be promoted through mindfulness-based interventions that teach individuals to
accept age-related changes and be compassionate towards themselves (Collins &
Kishita, 2018; Phillips & Ferguson, 2012; Xu, 2017).
Overall, the association between AARC gains and physical well-being was
negligible. This is not consistent with the broader literature on subjective ageing
which shows that more positive subjective ageing is associated with better physical

72
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

health (Bryant et al., 2012; Kavirajan et al., 2011). The inconsistency found between
AARC gains and other subjective ageing constructs with regard to physical well-
being may be due to AARC being the first concept that separates measurement of
gains and losses. This makes it possible to assess the coexistence of AARC gains
and losses, whereas other subjective ageing concepts are unidimensional. It may
therefore be that individuals with higher levels of AARC gains do not report better
physical health because of the coexistence of high levels of AARC losses. The
interplay between AARC gains and AARC losses may be an area of interest for
future research.
The negligible association between AARC gains and physical well-being may
suggest that when promoting good physical health in older individuals, enhancing
experiences of AARC gains may be less useful than targeting AARC losses.
However, this interpretation must remain cautious due to the small proportion of
individuals with low physical health included in the available studies exploring the
association between AARC and physical health. Moreover, associations between
AARC gains and physical well-being may emerge over longer time periods than
those observed in the included studies.
From the review, only one study found an association between AARC gains
and physical well-being (Brothers et al., 2019) and this may be due to the specific
sample of participants or to cultural differences. Indeed this study was conducted
with a sample of US residents while the remaining studies that investigated the
association between AARC gains and self-rated health were conducted either in
German samples (Dutt et al., 2016a, 2016b; Dutt & Wahl, 2018; Dutt et al., 2018;
Wahl et al., 2013) or in combined US and German samples (Brothers et al., 2016;
Brothers et al., 2017; Kaspar et al., 2019). The different finding of Brothers et al.
(2019) may be due to cultural differences, as US individuals have a more optimistic
response pattern on AARC questionnaires compared to German individuals
(Brothers et al., 2016). Country-specific differences exist also for other subjective
ageing concepts and their associations with health outcomes (e.g. Westerhof &
Barrett, 2005).
The moderate association between higher levels of AARC losses and lower
levels of self-rated health is also in line with the subjective ageing literature (Levy,

73
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

2003; Robertson et al., 2015) and this association may be due to lower engagement
in health-related behaviours by individuals with less positive subjective ageing (Levy
& Myers, 2004). Individuals with high levels of AARC losses may perceive their age-
related changes as physically limiting and as a consequence they may be more likely
to reduce engagement in certain health-related behaviours (e.g. physical activity,
social relations and hobbies) rather than adapting their behaviour according to age-
related changes (Dutt et al., 2016b). Individuals with high AARC losses may
therefore benefit from interventions helping them to cope with negative changes and
identify new goals to compensate for age-related changes (Baltes, 1997;
Brandtstädter & Rothermund, 2002).
Only two studies (English et al., 2019; Miche et al., 2014) included in the
systematic review reported significant associations between AARC and more
objective indicators of physical well-being, such as functional health and the
presence of chronic health conditions. In contrast, the remaining studies assessed
self-rated health as an indicator of physical well-being. The use of measures of self-
rated health as indicators of objective health has been questioned, as existing
empirical evidence shows that self-rated health remains stable despite declines in
physical and functional health (Jylha et al., 2001). Future research could further
explore whether the association between AARC losses and physical well-being
remains moderate when using more objective indicators of physical well-being.
Research on the association between AARC and objective indicators of physical
well-being, such as the presence of chronic conditions, may suggest ways of
promoting physical health in older age.
None of the studies included in the systematic review and meta-analysis
reported associations between AARC and indicators of cognitive functioning. As
subjective cognitive decline is considered to be an indicator of future cognitive
decline (Jessen et al., 2014), it may be that individuals’ AARC in relation to their
cognitive abilities reflects actual cognitive changes. Future research could therefore
investigate whether AARC gains and losses are associated with objective measures
of cognitive functioning. This would provide knowledge on whether individuals’ AARC
related to cognition is informative in relation to subtle cognitive changes, and hence
whether AARC could be used as an indicator of possible cognitive decline.

74
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

Lack of investigation of the association between AARC and cognitive


functioning was unexpected. Indeed, other subjective ageing concepts, such as
attitudes toward own ageing and age stereotypes, have been explored in relation to
cognitive functioning, and support the predictive role of positive subjective ageing in
relation to better cognitive functioning (Levy et al., 2002b; Opdebeeck et al., 2019;
Robertson et al., 2016; Robertson et al., 2015) and lower risk of Alzheimer’s-like
neuropathology (Levy et al., 2016). Studying whether AARC gains and losses predict
future cognitive abilities may suggest ways of promoting maintenance of cognitive
health in older people.
The interpretation of the results of this systematic review and meta-analysis
has to be considered in light of its limitations. For example, among studies included
in the systematic review, only two assessed physical well-being with objective
instruments. The presence of recall bias in subjective measures is well-documented,
making it difficult to interpret results related to subjective evaluations of physical
health and other domains (Althubaiti, 2016). However, the majority of studies used
valid and reliable measures of AARC and emotional well-being. Next, the
generalisability of study results is limited as the included studies explored the
associations of AARC with emotional well-being and/or physical well-being in four
cohorts including mainly white participants, with a majority of women, residing either
in the United States or in Germany, with above average education and self-reported
health. This highlights a gap in knowledge as levels of AARC are unknown for those
individuals who are less educated, have poor physical health status, and reside
outside United States or Germany, and whose ethnicity is not white.
In addition, only four studies explored the longitudinal associations of AARC
with emotional well-being and/or physical well-being. Longitudinal studies followed
participants for between 2.5 and 4.5 years. Even though 2.5 and 4.5 years may be
long enough to detect changes in levels of AARC (gains and losses) in older and
very old participants, associations of AARC with emotional well-being and physical
well-being in younger participants may need to be explored in studies with longer
follow-up periods. Among the subjective ageing literature, studies with longer follow-
ups found that positive subjective ageing predicts both better physical health and
emotional health (e.g. 23-year follow-up; Levy et al., 2002b; Stephan et al., 2011);

75
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

hence the association between AARC gains and physical health may need to be
explored with longer longitudinal studies.
A further limitation of the meta-analysis is the inclusion of only cross-sectional
associations of AARC with emotional and/or physical well-being, making it difficult to
establish the direction of associations. However, this was the most appropriate
approach as only two (Brothers et al., 2016; Dutt et al., 2018) of the six studies
included in the meta-analysis were longitudinal. Another limitation of the meta-
analysis lies in the between-study heterogeneity for the associations between AARC
gains and emotional well-being and between AARC losses and physical well-being.
This indicates that the studies included in the meta-analysis found different effects
for the association between AARC gains and emotional well-being and between
AARC losses and physical well-being; hence the estimated effects for these
associations should be interpreted with caution. Further research in this area could
help to determine the actual effect that physical and emotional well-being exerts on
AARC. The high heterogeneity observed in the meta-analysis for the associations
between AARC gains and emotional well-being and AARC losses and physical well-
being may be explained by the use of diverse measures to assess emotional and
physical well-being. For the association between AARC gains and emotional well-
being there was evidence of possible publication bias even though this was not
statistically significant. Therefore, it might be that the observed effects for the
association between AARC gains and emotional well-being are slightly inflated.
Among the key strengths of this systematic review and meta-analysis is the
comprehensive search that covered a wide number of databases including peer-
reviewed journal articles and literature. Searching grey literature allowed us to seek
out ongoing work on AARC that has not yet been published in peer-reviewed
journals and facilitated identification of one of the articles included in the review.
Searching grey literature was of particular importance as the concept of AARC is
relatively new (Diehl & Wahl, 2010) and interest in the topic of AARC is increasing.
During article screening, reviewers screened all abstracts relevant to subjective
ageing and all full-text versions of quantitative studies on AARC independently of the
outcomes mentioned in the abstract, thus increasing the likelihood of identifying all

76
Chapter 3: Associations of awareness of age-related change with emotional and

physical well-being: a systematic review and meta-analysis

relevant articles, including those assessing emotional and/or physical well-being as


covariates.

3.6 Conclusion
The moderate effect sizes we found for the associations of AARC gains and AARC
losses with emotional well-being, and of AARC losses with physical well-being,
indicate that AARC gains and losses can play a role in emotional well-being and that
AARC losses are associated with physical well-being. However, these results are
limited due to the specific characteristics of the participants included in the available
studies, indicating a need for future research to include more diverse samples
outside the United States and Germany. We identified several gaps in the literature:
the association of AARC gains and losses with cognitive functioning has not been
explored; there have been only a few longitudinal studies exploring the role of AARC
in relation to emotional and physical well-being, and these may not be of sufficient
length; there have been no trials of interventions based on AARC; and AARC has
mainly been explored in association with self-reported measures of well-being rather
than objective indicators. Further exploration of the AARC concept may be useful in
suggesting ways of promoting good emotional and physical well-being, and
preventing poor emotional and physical well-being, in older age.

77
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

Chapter 4: International relevance of two measures of awareness of age-


related change (AARC)

78
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

4.1 Summary
Chapter 3 found moderate associations for better mental and physical health with
lower levels of AARC losses and for better mental health with higher AARC gains
and a negligible association for better physical health with AARC gains. However,
available evidence was restricted to US and German participants. Even though
Chapter 3 also aimed to quantify the association of AARC with cognition; any study
exploring the association of AARC with cognition was identified. Hence, this thesis
focuses on the associations of AARC with cognitive, mental, and physical health
among UK individuals. As self-perceptions of ageing may vary across cultures (Voss
et al., 2018), Chapter 4 examines psychometric properties for two questionnaires
assessing AARC; one assessing global evaluations of AARC (AARC-10 SF; Kaspar
et al., 2019) and one assessing AARC specific to the cognitive domain (AARC-50
cognitive functioning subscale; Brothers et al., 2019) in the UK population aged 50
and over.
This paper has been published: Sabatini, S., Ukoumunne, O. C., Ballard, C.,
Brothers, A. F., Kaspar, R., Collins, R., Kim, S., Corbett, A., Aarsland, D.,
Hampshire, A., Brooker, H., & Clare, L. (2020). International relevance of two
measures of awareness of age-related change (AARC). BMC Geriatrics, 20(1), 359.
https://doi.org/10.1186/s12877-020-01767-6

Abstract
Background: A questionnaire assessing awareness of positive and negative age-
related changes (AARC gains and losses) was developed in the United States and
Germany. We validated the short form of the measure (AARC-10 SF) and the
cognitive functioning subscale from the 50-item version of the AARC (AARC-50
cognitive functioning subscale) questionnaire in the UK population aged 50 and over.
Method: Data from 9,410 participants (M (SD) age= 65.9 (7.1)) in the PROTECT
cohort were used to explore and confirm the psychometric properties of the AARC
measures including validity of the factor structure; reliability; measurement
invariance across men and women, individuals with and without a university degree,
and in middle-age, early old age, and advanced old age; and convergent validity with

79
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

measures of self-perception of ageing and mental, physical, and cognitive health.


We explored the relationship between sociodemographic variables
(age, sex, marital status, employment status, and university education) and AARC.
Results: We confirmed the two-factor structure (AARC gains and AARC losses) of
the AARC-10 SF and the AARC-50 cognitive functioning subscale. Both scales
showed good reliability and good convergent validity for AARC losses, but weak
convergent validity for AARC gains. For both scales metric invariance was held for
the two subgroups defined by education level and age. For the AARC-50 cognitive
functioning subscale, but not for the AARC-10 SF, strong invariance was also held
for the two subgroups defined by sex. Age, sex, marital status, employment status,
and university education predicted AARC gains and losses.
Conclusion: The AARC-10 SF and AARC-50 cognitive functioning subscale identify
UK individuals who perceive age-related changes in their mental, physical, and
cognitive health.

80
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

4.2 Introduction
Awareness of age-related change (AARC) is a useful concept that predicts a variety
of health-related outcomes such as depression and psychological and physical well-
being (Kaspar et al., 2019; Sabatini et al., 2020a) and could be used to motivate
engagement in healthy behaviours such as physical activity (Brothers & Diehl, 2017;
Diehl et al., 2014). AARC refers to “a person’s state of awareness that his or her
behaviour, level of performance, or way of experiencing life has changed as a
consequence of having grown older” (Diehl & Wahl, 2010, p. 342). AARC reflects the
observation that individuals’ experiences of ageing may vary across five life and
behavioural domains including health and physical functioning, cognitive functioning,
interpersonal relationships, socio-cognitive and socio-emotional functioning, and
lifestyle and life engagement. As the association between cognitive complaints and
cognitive performance is well-reported in the empirical literature (e.g. Amariglio et al.,
2018; Burmester et al., 2016), among the five AARC behavioural domains, the
cognitive functioning domain is potentially useful for detecting early stages of
cognitive decline. AARC captures awareness of both positive (AARC gains) and
negative (AARC losses) age-related changes and acknowledges that AARC gains
and losses can coexist, even in the same behavioural domain (Diehl & Wahl, 2010).
A questionnaire assessing AARC exists in three published versions of
differing length and across two languages (English and German). The 50-item
version (Brothers et al., 2019) and short 10-item version (AARC-10 SF; Kaspar et al.,
2019) are available in English. In the full 50-item version, out of the 50 items, half
represent perceived gains and half perceived losses. There are five gain- and loss-
related items representing each of the five theorised domains. The AARC-50
questionnaire has been shown to have good reliability in a sample of US residents
aged between 42 and 98 years old (Brothers et al., 2019), with Cronbach’s alpha (a)
coefficients ranging from .73 to .89 across all ten subscales.
The 10-item version of the AARC questionnaire is made up of selected items
from the AARC 50-item version. Reliability of the AARC-10 SF is good among US
and German residents aged 40 and over with Cronbach’s a coefficients ranging from
.49 to .75 across subscales (Kaspar et al., 2019). A 20-item modified version of the
AARC questionnaire adapted for daily use also exists (Neupert & Bellingtier, 2017).

81
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

The 20 items have been selected from the 50-item version of the AARC
questionnaire. However, each item stem, instead of asking participants to reflect on
their increasing age (“With my increasing age…”), invites participants to reflect on
their awareness of ageing in that specific day (“With my awareness of ageing
today…”). Psychometric properties of the AARC 20-item version have never been
explored.
AARC may be associated with cognitive functioning (e.g. Jessen et al., 2014).
As the AARC-10 SF includes only two items assessing AARC gains and AARC
losses respectively in the cognitive domain, the full 10-item subscale assessing
AARC gains taken from the AARC 50-item version of the questionnaire makes it
possible to more accurately explore the potential associations of AARC in the
cognitive domain with other indicators of cognitive functioning. The AARC-10 SF and
the AARC-50 cognitive functioning subscale may be particularly important when
thinking about new ways of preventing poor mental and physical health and cognitive
decline.
The AARC-10 SF (Kaspar et al., 2019) and the AARC-50 cognitive functioning
subscale (Brothers et al., 2019) are suitable to be used in long surveys or as
screening tools to identify those people at greater risk of poor mental and physical
health and/or cognitive decline (Jessen et al., 2014; Kaspar et al., 2019). In order to
use these measures in the United Kingdom, due to potential cross-cultural
differences in AARC, it is important to first explore their psychometric properties in
the UK population aged 50 and over (Voss et al., 2018). German participants, for
example, report fewer AARC gains, assessed with the AARC-10 SF, than US
participants (Brothers et al., 2016). Studying individuals aged 50 years and above is
considered appropriate as people in this age-group are old enough to be likely to
experience AARC. Individuals aged 50 years have previously reported experiencing
many age-related changes (e.g. Raz & Rodrigue, 2006; Shin et al., 2003; Siedlecki
et al., 2005) and shown concern about their physical health (Centre for Ageing
Better, 2015).
Amongst psychometric properties, it is important to test measurement
invariance to explore whether the AARC concept is interpreted consistently in the
same way across different population groups (e.g. defined by sex, level of education,

82
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

or age) (Brothers et al., 2016; English et al., 2019). Estimated reliable comparisons
of AARC scores among groups can, therefore, be potentially calculated (Carp &
Carp, 1983; Gregorich, 2006; Horn & McArdle, 1992; Mackinnon et al., 1999).
Regarding the AARC questionnaires, measurement invariance has so far been
tested only for the AARC-10 SF in relation to different age groups (Kaspar et al.,
2019).
Other measures of subjective ageing such as subjective age, which reflects
how old individuals feel they are (Barrett, 2003; Kotter-Grühn & Hess, 2012), and
attitudes toward own ageing (ATOA), which capture individuals’ evaluations of the
changes taking place in their lives as they age (Lawton, 1975), are suitable
measures to capture the way in which individuals experience ageing, albeit in a more
holistic manner compared to AARC (Diehl et al., 2014). These constructs, therefore,
were used as part of the exploration of convergent validity of existing AARC
questionnaires.
Moreover, as the AARC-10 SF covers awareness of changes in several
behaviours and life domains including socio-emotional, physical, and cognitive
functioning, investigating the associations of the AARC-10 SF with indicators of
mental, physical, and cognitive health provides information about the construct
validity of the AARC-10 SF. As part of the US and German validation of the AARC-
10 SF it has been found that AARC is associated with indicators of mental and
physical health including psychological well-being, satisfaction with life, depressive
symptoms, and functional and self-rated health (Kaspar et al., 2019). However,
despite age playing a role in levels of AARC (Brothers et al., 2016; Miche et al.,
2014), construct validity of the AARC-10 SF has not been explored in individuals
younger than 70 years. Moreover, research shows that higher levels of AARC losses
(measured with a 20-item version of the AARC questionnaire) are associated with
more negative affect (Miche et al., 2014; Neupert & Bellingtier, 2017) which is a key
component of anxiety. As common difficulties among older individuals, such as
poverty and diminished life expectations, are risk factors for anxiety (Butcher &
McGonigal-Kenney, 2005), the association between anxiety and AARC should also
be considered when exploring convergent validity of the AARC-10 SF. Finally,
construct validity of the AARC-50 cognitive functioning subscale (Brothers et al.,

83
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

2019) in relation to objective or subjective cognitive assessments has never been


explored.
Existing research in the United States and Germany, conducted by using the
50-item and 20-item versions of the AARC questionnaire, suggests that on average
individuals who are older, less well-educated, and/or woman have higher levels of
both AARC gains and AARC losses (Brothers et al., 2016; English et al., 2019).
These findings are in line with research conducted on other constructs that, similarly
to AARC, also capture individuals’ self-perceptions of ageing (Barrett, 2003; Kornadt
et al., 2019a; Steverink et al., 2001). Individuals who report a higher socioeconomic
status tend to experience more AARC gains and fewer AARC losses (assessed with
the 20-item version of the AARC questionnaire) than those with a lower
socioeconomic status (English et al., 2019). However, the role of sociodemographic
variables in the UK population is unexplored. Moreover, other sociodemographic
variables such as marital status and employment status have never been explored in
relation to AARC gains and losses even though existing literature suggests that they
may influence individuals’ perceptions of ageing (Barrett, 2005).
This study aims to: (a) confirm the two-factor structure (one factor for each of
gains and losses) and internal consistency of the AARC-10 SF (Kaspar et al., 2019)
and the AARC-50 cognitive functioning subscale (Brothers et al., 2019); (b) explore
measurement invariance for the AARC-10 SF and for the AARC-50 cognitive
functioning subscale among subgroups defined by sex, education level, and age; (c)
explore construct validity of the AARC-10 SF and the AARC-50 cognitive functioning
subscale by quantifying the associations of the AARC-10 SF with assessments of
subjective ageing, physical, mental, and cognitive health and of the AARC-50
cognitive functioning subscale with assessments of subjective ageing and cognitive
health; and (d) explore whether sociodemographic variables predict scores on the
AARC-10 SF and AARC-50 cognitive subscale gains and losses.

4.3 Methods
Study design and participants
The study was based on analyses of cross-sectional data collected through the
ongoing PROTECT (https://www.protectstudy.org.uk) study in 2019. PROTECT is a

84
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

25-year longitudinal study launched in 2015 that assesses participants every year on
measures of physical, mental, and cognitive health, lifestyle, and perceptions of
ageing through an online platform. Individuals are eligible to participate in the
PROTECT study if they are UK residents, English speakers, aged 50 years or over,
have access to a computer and internet, and do not have a clinical diagnosis of
dementia at the point of recruitment. Participants were recruited through national
publicity and via existing cohorts of older adults. Potential participants enrolled
through the PROTECT study website, downloaded the study information sheet, and
provided consent online. The PROTECT study has ethical approval from the London
Bridge NHS Research Ethics Committee and Health Research Authority (Ref:
13/LO/1578). Ethical approval for the data analyses was sought through the ethics
committee at the University of Exeter, School of Psychology (Application ID:
eCLESPsy000603 v1.0).
Between 1st January 2019 and 31st March 2019, 14,797 participants took part
in the PROTECT annual assessment. Among these, 9,410 participants completed
the AARC questionnaires and were therefore included in the present study
Instruments
Measures assessing subjective age, ATOA, mental and physical health, and
objective cognitive functioning were used to explore construct validity for the AARC-
10 SF. Measures assessing subjective age, ATOA, and objective, self-reported, and
informant-reported assessments of cognitive functioning were used to explore
construct validity for the AARC-50 cognitive functioning subscale. Demographic
variables (age, sex, marital status, employment status, and university education)
were assessed to explore their relationships with levels of AARC gains and losses
assessed both with the AARC-10 SF and with the AARC-50 cognitive functioning
subscale. Items included in the questionnaires were presented in the same order for
each person.

Demographic variables
Participants provided demographic information through the PROTECT platform at
baseline through an online assessment adapted from Office of National Statistics
measures, which included data on age, sex, ethnicity, marital status, employment

85
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

status, and university education. Ethnicity included the following categories: white,
mixed (included white and black Caribbean, white and black African, white and
Asian, any other mixed multiple ethnic background), Asian, black, or other ethnic
groups. Marital status was used as a dichotomous variable (individuals who were
married, in a civil partnership, or co-habiting were grouped together versus
individuals who were unmarried, divorced, separated, or widowed). Employment
status was used as a dichotomous variable (employed versus not employed).
University education was used as a dichotomous variable (university education
versus no university education). Individuals without a university education were
those participants that had completed secondary education (GCSE/O levels) or post-
secondary education (college, A-levels, NVQ3, or below). Individuals with a
university education were those participants that had completed vocational
qualifications (diploma, certificate, BTEC, NVQ4, and above), undergraduate
degrees (e.g. BA, BSc), post-graduate degrees (e.g. MA, MSc), or doctorates (PhD).

Awareness of age-related change (AARC)


The AARC-10 SF (Kaspar et al., 2019) is a brief tool for capturing perceived age-
related gains (AARC gains) and losses (AARC losses). It contains ten items, five
assessing AARC gains and five assessing AARC losses. Each of these five items
assesses a different AARC behavioural domain (health and physical functioning,
cognitive functioning, interpersonal relationships, socio-cognitive and socio-
emotional functioning, and lifestyle and life engagement). All ten items start with the
same stem “With my increasing age, I realise that…”. An example of an item
capturing AARC gains is “…I appreciate relationships and people much more”, while
an example of an item capturing AARC losses is “…I have less energy”.
Respondents rate how much each item applies to them on a five-point Likert scale
(1= not at all, 2= a little bit, 3= moderately, 4= quite a bit, and 5= very much). Scores
can be obtained for the AARC gains and AARC losses subscales by summing items
that fall into the respective scales. Scales scores range from a minimum of five to a
maximum of 25 with higher scores indicating higher levels of awareness of age-
related change.

86
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

The cognitive functioning subscale of the AARC-50 questionnaire (AARC-50


cognitive functioning subscale; Brothers et al., 2019) includes ten items, five
assessing AARC gains and five assessing AARC losses. An example item capturing
AARC gains in the cognitive domain is “With my increasing age, I realise that I have
become wiser”, while an item capturing losses is “With my increasing age, I realise
that I am more forgetful”. Respondents rate how much each item applies to them on
a five-point Likert scale (1= not at all, 2= a little bit, 3= moderately, 4= quite a bit, and
5= very much). Scores on the AARC- cognitive functioning gains and AARC-
cognitive functioning losses subscales are obtained by summing items that fall into
the respective subscales. Subscales scores range from a minimum of five to a
maximum of 25 and higher scores indicate higher levels of awareness of age-related
change in the cognitive domain.

Attitudes toward own ageing (ATOA)


The ATOA scale is a valid and reliable five-item scale assessing participants’
attitudes toward their own ageing taken from the Philadelphia Geriatric Center
Morale Scale (Lawton, 1975). For each statement respondents are asked to make
temporal comparisons about changes in energy level, perceived usefulness,
happiness, and quality of life and to respond on a binary response set (better versus
worse, yes versus no). An example item is “Things keep getting worse as I get
older”. A proportion-based score can be obtained by summing the participant’s item
scores and by dividing it by the number of responses, with a score of one indicating
that positive attitudes are implied in all answers and a score of zero indicating that a
negative response is implied in all answers.

Subjective age
Subjective age was assessed with a single-item question (adapted from the National
Survey of Midlife development in the United States; Barrett, 2003) asking
participants to write the age (in years) that they feel most of the time. A proportional
discrepancy score was calculated by subtracting the participants’ subjective age
from their chronological age, and by dividing this difference score by participants’

87
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

chronological age. A positive value indicates a youthful subjective age, whereas a


negative value indicates an older subjective age.

Cognitive functioning – Objective assessment


Cognitive functioning was measured with the PROTECT Cognitive Test Battery
(Corbett et al., 2015; Hampshire et al., 2012; Huntley et al., 2018) which includes
four tests: (1) the Grammatical Reasoning task assesses verbal reasoning
(Baddeley, 1968); (2) the Digit Span task (Huntley et al., 2017) assesses verbal
working memory; (3) the Self-ordered Search task measures spatial working memory
(Owen et al., 1990); and (4) the Paired Associate Learning task (Owen et al., 1993)
assesses visual episodic memory. For each task a summary score can be obtained
by subtracting the number of errors from the number of correct answers. Hence for
each task a higher score indicates a better performance. For digit span the summary
score can range from 0 to 20. For paired associate learning the summary score can
range from 0 to 16. For grammatical reasoning the summary score is also obtained
by subtracting the number of errors from the number of correct answers, but the
score has no set upper or lower limit as the participants can attempt as many trials
as they can manage within a specific timeframe. Finally, the summary score for the
self-ordered search task can range from 0 to 20.

Cognitive functioning - Informant rating and self-rating


The Informant Questionnaire on Cognitive Decline in the Elderly short form
(IQCODE-Informant; Jorm, 1994; Jorm & Jacomb, 1989) was administered to an
informant close to the participant. The IQCODE is a valid and reliable 16-item
questionnaire that asks respondents to rate the cognitive change of someone close
to them over the last ten years. Items describe both cognitive improvement and
cognitive decline (an example item is “Remembering things that have happened
recently”) and can be answered on a five-point scale (1= much improved, 2= a bit
improved, 3= not much change, 4= a bit worse, and 5= much worse). The final score
is the mean of the item scores. A parallel version of the IQCODE was administered
to the participant (IQCODE-Self; Jorm & Jacomb, 1989).

88
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

Mental health
The Patient Health Questionnaire-9 (PHQ-9; Kroenke et al., 2001) is a valid and
reliable nine-item scale capturing depressive symptoms over the previous two
weeks. It is based directly on the diagnostic criteria for major depressive disorder
described in the Diagnostic and Statistical Manual Fourth Edition (DSM IV; American
Psychiatric Association, 2000). Respondents are asked to indicate how frequently
they experience each symptom on a four-point Likert scale (1= not at all, 2= several
days, 3= more than half the days, and 4= nearly every day). The total score is the
sum of the item scores and can range from 9 to 36.
The Composite International Diagnostic Interview-Short Form (CIDI-SF;
Kessler et al., 1998) is a reliable and valid measure for assessing lifetime symptoms
of depression and anxiety. Nine items assess depressive symptoms, and eight items
assess anxiety symptoms. An example of a depressive symptom question is “Did
you lose interest in most things?”. For each item, participants can answer “yes” if
they have the symptom or “no” if they do not have the symptom. For both depressive
and anxiety symptoms a total score can be calculated by summing the items where
the participants answer yes. For depressive and anxiety symptoms the total score
can range from zero to nine and from zero to eight, respectively.
The Generalised Anxiety Disorder-7 (GAD-7; Spitzer et al., 2006) is a valid
and reliable seven-item measure assessing symptoms of generalised anxiety
disorder. Respondents are asked to indicate the frequency of occurrence of a list of
symptoms over the past two weeks on a four-point scale (1= not at all, 2= several
days, 3= more than half the days, and 4= nearly every day). The overall score is the
sum of the item scores and ranges from 7 to 28.

Instrumental activities of daily living


To measure individuals’ everyday functional ability at the present time a modified
version of the Lawton’s Instrumental Activities of Daily Living Scale was used
(Lawton & Brody, 1969). It assesses seven activities including preparing meals,
managing medications, and using the telephone. Two of the original eight items
(assessing laundry and housekeeping) were combined. For each activity
respondents rate how difficult they find doing the activity. For each activity

89
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

respondents have to rate how difficult they find performing the activity (0= no
difficulty, 1= some difficulty, and 2= great difficulty). The total score ranges from a
possible 0 to 14.

Self-rated health
We assessed self-rated health with a single-item question (taken from the SF-36;
Ware & Sherbourne, 1992) asking participants to rate their own health on a four-
point scale ranging from excellent to poor (1= excellent, 2= good, 3= fair, and 4=
poor).

Analyses
As the validation of the AARC-10 SF (Kaspar et al., 2019) in US and German
samples supported a two-factor structure (one factor for each of AARC gains and
AARC losses), we used confirmatory factor analysis (CFA) to confirm this structure
in the UK population. We tested whether the five items assessing gains and the five
items assessing losses (of the AARC-10 SF) are related to the respective
hypothesised underlying factors of AARC gains and AARC losses. The two factors
AARC gains and AARC losses were allowed to correlate in the CFA model. Error
terms were allowed to correlate for the pair of gains and losses items for the same
AARC behavioural domain (Figure 4.1a).
CFA was also conducted to confirm the two-factor structure of the AARC-50
cognitive functioning subscale (Brothers et al., 2019) (Figure 4.1b). For both the
AARC-10 SF and the AARC-50 cognitive functioning subscale, to confirm the need
for a two-factor model (above described), we also fitted a model in which a single
factor loaded on all ten items. For both the AARC-10 SF and the AARC-50 cognitive
functioning subscale, we compared goodness of fit indices (GoF) of the two-factor
model with those of one-factor model. Because the Chi-squared statistic is often
significant for well-fitting models in large samples (Bollen, 2014) alternative GoF
measures including the Comparative Fit index (CFI), the Tucker-Lewis index (TLI),
the Root Mean Square Error of Approximation (RMSEA), and the Standardised Root
Mean Square Residual (SRMR) were examined. Criteria for acceptable model fit
were CFI and TLI > .90, RMSEA < .08 (95% CI: between 0 and .08), and SRMR <

90
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

.06 (Byrne, 2012). The CFA models were fitted using the sem command in Stata.
Analyses included only participants that provided complete data on all items.
We used Cronbach’s alpha (α) to quantify reliability for the gains and losses
subscales of the AARC-10 SF and the AARC-50 cognitive functioning subscale
(Cronbach, 1951). We considered α values between .65 and .95 to be satisfactory
(Tavakol & Dennick, 2011).
For both the AARC-10 SF and the AARC-50 cognitive functioning subscale,
we used CFA to test measurement invariance (Acock, 2013; Gregorich, 2006;
Meredith, 1993) between men and women, between two groups characterised by
university education (vocational qualification, undergraduate degree, post-graduate
degree, or doctorate) and no university education (secondary or post-secondary
education, and among three age groups (middle-age= 50 to 65 years; early old age=
66 to 75 years, advanced old age ≥ 76 years). To explore measurement invariance,
we fitted three CFA models: (a) Model 1 placed no equality constraints across
groups on factor loadings, item intercepts, the error variances, the variances of the
latent variables, or the covariances of the latent variables (assumes configural
invariance); (b) Model 2 constrained the factor loadings to be identical across
subgroups (assumes metric invariance); (c) Model 3 constrained the factor loadings
and item intercepts to be identical across subgroups (assumes strong invariance).
In the analyses individuals who completed a vocational qualification (e.g.
diploma or certificate) were considered to have the same level of education as those
who completed a undergraduate degree, a master’s degree, or a doctorate.
However, several types of vocational qualifications exist, with some vocational
qualifications being comparable to a university level education whereas others are
not. We therefore conducted sensitivity analyses where we grouped education level
in an alternative way. Individuals without a university education comprised those
participants that had completed secondary education (GCSE/O levels), post-
secondary education (college, A-levels, NVQ3, or below), or vocational qualifications
(diploma, certificate, BTEC, NVQ4, and above). Individuals with a university
education comprised those participants that had completed undergraduate degrees
(e.g. BA, BSc), post-graduate degrees (e.g. MA, MSc), or doctorates (PhD).

91
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

To evaluate the fit of a model compared to a less restrictive one, the


traditional approach involves assessing the differences in the Chi-squared fit
statistics of the two examined CFA models by conducting likelihood ratio tests (LRT).
However, as LRTs often result in statistically significant differences in large samples
for models that are not markedly different in fit (Bollen, 2014) and alternative fit
indices are less sensitive to sample size (Cheung & Rensvold, 2002), we explored
model differences using alternative GoF indices including the Comparative Fit index
(CFI), the Root Mean Square Error of Approximation (RMSEA), and the
Standardised Root Mean Square Residual (SRMR). We concluded that a model had
a worse fit than a less constrained model when the difference in CFI (ΔCFI) was
larger than -.01 (Bentler, 1990; De Roover et al., 2014), the difference in RMSEA
(ΔRMSEA) was larger than .015 (Chen, 2007), and the difference in SRMR
(ΔSRMR) was larger than .03 (Chen, 2007).
Construct validity for the AARC-10 SF was explored by estimating correlations
between the AARC-10 SF and each of subjective age, ATOA, measures of mental
and physical health, and objective assessments of cognitive functioning
(grammatical reasoning, digit span, self-ordered search, and paired associate
learning). Construct validity for the AARC-50 cognitive functioning subscale was
explored by estimating correlations between the AARC-50 cognitive functioning
subscale and each of subjective age, ATOA, and objective (grammatical reasoning,
digit span, self-ordered search, and paired associate learning), self-reported, and
informant-reported assessments of cognitive functioning. We used Pearson’s r and
Spearman’s ρ to quantify correlations (Pearson, 1948). Correlation coefficients under
.10 were considered negligible, between .10 to .29 were considered small, between
.30 to .49 were considered moderate, and .50 or above were considered large
(Cohen, 1988).
To explore whether age, sex, marital status, employment status, and
university education explain variability in levels of AARC gains and/or AARC losses,
we fitted multiple linear regression models for each of the AARC-10 SF and the
AARC-50 cognitive functioning subscale gains and losses. We also conducted
simple regressions in which the predictive role of each demographic variable (age,
sex, marital status, employment status, and university education) on levels of AARC

92
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

gains and losses was explored without controlling for the predictive role of the
remaining demographic variables. All analyses were conducted in STATA version 16
(StataCorp, 2017).

3.4 Results
Descriptive statistics
Among the 9,410 study participants (mean (SD; range) age: 65.87 (7.05; 51,
95) years), only 0.4% reported having been diagnosed with mild cognitive
impairment. We estimated that a further 1.2% of participants had mild cognitive
impairment (as they scored 1.5 SDs below the mean study sample score in two or
more cognitive tasks). Those participants that we identified as having mild cognitive
impairment were kept in the analyses. However, participants with higher levels of
AARC losses on the AARC-10 SF and on the AARC-50 cognitive functioning
subscale had poorer scores on the four objective cognitive tasks, indicating that
participants were aware of their cognitive abilities (Table 4.1) and hence their
answers to the AARC-10 SF and the AARC-50 cognitive functioning subscale can be
deemed accurate. Scores on the cognitive tests were normally distributed (See
Figure 4.1a-d).

Figure 4.1a. Distribution of scores on the digit span task

93
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

Figure 4.1b. Distribution of scores on the paired associate learning task

Figure 4.1c. Distribution of scores on the grammatical reasoning task

94
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

Figure 4.1d. Distribution of scores on the self-ordered search task

Table 4.1. Means and standard deviations on the four objective cognitive tasks for
five levels of awareness of negative age-related cognitive change.
Level of negative N Digit span; Paired associate Grammatical Self-ordered
age-related M (SD) learning; M (SD) reasoning; M (SD) search; M (SD)
cognitive change
Not at all aware of 535 7.9 (1.9) 4.9 (1.0) 39.7 (10.3) 7.9 (2.7)
negative cognitive
change
A little aware of 5,331 7.7 (1.5) 4.8 (.9) 38.4 (10.2) 7.8 (2.6)
negative cognitive
change
Moderately aware 2,695 7.5 (1.5) 4.6 (.9) 36.2 (10.8) 7.5 (2.7)
of negative
cognitive change
Quite a bit aware of 672 7.3 (1.7) 4.5 (.9) 34.2 (10.9) 7.3 (2.7)
negative cognitive
change
Very much aware 177 6.9 (1.5) 4.5 (.8) 32.5 (10.0) 6.6. (3.0)
of negative
cognitive change
Note: N= number of participants. Not at all aware of negative cognitive changes=
Score between 0 and 5 on the AARC-50 cognitive functioning subscale. A little

95
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

aware of negative cognitive changes= Score between 6 and 10 on the AARC-50


cognitive functioning subscale. Moderately aware of negative cognitive changes=
Score between 11 and 15 on the AARC-50 cognitive functioning subscale. Quite a
bit aware of negative cognitive changes= Score between 16 and 20 on the AARC-50
cognitive functioning subscale. Very much aware of negative cognitive changes=
Score between 21 and 25 on the AARC-50 cognitive functioning subscale.
The majority of study participants was of white ethnicity (98.5% of
participants), married (79.1% of participants), completed a university education
(75.8% of participants) and was not retired (42.6% of participants). Demographic
characteristics for the study sample are reported in Table 4.2.

Table 4.2. Demographic characteristics of the study sample (N= 9,410).

Variables Statistics
Age (years), M (SD) 65.9 (7.1)
Range 51, 95
Sex (Women %) 79.9
Ethnicity (%)
White 98.5
Mixed .5
Asian .6
Black .1
Other ethnic groups .3
Marital status (%)
Married/ civil partnership/ co-habiting 79.1
Widowed/ separated/ divorced/ single 20.9
University education (Yes %) 75.8
Employment status (Yes %) 42.6

Means and standard deviations stratified by age, sex, and education level for
AARC gains and losses assessed both with the AARC-10 SF and the AARC-50
cognitive functioning subscale are reported in Tables 4.3a and 4.3b.

96
Chapter 4: International relevance of two measures of awareness of age-related change (AARC)

Table 4.3a. Levels of AARC gains and losses stratified by age (N= 9,410).

Age
Class 1: 50 to 65 Class 2: 66 to 75 Class 3: 76 and Class 1 vs 2 Class 1 vs 3 Class 2 vs 3 F statistic (df);
(N= 4,929) (N= 3,758) over (N= 723) p-value
Mean (SD); range Mean difference; (95% CI)

AARC-10 SF gains 18.1 (3.9); 5 - 25 17.8 (3.9); 5 - 25 17.4 (3.7); 6 - 25 -.3; (-.5, .1) -.7; (-1.0, -.3) -.4; (-.8, -.1) 12.1 (2); <.001
AARC-10 SF losses 9.4 (3.2); 5 - 25 10.2 (3.1); 5 - 25 12.1 (3.8); 5 - 25 .8; ( .6, .9) 2.7; (2.4, 3.0) 1.9; (1.7, 2.2) 243.4 (2); <.001
AARC-50 cognitive 14.3 (4.4); 5 - 25 13.6 (4.4); 5 - 25 13.3 (4.4); 5 - 25 -.7; (-.9, .5) -1.0; (-1.4, -.6) -.3; ( -.7, .1) 34.5 (2); <.001
functioning gains
AARC-50 cognitive 9.8 (3.7); 5 - 25 10.4 (3.5); 5 - 25 12.0 (4.1); 5 - 25 .6; (.4, .8) 2.2; (1.9, 2.5) 1.6; (1.2, 1.9) 123.0 (2); <.001
functioning losses

97
Chapter 4: International relevance of two measures of awareness of age-related change (AARC)

Table 4.3b. Levels of AARC gains and losses stratified by sex and educational level.
Sex
Women (N= 7,334) Men (N= 2,076)
M (SD); range M (SD); range t-statistics (df) p-value
AARC-10 SF gains 18.2 (3.8); 5 - 25 16.9 (4.0); 5 - 25 -14.2203 (9408) <.001
AARC-10 SF losses 9.7 (3.2); 5 - 25 10.5 (3.5); 5 - 25 9.2481 (9408) <.001
AARC-50 cognitive functioning gains 14.2 (4.4); 5 - 25 13.0 (4.4); 5 - 25 -11.6155 (9408) <.001
AARC-50 cognitive functioning losses 10.0 (3.6); 5 - 25 10.9 (3.9); 5 - 25 9.6811 (9408) <.001
Education level
No university education (N= 2,369) Completed university education (N= 7,041)
M (SD); range M (SD); range t-statistics (df) p-value
AARC-10 SF gains 18.1 (3.9); 5 - 25 17.9 (3.9); 5 - 25 2.5758 (9408) .006
AARC-10 SF losses 10.2 (3.5); 5 - 25 9.8 (3.2); 5 - 25 5.7887 (9408) <.001
AARC-50 cognitive functioning gains 14.4 (4.5); 5 - 25 13.8 (4.4); 5 - 25 5.3472 (9408) <.001
AARC-50 cognitive functioning losses 10.5 (3.8); 5 - 25 10.1 (3.6); 5 - 25 6.0893 (9408) <.001

98
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

A high proportion of participants perceived their health as good (54.1%) or


excellent (30.8%). On average participants did not report functional difficulties (IADL
M (SD) score= .16 (.77)). Participants had minimal levels of current depressive (M
(SD)= 11.5 (3.0)) and anxiety symptoms (M (SD)= 9.3 (8.5)), and low levels of both
lifetime depressive symptoms (M (SD)= 2.7 (3.3)) and lifetime anxiety symptoms (M
(SD)= 1.0 (2.1)). Compared to those who did not complete the AARC questionnaires
(N= 5,387), the study sample included a larger proportion of women (79.9% versus
71.3%) and participants who were better educated (75.8% versus 70.8%), and a
lower proportion of individuals who were employed (42.6% versus 54.7%).

Self-perceptions of ageing among the study sample


Distribution of scores on the AARC-10 SF and the AARC-50 cognitive functioning
subscale are reported in Figure 4.2a-d.

Figure 4.2a. Distribution of scores on the AARC-10 SF gains-total sample

99
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

Figure 4.2b. Distribution of scores on the AARC-10 SF losses-total sample

Figure 4.2c. Distribution of scores on the AARC-50 cognitive functioning subscale


gains-total sample

100
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

Figure 4.2d. Distribution of scores on the AARC-50 cognitive functioning subscale


losses-total sample

On the AARC-10 SF the majority of participants reported moderate (20.7%), quite a


bit of (47%), or a great deal of (27.8%) AARC gains and little (60.4%) or moderate
(29%) AARC losses. On the AARC-50 cognitive functioning subscale the majority of
participants reported moderate (39.6%) or quite a bit (28.0%) of AARC gains in their
cognition and little (56.6%) or moderate (28.6%) AARC losses in their cognition.
Further details about proportions of gains and losses perceived by participants on
the AARC-10 SF and the AARC-50 cognitive functioning subscale are reported in
Table 4.4. Participants felt 17% younger than their chronological age. Participants’
mean (SD) score on the ATOA scale was .52 (.16) indicating that participants
reported positive ATOA in some items but negative ATOA in others.

101
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

Table 4.4. Proportions of gains and losses reported by participants on the AARC-10
SF and the AARC-50 cognitive functioning subscale.
AARC-10 SF AARC-50 cognitive
functioning subscale
Gains Losses Gains Losses
Not being aware of age-related change .1% 3.8% 1.1% 5.7%
Little awareness of age-related change 4.4% 60.4% 23.4% 56.6%
Moderate awareness of age-related change 20.7% 29.0% 39.6% 28.6%
Quite a bit of awareness of age-related change 47.0% 6.0% 28.0% 7.4%
A great deal of awareness of age-related change 27.8% .8% 7.9% 1.9%

Psychometric properties of the AARC-10 SF and AARC-50 cognitive


functioning subscale
Confirmatory factor analysis
For the AARC-10 SF, compared to a one-factor model (RMSEA= .21; 95% CI: .00,
.00; CFI= .48; TLI= .33; SRMR= .18) the hypothesised two-factor model was a better
fit as indicated by GoF indices (RMSEA= .07; 95% CI: .07, .07; CFI= .94; TLI= .92;
SRMR= .05). Item characteristics for the ten items of the AARC-10 SF are displayed
in Table 4.5. The associations between factors and indicators were reasonably
strong for all items (Figure 4.3a). Factor loadings for the individual domain items on
the gains factor reflect greater heterogeneity of ageing experiences in the gains
compared to the losses factor.
For the AARC-50 cognitive functioning subscale, compared to a one-factor
model (RMSEA= .29; 95% CI: .00, .00; CFI= .42; TLI= .26; SRMR= .24) the
hypothesised two-factor model was a better fit as indicated by GoF indices
(RMSEA= .12; 95% CI: .12, .12; CFI= .90; TLI= .87; SRMR= .05). We calculated
modification indices for the AARC-50 two-factor model; however, modification
indices did not suggest any pattern that would have significantly improved the model.
Item characteristics for the 10 items of the AARC-50 cognitive functioning subscale
are displayed in Table 4.5 The associations between construct and indicators were
strong for all items (Figure 4.3b).

102
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

Figure 4.3a. Two-factor model of the AARC-10 SF. Note: Measurement model of
awareness of age-related change (AARC) for the AARC-10 SF. Fully standardised
coefficients are reported. AARC Domain abbreviations: PHY= Health and physical
functioning; COG= Cognitive functioning; INT= Interpersonal relations; SCSE=
Social-cognitive and social-emotional functioning; LIFE= Lifestyle and life
engagement; “+”= Positive domains; “-”= Negative domains.

Figure 4.3b. Two-factor model of the AARC-50 cognitive functioning subscale. Note:
Measurement model of awareness of age-related changes (AARC) for the AARC-50
cognitive functioning subscale. Fully standardised coefficients are reported. COG=
Cognition, “+”= Positive domains; “-”= Negative domains.

103
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

Reliability
For the AARC-10 SF item-to-total score correlations had values between .67 and
.78; hence all items reached satisfactory αs (Table 4.5). Cronbach’s α value was .77
for the AARC-10 SF gains scale and .80 for the AARC-10 SF losses scale. For the
AARC-50 cognitive functioning subscale all item-to-total score correlations reached
satisfactory values, ranging between .82 and .87 (Figure 4.3b). Cronbach’s α value
was .86 for the AARC-50 cognitive functioning subscale gains and .88 for the AARC-
50 cognitive functioning subscale losses.

104
Chapter 4: International relevance of two measures of awareness of age-related change (AARC)

Table 4.5 Item characteristics and Cronbach’s αs for the two subscales of the AARC-10 SF and the AARC-50 cognitive functioning
subscale.

AARC-10 SF domaina With my increasing age, I realise that… Item M (SD) Item-total correlation
PHYS- … I have less energy 2.8 1.1 .75
COG- …my mental capacity is declining 2.1 .9 .77
INT- …I feel more dependent on the help of others 1.5 .7 .76
SCSE- …I find it harder to motivate myself 1.7 .8 .78
LIFE- …I have to limit my activities 1.9 .9 .74
PHYS+ …I pay more attention to my health 3.1 1.1 .78
COG+ …I have more experience and knowledge to evaluate things and people 3.5 1.0 .71
INT+ …I appreciate relationship and people much more 3.8 1.1 .69
SCSE+ …I have a better sense of what is important for me 3.9 1.0 .67
LIFE+ …I have more freedom to live my days the way I want 3.7 1.2 .76
AARC-50 cognitive functioning With my increasing age, I realise that… Item M (SD) Item-total correlation
COG1 - …my mental capacity is declining 2.1 .9 .84
COG2 - ...I am slower in my thinking 1.8 .8 .84
COG3 - …I have a harder time concentrating 1.7 .8 .85
COG4 - …learning new things takes more time and effort 2.4 1.0 .87
COG5 - …I am more forgetful 2.2 1.0 .85
COG1 + …I have more experience and knowledge to evaluate things and people 3.5 1.0 .86
COG2 + …I have more foresight 2.7 1.1 .83
COG3 + …I have become wiser 2.6 1.2 .82
COG4 + …I think things through more carefully 2.6 1.1 .82
COG5 + …I gather more information before I make decisions 2.6 1.1 .85

105
Chapter 4: International relevance of two measures of awareness of age-related change (AARC)

Note: aAARC domain abbreviations: PHY= Health and physical functioning; COG = Cognitive functioning; INT = Interpersonal
relations; SCSE = Social-cognitive and social-emotional functioning; LIFE = Lifestyle and engagement; “+” = Positive domains; “-” =
Negative domains. M= mean. SD= Standard deviation.

106
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

Measurement invariance between sex groups for the AARC-10 SF and


the AARC-50 cognitive functioning subscale
With respect to measurement invariance for the AARC-10 SF between sex groups,
compared to the model with all parameters freely estimated (assuming configural
invariance), the model that restricted factor loadings to be the same across groups
(assuming metric invariance) did not substantially reduce the GoF (Table 4.6a).
Hence, the meaning of the concepts of AARC-10 SF gains and AARC-10 SF losses
appeared to be the same for men and women, allowing for valid representation of
AARC gains and losses in correlational studies across men and women. Restricting
item intercepts to be the same for men and women (assuming strong invariance)
substantially decreased model fit as indicated by GoF indices, meaning that men and
women interpret some items of the AARC-10 SF gains and the AARC-10 SF losses
subscales differently; hence scores on single items cannot be compared among men
and women. Since men and women systematically interpreted at least some items
differently, responses for men and women should not be compared without taking
this sex bias into account.
With respect to measurement invariance between sex groups for the AARC-
50 cognitive functioning subscale, compared to the model with all parameters freely
estimated in the men and women groups (assuming configural invariance),
restricting factor loadings to be the same across groups (assuming metric
invariance) did not substantially decrease model fit (Table 4.6a). Hence, the meaning
of the concept of AARC as captured by the AARC-50 cognitive functioning subscale
appeared to be the same for men and women. Restricting item intercepts and factor
loadings (assuming strong invariance) to be equal among men and women did not
substantially decrease GoF indices (RMSEA, CFI, and SRMR) (Table 4.6a),
meaning that men and women interpret items of the AARC-50 cognitive functioning
gains and losses subscales in the same way. Hence comparison of both observed
total scores across items and estimated factor means between sex groups is
possible.

107
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

Table 4.6a. Summary of the measurement invariance models for the AARC-10 SF
and the AARC-50 cognitive functioning subscale-sex groups.
AARC-10 SF
Models RMSEA; (95% CI) CFI SRMR
Model 1: Configural invariance .07; (.06, .07) .95 .05
Model 2: Metric invariance .06; (.06, .07) .95 .05
Model 3: Strong invariance .07; (.07, .08) .92 .71
AARC-50 cognitive functioning subscale
Models RMSEA (95% CI) CFI SRMR
Model 1: Configural invariance .12; (.12, .12) .90 .05
Model 2: Metric invariance .11; (.11, .12) .90 .05
Model 3: Strong invariance .11; (.11, .11) .90 .10
Note: RMSEA= Root mean square error of approximation. CFI= Comparative fit
index. SRMR= Standardised root mean square residual.

Measurement invariance between groups defined by education level for


the AARC-10 SF and the AARC-50 cognitive functioning subscale
With respect to measurement invariance for education level groups (university
education versus no university education) for the AARC-10 SF, compared to the
model with freely estimated parameters in the two groups (assuming configural
invariance), restricting factor loadings to be the same across groups (assuming
metric invariance) did not decrease model fit substantially (Table 4.6b). Hence, the
meaning of the concepts of AARC gains and AARC losses as captured by the
AARC-10 SF appeared to be the same for people with and without a university
education. Restricting item intercepts (assuming strong invariance) to be equal
among groups with a university education and without a university education did not
substantially decrease GoF indices (RMSEA, CFI, and SRMR) (Table 4.6b),
meaning that individuals with a university education interpret items of the AARC
gains and AARC losses subscales as captured by the AARC-10 SF similarly to their
counterparts without a university education. Hence for the AARC-10 SF comparison
of both observed total scores across items and estimated factor means between
education-based groups is possible.
With respect to measurement invariance for education level groups (university
education versus no university education) for the AARC-50 cognitive functioning

108
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

subscale, compared to the model that freely estimated parameters in the education
groups (assuming configural invariance), the model restricting factor loadings to be
the same (assuming metric invariance) across groups did not markedly decrease
model fit (Table 4.6b). Hence, the meaning of the concepts of AARC gains and
AARC losses in the cognitive domain as captured by the AARC-50 cognitive
functioning subscale appeared to be the same across people with a university
education and without a university education. Restricting item intercepts (strong
invariance) to be equal among participants with a university education and without a
university education did not substantially decrease GoF indices (RMSEA, CFI, and
SRMR) (Table 4.6b), meaning that people with a university education and people
without a university education interpret items of the AARC gains and AARC losses
subscales captured by the AARC-50 cognitive functioning subscale in the same way.
Hence for the AARC-50 cognitive functioning subscale comparison of both estimated
factor means and observed total scores across items between education-based
groups are possible.

Table 4.6b. Summary of the measurement invariance models for the AARC-10 SF
and the AARC-50 cognitive functioning subscale-educational level.

AARC-10 SF
Models RMSEA; (95% CI) CFI SRMR
Model 1: Configural invariance .07; (.07, .07) .95 .05
Model 2: Metric invariance .07; (.06, .07) .95 .05
Model 3: Strong invariance .06; (.06, .07) .94 .06
AARC-50 cognitive functioning subscale
Models RMSEA; (95% CI) CFI SRMR
Model 1: Configural invariance .12; (.12, .12) .91 .05
Model 2: Metric invariance .11; (.11, .12) .90 .99
Model 3: Strong invariance .11; (.11, .11) .90 .06
Note: RMSEA= Root mean square error of approximation. CFI= Comparative fit
index. SRMR= Standardised root mean square residual.

109
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

Sensitivity analyses show that grouping together individuals with a vocational


qualification with those without a university level degree, rather than with those with
a university level degree, does not change study results (see Table 4.6c).

Table 4.6c Summary of the measurement invariance models for the AARC-10 SF
and the AARC-50 cognitive functioning subscale - sensitivity analyses for
educational level

AARC-10 SF
Models RMSEA; (95% CI) CFI SRMR
Model 1: Configural invariance .07; (.07, .07) .95 .05
Model 2: Metric invariance .07; (.06, .07) .94 .06
Model 3: Strong invariance .07; (.06, .07) .94 .06
AARC-50 cognitive functioning subscale
Models RMSEA; (95% CI) CFI SRMR
Model 1: Configural invariance .12; (.12, .12) .91 .06
Model 2: Metric invariance .11; (.11, .12) .91 .06
Model 3: Strong invariance .11; (.11, .11) .90 .06
Note: RMSEA= Root mean square error of approximation. CFI= Comparative fit
index. SRMR= Standardised root mean square residual.

Measurement invariance among groups in middle-age, early old age,


and advanced old age for the AARC-10 SF and the AARC-50 cognitive
functioning subscale
With respect to measurement invariance among groups in middle-age, early old age,
and advanced old age (middle-age= 50 to 65 years; early old age= 66 to 75 years,
advanced old age ≥ 76 years) for the AARC-10 SF, compared to the model with
freely estimated parameters in the three age groups (assuming configural
invariance), restricting factor loadings to be the same across groups (assuming
metric invariance) did not decrease model fit substantially (Table 4.6d). Hence, the
meaning of the concepts of AARC gains and AARC losses as captured by the
AARC-10 SF appeared to be the same across middle-age, early old age, and
advanced old age. Restricting item intercepts (assuming strong invariance) to be
equal among age groups did not substantially decrease GoF indices (RMSEA, CFI,

110
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

and SRMR) (Table 4.6d), meaning that in middle-age, early old age, and advanced
old age individuals interpret items of the AARC gains and AARC losses subscales of
the AARC-10 SF similarly. Hence for the AARC-10 SF comparison of both observed
total scores across items and estimated factor means among age groups is possible.
With respect to measurement invariance among the three age groups for the
AARC-50 cognitive functioning subscale, compared to the model that freely
estimated parameters in the three age groups (assuming configural invariance), the
model restricting factor loadings to be the same (assuming metric invariance) across
the three age groups did not markedly decrease model fit (Table 4.6d). Hence, the
meaning of the concepts of AARC gains and AARC losses in the cognitive domain
as captured by the AARC-50 cognitive functioning subscale appeared to be the
same in middle-age, early old age, and advanced old age. Restricting item intercepts
(strong invariance) to be equal across the three age groups did not substantially
decrease GoF indices (RMSEA, CFI, and SRMR) (Table 4.6d), meaning that items
of the AARC gains and AARC losses subscales of the AARC-50 cognitive
functioning subscale are interpreted in the same way across middle-age, early old
age, and advanced old age. Hence for the AARC-50 cognitive functioning subscale
comparison of both estimated factor means and observed total scores across items
among age groups is possible.

Table 4.6d. Summary of the measurement invariance models for the AARC-10 SF
and the AARC-50 cognitive functioning subscale-age groups.

AARC-10 SF
Models RMSEA; (95% CI) CFI SRMR
Model 1: Configural invariance .07; (.07, .07) .94 .07
Model 2: Metric invariance .07; (.07, .07) .94 .07
Model 3: Strong invariance .08; (.08, .08) .90 .10
AARC-50 cognitive functioning subscale
Models RMSEA; (95% CI) CFI SRMR
Model 1: Configural invariance .12; (.12, .12) .91 .06
Model 2: Metric invariance .11; (.11, .11) .90 .06
Model 3: Strong invariance .11; (.11, .11) .89 .08
Note: RMSEA= Root mean square error of approximation. CFI= Comparative fit
index. SRMR= Standardised root mean square residual.

111
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

Validity of the AARC-10 SF in the over 50s UK population


Correlational evidence for validity of the AARC-10 SF is reported in Table 4.7a. As
expected, individuals who experience more AARC gains, assessed with the AARC-
10 SF, feel younger (r= .10; 95% CI: .08, .12) and have more positive ATOA (r= .12;
95% CI: .10, .14) compared to individuals who experience fewer AARC gains.
People who experience more AARC losses, assessed with the AARC-10 SF, feel
older (r= -.27; 95% CI: -.29, -.25) and have more negative ATOA (r= -.23; 95% CI: -
.25, -.21) compared to individuals who experience fewer AARC losses. Overall, we
found mixed and negligible correlations between AARC gains, assessed with the
AARC-10 SF, and indicators of mental and physical health. Individuals who
experience higher AARC losses, assessed with the AARC-10 SF, score higher on
measures assessing current symptoms of depression (r= .21; 95% CI: .19, .23) and
anxiety (r= .32; 95% CI: .30, .34), as well as lifetime symptoms of depression (r= .13;
95% CI: .12, .16) and anxiety (r= .16; 95% CI: .14, .18).
Participants with better functioning in activities of daily living and who rate
their health more positively experience higher levels of AARC gains, assessed with
the AARC-10 SF, but these correlations are negligible. Participants with better
functioning in activities of daily living (r= .23; 95% CI: .21, .24) and who rate their
health more positively (r= -.44; 95% CI: -.46, -.43) experience lower levels of AARC
losses, assessed with the AARC-10 SF, than participants with worse functional
health and who rate their health more negatively.
Correlations of the cognitive tasks digit span (r= -.01; 95% CI: -.04, .01) and
paired associate learning (r= -.01; 95% CI: -.04, .01) with AARC gains assessed with
the AARC-10 SF were not significant. The cognitive tasks grammatical reasoning (r=
-.04; 95% CI: -.07, -.02) and self-ordered search (r= -.05; 95% CI: -.08, -.03) showed
negative and negligible associations with AARC gains assessed with the AARC-10
SF. The cognitive tasks digit span (r= -.12; 95% CI: -.14, -.09), paired associate
learning (r= -.11; 95% CI: -.14, -.09), grammatical reasoning (r= -.15; 95% CI: -.18, -
.13), and self-ordered search (r= -.10; 95% CI: -.12, -.08) showed negative small
associations with AARC losses assessed with the AARC-10 SF.

112
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

Table 4.7a. Correlations between AARC-10 SF and measures of subjective ageing,


mental and physical health, and objective cognition.

Correlational evidence of validity of the AARC-10 SF


Variables AARC-10 SF Gains AARC-10 SF Losses
Pearson’s r (95% CI); p-value Pearson’s r (95% CI); p-value
Subjective age .10 ( .08, .12); <.001 -.27 (-.29, -.25); <.001
Lifetime depressive symptoms .07 ( .05, .09); <.001 .13 ( .12, .16); <.001
Lifetime anxiety symptoms .04 ( .02, .06); <.001 .16 ( .14, .18); <.001
Depressive symptoms -.08 (-.10, -.06); <.001 .32 ( .30, .34); <.001
Anxiety symptoms -.03 (-.05, -.01); .013 .21 ( .19, .23); <.001
Functional ability -.03 (-.05, -.01); <.001 .23 ( .21, .24); <.001
Self-rated health .09 ( .08, .12); <.001 -.44 (-.46, -.43); <.001
Digit span -.01 (-.04, .01); .338 -.12 (-.14, -.09); <.001
Paired associate learning -.01 (-.04, .01); .321 -.11 (-.14, -.09); <.001
Grammatical reasoning -.04 (-.07, -.02); .001 -.15 (-.18, -.13); <.001
Self-ordered search -.05 (-.08, -.03); <.001 -.10 (-.12, -.08); <.001
Spearman's ρ; p-value Spearman's ρ; p-value
Attitudes toward own ageing .13; <.001 -.25; <.001

Validity of the AARC-50 cognitive functioning subscale in the over 50s


UK population
Correlational evidence for validity of the AARC-50 cognitive functioning subscale is
reported in Table 4.7b. We found that those individuals who have higher awareness
of negative changes in their cognitive functioning also feel older and have more
negative ATOA than those individuals having lower AARC losses. Regarding the
correlations between the AARC-50 cognitive functioning subscale and objective
cognitive tasks, when compared to individuals with fewer AARC gains, individuals
with higher levels of awareness of positive changes score worse in tasks assessing
digit span, grammatical reasoning, and self-ordered search; estimate that their
cognitive abilities have increased over the past ten years; and a person close to
them also estimates that their abilities have increased over the past ten years.
However, most of the above-described correlations were of negligible size.
Compared to individuals with higher scores on cognitive tests, individuals with lower
scores on cognitive tests experience higher levels of negative age-related changes;

113
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

correlations among AARC losses and cognitive tasks were, however, either
negligible or small. Participants who report higher levels of negative age-related
changes in cognition also notice a decrease in their cognitive abilities over the past
ten years and this correlation was the strongest in size (r= .47; 95% CI: .45, .49).
However, participants’ awareness of negative age-related changes is not associated
with the judgment of an informant.
As the correlation between participants’ perceptions of negative age-related
changes in cognition and perceptions of a decrease in cognition over the past ten
years was the only correlation of moderate size, we further explored whether the size
of the correlation varies across different age-groups (middle-age= 50 to 65 years;
early old age= 66 to 75 years, advanced old age ≥ 76 years and over). We found that
the size of the correlation is similar among participants in middle-age (r= .48; 95%
CI: .45, .51); early old age (r= .47; 95% CI: .44, .50); and advanced old age (r= .42;
95% CI: .34, .91).

114
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

Table 4.7b. Correlations between AARC-50 cognitive functioning subscale and


measures of subjective ageing, subjective and objective cognition.

Correlational evidence of validity of the AARC-50 cognitive functioning subscale


Variables AARC-50 cognitive functioning AARC-50 cognitive functioning
gains losses
Pearson’s r (95% CI); p-value Pearson’s r (95% CI); p-value
Subjective age .08 ( .06, .10); <.001 -.19 (-.21, -.17); <.001
Digit span -.05 (-.08, -.03); <.001 -.10 (-.12, -.07); <.001
Paired associate learning -.02 (-.05, .00); .059 -.11 (-.14, -.09); <.001
Grammatical reasoning -.09 (-.12, -.07); <.001 -.16 (-.18, -.13); <.001
Self-ordered search -.07 (-.10, -.05); <.001 -.08 (-.11, -.06); <.001
IQCODE-Informant -.05 (-.07, -.03); <.001 -.01 (-.01, .03); .507
IQCODE-Self -.12 (-.15, -.10); <.001 .47 ( .45, .49); <.001
Spearman's ρ; p-value Spearman's ρ; p-value
Attitudes toward own ageing .04; <.001 -.14; <.001
Note: IQCODE-Informant= Informant Questionnaire on Cognitive Decline in the
Elderly short form asking informants to rate the cognitive change of someone close
to them over the last 10 years. IQCODE-Self= Informant Questionnaire on Cognitive
Decline in the Elderly short form asking participants to rate their own cognitive
change over the last 10 years.

115
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

Demographic variables as predictors of the AARC-10 SF and AARC-50


cognitive functioning subscale
From the two multiple regressions exploring the ability of demographic
variables to predict gains and losses measured on the AARC-10 SF (Tables 4.8a
and 4.8b), we found that, overall, being older, employed, and having a university
education significantly predict lower levels of AARC gains, whereas being a woman
significantly predicts higher levels of AARC gains. We also found that being a
woman, married, in a civil partnership, or co-habiting, and having a university
education significantly predict fewer AARC losses, whereas being older significantly
predicts more AARC losses.
From the multiple regressions exploring the ability of demographic variables to
predict gains and losses measured on the AARC-50 cognitive functioning subscale
(Tables 4.8c and 4.8d), we found that, overall, being older, married, in a civil
partnership, or co-habiting, and having a university education significantly predict
fewer AARC gains, whereas being a woman and employed significantly predict more
AARC gains. We also found that being a woman, employed, and having a university
education significantly predict fewer AARC losses, whereas being older significantly
predicts more AARC losses. Tables 4.8a to 4.8d also show the results of simple
regressions with each demographic variable (age, sex, marital status, employment
status, and university education) as a predictor of AARC gains and losses measured
with the AARC 10-SF and the AARC-50 cognitive functioning subscale.
Sensitivity analyses showed that grouping individuals with a vocational qualification
with those without a university level degree, rather than with those with a university
level degree, does not significantly change study results (see Table 4.9).

116
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

Table 4.8a. Simple and multiple regressions with demographic variables as


predictors of AARC gains scores on the AARC-10 SF.

(N= 8,639) Demographic variables as predictors of Demographic variables as predictors of


AARC gains: Simple regressions AARC gains: Multiple regression
AARC-10 SF gains
Variables B (95% CI); p-value ß B (95% CI); p-value ß
Age -.02 (-.04, -.01); <.001 -.05 -.03 (-.04, -.01); <.001 -.05
Sex 1.40 (1.21, 1.60); <.001 .15 1.31 (1.11, 1.51); <.001 .14
Marital status -.39 (-.59, -.19); <.001 -.04 -.27 (-.48, -.06); .010 -.03
Employment .04 (-.12, .21); .602 .01 -.23 (-.43, -.03); .020 -.03
status
University -.25 (-.44, -.06); .010 -.03 -.22 (-.41, -.03); .020 -.02
education
Total R2 .03
2
Adjusted R .02
Model F-test 44.61 (5, 8633); p<.001

Table 4.8b. Simple and multiple regressions with demographic variables as


predictors of AARC losses scores on the AARC-10 SF.
(N= 8,639) Demographic variables as predictors Demographic variables as predictors of
of AARC losses: Simple regressions AARC losses: Multiple regression
AARC-10 SF losses
Variables B (95% CI); p-value ß B (95% CI); p-value ß
Age .11 ( .10, .12); <.001 .24 .09 ( .08, .11); <.001 .20
Sex -.80 (-.96, -.64); <.001 -.10 -.60 (-.76, -.44); <.001 -.08
Marital status -.78 (-.95, -.61); <.001 -.10 -.52 (-.68, -.35); <.001 -.06
Employment -.96 (-1.1, -.82); <.001 -.15 -.11 (-.28, .05); .175 -.02
status
University -.45 (-.60, -.29); <.001 -.06 -.35 (-.50, -.19); <.001 -.05
education
Total R2 .07
Adjusted R2 .07
Model F-test 128.74 (5, 8633); p<.001

117
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

Table 4.8c. Simple and multiple regressions with demographic variables as


predictors of gains scores on the AARC-50 cognitive functioning subscale.

(N= 8,639) Demographic variables as predictors of Demographic variables as predictors of


AARC gains: Simple regressions AARC gains: Multiple regression
AARC-50 cognitive functioning gains
Variables B (95% CI); p-value ß B (95% CI); p-value ß
Age -.06 (-.07, -.05); <.001 -.10 -.04 (-.06, -.03); <.001 -.07
Sex 1.30 (1.08, 1.52); <.001 .12 1.06 ( .84, 1.29); <.001 .10
Marital status -.51 (-.74, -.28); <.001 -.05 -.51 (-.75, -.28); <.001 -.05
Employment .78 ( .59, .96); <.001 .09 .41 ( .18, .63); <.001 .05
status
University -.63 (-.84, -.41); <.001 -.06 -.63 (-.85, -.42); <.001 -.06
education
Total R2 .03
2
Adjusted R .03
Model F-test 51.36 (5, 8633); p<.001

Table 4.8d. Simple and multiple regressions with demographic variables as


predictors of losses scores on the AARC-50 cognitive functioning subscale.
(N= 8,639) Demographic variables as predictors of Demographic variables as predictors of
AARC losses: Simple regressions AARC losses: Multiple regressions
AARC-50 cognitive functioning losses
Variables B (95% CI); p-value ß B (95% CI); p-value ß
Age .08 ( .07, .09); <.001 .17 .07 ( .05, .08); <.001 .13
Sex -.88 (-1.06, -.70); <.001 -.11 -.76 (-.95, -.58); <.001 -.09
Marital status -.33 ( -.51, -.14); <.001 -.04 -.17 (-.36, .03); .092 -.02
Employment -.84 ( -.99, -.69); <.001 -.12 -.24 (-.43, -.06); .011 -.03
status
University -.47 ( -.64, -.30); <.001 -.06 -.41 (-.59, -.24); <.001 -.05
education
Total R2 .04
2
Adjusted R .04
Model F-test 70.07 (5, 8633); p<.001
Note: In the regression models we included only those participants that have no
missing data. ß= Standardised beta coefficients are calculated by subtracting the
mean from the variable and dividing it by its standard deviation.

118
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

Table 4.9. Simple and multiple regressions with education as predictor of gains and
losses scores on the AARC-10 SF and AARC-50 cognitive functioning subscale –
sensitivity analyses.
(N= 8,639) Simple regressions Multiple regressions
Variable B (95% CI); p-value ß B (95% CI); p-value ß
University education as predictor -.28 (-.43, -.12); .001 -.04 -.23 (-.43, -.03); .024 -.04
of AARC-10 SF gains
University education as predictor -.51 (-.65, -.38); < .001 -.08 -.11 (-.28, .05); .177 -.06
of AARC-10 SF losses
University education as predictor -.69 (-.87, -.51); < .001 -.08 .41 (.18, .64); < .001 -.08
of AARC-50 cognitive functioning
subscale gains
University education as predictor -.48 (-.62, -.33); < .001 -.06 -.43 (-.58, -.28); < .001 -.06
of AARC-50 cognitive functioning
subscale losses
ß= Standardised beta coefficients are calculated by subtracting the mean from the
variable and dividing it by its standard deviation.

119
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

4.5 Discussion
This was the first study exploring psychometric properties of the AARC-10 SF and
the AARC-50 cognitive functioning subscale in the UK population. We found that
both scales are valid and reliable measures of AARC gains and AARC losses in the
UK population aged 50 and over, that can be used in correlational studies and in
studies comparing AARC across men and women, across individuals with and
without a university degree, and across middle-age, early old age, and advanced old
age. However, some caution should be exercised when comparing the scores of
men and women on the AARC-10 SF. Both scales showed good convergent validity
for AARC losses, but weak convergent validity for AARC gains. Finally, we found
that age, sex, marital status, employment status, and university education explained
significant variability in levels of AARC gains and losses assessed with the AARC-10
SF and the AARC-50 cognitive functioning subscale.
Factor loadings for the two-factor model of the AARC-10 SF and for the two-
factor model of the AARC-50 cognitive functioning subscale were similar to those
found in the US and German validations of the measures (Brothers et al., 2019;
Kaspar et al., 2019), further supporting the use of these AARC measures in the
United Kingdom. Also in line with previous validations of AARC measures, we found
small and moderate overlap between AARC questionnaires (AARC-10 SF and
AARC-50 cognitive functioning subscale) and measures assessing the way in which
individuals experience ageing (subjective age and ATOA), supporting the conceptual
distinction of AARC from similar concepts (Brothers et al., 2019; Kaspar et al., 2019).
The partial overlap of AARC (assessed with both the AARC-10 SF and AARC-50
cognitive functioning subscale) with subjective age and ATOA suggests that AARC
may impact on the way in which individuals feel older or younger than they are or
how changes are reported or appreciated (Bordone & Arpino, 2015; Bowling et al.,
2005). However, subjective age and ATOA may also be associated with individuals’
perceptions of AARC, such as perceptions of age-related cognitive changes. A
recent longitudinal study showed that more negative ATOA predict greater perceived
cognitive decline (Siebert et al., 2020). Similarly, individuals with an older subjective
age perceive more age-related losses in the cognitive domain (Segel-Karpas &
Palgi, 2019). Overall, results relating to the construct validity of the AARC-10 SF and

120
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

the AARC-50 cognitive functioning subscale suggest that levels of AARC losses and
AARC gains are informative of individuals’ mental, physical, and cognitive health.
The correlation we found between higher levels of AARC-10 SF losses and more
symptoms of depression is in line with previous evidence describing the predictive
role of higher AARC losses over higher levels of depressive symptoms (e.g. Brothers
et al., 2019; Kaspar et al., 2019; Neupert & Bellingtier, 2017). Many of the variables
contributing to the aetiology of late-life onset of depression, such as poor physical
health (Fauth et al., 2012), increased dependence on others (Blazer, 2010), and
having little or no social support, are normative aspects of older age. This may
explain why the experiences of age-related losses that are captured in individuals’
perceptions of AARC losses are associated with current and future levels of
depressive symptoms (Dutt & Wahl, 2018; Dutt et al., 2018).
This was the first study exploring correlations between AARC gains and
losses and anxiety. Our results have shown that higher levels of AARC-10 SF losses
are correlated with anxiety. Symptoms of anxiety may be expected in older age and
may be a consequence of the negative changes that people experience in older age
(Butcher & McGonigal-Kenney, 2005). In contrast, the correlations we found
between AARC-10 SF gains and symptoms of depression and anxiety were mixed
and negligible. As similar findings were reported in previous studies (Dutt et al.,
2016a; Dutt & Wahl, 2018; Dutt et al., 2018), along with the newly identified
correlation with anxiety, it may be that when promoting mental health in older age
decreasing AARC losses is more important than increasing AARC gains.
The correlation we found between poorer functional ability and higher AARC
losses is in line with previous studies showing that individuals with poorer everyday
functioning report more AARC losses (Kaspar et al., 2019). Hence, individuals’
perceptions of AARC losses accurately reflect the negative changes that individuals
experience in their lives. The finding that individuals with higher AARC gains and/or
lower AARC losses (assessed with the AARC-10 SF) rate their health more
positively is also in line with previous evidence (Brothers et al., 2019; Kaspar et al.,
2019). Most correlations between AARC (assessed with the AARC-10 SF) and
indicators of health were small or moderate, suggesting the presence of multiple
factors alongside AARC gains and AARC losses that may contribute to experiences

121
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

of ageing and levels of mental and physical health. The size of the correlations of
AARC (assessed with the AARC-10 SF) with indicators of mental and physical health
were stronger for AARC losses than for AARC gains and this is in line with existing
literature on AARC (Dutt et al., 2016a; Sabatini et al., 2020a). It may be that mental
and physical health exert a greater influence on individuals’ perceptions of AARC
losses than on the perceptions of AARC gains (Sabatini et al., 2020a). Perceptions
of AARC gains may instead be more influenced by other factors such as personality
traits (Rupprecht et al., 2019), expectations for the future (Brothers et al., 2016), and
perceived social support (O’Brien & Sharifian, 2019).
This was the first study exploring correlations of AARC in the cognitive
functioning domain (assessed with the AARC-50 cognitive functioning subscale) with
objective, subjective, and informant-rated measures of cognition. We found that
higher levels of AARC losses in the cognitive domain reflect lower objective cognitive
performance and more negative self-evaluations of cognitive changes over 10 years.
Moreover, the correlation between AARC losses in the cognitive domain and
perceived cognitive change over the past ten years remained consistent across three
age groups (aged 50 to 65; aged 66 to 75; and aged 76 and over), suggesting that
the AARC-50 cognitive functioning subscale may detect across middle-age, early old
age, and advanced old age subclinical cognitive decline that is incorporated into
individuals’ ratings of their AARC (Kaspar et al., 2019). This finding is in line with
evidence supporting the value of subjective cognitive complaints in informing about
objective cognitive decline (Jessen et al., 2014; Okonkwo et al., 2009). We also
found that participants’ experience of negative age-related changes is not correlated
with informants’ rating of participants’ change in cognitive abilities over ten years. It
may therefore be that cognitively healthy individuals are aware of the subtle cognitive
changes they are experiencing but that such changes are unnoticed by people close
to them (Clare et al., 2010).
Interestingly, those individuals who performed more poorly on objective
cognitive tasks not only reported higher levels of AARC losses but also perceived
higher levels of AARC gains (assessed with both the AARC-10 SF and AARC-50
cognitive functioning subscale). It may be that in order to compensate for negative
changes in cognition individuals engage in new cognitively stimulating activities

122
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

(Buckner, 2004; Lövdén et al., 2010), resulting in increased self-perception of gains.


Alternatively, reporting high levels of gains alongside high levels of losses may be a
strategy of emotional coping: high levels of losses may cause mental distress which
can be compensated for by directing thoughts towards positive age-related change
(Lövdén et al., 2010). However, the strength of the correlations of AARC gains with
objective cognitive tasks and self-perceived cognitive decline were either of
negligible size or small; self-perceptions of age-related gains in cognition may be
influenced by individuals’ beliefs about ageing, more than by individuals’ actual
cognitive functioning. Overall, as most of the associations of AARC gains with
cognitive indicators were either small or of negligible size, evidence for convergent
validity for the AARC gains assessed with the AARC-50 cognitive functioning
subscale is weaker than evidence for AARC losses.
A secondary aim of this study was to explore whether demographic variables
(age, sex, marital status, employment status, and university education) predict
scores on the AARC-10 SF and AARC-50 cognitive subscale gains and losses. We
found that the demographic variables age, sex, marital status, employment status,
and university education explain some variability in levels of AARC. We found that
being older predicts fewer AARC gains and more AARC losses both in the AARC-10
SF and in the AARC-50 cognitive functioning subscale; the association of higher
AARC losses with being older is in line with previous evidence and with
gerontological literature reporting the greater salience of perceived losses among
older individuals (Baltes, 1987; Brothers et al., 2016).
The associations of older age with fewer AARC gains and more AARC losses
may be due to older individuals having a poorer health status than younger
individuals (Aarts et al., 2012; Barnett et al., 2012). The association between being
older and fewer AARC gains is not consistent with previous evidence reporting a
positive association between older age and higher levels of AARC gains (Brothers et
al., 2016); this discrepancy in results may be due to cultural differences as the
present study included UK residents whereas Brothers and colleagues (2016)
included US and German participants. Discrepancies in the association of age with
AARC gains between UK individuals and US and German individuals may be due to
underlying differences among these countries such as different retirement ages,

123
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

different healthcare systems, and cultural differences in outlook and attitudes


(Ackerman & Chopik, 2020; Barak, 2009; Kornadt et al., 2019b; North & Fiske, 2015;
Westerhof et al., 2011). Alternatively, the discrepancy between results found in the
current study with those of previous studies may be due to differences in the
composition of the study samples that are not due to cultural differences. For
instance, the current study sample may have included participants in poorer health
compared to participants included in previous studies.
We also found that being a man predicts fewer AARC gains and more AARC
losses both in the AARC-10 SF and in the AARC-50 cognitive functioning subscale.
This finding is also in line with existing evidence on sex differences in AARC and in
subjective well-being, showing that men report fewer AARC gains, higher AARC
losses, and lower levels of subjective well-being than women (Dolan et al., 2008;
English et al., 2019; Miche et al., 2014). This may be due to men being less actively
focused on positive changes compared to women (Kaminski & Hayslip, 2006; Moore,
2010). Indeed, research shows that positive experiences of ageing among women
outweigh negative experiences, despite women being aware of significant changes
in their body due to menopause (Hvas, 2006).
In line with existing literature (Bergland et al., 2014; Brown, 2006; Dolan et al.,
2008), we found that being married, in a civil relationship, or co-habiting predicts
fewer AARC losses assessed both with the AARC-10 SF and with the AARC-50
cognitive functioning subscale. However, we also found that being married, in a civil
partnership, or co-habiting predicts lower levels of awareness of positive age-related
change assessed both with the AARC-10 SF and with the AARC-50 cognitive
functioning subscale. Literature on the role of marriage in relation to cognitive
abilities is heterogeneous with some studies reporting lower cognitive abilities among
non-married individuals (Sommerlad et al., 2018) and conversely, others report a
non-significant association between marital status and cognition (Evans et al., 2019).
Our results suggest that working may have distinct effects on different AARC
life domains. Working predicted fewer AARC losses in cognition (as assessed with
the AARC-50 cognitive functioning subscale) compared to non-working and this may
be due to work stimulating cognition. Conversely, working predicted fewer AARC
gains in the remaining AARC life domains (assessed with the AARC-10 SF) and this

124
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

may be due to non-working individuals having more leisure time to enjoy hobbies
and friends compared to workers, resulting in increased likelihood of experiencing
age-related gains.
We found that people with a university education experience fewer AARC
losses, but at the same time also experience fewer AARC gains assessed both with
the AARC-10 SF and the AARC-50 cognitive functioning subscale compared to
individuals without a university education. The fewer AARC losses in the cognitive
domain experienced by those with a university education may be due to such
individuals experiencing lower objective cognitive decline (Grønkjær et al., 2019).
Indeed, education exerts a protective role against cognitive decline (Deary et al.,
2000; Yates et al., 2017). The lower score on the AARC-10 SF losses among those
with a university education may be due to more highly educated people being more
likely to engage in healthy behaviours and therefore to enjoy better physical health
(Craciun et al., 2017; Herd et al., 2007; Leopold & Engelhardt, 2013; Stephan et al.,
2019) and longer life expectancy (Kaplan et al., 2014).
An explanation for the lower levels of both AARC gains and losses on the
AARC-10 SF and on the AARC-50 cognitive functioning subscale reported by those
with a university education may be that individuals who experience low levels of age-
related losses are less likely to reflect on age-related changes and as a
consequence are less aware of positive age-related changes. However, we found
that for both the AARC-10 SF and the AARC-50 cognitive functioning subscale
correlations between AARC gains and AARC losses are negligible, indicating that
there is no overall AARC. The lower levels of AARC gains reported by individuals
with a university education may be due to more educated individuals attributing
positive changes to other causes rather than to their increased age.
The study has limitations that need to be acknowledged. The sample included
mainly white participants, women, individuals who were married (or in a civil
partnership or co-habiting) and who had above average education and self-reported
health. Among the 14,797 participants that took part in the PROTECT annual
assessment between 1st January 2019 and 31st March 2019, 9,410 participants
completed the AARC questionnaires. Compared to those who did not complete the
AARC questionnaires in 2019, the study sample included a larger proportion of

125
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

women and participants who were better educated and a lower proportion of
individuals who were employed. This selection bias may impact on study results
such as on the predictive role that we found for being a woman and having a
university level education over fewer AARC losses. However, there is no immediate
reason to believe that the relationship between the predictors and AARC losses is
different between those who provided data and those who did not.
Data for objective cognitive assessments were not collected on the same day
on which participants completed the AARC questionnaires but were completed within
two months of AARC completion. This was because completing a battery of cognitive
tasks is demanding, especially for older individuals, hence allowing participants to
complete objective cognitive assessments on a separate day from the remaining
measures decreased participants’ burden and increased the likelihood of collecting
accurate answers. Moreover, cognitive functions do not deteriorate or deteriorate
minimally in individuals without dementia over two months (e.g. Lövdén et al., 2004;
Salthouse, 2019). While cognitive abilities were assessed both through objective and
subjective measures, mental and physical health were assessed through self-report
measures only. Finally, individuals who completed a vocational qualification (e.g.
diploma or certificate) were considered to have the same level of education as
participants who completed a undergraduate degree, a master’s degree, or a
doctorate. This is a limitation as several types of vocational qualifications exist, with
some vocational qualifications being comparable to a university level education while
others are not. However, it was not possible to classify participants’ education in a
more detailed manner as PROTECT participants were not asked to specify the type
of vocational qualification they obtained.
Despite the above limitations this study has a large sample size including a
wide age range of UK participants. This is the first study testing content validity of the
AARC-50 cognitive functioning subscale with subjective and objective measures of
cognitive health. The quantification of psychometric properties based on a sample of
participants without a diagnosis of dementia is important because in this study AARC
losses in cognition (assessed with the AARC-50 cognitive functioning subscale) are
associated with objective measures of cognition and therefore could be useful to
identify early cognitive decline, which could in turn support efforts to prevent

126
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

dementia. The AARC-50 cognitive functioning subscale could therefore potentially be


used to identify those segments of the population at greater risk of cognitive decline
who require closer cognitive monitoring, and may benefit from early intervention such
as cognitive training programs (Kaspar et al., 2019; pg. 3) or interventions that help
them to accept their age-related changes and to minimise the negative impact of
age-related cognitive decline (Hahn & Lachman, 2015). The validation of the AARC-
50 cognitive functioning subscale also makes it possible to conduct future research
to better understand the cross-sectional relationship of subjective perceptions of
cognition with objective cognitive functioning, as well as the longitudinal association
with objective cognitive decline. Whereas much research studying perceived
cognitive decline as a predictor of objective cognitive decline exists (Jessen et al.,
2014; Pearman & Storandt, 2004; Tandetnik et al., 2015), the AARC-50 cognitive
functioning subscale is particularly useful as it makes it possible to explore for the
first time whether AARC gains convey protection against cognitive decline.
This is also the first study exploring content validity for the AARC-10 SF with a
measure assessing perceived symptoms of anxiety, in addition to symptoms of
depression, self-rated health, and functional ability that have been explored in the
US and German validations of the AARC-10 SF and the AARC-50 cognitive
functioning subscale (Brothers et al., 2019; Kaspar et al., 2019). Good psychometric
properties of the AARC-10 SF make it possible to use this scale to assess positive
and negative perceptions of age-related changes in several domains of one’s life in
clinical and research contexts within the United Kingdom.

4.6 Conclusion
The AARC-10 SF is a valid and reliable measure to identify segments of the
population that experience substantial change across multiple life domains as a
consequence of their ageing process. The brief measure may also be useful in
clinical and counselling settings within the United Kingdom to identify those
individuals who, because of higher levels of AARC losses and/or lower levels of
AARC gains, may benefit from interventions helping them to understand their age-
related changes, to adapt to age-related changes, or to engage in healthy
behaviours counteracting age-related losses (Kaspar et al., 2019).

127
Chapter 4: International relevance of two measures of awareness of age-related

change (AARC)

The AARC-50 cognitive functioning subscale – while capturing a more narrow


facet of the experience of ageing - also proved to be a valid and reliable measure
that could be used to identify those segments of the population at greater risk of
cognitive decline and that may require closer cognitive monitoring or may benefit
from early intervention such as cognitive training programs (Hudes et al., 2019;
Kaspar et al., 2019). Finally, as we found that demographic variables play a role in
the experience of AARC gains and AARC losses, future studies on AARC should
give a more detailed account of the mechanisms that foster the experience of age-
related gains or losses.

128
Chapter 5: Exploring awareness of age-related change among over 50s in the United

Kingdom: Qualitative and quantitative findings from the PROTECT study

Chapter 5: Exploring awareness of age-related changes among over 50s in the


United Kingdom: Qualitative and quantitative findings from the PROTECT
study

129
Chapter 5: Exploring awareness of age-related change among over 50s in the United

Kingdom: Qualitative and quantitative findings from the PROTECT study

5.1 Summary
Chapter 4 showed that two measures of AARC are valid and reliable to be used
among UK individuals aged 50 and over. Chapter 5 aims to identify additional
information about the validity and use of the AARC questionnaires in the UK
population. Chapter 5 also aims to better understand the conceptual framework of
AARC outlined by Diehl and Wahl (2010) by identifying new variables that
participants consider to be associated with their AARC. In order to achieve this, the
empirical study presented in Chapter 5 categorises the written thoughts that
participants provided while completing the AARC questionnaires.

Abstract
Introduction: Older people describe both positive and negative age-related
changes, but we do not know much about what contributes to making them aware of
these changes as they develop.
Methods: Participants (N= 609; aged 50 and over) completed a questionnaire
assessing awareness of age-related change (AARC) and responded to an open-
ended question asking them to comment on their responses. Using content analysis,
we categorised participants’ comments and explored the extent to which these
mapped onto theoretical conceptualisations of influences on AARC.
Results: While some of the emerging categories were in line with the existing
conceptual framework of AARC, for example experiencing positive and negative
changes or attitudes toward ageing, others were novel, for example engagement in
purposeful activities or involvement in roles that provide distraction from negative
age-related thoughts. Analyses also revealed some of the thought processes
involved in selecting responses to the questionnaire items, demonstrating different
ways in which people make sense of specific items.
Discussion and conclusion: These findings suggest ways of enriching the
theoretical conceptualisation of how awareness of age-related change develops
among over 50s. They also offer insights into interpretation of responses to
measures of AARC.

130
Chapter 5: Exploring awareness of age-related change among over 50s in the United

Kingdom: Qualitative and quantitative findings from the PROTECT study

5.2 Introduction
The global proportion of older individuals is increasing and this results in a growing
number of people experiencing poor cognitive, mental, and physical health (McGrath
et al., 2019; United Nations, 2019). Maintenance of good health throughout later life
is vital for quality of life and for the sustainability of the world health system and
economy. The study of subjective ageing may suggest ways of promoting health in
the second half of life. Indeed, those individuals scoring more positively on subjective
ageing measures are more likely to engage in healthy behaviours and to experience
better health (Jaconelli et al., 2017; Klusmann et al., 2017; Kornadt et al., 2019a;
Montepare, 2019; Westerhof et al., 2014) than those with more negative subjective
ageing. However, the way in which subjective ageing has been conceptualised and
measured so far may have resulted in a simplified understanding of the relation
between subjective ageing and health (Diehl et al., 2014).
Many existing measures of subjective ageing are unidimensional (e.g. Attitude
Toward Own Ageing Scale; Lawton, 1975). In this way respondents can report either
negative or positive subjective ageing, but not both. However, developmental gains
and losses coexist throughout the lifespan (Baltes, 1987; Miche et al., 2014).
Moreover, many measures provide global evaluations of subjective ageing (e.g.
subjective age; Barrett, 2003) even though individuals can experience different
perceptions of ageing in relation to distinct life domains (e.g. social, cognitive,
physical; Carstensen et al., 2011; Steverink et al., 2001; Timmer et al., 2002; Voss et
al., 2018). The concept of awareness of age-related change (AARC) aims to
overcome these limitations (Diehl & Wahl, 2010; Diehl et al., 2014) by capturing self-
perceptions of positive (gains) and negative (losses) age-related changes across five
life domains - health and physical functioning, cognition, interpersonal relationships,
socio-cognitive and socio-emotional functioning, and lifestyle and life engagement.
Perceptions of AARC gains and losses are assessed via questionnaire (e.g.
AARC-10 SF; Kaspar et al., 2019) and this has recently been validated within the UK
population (Sabatini et al., 2020b). Examples of items assessing AARC gains and
losses respectively are “With my increasing age, I realise that I have become wiser”
and “With my increasing age, I realise that I have less energy”. The unique
conceptual features of AARC may make it possible to advance our understanding of

131
Chapter 5: Exploring awareness of age-related change among over 50s in the United

Kingdom: Qualitative and quantitative findings from the PROTECT study

the mechanisms behind the association of subjective ageing with health. Based on
existing literature on subjective ageing Diehl and Wahl (2010) proposed a heuristic
framework placing the AARC construct into a broader conceptual framework
including potential antecedents, variables involved in the mental process, and
outcomes of AARC. Hypothesised distal antecedents of AARC comprise
sociodemographic, health-related, and psychological factors, whereas proximal
antecedents include personal goals, life limitations, age stereotypes, and life events.
Hypothesised factors involved in the mental process of AARC include personal
meaning-making and self-regulation of behaviour. Finally, hypothesised outcomes of
AARC are psychological health, physical health, and life engagement.
So far the relationships of AARC with some of the variables included in the
Diehl and Wahl (2010) conceptual framework have been widely researched,
whereas other potential relationships remain unexplored. The interest of the current
study is on factors influencing the formation and mental process of AARC. Examples
of known antecedents of AARC are sociodemographic variables and attitudes toward
own ageing (ATOA; Brothers et al., 2016; Brothers et al., 2020; Brothers et al., 2017;
English et al., 2019; Sabatini et al., 2020b); whereas examples of unexplored factors
are social engagement and lifestyle. An example of a variable related to the mental
process of AARC is engagement in self-regulatory behaviours (Dutt et al., 2016b),
whereas an example of an unexplored factor is the interpretation of changes as
being a consequence of ageing or of other factors such as illnesses (Rothermund et
al., 2020). Finally, factors related to AARC may vary among cultures (Sabatini et al.,
2020a).
In the present study we aim to understand what people take into consideration
when reporting instances of AARC. This can be achieved by qualitatively examining
the open-ended comments people make about their responses to the questionnaire
and considering these alongside the answers they give to the questionnaire items.
Indeed, qualitative accounts are useful in providing insights into factors that trigger
perceptions of AARC (Miche et al., 2014). However, the study by Miche et al. (2014)
is the only one exploring which factors individuals take into consideration when
reporting their AARC and is limited in that it focused exclusively on reports of daily
events; whether the formation of age-related thoughts is influenced by broader

132
Chapter 5: Exploring awareness of age-related change among over 50s in the United

Kingdom: Qualitative and quantitative findings from the PROTECT study

psychosocial factors has never been investigated with qualitative analysis. In


addition, Miche et al. (2014) included only participants aged 70 and over. However,
perceptions of age-related change become salient from middle-age (Kornadt et al.,
2018) and, due to the good health that generally characterises middle-aged
individuals (Balcombe & Sinclair, 2001), events related to AARC in this age group
may be different. It would be valuable to explore a wider range of factors associated
with AARC across the second half of life to reach a more comprehensive
understanding of how AARC is formed. Finally, as AARC has mostly been
researched among US and German residents, there is a need to understand what
influences instances of AARC in other cultures. Hence in this study we focus on
individuals living in the United Kingdom.
In sum, this study attempts to increase knowledge about what contributes to
making people aware of age-related changes as these develop in the second half of
life. This will be achieved by examining and categorising the written comments made
by UK individuals aged 50 and over when answering the AARC-10 SF and by
exploring how participants’ comments are related to their levels of AARC gains and
losses. In particular we want to understand the relative salience of daily life events
and broader psychosocial factors on AARC, and whether different types of thoughts
described by participants are differentially related to levels of AARC gains and
losses. For instance, individuals who have experienced a recent health issue may
report higher AARC losses and/or lower AARC gains compared to individuals who
do not report a recent health issue. We aim 1) to evaluate the degree to which the
comments that participants make while answering the AARC questionnaire map onto
the categories outlined in the Diehl and Wahl (2010) conceptual framework of AARC;
2) to identify what issues UK individuals report in relation to the items included in the
AARC questionnaire; and 3) to extend our understanding of why AARC is associated
with health outcomes.

5.3 Methods
Design and participants
Analyses were conducted on cross-sectional data collected as part of the PROTECT
study (https://www.protectstudy.org.uk). In the UK version of the PROTECT study

133
Chapter 5: Exploring awareness of age-related change among over 50s in the United

Kingdom: Qualitative and quantitative findings from the PROTECT study

participants are UK residents, English speakers, aged 50 years and over, with
access to the internet, and no clinical diagnosis of dementia at the time of
recruitment. The PROTECT study was publicised nationwide and among existing
cohorts of older adults (https://exetercrfnihr.org/about/exeter-10000/;
https://www.joindementiaresearch.nihr.ac.uk/; https://bdr.alzheimersresearchuk.org).
Participants provided their informed consent online through the PROTECT website.
Ethical approval for the PROTECT study was obtained from the London Bridge NHS
Research Ethics Committee and Health Research Authority (Ref: 13/LO/1578).
Ethical approval for the secondary data analysis was obtained by the Ethics
Committee of the School of Psychology, University of Exeter (Ref:
eCLESPsy000603v2.1).
On an annual basis PROTECT participants complete measures of health and
lifestyle through an online platform. As part of the PROTECT assessment in January
2019 participants were invited to complete additional questionnaires specific to this
study including the AARC-10 SF (Kaspar et al., 2019) and an open-ended question
asking participants to write any additional information they wished to report in
relation to their AARC. Inviting participants to write comments in relation to their
AARC makes it possible to explore their thoughts as these arise while answering the
AARC-10 SF (Teal et al., 2015), hence we deemed these textual data related to
AARC suitable for qualitative exploration. Among the 14,757 participants that took
part in the PROTECT study in 2019, 609 participants answered the AARC open-
ended question and hence were included in the current study. Mean differences in
demographic characteristics across the sample of individuals who answered the
AARC open-ended question and the sample who did not answer the open-ended
question are reported in Table 5.1.

Measures
Demographic variables included age, sex, ethnicity (white; mixed; Asian;
black; other), marital status (married versus not married), education level
(secondary; post-secondary; vocational qualification; undergraduate degree; post-
graduate degree; doctorate), and employment status (employed versus not
employed). AARC was assessed with the AARC-10 SF (Kaspar et al., 2019) which

134
Chapter 5: Exploring awareness of age-related change among over 50s in the United

Kingdom: Qualitative and quantitative findings from the PROTECT study

includes ten items, five measuring gains and five measuring losses across the five
AARC life domains. Respondents evaluate how much items apply to them on a five-
point scale (1= Not at all; 2= a little bit; 3= moderately; 4= quite a bit; 4= very much).
Scores for the gains and losses scales are obtained by summing items belonging to
the respective scales (5-25); higher scores indicate higher levels of AARC. In the UK
validation of the AARC-10 SF, Cronbach’s alpha (α) was .77 for the gains subscale
and .80 for the losses subscale (Sabatini et al., 2020b). The following open-ended
question was used to ask participants to add their thoughts regarding AARC: “If you
wish, please add any additional information or comments for this section in the box
below”. This question was asked only once after completion of both the AARC-10 SF
and the AARC-50 cognitive functioning subscale.

Data analyses
Content analysis was applied to participants’ answers to the AARC open-
ended question (Grbich, 2012; Pope et al., 2000). Content analysis was a suitable
method for this study as it can be applied to short written comments (Neuman,
2011). This approach made it possible to generate categories directly from the text
(Hsieh, 2005) and to analyse data both qualitatively and quantitatively.
The first author manually analysed the data using the following steps:

(1) Data familiarisation through reading and re-reading participants’ responses


(Polit & Beck, 2004);
(2) Definition of the categories and manual assignment of participants’ comments
to categories. When appropriate, participants’ statements were assigned to
more than one category (Richards, 2015);
(3) Visual presentation of categories in a diagram.
To maximise the reliability of the identified categories (Elliott et al., 1999), 10% of the
responses were analysed by a second researcher not otherwise involved in the
study. Differences among researchers were discussed until agreement was reached.
Cohens’ Kappa coefficient for the agreement between raters was .89%; indicating
acceptable agreement (Cohen, 1960). To demonstrate that the categories were
generated from the data (Whittemore et al., 2001), in the results section (under the

135
Chapter 5: Exploring awareness of age-related change among over 50s in the United

Kingdom: Qualitative and quantitative findings from the PROTECT study

section: comments associated with AARC judgments) descriptions of categories are


integrated with relevant extracts.
As quantitative data can contribute to the interpretation of meaning in qualitative
accounts (Sandelowski & Barroso, 2003), we calculated the number of participants
whose comments were assigned to each of the identified categories. We used the t-
test to explore whether those individuals whose comments were assigned to a
specific category had a statistically significantly different mean in AARC gains or
losses compared to those individuals whose comments were not assigned to the
same category. This enabled us to understand whether the experience of certain life
events and/or psychosocial factors makes individuals more likely to perceive higher
or lower levels of AARC gains and losses. Hedges’s g was used as indicator of effect
size. Analyses were conducted in STATA version 16 (StataCorp, 2017).

5.4 Results
Comments associated with AARC judgments
The average length of comments was 1.5 sentences per participant, although length
varied considerably and ranged from a short sentence of three words to ten
sentences. From analyses of participants’ comments about their AARC, ten
categories capturing comments associated with AARC were generated and grouped
into three higher-order categories (Figure 5.1). The first higher-order category covers
“subjective experiences while ageing” and includes five categories: “experiencing
negative changes”, “experiencing positive changes”, “attitudes toward ageing”,
“feeling unchanged”, and “purpose-driven experiences”. The second higher-order
category captures “factors impacting on evaluation of awareness of changes” and
includes three categories: “fluctuating levels of awareness of changes”, “disregarding
the impact of increased age”, and “difficulty when answering the AARC
questionnaire”. The third higher-order category describes the “process of awareness
of changes” and includes two categories: “healthy lifestyle and self-regulation” and
“social comparison”.

136
Chapter 5: Exploring awareness of age-related change among over 50s in the United Kingdom: Qualitative and quantitative findings

from the PROTECT study

Healthy lifestyle
and self-
regulation
Factors impacting on
Subjective experiences
evaluation of awareness of
while ageing:
changes:
Potential
1) Experiencing 1) Fluctuating levels of Attribution of interventions
Process of
negative changes Awareness awareness of changes changes to teaching to accept
awareness
2) Experiencing of changes 2) Disregarding the increased and/or to adapt to
of changes
positive changes impact of increased age (AARC) changes
3) Attitudes toward age
ageing 3) Difficulty when
4) Feeling unchanged answering the AARC
5) Purpose-driven questionnaire Social
experiences comparison

Figure 5.1. Thematic map illustrating predictors of AARC, factors explaining lack of accuracy in AARC, and the mental process of
AARC.

137
Chapter 5: Exploring awareness of age-related change among over 50s in the United

Kingdom: Qualitative and quantitative findings from the PROTECT study

Subjective experiences while ageing. Examples of participants’ comments for


the identified categories are presented in Table 5.2. The higher-order category
“Subjective experiences while ageing” describes factors that can increase or
decrease awareness of changes. Some participants (33.3% of the sample)
experienced negative changes across various life domains such as physical and
cognitive domains. Changes in the physical domain included the presence of a
chronic health condition, a recent injury or illness, falls, physical pain, decreased
energy, and changes in physical appearance. Changes in the cognitive domain
related to loss of memory or concentration. However, some participants also
experienced positive changes (18.2% of the sample) that, similar to awareness of
negative changes, happened in various domains including mental health (e.g. being
happier), physical health, cognitive abilities (e.g. increased knowledge), life
engagement (e.g. more time to pursue hobbies), and even characterised their
personality (e.g. becoming less impulsive).
For some participants awareness of changes was influenced by their attitudes
toward ageing and life in general (6.7% of participants). Some people experienced
older age as a positive period of life and perceived older people as competent,
whereas others viewed older age as a negative stage of life and believed that older
people are fragile and in need of help. For some participants, their attitudes toward
life changed while ageing; for instance, individuals become more aware of the
finitude of life or shift their focus from the future to the present.
Other participants instead felt unchanged or at least felt that some aspects of
their lives, attitudes, or behaviours had not changed with their increased age (28% of
participants). Finally, another factor that some participants (12.8% of participants)
described as being related to their awareness of changes is their level of
engagement in a variety of activities including work, hobbies (e.g. travelling), or
caring for someone living with an illness or for grandchildren. Being able to conduct
specific activities made some participants feel competent or conferred a sense of
purpose. For other individuals having several commitments left them little time to
reflect on the age-related changes they were experiencing, resulting in lower levels
of awareness of changes.

138
Chapter 5: Exploring awareness of age-related change among over 50s in the United

Kingdom: Qualitative and quantitative findings from the PROTECT study

Factors impacting on evaluation of awareness of changes. In the higher-order


category “Factors impacting on evaluation of awareness of changes” participants
remarked how several factors can influence the levels of changes they perceive.
First, a small proportion of participants (1.2% of participants) reported that their
AARC fluctuates depending on several factors such as the situation they are in.
Many individuals (31.3% of participants), despite acknowledging the presence of
relevant, specific changes in their life, clarified that such changes are not due to their
increased age. Instead, participants identified various factors that in their opinion
cause the experienced changes, including being a caregiver, a recent bereavement,
financial restrictions, societal changes, a physical condition, the menopause, and a
mental health issue. Finally, some individuals (6.2% of participants) had difficulty in
answering the AARC questionnaire due to being forced to choose an answer where
they would have preferred a non-applicable option, finding items to be vague or not
understanding the specific meaning. Overall, the above-described factors impacted
on the scores that a relevant proportion of participants obtained on the AARC
questionnaire. This may result in decreased accuracy of individuals’ scores on the
AARC-10 SF.
Process of awareness of changes. The higher-order category “Process of
awareness of changes” describes mental processes that can result in intention to
react or adapt to changes and mental processes that individuals use to make sense
of the changes they experience. A small proportion of people reacted or adapted to
AARC losses by abandoning physically demanding activities and hobbies and
engaging in new activities and/or in healthy behaviours (7.5% of participants). Few
individuals (1.5% of participants) used mental strategies such as comparing
themselves with other people that were in a less advantageous position than them or
emphasising the positive opinion that other individuals had about them. Frequencies
under each category for the study sample are presented in Table 5.3.

139
Chapter 5: Exploring awareness of age-related change among over 50s in the United

Kingdom: Qualitative and quantitative findings from the PROTECT study

Table 5.2. Examples of participants’ comments for each of the identified categories.

Higher- Categories Examples of participants’ comments per each of the identified categories
order
categories
Subjective Experiencing “I have recently acquired a nerve injury which affects my mobility” (Woman, 63
experiences negative years).
while changes “I have polymyalgia rheumatic diagnosed 2 years ago” (Woman, 63 years).
ageing “Osteo-arthritis in knees and (sometimes) pain in hips” (Woman, 69 years).
“I would say, I have fewer physical stamina” (Woman, 71 years).
“My appearance e.g. thinning hair affects how I feel about my age” (Woman, 62
years).
“Memory is failing” (Woman, 76 years).
“Sometimes I forget that I have put a teabag into the pot and do it again”
(Woman, 65 years).
Experiencing “I think I am happier than ever, and more aware of my good fortune in life. I can
positive take more time to enjoy myself and make more of every outing” (Woman, 71
changes years).
“My general physical and mental well-being improved dramatically after I have
retired 5 years ago. My knowledge base, my designing and planning to do
things, and my manual work improved many times over” (Man, 67 years).
“Life experience is quite a useful asset” (Woman, 91 years).
“I am more confident in my decisions” (Woman, 58 years).
Attitudes “Overall, I do not feel negative about my age, though it seems many others in
toward my age group appear to have negative thinking” (Woman, 70 years).
ageing “I am quite positive about growing older” (Woman, 56 years).
“Don’t believe in using age as a barrier” (Woman, 54 years).
“Don’t like getting old, the brain is still willing but the body at times say `No
chance mate`!” (Man, 65 years).
“We should all be allowed to book a place at the Vets, be given an injection, die
and be buried in our gardens” (Woman, 77 years).
“Getting older is a mixed blessing” (Woman, 79 years).
“Old age isn’t for the faint hearted” (Man, 75 years).
“I seem to have lost enthusiasm for life” (Woman, 73 years).
“More aware of mortality” (Woman, 56 years).
“I am realising, I am no longer immortal!” (Man, 65 years).
Feeling “I really don’t feel any different” (Woman, 63 years).
unchanged “I don’t often think or recognise my increasing age” (Woman, 68 years).
Purpose- “I am a competent administrator and a hard worker” (Woman, 71 years).
driven “Volunteering in two museums, in addition to a new range of hobbies and
experiences pastimes during retirement can be as demanding as working full-time. I enjoy
the challenge of new projects” (Woman, 72 years).

140
Chapter 5: Exploring awareness of age-related change among over 50s in the United

Kingdom: Qualitative and quantitative findings from the PROTECT study

“As a wife, mother of 5 children, with 8 grandchildren, active in my community, I


don’t yet have the time to focus on just my own needs nor to ‘have more
freedom to live my days the way I want’. But that is not something I aspire to. I
enjoy my life the way it is!” (Woman, 67 years).
“As I have children and grandchildren living with me, that keeps me fully
occupied” (Woman, 69 years).
Factors Fluctuating “I think my answers could vary from time to time” (Woman, 72 years).
impacting levels “These answers are all situation dependent” (Woman, 59 years).
on of “All very dependent on how I wake up in the morning!” (Woman, 83 years).
evaluation awareness of “My energy levels have become more markedly seasonal – low energy /
of changes motivation / activity in winter” (Woman, 65 years).
awareness Disregarding “My husband was very ill last year, and this has affected me in many ways”
of changes the impact of (Woman, 72 years).
increased “Being widowed has made me feel old” (Woman, 77 years).
age “Lack of money can be very limiting (Woman, 78 years).
“Some of the answers are affected by my physical condition rather than my
age” (Woman, 72 years).
“I think many issues may be due to menopause than age!” (Woman, 62 years).
“My bipolar disorder interferes at times with all these” (Woman, 60 years).
Difficulty “For some questions none of the answers apply. There should be a not
when applicable option” (Woman, 78 years).
answering “With my increasing age is too vague. Do you mean in the last five years, or
the AARC since I was a child, or since I turned 21, or since I retired?” (Man, 69 years).
questionnaire “Not sure what foresight means” (Man, 59 years).
Process of Attention to “I have started a beginners French language course to stimulate that part of the
awareness healthy brain” (Woman, 74 years).
of changes lifestyle and “I have researched the things I can do which may prevent or delay dementia,
adaptation to and I try to do all of them as often as possible. Since both my parents and
AARC sister had or have dementia, I am committed to a healthy lifestyle, mentally and
physically” (Woman, 68 years).
“I realise that as I get older, I need to take good care of myself” (Woman, 63
years).
“I am more inclined to pay to have other people do jobs such as house
alterations which I would previously do myself” (Man, 77 years).
“I have to do sports in a less extreme way so only windsurf in light winds, ride
quiet horses, less uphill walks” (Woman, 75 years).
“Rather than brisk walking, these days I prefer to cycle. Less wear and tear on
my knee and hip joints” (Man, 71 years).
Social “I feel very fortunate to have the degree of use of all of my faculties that I do
comparison have compared to many other people I see around me of the same age” (Man,
72 years).
“I have based some of these answers on what other people have told me”
(Woman, 77 years).

141
Chapter 5: Exploring awareness of age-related change among over 50s in the United

Kingdom: Qualitative and quantitative findings from the PROTECT study

Table 5.3. Number of observations and frequencies of percentages for the study
sample who answered the open-ended question.

Higher order categories Categories Number of observations


(frequencies) (N= 609)
Subjective experiences while Experiencing negative changes 203 (33.3 %)
ageing Experiencing positive changes 111 (18.2 %)
Attitudes toward ageing 41 ( 6.7 %)
Feeling unchanged 171 (28.0 %)
Purpose-driven experiences 78 (12.8 %)
Factors impacting on evaluation of Fluctuating levels of awareness of 7 ( 1.2 %)
awareness of changes changes
Disregarding the impact of 191 (31.3 %)
increased age
Difficulty when answering the 38 ( 6.2 %)
AARC questionnaire
Process of awareness of changes Attention to healthy lifestyle 46 ( 7.5 %)
and adaptation to AARC
Social comparison 9 ( 1.5 %)

Relations among categories


The ten categories are organised in a diagram showing how they are inter-related
(Figure 5.1). The diagram shows how different subjective experiences of ageing can
lead to varying degrees of awareness of changes, attempts to explain how
awareness of changes in some cases may not be interpreted as a consequence of
increased age, and how awareness of changes may trigger specific mental
processes. The left side of the diagram illustrates five categories (experiencing
positive changes, experiencing negative changes, ATOA, feeling unchanged, and
purpose-driven experiences) that may enhance or decrease awareness of changes
happening in individuals’ lives. The centre of the diagram illustrates variables that,
irrespective of the changes that individuals are experiencing, may impact on
individuals’ evaluation of awareness of changes, and hence on scores on the AARC
questionnaire. Indeed, some participants did not attribute changes to their increased
age, experienced fluctuations in levels of awareness of changes, and had difficulty in
understanding the AARC questionnaire. The right side of the diagram illustrates how

142
Chapter 5: Exploring awareness of age-related change among over 50s in the United

Kingdom: Qualitative and quantitative findings from the PROTECT study

being aware of changes may trigger mental processes related to the implementation
of self-regulatory strategies, attention to healthy lifestyles, and social comparison.
These mental processes could potentially be related to motivation to engage in
interventions teaching individuals to adapt to age-related changes.

Comparing levels of AARC gains and losses among individuals whose


comments were and were not assigned to a specific category
Scores on the AARC gains and AARC losses subscales were normally distributed.
Results from the t-test showing whether the extent of gains or losses for those
individuals who had comments assigned to the given category was different to the
extent of gains or losses reported by individuals whose comments were not assigned
to the same category are reported in Table 5.4 and 5.5. Those participants who felt
unchanged and/or found it difficult to answer the AARC questionnaire reported fewer
AARC gains than those participants whose comments were not assigned to these
categories. Participants who reported purpose-driven experiences and attention to a
healthy lifestyle and adaptation to age-related changes perceived more AARC gains
than those participants whose comments were not assigned to these categories.
Regarding the remaining seven categories, participants who commented on a
specific category did not differ significantly in levels of AARC gains from those
participants whose comments were not assigned to the same category.
Those participants who reported feeling unchanged or described purpose-
driven experiences perceived fewer AARC losses than those participants whose
comments were not assigned to these categories. Participants who reported
experiencing negative changes, fluctuating levels of awareness of changes, and
disregarding the impact of their age on the changes they are experiencing, perceived
more AARC losses than participants whose comments were not assigned to these
categories. Regarding the remaining five categories, participants whose comments
were assigned to a specific category did not differ significantly in levels of AARC
losses from those participants whose comments were not assigned to the same
category.

143
Chapter 5: Exploring awareness of age-related change among over 50s in the United Kingdom: Qualitative and quantitative findings

from the PROTECT study

Table 5.4. For each of the ten categories, the difference in the AARC gains mean between individuals whose comments were assigned to a
category and individuals whose comments were not assigned to the category.

Categories Answered or not to the category N of observations AARC gains, M (SD) t-test p-value Hedges’s g (95%CI)
Experiencing negative changes No 406 17.56 (4.17) -1.51 .132 -.13 (-.30, .04)
Yes 203 18.09 (3.78)
Experiencing positive changes No 498 17.64 (4.10) -1.27 .204 -.13 (-.34, .07)
Yes 111 18.18 (3.78)
Attitudes toward ageing No 568 17.66 (4.04) -1.71 .088 -.28 (-.59, .04)
Yes 41 18.78 (4.02)
Feeling unchanged No 438 18.08 (3.95) 3.33 .001 .30 ( .12, .48)
Yes 171 16.87 (4.17)
Purpose-driven experiences No 531 17.58 (4.04) -2.51 .012 -.30 (-.54, -.66)
Yes 78 18.81 (3.92)
Fluctuating levels of No 602 17.77 (4.04) 1.61 .107 -.61 (-.13, 1.36)
awareness of changes Yes 7 15.29 (4.23)
Disregarding the impact of No 418 17.59 (4.0) -1.36 .175 -.12 (-.29, -.05)
increased age Yes 191 18.07 (4.14)
Difficulty when answering the No 571 17.91 (3.96) 4.16 <.001 .70 ( .36, 1.03)
AARC questionnaire Yes 38 15.13 (4.50)
Attention to healthy lifestyle No 563 17.63 (4.01) -2.36 .019 -.36 (-.66, -.06)
and adaptation to AARC Yes 46 19.09 (4.29)
Social comparison No 600 17.72 (4.05) 17.40 .390 -.29 (-.95, .37)
Yes 9 18.89 (3.92)
Note: t-test = two-group mean comparison test. Hedges’s g = effect size. Degrees of Freedom (df) = 607.

144
Chapter 5: Exploring awareness of age-related change among over 50s in the United Kingdom: Qualitative and quantitative findings

from the PROTECT study

Table 5.2. For each of the ten categories, the difference in the AARC losses mean between individuals whose comments were
assigned to a category and individuals whose comments were not assigned to the category.

Categories Answered or not to the category N of observations AARC losses, M (SD) t-test p-value Hedges’s g (95%CI)

Experiencing negative No 406 10.10 (3.59) -2.37 .018 -.20 (-.37, -.03)
changes Yes 203 10.81 (3.39)
Experiencing positive changes No 498 10.42 (3.57) 1.37 .171 .14 (-.06, .35)
Yes 111 9.92 (3.38)
Attitudes toward ageing No 568 10.39 (3.57) 1.54 .123 .25 (-.07, .57)
Yes 41 9.51 (2.97)
Feeling unchanged No 438 10.85 (3.67) 5.95 <.001 .54 ( .36, .72)
Yes 171 9.01 (2.76)
Purpose-driven experiences No 531 10.53 (3.52) 3.60 <.001 .44 ( .20, .67)
Yes 78 9.0 (3.40)
Fluctuating levels of No 602 10.28 (3.50) -3.87 <.001 -1.47 (-2.22, -.72)
Awareness of changes Yes 7 15.43 (3.69)
Disregarding the impact of No 418 10.01 (3.42) -3.36 <.001 -.29 (-.46, -.12)
increased age Yes 191 11.04 (2.69)
Difficulty when answering the No 571 10.38 (3.58) 1.12 .262 .19 (-.14, .52)
AARC questionnaire Yes 38 9.71 (2.83)
Attention to healthy lifestyle No 563 10.32 (3.56) -.46 .647 -.07 (-.37, .23)
and adaptation to AARC Yes 46 10.57 (3.34)
Social comparison No 600 10.31 (3.52) -1.23 .218 -.41 (-1.07, .24)

145
Chapter 5: Exploring awareness of age-related change among over 50s in the United Kingdom: Qualitative and quantitative findings

from the PROTECT study

Yes 9 11.78 (4.66)


Note: t-test= two-group mean comparison test. Hedges’s g = effect size. 0.2 = small effect; 0.5 = medium effect; 0.8 = large effect.
Degrees of Freedom (df) = 607.

146
Chapter 5: Exploring awareness of age-related change among over 50s in the United

Kingdom: Qualitative and quantitative findings from the PROTECT study

5.5 Discussion
This study was the first to examine and categorise the open-ended comments made
by UK individuals aged 50 and over in relation to their responses on the AARC-10
SF. Levels of AARC gains and losses were shaped by a wide range of life events
and psychosocial factors. However, in some cases scores on the AARC-10 SF could
be influenced by some sources of bias related to completion of the AARC items.
Some of the factors identified in relation to perceptions of AARC were in line with the
categories outlined in the Diehl and Wahl (2010) conceptual framework of AARC,
whereas others were novel and had not been previously considered as potential
contributors to AARC; for example, engagement in purpose-driven activities or in
activities that distract from age-related thoughts. Among the identified sources of
inaccuracy related to the AARC questionnaire, perhaps the most important is that a
relevant proportion of participants scored high on AARC losses even though they did
not interpret the experienced changes as being a consequence of ageing. In the
following paragraphs the identified higher-order categories with their respective sub-
categories are discussed in relation to the Diehl and Wahl (2010) conceptual
framework of AARC. The higher-order category subjective experiences while ageing
will be discussed first, followed by mental process of awareness of changes and
factors impacting on evaluation of awareness of changes. Finally, we will consider
how study results provide possible interpretations for the associations of AARC with
health outcomes.
Most of the categories that we identified (experiencing negative changes,
experiencing positive changes, attitudes toward ageing, and purpose-driven
experiences) within the higher-order category subjective experiences while ageing
corroborate the hypothesised roles of health status, experience of limitations, current
life events, age stereotyping, and personal goals as factors influencing the formation
of AARC (Diehl & Wahl, 2010). However, our results suggest a more complex
picture. Whereas so far theoretical and empirical literature considered abandoned
goals and age-related roles as factors increasing awareness of age-related losses,
we found that those individuals engaged in pursuing life goals (e.g. volunteering)
perceived more gains and fewer losses than those not engaged in pursuing life
goals. Hence, future interventions could enhance engagement in life goals as a way

147
Chapter 5: Exploring awareness of age-related change among over 50s in the United

Kingdom: Qualitative and quantitative findings from the PROTECT study

of promoting more positive experiences of ageing. In line with Terror Management


Theory (Burke et al., 2010), it may be that undertaking roles and hobbies typically
assumed by older people increases awareness of mortality, negative emotions, and
negative self-perceptions (including higher AARC losses) only when individuals are
not engaged in activities that convey a high sense of purpose and self-efficacy.
Terror Management Theory (Burke et al., 2010; Greenberg et al., 1986) indeed
assumes that increasing age and death awareness give rise to anxiety only among
those individuals who do not have a sense of purpose and/or high levels of self-
efficacy.
Alternatively, it may be that the engagement in certain roles such as caring for
grandchildren acts as a full-time occupation that reduces individuals’ time to reflect
on and worry about negative changes (Bordone & Arpino, 2015). Finally, some
participants reported “feeling unchanged”; these individuals perceived fewer gains
and losses compared to those who did not feel unchanged. This empirically supports
the hypothesis that those individuals who are less self-reflective may maintain a
more consistent self-concept and hence report fewer AARC gains and losses,
regardless of the amount of objective changes experienced (Diehl & Wahl, 2010;
Kaufman, 1986). Future studies could explore whether levels of introspection
mediate the effects of antecedents on subjective perceptions of AARC.
The second higher-order category mental process of awareness of changes
corroborates Diehl and Wahls’ (2010) hypothesis that perceptions of AARC give rise
to a series of psychological processes involving self-regulation and meaning-making.
In line with the Two-Process Framework (Brandtstädter & Rothermund, 2002), for
some individuals AARC was a precondition for developing the intention to engage in
a “healthy lifestyle and/or in self-regulation” (e.g. abandonment of old hobbies for
less physically demanding ones). Moreover, these individuals scored higher on
AARC gains compared to those who did not engage in such behaviours, confirming
that perceived gains can be maintained through adaptation even when individuals
experience negative changes (Baltes, 1987; Brandtstädter & Rothermund, 2002).
For a small proportion of participants AARC triggered favourable “social comparison”
(Festinger, 1954), meaning that they compared themselves with someone
experiencing more negative changes than them. Age Stereotype Internalisation and

148
Chapter 5: Exploring awareness of age-related change among over 50s in the United

Kingdom: Qualitative and quantitative findings from the PROTECT study

Dissociation Theory posits that becoming aware of age-related changes can give
rise to concerns about meeting those negative age stereotypes that individuals have
internalised early in life. Consequently, to counteract negative thoughts and worries
individuals often distance themselves from their peers through favourable social
comparison (Weiss & Kornadt, 2018).
The last higher-order category factors impacting on evaluation of awareness
of changes incudes three categories covering potential sources of bias related to the
use of the AARC questionnaire. Almost one-third of participants disregarded the
impact of their increased age on the changes experienced, yet on average they
reported more AARC losses than participants whose comments did not fit into this
category. This means that several individuals reported high AARC losses despite
attributing their changes to other factors. As the concept of AARC (Diehl & Wahl,
2010) implies that people are both aware of changes and also attribute such
changes to their increased age, scores on the AARC questionnaire on some
occasions may not accurately capture individuals’ AARC. However, it may be that
those participants who answered the open-ended question in the current study were
more likely to be experiencing many age-related losses compared to those who did
not answer the open-ended question. Recent evidence confirms that the tendency to
causally attribute changes to increased age differs among individuals (Levy et al.,
2009a; Rothermund et al., 2020). As attributing changes to increased age enhances
awareness of limited time to live (Burke et al., 2010); the tendency of some
individuals with high levels of AARC losses to attribute changes to causes other than
their age may be an emotional reaction to decrease the negative thoughts
associated with awareness of mortality (Hobfoll, 2002; Leventhal et al., 2016;
Rosenstock, 1974).
Participants’ comments suggest that difficulty when answering the AARC
questionnaire could be reduced by including a “non-applicable” answer option,
specifying the comparative timeframe in the item stem, and for UK use, substitute
some American English words with more common alternatives in British-English.
Finally, some individuals experienced fluctuating levels of awareness of changes
depending on several aspects of their mental and physical health. Micro-longitudinal
research studies also document day to day fluctuation in levels of AARC gains and

149
Chapter 5: Exploring awareness of age-related change among over 50s in the United

Kingdom: Qualitative and quantitative findings from the PROTECT study

losses within the same person (Neupert & Bellingtier, 2017; Zhu & Neupert, 2020).
Daily fluctuation in AARC is related to fluctuation in other variables such as levels of
positive and negative affect (Neupert & Bellingtier, 2017) and control beliefs (Zhang
& Neupert, 2020). As daily fluctuation in AARC influences daily cognitive
performance (O’Brien et al., 2020; Zhu & Neupert, 2020), the fluctuating nature of
AARC gains and losses may limit the predictive accuracy of the AARC
questionnaire. It would therefore be important in future research to control for
variables (e.g. mood) that account for variability in AARC.
Study results may help to explain some of the existing associations of AARC
with health outcomes. First, in line with gerontological literature (Carstensen et al.,
2011; McGrath et al., 2019; Miche et al., 2014; Rubio et al., 2016; Segal et al., 2008)
participants’ comments show that with increasing age negative experiences mainly
relate to physical health (e.g. recent illnesses) whereas positive experiences often
relate to socio-emotional aspects (e.g. increased confidence and experience).
Moreover, those participants that commented on experiencing negative changes
scored higher on levels of AARC losses compared to those who did not commented
on this category. This pattern of results may explain why more AARC losses are
associated with worse mental and physical health whereas more AARC gains are
associated with higher mental health but not with physical health (Brothers et al.,
2019; Sabatini et al., 2020a; Sabatini et al., 2020c).
Second, those participants who engaged in purpose-driven experiences (e.g.
hobbies) reported more AARC gains and fewer AARC losses than those participants
not mentioning any engagement in activities. As high engagement in activities, high
perceived gains, and low perceived losses are all related to better mental health,
levels of engagement in activities may explain why individuals with high gains and
low losses maintain good mental health over time (Chippendale, 2013; Dutt & Wahl,
2018; Grossman & Gruenewald, 2017; Lindsay-Smith et al., 2019; Lindsay-Smith et
al., 2018; Simone & Haas, 2013). Third, the finding that some individuals attribute
changes to increased age whereas others attribute them to illnesses raises the
question of whether attributing changes to different factors leads to different
behavioural reactions. A recent study showed that those individuals who attribute
changes to ageing perceive lower levels of control over the experienced changes

150
Chapter 5: Exploring awareness of age-related change among over 50s in the United

Kingdom: Qualitative and quantitative findings from the PROTECT study

(Rothermund et al., 2020). As a consequence, it may be that these individuals are


more likely to opt for passive coping strategies (e.g. acceptance of changes). Future
studies on self-perceptions of ageing could explore whether the attribution of
changes to increased age (AARC) act as a moderator of the association of self-
perceptions of ageing with engagement in healthy and adaptive behaviours.

Strengths and limitations


This study has several limitations. First, demographic characteristics of participants
answering the AARC open-ended question differed slightly from those of participants
not answering the AARC open-ended question. However, participants answering the
AARC-related open-ended question may be the ones experiencing the highest levels
of losses and, because of this, they may feel the need of clarifying and describing
the losses they are facing. Second, as analyses are based on a self-selected
sample, study participants may have had strong positive or negative feelings when
answering the questionnaire (Ogletree & Katz, 2020). Third, the sample included a
majority of participants who were women, white, and well-educated. This study had
also several strengths including a large sample and the conduct of both qualitative
and quantitative analyses. Participants reported their spontaneous thoughts, and
hence participants' comments are not biased by the research questions or
hypotheses. Moreover, to maximise reliability of qualitative analyses an independent
researcher analysed 10% of the data.

5.6 Conclusion
A wide range of factors impact on perceptions of AARC. Some of these factors
corroborate the categories outlined in the Diehl and Wahl (2010) conceptual
framework of AARC. Others such as engagement in purpose-driven activities or in
activities that distract from age-related thoughts are novel and could be added to the
conceptual framework of AARC. Overall, results support the ability of the AARC
questionnaire to capture subjective experiences of age-related changes in physical,
mental, and cognitive health, engagement in social activities and in healthy and
adaptive behaviours. However, as we identified some issues related to use of the
AARC questionnaire, accuracy in assessing AARC could be increased through some

151
Chapter 5: Exploring awareness of age-related change among over 50s in the United

Kingdom: Qualitative and quantitative findings from the PROTECT study

minor adjustments to the questionnaire. Finally, study results raise new questions for
future research; for instance, the attribution of changes to increased age (AARC)
may act as a moderator of the associations of more general self-perceptions of
ageing with engagement in healthy and adaptive behaviours. It also remains to be
explored whether engagement in activities mediates the associations of higher
AARC gains and lower AARC losses with better health.

152
Chapter 6: Cross-sectional association between objective cognitive performance and
perceived age-related gains and losses in cognition

Chapter 6: Cross-sectional association between objective cognitive


performance and perceived age-related gains and losses in cognition

153
Chapter 6: Cross-sectional association between objective cognitive performance and
perceived age-related gains and losses in cognition

6.1 Summary
The systematic review conducted in Chapter 3 found that there were not studies
exploring the association of AARC with cognition. Chapter 6 therefore examines the
associations of AARC gains and losses with objective cognition. As AARC is the first
concept that makes it possible to assess the coexistence of awareness of positive
and negative age-related changes, this chapter tests whether the association of
AARC with objective cognition is stronger compared to those found between
objective cognition and well-established unidimensional measures of subjective
cognitive decline and subjective ageing. Finally, as psychological variables (Brothers
et al., 2019; Brothers et al., 2017; Kaspar et al., 2019; Mendonça et al., 2016;
Neupert & Bellingtier, 2017; Roehr et al., 2017; Siebert et al., 2020; Tandetnik et al.,
2017) are frequently associated with self-perceptions of cognition and subjective
ageing, Chapter 6 also explores whether variability in AARC cognitive gains and
losses is explained by anxiety, depression, ATOA, subjective age, and/or self-rated
health.

Abstract
Objectives: Although previous research showed that more negative self-perceptions
of cognition and ageing are associated with poorer cognition, indicators used were
unidimensional. We tested whether the bi-dimensional awareness of positive (AARC
gains) and negative (AARC losses) age-related change construct has greater utility
in linking self-perceptions to objective cognition compared to well-established
measures of self-perceptions of cognition and ageing. We examined the associations
of AARC with objective cognition, several psychological variables, and engagement
in cognitive training among three sub-groups of older adults.
Design: Cross-sectional observational study.
Participants: The sample comprised 6,056 cognitively healthy participants (M age=
66.0 years); divided into sub-groups representing middle-age (51-65 years), early old
age (66-75 years), and advanced old age (≥ 76 years).
Measurements: We used an online cognitive battery, a global assessment of AARC,
an assessment of AARC specific to the cognitive domain, and measures of

154
Chapter 6: Cross-sectional association between objective cognitive performance and
perceived age-related gains and losses in cognition

subjective cognitive change, ATOA, subjective age, depression, anxiety, and self-
rated health.
Results: Scores on the AARC measures were more strongly associated with
objective cognition compared to measures of self-perceptions of cognition and
ageing. Higher AARC gains and losses showed small and negligible associations
with poorer cognition across all sub-groups. Higher AARC losses were related to
higher levels of depressive and anxiety symptoms, more negative ATOA, an older
subjective age, and poorer self-rated health, but not to engagement in cognitive
training.
Conclusion: Assessing both positive and negative self-perceptions of cognition and
ageing is important when linking self-perceptions to cognitive functioning. Objective
cognition is one of the many variables -together with psychological variables- related
to perceived cognitive losses.

155
Chapter 6: Cross-sectional association between objective cognitive performance and

perceived age-related gains and losses in cognition

6.2 Introduction
Global estimations suggest that 50 million people are living with dementia (World
Health Organization, 2020a); a condition that creates a high social and economic
burden. It is therefore important to identify individuals with poorer cognition as they
may benefit the most from timely interventions aimed at preventing pathological
cognitive decline. Self-perceptions of cognition and subjective ageing may be one
way to identify such individuals. Self-reported cognitive limitations may be correlated
with poorer cognition and can often precede pathological cognitive decline (Amariglio
et al., 2012; Jessen et al., 2014; Mendonça et al., 2016; Reisberg & Gauthier, 2008;
Reisberg et al., 2008).
Subjective age (Barrett, 2003) and attitudes toward own ageing (ATOA;
Lawton, 1975) are two examples of concepts for measuring subjective ageing.
Subjective age captures how old individuals feel, which can differ from their
chronological age (Barrett, 2003) and ATOA assess older individuals’
evaluations of the changes happening in their lives (Lawton, 1975). Generally,
more positive ATOA and/or feeling younger than one’s chronological age are
associated with better cognition (Seidler & Wolff, 2017) and consequently to a
lower risk of dementia (Levy et al., 2016; Siebert et al., 2018) even among
individuals with higher risk of developing dementia (e.g. APOEƹ4 carriers;
Levy et al., 2018). Better cognitive functioning of those with more positive
ATOA and younger subjective age may be due to these individuals
experiencing less stress (Stephan et al., 2015), better mental and physical
health (Bryant et al., 2012), and being more engaged in preventive behaviours
(Hess, 2006). These specific factors have been identified as being protective
against cognitive decline (Anstey, 2013). Despite research supporting
connections between more positive subjective ageing and better cognition;
many available measures (e.g. ATOA and subjective age; Barrett, 2003;
Lawton, 1975) assess subjective ageing in a global way and do not capture
specific perceived age-related changes in cognition. Moreover, existing
evidence relies on bimodal (either positive or negative) indicators of subjective
ageing. These may provide an overly simplified picture. With ageing, for
example, some cognitive abilities generally decline (e.g. memory) but others

156
Chapter 6: Cross-sectional association between objective cognitive performance and
perceived age-related gains and losses in cognition

can improve (e.g. knowledge; Christensen, 2001). Therefore, when exploring the
associations of subjective ageing with cognitive abilities, it is important to assess the
coexistence of perceived age-related gains and losses in cognition.
The construct of awareness of age-related change (AARC) refers to
individuals’ awareness that their behaviour, performance, and/or way of experiencing
life has changed due to their increased age (Diehl & Wahl, 2010). AARC is the first
concept assessing the coexistence of perceived gains and losses across five life
domains - health and physical functioning, cognition, interpersonal relationships,
socio-cognitive and socio-emotional functioning, and lifestyle and life engagement.
Existing concepts (e.g. subjective cognitive decline; Jessen et al., 2014) capture self-
reported cognitive limitations that individuals attribute to pathology whereas
perceived losses in cognition measure cognitive limitations due to ageing. This
potentially makes it possible to detect perceptions of even subtle cognitive
limitations.
AARC is assessed via self-administered questionnaires; items capturing
perceived gains and losses derive from qualitative interviews and focus groups in
which middle-aged and older individuals reported both positive and negative aspects
of ageing (Brothers et al., 2019; Miche et al., 2014; Wahl et al., 2013). From both the
50-item (Brothers et al., 2019) and shorter 10-item (AARC-10 SF; Kaspar et al.,
2019) versions of the AARC questionnaire it is possible to obtain two global scores
representing AARC gains and losses across life domains. In addition, the 50-item
version of the questionnaire makes it possible to obtain ten domain-specific scores;
one for gains and one for losses in each of the five AARC life domains including the
cognitive domain (AARC-50 cognitive functioning subscale; Brothers et al., 2019).
The coexistence of AARC gains and losses in cognition is possible as they relate to
different aspects of cognition: gains capture perceived improvements in knowledge,
wisdom, and/or reflexivity whereas losses capture perceived limitations in processing
speed, memory, and/or mental capacity. Perceptions of cognitive gains are quite
independent from perceptions of cognitive losses (Sabatini et al., 2020b).
Although the current study relies on cross-sectional data, it adds important
facets to previous research as it explores for the first time the relation between
awareness of age-related change (AARC; Diehl & Wahl, 2010) and objective

157
Chapter 6: Cross-sectional association between objective cognitive performance and
perceived age-related gains and losses in cognition

cognitive performance. First, as the measures used in this study make it


possible to obtain both a global assessment of AARC and an assessment of
AARC specific to the cognitive domain; the current study is the first to test
whether an assessment of AARC gains and losses specific to the cognitive
domain is more strongly related to objective cognitive functioning, compared
to a global assessment of AARC and to three well-established bimodal
measures of self-perceptions of cognition and ageing (subjective cognitive
changes, ATOA, subjective age). Second, given its large sample, this study
makes it possible to robustly examine whether the associations of AARC
gains and losses in cognition with cognitive functioning vary among midlife,
early old age, and advanced old age. Third, this is the first study to explore
whether depression, anxiety, ATOA, subjective age, self-rated health, or level
of engagement in computerised cognitive training explain variability in levels
of AARC in cognition across age sub-groups. Psychological variables are
often associated with subjective ageing (Brothers et al., 2019; Brothers et al.,
2017; Kaspar et al., 2019; Mendonça et al., 2016; Neupert & Bellingtier, 2017;
Roehr et al., 2017; Sabatini et al., 2020a; Siebert et al., 2020; Tandetnik et al.,
2017) and engagement in online cognitive training has recently been linked to
more positive self-perceptions of cognition (Sullivan et al., 2020).
We hypothesise that the assessment of AARC gains and losses in the
cognitive domain will be more strongly associated with cognitive functioning
compared to a global assessment of AARC gains and losses, subjective
cognitive change, ATOA, and subjective age. Second, we hypothesise that
more AARC gains and fewer perceived losses in cognition are associated with
better cognitive performance across all sub-groups. Third, we hypothesise
that more severe depression and/or anxiety, negative ATOA, an older
subjective age, poorer self-rated health, or less engagement in computerised
cognitive training are associated with fewer perceived gains and more
perceived losses in cognition and that these associations become stronger in
older age.

158
Chapter 6: Cross-sectional association between objective cognitive performance and
perceived age-related gains and losses in cognition

6.3 Methods
Study design and participants
With the exception of frequency of cognitive training, this study used secondary data
collected from the ongoing PROTECT study (Platform for Research Online to
investigate Genetics and Cognition in Ageing; https://www.protectstudy.org.uk)
between November 2018 and March 2019. Frequency of cognitive training was
based on PROTECT data collected between 2015 and 2019. PROTECT participants
are UK residents, English speakers, aged 50 years or over, have access to the
internet, and did not have a clinical diagnosis of dementia at baseline (2015). To
recruit participants, the PROTECT study was publicised nationwide and among
existing research cohorts of older adults (Exeter 10,000
https://exetercrfnihr.org/about/exeter-10000/; Join Dementia Research
https://www.joindementiaresearch.nihr.ac.uk/; and Brains for Dementia Research
https://bdr.alzheimersresearchuk.org). At baseline, participants provided informed
online consent though the PROTECT platform. The PROTECT study has ethical
approval from the London Bridge NHS Research Ethics Committee and Health
Research Authority (Ref: 13/LO/1578). Ethical approval for the conduct of data
analyses was obtained from the ethics committee at the University of Exeter School
of Psychology (Ref: eCLESPsy000603 v1.0).
In PROTECT, participants undertake a self-administered annual assessment
through the PROTECT platform. Out of the 14,882 participants that took part in the
2019 annual assessment, 8,680 participants were excluded from the current study
as they did not complete either the cognitive tests or the AARC questionnaires. Even
though participants with possible mild cognitive impairment where included in the
validation of the AARC-10 SF and AARC-50 cognitive functioning subscale (Chapter
4) - due to the AARC questionnaires being useful to understand the losses perceived
by individuals with mild cognitive impairment - in the current study we excluded
individuals (N = 146 individuals) deemed to have pathological cognitive decline (mild
cognitive impairment or dementia). This is because individuals with pathological
cognitive decline can be inaccurate when evaluating their cognition (Lehrner et al.,
2015). Moreover, the current study investigated the association of AARC losses with
cognitive performance with the intent to explore whether the AARC questionnaire is

159
Chapter 6: Cross-sectional association between objective cognitive performance and
perceived age-related gains and losses in cognition

useful to identify those individuals that, because of poorer cognitive


functioning, are at higher risk of experiencing pathological cognitive decline.
These individuals could therefore benefit from interventions aiming to prevent
mild cognitive impairment and dementia. Individuals were deemed to have
mild cognitive impairment when they scored 1.5 standard deviations below the
mean study sample score in two or more cognitive tasks.
The cognitive tests were completed on a separate day (within two
months) to that on which participants answered the AARC questionnaires.
The current study sample resulted in 6,056 participants aged 51.4 to 95.9 (Age, M=
66.0 years, SD= 7.0 years); of which 76.2% were women and 98.6% were white.
Among study participants, 3,111 (61.5%) were middle-aged (51-65 years); 2,473
(48.9%) were in early old age (66-75 years), and 472 (9.3%) were in advanced old
age (≥76 years). The majority (68.8%) of middle-aged participants was working,
whereas a minority of early old (17.0%) and advanced old (6.8%) participants was
working. Participants excluded from study analyses had comparable demographic
variables, mental and self-rated health to the study sample but a higher proportion
was working.

Measures
Indicators of self-perceptions of ageing and cognition
To assess perceived gains and losses in the cognitive domain we used the AARC-50
cognitive functioning subscale taken from the 50-item version of the questionnaire
(Brothers et al., 2019). Participants rate how much each of the ten items (reported in
Appendix A) applies to them on a five-point scale (1= not at all, 2= a little bit, 3=
moderately, 4= quite a bit, and 5= very much). Scores can be obtained for AARC
gains and losses in cognition by summing the five items falling into the respective
subscale (score range: 5-25). Higher scores indicate higher AARC gains and losses
in cognition. The AARC-50 cognitive functioning subscale is reliable in the UK
population, with Cronbach’s αs of .86 for gains and of .88 for losses (Sabatini et al.,
2020b).
To assess global perceptions of gains and losses the AARC-10 SF
(Kaspar et al., 2019) was used as it has been shown to be reliable in the UK

160
Chapter 6: Cross-sectional association between objective cognitive performance and
perceived age-related gains and losses in cognition

population, with Cronbach’s αs of .77 for gains and of .80 for losses (Sabatini et al.,
2020b). The questionnaire contains ten items (Appendix A), five for gains and five for
losses, describing the five different life domains (health and physical functioning,
cognition, interpersonal relationships, socio-cognitive and socio-emotional
functioning, and lifestyle and life engagement). Participants rate how much each item
applies to them on a five-point scale (“not at all” to “very much”). Separate scores,
ranging from five to twenty-five, can be obtained for AARC gains and losses by
summing the five items falling into the respective subscale. Higher scores indicate
higher AARC gains and losses, respectively, across life domains.
The Informant Questionnaire on Cognitive Decline in the Elderly short-form
was used (IQCODE-Self; Jorm & Jacomb, 1989) to assess subjective cognitive
change over the last ten years. It includes sixteen items that are answered on a five-
point scale (1= much improved, 2= a bit improved, 3= not much change, 4= a bit
worse, and 5= much worse) The final score is the mean of the item scores; higher
scores indicate subjective cognitive decline whereas lower scores indicate subjective
cognitive improvement. Cronbach’ α for the IQCODE-Self in this sample is .86.
We used the ATOA scale (taken from the Philadelphia Geriatric Center
Morale Scale; Lawton, 1975) to assess attitudes toward own ageing. The score
ranges from zero to five; lower scores indicate more negative ATOA whereas higher
scores indicate more positive ATOA. To assess subjective age participants were
asked to write the age (in years) they feel most of the time (Barrett, 2003). A positive
value indicates a younger subjective age, whereas a negative value indicates an
older subjective age. To assess self-rated health, participants were asked to rate
their health on a four-point scale ("Excellent" to "poor"; Ware & Sherbourne, 1992).

Cognitive functioning
To assess cognitive functioning we used the PROTECT Cognitive Test Battery
(Corbett et al., 2015) which includes four tests: (1) Self-Ordered Search (Owen et al.,
1990) assessing spatial working memory (0-20); (2) Grammatical Reasoning
(Baddeley, 1968) assessing verbal reasoning (0-no upper limit); (3) Paired Associate
Learning (Owen et al., 1993) assessing visual episodic memory (0-16); and (4) Digit
Span (Huntley et al., 2017) assessing verbal working memory (0-20). For each test a

161
Chapter 6: Cross-sectional association between objective cognitive performance and
perceived age-related gains and losses in cognition

score is obtained by subtracting the number of errors from the number of correct
answers; a higher score indicates better performance on the test. For grammatical
reasoning, the score has no upper limit. This is because the number of trials within
the allocated time for the test varies depending on how rapidly participants respond
during the test. In PROTECT participants have access to online brain training games
though the PROTECT platform; the number of times participants played any brain
training game between 2015 and 2019 was used as an indicator of frequency of
engagement in cognitive training.

Mental health
The Patient Health Questionnaire-9 (Kroenke et al., 2001) was used to assess
depression; higher scores indicate more severe depression (9-36). Cronbach’s α on
the measure for this sample is .76. The Generalised Anxiety Disorder-7 (Spitzer et
al., 2006) was used to assess anxiety symptoms; higher scores indicate more severe
anxiety symptoms (7-28). Cronbach’s α on the measure for this sample is .76.

Demographic information
Demographic information collected included age, sex, employment status (employed
versus not employed), and education level (secondary education; post-secondary
education; vocational qualification; undergraduate degree; post-graduate degree;
doctorate).

Data-analytical design
For subjective age a proportional discrepancy score was calculated by subtracting
participants’ subjective age from their chronological age and dividing this difference
score by participants’ chronological age. To explore mean differences in the levels of
AARC gains and losses in cognition and across life domains between men and
women we conducted t-tests. To estimate the associations of participants’ scores on
the four objective cognitive tests (outcomes) with their scores on an assessment of
AARC gains and losses specific to the cognitive domain, a global assessment of
AARC gains and losses, subjective cognitive change, ATOA, and subjective age, we
fitted path analysis models. For the two assessments of AARC gains and losses we

162
Chapter 6: Cross-sectional association between objective cognitive performance and
perceived age-related gains and losses in cognition

included both gains and losses as predictors within the same model. Scores on
objective cognitive tests were allowed to intercorrelate. Sex, education, employment
status, depression, anxiety, and frequency of cognitive training were included in the
path analysis models as covariates. We treated depression, anxiety, and frequency
of cognitive training as covariates as they likely impact on AARC and on objective
cognition (Alexopoulos et al., 1993; Kassem et al., 2017). As the directions of all
these associations have not been empirically investigated, we also tested a model
excluding depression, anxiety, and frequency of cognitive training from covariates.
From path analysis models it is not possible to obtain partial coefficients of
determination. However, as regression coefficients obtained with multiple linear
regressions having gains and losses as predictors of scores on the cognitive tests
led to similar results to those obtained with path analysis models, we reported results
for the multiple regressions including information about the coefficients of
determination.
To examine whether the strength of the associations of perceived gains and
losses in cognition (predictors) with cognitive performance varies among age sub-
groups we estimated Pearson’s r correlation coefficients and three separate path
analysis models for individuals in middle-age, early old age, and advanced old age.
Sub-group analyses for men and women were not conducted as analysis of
measurement invariance for the AARC questionnaire (see Chapter 4) suggests that
scores for men and women on the AARC questionnaire cannot be compared.
However, we controlled for sex differences in the analyses. The Comparative Fit
index (CFI), the Tucker-Lewis index (TLI), the Root Mean Square Error of
Approximation (RMSEA), and the Standardised Root Mean Square Residual
(SRMR) were used to assess model fit. Values considered acceptable were CFI and
TLI > .90, RMSEA < .08 (95% CI: 0; .08), and SRMR < .06 (Byrne, 2012).
To test whether more depression and/or anxiety, more negative ATOA, an
older subjective age, and poorer self-rated health (predictors) are associated with
fewer gains in cognition and more losses in cognition across age sub-groups, we
fitted simple and multiple regressions. Finally, we estimated simple linear
regressions with frequency of cognitive training as the predictor of AARC gains or
losses in cognition across the age sub-groups. Standardised betas were used as

163
Chapter 6: Cross-sectional association between objective cognitive performance and
perceived age-related gains and losses in cognition

indicators of effect size. Associations ≤ .09 were considered negligible, .10–


.29 small, .30–.49 moderate and ≥.50 large (Cohen, 1988). Analyses were
conducted in STATA version 16 (StataCorp, 2017).

6.4 Results
Descriptive data
On average, participants perceived “a little bit” of AARC gains (M= 13.57; SD= 4.37)
and losses (M=10.10; SD= 3.60) specific to the cognitive domain; “quite a bit” of
AARC gains (M= 17.64; SD= 3.92) and “a little bit” of AARC losses (M= 9.81; SD=
3.21) in the global assessment of AARC. Men and women had significantly different
levels of AARC gains (M= 12.73; SD= 4.31 and M= 13.83; SD= 4.35 respectively; p
<.001) and losses (M= 10.73; SD= 3.82 and M= 9.91; SD= 3.50 respectively; p
<.001) in cognition. Men and women also had significantly different levels of AARC
gains (M= 16.77; SD= 4.01 and M= 17.91; SD= 3.85 respectively; p <.001) and
losses (M= 10.32; SD= 3.34 and M= 9.65; SD= 3.15 respectively; p <.001) across life
domains. On average participants reported subjective cognitive decline (M= 3.09;
SD= 0.22) and mixed ATOA (M= 2.62; SD= 0.79); felt 17% younger than their
chronological age, and had minimal levels of depressive (M= 11.41; SD= 2.89) and
anxiety (M= 8.42; SD= 2.42) symptoms. Frequency of engagement in cognitive
training varied greatly among participants (0-15, 942 times). A high proportion of
participants perceived their health as good (54.6%) or excellent (30.7%).
Characteristics for the overall study sample, sub-samples, and participants excluded
form analyses are reported in Table 6.1.

164
Chapter 6: Cross-sectional association between objective cognitive performance and perceived age-related gains and losses in

cognition

Table 6.1. Descriptive statistics of demographic variables and main study variables for the study sample, study sub-samples, and
participants not included in the study.

Variables Study sample Participants not Participants in Participants in Participants in p-value


(N= 6,056) included in the middle-age early old age advanced old
study (N= (N= 3,111) (N= 2,473) age (N= 472)
8,826)
Age (years), M (SD; range) 66.0 (6.9; 65.05 (7.2; 60.5 (3.5; 70.3 (2.7; 79.9 (3.5; Not applicable
51.4, 95.9) 50.2, 103.5) 51, 65) 66, 75) 76, 95)
Sex (Women %) 4,615 (76.2) 6,532 (75.1) 2,532 (81.4) 1,789 (72.3) 294 (62.3) <.001
Education level (%)
Secondary 812 (13.4) 1,391 (16.0) 336 (10.8) 386 (15.6) 90 (19.1) <.001
Post-secondary 682 (11.3) 1,048 (12.0) 355 (11.4) 269 (10.9) 58 (12.3)
Vocational qualification 1,209 (20) 1,763 (20.3) 596 (19.2) 519 (20.6) 103 (21.8)
Undergraduate degree 2,062 (34.1) 2,780 (32.0) 1,151 (37.0) 768 (31.1) 143 (30.3)
Post-graduate degree 1,506 (17.4) 1,459 (16.8) 580 (18.6) 431 (17.4) 45 (9.5)
Doctorate 235 (3.9) 260 (3.0) 93 (3.0) 109 (4.4) 33 (7.0)
Employment status (Employed %) 2,496 (42.2) 4,233 (50.3) 2,046 (68.8) 418 (17.0) 32 (6.8) <.001
AARC gains in cognition, M (SD) 13.6 (4.4) Not applicable 13.9 (4.4) 13.3 (4.3) 13.1 (4.3) <.001
AARC losses in cognition, M (SD) 10.1 (3.6) Not applicable 9.7 (3.6) 10.3 (3.4) 11.8 (4.0) <.001
AARC gains across life domains, M (SD) 17.6 (3.9) Not applicable 17.8 (4.0) 17.6 (3.9) 17.3 (3.8) .013
AARC losses across life domains, M 9.8 (3.2) Not applicable 9.3 (3.1) 10.1 (3.1) 12 (3.6) <.001
(SD)
Self-ordered search, M (SD) 7.8 (2.5) Not applicable 8.1 (2.5) 7.5 (2.5) 6.7 (2.5) <.001

165
Chapter 6: Cross-sectional association between objective cognitive performance and perceived age-related gains and losses in
cognition

Paired associate learning, M (SD) 4.7 (.9) Not applicable 4.9 (.91) 4.7 (.9) 4.3 (.9) <.001
Grammatical reasoning, M (SD) 37.8 (10.2) Not applicable 39.8 (10.4) 36.6 (9.6) 31.8 (9.3) <.001
Digit span, M (SD) 7.6 (1.5) Not applicable 7.8 (1.5) 7.6 (1.5) 7.2 (1.5) <.001
Depression, M (SD) 11.4 (2.9) 11.8 (3.4) 11.7 (3.1) 11.1 (2.6) 11.2 (2.6) <.001
Anxiety, M (SD) 8.4 (2.4) 8.7 (2.8) 8.7 (2.6) 8.2 (2.2) 8.0 (2) <.001
Self-rated health (%)
Poor 120 (2.0) 89 (2.7) 65 (2.1) 49 (2.0) 6 (1.3) .557
Fair 765 (12.7) 453 (13.5) 396 (12.8) 305 (12.4) 64 (13.6)
Good 3,302 (54.6) 1,774 (53.0) 1,663 (53.5) 1,377 (55.8) 262 (55.6)
Excellent 1,858 (30.7) 1,029 (30.8) 982 (31.6) 737 (29.9) 139 (29.5)
Subjective cognitive change, M (SD) 3.1 (.2) Not applicable 3.08 (.2) 3.1 (.2) 3.1 (.3) <.001
Attitudes toward own ageing, M (SD) 2.6 (.8) Not applicable 2.61 (.8) 2.6 (.8) 2.6 (.9) .113
Subjective age, M (SD) .2 (.1) Not applicable .17 (.2) .17 (.1) .2 (.2) .688
Frequency of cognitive training over 5 399.7 (993.1; Not applicable 307.1 (759.6; 474.5 (1125; 617.9 (1446; <.001
years; M (SD; range) 0, 15942) 0, 11602) 0, 13711) 0, 15942)
Note: p-values indicate whether mean scores in the specific variable are statistically different among the three age sub-groups. P-
values have been calculated with ANOVA for continuous variables and with x2 tests for categorical variables.

166
Chapter 6: Cross-sectional association between objective cognitive performance and

perceived age-related gains and losses in cognition

AARC gains and losses in the cognitive domain, global levels of AARC gains
and losses, subjective cognitive change, ATOA, and subjective age as
predictors of cognitive functioning
The associations of AARC gains and losses in the cognitive domain, global levels of
AARC gains and losses, subjective cognitive change, ATOA, and subjective age with
cognitive performance are reported in Tables 6.2 and 6.3 and Figure 6.1a-e. Overall,
both AARC gains and losses in cognition were associated with poorer cognitive
performance. AARC gains and losses in cognition and global levels of AARC gains
and losses showed the strongest associations with cognitive performance, even
though these associations were either small or negligible. The global assessment of
AARC gains and losses explained slightly more variance in cognitive functioning
than the assessment of AARC gains and losses specific to the cognitive domain.

167
Chapter 6: Cross-sectional association between objective cognitive performance and perceived age-related gains and losses in

cognition

Table 6.2. Path analysis model exploring AARC gains and losses in cognition, AARC gains and losses across life domains,
subjective cognitive change, ATOA, and subjective age as predictors of cognition in the overall study sample while controlling for
sex, education, employment status, depressive and anxiety symptoms, and frequency of cognitive training.
ß (95% CI); p- value
Predictors Self-ordered search Grammatical reasoning Paired associate learning Digit span
AARC gains in cognition -.06 (-.09, -.04); <.001 -.09 (-.11, -.07); <.001 -.02 (-.05, .00); .083 -.04 (-.07, -.02); .001
AARC losses in cognition -.05 (-.08, -.03); <.001 -.12 (-.14, -.09); <.001 -.08 (-.11, -.06); <.001 -.06 (-.09, -.04); .001
2
RMSEA (90%CI): .07 (.05, .09); CFI: .99; TLI: .53; SRMR: .01; R : 19%

AARC gains across life domains -.04 (-.07, -.02); .002 -.04 (-.06, -.01); .002 -.02 (-.04, .01); .160 -.01 (-.03, .02); .648
AARC losses across life domains -.07 (-.10, -.04); <.001 -.12 (-.14, -.09); <.001 -.08 (-.11, -.05); <.001 -.09 (-.11, -.06); <.001
2
RMSEA (90%CI): .06 (.04, .09); CFI: .99; TLI: .62; SRMR: .01; R : 21%

Subjective cognitive change .02 (-.01, .05); .129 -.004 (-.03, .02); .770 -.03 (-.06, -.00); .045 -.01 (-.04, .02); .447
2
RMSEA (90%CI): .00 (.00, .04); CFI: 1.0; TLI: 1.0; SRMR: .00; R : 12%

Attitudes toward own ageing .01 (-.01, .04); .409 .01 (-.02, .03); .519 -.0001 (-.03, .03); .996 .001 (-.03, .03); .960
2
RMSEA (90%CI): .00 (.00, .00); CFI: 1.0; TLI: 1.0; SRMR: .00; R : 11%

Subjective age .01 (-.02, .03); .665 .02 (-.00, .05); .070 .04 ( .01, .06); .004 .03 ( .00, .05); .030
2
RMSEA (90%CI): .00 (.00, .00); CFI: 1.0; TLI: 1.0; SRMR: .00; R : 9%

Note: RMSEA= Root mean square error of approximation. CFI= Comparative fit index. TLI= Tucker-Lewis index. SRMR= Standard
root mean square residual. R2= R-squared/coefficient of determination.

168
Chapter 6: Cross-sectional association between objective cognitive performance and perceived age-related gains and losses in
cognition

Confounders: Gender,
education level,
Confounders: Gender, employment status,
education level, depression, anxiety,
employment status, frequency of cognitive
depression, anxiety, training
frequency of cognitive
training Self-ordered
p< .0
01 search
06
ß = -.

ß= -.09 p< .001 Grammatical


AARC gains
in cognition ß = -.
reasoning
1 02 p=
.083
.00 .001
p< 2p
<
- .05 ß= -.1 Paired
ß = .001
AARC ß= -.08 p< associate
ß=
losses in ß = -. -. 0
4p learning
06 p =.
cognition = .001 00
1
Study sample N = 6,056
RMSEA (90%CI): .07 (.05, .09)
Digit span CFI: .99
Confounders: Gender, education TLI: .53
level, employment status, SRMR: .01
depression, anxiety, frequency of R2: 19%
cognitive training

Figure 6.1a. Path analysis model exploring AARC gains and losses in cognition as predictors of cognition in the overall study
sample while controlling for sex, education, employment status, depressive and anxiety symptoms, and frequency of cognitive
training.

169
Chapter 6: Cross-sectional association between objective cognitive performance and perceived age-related gains and losses in
cognition

Confounders: Gender,
education level,
Confounders: Gender, employment status,
education level, depression, anxiety,
employment status, frequency of cognitive
depression, anxiety, training
frequency of cognitive
training Self-ordered
4 p=
.002 search
ß = -. 0

AARC gains ß= -.04 p= .002


Grammatical
across life reasoning
ß = -.
domains 00
1 02 p=
p<. . 001 .160
7 p <
- .0 - . 12
ß= ß= Paired
AARC losses ß= -.08 p< .0
01
associate
across life ß=
-. 0 learning
ß = -. 1p
domains 09 p
< .00
1
=.
64
8
Study sample N = 6,056
RMSEA (90%CI): .06 (.04, .09)
Digit span CFI: .99
Confounders: Gender, TLI: .62
education level, employment SRMR: .01
status, depression, anxiety, R2: 21%
frequency of cognitive
training

Figure 6.1b. Path analysis model exploring AARC gains and losses across life domains as predictors of cognition in the overall
study sample while controlling for sex, education, employment status, depressive and anxiety symptoms, and frequency of
cognitive training.

170
Chapter 6: Cross-sectional association between objective cognitive performance and perceived age-related gains and losses in
cognition

Confounders:
Gender, education
Confounders: Gender, level, employment
education level, status, depression,
employment status, anxiety, frequency of
depression, anxiety, cognitive training
frequency of cognitive
training Self-ordered
search
p= .129
.02
ß=
4 p= .7
70 Grammatical
Subjective ß= -.00
reasoning
cognitive ß= -.03 p=
.045
decline
ß=
-. 0
Paired
1p
=.
447 associate
learning
Confounders: Gender,
education level,
employment status, Study sample N= 6,056
depression, anxiety, RMSEA (90%CI): .00 (.00, .00)
Digit span CFI: 1.0
frequency of cognitive
training TLI: 1.0
SRMR: .00
R2: 12%

Figure 6.1c. Path analysis model exploring subjective cognitive decline as predictor of cognition in the overall study sample while
controlling for sex, education, employment status, depressive and anxiety symptoms, and frequency of cognitive training.

171
Chapter 6: Cross-sectional association between objective cognitive performance and perceived age-related gains and losses in
cognition

Confounders:
Gender, education
Confounders: Gender, level, employment
education level, status, depression,
employment status, anxiety, frequency of
depression, anxiety, cognitive training
frequency of cognitive
training Self-ordered
search
409
p= .
= .01
ß

ß= .01 p= .519
Grammatical
Attitudes reasoning
toward own ß= -.000
1 p= .99
ageing 6
ß= Paired
.00
1p
= .9 associate
60
learning
Confounders: Gender,
education level,
employment status, Study sample N= 6,056
depression, anxiety, RMSEA (90%CI): .00 (.00, .00)
Digit span CFI: 1.0
frequency of cognitive
training TLI: 1.0
SRMR: .00
R2: 11%

Figure 6.1d. Path analysis model exploring attitudes toward own ageing as predictor of cognition in the overall study sample while
controlling for sex, education, employment status, depressive and anxiety symptoms, and frequency of cognitive training.

172
Chapter 6: Cross-sectional association between objective cognitive performance and perceived age-related gains and losses in
cognition

Confounders:
Gender, education
Confounders: Gender, level, employment
education level, status, depression,
employment status, anxiety, frequency of
depression, anxiety, cognitive training
frequency of cognitive
training Self-ordered
search
665
1 p= .
ß = .0

ß= .02 p
= .070 Grammarical
Subjective reasoning
age ß= .04
p= .004

ß= Paired
.03
p=
.03 associate
0
learning
Confounders: Gender,
education level,
employment status, Study sample N= 6,056
depression, anxiety, RMSEA (90%CI): .00 (.00, .00)
Digit span CFI: 1.0
frequency of cognitive
training TLI: 1.0
SRMR: .00
R2: 9%

Figure 6.1e. Path analysis model exploring subjective age as predictor of cognition in the overall study sample while controlling for
sex, education, employment status, depressive and anxiety symptoms, and frequency of cognitive training.

173
Chapter 6: Cross-sectional association between objective cognitive performance and perceived age-related gains and losses in
cognition

Table 6.3. Multiple linear regressions with AARC gains and losses in cognition and AARC gains and losses across life domains as
predictors of scores on the cognitive tasks in the overall study sample.

AARC-50 cognitive functioning subscale


Self-ordered search Grammatical reasoning Paired associate learning Digit span
Predictors ß (95% CI); p- value Partial ß (95% CI); p- value Partial ß (95% CI); p- value Partial ß (95% CI); p- value Partial
2 2 2
R R R R2
AARC gains in -.07 (-.09, -.04); <.001 .04% -.10 (-.12, -.07); <.001 1% -.02 (-.05, -.00); .030 .01% -.05 (-.07, -.02); <.001 .02%
cognition
AARC losses -.06 (-.08, -.03); <.001 .03% -.12 (-.15, -.10); <.001 1% -.09 (-.11, -.06); <.001 1% -.07 (-.09, -.04); <.001 .04%
in cognition
AARC-10 SF
Self-ordered search Grammatical reasoning Paired associate learning Digit span
Predictors ß (95% CI); p- value Partial ß (95% CI); p- value Partial ß (95% CI); p- value Partial ß (95% CI); p- value Partial
2 2 2
R R R R2
AARC gains -.04 (-.07, -.02); <.001 .02% -.05 (-.06, -.01); <.001 .02% -.02 (-.04, .01); .074 .01% -.01 (-.04, .01); .388 0%
across life
domains
AARC losses -.07 (-.10, -.05); <.001 1% -.12 (-.14, -.09); <.001 1% -.08 (-.11, -.06); <.001 1% -.09 (-.11, -.06); <.001 1%
across life
domains
Note: Sex, education, employment status, depression, anxiety, and frequency of cognitive training are included as covariates in the
regression models.

174
Chapter 6: Cross-sectional association between objective cognitive performance and

perceived age-related gains and losses in cognition

Associations of AARC gains and losses in the cognitive domain with


cognitive performance across age sub-groups
Estimates for correlations, path analysis models, and multiple linear regressions
exploring the associations of AARC gains and losses in cognition with scores on
cognitive tasks across age sub-groups are reported in Tables 6.4-6.8 and Figure
6.2a-c. Overall, higher AARC gains in cognition were associated with poorer scores
on cognitive tests among participants in middle-age and early old age. Although
higher AARC gains were also related to poorer scores on cognitive tasks in
advanced old age, these associations were not statistically significant. Higher AARC
losses in cognition were associated with poorer performance on most cognitive
tasks, especially in grammatical reasoning. These associations were consistent
across all age sub-groups but strongest in size in early old and advanced old age.
For both AARC gains and losses in cognition associations with scores on the
cognitive tasks were either negligible or small; even though associations were
slightly stronger in size for AARC losses.

175
Chapter 6: Cross-sectional association between objective cognitive performance and perceived age-related gains and losses in

cognition

Table 6.4. Path analysis model exploring AARC gains and losses in cognition as predictors of cognition in the three age sub-groups
while controlling for sex, education, employment status, depressive and anxiety symptoms, and frequency of cognitive training.

Predictors ß (95% CI); p- value


Participants aged 51 to 65
Self-ordered search Grammatical reasoning Paired associate learning Digit span
AARC gains in cognition -.08 (-.12, -.05); <.001 -.08 (-.11, -.04); <.001 -.05 (-.09, -.02); .004 -.05 (-.09, -.02); .004
AARC losses in cognition -.05 (-.08, -.01); .013 -.09 (-.12, -.05); <.001 -.04 (-.07, .00); .052 -.04 (-.08, -.00); .038
2
RMSEA (90%CI): .07 (.05, .11); CFI: .99; TLI: .40; SRMR: .01; R : 16%
Participants aged 66 to 75
Self-ordered search Grammatical reasoning Paired associate learning Digit span
AARC gains in cognition -.05 (-.09, -.02); .007 -.13 (-.17, -.09); <.001 .001 (-.04, .04); .963 -.04 (-.08, -.001); .048
AARC losses in cognition -.02 (-.06, .02); .370 -.10 (-.14, -.06); <.001 -.09 (-.14, -.05); <.001 -.07 (-.11, -.03); .001
RMSEA (90%CI): .07 (.04, .11); CFI: .99; TLI: .35; SRMR: .01; R2: 17%
Participants aged 76 and over
Self-ordered search Grammatical reasoning Paired associate learning Digit span
AARC gains in cognition -.04 (-.13, .05); .350 -.08 (-.17, .01); .066 -.003 (-.09, .09); .946 -.05 (-.14, .04); .268
AARC losses in cognition .02 (-.09, .12); .767 -.20 (-.30, -.10); <.001 -.10 (-.20, -.0002); .049 -.05 (-.15, .05); .362
RMSEA (90%CI): .00 (.00, .12); CFI: 1.0; TLI: 1.0; SRMR: .00; R2: 27%
Note: RMSEA= Root mean square error of approximation. CFI= Comparative fit index. TLI= Tucker-Lewis index. SRMR= Standard
root mean square residual. R2= R-squared/coefficient of determination.

176
Chapter 6: Cross-sectional association between objective cognitive performance and perceived age-related gains and losses in

cognition

Confounders: Gender,
education level,
employment status,
Confounders: Gender, depression, anxiety,
education level, frequency of cognitive
employment status, training
depression, anxiety,
frequency of cognitive
training Self-ordered
=p<.0.0
201 search
-.408p
=-.0
ß=
ß

AARC gains ß= -.08 p< .001 Grammatical


in cognition reasoning
3 ß= -.0
5 p=
= .01 < .001 .004
5 p 09 p
.0 - .
ß=
- ß = Paired
52
AARC ß= -.04 p= .0 associate
losses in ß = -.
ß=
-. 0 learning
04 p= 5p
cognition .038 =.
00
4
Participants aged 51 to 65
years N = 3,111
Digit span RMSEA (90%CI): .07 (.05, .11)
Confounders: Gender,
education level, CFI: .99
employment status, TLI: .40
depression, anxiety, SRMR: .01
R2: 16%
frequency of cognitive
training

Figure 6.2a. Path analysis model exploring AARC gains and losses as predictors of cognition in the sub-sample of participants
aged 51 to 65 years while controlling for sex, education, employment status, depressive and anxiety symptoms, and frequency of
cognitive training.

177
Chapter 6: Cross-sectional association between objective cognitive performance and perceived age-related gains and losses in

cognition

Confounders: Gender,
education level,
employment status,
Confounders: Gender, depression, anxiety,
education level, frequency of cognitive
employment status, training
depression, anxiety,
frequency of cognitive
training Self-ordered
=p=.0.0
207 search
-.405p
ß=-.0
=
ß

AARC gains ß= -.13 p< .002 Grammatical


in cognition reasoning
0 ß= .0
01 p=
= .37 < . 001 .963
2 p 10 p
0 - .
ß=
- . ß= Paired
01
AARC ß= -.09 p< .0 associate
losses in ß = -.
ß=
-. 0 learning
07 p= 4p
cognition .001 =.
04
8
Participants aged 66 to 75
years N = 2,473
Digit span RMSEA (90%CI): .07 (.04, .11)
Confounders: Gender,
education level, CFI: .99
employment status, TLI: .35
depression, anxiety, SRMR: .01
R2: 17%
frequency of cognitive
training

Figure 6.2b. Path analysis model exploring AARC gains and losses as predictors of cognition in the sub-sample of participants
aged 66 to 75 years while controlling for sex, education, employment status, depressive and anxiety symptoms, and frequency of
cognitive training.

178
Chapter 6: Cross-sectional association between objective cognitive performance and perceived age-related gains and losses in
cognition

Confounders: Gender,
education level,
employment status,
Confounders: Gender, depression, anxiety,
education level, frequency of cognitive
employment status, training
depression, anxiety,
frequency of cognitive
training Self-ordered
=p=.0.03
250 search
-.404p
ß=-.0
=
ß

AARC gains ß= -.08 p= .066 Grammatical


in cognition reasoning
ß= -.0
7 03 p=
. 76 . 001 .946
2 p= 20 p<
0 - .
ß=
- . ß = Paired
49
AARC ß= -.10 p= .0 associate
losses in ß = -.
ß=
-. 0 learning
05 p= 5p
cognition .362 =.
26
8
Participants aged 76 years and
over N = 472
Digit span RMSEA (90%CI): .00 (.00, .12)
Confounders: Gender,
education level, CFI: 1.0
employment status, TLI: .1.0
depression, anxiety, SRMR: .00
R2: 27%
frequency of cognitive
training

Figure 6.2c. Path analysis model exploring AARC gains and losses as predictors of cognition in the sub-sample of participants
aged 76 years and over while controlling for sex, education, employment status, depressive and anxiety symptoms, and frequency
of cognitive training.

179
Chapter 6: Cross-sectional association between objective cognitive performance and perceived age-related gains and losses in

cognition

Table 6.5. Correlations for AARC gains in cognition, AARC losses in cognition, and scores on the objective cognitive tasks.

Age sub-groups Variables AARC gains in cognition AARC losses in cognition


r (95% CI); p-value r (95% CI); p-value
Participants aged 51 to 65 Self-ordered search -.10 (-.13, -.06); <.001 -.07 (-.09, -.02); .002
Paired associate learning -.06 (-.09, -.02); .002 -.06 (-.15, -.08); .003
Grammatical reasoning -.09 (-.12, -.05); <.001 -.12 (-.10, -.03); <.001
Digit span -.06 (-.09, -.20); .002 -.06 (-.10, -.03); <.001
AARC losses in cognition .06 ( .03, .10); <.001
Participants aged 66 to 75 Self-ordered search -.07 (-.11, -.03); <.001 -.03 (-.07, .01); <.001
Paired associate learning -.01 (-.05, .03); .794 -.11 (-.15, -.07); <.001
Grammatical reasoning -.14 (-.18, -.10); <.001 -.11 (-.15, -.07); <.001
Digit span -.05 (-.09, -.01); .010 -.09 (-.13, -.05); <.001
AARC losses in cognition .06 ( .02, .10); .003
Participants aged 76 and over Self-ordered search -.05 (-.14, .05); .328 -.00 (-.09, .09); .981
Paired associate learning .004 (-.09, .10); .926 -.09 (-.17, .01); .064
Grammatical reasoning -.10 (-.19, -.01); .031 -.16 (-.25, -.07); <.001
Digit span -.06 (-.15, .04); .229 -.02 (-.11, .07); .699
AARC losses in cognition .02 (-.07, .11); .687

180
Chapter 6: Cross-sectional association between objective cognitive performance and perceived age-related gains and losses in

cognition

Table 6.6. Path analysis model exploring AARC gains and losses in cognition as predictors of cognition while controlling for sex,
education, employment status.
Predictors ß (95% CI); p- value
Participants aged 51 to 65
Self-ordered search Grammatical reasoning Paired associate learning Digit span
AARC-50 cognitive gains -.09 (-.12, -.05); <.001 -.08 (-.12, -.05); <.001 -.05 (-.09, -.02); .005 -.05 (-.09, -.02); .005
AARC-50 cognitive losses -.07 (-.10, -.03); <.001 -.10 (-.13, -.06); <.001 -.06 (-.09, -.02); .002 -.06 (-.09, -.02); .002
RMSEA (90%CI): .07 (.04, .10); CFI: .99; TLI: .61; SRMR: .01; R2: 7%
Participants aged 66 to 75
Self-ordered search Grammatical reasoning Paired associate learning Digit span
AARC-50 cognitive gains -.05 (-.09, -.01); .008 -.13 (-.17, -.09); <.001 -.0001 (-.04, .04); .995 -.04 (-.08, .001); .050
AARC-50 cognitive losses -.03 (-.07, .01); .105 -.08 (-.12, -.05); <.001 -.10 (-.14, -.06); <.001 -.08 (-.12, -.04); <.001
2
RMSEA (90%CI): .07 (.04, .11); CFI: .98; TLI: .49; SRMR: .01; R : 7%
Participants aged 76 and over
Self-ordered search Grammatical reasoning Paired associate learning Digit span
AARC-50 cognitive gains -.05 (-.14, .04); .350 -.09 (-.18, -.002); .046 .0002 (-.09, .09); .996 -.04 (-.13, .05); .261
AARC-50 cognitive losses -.03 (-.12, .06); .920 -.15 (-.24, -.06); .001 -.08 (-.18, .01); .068 -.01 (-.10, .09); .556
2
RMSEA (90%CI): .00 (.00, .10); CFI: 1.0; TLI: 1.2; SRMR: .00; R : 9%
Note: RMSEA= Root mean square error of approximation. CFI= Comparative fit index. TLI= Tucker-Lewis index. SRMR= Standard
root mean square residual. R2= R-squared/coefficient of determination. Sex, education, employment status, depression, anxiety,
and frequency of cognitive training are included as covariates in the regression models.

181
Chapter 6: Cross-sectional association between objective cognitive performance and perceived age-related gains and losses in

cognition

Table 6.7. Multiple linear regressions exploring AARC gains and losses in cognition as predictors of cognition while controlling for
sex, education, employment status, depressive and anxiety symptoms and frequency of cognitive training.

Predictors Self-ordered search Grammatical reasoning Paired associate learning Digit span
ß (95% CI); p- value R2 ß (95% CI); p- value R2 ß (95% CI); p- value R2 ß (95% CI); p- value R2
Participants aged 51 to 65
AARC gains in -.09 (-.12, -.05); <.001 .08% -.09 (-.12, -.05); <.001 .07% -.06 (-.09, -.02); .002 .03% -.06 (-.09, -.02); .002 .03%
cognition
AARC losses -.05 (-.09, -.01); .005 .03% -.09 (-.13, -.06); <.001 1% -.04 (-.08, -.001); .030 .02% -.04 (-.08, -.01); .022 .02%
in cognition
Participants aged 66 to 75
AARC gains in -.06 (-.10, -.02); .005 .03% -.13 (-.17, -.10); <.001 2% -.01 (-.05, .03); .770 0% -.05 (-.08, -.01); .026 .02%
cognition
AARC losses -.02 (-.06, .02); .270 .01% -.11 (-.15, -.07); <.001 1% -.09 (-.14, -.05); <.001 .08% -.07 (-.11, -.03); .001 .05%
in cognition
Participants aged 76 and over
AARC gains in -.04 (-.13, .05); .356 .02% -.09 (-.18, -.001); .047 1% -.01 (-.10, .08); .880 0% -.05 (-.14, .04); .251 .03%
cognition
AARC losses .01 (-.09, .11); .797 0% -.20 (-.30, -.11); <.001 3% -.10 (-.20, -.001); .048 1% -.05 (-.15, .05); .337 2%
in cognition
Note: ß= Standardised regression coefficient. R2= Partial R-squared/ coefficient of determination. Sex, education, employment
status, depression, anxiety, and frequency of cognitive training are included as covariates in the regression models.

182
Chapter 6: Cross-sectional association between objective cognitive performance and perceived age-related gains and losses in

cognition

Table 6.8. Multiple linear regressions exploring AARC gains and losses in cognition as predictors of cognition while controlling for
sex, education, employment status.

Self-ordered search Grammatical reasoning Paired associate learning Digit span


Predictors ß (95% CI); p- value R2 ß (95% CI); p- value R2 ß (95% CI); p- value R2 ß (95% CI); p- value R2
Participants aged 51 to 65
AARC gains in -.09 (-.13, -.06); <.001 1% -.09 (-.12, -.05); <.001 1% -.06 (-.09, -.02); .002 .03% -.06 (-.09, -.02); .002 .03%
cognition
AARC losses -.07 (-.11, -.04); <.001 1% -.10 (-.14, -.07); <.001 1% -.06 (-.10, -.02); <.001 .04% -.06 (-.10, -.02); <.001 .04%
in cognition
Participants aged 66 to 75
AARC gains in -.06 (-.10, -.02); .005 .03% -.14 (-.18, -.10); <.001 2% -.01 (-.05, .03); .705 0% -.05 (-.09, -.01); .024 .02%
cognition
AARC losses -.04 (-.08, .003); .069 .01% -.09 (-.13, -.06); <.001 1% -.10 (-.14, -.06); <.001 1% -.09 (-.13, -.05); <.001 .07%
in cognition
Participants aged 76 and over
AARC gains in -.04 (-.13, .05); .348 .02% -.09 (-.18, -.01); .038 1% -.002 (-.09, .09); .963 0% -.05 (-.14, .04); .253 .03%
cognition
AARC losses -.01 (-.10, .09); .899 0% -.16 (-.24, -.07); <.001 2% -.08 (-.18, .01); .066 1% -.03 (-.12, .06); .535 .01%
in cognition
Note: ß= Standardised regression coefficient. R2= Partial R-squared/coefficient of determination. Sex, education, employment
status, depression, anxiety, and frequency of cognitive training are included as covariates in the regression models.

183
Chapter 6: Cross-sectional association between objective cognitive performance and

perceived age-related gains and losses in cognition

Associations of psychological variables and frequency of


cognitive training with AARC gains and losses in the cognitive domain
across age sub-groups
Fewer AARC gains in cognition were associated with a younger subjective age in
early old age whereas the associations of AARC gains in cognition with depressive
and anxiety symptoms, ATOA, and self-rated health were either negligible or non-
significant; see Table 6.9. Among participants in middle and early old age, more
severe depressive and anxiety symptoms, more negative ATOA, an older subjective
age, and poorer self-rated health were associated with more AARC losses in
cognition. Among participants in advanced old age, more severe depressive and
anxiety symptoms, an older subjective age, and poorer self-rated health showed
small to moderate associations with more AARC losses in cognition (see Table 5).
Higher engagement in computerised cognitive training was associated with higher
AARC gains in cognition in middle-age only and was not associated with AARC
losses in any sub-group.

184
Chapter 6: Cross-sectional association between objective cognitive performance and perceived age-related gains and losses in

cognition

Table 6.9. Associations of psychological variables and frequency of cognitive training with AARC gains and losses in cognition
across three age sub-groups.

Participants aged 51 to 65
AARC gains in cognition AARC losses in cognition
Simple regressions Multiple regression Simple regressions Multiple regression
Predictors ß (95% CI); p-value R2 ß (95% CI); p-value ß (95% CI); p-value R2 ß (95% CI); p-value
Depression -.03 (-.06, .01); .154 .10% -.04 (-.08, .01); .137 .27 ( .24, .31); <.001 7% .16 ( .11, .20); <.001
Anxiety .01 (-.03, .04); .653 .01% .04 (-.01, .09); .078 .23 ( .20, .26); <.001 5% .09 ( .05, .14); <.001
Attitudes toward own ageing .02 (-.02, .05); .296 0% .01 (-.03, .04); .808 -.15 (-.18, -.11); <.001 2% -.06 (-.09, -.02); <.001
Subjective age .07 ( .03, .10); <.001 .10% .06 ( .02, .10); .002 -.21 (-.24, -.17); <.001 4% -.14 (-.18, -.11); <.001
Self-rated health .05 ( .02, .09); .004 .03% .03 (-.01, .07); .137 -.24 (-.27, -.20); <.001 6% -.12 (-.16, -.09); <.001
Frequency of cognitive training -.04 (-.08, -.01); .025 .01% -.02 (-.05, .02); .397 0%
Participants aged 66 to 75
AARC gains in cognition AARC losses in cognition
Simple regressions Multiple regression Simple regressions Multiple regression
2 2
Predictors ß (95% CI); p-value R ß (95% CI); p-value ß (95% CI); p-value R ß (95% CI); p-value
Depression .003 (-.04, .04); .897 0% -.02 (-.07, .03); .498 .28 ( .24, .31); <.001 8% .20 ( .15, .25); <.001
Anxiety .03 (-.01, .07); .096 .01% .06 ( .01, .11); .022 .21 ( .18, .25); <.001 5% .07 ( .02, .11); .006
Attitudes toward own ageing .06 ( .03, .10); <.001 .04% .06 ( .02, .10); .005 -.10 (-.14, -.06); <.001 1% -.03 (-.07, .00); .081
Subjective age .11 ( .07, .15); <.001 .10% .11 ( .07, .15); <.001 -.18 (-.21, -.14); <.001 3% -.12 (-.16, -.08); <.001
Self-rated health .03 (-.01, .07); .203 .01% -.01 (-.05, .04); .743 -.21 (-.25, -.17); <.001 4% -.11 (-.15, -.07); <.001
Frequency of cognitive training -.03 (-.07, .01); .188 .01% .02 (-.02, .06); .401 0%

185
Chapter 6: Cross-sectional association between objective cognitive performance and perceived age-related gains and losses in
cognition

Participants aged 76 and over


AARC gains in cognition AARC losses in cognition
Simple regressions Multiple regression Simple regressions Multiple regression
Predictors ß (95% CI); p-value R2 ß (95% CI); p-value ß (95% CI); p-value R2 ß (95% CI); p-value
Depression -.04 (-.13, .05); .424 .01% -.12 (-.22, -.01); .034 .33 ( .25, .41); <.001 11% .25 ( .16, .35); <.001
Anxiety .01 (-.08, .10); .844 0% .06 (-.04, .17); .227 .32 ( .24, .40); <.001 10% .18 ( .09, .28); <.001
Attitudes toward own ageing .03 (-.06, .12); .465 .01% .04 (-.05, .13); .429 -.06 (-.15, .03); .228 .03% -.01 (-.09, .07); .825
Subjective age .09 ( .00, .18); .051 1% .08 (-.01, .17); .084 -.19 (-.27, -.10); <.001 4% -.16 (-.24, -.08); <.001
Self-rated health -.07 (-.16, .02); .117 1% -.11 (-.20, -.02); .019 -.12 (-.21, -.03); .010 1% -.04 (-.12, .05); .417
Frequency of cognitive training -.02 (-.12, .06); .541 .01% .01 (-.08, .10); .854 0%
Note: ß= Standardised regression coefficient. Partial R2= Partial R-squared/coefficient of determination. Sex, education,
employment status, depression, anxiety, and frequency of cognitive training are included as covariates in the regression models.

186
Chapter 6: Cross-sectional association between objective cognitive performance and

perceived age-related gains and losses in cognition

6.5 Discussion
This is the first study testing whether the AARC-50 cognitive functioning subscale
has greater utility in linking self-perceptions of ageing to objective cognition
compared to the AARC-10 SF and other well-established measures of self-
perceptions of cognition and ageing. This study is also the first to explore whether
the coexistence of AARC gains and losses in cognition is associated with objective
cognitive functioning, a range of psychological variables, or frequency of
engagement in cognitive training. In a sample of older individuals with normal
cognitive function we found that, compared to bimodal measures of self-perceptions
of ageing, measures capturing the coexistence of positive and negative age-related
changes are more strongly associated with cognitive performance. However, the
global assessment of AARC -encompassing perceptions of age-related changes
across several life domains- is more strongly associated with objective cognition than
a domain-specific assessment of AARC in cognition. Unexpectedly, both AARC
gains and losses in cognition were associated with poorer cognitive performance.
Higher AARC losses in cognition, but not AARC gains, were related to more
depressive and anxiety symptoms, more negative self-perceptions of ageing, and
poorer self-rated health suggesting that poorer cognitive functioning may be one of
the many variables related to perceptions of AARC losses in cognition. AARC losses
and AARC gains in cognition respectively showed non-significant and negligible
associations with engagement in cognitive training.
The findings of this study support the importance of assessing the
coexistence of perceived gains and losses when relating self-perceptions of
cognition and/or ageing to objective cognitive functioning. However, in contrast to our
hypothesis, a global assessment of AARC gains and losses may be more
informative of objective cognition compared to a domain-specific assessment of
AARC in cognition. This may be due to the global assessment of AARC capturing
individuals’ perceptions of declines in their mental, physical, and social functioning,
alongside cognition, and these are all domains related to objective cognitive
functioning (Anstey, 2013). Even though it has previously been suggested that
domain-specific measures of self-perceptions of ageing have additional value in

187
Chapter 6: Cross-sectional association between objective cognitive performance and
perceived age-related gains and losses in cognition

predicting matched outcomes (Levy & Leifheit-Limson, 2009); study results


suggest that this may not be true for the cognitive domain.
The higher gains in cognition reported by those with poorer objective cognition
may be due to these individuals making less accurate appraisals of their cognitive
performance. This result is in contrast to research showing that cognitively healthy
people are generally accurate appraisers of their performance on cognitive
tests (Clare et al., 2010). In the current study participants were asked to report
perceived cognitive abilities in general, rather than evaluating their
performance before and/or after having completed a specific cognitive test;
this may explain the difference between our findings and existing literature.
However, as we found that more AARC losses in cognition were associated
with worse performance on all cognitive tests, this suggests that participants
are at least somewhat accurate in their perceptions of their cognitive abilities.
Alternatively, the finding that those with poorer cognitive performance
reported higher AARC gains in cognition may be due to individuals with
poorer cognition paying more attention to their cognitive gains. Thinking about
perceived cognitive gains may be an adaptive coping strategy which promotes
a more positive emotional state, re-establishes self-efficacy, and facilitates
acceptance of negative changes (Allen et al., 2015; Loidl & Leipold, 2019).
However, interestingly we found that higher levels of AARC gains in cognition
were fairly independent from more positive self-perceptions of ageing, mental,
and physical health. Hence, individuals perceiving more AARC gains in
cognition may not have a general tendency to be more positive in their self-
evaluations and may only show this tendency when rating their cognition.
Finally, the counterintuitive association of higher AARC gains and worse
cognitive functioning may be due to the nature of the items used to assess
AARC. Whereas AARC losses items capture perceived cognitive decline in
domains, such as memory and processing speed, that can be compared to
objective performance in tasks assessing the same domains, the gains items
capture social cognition and wisdom which may not be suitable for
comparison with performance on objective cognitive tasks. Moreover, some of
the items assessing AARC gains (e.g. “With my increasing age, I think things

188
Chapter 6: Cross-sectional association between objective cognitive performance and
perceived age-related gains and losses in cognition

through more carefully”) may be interpreted by some respondents as a sign of


cognitive slowing and hence these items may assume a negative valence. In
sum, whereas some items assessing AARC gains (e.g. “with my increasing
age, I have become wiser”) may represent age-related cognitive gains that
people would wish for themselves, others (e.g. “with my increasing age, I gather
more information before I make decisions” or “with my increasing age, I think things
through more carefully”) may capture cognitive changes that people may not wish to
experience while growing older.
This study found that individuals perceiving high losses in cognition may be
experiencing poorer cognition; especially poorer memory and verbal reasoning. In
line with international evidence on the associations of AARC with mental and
physical health (Sabatini et al., 2020a; Sabatini et al., 2020c), associations of
objective cognition with AARC losses were stronger than associations with AARC
gains. Among cognitive tests, perceived cognitive losses were most strongly
associated with grammatical reasoning. This finding was consistent across all age
sub-groups but strongest in advanced old age; supporting the greater accuracy of
older individuals in reporting cognitive difficulties (Jessen et al., 2014). The stronger
associations between AARC and cognitive performance found in the oldest age
group may be due to these individuals experiencing greater cognitive loss. Indeed, in
the current study individuals in advanced-old age showed poorer cognitive
performance compared to middle-aged and early-old individuals.
A recent study examining daily within-person variability in AARC and
cognitive performance showed that AARC losses predict within-person decreases in
inductive reasoning on the same day and decreases from one day to the next (Zhu &
Neupert, 2020). Despite the methodological differences between this study and ours,
both found that amongst several cognitive domains AARC is most strongly
associated with reasoning. This may be due to reasoning being vulnerable to age-
related decline (Christensen, 2001). Even though the current study only explored the
association of AARC losses with cognitive performance in the domains of memory
and verbal reasoning, recent evidence suggests that the association of AARC with
cognitive performance is consistently of small size across several cognitive domains
(O’Brien et al., 2020; Zhu & Neupert, 2020).

189
Chapter 6: Cross-sectional association between objective cognitive performance and
perceived age-related gains and losses in cognition

Our results are line with evidence supporting an association between


more negative self-perceptions of cognition and poorer objective cognition
(Amariglio et al., 2012; Reisberg et al., 2008) but documenting that self-
perceptions of cognition explain a small amount of variability in levels of
cognitive functioning. The small size of the associations of higher AARC
losses in cognition with poorer scores on objective cognitive tasks may be due
to perceived cognitive losses reflecting individuals’ experience of a trajectory
of subtle cognitive decline that is not captured with cross-sectional
assessment of objective cognition (Caselli et al., 2014). Research shows that,
although self-rated health generally does not match with objective measures
of health, it can be a better predictor of future levels of health than objective
measures of health (Idler & Benyamini, 1997). Similarly, AARC losses in
cognition may be more strongly associated with objective cognition at the
longitudinal level than at the cross-sectional level.
However, the small associations of higher AARC losses in cognition
with poorer scores on objective cognitive tasks may be due to AARC losses
and cognitive tasks capturing different types of information. Whereas scores
on the cognitive tasks are informative of participants’ cognitive ability, AARC
losses capture individuals’ perceptions of their cognitive performance which
do not necessarily match with their objective cognitive ability. Indeed, a large
amount of literature shows that self-perceptions of cognition can be influenced
by psychosocial factors (Buckley et al., 2013; Mendonça et al., 2016; Segel-
Karpas & Palgi, 2019; Siebert et al., 2020; Tandetnik et al., 2017; Witherby et
al., 2019). In line with broader literature on self-perceptions of cognition, in our
study higher levels of AARC losses in cognition are associated with more
severe depressive and anxiety symptoms, more negative ATOA, an older
subjective age, and poorer self-rated health. Interestingly, we found that with
increasing age depressive and anxiety symptoms are more strongly
associated with higher levels of AARC losses in cognition, whereas poorer
self-rated health and negative ATOA are most strongly related to AARC
losses in middle-age. These findings are aligned with literature documenting
the cooccurrence of depression, anxiety, poorer cognitive and physical health

190
Chapter 6: Cross-sectional association between objective cognitive performance and
perceived age-related gains and losses in cognition

in older age (Alexopoulos et al., 1993; Kassem et al., 2017; Roehr et al., 2017) but
are inconsistent with research supporting the greater self-relevance of ATOA in older
age (Kornadt & Rothermund, 2012).
The association of more AARC losses in cognition with an older subjective
age is in line with research reporting that those individuals who feel older than their
chronological age pay more attention to age-related losses in memory compared to
those people who feel their age or younger than their age (Segel-Karpas & Palgi,
2019). Overall, the small associations of AARC losses in cognition with objective
cognition and the small to moderate associations of AARC losses with more negative
scores on psychological variables suggest that perceived cognitive losses may be
somewhat influenced, not only by individuals’ objective cognitive ability, but also by
their interpretation of the cognitive changes they experience. The way in which older
individuals interpret their cognitive changes may be shaped by their beliefs about
age-related changes in cognition and their current emotional state (Brothers et al.,
2020; Weiss & Kornadt, 2018). Hence, individuals perceiving more cognitive losses
may benefit more from interventions promoting psychological health (Siebert et al.,
2020; Witherby et al., 2019), rather than from cognitive training.

Strengths and limitations


This study has several limitations. First, the sample included cognitively healthy
individuals aged 51 and over, so participants may not have been old enough to
perceive high levels of age-related losses in cognition. Future studies could consider
categorising cognitive scores (e.g. impaired versus not impaired) in order to explore
whether the associations of AARC gains and losses with cognitive performance differ
between those who are cognitively impaired and those who are not cognitively
impaired. Second, analyses are based on a selective group of participants. Indeed,
out of the 14,882 participants that took part in the PROTECT annual assessment
between 1st January and 31st March 2019, 8,826 were excluded from the current
study analyses. However, participants that were excluded from study analyses had
similar demographic profiles and mental and physical health to the study sample.
Third, the study sample included a majority of participants who were women, well-
educated, and who rated their health as good or excellent; hence extrapolation of

191
Chapter 6: Cross-sectional association between objective cognitive performance and
perceived age-related gains and losses in cognition

results to a broader population should be considered with caution. Fourth, analyses


were based on cross-sectional data; hence causality for the associations of
perceived cognitive gains and losses with cognitive functioning and psychological
variables cannot be inferred. Fifth, even though the AARC questionnaire captures
awareness of changes, we explored it in association with current objective cognition
instead of cognitive change. Nonetheless, we deemed the AARC-50 cognitive
subscale suitable to assess current self-perceptions of cognition as it is reasonable
to assume that current self-perceptions of cognition are on average more positive for
those who perceive more AARC gains and more negative for those who perceive
more AARC losses.
Sixth, as cognitive tests were self-administered online, those
participants who are less familiar with technology may perform more poorly on
cognitive tests compared to when assessed by a researcher. However, in
PROTECT all participants were familiar with the online cognitive tests from
previous assessments. Moreover, the use of an online cognitive assessment
provides a means of assessing cognition in a standardised way across
participants, which may not happen when in-person assessment is conducted.
Seventh, even though PROTECT participants are invited to repeat the
completion of the cognitive tests in three sessions within a week and then the
average score is calculated, numerous participants did not complete the
cognitive tests over three sessions. In order to optimise use of data across the
cohort, we used only data from the first session. Eighth, the cognitive tests
were completed on a separate day (within two months) to that on which
participants answered the AARC-50 cognitive functioning subscale. This is a
limitation as levels of perceived cognitive gains and losses can vary on a daily
basis (Neupert & Bellingtier, 2017; Zhu & Neupert, 2020). However, cognitive
functioning among individuals without dementia is generally stable over two
months (e.g. Lövdén et al., 2004). Ninth, items assessing perceived cognitive
losses may overlap with symptoms of depression and anxiety (Jessen et al.,
2007) and those individuals who are more introspective may score high on
perceived cognitive losses, depressive and anxiety symptoms (Roberts et al.,
2009). Tenth, this study only considered frequency of engagement in

192
Chapter 6: Cross-sectional association between objective cognitive performance and
perceived age-related gains and losses in cognition

computerised online cognitive training available as part of the PROTECT study; this
is a limitation as individuals could have been cognitively engaged (e.g. doing paper-
and-pencil crosswords) in many other ways not recorded in this study.
The large sample size is both a strength and a limitation of this study; even
though a large sample enabled us to test associations across three age groups; the
large sample size may have made it possible to detect even small effects that may
not be of clinical relevance. The use of valid measures assessing the coexistence of
perceived gains and losses across several domains (Brothers et al., 2019; Kaspar et
al., 2019; Sabatini et al., 2020b) is a strength as this made it possible to advance
knowledge on self-perceptions of cognition by showing that more perceived gains
and losses both in cognition and across life domains may be associated with poorer
cognitive performance.

6.6 Conclusion
This study adds several contributions to existing research. First, when examining the
association of self-perceptions of ageing with objective cognition it is important to
assess the coexistence of positive and negative self-perceptions of ageing across
several life domains. Second, both higher levels of AARC gains and losses in
cognition may be indicative of poorer cognitive functioning, even though associations
are either negligible or small and the reasons underlying the association of higher
perceived gains in cognition and poorer cognitive functioning need to be investigated
with future research. Third, whereas perceived cognitive gains are minimally related
to psychological variables, AARC losses in cognition are associated with more
severe depressive and anxiety symptoms, more negative ATOA, older subjective
age, and poorer self-rated health. Overall, poorer cognitive functioning may be only
one of the many variables related to AARC losses in cognition.

193
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

Chapter 7: Differences in health among people with different profiles of


awareness of positive and negative age-related changes

194
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

7.1 Summary
The systematic review presented in chapter 3 showed that even though the AARC
concept makes it possible to assess the coexistence of perceived gains and losses,
the coexistence of AARC gains and losses have never been explored in relation with
health outcomes. Hence Chapter 7 aims to identify the number and types of profiles
of individuals characterised by different combinations of levels of AARC gains and
losses and to explore whether the identified profiles differ in levels of cognitive,
mental, and physical health, and demographic characteristics.

Abstract
Background: Higher awareness of positive age-related changes (AARC gains) is
related to better mental health, whereas higher awareness of negative age-related
changes (AARC losses) is related to poorer mental and physical health. So far
perceived gains and losses have been explored separately, but people report gains
and losses concurrently in varying degrees, and different combinations of gains and
losses may be differentially associated with health. We identified profiles of gains
and losses and explored whether different profiles differed in cognitive, mental, and
physical health and sociodemographic characteristics.
Methods: We used cross-sectional data from the PROTECT study (N= 6,192; M
(SD) age= 66.1 (7.0)).
Results: Using latent profile analysis, a four-class solution showed the best model
fit. We found that 45% of people reported many AARC gains and few AARC losses
(Class 1); 24% reported moderate AARC gains and few AARC losses (Class 2); 24%
reported many AARC gains and moderate AARC losses (Class 3); 7% reported
many AARC gains and many AARC losses (Class 4). ANOVA and Chi-squared
testing showed that Class 1 had relatively better cognitive, mental, and physical
health, followed by Classes 2, 3, and 4. Sociodemographic characteristics differed
among classes.
Conclusion: Experiencing one’s ageing to a high degree as gain may be related to
better health only when individuals interpret ageing as involving low levels of loss
across several life domains. Risk in terms of poorer health emerged in those who
perceived both high gains and losses. Considering gains and losses in parallel,

195
Chapter 6: Cross-sectional association between objective cognitive performance and
perceived age-related gains and losses in cognition

rather than separately, may lead to a more fine-tuned understanding of relations with
health.

196
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

7.2 Introduction
Subjective evaluations of ageing (subjective ageing) comprise an important part of
the ageing process and provide valuable information about current and future levels
of health (Idler & Benyamini, 1997; Mendonça et al., 2016; Westerhof et al., 2014).
The assessment of subjective ageing can even be more informative of future health
than scores obtained through objective measures of health (Sargent-Cox et al.,
2012a). For instance the age individuals feel they are (subjective age) can be a
better predictor of cognitive functioning than well-established risk factors for cognitive
decline such as body mass index (Stephan et al., 2015). Individuals who perceive
their ageing more positively tend to have better cognitive, mental, and physical
health compared to those individuals who evaluate their own ageing more negatively
(Levy et al., 2018; Siebert et al., 2018; Siebert et al., 2016; Westerhof et al., 2014).
This association may be due to those individuals with positive subjective ageing
being more engaged in health-related behaviours (Bousquet et al., 2015). Subjective
ageing can therefore be useful to identify those individuals that could benefit the
most from health-promoting interventions (Brothers & Diehl, 2017).
A limitation of many existing studies assessing subjective ageing in relation to
health is that they used measures capturing global ratings of subjective ageing (Diehl
et al., 2014), such as asking individuals to report the age they feel they are
(subjective age; Barrett, 2003). However, when evaluating their own ageing, people
often take into account their different experiences across several life domains (e.g.
mental and physical domains; Steverink et al., 2001; Voss et al., 2018). A further
limitation of existing evidence is that it relies on constructs (e.g. attitudes toward own
ageing; Lawton, 1975) that assess subjective evaluations of ageing in a
unidimensional way. As a consequence, individuals can evaluate their own ageing
either positively (gains) or negatively (losses), but not both. However, individuals’
evaluations of their own ageing are often mixed and can include varying amounts of
gains and losses (Miche et al., 2014). On one hand, older individuals often
experience unwelcome changes that, for instance, can be the result of disability,
complex co-morbidities, and chronic health conditions. These changes can impact
negatively on subjective evaluations of ageing (Barnett et al., 2012; Kingston et al.,
2018; Royall et al., 2007). On the other hand, older individuals can contribute greatly

197
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

to the workforce, providing knowledge and experience, and engage in volunteer


activity. This, together with the experience of valuable social relations, increased
leisure time, and accumulated knowledge and life experience may foster positive
subjective evaluations of ageing (Carstensen, 1993; Carstensen et al., 2011;
Steverink et al., 2001; Timmer et al., 2002). In summary, when exploring the
association of subjective ageing with health, it is important to consider perceptions of
both developmental gains and losses and their possible coexistence in multiple life
domains (Baltes, 1987; Heckhausen et al., 1989; Steverink et al., 2001).
Compared to existing concepts, the awareness of age-related changes
concept (AARC; Diehl & Wahl, 2010) strives to better capture the multifaceted way in
which individuals experience their own ageing. AARC is anchored in lifespan
research (e.g. Baltes, 1987; Brandtstädter & Rothermund, 2002) and refers to “a
person’s state of awareness that his or her behaviour, level of performance, or way
of experiencing life has changed as a consequence of having grown older” (Diehl &
Wahl, 2010; pg. 342). AARC is a two-dimensional construct comprising awareness
of positive (AARC gains) and negative (AARC losses) age-related change. AARC
makes it possible to capture perceived age-related changes across five life domains
(health and physical functioning, cognitive functioning, interpersonal relationships,
socio-cognitive and socio-emotional functioning, and lifestyle and life engagement).
AARC assumes that gains and losses can occur simultaneously. Moreover, AARC
gains (e.g. wisdom) and losses (e.g. processing speed) capture different changes
that can happen within the same life domain and hence a person can potentially
perceive both high (or low) gains and losses. The perception of AARC gains is quite
independent from the perception of AARC losses; indeed correlations between
perceived gains and losses are small and positive (Kaspar et al., 2019). This is
inconsistent with results from some developmental studies showing that there is a
relation between the experience of gains and losses in older age. For instance, older
individuals are less focused on striving for gains and more focused on the prevention
of losses (Ebner et al., 2006; Miche et al., 2014).
Self-reported levels of AARC gains and losses explain variability in health-
related outcomes (Sabatini et al., 2020a). Higher levels of both AARC gains and
losses are correlated with worse cognitive performance (Sabatini et al., 2020b).

198
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

Moreover, higher AARC gains and AARC losses are associated with better and
worse mental health, respectively (Brothers et al., 2016; Dutt et al., 2016a). Finally,
higher AARC losses are associated with worse physical health (Brothers et al.,
2019). Individuals having fewer AARC gains and/or more AARC losses are less
likely to use self-regulatory strategies (Dutt et al., 2016b), to follow a healthy lifestyle
(Brothers & Diehl, 2017), and to have high self-efficacy (Dutt & Wahl, 2018), which
can also contribute to poor physical and mental health (Brandtstädter & Rothermund,
2002). As health is more strongly associated with AARC losses than AARC gains
(Sabatini et al., 2020a), individuals’ current cognitive functioning may be captured by
perceived losses, irrespective of the coexistence of perceived gains. Evidence on
AARC is limited to the study of the separate associations of gains and losses with
health; as a consequence, how the coexistence of different levels of gains and
losses relates to health indicators is unknown.
Existing research so far has reported on generalised average population
levels for AARC gains and AARC losses. In the United States, Germany, and United
Kingdom on average middle-aged and older adults reported “quite a bit” of AARC
gains and “a little bit” of AARC losses (Brothers et al., 2019; Kaspar et al., 2019;
Sabatini et al., 2020b). It may be that within a population there are different profiles
of individuals having varying levels of AARC gains and losses. Indeed, among
middle-aged and older individuals there is great variability in levels of cognitive,
mental, and physical health (Deary et al., 2009; Gow et al., 2007; Jagger et al.,
2009) and this may be reflected in individuals’ perceptions of AARC (Miche et al.,
2014) and in the combination of perceived levels of age-related gains and losses.
Finally, different AARC profiles may be related to specific sociodemographic
characteristics. In the United States and Germany individuals who are older, less
well-educated, and/or women report both more AARC gains and losses (Brothers et
al., 2016; English et al., 2019; Miche et al., 2014). In the United Kingdom, individuals
who are older and/or men report fewer AARC gains and more AARC losses; people
who are married, in a civil relationship, or co-habiting, and/or who completed a
university education experience both fewer AARC gains and losses; and people who
are working have fewer AARC gains (Sabatini et al., 2020b). However, how the

199
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

coexistence of gains and losses relates to sociodemographic characteristics is


unknown.
The present study aims for the first time to identify the profiles of AARC gains
and losses using a large sample of UK individuals aged 50 and over. The study
examines how participants’ profiles of AARC gains and losses relate to a broad
range of health indicators including cognitive, mental, and physical functioning. The
role of sociodemographic characteristics will also be considered. We hypothesised
that people can be divided into four classes of individuals with different profiles of
AARC gains and losses by crossing the two dimensions based on high versus low
scores on AARC gains and losses. We expected to find classes with the following
profiles: few gains and few losses, many gains and few losses, few gains and many
losses, and many gains and many losses. We also hypothesised that individuals in
different classes have different levels of cognitive, mental, and physical health.
Specifically, we hypothesised that individuals perceiving many gains and few losses
will obtain the best scores on health measures, followed by individuals perceiving
few gains and few losses, many gains and many losses, and few gains and many
losses (Sabatini et al., 2020a). Finally, we expected the classes to differ in terms of
sociodemographic characteristics, but as the strength and direction of associations
of AARC with sociodemographic variables varies between AARC gains and losses
(English et al., 2019; Sabatini et al., 2020b), it is difficult to predict exactly how the
combination of levels of AARC gains and losses is associated with
sociodemographic variables.

7.3 Methods
Study Design and Participants
We used cross-sectional data collected through the ongoing PROTECT
(Platform for Research Online to investigate Genetics and Cognition in Ageing
https://www.protectstudy.org.uk) study in 2019. PROTECT is a 25-year longitudinal
study launched in 2015 that assesses participants yearly on measures of cognitive,
mental, and physical health through an online platform.
Individuals are eligible to participate in the PROTECT study if they are UK residents,
English speakers, aged 50 years or over, have access to internet, and do not have a

200
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

clinical diagnosis of dementia at the point of recruitment. In PROTECT, participants


were recruited through national publicity and via existing cohorts of older adults
(Exeter 10,000 https://exetercrfnihr.org/about/exeter-10000/; Join Dementia
Research https://www.joindementiaresearch.nihr.ac.uk/; and Brains for Dementia
Research https://bdr.alzheimersresearchuk.org). Potential participants enrolled
through the PROTECT study website and provided informed consent online. As part
of the 2019 annual assessment, PROTECT participants were invited to complete
additional measures on self-perceptions of ageing and self-rated health. Analyses for
this study are based only on those participants who completed the additional
measures between 1st January 2019 and 31st March 2019. The PROTECT study
obtained ethical approval from the London Bridge NHS Research Ethics Committee
and Health Research Authority (Ref: 13/LO/1578). Ethical approval for the data
analyses was sought through the ethics committee at the University of Exeter,
School of Psychology (Application ID: eCLESPsy000603v1.0).
The study sample comprised 6,192 participants. The mean (SD; range) age
was 66.1 years (7.0 years; 51 to 95 years) and 76.0% were women. Further
demographic characteristics are reported in Table 7.1. Only 0.37% of participants
reported having mild cognitive impairment. We estimated that 5.6% of participants
might experience age-associated cognitive decline (as they scored 1 SD below the
mean sample score in two or more cognitive tasks) and 1.68% might have mild
cognitive impairment (as they scored 1.5 SD below the mean sample score in two or
more cognitive tasks).

Table 7.1. Descriptive statistics of demographic variables and main study variables
(N= 6,192).
Variables Statistics
Age in years, M (SD; Range) 66.1 (7.0; 51.4, 95.9)
Women, % 75.9
Marital status, %
Married/ civil partnership/ co-habiting 78.6
Widowed/ separated/ divorced/ single 21.4
Education level, %
High education 75.0

201
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

Low education 25.0


Employment status, (Employed %) 42.1
AARC gains, %
Not at all .1
A little bit 5.1
Moderately 22.2
Quite a bit 47.6
Very much 25.0
AARC losses, %
Not at all 3.8
A little bit 61.0
Moderately 28.7
Quite a bit 5.7
Very much .8
Digit span, M (SD) 7.6 (1.5)
Paired associate learning, M (SD) 4.7 (.94)
Grammatical reasoning, M (SD) 37.5 (10.5)
Self-ordered search, M (SD) 7.7 (2.6)
Self-rated health, %
Poor 2.0
Fair 12.9
Good 54.5
Excellent 30.6
Depressive symptoms, M (SD) 11.4 (2.9)
Anxiety symptoms, M (SD) 8.4 (2.4)
Functional ability, M (SD) .2 (.8)

Measures
Sociodemographic variables
Participants reported their age, sex, marital status, education level, and employment
status. Marital status was analysed as a dichotomous variable (the categories
married, in a civil partnership, and cohabiting were grouped together versus the
categories unmarried, divorced, separated, and widowed). Education level was
analysed as a dichotomous variable (high versus low education). Individuals with
high education had completed vocational qualifications (diploma, certificate, BTEC,
NVQ4, and above), undergraduate degrees (e.g. BA), post-graduate degrees (e.g.

202
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

MA), or doctorates. Individuals with low education had completed secondary


education (GCSE/O levels) or post-secondary education (college, A-levels, NVQ3, or
below). Employment status was analysed as a dichotomous variable (employed
versus not employed).

Awareness of age-related change (AARC)


The AARC-10 SF (Kaspar et al., 2019) contains ten items; five assessing AARC
gains and five assessing AARC losses. Each of these five items assesses a different
AARC life domain. All items start with the same stem: “With my increasing age, I
realise that…”. An example of an item capturing AARC gains is “…I appreciate
relationships and people much more”, while an example of an item capturing AARC
losses is “…I have less energy”. Respondents rate how much each item applies to
them on a five-point Likert scale (1= not at all, 2= a little bit, 3= moderately, 4= quite
a bit, and 5= very much). Scores can be obtained for the AARC gains and AARC
losses subscales by summing items that fall into the respective scales. Scales
scores range from a minimum of five to a maximum of 25; higher scores indicate
higher levels of AARC. The AARC-10 SF has internal consistency for the UK
population with Cronbach’s α value of .77 for the AARC gains scale and .80 for the
AARC losses scale (Sabatini et al., 2020b).

Cognitive functioning
Cognitive function was measured with the PROTECT Cognitive Test Battery (Corbett
et al., 2015; Hampshire et al., 2012; Huntley et al., 2018) which includes four tasks:
Digit Span; Paired Associate Learning; Grammatical Reasoning; and Self-ordered
Search. For each task a summary score can be obtained by subtracting the number
of errors from the number of correct answers; a higher score indicates better
performance. For digit span the summary score can range from 0 to 20. For paired
associate learning the summary score can range from 0 to 16. For grammatical
reasoning the summary score is also obtained by subtracting the number of errors
from the number of correct answers, but the score has no upper or lower limit due to
the fact that respondents can make attempts on as may trials as they can in the
available time (three minutes). Finally, the summary score for the self-ordered

203
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

search task can range from 0 to 20. In PROTECT, test-retest reliability for the
cognitive tasks was obtained by calculating Pearson’s r correlation coefficients
among participants’ age and scores on the same cognitive tasks completed three
times within a week. Test-retest reliability for the digit span task is r= -.06 at session
one, r= -.07 at session two, and r= -.07 at session three. Test-retest reliability for the
paired associate learning task is r= -.09 at session one, r= -.1 at session two, and r=
-.11 at session three. Test-retest reliability for the grammatical reasoning task is r= -
.13 at session one, r= -.14 at session two, and r= -.17 at session three. Finally, test-
retest reliability for the self-ordered search task is r= -.06 at session one, r= -.11 at
session two, and r= -.14 at session three.

Mental health
The Patient Health Questionnaire-9 (PHQ-9; Kroenke et al., 2001) is a nine-item
scale capturing depressive symptoms over the previous two weeks. Respondents
are asked to indicate how frequently they experience each symptom on a four-point
scale (1= not at all, 2= several days, 3= more than half the days, and 4= nearly every
day). The total score is the sum of the item scores and can range from a possible
nine to 36; higher scores indicate the presence of more depressive symptoms. The
PHQ-9 has excellent internal consistency with Cronbach’s α coefficient of .84 in the
normative sample (Kroenke et al., 2001) and of .76 in the sample of the current
study.
The Generalised Anxiety Disorder-7 (GAD-7; Spitzer et al., 2006) is a seven-
item measure asking respondents to indicate the frequency of occurrence of a list of
symptoms of generalised anxiety disorder on a four-point scale (1= not at all, 2=
several days, 3= more than half the days, and 4= nearly every day). The scale score
is the sum of the item scores and ranges from a possible seven to 28, higher scores
indicate greater presence of anxiety symptoms. The GAD-7 is an internally
consistent measure with a Cronbach’s a of .92 in the normative sample (Spitzer et
al., 2006) and of .86 in the sample of the current study.

204
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

Indicators of physical and functional ability


To measure individuals’ everyday functional ability at the present time a
modified version of the Lawton’s Instrumental Activities of Daily Living Scale was
used (IADL; Lawton & Brody, 1969). It assesses seven activities including preparing
meals, managing medications, and using the telephone. Two of the original eight
items (assessing laundry and housekeeping) were combined. Respondents have to
rate how difficult they find performing the activity (0= no difficulty, 1= some difficulty,
and 2= great difficulty). The total score varies from a possible 0 to 14, with higher
scores indicating greater functional difficulty. Inter-rater reliability for the IADL scale
resulted in a Pearson’s correlation of .85 (Lawton & Brody, 1969).
Self-rated health was assessed with a single-item question (taken from the
SF-36; Ware & Sherbourne, 1992) asking participants to rate their own health on a
four-point scale (1= excellent, 2= good, 3= fair, and 4= poor).

Analyses
To explore whether the population can be divided into classes of individuals
characterised by different profiles of levels of AARC gains and losses we conducted
latent profile analysis (LPA), using the gsem command in STATA version 16
(StataCorp, 2017). We fit the latent profile models based on manifest variables
representing responses to the ten items of the AARC-10 SF (5 gain items and 5 loss
items). To identify the model with the optimal number of classes, we fit a two-class
model and systematically increased the number of classes by one until adding more
classes no longer resulted in an improvement in model fit and did not compromise
the parsimony of the model. To identify the best fitting model, we compared
goodness of fit (GoF) information among models using Akaike’s information criterion
(AIC) and the Bayesian information criterion (BIC) (Nylund et al., 2007). The best
fitting model is one which has a low value on the information criterion while being
parsimonious in the number of identified classes. Having identified the best fitting
model, we reported estimates of the percentage of the population falling in each
class and estimates of the mean and standard deviation for each of the 10 AARC
items for each class. The latter estimates were the basis for assigning names that
characterise the profile of responses in each class.

205
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

Analysis of variance and Chi-square tests were conducted to compare


cognitive (scores on tasks assessing digit span, paired associate learning,
grammatical reasoning, and self-ordered search), mental (depressive and anxiety
symptoms), and physical functioning and sociodemographic characteristics (age,
sex, marital status, education level, and employment status) across the classes
identified in the latent profile analysis. For the analysis of variance comparing
cognitive, mental, physical, and functional health, we tested two models: one
unadjusted and one adjusted for the effects of age, sex, marital status, education
level, and employment status. For these analyses study participants were allocated
to the class for which they had the greatest probability of membership. For the
results from the analyses of variance, the effect size was calculated using eta
squared (η2). We interpreted effect sizes between .01 and .05 as small, between .06
and .13 as moderate and ≥.14 as large (Cohen, 1988).
In the analyses individuals who completed a vocational qualification (e.g.
diploma or certificate) were considered to have the same level of education as
participants who completed a undergraduate degree, a master’s degree, or a
doctorate. However, several types of vocational qualifications exist, with some
vocational qualifications being comparable to a university level education whereas
others are not. We therefore conducted sensitivity analyses where we grouped
individuals with a vocational qualification (diploma, certificate, BTEC, NVQ4, and
above) together with those who completed secondary education (GCSE/O levels)
and post-secondary education (college, A-levels, NVQ3, or below). All analyses were
conducted in STATA version 16 (StataCorp, 2017).

8.4 Results

Descriptive analyses
Participants had minimal levels of depressive (M= 11.42; SD= 2.91) and anxiety
symptoms (M= 8.43; SD= 2.43). A high proportion of participants perceived their
health as good (54.6%) or excellent (30.1%) and reported no functional difficulties.
The majority of participants perceived “a little bit” (61.0% of participants) or a
“moderate” level (28.7% of participants) of AARC losses and “moderate” (22.2% of

206
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

participants), “quite a bit” (47.6% of participants), or “very much” (25% of


participants) AARC gains. Further details of descriptive analyses are reported in
Table 7.1.

Pattern of AARC gains and losses profiles


The first objective of this study was to identify the classes that represent different
profiles of AARC gains and losses. Among the tested models (two-class, three-class,
four-class, and five-class models), we selected the four-class model as it was a
parsimonious model that also showed good fit (see Table 7.2 for goodness of fit
statistics). In the selected model, participants in Class 1 reported many gains and
few losses (45% of participants); participants in Class 2 reported moderate gains and
few losses (24% of participants); participants in Class 3 reported many gains and
moderate losses (24% of participants); and participants in Class 4 reported many
gains and many losses (7% of participants). The means of the items is shown for
each class in Table 7.3. Based on allocating the sample participants to the class to
which they had the highest probability of belonging, 2,833 participants were allocated
to Class 1, 1,493 to Class 2, 1,420 to Class 3, and 446 to Class 4.

Table 7.2. Goodness of fit indices for the four latent profile models tests.

Model Akaike’s information criterion Bayesian information criterion


Two-class model 164047.3 164256
Three-class model 157715.2 157997.9
Four-class model 156061.93 156418.7
Five-class model 154819.7 155250.5

207
Chapter 7: Differences in health among people with different profiles of awareness of positive and negative age-related changes

Average levels of gains and losses among the four classes


5
4,4
4,5 4,2 4,2 4,2
4,1 4,1
4 3,8
3,7
3,8 3,8
3,6 3,6 3,6
3,4 3,4
3,5 3,2
3,3
3,2
3,3 3,3
3
3 2,7
2,5 2,5 2,5 2,5
2,4 2,4 2,4
2,5 2,3
2
1,9
2 1,7
1,8
1,7
1,6
1,5
1,5 1,3 1,3
1,1
1

0,5

0
Physical gains Cognitive gains Interpersonal Socio-cog. Lifestyle gains Physical losses Cognitive losses Interpersonal Socio-cog. Lifestyle losses
gains socio-emot. losses socio-emot.
gains losses

Class 1 Class 2 Class 3 Class 4

Figure 7.1. Estimated average scores for each AARC item by class membership. Note: Y axis= scores on the AARC items ranging
from 0 to 5. Socio-cog socio-emot= socio-cognitive socio-emotional.

208
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

Table 7.3. Estimated means for each AARC item by class membership.

Class 1 Class 2 Class 3 Class 4


(many gains and (moderate gains (many gains and (many gains and
few losses) and few losses) moderate losses) many losses)

AARC gains, M (95% CI)


Physical 3.21 (3.16, 3.25) 2.48 (2.42, 2.53) 3.42 (3.36, 3.48) 3.34 (3.24, 3.44)
Cognitive 3.77 (3.73, 3.81) 2.46 (2.41, 2.51) 3.62 (3.56, 3.68) 3.36 (3.26, 3.45)
Interpersonal 4.20 (4.16, 4.23) 2.44 (2.39, 4.49) 4.10 (4.05, 4.16) 3.71 (3.62, 3.80)
Socio-cognitive 4.38 (4.34, 4.41) 2.54 (2.48, 2.59) 4.18 (4.13, 4.23) 3.77 (3.69, 3.86)
socio-emotional
Lifestyle 4.08 (4.04, 4.13) 2.73 (2.67, 2.79) 3.76 (3.70, 3.83) 3.20 (3.09, 3.32)
AARC losses, M (95% CI)
Physical 2.32 (2.28, 2.36) 2.44 (2.39, 2.49) 3.57 (3.51, 3.62) 4.20 (4.12, 4.29)
Cognitive 1.73 (1.69, 1.76) 1.87 (1.82, 1.91) 2.51 (2.46, 2.56) 3.26 (3.17, 3.34)
Interpersonal 1.14 (1.12, 1.16) 1.25 (1.22, 1.28) 1.79 (1.73, 1.84) 3.04 (2.97, 3.11)
Socio-cognitive 1.29 (1.26, 1.32) 1.56 (1.52, 1.59) 1.93 (1.88, 1.98) 3.25 (3.16, 3.34)
socio-emotional
Lifestyle 1.53 (1.50, 1.56) 1.70 (1.67, 1.74) 2.37 (2.31, 2.42) 3.59 (3.51, 3.67)

Figure 7.1 and Table 7.3 present the percentage of perceived gains and losses
reported in each of the five life domains captured by the multidimensional AARC
construct. For all four classes, physical losses rank higher than losses in other
domains; not surprisingly, physical losses are the highest in class 3 and 4 than 1 and
2. However, all four classes report losses across all life domains which suggests that
the pattern of overall findings is not driven disproportionately by physical-related
losses.

Differences in health among the four classes of AARC gain and loss
profiles
Using analysis of variance tests, we found that the four classes differed significantly
in scores obtained on the cognitive tasks digit span (p< .001), paired associate
learning (p< .001), grammatical reasoning (p< .001), and self-ordered search (p<
.001; Table 3), depressive symptoms (p< .001), anxiety symptoms (p< .001), and

209
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

mean levels of functional ability (p< .001). The class with the best cognitive, mental,
and physical health was the one with “many gains and few losses” (Class 1),
followed by the classes with “moderate gains and few losses” (Class 2), “many gains
and moderate losses” (Class 3), and “many gains and many losses” (Class 4; Table
3). Differences in health outcomes among classes remained consistent after
controlling for sociodemographic variables. Using Chi-square tests, we found that the
four classes were characterised by different proportions of participants who rated
their health as poor (p< .001; Table 5.4). The class of participants with “many gains
and few losses” (Class 1) perceived their health most positively, followed by those
who reported “moderate gains and few losses” (Class 2), “many gains and moderate
losses” (Class 3), and “many gains and many losses” (Class 4).

210
Chapter 7: Differences in health among people with different profiles of awareness of positive and negative age-related changes

Table 7.4. Comparison of cognitive, mental, and physical health outcomes across the four profiles of AARC gains and losses.

Variables Class 1 Class 2 Class 3 Class 4 Unadjusted Adjusted


p-value η2 p-value η2
Digit span, M (SD) 7.68 ( 1.56) 7.70 ( 1.46) 7.46 ( 1.49) 7.14 ( 1.53) <.001 .01 <.001 .01
Paired associate learning, M (SD) 4.78 ( .94) 4.78 ( .92) 4.61 ( .93) 4.49 ( .91) <.001 .01 <.001 .01
Self-ordered search, M (SD) 7.75 ( 2.57) 7.92 ( 2.64) 7.43 ( 2.60) 7.04 ( 2.92) <.001 .01 <.001 .003
Grammatical reasoning, M (SD) 38.13 (10.33) 38.53 (10.51) 36.27 (10.23) 33.39 (11.40) <.001 .02 <.001 .01
Depressive symptoms, M (SD) 10.83 ( 2.28) 11.38 ( 2.84) 11.89 ( 3.09) 13.80 ( 4.29) <.001 .07 <.001 .05
Anxiety symptoms, M (SD) 8.14 ( 2.10) 8.30 ( 2.23) 8.74 ( 2.76) 9.62 ( 3.31) <.001 .03 <.001 .03
Functional ability, M (SD) .05 ( .39) .14 ( .67) .19 ( .86) .79 ( 1.75) <.001 .06 <.001 .05
Self-rated health, %
Poor .5 .8 2.2 15.5
Fair 5.9 11.7 20.0 38.1
Good 51.6 58.9 61.1 36.1
Excellent 42.0 28.4 16.7 9.4
Note: Due to missing data in health-related variables and sociodemographic variable the sample size ranged. In Class 1 the sample
size ranged from 2,820 to 2,833; in Class 2 the sample size ranged from 1,487 to 1,493; in Class 3 the sample size ranged from
1,413 to 1,420; in Class 4 the sample size ranged from 442 to 446. Participants in Class 1 reported many AARC gains and few
AARC losses. Participants in Class 2 reported moderate gains and few losses. Participants in Class 3 perceived many gains and
moderate losses. Participants in Class 4 perceived many gains and many losses. η2= Eta-squared is the effect size. Adjusted for
age, sex, marital status, education level, and employment status.

211
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

Sociodemographic differences among the four AARC gains and losses


profiles

The four classes differed in age (p< .001) and were characterised by different
proportions of participants who were women (p< .001), married (p< .001), working
(p< .001), and achieved high education (p< .001; Table 7.5). The class with the
oldest participants (M= 69.74 years; SD= 8.76 years) was the one reporting “many
gains and many losses” (Class 4), whereas the class with the youngest participants
(M= 64.8 years; SD= 6.4 years) reported “many gains and few losses” (Class 1). The
“many gains and few losses” class (Class 1) had the highest proportion of
participants who were women (82.7% of participants), married (75.8% of
participants), and working (46% of participants); the “moderate gains and few losses”
class (Class 2) included the lowest proportion of women (66% of participants) and
married participants (10.1% of participants), and the “many gains and many losses”
class (Class 4) included the lowest proportion of individuals who were working
(28%). The “many gains and few losses” class (Class 1) had the highest proportion
of participants with a high education (76.8% of participants), whereas the “many
gains and many losses” class (Class 4) had the lowest proportion of individuals with
a high education (68.2% of participants). However, from sensitivity analyses
grouping individuals who completed a vocational qualification (e.g. diploma or
certificate) with those who completed secondary education and post-secondary
education (rather than with those who completed a undergraduate degree, a
master’s degree, or a doctorate), the “many gains and moderate losses” class (Class
3) was the one with the lowest proportion of individuals with high education.

212
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

Table 7.5. Comparison of sociodemographic variables and self-rated health across


the four profiles of AARC gains and losses.

Variables Class 1 Class 2 Class 3 Class 4 p-value


Women, % 82.7 66.0 75.4 69.1 <.001
Married, % 75.8 10.1 69.9 62.1 <.001
Employed, % 46.1 40.7 34.9 28.0 <.001
High education, % 76.8 75.6 72.8 68.2 <.001
High education 58.7 56.6 50.0 52.5 <.001
(without those with
a vocational
qualification), %
Age, M (SD) 64.9 (6.4) 66.2 (7.0) 67.4 (7.2) 69.7 (8.8) <.001
Note: In Class 1 the sample size ranged from 2,666 to 2,893; in Class 2 the sample
size ranged from 1,403 to 1,493; in Class 3 the sample size ranged from 1,325 to
1,420; in Class 4 the sample size ranged from 417 to 446. Participants in Class 1
reported many AARC gains and few AARC losses. Participants in Class 2 reported
moderate gains and few losses. Participants in Class 3 perceived many gains and
moderate losses. Participants in Class 4 perceived many gains and many losses.

7.5 Discussion
Using a large sample of UK individuals aged 50 and over, this is the first study to
identify profiles of AARC gains and losses and showing how different profiles relate
to participants’ cognitive, mental, and physical health and sociodemographic
characteristics. As hypothesised, we identified four profiles (Class 1 reported many
gains and few losses; Class 2 reported moderate gains and few losses; Class 3
reported many gains and moderate losses; and Class 4 reported many gains and
many losses), demonstrating that there is variability in the combination of AARC
gains and losses that individuals experience. However, only two profiles (Class 1
with “many gains and few losses” and Class 4 with “many gains and many losses”)
were consistent with our hypothesis. As hypothesised, the four classes differed in
terms of cognitive, mental, and physical health and sociodemographic variables. The
size of the effects for depression and functional ability were moderate whereas the
size of the effects for the remaining health indicators were small and may therefore

213
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

not be clinically meaningful. Overall, the “many gains and many losses” class had
the worst physical, mental, and cognitive health, followed by the classes with “many
gains and moderate losses”, “moderate gains and few losses”, and “many gains and
few losses”. This pattern of results suggests that considering the coexistence of
gains and losses is important for a comprehensive understanding of how AARC
relates to health.
In our sample, participants in the four groups perceived either moderate or
many positive age-related changes; what differs the most among classes is the level
of perceived negative age-related changes. Out of four classes, one perceived
moderate levels of losses and one perceived high levels of losses. These two
classes together only accounted for 31% of the population which is in contrast to
existing literature on other subjective ageing concepts reporting that, for example,
the majority of individuals have negative attitudes toward ageing. This discrepancy
may be due to attitudes toward ageing capturing individuals’ generalised beliefs
about ageing, whereas AARC capturing what people have experienced as they grow
older. It may be that many individuals have negative expectations about ageing that
are not always met and hence do not come into play when evaluating their own
ageing. However, the small proportion of participants reporting moderate or many
levels of AARC losses may also be due to our sample including individuals that on
average had good health.
The better cognitive, mental, and physical health in the group experiencing
many gains and few losses is consistent with literature reporting associations for
more gains and lower losses with better mental and physical health (Sabatini et al.,
2020a). The class having “many gains and few losses” had better cognitive, physical,
and mental health than the class experiencing “moderate gains and few losses”,
suggesting that the presence of gains enhances levels of cognitive, physical, and
mental health. However, the class reporting many gains but also many concomitant
losses showed the worst health status. Hence higher AARC gains may be related to
better health only when accompanied by low levels of AARC losses.
The worst cognitive performance showed by the group reporting many gains and
many losses may be due to some of the items assessing AARC gains (such as “With
my increasing age, I think things through more carefully”) being interpreted by some

214
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

respondents as negative age-related changes. Interventions aiming to increase


positive self-perceptions of ageing and engagement in health-related behaviours
(e.g. physical activity) are effective in enhancing perceived gains, reducing perceived
losses, and promoting healthy behaviours (Beyer et al., 2019; Brothers & Diehl,
2017).
Overall, study results suggest that individuals experiencing many losses are
likely to perceive their health as poor, irrespective of whether they experience high or
low levels of AARC gains. Promoting positive views of ageing among people with
high AARC losses (Beyer et al., 2019; Brothers & Diehl, 2017) may not be as
effective as decreasing negative views of ageing (Wolff et al., 2014). People who
experience many AARC losses may instead benefit from health-promoting
interventions (e.g. promoting engagement in physical activity; Brothers & Diehl,
2017) that are effective in decreasing the experience and perception of AARC
losses.
Class 1 (many gains and few losses) was the one including the youngest and
best educated participants. This is consistent with literature reporting that with
ageing individuals tend to experience an increasing number of changes (Baltes,
1987). The highest level of education characterising individuals reporting many gains
and few losses may be due to more educated people experiencing better cognitive,
physical, and mental health due to a greater likelihood of engaging in healthy
behaviours and cognitively stimulating lifestyles (Deary et al., 2000; Grønkjær et al.,
2019; Kaplan et al., 2014; Stephan et al., 2019; Yates et al., 2017).
The class with “many gains and few losses” had the highest proportion of
participants who were women whereas the class with “moderate gains and few
losses” had the lowest proportion of women. This finding may be due to women
being more likely to focus on positive changes (Dolan et al., 2008; English et al.,
2019; Miche et al., 2014), and/or to maintain or increase their family identity in later
life (Barrett, 2005). Finally, the class with “many gains and few losses” had the
highest proportion of participants who were working whereas the class with
“moderate gains and many losses” had the lowest proportion of working individuals.
It may be that working individuals are more physically fit, functionally able, and have
better cognitive abilities and hence are more likely to experience and report more

215
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

gains and fewer losses (Alavinia & Burdorf, 2008; Rohwedder & Willis, 2010; Zhan et
al., 2009). There is however some evidence reporting an increase in self-rated health
after retirement (Rohwedder & Willis, 2010). Retirement can have a negative effect
on the mental health (e.g. more depressive symptoms) of older individuals (Banks &
Smith, 2006) and this may be due to older individuals experiencing role loss, loss of
meaning, reduced self-esteem and decreased sense of mastery (Gallo, 2013; Kim &
Moen, 2002). The relationship between health and work may be complex,
encompassing directional effects of health on work and vice versa. Moreover, there
may be common variables that impact on both health and work (Lindeboom &
Kerkhofs, 2009).
Of note, the class reporting “moderate gains and few losses” included only
10% of participants who were married, whereas the remaining three classes reported
more levels of AARC gains and included a majority of participants who were married.
It may be that being married is important for the experience of positive age-related
changes. Indeed, literature reports that married individuals have better physical
health and lower chronic health conditions than those who are not married (Hughes
& Waite, 2009; Pienta et al., 2000). This may be due to the emotional and social
support that a spouse can provide (Barrett, 1999). Indeed, individuals who perceive
higher emotional and social support report higher AARC gains (O’Brien & Sharifian,
2019). However, it is not necessarily the case that those individuals who are
unmarried, divorced, or widowed lack social support or are socially isolated (Victor et
al., 2000).

Strengths and limitations


The study has some limitations. As the sample included mainly white participants,
women, individuals who were married (or in a civil partnership or co-habiting), well-
educated, and in good health; results cannot be generalised to the broader
population of middle-aged and older individuals. The cognitive tasks were not
completed on the same day on which participants completed the AARC
questionnaire but within two months of that date. However, cognitive abilities
generally do not deteriorate or deteriorate minimally in cognitively healthy individuals
over two months (e.g. Lövdén et al., 2004; Salthouse, 2019) and giving participants

216
Chapter 7: Differences in health among people with different profiles of awareness of

positive and negative age-related changes

the opportunity to complete the objective cognitive tasks on a separate day from the
remaining measures decreased participants’ burden and increased the likelihood of
collecting accurate answers. Mental and physical health were assessed through self-
report measures, hence recall bias may have occurred (Althubaiti, 2016).
This study has also several strengths. First, the large sample size made it
possible to conduct latent profile analysis and therefore to divide the sample into
subgroups of participants with different combinations of AARC gains and losses.
Second, this is the first study looking at the co-existence of gains and losses and
how this is related to levels of health and sociodemographic variables. Third, this
study considered a wide range of sociodemographic variables and several indicators
of health including cognitive functioning, functional ability, perceived health, and
symptoms of depression and anxiety. These strengths made it possible to explore
how the identified profiles of individuals with different levels of AARC gains and
losses differ in a wide range of health-related and sociodemographic variables.

7.6 Conclusion
This is the first study exploring the potential coexistence of AARC gains and losses
and identifying which profiles of AARC gains and losses are likely to occur in the
population. Among middle-aged and older individuals with intact cognitive abilities
and above average perceived physical health, there is variability in the coexistence
of AARC gains and losses. Most frequently, individuals perceive many age-related
gains and few age-related losses whereas the experience of both many gains and
losses is less frequent. This is also the first study exploring whether different profiles
of AARC gains and losses are related to varying levels of cognitive, mental, and
physical health and individuals’ sociodemographic characteristics. Profiles with
different combinations of AARC gains and losses are related to different levels of
cognitive, mental, and physical health, suggesting that assessing the coexistence of
gains and losses is important when relating AARC to health.

217
Chapter 8: Discussion

Chapter 8: Discussion

218
Chapter 8: Discussion

8.1 Introduction
The research included in this thesis relied on the multi-dimensional concept of
awareness of positive (AARC gains) and negative (AARC losses) age-related
change (Diehl & Wahl, 2010) to explore whether existing measures of AARC are
valid, reliable, and useful to assess perceived age-related changes in the UK
population (question 1) and whether levels of AARC gains and losses are associated
with indicators of cognitive (question 2), mental (question 3) and physical (question
4) health and sociodemographic variables (question 5).
Chapter 3 synthesised and quantified existing empirical work on the
associations of AARC with emotional and physical well-being. Even though one of
the aims of Chapter 3 was to quantify the relationship of AARC with cognition, no
study examining this association was found. Moreover, Chapter 3 showed that
existing studies on AARC were based only on four different cohorts including US
and/or German participants; this thesis therefore explored AARC and its associations
with cognitive, mental, and physical health and sociodemographic variables in a
sample of UK individuals aged 50 and over. As subjective ageing may vary across
cultures (Voss et al., 2018), the first empirical study (Chapter 4) examined the
psychometric properties of two measures of AARC, the AARC-10 SF and the AARC-
50 cognitive functioning subscale (Brothers et al., 2019; Kaspar et al., 2019), in the
UK population. In addition, Chapter 4 explored whether levels of AARC gains and
losses vary in relation to sociodemographic variables. Chapter 5 examined the
comments that participants wrote while completing the AARC questionnaires and, in
doing so, it provided additional information on the use of the AARC questionnaires in
the United Kingdom. Chapter 6 examined for the first time the association of AARC
with cognition and tested whether AARC gains and losses are more strongly
associated with objective cognition compared to well-established measures of
subjective cognitive decline and subjective ageing. In addition, this study estimated
the associations of perceived age-related cognitive changes with depressive and
anxiety symptoms, ATOA, subjective age, self-rated health, and engagement in
cognitive training. Unlike previous studies, Chapter 7 identified four profiles of
individuals characterised by different combinations of levels of AARC gains and
losses and estimated differences in their cognitive, mental, and physical health, and
sociodemographic characteristics. Overall, this research found that perceived age-
related gains and losses can reliably be assessed in the UK population with the

219
Chapter 8: Discussion

AARC questionnaires. It therefore validated two measures that, differently from


existing subjective ageing measures (e.g. ATOA and subjective age; Barrett, 2003;
Lawton, 1975) make it possible to capture the complexity of subjective experiences
of ageing (Miche et al., 2014; Steverink et al., 2001; Voss et al., 2018) and, as a
consequence, advance research on the topic in the United Kingdom. Moreover, this
research showed that levels of AARC gains and losses are related to levels of
cognitive, mental, and physical health and sociodemographic characteristics. This is
important as self-perceptions of ageing are potentially modifiable and intervention
programs aiming to decrease negative self-perceptions of ageing and/or to increase
positive self-perceptions of ageing appear to be effective in promoting health
maintenance in the second half of life (Brothers & Diehl, 2017; Burnes et al., 2019;
Wolff et al., 2014).
In the following sections the five research questions will be discussed in turn.
Findings from the examination of previous research in Chapter 3 and the empirical
studies in Chapters 4-7 will be summarised and discussed in relation to the wider
literature. Methodological limitations will be considered, followed by suggestions for
future research, and finally the theoretical and practical implications of this work will
be discussed.

8.2 Research question one


Are the AARC-10 SF and the AARC-50 cognitive functioning subscale valid, reliable,
and useful measures to assess perceived age-related gains and losses among UK
individuals aged 50 and over?

The research conducted in Chapters 4, 5, and 6 answers question one. First,


Chapter 4 tested whether two measures, one assessing global AARC (AARC-10 SF;
Kaspar et al., 2019) and one assessing AARC specific to the cognitive domain
(AARC-50 cognitive functioning subscale; Brothers et al., 2019) are valid and reliable
in the UK population aged 50 and over. Further evidence on the validity of the AARC
questionnaires is provided in Chapter 5 where the comments that participants wrote
while answering the AARC questionnaires were examined. Finally, Chapter 6 tested
whether a measure assessing AARC specific to the cognitive domain is needed, as
well as the utility of assessing the coexistence of perceived age-related gains and
losses.

220
Chapter 8: Discussion

Overall, both the AARC-10 SF and the AARC-50 cognitive functioning subscale
are valid and reliable measures that allow to assess perceived age-related changes
in the UK population. However, examination of participants’ comments in Chapter 5
suggests that the AARC questionnaires could increase their accuracy if minor
adjustments are implemented in future studies. For instance, the AARC
questionnaire could benefit from inclusion of a “non-applicable” answer option,
specification of the comparative timeframe in the item stem, and for UK use,
substitution of some words with more common alternatives in British-English.
Moreover, even though both the AARC-10 SF and the AARC-50 cognitive
functioning subscale show good psychometric properties in the United Kingdom,
Chapter 6 suggests that, when investigating the associations of AARC gains and
losses with objective cognitive functioning, a global measure of AARC may be more
useful compared to a measure of AARC specific to the cognitive domain. Indeed, this
study found that scores on the AARC-10 SF were more strongly associated with
objective cognition compared to scores obtained with the AARC-50 cognitive
functioning subscale. The greater predictive value of the AARC-10 SF over cognitive
functioning may be due to this measure capturing, in addition to perceived cognitive
changes, AARC gains and losses across several domains that are all related to
maintenance of cognitive health (Anstey, 2013). Examples of such domains are
mental and physical health, lifestyle, and social engagement.
Findings from Chapter 6 support the additional utility of assessing both
perceived gains and losses when linking self-perceptions of ageing to health
outcomes. Indeed, scores on the AARC questionnaires (AARC-10 SF and the
AARC-50 cognitive functioning subscale) were more strongly associated with
objective cognition in comparison to well-established unidimensional measures of
subjective ageing. The greater predictive value of the AARC measures may be due
to them capturing the coexistence of gains and losses, whereas the measures of
subjective ageing used in this research (ATOA and subjective age) categorise
individuals as being either positive or negative about their ageing. However, as
showed in Chapter 7, those individuals who perceive high losses do not necessarily
perceive low gains. Recent research on subjective ageing also emphasises the
importance of calculating separate scores for positive and negative age-related
experiences (Hooker et al., 2019; Robertson & Kenny, 2016a; Turner et al., 2020).

221
Chapter 8: Discussion

Finally, this thesis, together with existing evidence on AARC and subjective
ageing, suggests that AARC gains and losses are related to different aspects of
ageing (Brothers et al., 2020; Meisner, 2012; Zhu & Neupert, 2020). Indeed,
whereas AARC losses may be an indicator of poorer cognitive, mental, and physical
health, AARC gains may not reflect cognitive and physical health. Instead, AARC
gains appear to be related to social and psychological factors including perceived
emotional and instrumental support (O’Brien & Sharifian, 2019), specific personality
traits such as openness, conscientiousness, and neuroticism (Rupprecht et al.,
2019), personal growth, and social relations (Brothers et al., 2019). As AARC gains
and losses may be informative of different developmental outcomes, this highlights
one more time the importance of calculating separate scores for perceived gains and
losses. In conclusion, this thesis corroborates the validity and reliability of two
measures of AARC in UK population and their ability to provide a more fine-tuned
assessment of self-perceptions of ageing.

8.3 Research question two


Are AARC gains and losses associated with cognitive functioning?

The associations of AARC gains and losses with cognitive functioning are addressed
in Chapters 4, 6, and 7. Chapter 4 explored the correlations of AARC gains and
losses with self-reported, informant-reported, and objective indicators of cognition.
Chapter 6 examined the associations of AARC gains and losses with cognitive
functioning across individuals in middle-age, early old age, and advanced old age.
Chapter 7 showed that four sub-groups (profiles) of individuals characterised by
different levels of AARC gains and losses differ in their cognitive performance.
Unexpectedly, Chapters 4 and 6 found that among middle-aged and early old
individuals those with poorer cognition report more AARC gains. Moreover, Chapter
4 showed that higher AARC gains are related to perceived cognitive improvement.
Individuals with higher AARC gains may therefore be unaware of their poorer
cognitive ability. As existing literature shows that cognitively healthy older people are
generally accurate when evaluating their performance before and after having
completed a specific cognitive test (Clare et al., 2010), perhaps older adults are less
accurate appraisers of their overall cognitive ability compared to when they are
asked to estimate their performance on a specific cognitive task.

222
Chapter 8: Discussion

An alternative explanation for the counterintuitive association between higher


AARC gains and poorer cognitive performance may be that the experience of
cognitive losses (e.g. decline in memory) makes individuals more prone to reflect on
their cognitive gains (e.g. increased wisdom). Developmental theories, such as
Stereotype Internalisation and Dissociation Model (SIDI; Weiss & Kornadt, 2018) and
Terror Management Theory (Burke et al., 2010; Greenberg et al., 1986), suggest that
the experience of age-related losses, such as cognitive decline, confirms the
negative age stereotypes that individuals have internalised earlier in life and makes
them increasingly aware of their mortality (Burke et al., 2010). Thinking about gains
may be a strategy to counterbalance negative thoughts, re-establish a more positive
emotional state, and facilitate acceptance of negative changes (Allen et al., 2015;
Loidl & Leipold, 2019). Stereotype Internalisation and Dissociation Model (SIDI;
Weiss & Kornadt, 2018) theorises that throughout their lives individuals internalise
negative stereotypes about ageing and, when reaching older age, negative age-
related stereotypes become integrated into a person’s self-concept (internalisation).
As a way of distancing themselves from negative age-related stereotypes, older
individuals frequently dissociate themselves from their age group (dissociation).
Terror Management Theory (Burke et al., 2010; Greenberg et al., 1986) instead
assumes that the increased death awareness that characterises older people gives
rise to anxiety. However, a high sense of meaning in life and good self-esteem are
psychological resources that secure older individuals from concerns stemming from
awareness of mortality. This is because death anxiety becomes less salient when
people believe that some valuable aspects of themselves will continue to live, either
literally or symbolically, after cessation of their biological body.
Finally, Chapter 7 suggests a more complex relation between AARC gains
and objective cognition compared to that found in Chapters 4 and 6. Indeed, higher
AARC gains were related to better cognition only when individuals experienced
ageing as involving low losses but not when they perceived high levels of losses.
Hence, the same level of gains may have different meanings depending on the
levels of losses perceived by the individual.
In Chapters 4, 6, and 7 higher AARC losses were associated with poorer
objective cognition and this association was consistent irrespective of whether
participants experienced high or low concomitant gains. Hence, perceived cognitive
losses may be informative of current cognitive ability. This may be particularly true

223
Chapter 8: Discussion

for older people as the association of objective cognition with AARC losses was
strongest in size among individuals in advanced old age. It may be that, due to
negative age stereotypes becoming increasingly salient with ageing, when
individuals in advanced old age experience cognitive decline they are more likely to
perceive themselves in a negative way compared to younger individuals (Hicks,
2017; Levy, 2009; Meisner, 2012). However, AARC losses explained a small amount
of variability over objective cognition and in Chapter 4 self-perceptions of age-related
cognitive losses were not related to the judgment of a person close to the individual.
Hence, measures of AARC losses, objective cognitive tasks, and the judgment of an
informant may all provide different information and should not be used
interchangeably.
Perceived age-related changes in cognition also appear to be related to
individuals’ psychological well-being, as well as to their beliefs and attitudes toward
cognition in older age. Indeed, in Chapter 6 higher levels of AARC losses in
cognition were associated with greater depressive and anxiety symptoms, more
negative ATOA, an older subjective age, and poorer self-rated health. Similarly,
literature on subjective cognitive decline has consistently reported small associations
between self-perceptions of cognition and objective cognition and between
subjective cognitive decline and more negative emotional states (Chapman et al.,
2019; Crane et al., 2007; Hill et al., 2016; Lubitz et al., 2018; Mendonça et al., 2016;
Montejo et al., 2011; Schmidtke et al., 2008; Siebert et al., 2020; Yates et al., 2015;
Zlatar et al., 2018). Hence, even though at theoretical level the AARC concept
advances some of the limitations of existing conceptualisations of subjective
cognitive decline (see Chapter 6), results of this thesis are similar to those reported
in previous studies on subjective cognitive decline (Amariglio et al., 2012; Koppara et
al., 2015; Peter et al., 2014; Reisberg et al., 2008; Rönnlund et al., 2015).
In sum, AARC losses in cognition may be somewhat influenced by individuals’
objective cognitive ability but also by their interpretation of the cognitive changes
they experience. The way in which older individuals interpret their cognitive changes
may be shaped by their beliefs about age-related changes in cognition and by their
current emotional state (Brothers et al., 2020; Brothers et al., 2017; Levy, 2003;
Weiss & Kornadt, 2018). However, findings need to be interpreted with caution as
this research is the first to explore AARC in relation to cognition and it is entirely
based on cross-sectional analyses. As individuals with high AARC losses on

224
Chapter 8: Discussion

average have more negative ATOA, lower emotional well-being, and poorer
objective cognition, and these variables are all predictors of future cognitive ability,
the association of AARC losses with cognitive functioning may be stronger if
investigated with longitudinal studies compared to cross-sectional study designs. In
support of this hypothesis a recent micro-longitudinal study, using a modified version
of the AARC measure adapted for daily use, showed that AARC losses predict
within-person decreases in objective cognition from one day to the next (Zhu &
Neupert, 2020). Longitudinal analyses could be useful to better understand the
unexpected association between higher perceived cognitive gains and poorer
cognitive functioning.

8.4 Research question three


Are AARC gains and losses associated with mental health and emotional well-
being?

The associations of AARC gains and losses with emotional well-being are addressed
in chapters 3, 4, 6, and 7. The correlational meta-analysis conducted in Chapter 3
found that those individuals perceiving more age-related gains and fewer age-related
losses on average have better emotional well-being. However, studies included only
US and German participants. Moreover, from analysis of previous literature it was
not clear whether higher AARC gains act as a protective factor against depressive
symptoms. Indeed, whereas some studies reported a significant association between
higher levels of AARC gains and fewer depressive symptoms (Kaspar et al., 2019),
others did not (Dutt et al., 2016a; Dutt & Wahl, 2018; Dutt et al., 2018). In addition,
even though anxiety symptoms often arise from the experience of age-related losses
(Abramson et al., 1989; Banazak, 1997), the systematic search of the literature
conducted in Chapter 3 did not identify any study exploring the association of anxiety
with AARC. Hence Chapters 4 and 6 examined in a sample of UK individuals the
associations of AARC gains and losses with symptoms of depression and anxiety.
Chapter 7 focused on the coexistence of gains and losses and tested whether the
severity of depressive and anxiety symptoms varies across four profiles of
individuals having different combinations of levels of AARC gains and losses.
Overall, in Chapters 4 and 6 less severe depressive and anxiety symptoms
showed negligible associations with more AARC gains, meaning that among UK

225
Chapter 8: Discussion

individuals higher AARC gains may not protect from the experience of symptoms of
depression and anxiety. This finding may however be due to the low levels of
depression and anxiety experienced by PROTECT participants. In the systematic
review (Chapter 3) AARC gains appeared to be a protective factor against
depressive symptoms among individuals in advanced old age (Kaspar et al., 2019)
but not among younger individuals (Dutt et al., 2016a; Dutt & Wahl, 2018). However,
in this thesis (Chapter 6) the association of AARC gains with depression was
consistently of negligible size across individuals in middle-age, early old age, and
advanced old age. Hence, the mixed results found in previous studies exploring the
relationship of AARC gains with depression may be due to other factors, rather than
to differences in the mean age of their samples. For instance, findings of Chapter 7
show that in some cases high AARC gains are related to poor mental health due to
these individuals experiencing both high AARC gains and high AARC losses and
negative age-related evaluations have a greater influence on mental health
compared to positive age-related evaluations (Kornadt & Rothermund, 2012).
Nonetheless, experiencing one’s ageing as involving many gains appeared to
enhance mental health among those individuals experiencing few age-related
losses. It may be that the experience of positive changes helps older individuals to
appreciate and accept themselves and their lives. This may facilitate the
achievement of what Erikson called “ego-integrity” (Erikson, 1959), which is
important for maintenance of mental health in older age.
Regarding AARC losses, Chapters 4 and 6 showed that those UK individuals
perceiving more age-related losses experience more severe depressive and anxiety
symptoms. This is not surprising as many of the risk factors (e.g. poor physical
health) for depression and anxiety in older age are captured in the items assessing
AARC losses (Blazer, 2010; Butcher & McGonigal-Kenney, 2005; Fauth et al., 2012).
Hopelessness Theory of Depression (Abramson et al., 1989) assumes that the
interaction between negative life events and negative cognitive appraisal leads to
hopelessness which is a symptom of depression. In line with Hopelessness Theory
of Depression the experience of negative changes and the interpretation of such
changes as being a consequence of ageing may lead to hopelessness about the
future. In addition, Terror Management Theory (Burke et al., 2010) suggests that
interpreting negative changes as being a consequence of ageing may lead to worry,
which is a symptom of anxiety. In support of this reasoning Dutt and Wahl (2018),

226
Chapter 8: Discussion

showed that the experience of age-related losses leads to a decrease in self-efficacy


and to a more negative vision of the future; these in turn facilitate the insurgence of
depressive symptoms.
However, among those reporting high AARC losses, some individuals may be
more likely to experience depressive symptoms than others. Indeed those individuals
with high AARC losses and that are also likely to ruminate have a greater risk of
experiencing depressive symptoms compared to those with high AARC losses but
that do not have the tendency to ruminate over the experienced losses (Dutt et al.,
2018). Future research could test whether the mediating roles of self-efficacy and
future time perspective and the moderating role of rumination are present also in the
newly identified association of AARC with anxiety symptoms. Moreover, it may be
that the negative emotions that arise from the experience of age-related losses are
more accentuated among those individuals holding negative attitudes toward own
ageing (Brothers et al., 2020); this could also be tested in future studies. Finally,
even though the findings of this thesis advance our understanding of the
associations of AARC with mental health, longitudinal research is needed to test the
direction of the associations of AARC with mental health and to identify other
variables (e.g. loneliness, social engagement, social networks) that may moderate
and mediate the associations of AARC with emotional well-being.

8.5 Research question four


Are AARC gains and losses associated with indicators of physical health and well-
being?

The associations of AARC gains and losses with physical health are addressed in
Chapters 3, 4, and 7. Chapter 3 synthesised and quantified existing work on AARC
and physical well-being. Chapter 4 examined the associations of AARC gains and
losses with self-rated health and functional ability. Chapter 7 took into account the
coexistence of AARC gains and losses and tested whether four profiles of individuals
characterised by varying combinations of levels of AARC gains and losses differ in
their physical health.
Chapters 3 and 4 show that among US, German, and UK individuals, there is
a negligible association between higher levels of AARC gains and better physical
health; meaning that levels of AARC gains are not informative of physical health

227
Chapter 8: Discussion

status. This finding is inconsistent with the broader literature reporting that those
individuals with more positive subjective ageing experience better physical health
(Bryant et al., 2012; Kavirajan et al., 2011). The different results obtained for AARC
gains compared to other subjective ageing constructs may be due to AARC being
the first concept that makes it possible to obtain two separates scores for perceived
gains and losses. It may be that in some cases individuals with high AARC gains do
not report better physical health because they are also experiencing many age-
related losses; findings of Chapter 7 support this hypothesis. Future studies could
explore the moderating role of AARC losses in the associations of AARC gains with
physical health outcomes. Conservation Resources Theory (Hobfoll, 2002) assumes
that people seek to obtain, retain, and protect resources and that positive emotions
occur when people gain resources. In contrast, when people fail to gain resources or
when resources are threatened or lost, stress occurs. Hence, Conservation
Resources Theory suggests that, even though AARC gains may not have a strong
impact on physical health, perceiving age-related gains in the presence of negative
changes may provide emotional respite, and may motivate individuals to sustain goal
pursuit. Future research could explore whether the presence of AARC gains helps
those with poorer physical health to maintain a better emotional well-being.
The meta-analysis conducted in Chapter 3 found that among US and German
individuals, higher levels of AARC losses are moderately associated with poorer
scores on several indicators of physical health (e.g. self-rated health, functional
ability, number of chronic health conditions). Chapter 4 confirmed this pattern of
results among UK individuals aged 50 and over. Moreover, Chapter 7 showed that
levels of AARC losses are informative of physical health irrespective of the perceived
levels of gains. Hence, AARC losses seem to accurately reflect the negative
changes in physical health that individuals experience. Even though this thesis
focused on the cross-sectional association of AARC and physical health, recent
evidence suggests that higher AARC losses predict poorer health over time
(Brothers et al., 2019; Brothers et al., 2020).
Stereotype Embodiment Theory (Levy, 2009; described in Chapter 1, Section
1.9) provides an explanation for the detrimental influence of AARC losses over
physical health. Stereotype Embodiment Theory posits that the behavioural,
psychological, and/or physiological features of those individuals with more positive
subjective ageing explain why they are more likely to experience better cognitive,

228
Chapter 8: Discussion

mental, and physical health. In line with Levy’s theory high AARC losses may
negatively impact on physical health through three pathways: behavioural,
psychological, and physiological. In support of the behavioural pathway, those
individuals with higher AARC losses tend to engage less in physical activity and in
coping behaviours (Brothers & Diehl, 2017; Dutt et al., 2016b), which are important
for maintenance of physical health. The greater inactivity that characterises those
with higher AARC losses may be due to them interpreting age-related losses as
being an inevitable consequence of ageing (Kornadt et al., 2019a; Kornadt &
Rothermund, 2012; Rothermund, 2005; Rothermund & Kornadt, 2015; Stewart et al.,
2011; Wolff et al., 2017). In support of the psychological pathway, those with high
AARC losses have lower self-efficacy and a more negative view of the future (Dutt &
Wahl, 2018; Zhang & Neupert, 2020), which are also factors related to less
engagement in healthy behaviours and poorer physical health. Finally, even though
the limited evidence available on AARC has not yet tested the physiological
pathway, literature on subjective ageing suggests that those individuals with higher
AARC losses are more likely to experience physiological changes related to poorer
physical health such as heightened cardiovascular stress response (Chida &
Steptoe, 2010; Levy et al., 2000; Panaite et al., 2015; Stephan et al., 2015).
Overall, this thesis suggests that among UK individuals high AARC losses, but
not low AARC gains, may be an indicator of poorer physical health. However, as this
research relied on cross-sectional analyses and the few available longitudinal
studies treated AARC as the predictor of physical health (Brothers et al., 2019;
Brothers et al., 2020), the bidirectional associations of AARC gains and losses with
physical health remain to be addressed in future longitudinal studies. This would be
important as literature on subjective ageing indicates that changes in physical health
and AARC losses likely impact on each other (Baali et al., 2012; Benyamini & Burns,
2019; Diehl & Wahl, 2010; Giles et al., 2010; Karp, 1988; Kleinspehn-Ammerlahn et
al., 2008; Kotter-Grühn et al., 2009; Lin et al., 2004; Nilsson et al., 2000; Rippon &
Steptoe, 2018; Sabatini et al., 2020a; Sargent-Cox et al., 2012a; Sherman, 1994;
Spuling et al., 2013; Steverink et al., 2001; Westerhof et al., 2014).

229
Chapter 8: Discussion

8.6 Research question five


Are AARC gains and losses associated with sociodemographic variables?

Chapters 4 and 7 investigated the associations of AARC gains and AARC losses
with age, sex, marital status, employment status, and education level. Some of the
sociodemographic variables explored, such as employment status, had never been
investigated in relation to AARC. The remaining factors had been studied only
among US and German samples and were mainly treated as covariates (with the
exceptions of English et al., 2019; Miche et al., 2014). Chapter 4 explored whether
sociodemographic variables predict variability in levels of AARC gains and losses.
Chapter 7 examined whether four profiles of individuals with different combinations
of levels of AARC gains and losses differ in their sociodemographic characteristics.
Age. In line with evidence exploring AARC among US and German
individuals (Brothers et al., 2016; English et al., 2019; Miche et al., 2014), compared
to younger individuals, Chapter 4 found that on average older individuals living in the
United Kingdom experience higher AARC losses. However, compared to younger
individuals, older individuals living in the Uniters States or Germany experience
higher AARC gains (Brothers et al., 2016; English et al., 2019; Miche et al., 2014),
whereas older individuals living in the United Kingdom perceive fewer gains (Sabatini
et al., 2020b). More international research is therefore needed to clarify how levels of
AARC gains and losses change with increasing age.
Sex. The research conducted in this thesis, together with previous research,
suggests that US, German, and UK women experience more AARC gains than men
(Brothers et al., 2016; English et al., 2019; Miche et al., 2014). Moreover, in the
United Kingdom women report fewer AARC losses than men. There are several
possible explanations for the higher levels of gains and lower levels of losses
experienced by women. First, women are more prone to foster strong and satisfying
social relationships and to be involved in the family (Barrett, 2005), which are
important factors for maintenance of emotional well-being in older age (Carstensen,
1992, 1993, 2006). Second, as women enter older age already having navigated life
as marginalised gender (Bell, 1970; Lytle et al., 2018; Sontag, 1997), they may have
higher levels of adaptation and resilience which make them face and interpret age-
related changes in a less negative way. Third, women may be more optimistic in
general and their greater positivity may be reflected in their evaluations of age-

230
Chapter 8: Discussion

related changes. Indeed, on average women have less ageist attitudes, a more
elongated views of the life course, and younger age identities compared to men
(Barrett, 2005; Dolan et al., 2008; English et al., 2019; Kaminski & Hayslip, 2006;
Miche et al., 2014; Moore, 2010). Fourth, men and women give greater importance
to different age-related changes (Settersten, 2017). For instance, women tend to
worry about changes in their physical appearance whereas men are concerned
about maintenance of physical strength and energy (Calasanti, 2010; Calasanti &
King, 2018; Calasanti et al., 2006). As items assessing AARC losses capture
decreases in strength and energy, but do not capture a decline in one’s physical
appearance, this may also explain why men score higher on AARC losses than
women.
Whereas this thesis found that UK women perceive fewer losses than men,
US and German women on average perceive higher losses than men (Brothers et
al., 2016; English et al., 2019; Miche et al., 2014). Discrepancy in results may be due
to cultural differences in the way in which middle-aged and older women view
themselves and are viewed by their society (McConatha et al., 2003; Settersten,
2017). In sum, the way in which men and women experience age-related changes is
likely to differ. However, this research and the inconsistent findings reported in the
broader subjective ageing literature (Barrett, 2005; Kornadt et al., 2013; Levy et al.,
2002a; Schafer & Shippee, 2009; Turner et al., 2020) do not make it possible to infer
which gender experiences ageing more positively. It may be that women perceive
some aspects of ageing more positively, whereas men experience others more
positively. The multidimensionality of the AARC concept could be particularly useful
to explore whether sex differences in self-perceptions of ageing vary across different
life domains (Kornadt et al., 2020). Further understanding of sex differences in older
age may also be achieved by investigating those factors (e.g. social roles, levels of
introspection, and health status) likely to account for sex differences in subjective
ageing.
Marital status. Consistently with literature on subjective ageing Chapters 4
and 7 found that those individuals who are married, in a civil relationship, or co-
habiting experience fewer AARC losses compared to those who are not married
(Bergland et al., 2014; Brown, 2006; Dolan et al., 2008). This may be due to the
better physical health that married individuals generally have compared to those who
are not married (Hughes & Waite, 2009; Mejía et al., 2020; Pienta et al., 2000).

231
Chapter 8: Discussion

Moreover, when facing a negative event married individuals may benefit from the
social support of their spouse and this may attenuate the negative impact that age-
related losses can have on self-perceptions (Barrett, 1999; Kim et al., 2018). Indeed,
individuals who receive high social and emotional support report low AARC losses
(O’Brien & Sharifian, 2019). However, it is not necessarily the case that those
individuals who are not living with a spouse or partner lack social support or are
socially isolated (Victor et al., 2000).
Employment status. Chapter 4 found that working individuals experience
more AARC gains and fewer AARC losses compared to non-working individuals. It
may be that working individuals are more physically fit, functionally able, and have
better cognitive abilities, and hence are more likely to experience and report more
gains and fewer losses (Alavinia & Burdorf, 2008; Rohwedder & Willis, 2010; Zhan et
al., 2009). Moreover, as the transition to retirement can have a negative effect on the
mental health (e.g. more depressive symptoms) of older individuals (Banks & Smith,
2006; Gallo, 2013; Kim & Moen, 2002), this may also explain the higher losses
perceived by those who are not working. However, it has to be noted that the
protective role of working over healthy ageing may depend on the characteristics of
the job. Indeed, those individuals who experience ageism in the work environment
and/or who conduct physically demanding jobs may be less likely to report positive
subjective ageing (Staudinger et al., 2016; Weiss & Perry, 2020). In addition, the
type of job that individuals have may influence the association of AARC with
employment status. Individuals whose work is physically demanding or poorly paid
may be more likely to experience high levels of losses compared to those individuals
who are still in the workforce but covering less physical demanding or better
remunerated jobs. Future studies could investigate whether levels of AARC gains
and losses vary in relation to different types of jobs.
Education. As presented in Chapter 4, in the United Kingdom better
educated individuals reported fewer AARC gains and losses compared to those with
poorer education. The higher levels of AARC losses experienced by individuals with
lower education may be due to these individuals having started working at an earlier
age and hence having been working for longer. The fewer AARC losses experienced
by more highly educated individuals may also be due to them being more engaged in
healthy behaviours and therefore being more likely to enjoy good cognitive and
physical health (Craciun et al., 2017; Deary et al., 2000; Grønkjær et al., 2019; Herd

232
Chapter 8: Discussion

et al., 2007; Kaplan et al., 2014; Leopold & Engelhardt, 2013; Stephan et al., 2019;
Yates et al., 2017). In contrast, in the United States and Germany better educated
individuals reported both higher AARC gains and losses (Brothers et al., 2016;
English et al., 2019). However, associations between AARC and educational
attainment were of negligible size for UK, US, and German participants; meaning
that education has a minimal influence on AARC. Nonetheless, in order to detect
potential cultural differences in the association of AARC with education, cross-
cultural studies are needed.
Taken together, the above reported results suggest that levels of AARC gains and
losses are influenced by a wide range of sociodemographic characteristics, even
though the role of some of these factors needs to be better understood. Increasing
knowledge on the relation of sociodemographic variables with AARC would be
particularly important as it could help to identify those groups of individuals that may
benefit the most from interventions aiming to promote positive experiences of
ageing.

8.7 Methodological considerations


Exploring the associations of AARC gains and losses with cognitive, mental, and
physical health involved several methodological challenges that should be
considered while interpreting the results of this thesis. First, analyses were based on
secondary data collected as part of the PROTECT study. One of the many benefits
of using data from the PROTECT study was the possibility of collecting longitudinal
data; however, due to COVID-19 data collection in PROTECT was postponed and I
have not received the longitudinal data set by the end of my PhD. Hence, even
though this thesis supports the associations of AARC with cognitive, mental, and
physical health, the lack of longitudinal analyses does not make it possible to infer
the direction of such associations. Longitudinal data would have been particularly
important as items capturing AARC gains and losses invite respondents to reflect
upon age-related changes. Even though the research presented in this thesis
explored the associations of AARC gains and losses with a wide range of
sociodemographic variables, it was not possible to explore the association of AARC
with indicators of socioeconomic status due to this information not having been
collected as part of the PROTECT study.

233
Chapter 8: Discussion

One limitation of most studies collecting information about health and lifestyle
among older individuals and that is also relevant to the PROTECT study is that older
adults enrolled in surveys tend to differ from the general population in terms of
certain characteristics related to adaptive ageing. For instance many people decline
to take part in research studies as they are either too healthy and do not have the
time to participate, or are too ill and therefore unable to participate (Minder et al.,
2002; Ogletree & Katz, 2020). The cohort overall is skewed and over-represents
white participants, women, individuals who were married, in a civil partnership, or co-
habiting and who are well-educated. However, in the analyses we controlled for
sociodemographic characteristics; therefore, the statistically significant results found
in this thesis are unlikely to be due to the sociodemographic characteristics of the
samples.
For the purpose of the research included in this thesis, PROTECT participants
were invited to complete some measures related to their perceptions of ageing and
health, in addition to the main PROTECT questionnaires. Among the 14,797
participants that took part in the PROTECT annual assessment between 1st January
2019 and 31st March 2019, 9,410 participants completed the additional measures.
Hence, the analyses conducted in this thesis are based on a self-selected sample.
Participants who answered the additional measures may be more prone to reflect on
their age-related changes. Compared to those who did not complete the AARC
questionnaires in 2019, the sample who filled in the additional measures relevant for
this thesis included a larger proportion of women, participants who were better
educated and non-working. This selection bias may impact on the significant
associations found between AARC and sociodemographic variables. However, there
is no immediate reason to believe that the relationship between AARC and
sociodemographic variables is different between those who provided data and those
who did not.
Another limitation related to the use of secondary data is that since the
beginning of the PROTECT study in 2015 participants took part in several projects
that may have shaped the way they perceive their ageing. Moreover, participants
enrolled in the PROTECT study are encouraged to engage in brain training games.
As a consequence, many participants in PROTECT have been engaged in
computerised brain training games and may therefore show better cognitive
functioning as a consequence of having benefitted from a practice effect. Hence the

234
Chapter 8: Discussion

study sample may not be representative of the general population. However, in order
to account for this potential source of bias, when exploring the associations of AARC
gains and losses with objective cognition in the analyses we controlled for the
frequency with which individuals engaged in computerised brain training games.
The objective cognitive assessments were not undertaken on the same day
on which participants completed the AARC questionnaire but within two months of
AARC completion and this may be a limitation as AARC may fluctuate over time
(Neupert & Bellingtier, 2017). However, as completing a battery of cognitive tasks is
demanding, especially for older individuals, allowing participants to complete
objective cognitive assessments on a separate day from the remaining measures
decreased participants’ burden and increased the likelihood of collecting accurate
answers. Moreover, cognitive functions do not deteriorate or deteriorate minimally in
individuals without dementia over two months. While cognitive abilities were
assessed through both objective and subjective measures, mental and physical
health were assessed through self-report measures only. The use of self-rated
health as an indicator of physical health is a limitation of this research as a person
may report good self-rated health despite having chronic health conditions and
functional impairments (Pinquart, 2001). Hence the association that was found
between AARC and self-rated health cannot be generalised to objective physical
health. Finally, in PROTECT participants were not asked to specify the type of
vocational qualification they obtained. Hence in this thesis individuals who completed
a vocational qualification (e.g. diploma or certificate) were considered to have the
same level of education as participants who completed an undergraduate degree, a
master’s degree, or a doctorate. This is a limitation as in the United Kingdom several
types of vocational qualifications exist, with some being comparable to a university
level education while others are not.
As a strategy to avoid missing data in PROTECT participants have to answer
all items of a questionnaire before being allowed to continue with the next
assessment. This is a strength of the current research as it made it possible to avoid
missing answers within the same questionnaire. However this may also limit the
validity of results as participants are forced to choose an answer even when they do
not identify themselves with any of the answer options. Leaving participants free to
skip some items could have been very informative when validating the AARC
measures in the United Kingdom. Indeed, patterns of missing answers could have

235
Chapter 8: Discussion

indicated lack of utility or lack of understanding of a specific item. However, this


problem was partially solved by allowing participants to write any additional
comments they had in relation to the AARC questionnaires; analyses of these
comments (reported in Chapter 5) made it possible to identify issues related to some
AARC items.
The large sample used to conduct analyses for the four empirical studies is
both a strength and a limitation of this research. On one hand, the large sample size
may have made it possible to detect small associations between AARC and health
outcomes that may not have clinical relevance. On the other hand, a large sample
made it possible to reliably conduct subgroup analyses (e.g. to examine
measurement invariance across sex, education level, and age groups). A strength of
the thesis is that it explored for the first time the association of AARC with objective,
self-reported, and informant-reported measures of cognition. Another methodological
strength of the research presented in this thesis was controlling for several
covariates in the analyses. However, other confounding variables such as the
presence of chronic health conditions (English et al., 2019) may exist but have not
been assessed in the research presented in this thesis. A further strength was the
conduct of both qualitative and quantitative analyses, which made it possible to
better explore the validity of the AARC measures in the United Kingdom and
understand the associations of AARC with health outcomes. Finally, the use of latent
profile analysis is also a strength of this thesis as it made it possible for the first time
to identify profiles of individuals having different degrees of AARC gains and losses
and to examine how coexistence of ageing and losses is related to health outcomes.

8.8 Future research directions


Overall, the work presented in this thesis extends previous evidence on the
associations of AARC with health outcomes and sociodemographic variables in the
UK population and provides knowledge for the association of AARC with the newly
investigated outcomes of cognition and anxiety. However, many aspects of these
associations remain unknown, and this gives rise to future directions. As evidence on
subjective ageing suggests that AARC and health outcomes may influence each
other over time, there is the need to use longitudinal study designs to examine the
bidirectional associations of AARC gains and losses with cognitive, mental, and
physical health (Dutt et al., 2016a; Rippon & Steptoe, 2018; Sargent-Cox et al.,

236
Chapter 8: Discussion

2012a, 2012b; Seidler & Wolff, 2017). Even though I planned to explore the causal
pathways between AARC and health outcomes during my PhD, this was not possible
due to a delay in the collection of follow-up data. However, when longitudinal data for
AARC and indicators of cognitive, mental, and physical health are available in
PROTECT, I will conduct longitudinal analyses as follows. Initially I will explore the
strength and direction of the associations of AARC gains and losses with cognition,
mental and, physical health by fitting cross-lagged panel models (Kearney, 2018). I
will conduct the same analyses across three age sub-groups: individuals in middle-
age, early old age, and advanced old age.
Once the direction of the associations has been identified, I will explore
several mediation pathways. First, I will test whether depressive and anxiety
symptoms mediate the longitudinal association of AARC (gains and losses) with
cognitive performance. Second, as for the purpose of this research project I collected
through the PROTECT study longitudinal data on rumination, I will examine whether
follow-up levels of rumination mediate the associations of AARC gains and losses
with symptoms of depression and anxiety. So far only one study has shown that
rumination plays a role in the association of AARC losses with depression (Dutt et
al., 2018) and the potential role played by rumination in the association of AARC with
anxiety has never been explored.
Third, I will explore the mediating roles of several health-related behaviours
(such as level of engagement in physical activity, following a healthy diet, and level
of engagement in social and cognitively stimulating activities) in the association of
AARC (gains and losses) with physical health (self-rated health and functional
ability). However, many other variables not assessed in the PROTECT study (such
as personal and social resources) may mediate the associations of AARC with
cognitive, mental, and physical health and could therefore be explored in future
studies (O’Brien & Sharifian, 2019). Finally, as perceived gains may help to
counteract and attenuate the negative effect of losses over health, it would be
important to explore the interaction of AARC gains and AARC losses as a predictor
of health outcomes (Brothers & Diehl, 2017; Dutt et al., 2016b).
As follow-up data for AARC and health-outcomes has been collected during
the COVID-19 pandemic, when conducting longitudinal analyses, I will take into
account and acknowledge the impact that COVID-19 may have had on the data
collected. Indeed, it is likely that the lockdowns and social restrictions that have been

237
Chapter 8: Discussion

put in place during the COVID-19 pandemic impacted both on older individuals’
mental health (Robb et al., 2020) and on their self-perceptions of ageing
(Terracciano et al., 2021).
In addition to considering AARC either as a predictor or as an outcome of
health indicators, future research could explore AARC as a mediating and/or
moderating factor in the associations of other psychological constructs with health
outcomes. For instance, a cross-sectional study showed that higher AARC losses
mediate the associations of poorer perceived physical health and memory with
poorer control beliefs (Zhang & Neupert, 2020). Another study (Brothers et al., 2020)
showed that AARC losses mediate the effect of ATOA on physical health and that
both AARC gains and AARC losses mediate the effect of ATOA on mental health.
However, this study focused on US and German participants only and did not
consider cognition as an outcome. Whether AARC acts as a mediator in the
association of ATOA with cognitive functioning can be explored when longitudinal
data on AARC are available in PROTECT.
As findings of this and previous research (Dutt et al., 2016a; Kaspar et al.,
2019; Zhu & Neupert, 2020) showed that health outcomes are most strongly
associated with AARC losses compared to AARC gains, there is the need to better
understand what is captured with AARC gains. Results of this thesis show (Chapter
7, Table 7.3) that, among the five AARC life domains, participants experience more
gains in the socio-cognitive socio-emotional domain and in their interpersonal
relations compared to the remaining domains. Losses instead are more commonly
experienced in the health and physical functioning domain compared to the
remaining AARC domains. By assessing the associations of AARC with cognitive
and physical health, this thesis may have tapped mainly into the negative aspects of
ageing. As gains are more prevalent in the socio-emotional domain, AARC gains
may be more strongly linked to scores on measures assessing social relations,
social support, and loneliness (Carstensen, 1992, 1993, 2006; Hobfoll, 2002; Kaspar
et al., 2019) compared to health outcomes. So far there are only two studies that
supported the association of AARC with socio-emotional outcomes such as social
support (O’Brien & Sharifian, 2019), personal growth, and personal relations (Kaspar
et al., 2019). There are no studies available exploring the associations of AARC
gains and losses with levels of social isolation and loneliness. Exploring these
associations would be important to better understand the association that has been

238
Chapter 8: Discussion

found in this thesis between marital status and AARC. As levels of loneliness and
social isolation have already been explored among PROTECT participants; future
studies could extend existing analyses undertaken with the PROTECT dataset to
explore whether levels of social isolation and loneliness influence perceptions of
AARC gains and losses. In sum, more research on the associations of AARC gains
and socio-emotional variables would enable better understanding of what is captured
with AARC gains.
Some of the associations found for AARC in relation to sociodemographic
variables among UK individuals differ from those reported for US and German
samples. It would be important to conduct international studies on the associations of
AARC with sociodemographic variables (Ackerman & Chopik, 2020; Staudinger,
2015). Indeed, in support of cultural differences in the way in which individuals
experience ageing a recent study, examining differences in subjective age, ATOA,
and age bias among 68 different countries, showed that UK individuals report the
youngest subjective age, followed by US and German individuals (Ackerman &
Chopik, 2020).
The systematic review showed that the association of AARC with physical
health has mainly been explored with subjective measures of physical health (e.g.
self-rated health) and the research included in this thesis also relied on self-reported
measures. As subjective evaluations of health can differ greatly from scores obtained
with objective measures of health (Carstensen, 1992, 1993, 2006; Chan et al., 2007;
Idler & Benyamini, 1997; Jylha et al., 2001), it would be appropriate to explore AARC
in relation to more objective measures (e.g. presence of chronic health conditions,
assessments of physical strength, medications used, time spent in hospital, and
biomarkers of ageing). As in 2020 data for co-morbidity was collected as part of the
PROTECT main assessment, when data are available I will explore the cross-
sectional associations of AARC gains and losses with two more objective indicators
of physical health that are number of chronic health conditions and body mass index.
Finally, analyses included in this thesis were based on a sample of individuals
who were cognitively healthy, with above average self-reported health and
education. It would therefore be important to replicate the research conducted in this
thesis in samples of individuals who are less healthy as they may perceive higher
levels of AARC losses and show stronger associations with health outcomes.
Similarly, as people with mild cognitive impairment are likely to experience significant

239
Chapter 8: Discussion

changes in many domains of their lives (Cook & Marsiske, 2006), future research
could explore whether the concept of AARC is of use among individuals with mild
cognitive impairment. The multidimensionality of the AARC questionnaire may make
it possible in clinical contexts to understand how individuals with mild cognitive
impairment perceive the losses they are experiencing across several domains of
their lives (Caddell & Clare, 2010; Corner & Bond, 2006). Completing the AARC
questionnaire may also facilitate self-awareness about the changes that individuals
are experiencing and foster motivation to engage in adaptive behaviours (Baltes,
1997; Brandtstädter & Rothermund, 2002; Buschkuehl et al., 2008; Clare et al.,
2010; Clare & Woods, 2004; Feher et al., 1992; Freund & Baltes, 1998; Hudes et al.,
2019; Jessen et al., 2014; Loidl & Leipold, 2019; Meléndez et al., 2018; Michon et
al., 1994).

8.9 Implications of the research findings


The findings of this thesis have theoretical and practical implications for research,
clinical practice, and social policies.

Theoretical implications. Overall, the research included in this thesis


emphasises the complexity of self-perceptions of ageing. The identification of four
different profiles of individuals having varying combinations of degrees of AARC
gains and losses corroborates the utility of a construct, such as AARC, that assesses
both gains and losses. This research also provided empirical support for several
health outcomes outlined in the Diehl and Wahl (2010) conceptual framework of
AARC, including the newly investigated outcomes of cognitive performance and
anxiety. New factors potentially related to AARC, such as engagement in purposeful
activities and attribution of changes to health versus to ageing, were also identified in
Chapter 5 and they could be added to the AARC conceptual framework. However,
the association between higher perceived gains and poorer cognitive performance
does not support theoretical reasoning on AARC. If longitudinal research confirms
this finding, the AARC theoretical framework may benefit from amendment.
Implications for research. At a methodological level, this thesis provided the
psychometric properties for the AARC-10 SF and the AARC-50 cognitive functioning
subscale, enabling future research within the United Kingdom to use these measures
to assess perceived age-related changes in cognitive, mental, and physical health.

240
Chapter 8: Discussion

However, findings of this research also suggest that future studies using the AARC
questionnaires could consider including a “non-applicable” answer option to AARC
items, specifying the comparative timeframe in the item stem, and for UK use,
substitute some American-English words with more common alternatives in British-
English. As a multidomain measure of AARC (AARC-10 SF) appeared to be a better
indicator of objective cognitive functioning compared to a measure of AARC specific
to the cognitive domain (AARC-50 cognitive functioning subscale), future studies
may benefit from the use of multidimensional measures when relating self-
perceptions of ageing to objective cognition. Finally, as AARC varies in relation to
sociodemographic variables, future studies that do not have the primary aim of
exploring the relation between sociodemographic characteristics and AARC should
control for sociodemographic variables in their analyses.
Some of the knowledge produced in this thesis could be transferred to those
constructs that share similarities with AARC such as other subjective ageing
concepts, subjective cognitive decline, and self-rated health. I have extended some
of the findings of this research in several empirical studies. First, as higher levels of
depression and anxiety are related to more AARC cognitive losses, I applied this
knowledge to the concepts of subjective cognitive decline and subjective memory
decline. By using secondary data from the Cognitive Function & Ageing Study
(CFAS-Wales Study; http://www.cfas.ac.uk/cfas-wales/) I found that in a group of
cognitively healthy individuals aged 65 and over subjective cognitive decline and
subjective memory decline are associated with more anxiety and depression.
Moreover, greater subjective memory decline predicts greater decline in memory
(Sabatini et al., 2021). Second, as in Chapter 5 participants commented that their
AARC is related to a variety of health issues, including levels of pain and subjective
sleep difficulties, I tested whether AARC gains and losses are associated with pain
and subjective sleep difficulties. I found that those individuals experiencing higher
levels of pain and/or of sleep difficulties perceive fewer age-related gains and more
losses.
Third, I explored whether some of the knowledge produced in this thesis for
cognitively healthy older individuals could be translated to people with cognitive
impairment. By combining data for people living with mild-to-moderate dementia
(from the Improving the experience of Dementia and Enhancing Active Life, IDEAL

241
Chapter 8: Discussion

study) with data for cognitively healthy individuals (collected as part of my PhD
through the PROTECT study) I tested whether levels of ATOA vary between people
with and without dementia. Interestingly, I found that there are no differences in
ATOA between people with and without dementia. Consistent with results of this
thesis, those people with dementia with more positive ATOA and/or a younger
subjective age enjoyed better quality of life, higher satisfaction with life, greater
psychological well-being, and were less likely to experience depression. Hence,
positive subjective ageing may not only facilitate health maintenance across healthy
older adults, as has been showed in this thesis, but it may also be a psychological
resource that helps those older individuals who develop illnesses to live better with
the condition. This suggests that findings of this thesis, despite being based on a
sample of individuals with above average physical health, may also apply to those
with poorer health.
Clinical implications. The small association found between more AARC
losses and poorer current cognitive functioning, as well as the association found
between more AARC gains and poorer current cognitive functioning, suggest that
AARC losses should not be used as an indicator of cognitive decline. The AARC
questionnaires may instead be useful in clinical and counselling settings within the
United Kingdom to identify those individuals who, because of higher levels of AARC
losses, may be experiencing poor mental and physical health and would therefore
benefit from interventions promoting adaptation to age-related changes. Depending
on the health status of the individual, either passive or active adaptation to age-
related changes could be promoted (Baltes, 1997; Brothers et al., 2020; Hahn &
Lachman, 2015; Wrosch et al., 2003).
Indeed, on one hand some individuals can experience severe acute illnesses
or debilitating chronic health conditions that negatively impact on several aspects of
their lives and that cannot be improved through treatment of underlying causes. In
these cases, interpreting poorer health as a consequence of increased age (high
AARC losses) may be a realistic understanding of the changes that can come with
ageing. For these individuals disengaging from unachievable goals may be more
adaptive than engaging in active coping (Baltes, 1987, 1997; Brandtstädter &
Rothermund, 2002; Heckhausen & Schulz, 1995; Wrosch et al., 2003). Hence, the
emotional well-being of these individuals may benefit from psychological
interventions, such as mindfulness and psychotherapy, promoting passive coping

242
Chapter 8: Discussion

strategies such as acceptance of negative age-related changes and ego-integrity


(Brothers et al., 2016; Carstensen, 1993; Collins & Kishita, 2018; Dutt et al., 2016a;
Dutt et al., 2018; Erikson, 1959; Greenberg et al., 1986; Phillips & Ferguson, 2012;
Wahl & Ehni, 2020; Westerhof et al., 2015; Xu, 2017; Zanini et al., 2018).
On the other hand, despite older people being more likely to experience a
decline in physical and functional health (Aarts et al., 2012; Barnett et al., 2012;
Brayne et al., 2001; Evandrou & ESRC SAGE Research Group, 2005), for some
individuals levels of physical health can potentially be increased. For these
individuals attributing poorer physical health to something out of one’s control, such
as ageing, can be maladaptive and result in inactivity (Dunbar-Jacob et al., 1995;
Goodwin et al., 1999; Jensen & Karoly, 1991; Turner et al., 2000). Hence, these
individuals may benefit from interventions that promote both more positive views of
ageing and engagement in healthy behaviours that counteract the age-related losses
experienced. So far only six interventions targeting subjective ageing have been
implemented (Beyer et al., 2019; Brothers & Diehl, 2017; Klusmann et al., 2012;
Levy et al., 2014a; Sarkisian et al., 2007; Wolff et al., 2014). The majority of
interventions aiming to change perceptions of ageing promoted both more positive
views on ageing and higher engagement in physical activity (Brothers & Diehl, 2017;
Klusmann et al., 2012; Levy et al., 2014a; Sarkisian et al., 2007; Wolff et al., 2014).
However, one intervention study focused solely on the promotion of more positive
views on ageing (Beyer et al., 2019). Moreover, self-perceptions of ageing can be
changed either indirectly by addressing age-stereotypes (Levy et al., 2014a; Wolff et
al., 2014) or directly by addressing self-perceptions of ageing themselves (Beyer et
al., 2019). In addition, whereas some intervention studies showed that it is possible
to decrease negative views on ageing (Wolff et al., 2014); others showed that it is
possible to promote positive views on ageing (Beyer et al., 2019). In sum, the results
of existing interventions are promising as they show that more positive subjective
ageing can be promoted and result in higher engagement in physical activity and
better physical and mental health. These interventions appear to be effective even
among those older individuals with poor functional health (Beyer et al., 2019).
However, existing intervention studies are scarce and have several
limitations. First, no study has explored whether the positive effects of existing
interventions targeting self-perceptions of ageing over physical health are
transferable to the cognitive domain. Second, only three interventions targeting

243
Chapter 8: Discussion

views on ageing were randomised controlled trials (Beyer et al., 2019; Klusmann et
al., 2012; Wolff et al., 2014), hence there is the need for more rigorous examination
of the efficacy of available interventions targeting views of ageing. Third, the duration
of interventions ranged between 4 weeks to 10 months. There is the need to conduct
studies with longer follow-ups in order to test whether the effects of existing
interventions are long-lasting. Fourth, evidence on the efficacy of interventions
targeting views on ageing in other countries than the US or Germany is lacking
(Beyer et al., 2019; Brothers & Diehl, 2017; Klusmann et al., 2012; Levy et al.,
2014a; Sarkisian et al., 2007; Wolff et al., 2014). Fifth, Intervention studies mainly
focused on older individuals. Only the intervention conducted by Brothers and
colleagues (2017) included individuals aged 50 and over, whereas the remaining
intervention studies targeted individuals aged 61 and over (Levy et al., 2014a), 65
and over (Sarkisian et al., 2007; Wolff et al., 2014), 66 and over (Beyer et al., 2019),
or 70 and over (Klusmann et al., 2012). As the effects of negative views on ageing
and disengagement in health-related behaviours are cumulative over time,
interventions tailored to middle-aged individuals are needed.
Results of this thesis provide some considerations for interventions targeting
subjective ageing. Indeed, among existing interventions, some aimed to decrease
negative views on ageing (Wolff et al., 2014), whereas others aimed to promote
positive views on ageing (Beyer et al., 2019). The research presented in this thesis
(Chapter 7) suggests that decreasing negative views of ageing may be more
effective than increasing positive views of ageing when working with individuals in
poor health. However, targeting both positive and negative views on ageing, as done
by Brothers et al. (2017) may be the preferred strategy when promoting health
maintenance among healthier groups. Moreover, as sociodemographic variables are
related to perceptions of gains and losses, future interventions could target specific
groups at higher risk of high AARC losses and/or be tailored in line with participants’
sociodemographic characteristics. For instance, gendered interventions, such as that
developed by Klusmann et al. (2012), may be more efficacious given the sex
differences found in AARC and subjective ageing (Barrett, 2005; Dolan et al., 2008;
English et al., 2019; Kaminski & Hayslip, 2006; Miche et al., 2014; Moore, 2010).
As middle-aged and older individuals with higher levels of AARC losses are at
higher risk of poor psychological well-being, social prescribing (Brandling & House,
2009) may be useful to promote more positive experiences of ageing among these

244
Chapter 8: Discussion

individuals. Indeed, social prescribing leads to increased mental well-being and


positive mood, self-esteem, social interaction, health and quality of life, reduction in
anxiety and depression, and reduction in the use of primary care consultations
(Brandling & House, 2009; Chatterjee et al., 2018; Dayson & Bashir, 2014; Husk et
al., 2019; Mulligan et al., 2020). Social prescribing is a formal mechanism for linking
patients with non-medical sources of support and services within the community
(City and Hackney Clinical Commissioning Group and University of East London,
2015). Examples of sources of support are voluntary work agencies, further
education, libraries, social or lunch clubs, self-help groups, befriending
organisations, horticulture, sport clubs, nature conservation, and book groups. One
of the gaps in literature is to identify those individuals that may benefit the most from
social prescribing (Brandling & House, 2009); findings of this thesis suggest that
individuals perceiving high AARC losses are a group that may receive greater benefit
from social prescribing (Mulligan et al., 2020).
Policies. As findings of this and existing research on AARC suggest that
negative ATOA predict more AARC losses and fewer gains (Brothers et al., 2020),
promoting positive views of ageing at societal level may promote more positive and
less negative self-perceptions of ageing. Levy (2017) discussed in detail how several
strategies may be useful to optimise views of ageing. Examples of policies could be
abolishing advertisements based on age stereotypes, fighting discrimination in hiring
based on people’s age, modifying the education system by including curricula on
ageing, and promoting contexts that generate intergenerational relationships (Burnes
et al., 2019). Currently some initiatives and programs aiming to reduce ageism are
already in place (e.g. an example is the Reframing Ageing Project, see
https://www.geron.org/programs-services/reframing-aging-initiative).

8.10 Conclusion
The work included in this thesis explored for the first time levels of awareness of
positive and negative age-related change among UK individuals aged 50 and over. It
also related AARC gains and losses to cognitive, mental, and physical health, and
sociodemographic variables. In sum, a global assessment of AARC and an
assessment of AARC specific to the cognitive domain have good psychometric
properties in the United Kingdom. There is great variability in the combination of
levels of AARC gains and losses that individuals experience, and this reflects the

245
Chapter 8: Discussion

heterogeneity that exists in the cognitive, mental, and physical health of middle-aged
and older individuals (Deary et al., 2009; Gow et al., 2007; Jagger et al., 2009).
Levels of AARC gains and losses can also vary in relation to individuals’
sociodemographic characteristics. Even though more perceived losses were related
to a variety of indicators of poorer cognitive, mental, and physical health, the size of
the associations (especially for cognition) was small. Hence, AARC losses somehow
reflect the negative changes in cognitive, mental and physical health that individuals
experience in their lives. AARC gains instead showed negligible associations with
better mental and physical health and with poorer cognition compared to AARC
losses. Moreover, levels of gains may not be informative of health when individuals
experience high losses but only when they experience low losses. Literature
suggests that AARC gains may be related to the social aspects of ageing that have
not been investigated in this thesis (Carstensen, 1992, 1993, 2006; Hobfoll, 2002);
however future research could further test this hypothesis. In sum, the AARC
questionnaire may be useful in the United Kingdom to identify those individuals that
because they report more AARC losses and fewer AARC gains may be experiencing
poorer cognitive, mental, and/or physical health. These individuals could therefore
benefit from interventions helping them to adapt to age-related changes and to
develop more positive perceptions of themselves and of their life while ageing. As
this thesis found that AARC is associated with cognitive, mental, and physical health
at cross-sectional level, further work with longitudinal data is needed to provide more
definite information on the direction of the associations of AARC with health
outcomes and the potential use of AARC to promote health maintenance in older
age.

246
References

References
Aarts, S., Akker, D. M., Bosma, H., Tan, F., Verhey, F., Metsemakers, J., & van
Boxtel, M. (2012). The effect of multimorbidity on health related functioning:
Temporary or persistent? Results from a longitudinal cohort study. Journal of
Psychosomatic Research, 73(3), 211-217.
https://doi.org/10.1016/j.jpsychores.2012.05.014

Abramson, L. Y., Metalsky, G. I., & Alloy, L. B. (1989). Hopelessness depression: A


theory-based subtype of depression. Psychological Review, 96(2), 358-372.
https://doi.org/10.1037/0033-295X.96.2.358

Ackerman, L. S., & Chopik, W. J. (2020). Cross-cultural comparisons in implicit and


explicit age bias. Personality and Social Psychology Bulletin,
0146167220950070. https://doi.org/10.1177/0146167220950070

Acock, A. C. (2013). Discovering structural equation modeling using Stata: Revised


edition. Stata Press.

Age UK. (2019). Later life in the United Kingdom 2019.


https://www.ageuk.org.uk/globalassets/age-uk/documents/reports-and-
publications/later_life_uk_factsheet.pdf

Alavinia, S. M., & Burdorf, A. (2008). Unemployment and retirement and ill-health: A
cross-sectional analysis across European countries. International Archives of
Occupational and Environmental Health, 82(1), 39-45.
https://doi.org/10.1007/s00420-008-0304-6

Alexopoulos, G. S. (2005). Depression in the elderly. The Lancet, 365(9475), 1961-


1970. https://doi.org/10.1016/S0140-6736(05)66665-2

247
References

Alexopoulos, G. S., Meyers, B. S., Young, R. C., Mattis, S., & Kakuma, T. (1993).
The course of geriatric depression with "reversible dementia": A controlled
study. The American Journal of Psychiatry, 150, 1693-1699.
https://doi.org/10.1176/ajp.150.11.1693

Allen, P. M., Mejía, S. T., & Hooker, K. (2015). Personality, self-perceptions, and
daily variability in perceived usefulness among older adults. Psychology and
Aging, 30(3), 534-543. https://doi.org/10.1037/pag0000039

Althubaiti, A. (2016). Information bias in health research: Definition, pitfalls, and


adjustment methods. Journal of Multidisciplinary Healthcare, 9, 211-217.
https://doi.org/10.2147/JMDH.S104807

Amariglio, R. E., Becker, J. A., Carmasin, J., Wadsworth, L. P., Lorius, N., Sullivan,
C., Maye, J. E., Gidicsin, C., Pepin, L. C., Sperling, R. A., Johnson, K. A., &
Rentz, D. M. (2012). Subjective cognitive complaints and amyloid burden in
cognitively normal older individuals. Neuropsychologia, 50(12), 2880-2886.
https://doi.org/10.1016/j.neuropsychologia.2012.08.011

Amariglio, R. E., Buckley, R. F., Mormino, E. C., Marshall, G. A., Johnson, K. A.,
Rentz, D. M., & Sperling, R. A. (2018). Amyloid-associated increases in
longitudinal report of subjective cognitive complaints. Alzheimer's & Dementia,
4, 444-449. https://doi.org/10.1016/j.trci.2018.08.005

American Psychiatric Association. (2000). Diagnostic and Statistical Manual of


Mental Disorders (4th., text rev. ed.). American Psychiatric Association.

American Psychiatric Association. (2013). Diagnostic and Statistical Manual of


Mental Disorders (DSM-5®). American Psychiatric Association.

248
References

Andreas, S., Schulz, H., Volkert, J., Dehoust, M., Sehner, S., Suling, A., Ausín, B.,
Canuto, A., Crawford, M., Da Ronch, C., Grassi, L., Hershkovitz, Y., Muñoz,
M., Quirk, A., Rotenstein, O., Santos-Olmo, A., Shalev, A., Siegert, J., Weber,
K., & Härter, M. (2017). Prevalence of mental disorders in elderly people: The
European MentDis-ICF65+ study. The British Journal of Psychiatry, 210(2),
125-131. https://doi.org/10.1192/bjp.bp.115.180463

Andrews, F. M., & McKennell, A. C. (1980). Measures of self-reported well-being:


Their affective, cognitive, and other components. Social Indicators Research,
8(2), 127-155. https://doi.org/10.1007/bf00286474

Anstey, K. J. (2013). Optimizing cognitive development over the life course and
preventing cognitive decline: Introducing the Cognitive Health Environment
Life Course Model (CHELM). International Journal of Behavioral
Development, 38(1), 1-10. https://doi.org/10.1177/0165025413512255

Antaramian, S. P., Scott Huebner, E., Hills, K. J., & Valois, R. F. (2010). A dual-
factor model of mental health: Toward a more comprehensive understanding
of youth functioning. The American Journal of Orthopsychiatry, 80(4), 462-
472. https://doi.org/10.1111/j.1939-0025.2010.01049.x

Baali, A., Anstey, K. J., & Luszcz, M. A. (2012). The relationship between change in
self-perceptions of aging and physical functioning in older adults. Psychology
and Aging, 27(3), 750-760. https://doi.org/10.1037/a0027578

Baddeley, A. D. (1968). A 3 min reasoning test based on grammatical


transformation. Psychonomic Science, 10(10), 341-342.
https://doi.org/10.3758/BF03331551

249
References

Balcombe, N. R., & Sinclair, A. (2001). Ageing: Definitions, mechanisms and the
magnitude of the problem. Best Practice & Research Clinical
Gastroenterology, 15(6), 835-849. https://doi.org/10.1053/bega.2001.0244

Baltes, P. B. (1987). Theoretical propositions of life-span developmental psychology:


On the dynamics between growth and decline. Developmental Psychology,
23(5), 611-626. https://doi.org/10.1037/0012-1649.23.5.611

Baltes, P. B. (1997). On the incomplete architecture of human ontogeny: Selection,


optimization, and compensation as foundation of developmental theory.
American Psychologist, 52(4), 366-380. https://doi.org/10.1037/0003-
066X.52.4.366

Baltes, P. B., Reese, H. W., & Nesselroade, J. R. (1977). Life-span developmental


psychology: Introduction to research methods. Psychology Press.

Banazak, D. A. (1997). Anxiety disorders in eldelry patients. Clinical review. Journal


of the American Board Family Practice 10(4), 280-289.
https://doi.org/10.3122/jabfm.10.4.280

Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change.


Psychological Review, 84(2), 191-215. https://doi.org/10.1037/0033-
295X.84.2.191

Banks, J., & Smith, S. C. (2006). Retirement in the UK. Oxford Review of Economic
Policy, 22(1), 40-56. https://doi.org/10.1093/oxrep/grj003

Barak, B. (1987). Cognitive age: A new multidimensional approach to measuring age


identity. The International Journal of Aging and Human Development, 25(2),
109-128. https://doi.org/10.2190/RR3M-VQT0-B9LL-GQDM

250
References

Barak, B. (2009). Age identity: A cross-cultural global approach. International Journal


of Behavioral Development, 33(1), 2-11.
https://doi.org/10.1077/0165025408099485

Barnett, K., Mercer, S. W., Norbury, M., Watt, G., Wyke, S., & Guthrie, B. (2012).
Epidemiology of multimorbidity and implications for health care, research, and
medical education: A cross-sectional study. The Lancet, 380(9836), 37-43.
https://doi.org/10.1016/S0140-6736(12)60240-2

Baron, E. C., Davies, T., & Lund, C. (2017). Validation of the 10-item Centre for
Epidemiological Studies Depression Scale (CES-D-10) in Zulu, Xhosa and
Afrikaans populations in South Africa. BMC Psychiatry, 17(1), 6.
https://doi.org/10.1186/s12888-016-1178-x

Barrett, A. E. (1999). Social support and life satisfaction among the never married:
Examining the effects of age. Research on Aging, 21(1), 46-72.
https://doi.org/10.1177/0164027599211003

Barrett, A. E. (2003). Socioeconomic status and age identity: The role of dimensions
of health in the subjective construction of age. The Journals of Gerontology,
Series B: Psychological Sciences and Social Sciences, 58(2), 101-109.
https://doi.org/10.1093/geronb/58.2.S101

Barrett, A. E. (2005). Gendered experiences in midlife: Implications for age identity.


Journal of Aging Studies, 19(2), 163-183.
https://doi.org/10.1016/j.jaging.2004.05.002

Beekman, A. T., De Beurs, E., Van Balkom, A. J., Deeg, D. J., Van Dyck, R., & Van
Tilburg, W. (2000). Anxiety and depression in later life: Co-occurrence and

251
References

communality of risk factors. The American Journal of Psychiatry, 157(1), 89-


95. https://doi.org/10.1176/ajp.157.1.89

Bell, I. P. (1970). The double standard. Transaction, 8(1–2), 75-80.

Bennett, R., & Eckman, J. (1973). Attitudes toward aging: A critical examination of
recent literature and implications for future research. In C. Eisdorfer & M. P.
Lawton (Eds.), The Psychology of Adult Development and Aging (pp. 575-
597). American Psychological Association. https://doi.org/10.1037/10044-018

Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological


Bulletin, 107(2), 238-246. https://doi.org/10.1037/0033-2909.107.2.238

Benyamini, Y., & Burns, E. (2019). Views on aging: Older adults’ self-perceptions of
age and of health. European Journal of Ageing.
https://doi.org/10.1007/510433-019-00528-8

Bergland, A., Nicolaisen, M., & Thorsen, K. (2014). Predictors of subjective age in
people aged 40–79 years: A five-year follow-up study. The impact of mastery,
mental and physical health. Aging & Mental Health, 18(5), 653-661.
https://doi.org/10.1080/13607863.2013.869545

Beyer, A. K., Wolff, J. K., Freiberger, E., & Wurm, S. (2019). Are self-perceptions of
ageing modifiable? Examination of an exercise programme with vs. without a
self-perceptions of ageing-intervention for older adults. Psychology & Health,
34(6), 661-676. https://doi.org/10.1080/08870446.2018.1556273

Beyer, A. K., Wolff, J. K., Warner, L. M., Schuz, B., & Wurm, S. (2015). The role of
physical activity in the relationship between self-perceptions of ageing and
self-rated health in older adults. Psychology & Health, 30(6), 671-685.
https://doi.org/10.1080/08870446.2015.1014370

252
References

Bhalla, R. K., Butters, M. A., Becker, J. T., Houck, P. R., Snitz, B. E., Lopez, O. L.,
Aizenstein, H. J., Raina, K. D., DeKosky, S. T., & Reynolds, C. F. (2009).
Patterns of mild cognitive impairment after treatment of depression in the
elderly. The American Journal of Geriatric Psychiatry, 17(4), 308-316.
https://doi.org/10.1097/JGP.0b013e318190b8d8

Blazer, D. G. (2003). Depression in late life: Review and commentary. The Journals
of Gerontology, Series A: Biological Sciences and Medical Sciences, 58(3),
M249-M265. https://doi.org/10.1093/gerona/58.3.M249

Blazer, D. G. (2010). Protection from late life depression. International


Psychogeriatrics, 22(2), 171-173.
https://doi.org/10.1017/S1041610209990895

Bollen, K. A. (2014). Structural equations with latent variables (Vol. 210). John Wiley
& Sons.

Bordone, V., & Arpino, B. (2015). Do grandchildren influence how old you feel?
Journal of Aging and Health, 28(6), 1055-1072.
https://doi.org/10.1177/0898264315618920

Bordone, V., Arpino, B., & Rosina, A. (2019). Forever young? An analysis of the
factors influencing perceptions of ageing. Ageing & Society, 1669-1693.
https://doi.org/10.1017/S0144686X19000084

Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R. (2005).


Comprehensive meta-analysis. In (Version 2) Biostat.

253
References

Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R. (2009).


Introduction to meta-analysis. John Wiley & Sons, Ltd.
https://doi.org/10.1002/9780470743386

Bousquet, J., Kuh, D., Bewick, M., Standberg, T., Farrell, J., Pengelly, R., Joel, M.
E., Rodriguez, M. L., Mercier, J., Bringer, J., Camuzat, T., Bourret, R.,
Bedbrook, A., Kowalski, M. L., Samolinski, B., Bonini, S., Brayne, C., Michel,
J. P., Venne, J., Viriot-Durandal, P., Alonso, J., Avignon, A., Ben-Shlomo, Y.,
Bousquet, P. J., Combe, B., Cooper, R., Hardy, R., Iaccarino, G., Keil, T.,
Kesse-Guyot, E., Momas, I., Ritchie, K., Robine, J. M., Thijs, C., Tischer, C.,
Vellas, B., Zaidi, A., Alonso, F., K., A. R., Andreeva, V., Ankri, J., Arnavielhe,
S., Arshad, H., Augé, P., Berr, C., Bertone, P., Blain, H., Blasimme, A., Buijs,
G. J., Caimmi, D., Carriazo, A., Cesario, A., Coletta, J., Cosco, T., Criton, M.,
Cuisinier, F., Demoly, P., Fernandez-Nocelo, S., Fougère, B., Garcia-
Aymerich, J., Goldberg, M., Guldemond, N., Gutter, Z., Harman, D., Hendry,
A., Heve, D., Illario, M., Jeande, C., Krauss-Etschmann, S., Krys, O., Kula, D.,
Laune, D., Lehmann, S., Maier, D., Malva, J., Matignon, P., Melen, E.,
Mercier, G., Moda, G., Nizinkska, A., Nogues, M., O’Neill, M., Pelissier, J. Y.,
Poethig, D., Porta, D., Postma, D., Puisieux, F., Richards, M., Robalo-
Cordeiro, C., Romano, V., Roubille, F., Schulz, H., Scott, A., Senesse, P.,
Slagter, S., Smit, H. A., Somekh, D., Stafford, M., Suanzes, J., Todo-Bom, A.,
Touchon, J., Traver-Salcedo, V., Van Beurden, M., Varraso, R., Vergara, I.,
Villalba-Mora, E., Wilson, N., Wouters, E., & Zins, M. (2015). Operational
definition of active and healthy ageing (AHA): A conceptual framework. The
Journal of Nutrition, Health & Aging, 19(9), 955-960.
https://doi.org/10.1007/s12603-015-0589-6

Bowling, A., See-Tai, S., Ebrahim, S., Gabriel, Z., & Solanki, P. (2005). Attributes of
age-identity. Ageing & Society, 25(4), 479-500.
https://doi.org/10.1017/S0144686X05003818

254
References

Boyles, A. L., Harris, S. F., Rooney, A. A., & Thayer, K. A. (2011). Forest Plot
Viewer: A new graphing tool. Epidemiology, 22(5), 746-747.
https://doi.org/10.1097/EDE.0b013e318225ba48

Braam, A. W., Copeland, J. R., Delespaul, P. A., Beekman, A. T., Como, A., Dewey,
M., Fichter, M., Holwerda, T. J., Lawlor, B. A., Lobo, A., Magnusson, H.,
Prince, M. J., Reischies, F., Wilson, K. C., & Skoog, I. (2014). Depression,
subthreshold depression and comorbid anxiety symptoms in older Europeans:
Results from the EURODEP concerted action. Journal of Affective Disorders,
155, 266-272. https://doi.org/10.1016/j.jad.2013.11.011

Bradburn, N. M. (1969). The structure of psychological well-being. Aldine.

Brandling, J., & House, W. (2009). Social prescribing in general practice: adding
meaning to medicine. British Journal of General Practice, 59(563), 454-456.

Brandtstädter, J., & Rothermund, K. (2002). The life-course dynamics of goal pursuit
and goal adjustment: A two-process framework. Developmental Review,
22(1), 117-150. https://doi.org/10.1006/drev.2001.0539

Brayne, C., Matthews, F. E., McGee, M. A., & Jagger, C. (2001). Health and ill-health
in the older population in England and Wales. The Medical Research Council
Cognitive Function and Ageing Study (MRC CFAS). Age and Ageing, 30(1),
53-62. https://doi.org/10.1093/ageing/30.1.53

Brewster, P. W. H., Melrose, R. J., Marquine, M. J., Johnson, J. K., Napoles, A.,
MacKay-Brandt, A., Farias, S. T., Reed, B., & Mungas, D. (2014). Life
experience and demographic influences on cognitive function in older adults.
Neuropsychology, 28(6), 846-858. https://doi.org/10.1037/neu0000098

255
References

Brodaty, H., Luscombe, G., Parker, G., Wilhelm, K., Hickie, I., Austin, M. P., &
Mitchell, P. (2001). Early and late onset depression in old age: Different
aetiologies, same phenomenology. Journal of Affective Disorders, 66(2-3),
225-236. https://doi.org/10.1016/S0165-0327(00)00317-7

Brothers, A. F. (2016). Awareness of age-related change (AARC): Measurement,


conceptual status, and role for promoting successful aging [Doctoral
dissertation, Colorado State University]. Forth Collins, Colorado.

Brothers, A. F., & Diehl, M. K. (2017). Feasibility and efficacy of the Aging(Plus)
program: Changing views on aging to increase physical activity. Journal of
Aging and Physical Activity, 25(3), 402-411. https://doi.org/10.1123/japa.2016-
0039

Brothers, A. F., Gabrian, M., Wahl, H.-W., & Diehl, M. K. (2016). Future time
perspective and awareness of age-related change: Examining their role in
predicting psychological well-being. Psychology and Aging, 31(6), 605-617.
https://doi.org/10.1037/pag0000101

Brothers, A. F., Gabrian, M., Wahl, H.-W., & Diehl, M. K. (2019). A new
multidimensional questionnaire to assess awareness of age-related change
(AARC). The Gerontologist, 59(3), e141-e151.
https://doi.org/10.1093/geront/gny006

Brothers, A. F., Kornadt, A. E., Nehrkorn-Bailey, A., Wahl, H.-W., & Diehl, M. K.
(2020). The effects of age stereotypes on physical and mental health are
mediated by self-perceptions of aging. The Journals of Gerontology, Series B:
Psychological Sciences and Social Sciences.
https://doi.org/10.1093/geronb/gbaa176

256
References

Brothers, A. F., Miche, M., Wahl, H.-W., & Diehl, M. K. (2017). Examination of
associations among three distinct subjective aging constructs and their
relevance for predicting developmental correlates. The Journals of
Gerontology, Series B: Psychological Sciences and Social Sciences, 72(4),
547-560. https://doi.org/10.1093/geronb/gbv085

Brown, T. A. (2006). Confirmatory factor analysis for applied research. The Guilford
Press.

Bruce, M. L., McAvay, G. J., Raue, P. J., Brown, E. L., Meyers, B. S., Keohane, D.
J., Jagoda, D. R., & Weber, C. (2002). Major depression in elderly home
health care patients. The American Journal of Psychiatry, 159(8), 1367-1374.
https://doi.org/10.1176/appi.ajp.159.8.1367

Bryant, C., Bei, B., Gilson, K., Komiti, A., Jackson, H., & Judd, F. (2012). The
relationship between attitudes to aging and physical and mental health in
older adults. International Psychogeriatrics, 24(10), 1674-1683.
https://doi.org/10.1017/s1041610212000774

Bryant, F. B., & Veroff, J. (1982). The structure of psychological well-being: A


sociohistorical analysis. Journal of Personality and Social Psychology, 43(4),
653-673. https://doi.org/10.1037/0022-3514.43.4.653

Buchtemann, D., Luppa, M., Bramesfeld, A., & Riedel-Heller, S. (2012). Incidence of
late-life depression: A systematic review. Journal of Affective Disorders,
142(1-3), 172-179. https://doi.org/10.1016/j.jad.2012.05.010

Buckley, R., Saling, M. M., Ames, D., Rowe, C. C., Lautenschlager, N. T., Macaulay,
S. L., Martins, R. N., Masters, C. L., O'Meara, T., & Savage, G. (2013).
Factors affecting subjective memory complaints in the AIBL aging study:

257
References

Biomarkers, memory, affect, and age. International Psychogeriatrics, 25(8),


1307-1315. https://doi.org/10.1017/S1041610213000665

Buckner, R. L. (2004). Memory and executive function in aging and AD: Multiple
factors that cause decline and reserve factors that compensate. Neuron,
44(1), 195-208. https://doi.org/10.1016/j.neuron.2004.09.006

Bugental, E. K., & Bugental, J. F. (1984). Dispiritedness: A new perspective on a


familiar state. 24, 49-67. https://doi.org/10.1177/0022167884241004

Bullinger, M. (1995). German translation and psychometric testing of the SF-36


Health Survey: Preliminary results from the IQOLA project. International
quality of life assessment. Social Science & Medicine 41(10), 1359-1366.
https://doi.org/10.1016/0277-9536(95)00115-N

Bunce, D., Batterham, P. J., Christensen, H., & Mackinnon, A. J. (2014). Causal
associations between depression symptoms and cognition in a community-
based cohort of older adults. The American Journal of Geriatric Psychiatry,
22(12), 1583-1591. https://doi.org/10.1016/j.jagp.2014.01.004

Burke, B. L., Martens, A., & Faucher, E. H. (2010). Two decades of terror
management theory: A meta-analysis of mortality salience research.
Personality and Social Psychology Review, 14(2), 155-195.
https://doi.org/10.1177/1088868309352321

Burmester, B., Leathem, J., & Merrick, P. (2016). Subjective cognitive complaints
and objective cognitive function in aging: A systematic review and meta-
analysis of recent cross-sectional findings. Neuropsychology Review, 26(4),
376-393. https://doi.org/10.1007/s11065-016-9332-2

258
References

Burnes, D., Sheppard, C., Henderson Jr, C. R., Wassel, M., Cope, R., Barber, C., &
Pillemer, K. (2019). Interventions to reduce ageism against older adults: A
systematic review and meta-analysis. American Journal of Public Health,
109(8), e1-e9. https://doi.org/10.2105/AJPH.2019.305123

Buschkuehl, M., Jaeggi, S. M., Hutchison, S., Perrig-Chiello, P., Däpp, C., Müller, M.,
Breil, F., Hoppeler, H., & Perrig, W. J. (2008). Impact of working memory
training on memory performance in old-old adults. Psychology and Aging,
23(4), 743-753. https://doi.org/10.1037/a0014342

Butcher, H., & McGonigal-Kenney, M. (2005). Depression & dispiritedness in later


life: A "gray drizzle of horror" isn't inevitable. American Journal of Nursing,
105, 52-61. https://doi.org/10.1097/00000446-200512000-00032

Byrne, B. M. (2012). Structural equation modeling with Mplus: Basic concepts,


applications, and programming. Routledge/Taylor & Francis Group.

Caddell, L. S., & Clare, L. (2010). The impact of dementia on self and identity: A
systematic review. Clinical Psychology Review, 30(1), 113-126.
https://doi.org/10.1016/j.cpr.2009.10.003

Calasanti, T. (2010). Gender relations and applied research on aging. The


Gerontologist, 50(6), 720-734. https://doi.org/10.1093/geront/gnq085

Calasanti, T., & King, N. (2018). The dynamic nature of gender and aging bodies.
Journal of Aging Studies, 45, 11-17.
https://doi.org/10.1016/j.jaging.2018.01.002

Calasanti, T., Slevin, K. F., & King, N. (2006). Ageism and feminism: From" et
cetera" to center. NWSA journal, 13-30.
https://doi.org/10.1353/nwsa.2006.0004

259
References

Carp, F. M., & Carp, A. (1983). Structural stability of well-being factors across age
and gender, and development of scales of well-being unbiased for age and
gender. Journal of Gerontology, 38(5), 572-581.
https://doi.org/10.1093/geronj/38.5.572

Carstensen, L. L. (1992). Social and emotional patterns in adulthood: Support for


socioemotional selectivity theory. Psychology and Aging, 7(3), 331-338.
https://doi.org/10.1037/0882-7974.7.3.331

Carstensen, L. L. (1993). Motivation for social contact across the life span: A theory
of socioemotional selectivity. In J. E. Jacobs (Ed.), Current Theory and
Research in Motivation (Vol. 40, pp. 209-254). University of Nebraska Press.

Carstensen, L. L. (2006). The influence of a sense of time on human development.


Science, 312(5782), 1913-1915. https://doi.org/10.1126/science.1127488

Carstensen, L. L., Turan, B., Scheibe, S., Ram, N., Ersner-Hershfield, H., Samanez-
Larkin, G. R., Brooks, K. P., & Nesselroade, J. R. (2011). Emotional
experience improves with age: Evidence based on over 10 years of
experience sampling. Psychology and Aging, 26(1), 21-33.
https://doi.org/10.1037/a0021285

Caselli, R. J., Chen, K., Locke, D. E., Lee, W., Roontiva, A., Bandy, D., Fleisher, A.
S., & Reiman, E. M. (2014). Subjective cognitive decline: Self and informant
comparisons. Alzheimer's & Dementia, 10(1), 93-98.
https://doi.org/10.1016/j.jalz.2013.01.003

Centre for Ageing Better. (2015). Retrieved 02/07/2020 from http://laterlife.ageing-


better.org.uk/

260
References

Chan, A., Fung, H., & Cheng, S.-T. (2007). Maintaining self-rated health through
social comparison in old age. The Journals of Gerontology, Series B:
Psychological Sciences and Social Sciences, 62(5), P277-P285.
https://doi.org/10.1093/geronb/62.5.P277

Chapman, S., Weiss, D., Broulíková, H. M., Joyce, J. L., Sunderaraman, P.,
McKeague, I. W., & Cosentino, S. (2019). AGE STEREOTYPES INFLUENCE
SUBJECTIVE COGNITIVE DECLINE VIA DEPRESSION. Alzheimer's &
Dementia, 15(7), P1366.

Chapman, S. A. (2005). Theorizing about aging well: Constructing a narrative.


Canadian Journal on Aging, 24(1), 8-18. https://doi.org/10.1353/cja.2005.0004

Chatterjee, H. J., Camic, P. M., Lockyer, B., & Thomson, L. J. M. (2018). Non-clinical
community interventions: A systematised review of social prescribing
schemes. Arts & Health, 10(2), 97-123.
https://doi.org/10.1080/17533015.2017.1334002

Chen, F. F. (2007). Sensitivity of goodness of fit indexes to lack of measurement


invariance. Structural Equation Modeling: A Multidisciplinary Journal, 14(3),
464-504. https://doi.org/10.1080/10705510701301834

Cheung, G. W., & Rensvold, R. B. (2002). Evaluating goodness-of-fit indexes for


testing measurement invariance. Structural Equation Modeling, 9(2), 233-255.
https://doi.org/10.1207/S15328007SEM0902_5

Chida, Y., & Steptoe, A. (2010). Greater cardiovascular responses to laboratory


mental stress are associated with poor subsequent cardiovascular risk status:
A meta-analysis of prospective evidence. Hypertension, 55(4), 1026-1032.
https://doi.org/10.1161/hypertensionaha.109.146621

261
References

Chippendale, T. (2013). Factors associated with depressive symptoms among elders


in senior residences: The importance of feeling valued by others. Clinical
Gerontologist, 36(2), 162-169. https://doi.org/10.1080/07317115.2012.749321

Choi, N. G., & DiNitto, D. M. (2014). Felt age and cognitive-affective depressive
symptoms in late life. Aging & Mental Health, 18(7), 833-837.
https://doi.org/10.1080/13607863.2014.886669

Christensen, H. (2001). What cognitive changes can be expected with normal


ageing? Australian and New Zealand Journal of Psychiatry, 35(6), 768-775.
https://doi.org/10.1046/j.1440-1614.2001.00966.x

City and Hackney Clinical Commissioning Group and University of East London.
(2015). Shine 2014 final report. Social Prescribing: Integrating GP and
Community Assets for Health.

Clare, L., Whitaker, C. J., & Nelis, S. M. (2010). Appraisal of memory functioning and
memory performance in healthy ageing and early-stage Alzheimer's Disease.
Aging, Neuropsychology, and Cognition, 17(4), 462-491.
https://doi.org/10.1080/13825580903581558

Clare, L., & Woods, R. T. (2004). Cognitive training and cognitive rehabilitation for
people with early-stage Alzheimer's disease: A review. Neuropsychological
Rehabilitation, 14(4), 385-401. https://doi.org/10.1080/09602010443000074

Clare, L., Wu, Y.-T., Teale, J. C., MacLeod, C., Matthews, F., Brayne, C., Woods, B.,
& CFAS-Wales study team. (2017). Potentially modifiable lifestyle factors,
cognitive reserve, and cognitive function in later life: A cross-sectional study.
PLoS medicine, 14(3), e1002259-e1002259.
https://doi.org/10.1371/journal.pmed.1002259

262
References

Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and


Psychological Measurement, 20(1), 37-46.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Lawrence
Earlbaum Associates.

Collins, R. N., & Kishita, N. (2018). The effectiveness of mindfulness- and


acceptance-based interventions for informal caregivers of people with
dementia: A meta-analysis. The Gerontologist, e363-e379.
https://doi.org/10.1093/geront/gny024

Connidis, I. (1989). The subjective experience of aging: correlates of divergent


views. Canadian Journal on Aging, 8.
https://doi.org/10.1017/S0714980800011168

Conwell, Y., Duberstein, P. R., & Caine, E. D. (2002). Risk factors for suicide in later
life. Biological Psychiatry, 52(3), 193-204. https://doi.org/10.1016/S0006-
3223(02)01347-1

Cook, S. M., & Marsiske, M. (2006). Subjective memory beliefs and cognitive
performance in normal and mildly impaired older adults. Aging & Mental
Health, 10(4), 413-423. https://doi.org/10.1080/13607860600638487

Corbett, A., Owen, A., Hampshire, A., Grahn, J., Stenton, R., Dajani, S., Burns, A.,
Howard, R., Williams, N., Williams, G., & Ballard, C. (2015). The effect of an
online cognitive training package in healthy older adults: An online
randomized controlled trial. Journal of the American Medical Directors
Association, 16(11), 990-997. https://doi.org/10.1016/j.jamda.2015.06.014

263
References

Corner, L., & Bond, J. (2006). The impact of the label of mild cognitive impairment on
the individual's sense of self. Philosophy, Psychiatry, & Psychology, 13(1), 3-
12. https://doi.org/10.1353/ppp.2006.0036

Cosco, T. D., Armstrong, J. J., Stephan, B. C. M., & Brayne, C. (2015a). Successful
aging and frailty: Mutually exclusive paradigms or two ends of a shared
continuum? Canadian Geriatrics Journal: CGJ, 18(1), 35-36.
https://doi.org/10.5770/cgj.18.131

Cosco, T. D., Prina, A. M., Perales, J., Stephan, B. C. M., & Brayne, C. (2013). Lay
perspectives of successful ageing: A systematic review and meta-
ethnography. BMJ Open, 3(6). https://doi.org/10.1136/bmjopen-2013-002710

Cosco, T. D., Prina, A. M., Perales, J., Stephan, B. C. M., & Brayne, C. (2014).
Operational definitions of successful aging: A systematic review. International
Psychogeriatrics, 26(3), 373-381.
https://doi.org/10.1017/S1041610213002287

Cosco, T. D., Stephan, B. C. M., & Brayne, C. (2015b). Validation of an a priori,


index model of successful aging in a population-based cohort study: The
successful aging index. International Psychogeriatrics, 27(12), 1971-1977.
https://doi.org/10.1017/S1041610215000708

Costa, P. T., & McCrae, R. R. (1992). Normal personality assessment in clinical


practice: The NEO Personality Inventory. Psychological Assessment, 4(1), 5-
13. https://doi.org/10.1037/1040-3590.4.1.5

Craciun, C., Gellert, P., & Flick, U. (2017). Aging in precarious circumstances: Do
positive views on aging make a difference? The Gerontologist, 57(3), 517-
528. https://doi.org/10.1093/geront/gnv135

264
References

Crane, M. K., Bogner, H. R., Brown, G. K., & Gallo, J. J. (2007). The link between
depressive symptoms, negative cognitive bias and memory complaints in
older adults. Aging & Mental Health, 11(6), 708-715.
https://doi.org/10.1080/13607860701368497

Critical Appraisal Skills Programme. (2018). CASP (cohort study checklist).


https://casp-uk.net/casp-tools-checklists/

Cronbach, L. J. (1951). Coefficient Alpha and the internal structure of tests.


Psychometrika, 16, 297-334.

D' Mello, D. A. (2003). Epidemiology of late-life depression. In J. E. Ellison & S.


Verma (Eds.), Depression in Later Life (pp. 1-26). CRC Press.

Daviglus, M. L., Bell, C. C., Berrettini, W., Bowen, P. E., Connolly, E. S. J., Cox, N.
J., Dunbar-Jacob, J. M., Granieri, E. C., Hunt, G., McGarry, K., Patel, D.,
Potosky, A. L., Sanders-Bush, E., Silberberg, D., & Trevisan, M. (2010, Apr
28). NIH state-of-the-science conference statement: Preventing Alzheimer's
disease and cognitive decline. NIH Consensus and State of the Science
Statements, 27(4), 1-30.

Dayson, C., & Bashir, N. (2014). The social and economic impact of the Rotherham
Social Prescribing Pilot: Main evaluation report.

De Roover, K., Timmerman, M. E., De Leersnyder, J., Mesquita, B., & Ceulemans,
E. (2014). What's hampering measurement invariance: Detecting non-
invariant items using clusterwise simultaneous component analysis. Frontiers
in Psychology, 5, 604. https://doi.org/10.3389/fpsyg.2014.00604

Deary, I. J., Corley, J., Gow, A. J., Harris, S. E., Houlihan, L. M., Marioni, R. E.,
Penke, L., Rafnsson, S. B., & Starr, J. M. (2009). Age-associated cognitive

265
References

decline. British Medical Bulletin, 92, 135-152.


https://doi.org/10.1093/bmb/ldp033

Deary, I. J., Whalley, L. J., Lemmon, H., Crawford, J. R., & Starr, J. M. (2000). The
stability of individual differences in mental ability from childhood to old age:
Follow-up of the 1932 Scottish Mental Survey. Intelligence, 28(1), 49-55.
https://doi.org/10.1016/S0160-2896(99)00031-8

Denkinger, M., Weyerhäuser, K., Nikolaus, T., & Coll-Planas, L. (2008). Reliability of
the abbreviated version of the Late Life Function and Disability Instrument - a
meaningful and feasible tool to assess physical function and disability in the
elderly. Zeitschrift für Gerontologie und Geriatrie, 42(1), 28-38.
https://doi.org/10.1007/s00391-008-0550-y

DerSimonian, R., & Laird, N. (1986). Meta-analysis in clinical trials. Controlled


Clinical Trials, 7(3), 177-188. https://doi.org/10.1016/0197-2456(86)90046-2

Diehl, M. K., & Wahl, H.-W. (2010). Awareness of age-related change: Examination
of a (mostly) unexplored concept. The Journals of Gerontology, Series B:
Psychological Sciences and Social Sciences, 65B(3), S340-S350.
https://doi.org/10.1093/geronb/gbp110

Diehl, M. K., Wahl, H.-W., Barrett, A. E., Brothers, A. F., Miche, M., Montepare, J.
M., Westerhof, G. J., & Wurm, S. (2014). Awareness of aging: Theoretical
considerations on an emerging concept. Developmental Review, 34(2), 93-
113. https://doi.org/10.1016/j.dr.2014.01.001

Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The Satisfaction With
Life Scale. Journal of Personality Assessment, 49(1), 71-75.
https://doi.org/10.1207/s15327752jpa4901_13

266
References

Dolan, P., Peasgood, T., & White, M. (2008). Do we really know what makes us
happy? A review of the economic literature on the factors associated with
subjective well-being. Journal of Economic Psychology, 29(1), 94-122.
https://doi.org/10.1016/j.joep.2007.09.001

Dooley, E., & Kunik, M. E. (2017). Depression and anxiety across the age spectrum.
In A. Sharafkhaneh, A. M. Yohannes, N. A. Hanania, & M. E. Kunik (Eds.),
Depression and Anxiety in Patients with Chronic Respiratory Diseases (pp.
11-31). Springer. https://doi.org/10.1007/978-1-4939-7009-4_2

Dunbar-Jacob, J., Burke, L. E., & Puczynski, S. (1995). Clinical assessment and
management of adherence to medical regimens. American Psychological
Association. https://doi.org/10.1037/10511-009

Dutt, A. J., Gabrian, M., & Wahl, H.-W. (2016a). Awareness of age-related change
and depressive symptoms in middle and late adulthood: Longitudinal
associations and the role of self-regulation and calendar age. The Journals of
Gerontology, Series B: Psychological Sciences and Social Sciences, 73(6),
944-953. https://doi.org/10.1093/geronb/gbw095

Dutt, A. J., Gabrian, M., & Wahl, H.-W. (2016b). Developmental regulation and
awareness of age-related change: A (mostly) unexplored connection. The
Journals of Gerontology, Series B: Psychological Sciences and Social
Sciences, 73(6), 934-943. https://doi.org/10.1093/geronb/gbw084

Dutt, A. J., & Wahl, H.-W. (2018). Future time perspective and general self-efficacy
mediate the association between awareness of age-related losses and
depressive symptoms. European Journal of Ageing, 16(2), 227-236.
https://doi.org/10.1007/s10433-018-0482-3

267
References

Dutt, A. J., Wahl, H.-W., & Rupprecht, F. (2018). Mindful vs. mind full: Processing
strategies moderate the association between subjective aging experiences
and depressive symptoms. Psychology and Aging, 33, 630-642.
https://doi.org/10.1037/pag0000245

Ebner, N. C., Freund, A. M., & Baltes, P. B. (2006). Developmental changes in


personal goal orientation from young to late adulthood: From striving for gains
to maintenance and prevention of losses. Psychology and Aging, 21(4), 664-
678. https://doi.org/10.1037/0882-7974.21.4.664

Elliott, R., Fischer, C. T., & Rennie, D. L. (1999). Evolving guidelines for publication
of qualitative research studies in psychology and related fields. British Journal
of Clinical Psychology, 38(3), 215-229.
https://doi.org/10.1348/014466599162782

English, A. N., Bellingtier, J. A., & Neupert, S. D. (2019). It’s “the Joneses”: The
influence of objective and subjective socioeconomic status on subjective
perceptions of aging. European Journal of Ageing, 16(1), 121-128.
https://doi.org/10.1007/s10433-018-0475-2

Erikson, E. (1959). Identity and the life cycle. International University Press.

Evandrou, M., & ESRC SAGE Research Group. (2005). Health and well-being
amongst older people in Britain at the start of the twenty-first century.

Evans, I. E. M., Llewellyn, D. J., Matthews, F. E., Woods, R. T., Brayne, C., & Clare,
L. (2019). Living alone and cognitive function in later life. Archives of
Gerontology and Geriatrics, 81, 222-233.
https://doi.org/10.1016/j.archger.2018.12.014

268
References

Faul, F., Erdfelder, E., Buchner, A., & Lang, A.-G. (2009). Statistical power analyses
using G*Power 3.1: Tests for correlation and regression analyses. Behavior
Research Methods, 41, 1149-1160. https://doi.org/10.3758/BRM.41.4.1149

Fauth, E. B., Gerstorf, D., Ram, N., & Malmberg, B. (2012). Changes in depressive
symptoms in the context of disablement processes: Role of demographic
characteristics, cognitive function, health, and social support. The Journals of
Gerontology, Series B: Psychological Sciences and Social Sciences, 67(2),
167-177. https://doi.org/10.1093/geronb/gbr078

Feher, E. P., Larrabee, G. J., & Crook, T. H. r. (1992). Factors attenuating the
validity of the Geriatric Depression Scale in a dementia population. Journal of
the American Geriatrics Society, 40(9), 906-909.
https://doi.org/10.1111/j.1532-5415.1992.tb01988.x

Festinger, L. (1954). A theory of social comparison processes. Human Relations,


7(2), 117-140.

Fiske, A., Wetherell, J. L., & Gatz, M. (2009). Depression in older adults. Annual
Review of Clinical Psychology, 5, 363-389.
https://doi.org/10.1146/annurev.clinpsy.032408.153621

Forman, D. E., Berman, A. D., McCabe, C. H., Baim, D. S., & Wei, J. Y. (1992).
PTCA in the elderly: The “young-old” versus the “old-old”. Journal of the
American Geriatrics Society, 40(1), 19-22. https://doi.org/10.1111/j.1532-
5415.1992.tb01823.x

Freeman, A. T., Santini, Z. I., Tyrovolas, S., Rummel-Kluge, C., Haro, J. M., &
Koyanagi, A. (2016). Negative perceptions of ageing predict the onset and
persistence of depression and anxiety: Findings from a prospective analysis

269
References

of the Irish Longitudinal Study on Ageing (TILDA). Journal of Affective


Disorders, 199, 132-138. https://doi.org/10.1016/j.jad.2016.03.042

French, D. J., Sargent-Cox, K., & Luszcz, M. A. (2012). Correlates of subjective


health across the aging lifespan: Understanding self-rated health in the oldest
old. Journal of Aging and Health, 24(8), 1449-1469.
https://doi.org/10.1177/0898264312461151

Freund, A. M., & Baltes, P. B. (1998). Selection, optimization, and compensation as


strategies of life management: Correlations with subjective indicators of
successful aging. Psychology and Aging, 13(4), 531-543.
https://doi.org/10.1037/0882-7974.13.4.531

Fries, J. F. (2002). Aging, natural death, and the compression of morbidity. Bulletin
of the World Health Organization, 80, 245-250.
https://doi.org/10.1056/NEJM198007173030304

Furstenberg, A.-L. (2002). Trajectories of aging: Imagined pathways in later life. The
International Journal of Aging and Human Development, 55(1), 1-24.
https://doi.org/10.2190/3H62-78VR-6D9D-PLMH

Gallo, J. J., Bogner, H. R., Morales, K. H., Post, E. P., Ten Have, T., & Bruce, M. L.
(2005). Depression, cardiovascular disease, diabetes, and two-year mortality
among older, primary-care patients. The American Journal of Geriatric
Psychiatry 13(9), 748-755. https://doi.org/10.1176/appi.ajgp.13.9.748

Gallo, W. T. (2013). The association of retirement with physical and behavioral


health. In M. Wang (Ed.), The Oxford handbook of retirement (pp. 325–338).
Oxford University Press.

270
References

Gatz, M., Reynolds, C. A., Fratiglioni, L., Johansson, B., Mortimer, J. A., Berg, S.,
Fiske, A., & Pedersen, N. L. (2006). Role of genes and environments for
explaining Alzheimer disease. Archives of General Psychiatry, 63(2), 168-174.
https://doi.org/10.1001/archpsyc.63.2.168

Gaugler, J. E., Kane, R. L., Kane, R. A., Clay, T., & Newcomer, R. (2003).
Caregiving and institutionalization of cognitively impaired older people:
Utilizing dynamic predictors of change. The Gerontologist, 43(2), 219-229.
https://doi.org/10.1093/geront/43.2.219

Gauthier, S., Reisberg, B., Zaudig, M., Petersen, R. C., Ritchie, K., Broich, K.,
Belleville, S., Brodaty, H., Bennett, D., Chertkow, H., Cummings, J. L., de
Leon, M., Feldman, H., Ganguli, M., Hampel, H., Scheltens, P., Tierney, M.
C., Whitehouse, P., & Winblad, B. (2006). Mild cognitive impairment. The
Lancet, 367(9518), 1262-1270. https://doi.org/10.1016/S0140-6736(06)68542-
5

Giles, H., McIlrath, M., Mulac, A., & McCann, R. M. (2010). Expressing age salience:
Three generations' reported events, frequencies, and valences. International
Journal of the Sociology of Language, 2010(206), 73-91.
https://doi.org/10.1515/ijsl.2010.049

Glaesmer, H., Grande, G., Braehler, E., & Roth, M. (2011). The German version of
the Satisfaction with Life Scale (SWLS): Psychometric properties, validity, and
population-based norms. European Journal of Psychological Assessment,
27(2), 127-132. https://doi.org/10.1027/1015-5759/a000058

Goodwin, J. S., Black, S. A., & Satish, S. (1999). Aging versus disease: The opinions
of older black, Hispanic, and non-Hispanic white Americans about the causes
and treatment of common medical conditions. Journal of the American
Geriatrics Society, 47(8), 973-979. https://doi.org/10.1111/j.1532-
5415.1999.tb01293.x

271
References

Gow, A. J., Pattie, A., Whiteman, M. C., Whalley, L. J., & Deary, I. J. (2007). Social
support and successful aging: Investigating the relationships between lifetime
cognitive change and life satisfaction. Journal of Individual Differences, 28(3),
103-115. https://doi.org/10.1027/1614-0001.28.3.103

Grbich, C. (2012). Qualitative data analysis: An introduction. Sage.

Greenberg, J., Pyszczynski, T., & Solomon, S. (1986). The causes and
consequences of a need for self-esteem: A terror management theory. In R.
F. Baumeister (Ed.), Public Self and Private Self (Baumeister R.F. ed., pp.
189-192). Springer. https://doi.org/10.1007/978-1-4613-9564-5_10

Greer, S., Morris, T., Pettingale, K. W., & Haybittle, J. L. (1990). Psychological
response to breast cancer and 15-year outcome. The Lancet, 335(8680), 49-
50. https://doi.org/10.1016/0140-6736(90)90173-3

Gregorich, S. E. (2006). Do self-report instruments allow meaningful comparisons


across diverse population groups? Testing measurement invariance using the
confirmatory factor analysis framework. Medical Care, 44(11 Suppl 3), S78-
S94. https://doi.org/10.1097/01.mlr.0000245454.12228.8f

Grønkjær, M., Osler, M., Flensborg-Madsen, T., Sørensen, H. J., & Mortensen, E. L.
(2019). Associations between education and age-related cognitive changes
from early adulthood to late midlife. Psychology and Aging, 34(2), 177-186.
https://doi.org/10.1037/pag0000332

Grossman, M. R., & Gruenewald, T. L. (2017). Caregiving and perceived


generativity: A positive and protective aspect of providing care? Clinical
Gerontologist, 40(5), 435-447.
https://doi.org/10.1080/07317115.2017.1317686

272
References

Hahn, E. A., & Lachman, M. E. (2015). Everyday experiences of memory problems


and control: The adaptive role of selective optimization with compensation in
the context of memory decline. Neuropsychology, Development, and
Cognition. Section B, Aging, Neuropsychology and Cognition, 22(1), 25-41.
https://doi.org/10.1080/13825585.2014.888391

Hampshire, A., Highfield, R. R., Parkin, B. L., & Owen, A. M. (2012). Fractionating
human intelligence. Neuron, 20(76), 1225-1237.
https://doi.org/10.1016/j.neuron.2012.06.022

Harris, L. A., & Dollinger, S. M. C. (2003). Individual differences in personality traits


and anxiety about aging. Personality and Individual Differences, 34(2), 187-
194. https://doi.org/10.1016/S019-8869(02)00019-3

Heckhausen, J., Dixon, R. A., & Baltes, P. B. (1989). Gains and losses in
development throughout adulthood as perceived by different adult age
groups. Developmental Psychology, 25(1), 109-121.
https://doi.org/10.1037/0012-1649.25.1.109

Heckhausen, J., & Schulz, R. (1995). A life-span theory of control. Psychological


Review, 102(2), 284-304. https://doi.org/10.1037/0033-295x.102.2.284

Hedden, T., & Gabrieli, J. D. E. (2004). Insights into the ageing mind: A view from
cognitive neuroscience. Nature Reviews Neuroscience, 5, 87-97.
https://doi.org/10.1038/nrn1323

Hendrie, H. C., Albert, M. S., Butters, M. A., Gao, S., Knopman, D. S., Launer, L. J.,
Yaffe, K., Cuthbert, B. N., Edwards, E., & Wagster, M. V. (2006). The NIH
Cognitive and Emotional Health Project. Report of the critical evaluation study

273
References

committee. Alzheimer's & Dementia, 2(1), 12-32.


https://doi.org/10.1016/j.jalz.2005.11.004

Herd, P., Goesling, B., & House, J. S. (2007). Socioeconomic position and health:
The differential effects of education versus income on the onset versus
progression of health problems. Journal of Health and Social Behavior, 48(3),
223-238. https://doi.org/10.1177/002214650704800302

Hess, T. M. (2006). Attitudes toward aging and their effects on behavior. In UCLA
Center on Aging (Ed.), Handbook of the Psychology of Aging (pp. 379-406).
Elsevier.

Heun, R., Papassotiropoulos, A., Jessen, F., Maier, W., & Breitner, J. C. (2001). A
family study of Alzheimer disease and early- and late-onset depression in
elderly patients. Archives of General Psychiatry, 58(2), 190-196.
https://doi.org/10.1001/archpsyc.58.2.190

Hicks, S. A. (2017). Examining the influence of perceptions of aging on objective


physical functioning and self-rated health [Doctoral dissertation, Fordham
University]. ProQuest Dissertations & Theses Global. Ann Arbor.
https://search.proquest.com/docview/1914323863?accountid=10792

Higgins, J. P., Thompson, S. G., Deeks, J. J., & Altman, D. G. (2003). Measuring
inconsistency in meta-analyses. BMJ, 327(7414), 557-560.
https://doi.org/10.1136/bmj.327.7414.557

Hill, N. L., Mogle, J., Wion, R., Munoz, E., DePasquale, N., Yevchak, A. M., & Parisi,
J. M. (2016). Subjective cognitive impairment and affective symptoms: A
systematic review. The Gerontologist, 56(6), e109-e127.
https://doi.org/10.1093/geront/gn-w091

274
References

Ho, C., Feng, L., Fam, J., Mahendran, R., Kua, E. H., & Ng, T. P. (2014). Coexisting
medical comorbidity and depression: Multiplicative effects on health outcomes
in older adults. International Psychogeriatrics, 26(7), 1221-1229.
https://doi.org/10.1017/s1041610214000611

Hobfoll, S. E. (2002). Social and psychological resources and adaptation. Review of


general psychology, 6(4), 307-324. https://doi.org/10.1037/1089-2680.6.4.307

Hooker, K., Mejía, S. T., Phibbs, S., Tan, E. J., & Stevens, J. (2019). Effects of age
discrimination on self-perceptions of aging and cancer risk behaviors. The
Gerontologist, 59(Supplement_1), S28-S37.
https://doi.org/10.1093/geront/gny183

Horn, J. L., & McArdle, J. J. (1992). A practical and theoretical guide to


measurement invariance in aging research. Experimental Aging Research,
18(3), 117-144. https://doi.org/10.1080/03610739208253916

Hsieh, C.-M. (2005). Age and relative importance of major life domains. Journal of
Aging Studies, 19(4), 503-512. https://doi.org/10.1016/j.jaging.2005.07.001

Huber, M., van Vliet, M., Giezenberg, M., Winkens, B., Heerkens, Y., Dagnelie, P.
C., & Knottnerus, J. A. (2016). Towards a ‘patient-centred’ operationalisation
of the new dynamic concept of health: A mixed methods study. BMJ Open,
6(1), e010091. https://doi.org/10.1136/bmjopen-2015-010091

Hubley, A. M., & Russell, L. B. (2009). Prediction of subjective age, desired age, and
age satisfaction in older adults: Do some health dimensions contribute more
than others? International Journal of Behavioral Development, 33(1), 12-21.
https://doi.org/10.1177/0165025408099486

275
References

Hudes, R., Rich, J. B., Troyer, A. K., Yusupov, I., & Vandermorris, S. (2019). The
impact of memory-strategy training interventions on participant-reported
outcomes in healthy older adults: A systematic review and meta-analysis.
Psychology and Aging, 34(4), 587-597. https://doi.org/10.1037/pag0000340

Hughes, M. E., & Waite, L. J. (2009). Marital biography and health at mid-life.
Journal of Health and Social Behavior, 50(3), 344-358.
https://doi.org/10.1177/002214650905000307

Huntley, J. D., Corbett, A., Wesnes, K., Brooker, H., Stenton, R., Hampshire, A., &
Ballard, C. (2018). Online assessment of risk factors for dementia and
cognitive function in healthy adults. International Journal of Geriatric
Psychiatry, 33(2), e286-e293. https://doi.org/10.1002/gps.4790

Huntley, J. D., Hampshire, A., Bor, D., Owen, A., & Howard, R. J. (2017). Adaptive
working memory strategy training in early Alzheimer's disease: Randomised
controlled trial. The British Journal of Psychiatry, 210(1), 61-66.
https://doi.org/10.1192/bjp.bp.116.182048

[Record #1722 is using a reference type undefined in this output style.]

Hvas, L. (2006). Menopausal women's positive experience of growing older.


Maturitas, 54(3), 245-251. https://doi.org/10.1016/j.maturitas.2005.11.006

Idler, E. L., & Benyamini, Y. (1997). Self-rated health and mortality: A review of
twenty-seven community studies. Journal of Health and Social Behavior,
38(1), 21-37. https://doi.org/10.2307/2955359

Jaconelli, A., Terracciano, A., Sutin, A. R., Sarrazin, P., Raffard, S., & Stephan, Y.
(2017). Subjective age and dementia. Clinical Gerontologist, 40(2), 106-113.
https://doi.org/10.1080/07317115.2016.1187695

276
References

Jagger, C., Matthews, R., Lindesay, J., Robinson, T., Croft, P., & Brayne, C. (2009).
The effect of dementia trends and treatments on longevity and disability: A
simulation model based on the MRC Cognitive Function and Ageing Study
(MRC CFAS). Age and Ageing, 38(3), 319-325.
https://doi.org/10.1093/ageing/afp016

Jang, Y., Haley, W. E., Small, B. J., & Mortimer, J. A. (2002). The role of mastery
and social resources in the associations between disability and depression in
later life. The Gerontologist, 42(6), 807-813.
https://doi.org/10.1093/geront/42.6.807

Jensen, M. P., & Karoly, P. (1991). Control beliefs, coping efforts, and adjustment to
chronic pain. Journal of Consulting and Clinical Psychology, 59(3), 431-438.
https://doi.org/10.1037//0022-006x.59.3.431

Jessen, F., Amariglio, R. E., van Boxtel, M., Breteler, M., Ceccaldi, M., Chetelat, G.,
Dubois, B., Dufouil, C., Ellis, K. A., van der Flier, W. M., Glodzik, L., van
Harten, A. C., de Leon, M. J., McHugh, P., Mielke, M. M., Molinuevo, J. L.,
Mosconi, L., Osorio, R. S., Perrotin, A., Petersen, R. C., Rabin, L. A., Rami,
L., Reisberg, B., Rentz, D. M., Sachdev, P. S., de la Sayette, V., Saykin, A. J.,
Scheltens, P., Shulman, M. B., Slavin, M. J., Sperling, R. A., Stewart, R.,
Uspenskaya, O., Vellas, B., Visser, P. J., & Wagner, M. (2014). A conceptual
framework for research on subjective cognitive decline in preclinical
Alzheimer's disease. Alzheimer's & Dementia, 10(6), 844-852.
https://doi.org/10.1016/j.jalz.2014.01.001

Jessen, F., Wiese, B., Cvetanovska, G., Fuchs, A., Kaduszkiewicz, H., Kölsch, H.,
Luck, T., Mösch, E., Pentzek, M., Riedler-Heller, S. G., Werle, J., Weyerer, S.,
Zimmermann, T., Maier, W., & Bickel, H. (2007). Patterns of subjective
memory impairment in the elderly: Association with memory performance.

277
References

Psychological Medicine, 1753-1762.


https://doi.org/10.1017/S0033291707001122

Jorm, A. F. (1994). A short form of the Informant Questionnaire on Cognitive Decline


in the Elderly (IQCODE): Development and cross-validation. Psychological
Medicine, 24(1), 145-153. https://doi.org/10.1017/S003329170002691X

Jorm, A. F., & Jacomb, P. A. (1989). The Informant Questionnaire on Cognitive


Decline in the Elderly (IQCODE): Socio-demographic correlates, reliability,
validity and some norms. Psychological Medicine, 19(4), 1015-1022.
https://doi.org/10.1017/S0033291700005742

Jylha, M. (2009). What is self-rated health and why does it predict mortality?
Towards a unified conceptual model. Social Science & Medicine 69(3), 307-
316. https://doi.org/10.1016/j.socscimed.2009.05.013

Jylha, M., Guralnik, J. M., Balfour, J., & Fried, L. P. (2001). Walking difficulty, walking
speed, and age as predictors of self-rated health: The women's health and
aging study. The Journals of Gerontology, Series A: Biological Sciences and
Medical Sciences, 56(10), M609-617.
https://doi.org/10.1093/gerona/56.10.M609

Kaminski, P. L., & Hayslip, B. (2006). Gender differences in body esteem among
older adults. Journal of Women & Aging, 18(3), 19-35.
https://doi.org/10.1300/J074v18n03_03

Kaplan, R. M., Spittel, M. L., & Zeno, T. L. (2014). Educational attainment and life
expectancy. Health and Well-Being, 1(1), 189-194.
https://doi.org/10.1177/2372732214549754

278
References

Karp, D. A. (1988). A decade of reminders: Changing age consciousness between


fifty and sixty years old. The Gerontologist, 28(6), 727-738.
https://doi.org/10.1093/geront/28.6.727

Kasckow, J. W., Karp, J. F., Whyte, E., Butters, M., Brown, C., Begley, A., Bensasi,
S., & Reynolds, C. F., 3rd. (2013). Subsyndromal depression and anxiety in
older adults: Health related, functional, cognitive and diagnostic implications.
Journal of Psychiatric Research, 47(5), 599-603.
https://doi.org/10.1016/j.jpsychires.2013.01.017

Kaspar, R., Gabrian, M., Brothers, A. F., Wahl, H.-W., & Diehl, M. K. (2019).
Measuring awareness of age-related change: Development of a 10-Item short
form for use in large-scale surveys. The Gerontologist, 59(3), e130-e140.
https://doi.org/10.1093/geront/gnx213

Kassem, A. M., Ganguli, M., Yaffe, K., Hanlon, J. T., Lopez, O. L., Wilson, J. W., &
Cauley, J. A. (2017). Anxiety symptoms and risk of cognitive decline in older
community-dwelling men. International Psychogeriatrics, 29(7), 1137-1145.
https://doi.org/10.1017/S104161021700045X

Kastenbaum, R., Derbin, V., Sabatini, P., & Artt, S. (1972). “The ages of me”:
Toward personal and interpersonal definitions of functional aging. The
International Journal of Aging and Human Development, 3(2), 197-211.
https://doi.org/10.2190/tujr-wtxk-866q-8qu7

Kaufman, S. R. (1986). The ageless self: Sources of meaning in late life. University
of Wisconsin Press.

Kavirajan, H., Vahia, I. V., Thompson, W. K., Depp, C., Allison, M., & Jeste, D. V.
(2011). Attitude toward own aging and mental health in post-menopausal

279
References

women. Asian Journal of Psychiatry, 4(1), 26-30.


https://doi.org/10.1016/j.ajp.2011.01.006

Kearney, M. W. (2018). Cross-Lagged Panel Analysis. Sage.


https://doi.org/10.4135/9781483381411

Keller, M. L., Leventhal, E. A., & Larson, B. (1989). Aging: The lived experience. The
International Journal of Aging and Human Development, 29(1), 67-82.
https://doi.org/10.2190/DEQQ-AAUV-NBU0-3RMY

Kemp, A. H., Hopkinson, P. J., Stephan, B. C. M., Clark, C. R., Gordon, E., Bryant,
R. A., & Williams, L. M. (2006). Predicting severity of non-clinical depression:
Preliminary findings using an integrated approach. Journal of Integrative
Neuroscience, 05(01), 89-110. https://doi.org/10.1142/S0219635206001069

Kemp, A. H., Stephan, B. C. M., Hopkinson, P. J., Sumich, A. L., Paul, R. H., Richard
Clark, C., Gordon, E., Bryant, R. A., & Williams, L. M. (2005). Toward an
integrated profile of depression: Evidence from the brain resource
international database. Journal of Integrative Neuroscience, 04(01), 95-106.
https://doi.org/10.1142/S0219635205000665

Kessler, R. C., Andrews, G., Mroczek, D., Ustun, B., & Wittchen, H.-U. (1998). The
World Health Organization Composite International Diagnostic Interview
short-form (CIDI-SF). International Journal of Methods in Psychiatric
Research, 7(4), 171-185. https://doi.org/10.1002/mpr.47

Kim, J. E., & Moen, P. (2002). Retirement transitions, gender, and psychological
well-being: A life-course, ecological model. The Journals of Gerontology,
Series B: Psychological Sciences and Social Sciences, 57(3), P212-P222.
https://doi.org/10.1093/geronb/57.3.P212

280
References

Kim, S. H. (2009). Older people’s expectations regarding ageing, health-promoting


behaviour and health status. Journal of Advanced Nursing, 65(1), 84-91.
https://doi.org/10.1111/j.1365-2648.2008.04841.x

Kim, Y., Kim, K., Boerner, K., & Han, G. (2018). Aging together: Self-perceptions of
aging and family experiences among Korean baby boomer couples. The
Gerontologist, 58(6), 1044-1053. https://doi.org/10.1093/geront/gnx132

Kingston, A., Robinson, L., Booth, H., Knapp, M., & Jagger, C. (2018). Projections of
multi-morbidity in the older population in England to 2035: Estimates from the
Population Ageing and Care Simulation (PACSim) model. Age and Ageing,
47(3), 374-380. https://doi.org/10.1093/ageing/afx201

Kingston, A., Wohland, P., Wittenberg, R., Robinson, L., Brayne, C., Matthews, F. E.,
Jagger, C., Green, E., Gao, L., & Barnes, R. (2017). Is late-life dependency
increasing or not? A comparison of the Cognitive Function and Ageing
Studies (CFAS). The Lancet, 390(10103), 1676-1684.
https://doi.org/10.1016/S0140-6736(17)31575-1

Kleinspehn-Ammerlahn, A., Kotter-Grühn, D., & Smith, J. (2008). Self-perceptions of


aging: Do subjective age and satisfaction with aging change during old age?
The Journals of Gerontology, Series B: Psychological Sciences and Social
Sciences, 63(6), S377-S385. https://doi.org/10.1093/geronb/63.6.

Klusmann, V., Evers, A., Schwarzer, R., & Heuser, I. (2012). Views on aging and
emotional benefits of physical activity: Effects of an exercise intervention in
older women. Psychology of Sport and Exercise, 13(2), 236-242.
https://doi.org/10.1016/j.psychsport.2011.11.001

Klusmann, V., Sproesser, G., Wolff, J. K., & Renner, B. (2017). Positive self-
perceptions of aging promote healthy eating behavior across the life span via

281
References

social-cognitive processes. The Journals of Gerontology, Series B:


Psychological Sciences and Social Sciences, 1-10.
https://doi.org/10.1093/geronb/gbx139

Knapp, M., & Wong, G. (2020). Economics and mental health: The current scenario.
World Psychiatry, 19(1), 3-14. https://doi.org/10.1002/wps.20692

Koppara, A., Wagner, M., Lange, C., Ernst, A., Wiese, B., König, H.-H.,
Brettschneider, C., Riedel-Heller, S., Luppa, M., & Weyerer, S. (2015).
Cognitive performance before and after the onset of subjective cognitive
decline in old age. Alzheimer's & Dementia, 1(2), 194-205.
https://doi.org/10.1016/j.dadm.2015.02.005

Kornadt, A. E., Hess, T. M., & Rothermund, K. (2020). Domain-specific views on


aging and preparation for age-related changes—Development and validation
of three brief scales. The Journals of Gerontology, Series B: Psychological
Sciences and Social Sciences, 75(2), 303-307.
https://doi.org/10.1093/geronb/gby055

Kornadt, A. E., Hess, T. M., Voss, P., & Rothermund, K. (2018). Subjective age
across the life span: A differentiated, longitudinal approach. The Journals of
Gerontology, Series B: Psychological Sciences and Social Sciences, 73(5),
767-777. https://doi.org/10.1093/geronb/gbw072

Kornadt, A. E., Kessler, E.-M., Wurm, S., Bowen, C. E., Gabrian, M., & Klusmann, V.
(2019a). Views on ageing: A lifespan perspective. European Journal of
Ageing. https://doi.org/10.1007/s10433-019-00535-9

Kornadt, A. E., & Rothermund, K. (2011). Contexts of aging: Assessing evaluative


age stereotypes in different life domains. The Journals of Gerontology, Series

282
References

B: Psychological Sciences and Social Sciences, 66(5), 547-556.


https://doi.org/10.1093/geronb/gbr036

Kornadt, A. E., & Rothermund, K. (2012). Internalization of age stereotypes into the
self-concept via future self-views: A general model and domain-specific
differences. Psychology and Aging, 27(1), 164-172.
https://doi.org/10.1037/a0025110

Kornadt, A. E., Voss, P., Fung, H. H., Hess, T. M., & Rothermund, K. (2019b).
Preparation for old age: The role of cultural context and future perceptions.
The Journals of Gerontology, Series B: Psychological Sciences and Social
Sciences, 74(4), 609-619. https://doi.org/10.1093/geronb/gby075

Kornadt, A. E., Voss, P., & Rothermund, K. (2013). Multiple standards of aging:
Gender-specific age stereotypes in different life domains. European Journal of
Ageing, 10(4), 335-344. https://doi.org/10.1007/s10433-013-0281-9

Kotter-Grühn, D., & Hess, T. M. (2012). The impact of age stereotypes on self-
perceptions of aging across the adult lifespan. The Journals of Gerontology,
Series B: Psychological Sciences and Social Sciences, 67, 563-571.
https://doi.org/10.1093/geronb/gbr153

Kotter-Grühn, D., Kleinspehn-Ammerlahn, A., Gerstorf, D., & Smith, J. (2009). Self-
perceptions of aging predict mortality and change with approaching death: 16-
year longitudinal results from the Berlin Aging Study. Psychology and Aging,
24(3), 654-667. https://doi.org/10.1037/a0016510

Kotter-Grühn, D., Kornadt, A. E., & Stephan, Y. (2016). Looking beyond


chronological age: Current knowledge and future directions in the study of
subjective age. Gerontology, 62, 86-93. https://doi.org/10.1159/000438671

283
References

Kroenke, K., Spitzer, R. L., & Williams, J. B. (2001). The PHQ-9: Validity of a brief
depression severity measure. Journal of General Internal Medicine, 16(9),
606-613. https://doi.org/10.1046/j.1525-1497.2001.016009606.x

Kuh, D., Karunananthan, S., Bergman, H., & Cooper, R. (2014). A life-course
approach to healthy ageing: Maintaining physical capability. The Proceedings
of the Nutrition Society, 73(2), 237-248.
https://doi.org/10.1017/s0029665113003923

Kvaal, K., McDougall, F. A., Brayne, C., Matthews, F. E., & Dewey, M. E. (2008). Co-
occurrence of anxiety and depressive disorders in a community sample of
older people: Results from the MRC CFAS (Medical Research Council
Cognitive Function and Ageing Study). International Journal of Geriatric
Psychiatry, 23(3), 229-237. https://doi.org/10.1002/gps.1867

Kwak, M., Ingersoll-Dayton, B., & Burgard, S. (2014). Receipt of care and depressive
symptoms in later life: The importance of self-perceptions of aging. The
Journals of Gerontology, Series B: Psychological Sciences and Social
Sciences, 69(2), 325-335. https://doi.org/10.1093/geronb/gbt128

Kwak, S., Kim, H., Chey, J., & Youm, Y. (2018). Feeling how old I am: Subjective
age is associated with estimated brain age. Frontiers in Aging Neuroscience
10, 1-11. https://doi.org/10.3389/fnagi.2018.00168

Lachman, M. E., Andreoletti, C., & Pearman, A. (2006). Memory control beliefs: How
are they related to age, strategy use and memory improvement? Social
Cognition, 24(3), 359-385. https://doi.org/10.1521/soco.2006.24.3.359

Lara, J., Godfrey, A., Evans, E., Heaven, B., Brown, L. J., Barron, E., Rochester, L.,
Meyer, T. D., & Mathers, J. C. (2013). Towards measurement of the healthy

284
References

ageing phenotype in lifestyle-based intervention studies. Maturitas, 76(2),


189-199. https://doi.org/10.1016/j.maturitas.2013.07.007

Lawton, M. P. (1975). The Philadelphia Geriatric Center Morale Scale: A revision.


Journal of Gerontology, 30(1), 85-89.

Lawton, M. P., & Brody, E. M. (1969). Assessment of oder people: Self-maintaining


and instrumental activities of daily living. The Gerontologist, 9(3), 179-186.
https://doi.org/10.1093/guront/9.3Part1.179

Le Roux, H., Gatz, M., & Wetherell, J. L. (2005). Age at onset of generalized anxiety
disorder in older adults. The American Journal of Geriatric Psychiatry, 13(1),
23-30. https://doi.org/10.1176/appi.ajgp.13.1.23

Lehrner, J., Kogler, S., Lamm, C., Moser, D., Klug, S., Pusswald, G., Dal-Bianco, P.,
Pirker, W., & Auff, E. (2015). Awareness of memory deficits in subjective
cognitive decline, mild cognitive impairment, Alzheimer's disease and
Parkinson's disease. International Psychogeriatrics, 27(3), 357-366.
https://doi.org/10.1017/s1041610214002245

Leopold, L., & Engelhardt, H. (2013). Education and physical health trajectories in
old age. Evidence from the Survey of Health, Ageing and Retirement in
Europe (SHARE). International Journal of Public Health, 58(1), 23-31.
https://doi.org/10.1007/s00038-012-0399-0

Lerner, R. M. (1984). On the nature of human plasticity. Cambridge University Press.


https://doi.org/10.1017/CBO9780511666988

Leventhal, E. A., & Prohaska, T. R. (1986). Age, symptom interpretation, and health
behavior. Journal of the American Geriatrics Society, 34(3), 185-191.

285
References

Leventhal, H., Phillips, L. A., & Burns, E. (2016). The Common-Sense Model of Self-
Regulation (CSM): A dynamic framework for understanding illness self-
management. Journal of Behavioral Medicine, 39(6), 935-946.
https://doi.org/10.1007/s10865-016-9782-2

Levy, B. R. (2003). Mind matters: Cognitive and physical effects of aging self-
stereotypes. The Journals of Gerontology, Series B: Psychological Sciences
and Social Sciences, 58(4), P203-P211.
https://doi.org/10.1093/geronb/58.4.P203

Levy, B. R. (2009). Stereotype embodiment: A psychosocial approach to aging.


Current Directions in Psychological Science, 18(6), 332-336.
https://doi.org/10.1111/j.1467-8721.2009.01662.x

Levy, B. R. (2017). Age-stereotype paradox: Opportunity for social change. The


Gerontologist, 57(suppl_2), S118-S126. https://doi.org/10.1093/geront/gnx059

Levy, B. R., Ashman, O., & Slade, M. D. (2009a). Age attributions and aging health:
Contrast between the United States and Japan. The Journals of Gerontology,
Series B: Psychological Sciences and Social Sciences, 64(3), 335-338.
https://doi.org/10.1093/geronb/gbp002

Levy, B. R., Ferrucci, L., Zonderman, A. B., Slade, M. D., Troncoso, J., & Resnick, S.
M. (2016). A culture-brain link: Negative age stereotypes predict Alzheimer's
disease biomarkers. Psychology and Aging, 31(1), 82-88.
https://doi.org/10.1037/pag0000062

Levy, B. R., Hausdorff, J. M., Hencke, R., & Wei, J. Y. (2000). Reducing
cardiovascular stress with positive self-stereotypes of aging. The Journals of

286
References

Gerontology, Series B: Psychological Sciences and Social Sciences, 55(4),


205-213. https://doi.org/10.1093/geronb/55.4.p205

Levy, B. R., & Leifheit-Limson, E. (2009). The stereotype-matching effect: Greater


influence on functioning when age stereotypes correspond to outcomes.
Psychology and Aging, 24(1), 230-233. https://doi.org/10.1037/a0014563

Levy, B. R., & Myers, L. M. (2004). Preventive health behaviors influenced by self-
perceptions of aging. Preventive Medicine, 39(3), 625-629.
https://doi.org/10.1016/j.ypmed.2004.02.029

Levy, B. R., Pilver, C., Chung, P. H., & Slade, M. D. (2014a). Subliminal
strengthening: Improving older individuals' physical function over time with an
implicit-age-stereotype intervention. Psychological Science, 25(12), 2127-
2135. https://doi.org/10.1177/0956797614551970

Levy, B. R., Pilver, C. E., & Pietrzak, R. H. (2014b). Lower prevalence of psychiatric
conditions when negative age stereotypes are resisted. Social Science &
Medicine, 119, 170-174. https://doi.org/10.1016/j.socscimed.2014.06.046

Levy, B. R., Slade, M. D., Chang, E.-S., Kannoth, S., & Wang, S.-Y. (2020a). Ageism
amplifies cost and prevalence of health conditions. The Gerontologist, 60(1),
174-181. https://doi.org/10.1093/geront/gny131

Levy, B. R., Slade, M. D., Chung, P. H., & Gill, T. M. (2015). Resiliency over time of
elders' age stereotypes after encountering stressful events. The Journals of
Gerontology, Series B: Psychological Sciences and Social Sciences, 70(6),
886-890. https://doi.org/10.1093/geronb/gbu082

Levy, B. R., Slade, M. D., & Kasl, S. V. (2002a). Longitudinal benefit of positive self-
perceptions of aging on functional health. The Journals of Gerontology, Series

287
References

B: Psychological Sciences and Social Sciences, 57(5), P409-P417.


https://doi.org/10.1093/geronb/57.5.p409

Levy, B. R., Slade, M. D., Kunkel, S. R., & Kasl, S. V. (2002b). Longevity increased
by positive self-perceptions of aging. Journal of Personality and Social
Psychology, 83(2), 261-270. https://doi.org/10.1037/0022-3514.83.2.261

Levy, B. R., Slade, M. D., Pietrzak, R. H., & Ferrucci, L. (2018). Positive age beliefs
protect against dementia even among elders with high-risk gene. PLoS One,
13(2), e0191004. https://doi.org/10.1371/journal.pone.0191004

Levy, B. R., Slade, M. D., Pietrzak, R. H., & Ferrucci, L. (2020b). When culture
influences genes: Positive age beliefs amplify the cognitive-aging benefit of
APOE ϵ2. The Journals of Gerontology, Series B: Psychological Sciences and
Social Sciences.

Levy, B. R., Zonderman, A. B., Slade, M. D., & Ferrucci, L. (2009b). Age stereotypes
held earlier in life predict cardiovascular events in later life. Psychological
Science, 20(3), 296-298. https://doi.org/10.1111/j.1467-9280.2009.02298.x

Lewis, S., & Clarke, M. (2001). Forest plots: Trying to see the wood and the trees.
BMJ, 322(7300), 1479-1480. https://doi.org/10.1136/bmj.322.7300.1479

Lin, M.-C., Hummert, M. L., & Harwoodc, J. (2004). Representation of age identities
in on-line discourse. Journal of Aging Studies, 1325-1334.
https://doi.org/10.1016/j.jaging.2004.03.006

Lindeboom, M., & Kerkhofs, M. (2009). Health and work of the elderly: Subjective
health measures, reporting errors and endogeneity in the relationship
between health and work. Journal of Applied Econometrics, 24(6), 1024-1046.
https://doi.org/10.1002/jae.1077

288
References

Lindsay-Smith, G., Eime, R., O’Sullivan, G., Harvey, J., & van Uffelen, J. G. Z.
(2019). A mixed-methods case study exploring the impact of participation in
community activity groups for older adults on physical activity, health and
wellbeing. BMC Geriatrics, 19(1), 243. https://doi.org/10.1186/s12877-019-
1245-5

Lindsay-Smith, G., O’Sullivan, G., Eime, R., Harvey, J., & Van Uffelen, J. G. Z.
(2018). A mixed methods case study exploring the impact of membership of a
multi-activity, multicentre community group on social wellbeing of older adults.
BMC Geriatrics, 18(1), 226. https://doi.org/10.1186/s12877-018-0913-1

Little, W. (2014). Aging and the elderly. Chapter 13. In O. College (Ed.), Introduction
to Sociology-1st Canadian Edition. BC Open Textbook Project.

Liverman, C. T., Yaffe, K., & Blazer, D. G. (2015). Cognitive aging: Progress in
understanding and opportunities for action. The National Academies Press.
https://doi.org/10.17226/21693

Livingston, G., Huntley, J., Sommerlad, A., Ames, D., Ballard, C., Banerjee, S.,
Brayne, C., Burns, A., Cohen-Mansfield, J., & Cooper, C. (2020). Dementia
prevention, intervention, and care: 2020 report of the Lancet Commission.
The Lancet, 396(10248), 413-446. https://doi.org/10.1016/S0140-
6736(20)30367-6

Loidl, B., & Leipold, B. (2019). Facets of accommodative coping in adulthood.


Psychology and Aging, 34(5), 640-654. https://doi.org/10.1037/pag0000378

Lövdén, M., Backman, L., Lindenberger, U., Schaefer, S., & Schmiedek, F. (2010). A
theoretical framework for the study of adult cognitive plasticity. Psychological
Bulletin, 136(4), 659-676. https://doi.org/10.1037/a0020080

289
References

Lövdén, M., Rönnlund, M., Wahlin, Å., Bäckman, L., Nyberg, L., & Nilsson, L.-G.
(2004). The extent of stability and change in episodic and semantic memory in
old age: Demographic predictors of level and change. The Journals of
Gerontology, Series B: Psychological Sciences and Social Sciences, 59(3),
P130-P134. https://doi.org/10.1093/geronb/59.3.P130

Lubitz, A. F., Eid, M., & Niedeggen, M. (2018). Complainer Profile Identification
(CPI): Properties of a new questionnaire on subjective cognitive complaints.
Aging, Neuropsychology, and Cognition, 25(1), 99-121.
https://doi.org/10.1080/13825585.2016.1267325

Lytle, A., Apriceno, M., Dyar, C., & Levy, S. R. (2018). Sexual orientation and gender
differences in aging perceptions and concerns among older adults. Innovation
in Aging, 2(3), igy036. https://doi.org/10.1093/geroni/igy036

Mackinnon, A., Jorm, A. F., Christensen, H., Korten, A. E., Jacomb, P. A., &
Rodgers, B. (1999). A short form of the Positive and Negative Affect
Schedule: Evaluation of factorial validity and invariance across demographic
variables in a community sample. Personality and Individual Differences,
27(3), 405-416. https://doi.org/10.1016/S0191-8869(98)00251-7

Maier, H., & Smith, J. (1999). Psychological predictors of mortality in old age. The
Journals of Gerontology, Series B: Psychological Sciences and Social
Sciences, 54(1), 44-54. https://doi.org/10.1093/geronb/54b.1.p44

Matthews, F. E., Stephan, B. C. M., Bond, J., & McKeith, I. G. (2007).


Operationalisation of mild cognitive impairment: A graphical approach. PLoS
medicine, 4(10), 1615-1619. https://doi.org/10.1371/journal.pmed.0040304

290
References

Matthews, F. E., Stephan, B. C. M., McKeith, I. G., Bond, J., Brayne, C., Function, M.
R. C. C., & Ageing Study. (2008). Two-year progression from mild cognitive
impairment to dementia: To what extent do different definitions agree? Journal
of the American Geriatrics Society, 56(8), 1424-1433.
https://doi.org/10.1111/j.1532-5415.2008.01820.x

McConatha, J. T., Schnell, F., Volkwein, K., Riley, L., & Leach, E. (2003). Attitudes
toward aging: A comparative analysis of young adults from the United States
and Germany. The International Journal of Aging and Human Development,
57(3), 203-215. https://doi.org/10.2190/K8Q8-5549-0Y4K-UGG0

McCrone, P., Dhanasiri, S., Patel, A., Knapp, M., & Lawton-Smith, S. (2008). Paying
the price: The cost of mental health care in England to 2026. K. s. Fund.
https://www.kingsfund.org.uk/sites/default/files/Paying-the-Price-the-cost-of-
mental-health-care-England-2026-McCrone-Dhanasiri-Patel-Knapp-Lawton-
Smith-Kings-Fund-May-2008_0.pdf

McGrath, R., Al Snih, S., Markides, K., Hall, O., & Peterson, M. (2019). The burden
of health conditions for middle-aged and older adults in the United States:
Disability-adjusted life years. BMC Geriatrics, 19(1), 100.
https://doi.org/10.1186/s12877-019-1110-6

Meeks, T. W., Vahia, I. V., Lavretsky, H., Kulkarni, G., & Jeste, D. V. (2011). A tune
in "a minor" can "b major": A review of epidemiology, illness course, and
public health implications of subthreshold depression in older adults. Journal
of Affective Disorders, 129(1-3), 126-142.
https://doi.org/10.1016/j.jad.2010.09.015

Meisner, B. A. (2012). A meta-analysis of positive and negative age stereotype


priming effects on behavior among older adults. The Journals of Gerontology,
Series B: Psychological Sciences and Social Sciences, 67(1), 13-17.
https://doi.org/10.1093/geronb/gbr062

291
References

Mejía, S. T., Giasson, H. L., Smith, J., & Gonzalez, R. (2020). Concurrent and
enduring associations between married partners’ shared beliefs and markers
of aging. Psychology and Aging. https://doi.org/10.1037/pag0000515

Meléndez, J. C., Satorres, E., Redondo, R., Escudero, J., & Pitarque, A. (2018).
Wellbeing, resilience, and coping: Are there differences between healthy older
adults, adults with mild cognitive impairment, and adults with Alzheimer-type
dementia? Archives of Gerontology and Geriatrics, 77, 38-43.
https://doi.org/10.1016/j.archger.2018.04.004

Mendonça, M. D., Alves, L., & Bugalho, P. (2016). From subjective cognitive
complaints to dementia: Who is at risk?: A systematic review. American
Journal of Alzheimer's Disease & Other Dementias, 31(2), 105-114.
https://doi.org/10.1177/1533317515592331

Meredith, W. (1993). Measurement invariance, factor analysis and factorial


invariance. Psychometrika, 58(4), 525-543.
https://doi.org/10.1007/bf02294825

Miche, M., Brothers, A. F., Diehl, M. K., & Wahl, H.-W. (2015). The role of subjective
aging within the changing ecologies of aging: Perspectives for research and
practice. In M. K. Diehl & H.-W. Wahl (Eds.), Annual Review of Gerontology
and Geriatrics (Vol. 35, pp. 211-245). Springer Publishing Company.
https://doi.org/10.1891/0198-8794.35.211

Miche, M., Wahl, H.-W., Diehl, M. K., Oswald, F., Kaspar, R., & Kolb, M. (2014).
Natural occurrence of subjective aging experiences in community-dwelling
older adults. The Journals of Gerontology, Series B: Psychological Sciences
and Social Sciences, 69(2), 174-187. https://doi.org/10.1093/geronb/gbs164

292
References

Michon, A., Deweer, B., Pillon, B., Agid, Y., & Dubois, B. (1994). Relation of
anosognosia to frontal lobe dysfunction in Alzheimer's disease. Journal of
Neurology, Neurosurgery, and Psychiatry, 57(7), 805-809.
https://doi.org/10.1136/jnnp.57.7.805

Minder, C. E., Müller, T., Gillmann, G., Beck, J. C., & Stuck, A. E. (2002). Subgroups
of refusers in a disability prevention trial in older adults: Baseline and follow-
up analysis. American Journal of Public Health, 92(3), 445-450.

Minino, A. M., Arias, E., Kochanek, K. D., Murphy, S. L., & Smith, B. L. (2002).
Deaths: Final data for 2000. National vital statistics reports.

Mitchell, A. J., & Shiri-Feshki, M. (2009). Rate of progression of mild cognitive


impairment to dementia-meta-analysis of 41 robust inception cohort studies.
Acta Psychiatrica Scandinavica, 119(4), 252-265.
https://doi.org/10.1111/j.1600-0447.2008.01326.x

Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. G. (2009). Preferred reporting items
for systematic reviews and meta-analyses: The PRISMA statement. PLoS
medicine, 6(7), e1000097. https://doi.org/10.1371/journal.pmed.1000097

Montejo, P., Montenegro, M., Fernandez, M. A., & Maestu, F. (2011). Subjective
memory complaints in the elderly: Prevalence and influence of temporal
orientation, depression and quality of life in a population-based study in the
city of Madrid. Aging & Mental Health, 15(1), 85-96.
https://doi.org/10.1080/13607863.2010.501062

Montepare, J. M. (2009). Subjective age: Toward a guiding lifespan framework.


International Journal of Behavioral Development, 33(1), 42-46.
https://doi.org/10.1177/0165025408095551

293
References

Montepare, J. M. (2019). An exploration of subjective age, actual age, age


awareness, and engagement in everyday behaviors. European Journal of
Ageing, 299-307. https://doi.org/10.1007/s10433-019-00534-w

Montepare, J. M., & Lachman, M. E. (1989). "You're only as old as you feel": Self-
perceptions of age, fears of aging, and life satisfaction from adolescence to
old age. Psychology and Aging, 4(1), 73-78. https://doi.org/10.1037/0882-
7974.4.1.73

Moore, S. E. H. (2010). Is the healthy body gendered? Toward a feminist critique of


the new paradigm of health. Body & Society, 16(2), 95-118.
https://doi.org/10.1177/1357034X10364765

Morrison, V. (2008). Ageing and physical health. In R. T. Woods & L. Clare (Eds.),
Handbook of the Clinical Psychology of Ageing. Wiley. J. & Sons.

Moser, C., Spagnoli, J., & Santos-Eggimann, B. (2011). Self-perception of aging and
vulnerability to adverse outcomes at the age of 65-70 years. The Journals of
Gerontology, Series B: Psychological Sciences and Social Sciences, 66, 675-
680. https://doi.org/10.1093/geronb/gbr052

Mulligan, K., Bhatti, S., Rayner, J., & Hsiung, S. (2020). Social prescribing: Creating
pathways towards better health and wellness. Journal of the American
Geriatrics Society, 68(2), 426-428.

Neuman, W. L. (2011). Social research methods: Qualitative and quantitative


approaches. Pearson. https://books.google.co.uk/books?id=rKL9lgWN_1gC

Neupert, S. D., & Bellingtier, J. A. (2017). Aging attitudes and daily awareness of
age-related change interact to predict negative affect. The Gerontologist,
57(suppl_2), S187-S192. https://doi.org/10.1093/geront/gnx055

294
References

Nilsson, M., Sarvimäki, A., & Ekman, S.-L. (2000). Feeling old: Being in a phase of
transition in later life. Nursing Inquiry, 7(1), 41-49.
https://doi.org/10.1046/j.1440-1800.2000.00049.x

Nolen-Hoeksema, S., Wisco, B. E., & Lyubomirsky, S. (2008). Rethinking rumination.


Perspectives on Psychological Science, 3(5), 400-424.
https://doi.org/10.1111/j.1745-6924.2008.00088.x

Nordhus, I. H. (2008). Manifestations of depression and anxiety in older adults. In R.


T. Woods & L. Clare (Eds.), Handbook of the Clinical Psychology of Ageing
(2nd ed., pp. 99-110). Wiley. J. & Sons.

North, M. S., & Fiske, S. T. (2015). Modern attitudes toward older adults in the aging
world: A cross-cultural meta-analysis. Psychological Bulletin, 141(5), 993-
1021. https://doi.org/10.1037/a0039469

Nylund, K. L., Asparouhov, T., & Muthén, B. O. (2007). Deciding on the number of
classes in latent class analysis and growth mixture modeling: A Monte Carlo
simulation study. Structural Equation Modeling: A Multidisciplinary Journal,
14(4), 535-569. https://doi.org/10.1080/10705510701575396

O’Brien, E. L., & Sharifian, N. (2019). Managing expectations: How stress, social
support, and aging attitudes affect awareness of age-related changes. Journal
of Social and Personal Relationships, 37(3), 986-1007.
https://doi.org/10.1177/0265407519883009

O’Brien, E. L., Torres, G. E., & Neupert, S. D. (2020). Cognitive interference in the
context of daily stressors, daily awareness of age-related change, and general
aging attitudes. The Journals of Gerontology, Psychological Sciences: Series
B. https://doi.org/10.1093/geronb/gbaa155

295
References

Ogletree, A., & Katz, B. (2020). How do older adults recruited using MTurk differ
from those in a national probability sample? The International Journal of Aging
and Human Development, 0091415020940197.
https://doi.org/10.1177/0091415020940197

Okonkwo, O. C., Griffith, H. R., Vance, D. E., Marson, D. C., Ball, K. K., & Wadley, V.
G. (2009). Awareness of functional difficulties in mild cognitive impairment: A
multidomain assessment approach. Journal of the American Geriatrics
Society, 57(6), 978-984. https://doi.org/10.1111/j.1532-5415.2009.02261.x

Opdebeeck, C., Yates, J. A., Kudlicka, A., & Martyr, A. (2019). What are subjective
cognitive difficulties and do they matter? Age and Ageing, 48(1), 122-127.
https://doi.org/10.1093/ageing/afy148

Owen, A. M., Beksinska, M., James, M., Leigh, P. N., Summers, B. A., Marsden, C.
D., Quinn, N. P., Sahakian, B. J., & Robbins, T. W. (1993). Visuospatial
memory deficits at different stages of Parkinson's disease. Neuropsychologia,
31(7), 627-644. https://doi.org/10.1016/0028-3932(93)90135-M

Owen, A. M., Downes, J. J., Sahakian, B. J., Polkey, C. E., & Robbins, T. W. (1990).
Planning and spatial working memory following frontal lobe lesions in man.
Neuropsychologia, 28(10), 1021-1034. https://doi.org/10.1016/0028-
3932(90)90137-D

Palsson, S. P., Ostling, S., & Skoog, I. (2001). The incidence of first-onset
depression in a population followed from the age of 70 to 85. Psychological
Medicine, 31(7), 1159-1168. https://doi.org/10.1017/S0033291701004524

Panaite, V., Salomon, K., Jin, A., & Rottenberg, J. (2015). Cardiovascular recovery
from psychological and physiological challenge and risk for adverse

296
References

cardiovascular outcomes and all-cause mortality. Psychosomatic Medicine,


77(3), 215-226. https://doi.org/10.1097/psy.0000000000000171

Park, D. C., & Reuter-Lorenz, P. (2009). The adaptive brain: Aging and
neurocognitive scaffolding. Annual Review of Psychology, 60, 173-196.
https://doi.org/10.1146/annurev.psych.59.103006.093656

Patterson, C. (2018). World Alzheimer Report 2018. The state of the art of dementia
research: New frontiers.
https://www.alz.co.uk/research/WorldAlzheimerReport2018.pdf

Paukert, A. L., Pettit, J. W., Kunik, M. E., Wilson, N., Novy, D. M., Rhoades, H. M.,
Greisinger, A. J., Wehmanen, O. A., & Stanley, M. A. (2010). The roles of
social support and self-efficacy in physical health's impact on depressive and
anxiety symptoms in older adults. Journal of Clinical Psychology in Medical
Settings, 17(4), 387-400. https://doi.org/10.1007/s10880-010-9211-6

Pearman, A., & Storandt, M. (2004). Predictors of subjective memory in older adults.
The Journals of Gerontology, Series B: Psychological Sciences and Social
Sciences, 59(1), P4-P6. https://doi.org/10.1093/geronb/59.1.p4

Pearson, K. (1948). Karl Pearson's early statistical papers. Cambridge: University


Press.

Peter, J., Scheef, L., Abdulkadir, A., Boecker, H., Heneka, M., Wagner, M., Koppara,
A., Kloppel, S., & Jessen, F. (2014). Gray matter atrophy pattern in elderly
with subjective memory impairment. Alzheimer's & Dementia, 10(1), 99-108.
https://doi.org/10.1016/j.jalz.2013.05.1764

297
References

Phillips, W. J., & Ferguson, S. J. (2012). Self-compassion: A resource for positive


aging. The Journals of Gerontology, Series B: Psychological Sciences and
Social Sciences, 68(4), 529-539. https://doi.org/10.1093/geronb/gbs091

Pienta, A. M., Hayward, M. D., & Jenkins, K. R. (2000). Health consequences of


marriage for the retirement years. Journal of Family Issues, 21(5), 559-586.
https://doi.org/10.1177/019251300021005003

Pietrzak, R. H., Maruff, P., Woodward, M., Fredrickson, J., Fredrickson, A., Krystal,
J. H., Southwick, S. M., & Darby, D. (2012). Mild worry symptoms predict
decline in learning and memory in healthy older adults: A 2-year prospective
cohort study. The American Journal of Geriatric Psychiatry, 20(3), 266-275.
https://doi.org/10.1097/JGP.0b013e3182107e24

Pinquart, M. (2001). Correlates of subjective health in older adults: A meta-analysis.


Psychology and Aging, 16(3), 414-426. https://doi.org/10.1037/0882-
7974.16.3.414

Polit, D. F., & Beck, C. T. (2004). Nursing research: Principles and methods (7 ed.).
Lippincott Williams and Wilkins.

Pope, C., Ziebland, S., & Mays, N. (2000). Analysing qualitative data. BMJ,
320(7227), 114-126. https://doi.org/10.1136/bmj.320.7227.114

Porensky, E. K., Dew, M. A., Karp, J. F., Skidmore, E., Rollman, B. L., Shear, M. K.,
& Lenze, E. J. (2009). The burden of late-life generalized anxiety disorder:
Effects on disability, health-related quality of life, and healthcare utilization.
The American Journal of Geriatric Psychiatry, 17(6), 473-482.
https://doi.org/10.1097/JGP.0b013e31819b87b2

298
References

Prince, M., Knapp, M., Guerchet, M., McCrone, P., Prina, M., Comas-Herrera, A.,
Wittenberg, R., Adelaja, B., Hu, B., King, D., Rehill, A., & Salimkumar, D.
(2014). Dementia UK: Update. alzheimers.org.uk/dementiauk

Prince, M., Wimo, A., Guerchet, M., Ali, G. C., Wu, Y.-T., Prina, A. M., & Alzheimer's
Disease International. (2015). World Alzheimer report 2015: The global impact
of dementia Alzheimer's Disease International.
https://www.alz.co.uk/research/world-report-2015

Public Health England. (2018). Chapter 1: Population change and trends in life
expectancy. Retrieved 21-11-2010 from
https://www.gov.uk/government/publications/health-profile-for-england-
2018/chapter-1-population-change-and-trends-in-life-expectancy

Radloff, L. S. (1977). The CES-D Scale: A self-report depression scale for research
in the general population. Applied Psychological Measurement, 1(3), 385-401.
https://doi.org/10.1177/014662167700100306

Rakesh, G., Szabo, S. T., Alexopoulos, G. S., & Zannas, A. S. (2017). Strategies for
dementia prevention: Latest evidence and implications. Therapeutic Advances
in Chronic Disease, 8(8-9), 121-136.
https://doi.org/10.1177/2040622317712442

Rattan, S. I. (2013). Healthy ageing, but what is health? Biogerontology, 14(6), 673-
677. https://doi.org/10.1007/s10522-013-9442-7

Raz, N., & Rodrigue, K. M. (2006). Differential aging of the brain: Patterns, cognitive
correlates and modifiers. Neuroscience & Biobehavioral Reviews, 30(6), 730-
748. https://doi.org/10.1016/j.neubiorev.2006.07.001

299
References

Reijmer, Y., Van den Berg, E., de Bresser, J., Kessels, R., Kappelle, L., Algra, A., &
Biessels, G. (2011). Accelerated cognitive decline in patients with type 2
diabetes: MRI correlates and risk factors. Diabetes Metabolism Research
Review, 27, 195-202. https://doi.org/10.1002/dmrr.1163

Reisberg, B., & Gauthier, S. (2008). Current evidence for subjective cognitive
impairment (SCI) as the pre-mild cognitive impairment (MCI) stage of
subsequently manifest Alzheimer's disease. International Psychogeriatrics,
20(1), 1-16. https://doi.org/10.1017/S1041610207006412

Reisberg, B., Prichep, L., Mosconi, L., John, E. R., Glodzik-Sobanska, L., Boksay, I.,
Monteiro, I., Torossian, C., Vedvyas, A., & Ashraf, N. (2008). The pre–mild
cognitive impairment, subjective cognitive impairment stage of Alzheimer’s
disease. Alzheimer's & Dementia, 4(1), S98-S108.
https://doi.org/10.1016/j.jalz.2007.11.017

Richards, L. (2015). Handling qualitative data: A practical guide (3 ed.). Sage.

Rippon, I., & Steptoe, A. (2018). Is the relationship between subjective age,
depressive symptoms and activities of daily living bidirectional? Social
Science & Medicine, 214, 41-48.
https://doi.org/10.1016/j.socscimed.2018.08.013

Robb, C. E., de Jager, C. A., Ahmadi-Abhari, S., Giannakopoulou, P., Udeh-Momoh,


C., McKeand, J., Price, G., Car, J., Majeed, A., & Ward, H. (2020).
Associations of social isolation with anxiety and depression during the early
COVID-19 pandemic: A survey of older adults in London, UK. Frontiers in
Psychiatry, 11. https://doi.org/10.3389/fpsyt.2020.591120

Roberts, J. L., Clare, L., & Woods, R. T. (2009). Subjective memory complaints and
awareness of memory functioning in mild cognitive impairment: A systematic

300
References

review. Dementia and Geriatric Cognitive Disorders, 28(2), 95-109.


https://doi.org/10.1159/000234911

Robertson, D. A., & Kenny, R. A. (2016a). “I'm too old for that”—The association
between negative perceptions of aging and disengagement in later life.
Personality and Individual Differences, 100, 114-119.
https://doi.org/10.1016/j.paid.2016.03.096

Robertson, D. A., & Kenny, R. A. (2016b). Negative perceptions of aging modify the
association between frailty and cognitive function in older adults. Personality
and Individual Differences, 100, 120-125.
https://doi.org/10.1016/j.paid.2015.12.010

Robertson, D. A., King-Kallimanis, B. L., & Kenny, R. A. (2016). Negative


perceptions of aging predict longitudinal decline in cognitive function.
Psychology and Aging, 31(1), 71-81. https://doi.org/10.1037/pag0000061

Robertson, D. A., Savva, G. M., King-Kallimanis, B. L., & Kenny, R. A. (2015).


Negative perceptions of aging and decline in walking speed: A self-fulfilling
prophecy. PLoS One, 10(4), e0123260.
https://doi.org/10.1371/journal.pone.0123260

Robinson-Whelen, S., Kim, C., MacCallum, R. C., & Kiecolt-Glaser, J. K. (1997).


Distinguishing optimism from pessimism in older adults: Is it more important to
be optimistic or not to be pessimistic? Journal of Personality and Social
Psychology, 73(6), 1345-1353. https://doi.org/10.1037/0022-3514.73.6.1345

Roehr, S., Luck, T., Pabst, A., Bickel, H., König, H.-H., Lühmann, D., Fuchs, A.,
Wolfsgruber, S., Wiese, B., & Weyerer, S. (2017). Subjective cognitive decline
is longitudinally associated with lower health-related quality of life.

301
References

International Psychogeriatrics, 29(12), 1939-1950.


https://doi.org/10.1017/S1041610217001399

Rohwedder, S., & Willis, R. J. (2010). Mental retirement. Journal of Economic


Perspectives, 24(1), 119-138. https://doi.org/10.1257/jep.24.1.119

Rönnlund, M., Sundström, A., Adolfsson, R., & Nilsson, L.-G. (2015). Subjective
memory impairment in older adults predicts future dementia independent of
baseline memory performance: Evidence from the Betula prospective cohort
study. Alzheimer's & Dementia, 11(11), 1385-1392.
https://doi.org/10.1016/j.jalz.2014.11.006

Rosenstock, I. M. (1974). Historical origins of the Health Belief Model. Health


Education Monographs, 2(4), 328-335.
https://doi.org/10.1177/109019817400200403

Rothermund, K. (2005). Effects of age stereotypes on self-views and adaptation. In


W. Greve, K. Rothermund, & D. Ventura (Eds.), The Adaptive Self: Personal
Continuity and Intentional Self-development (pp. 223-242). Hogrefe & Huber
Publishers.

Rothermund, K., De Paula Couto, M., Fung, H. H., Graf, S., Hess, T. M., Liou, S., &
Nikitin, J. (2020). Age-related attributions of experienced changes in life:
Origins and implications. The Journals of Gerontology, Series B:
Psychological Sciences and Social Sciences.
https://doi.org/10.1093/geronb/gbaa160

Rothermund, K., & Kornadt, A. E. (2015). Views on aging: Domain-specific


approaches and implications for developmental regulation. Annual Review of
Gerontology and Geriatrics, 35(1), 121-144. https://doi.org/10.1891/0198-
8794.35.121

302
References

Rowe, J. W., & Kahn, R. L. (1987). Human aging: Usual and successful. Science,
237(4811), 143-149.

Rowe, J. W., & Kahn, R. L. (1997). Successful aging. The Gerontologist, 37(4), 433-
440. https://doi.org/10.1093/geront/37.4.433

Royall, D. R., Lauterbach, E. C., Kaufer, D., Malloy, P., Coburn, K. L., & Black, K. J.
(2007). The cognitive correlates of functional status: A review from the
Committee on Research of the American Neuropsychiatric Association. The
Journal of Neuropsychiatry and Clinical Neurosciences, 19(3), 249-265.
https://doi.org/10.1176/jnp.2007.19.3.249

Rubin, D. C., & Berntsen, D. (2006). People over forty feel 20% younger than their
age: Subjective age across the lifespan. Psychonomic Bulletin & Review,
13(5), 776-780. https://doi.org/10.3758/bf03193996

Rubio, L., Dumitrache, C., Cordon-Pozo, E., & Rubio-Herrera, R. (2016). Coping:
Impact of gender and stressful life events in middle and in old age. Clinical
Gerontologist, 39(5), 468-488.
https://doi.org/10.1080/07317115.2015.1132290

Rupprecht, F. S., Dutt, A. J., Wahl, H.-W., & Diehl, M. K. (2019). The role of
personality in becoming aware of age-related changes. The Journal of
Gerontopsychology and Geriatric Psychiatry, 32(2), 57-67.
https://doi.org/10.1024/1662-9647/a000204

Ryff, C. D., & Keyes, C. L. (1995). The structure of psychological well-being


revisited. Journal of Personality and Social Psychology, 69(4), 719-727.
https://doi.org/10.1037/0022-3514.69.4.719

303
References

Sabatini, S., Silarova, B., Martyr, A., Collins, R., Ballard, C., Anstey, K. J., Kim, S., &
Clare, L. (2020a). Associations of awareness of age-related change with
emotional and physical well-being: A systematic review and meta-analysis.
The Gerontologist, 60(6), e477-e490. https://doi.org/10.1093/geront/gnz101

Sabatini, S., Ukoumunne, O. C., Ballard, C., Brothers, A. F., Kaspar, R., Collins, R.,
Kim, S., Corbett, A., Aarsland, D., Hampshire, A., Brooker, H., & Clare, L.
(2020b). International relevance of two measures of awareness of age-related
change (AARC). BMC Geriatrics, 20(1), 359. https://doi.org/10.1186/s12877-
020-01767-6

Sabatini, S., Ukoumunne, O. C., Ballard, C., Collins, R., Corbett, A., Brooker, H., &
Clare, L. (2020c). The cross-sectional relationship between pain and
awareness of age-related changes. British Journal of Pain.
https://doi.org/10.1177/2049463720961798

Sabatini, S., Woods, R. T., Ukoumunne, O. C., Ballard, C., Collins, R., & Clare, L.
(2021). Associations of subjective cognitive and memory decline with
depression, anxiety, and two-year change in objectively-assessed global
cognition and memory. Aging, Neuropsychology, and Cognition, 1-27.
https://doi.org/10.1080/13825585.2021.1923634

Sable, J. A., Dunn, L. B., & Zisook, S. (2002). Late-life depression. How to identify its
symptoms and provide effective treatment. Geriatrics, 57(2), 31-36.
https://doi.org/10.1177/171516350513800106

Salthouse, T. A. (2019). Trajectories of normal cognitive aging. Psychology and


Aging, 34(1), 17-24. https://doi.org/10.1037/pag0000288

304
References

Sandelowski, M., & Barroso, J. (2003). Classifying the findings in qualitative studies.
Qualitative Health Research, 13(7), 905-923.
https://doi.org/10.1177/1049732303253488

Sargent-Cox, K. A., Anstey, K. J., & Luszcz, M. A. (2012a). Change in health and
self-perceptions of aging over 16 years: The role of psychological resources.
Health Psychology, 31(4), 423-432. https://doi.org/10.1037/a0027464

Sargent-Cox, K. A., Anstey, K. J., & Luszcz, M. A. (2012b). The relationship between
change in self-perceptions of aging and physical functioning in older adults.
Psychology and Aging, 27(3), 750-760. https://doi.org/10.1037/a0027578

Sargent-Cox, K. A., Anstey, K. J., & Luszcz, M. A. (2014). Longitudinal change of


self-perceptions of aging and mortality. The Journals of Gerontology, Series
B: Psychological Sciences and Social Sciences, 69(2), 168-173.
https://doi.org/10.1093/geronb/gbt005

Sarkisian, C. A., Liu, H., Ensrud, K. E., Stone, K. L., & Mangione, C. M. (2001).
Correlates of attributing new disability to old age. Study of Osteoporotic
Fractures Research Group. Journal of the American Geriatrics Society, 49(2),
134-141. https://doi.org/10.1046/j.1532-5415.2001.49033.x

Sarkisian, C. A., Prohaska, T. R., Davis, C., & Weiner, B. (2007). Pilot test of an
attribution retraining intervention to raise walking levels in sedentary older
adults. Journal of the American Geriatrics Society, 55, 1842-1846.
https://doi.org/10.1111/j.1532-5415.2007.01427.x

Sarkisian, C. A., Prohaska, T. R., Wong, M. D., Hirsch, S., & Mangione, C. M.
(2005). The relationship between expectations for aging and physical activity
among older adults. Journal of General Internal Medicine, 20(10), 911-915.
https://doi.org/10.1111/j.1525-1497.2005.0204.x

305
References

Savva, G. M., Stephan, B. C. M., & Alzheimer’s Society Vascular Dementia


Systematic Review Group. (2010). Epidemiological studies of the effect of
stroke on incident dementia: A systematic review. Stroke, 41(1), e41-e46.
https://doi.org/10.1161/STROKEAHA.109.559880

Scarborough, P., Bhatnagar, P., Wickramasinghe, K. K., Allender, S., Foster, C., &
Rayner, M. (2011). The economic burden of ill health due to diet, physical
inactivity, smoking, alcohol and obesity in the UK: An update to 2006-07 NHS
costs. Journal of Public Health 33(4), 527-535.
https://doi.org/10.1093/pubmed/fdr033

Schafer, M. H., & Shippee, T. P. (2009). Age identity, gender, and perceptions of
decline: Does feeling older lead to pessimistic dispositions about cognitive
aging? The Journals of Gerontology, Series B: Psychological Sciences and
Social Sciences, 65B(1), 91-96. https://doi.org/10.1093/geronb/gbp046

Schmidtke, K., Pohlmann, S., & Metternich, B. (2008). The syndrome of functional
memory disorder: Definition, etiology, and natural course. The American
Journal of Geriatric Psychiatry, 16(12), 981-988.
https://doi.org/10.1097/JGP.0b013e318187ddf9

Schoevers, R. A., Deeg, D. J., van Tilburg, W., & Beekman, A. T. (2005). Depression
and generalized anxiety disorder: Co-occurrence and longitudinal patterns in
elderly patients. The American Journal of Geriatric Psychiatry, 13(1), 31-39.
https://doi.org/10.1176/appi.ajgp.13.1.31

Schuurmans, J., & Van Balkom, A. (2011). Late-life anxiety disorders: A review.
Current Psychiatry Reports, 13(4), 267-273. https://doi.org/10.1007/s11920-
011-0204-4

306
References

Segal, D. L., Lebenson, S., & Coolidge, F. L. (2008). Global self-rated health status
predicts reasons for living among older adults. Clinical Gerontologist, 31(4),
122-132. https://doi.org/10.1080/07317110802144006

Segel-Karpas, D., & Palgi, Y. (2019). ‘It is nothing more than a senior moment’: The
moderating role of subjective age in the effect of change in memory on self-
rated memory. Aging & Mental Health, 23(2), 272-276.
https://doi.org/10.1080/13607863.2017.1399350

Seidler, A. L., & Wolff, J. K. (2017). Bidirectional associations between self-


perceptions of aging and processing speed across 3 years. The Journal of
Gerontopsychology and Geriatric Psychiatry, 30, 49-59.
https://doi.org/10.1024/1662-9647/a000165

Settersten, R. A., Jr. (2017). Some things I have learned about aging by studying the
life course. Innovation in Aging, 1(2). https://doi.org/10.1093/geroni/igx014

Sheikh, J. I., Swales, P. J., Carlson, E. B., & Lindley, S. E. (2004). Aging and panic
disorder: Phenomenology, comorbidity, and risk factors. The American
Journal of Geriatric Psychiatry, 12(1), 102-109.
https://doi.org/10.1097/00019442-200401000-00013

Sheldon, J. H. (1947). The social medicine of old age (Vol. 61). Oxford University
Press. https://doi.org/10.1016/S0033-3506(47)80243-4

Sherman, S. R. (1994). Changes in age identity: Self perceptions in middle and late
life. Journal of Aging Studies, 8(4), 397-412. https://doi.org/10.1016/0890-
4065(94)90011-6

307
References

Shin, K. R., Kim, M. Y., & Kim, Y. H. (2003). Study on the lived experience of aging.
Nursing & Health Sciences, 5(4), 245-252. https://doi.org/10.1046/j.1442-
2018.2003.00161.x

Siebert, J. S., Braun, T., & Wahl, H. W. (2020). Change in attitudes toward aging:
Cognitive complaints matter more than objective performance. Psychology
and Aging, 35(3), 357-368. https://doi.org/10.1037/pag0000451

Siebert, J. S., Wahl, H.-W., Degen, C., & Schröder, J. (2018). Attitude toward own
aging as a risk factor for cognitive disorder in old age: 12-year evidence from
the ILSE study. Psychology and Aging, 33(3), 461-472.
https://doi.org/10.1037/pag0000252

Siebert, J. S., Wahl, H.-W., & Schröder, J. (2016). The role of attitude toward own
aging for fluid and crystallized functioning: 12-year evidence from the ILSE
study. The Journals of Gerontology, Series B: Psychological Sciences and
Social Sciences, 73(5), 836-845. https://doi.org/10.1093/geronb/gbw050

Siedlecki, K. L., Salthouse, T. A., & Berish, D. E. (2005). Is there anything special
about the aging of source memory? Psychology and Aging, 20(1), 19-32.
https://doi.org/10.1037/0882-7974.20.1.19

Simone, P. M., & Haas, A. L. (2013). Frailty, leisure activity and functional status in
older adults: Relationship with subjective well being. Clinical Gerontologist,
36(4), 275-293. https://doi.org/10.1080/07317115.2013.788114

Sommerlad, A., Ruegger, J., Singh-Manoux, A., Lewis, G., & Livingston, G. (2018).
Marriage and risk of dementia: Systematic review and meta-analysis of
observational studies. Journal of Neurology, Neurosurgery & Psychiatry,
89(3), 231-238. https://doi.org/10.1136/jnnp-2017-316274

308
References

Sontag, S. (1997). The double standard of aging. Routledge.

Soper, D. Calculator: A-priori sample size for structural equation models.


https://www.danielsoper.com/statcalc/calculator.aspx?id=89

Spitzer, R. L., Kroenke, K., Williams, J. B., & Lowe, B. (2006). A brief measure for
assessing generalized anxiety disorder: The GAD-7. Archives of Internal
Medicine, 166(10), 1092-1097. https://doi.org/10.1001/archinte.166.10.1092

Spuling, S. M., Miche, M., Wurm, S., & Wahl, H.-W. (2013). Exploring the causal
interplay of subjective age and health dimensions in the second half of life: A
cross-lagged panel analysis. Zeitschrift für Gesundheitspsychologie, 21(1), 5-
15. https://doi.org/10.1026/0943-8149/a000084

StataCorp. (2017). Stata Statistical Software: Release 16. College Station, TX2016.

Staudinger, U. M. (2015). Images of aging: Outside and inside perspectives. Annual


Review of Gerontology and Geriatrics, 35(1), 187-209.
https://doi.org/10.1891/0198-8794.35.187

Staudinger, U. M., Finkelstein, R., Calvo, E., & Sivaramakrishnan, K. (2016). A


global view on the effects of work on health in later life. The Gerontologist,
56(Suppl_2), S281-S292. https://doi.org/10.1093/geront/gnw032

Stephan, B. C. M., & Brayne, C. (2008). Vascular factors and prevention of


dementia. International Review of Psychiatry, 20(4), 344-356.
https://doi.org/10.1080/09540260802094456

Stephan, Y., Caudroit, J., & Chalabaev, A. (2011). Subjective health and memory
self-efficacy as mediators in the relation between subjective age and life

309
References

satisfaction among older adults. Aging & Mental Health, 15(4), 428-436.
https://doi.org/10.1080/13607863.2010.536138

Stephan, Y., Caudroit, J., Jaconelli, A., & Terracciano, A. (2014). Subjective age and
cognitive functioning: A 10-year prospective study. The American Journal of
Geriatric Psychiatry, 22(11), 1180-1187.
https://doi.org/10.1016/j.jagp.2013.03.007

Stephan, Y., Sutin, A. R., Caudroit, J., & Terracciano, A. (2016a). Subjective age and
changes in memory in older adults. The Journals of Gerontology, Series B:
Psychological Sciences and Social Sciences, 71(4), 675-683.
https://doi.org/10.1093/geronb/gbv010

Stephan, Y., Sutin, A. R., Kornadt, A., & Terracciano, A. (2019). Polygenic scores for
education, health, and personality as predictors of subjective age among older
individuals of European ancestry: Evidence from the Health and Retirement
Study. Psychology and Aging, 34(1), 139-144.
https://doi.org/10.1037/pag0000283

Stephan, Y., Sutin, A. R., & Terracciano, A. (2015). Subjective age and personality
development: A 10-year study. Journal of Personality, 83(2), 142-154.
https://doi.org/10.1111/jopy.12090

Stephan, Y., Sutin, A. R., & Terracciano, A. (2016b). Feeling older and risk of
hospitalization: Evidence from three longitudinal cohorts. Health Psychology,
35(6), 634-637. https://doi.org/10.1037/hea0000335

Steverink, N., Westerhof, G. J., Bode, C., & Dittmann-Kohli, F. (2001). The personal
experience of aging, individual resources, and subjective well-being. The
Journals of Gerontology, Series B: Psychological Sciences and Social
Sciences, 56(6), 364-373. https://doi.org/10.1093/geronb/56.6.P364

310
References

Stewart, R., Godin, O., Crivello, F., Maillard, P., Mazoyer, B., Tzourio, C., & Dufouil,
C. (2011). Longitudinal neuroimaging correlates of subjective memory
impairment: 4-year prospective community study. The British Journal of
Psychiatry, 198(3), 199-205. https://doi.org/10.1192/bjp.bp.110.078683

Sullivan, K. W., Law, W. A., Loyola, L., Knoll, M. A., Shub, D. E., & French, L. M.
(2020). A novel intervention Platform for Service Members With Subjective
Cognitive Complaints: Implementation, patient participation, and satisfaction.
Military medicine, 185(Supplement_1), 326-333.
https://doi.org/10.1093/milmed/usz218

Szanto, K., Dombrovski, A. Y., Sahakian, B. J., Mulsant, B. H., Houck, P. R.,
Reynolds, C. F., & Clark, L. A. (2012). Social emotion recognition, social
functioning, and attempted suicide in late-life depression. The American
Journal of Geriatric Psychiatry, 20(3), 257-265.
https://doi.org/10.1097/JGP.0b013e31820eea0c

Tandetnik, C., Farrell, M. T., Cary, M. S., Cines, S., Emrani, S., Karlawish, J., &
Cosentino, S. (2015). Ascertaining subjective cognitive decline: A comparison
of approaches and evidence for using an age-anchored reference group.
Journal of Alzheimer's Disease, 48(s1), S43-S55. https://doi.org/10.3233/JAD-
150251

Tandetnik, C., Hergueta, T., Bonnet, P., Dubois, B., & Bungener, C. (2017).
Influence of early maladaptive schemas, depression, and anxiety on the
intensity of self-reported cognitive complaint in older adults with subjective
cognitive decline. International Psychogeriatrics, 29(10), 1657-1667.
https://doi.org/10.1017/S1041610217001119

Tavakol, M., & Dennick, R. (2011). Making sense of Cronbach's Alpha. International
Journal of Medical Education, 2, 53-55. https://doi.org/10.5116/ijme.4dfb.8dfd

311
References

Teal, R., Enga, Z., Diehl, S. J., Rohweder, C. L., Kim, M., Dave, G., Durr, A., Wynn,
M., Isler, M. R., Corbie-Smith, G., & Weiner, B. J. (2015). Applying cognitive
interviewing to inform measurement of partnership readiness: A new
approach to strengthening community-academic research. Progress in
Community Health Partnerships: Research, Education, and Action, 9(4), 513-
519. https://doi.org/10.1353/cpr.2015.0083

Terracciano, A., Stephan, Y., Aschwanden, D., Lee, J. H., Sesker, A. A.,
Strickhouser, J. E., Luchetti, M., & Sutin, A. R. (2021). Changes in subjective
age during COVID-19. The Gerontologist, 61(1), 13-22.
https://doi.org/10.1093/geront/gnaa104

Teuscher, U. (2009). Subjective age bias: A motivational and information processing


approach. International Journal of Behavioral Development, 33(1), 22-31.
https://doi.org/10.1177/0165025408099487

Thomae, H. (1979). The concept of development and life-span developmental


psychology (Vol. 2). Academic Press.

Timmer, E., Steverink, N., & Dittmann-Kohli, F. (2002). Cognitive representations of


future gains, maintenance, and losses in the second half of life. The
International Journal of Aging and Human Development, 55(4), 321-339.
https://doi.org/10.1177/009141500205500401

Turner, J. A., Jensen, M. P., & Romano, J. M. (2000). Do beliefs, coping, and
catastrophizing independently predict functioning in patients with chronic
pain? Pain, 85(1-2), 115-125. https://doi.org/10.1016/s0304-3959(99)00259-6

312
References

Turner, S. G., Hooker, K., & Geldhof, G. J. (2020). Self-perceptions of aging:


Factorial structure and invariance by gender. The Gerontologist.
https://doi.org/10.1093/geront/gnaa059

United Nations. (2019). Levels of trends in population ageing. Retrieved 21/11/2019


from http://www.un.org/en/sections/issues-depth/ageing/

Uotinen, V., Rantanen, T., & Suutama, T. (2005). Perceived age as a predictor of old
age mortality: A 13-year prospective study. Age and Ageing, 34(4), 368-372.
https://doi.org/10.1093/ageing/afi091

Van der Weele, G. M., Gussekloo, J., De Waal, M. W. M., De Craen, A. J. M., & Van
der Mast, R. C. (2009). Co-occurrence of depression and anxiety in elderly
subjects aged 90 years and its relationship with functional status, quality of life
and mortality. International Journal of Geriatric Psychiatry, 24(6), 595-601.
https://doi.org/10.1002/gps.2162

Vassilaki, M., Aakre, J. A., Cha, R. H., Kremers, W. K., St. Sauver, J. L., Mielke, M.
M., Geda, Y. E., Machulda, M. M., Knopman, D. S., Petersen, R. C., &
Roberts, R. O. (2015). Multimorbidity and risk of mild cognitive impairment.
Journal of the American Geriatrics Society, 63(9), 1783-1790.
https://doi.org/10.1111/jgs.13612

Victor, C., Scambler, S., Bond, J., & Bowling, A. (2000). Being alone in later life:
Loneliness, social isolation and living alone. Reviews in Clinical Gerontology,
10(4), 407-417. https://doi.org/10.1017/S0959259800104101

Voss, P., Kornadt, A. E., Hess, T. M., Fung, H. H., & Rothermund, K. (2018). A world
of difference? Domain-specific views on aging in China, the US, and
Germany. Psychology and Aging, 33(4), 595-606.
https://doi.org/10.1037/pag0000237

313
References

Wahl, H.-W., & Ehni, H.-J. (2020). Advanced old age as a developmental dilemma:
An in-depth comparison of established fourth age conceptualizations. Journal
of Aging Studies, 55, 100896.

Wahl, H.-W., Konieczny, C., & Diehl, M. K. (2013). Experiencing age-related change
in adulthood: An exploratory study based on the concept of “Awareness of
Age-Related Change” (AARC). Zeitschrift für Entwicklungspsychologie und
Pädagogische Psychologie, 45(2), 66-76. https://doi.org/10.1026/0049-
8637/a000081

Ware, J. E. J., & Sherbourne, C. D. (1992). The MOS 36-item short-form health
survey (SF-36). Conceptual framework and item selection. Medical Care,
30(6), 473-483. https://doi.org/10.1177/016327879702000102

Ware, J. J. E., Kosinski, M., Bjorner, J. B., Turner-Bowker, D. M., Gandek, B., &
Maruish, M. E. (2007). User’s manual for the 36v2® Health Survey. 2nd
edition. Quality Metric,.

Watson, D., Clark, L. A., Tellegen, A., & Cook, R. D. (1988). Development and
validation of brief measures of positive and negative affect: The PANAS
scales. Journal of Personality and Social Psychology, 54(6), 1063-1070.
https://doi.org/10.1037/0022-3514.54.6.1063

Weiss, D., & Kornadt, A. E. (2018). Age-stereotype internalization and dissociation:


Contradictory processes or two sides of the same coin? Current Directions in
Psychological Science, 27(6), 477-483.
https://doi.org/10.1177/0963721418777743

Weiss, D., & Perry, E. L. (2020). Implications of generational and age


metastereotypes for older adults at work: The role of agency, stereotype

314
References

threat, and job search self-efficacy. Work, Aging and Retirement, 6(1), 15-27.
https://doi.org/10.1093/workar/waz010

Westerhof, G. J., & Barrett, A. E. (2005). Age identity and subjective well-being: A
comparison of the United States and Germany. The Journals of Gerontology,
Series B: Psychological Sciences and Social Sciences, 60(3), S129- S136.
https://doi.org/10.1093/geronb/60.3.S129

Westerhof, G. J., Barrett, A. E., & Steverink, N. (2003). Forever young?: A


comparison of age identities in the United States and Germany. Research on
Aging, 25(4), 366-383. https://doi.org/10.1177/0164027503025004002

Westerhof, G. J., Bohlmeijer, E. T., & McAdams, D. P. (2015). The relation of ego
integrity and despair to personality traits and mental health. The Journals of
Gerontology, Series B: Psychological Sciences and Social Sciences, 72(3),
400-407. https://doi.org/10.1093/geronb/gbv062

Westerhof, G. J., Gabrian, M., Brothers, A. F., Barrett, A. E., Diehl, M. K.,
Montepare, J. M., Wahl, H.-W., & Wurm, S. (2014). The influence of
subjective aging on health and longevity: A meta-analysis of longitudinal data.
Psychology and Aging, 29, 793-802. https://doi.org/10.1037/a0038016

Westerhof, G. J., Whitbourne, S. K., & Freeman, G. P. (2011). The aging self in a
cultural context: The relation of conceptions of aging to identity processes and
self-esteem in the United States and the Netherlands. The Journals of
Gerontology, Series B: Psychological Sciences and Social Sciences, 67B(1),
52-60. https://doi.org/10.1093/geronb/gbr075

Weyerer, S., Eifflaender-Gorfer, S., Wiese, B., Luppa, M., Pentzek, M., Bickel, H.,
Bachmann, C., Scherer, M., Maier, W., & Riedel-Heller, S. G. (2013).
Incidence and predictors of depression in non-demented primary care

315
References

attenders aged 75 years and older: Results from a 3-year follow-up study.
Age and Ageing, 42(2), 173-180. https://doi.org/10.1093/ageing/afs184

Whittemore, R., Chase, S. K., & Mandle, C. L. (2001). Validity in qualitative research.
Qualitative Health Research, 11(4), 522-537.
https://doi.org/10.1177/104973201129119299

Williams, K. N., & Kemper, S. (2010). Interventions to reduce cognitive decline in


aging. Journal of Psychosocial Nursing and Mental Health Services, 48(5),
42-51. https://doi.org/10.3928/02793695-20100331-03

Wilson, R. S., Beckett, L. A., Barnes, L. L., Schneider, J. A., Bach, J., Evans, D. A.,
& Bennett, D. A. (2002). Individual differences in rates of change in cognitive
abilities of older persons. Psychology and Aging, 17(2), 179-193.
https://doi.org/10.1037/0882-7974.17.2.179

Wilson, R. S., Hebert, L. E., Scherr, P. A., Barnes, L. L., Mendes de Leon, C. F., &
Evans, D. A. (2009). Educational attainment and cognitive decline in old age.
Neurology, 72(5), 460-465.
https://doi.org/10.1212/01.wnl.0000341782.71418.6c

Wilton-Harding, B., & Windsor, T. D. (2021). Awareness of age-related change,


future time perspective, and implications for goal adjustment in older
adulthood. Aging & Mental Health, 1-9.
https://doi.org/10.1080/13607863.2021.1893269

Winblad, B., Palmer, K., Kivipelto, M., Jelic, V., Fratiglioni, L., Wahlund, L. O.,
Nordberg, A., Backman, L., Albert, M., Almkvist, O., Arai, H., Basun, H.,
Blennow, K., De Leon, M., DeCarli, C., Erkinjuntti, T., Giacobini, E., Graff, C.,
Hardy, J., Jack, C., Jorm, A., Ritchie, K., van Duijn, C., Visser, P. J., &
Petersen, R. C. (2004). Mild cognitive impairment--beyond controversies,

316
References

towards a consensus: Report of the International Working Group on Mild


Cognitive Impairment. Journal of Internal Medicine, 256(3), 240-256.
https://doi.org/10.1111/j.1365-2796.2004.01380.x

Witherby, A., Tauber, S., Rhodes, M., & Castel, A. (2019). Aging and forgetting:
Forgotten information is perceived as less important than is remembered
information. Psychology and Aging, 34(2), 228-241.
https://doi.org/10.1037/pag0000322

Wolff, J. K., Schuz, B., Ziegelmann, J. P., Warner, L. M., & Wurm, S. (2017). Short-
term buffers, but long-term suffers? Differential effects of negative self-
perceptions of aging following serious health events. The Journals of
Gerontology, Series B: Psychological Sciences and Social Sciences, 72(3),
408-414. https://doi.org/10.1093/geronb/gbv058

Wolff, J. K., Warner, L. M., Ziegelmann, J. P., & Wurm, S. (2014). What do targeting
positive views on ageing add to a physical activity intervention in older adults?
Results from a randomised controlled trial. Psychology & Health, 29(8), 915-
932. https://doi.org/10.1080/08870446.2014.896464

World Health Organization. (2008a). 2008-2013 Action plan for the global strategy
for the prevention and control of non communicable diseases. World Health
Organization. http://wwwwhoint/nmh/Actionplan-PC-NCD-2008pdf

World Health Organization. (2008b). The global burden of disease: 2004 update.
https://www.who.int/healthinfo/global_burden_disease/GBD_report_2004upda
te_full.pdf

World Health Organization. (2015). The epidemiology and impact of dementia.


Current state and future trends. Retrieved 27/04/2020 from

317
References

https://www.who.int/mental_health/neurology/dementia/dementia_thematicbri
ef_epidemiology.pdf

World Health Organization. (2020a). Retrieved 27/04/2020 from


https://www.who.int/news-room/fact-sheets/detail/dementia

World Health Organization. (2020b). Mental health of older adults. Retrieved 21-11-
2020 from http://www.who.int/news-room/fact-sheets/detail/mental-health-of-
older-adults

Wrosch, C., Scheier, M. F., Carver, C. S., & Schulz, R. (2003). The importance of
goal disengagement in adaptive self-regulation: When giving up is beneficial.
Self and Identity, 2(1), 1-20. https://doi.org/10.1080/15298860309021

Wurm, S., & Benyamini, Y. (2014). Optimism buffers the detrimental effect of
negative self-perceptions of ageing on physical and mental health.
Psychology & Health, 29(7), 832-848.
https://doi.org/10.1080/08870446.2014.891737

Wurm, S., Tomasik, M. J., & Tesch-Römer, C. (2008). Serious health events and
their impact on changes in subjective health and life satisfaction: The role of
age and a positive view on ageing. European Journal of Ageing, 5(2), 117-
127. https://doi.org/10.1007/s10433-008-0077-5

Wurm, S., Tomasik, M. J., & Tesch-Römer, C. (2010). On the importance of a


positive view on ageing for physical exercise among middle-aged and older
adults: Cross-sectional and longitudinal findings. Psychology & Health, 25(1),
25-42. https://doi.org/10.1080/08870440802311314

318
References

Wurm, S., Warner, L. M., Ziegelmann, J. P., Wolff, J. K., & Schuz, B. (2013). How do
negative self-perceptions of aging become a self-fulfilling prophecy?
Psychology and Aging, 28(4), 1088-1097. https://doi.org/10.1037/a0032845

Xu, J. (2017). A tripartite function of mindfulness in adjustment to aging: Acceptance,


integration, and transcendence. The Gerontologist, 58(6), 1009-1015.
https://doi.org/10.1093/geront/gnx100

Yates, J. A., Clare, L., & Woods, R. T. (2017). What is the relationship between
health, mood, and mild cognitive impairment? Journal of Alzheimer's Disease,
55(3), 1183-1193. https://doi.org/10.3233/jad-160611

Yates, J. A., Clare, L., Woods, R. T., & Matthews, F. E. (2015). Subjective memory
complaints are involved in the relationship between mood and mild cognitive
impairment. Journal of Alzheimer's Disease, 48(s1), S115-S123.
https://doi.org/10.3233/JAD-150371

Zanini, S., Voltolini, A., Gragnano, G., Fumagalli, E., & Pagnini, F. (2018). Changes
in pain perception following psychotherapy: The mediating role of
psychological components. Pain Research & Management, 2018, 8713084-
8713084. https://doi.org/10.1155/2018/8713084

Zhan, Y., Wang, M., Liu, S., & Shultz, K. S. (2009). Bridge employment and retirees’
health: A longitudinal investigation. Journal of Occupational Health
Psychology, 14(4), 374-389. https://doi.org/10.1037/a0015285

Zhang, S., & Neupert, S. D. (2020). Within-and between-person relationships among


health, awareness of aging, and control beliefs: A micro-longitudinal study.
The Journals of Gerontology, Series B: Psychological Sciences and Social
Sciences. https://doi.org/10.1093/geronb/gbaa180

319
References

Zhu, X., & Neupert, S. D. (2020). Dynamic awareness of age-related losses predict
concurrent and subsequent changes in daily inductive reasoning
performance. British Journal of Developmental Psychology.
https://doi.org/10.1111/bjdp.12344

Zizza, C. A., Ellison, K. J., & Wernette, C. M. (2009). Total aater intakes of
community-living middle-old and oldest-old adults. The Journals of
Gerontology, Series A: Biological Sciences and Medical Sciences, 64A(4),
481-486. https://doi.org/10.1093/gerona/gln045

Zlatar, Z. Z., Muniz, M., Galasko, D., & Salmon, D. P. (2018). Subjective cognitive
decline correlates with depression symptoms and not with concurrent
objective cognition in a clinic-based sample of older adults. The Journals of
Gerontology, Series B: Psychological Sciences and Social Sciences, 73(7),
1198-1202. https://doi.org/10.1093/geronb/bgw207

320
Appendices

Appendices

321
Appendices

322
Appendices

Appendix A: Questions added to the main PROTECT platform

Project: “How we see ourselves: changes over time”


Assessment: Self-rated health
1.1 “Would you say that for someone of your age, your own health in general is:
Excellent/good/fair/poor/don’t know?”

Assessment: Feelings and attitudes toward ageing


This section contains statements on how a person may feel about getting older. We
are interested in how YOU feel about getting older and what you can tell us about
your own experiences. We are mostly interested in whether certain experiences
apply to you and to what extent you can relate to them given that you may have
experienced them yourself as the years have passed. Please read each statement
carefully and answer in the following way:
• If a statement very much reflects your own experience, then you would fill in
the circle under the number “5”, which stands for “Very much.”
• If a statement does not reflect your own experience at all, then you indicate
that by filling in the circle under the number “1”, which stands for “Not at all.”
• If a statement reflects your experiences more than “a little bit” but less than
“quite a bit,” then you should mark the circle under the number “3”, which
stands for “Moderately.”
There are no “right” or “wrong” answers to the statements on this questionnaire. We
are simply interested in your own personal experiences and your honest opinion.

1.1 With my increasing age, I realise that I pay more attention to my health.
1.2 With my increasing age, I realise that I have more experience and knowledge to
evaluate things and people.
1.3 With my increasing age, I realise that I appreciate relationships and people much
more.
1.4 With my increasing age, I realise that I have a better sense of what is important
for me.
1.5 With my increasing age, I realise that I have more freedom to live my days the
way I want.

323
Appendices

1.6 With my increasing age, I realise that I have less energy.


1.7 With my increasing age, I realise that my mental capacity is declining.
1.8 With my increasing age, I realise that I feel more dependent on the help of
others.
1.9 With my increasing age, I realise that I find it harder to motivate myself.
1.10 With my increasing age, I realise that I have to limit my activities.
1.11 With my increasing age, I realise that I am slower in my thinking
1.12 With my increasing age, I realise that I have a harder time concentrating.
1.13 With my increasing age, I realise that I have more foresight.
1.14 With my increasing age, I realise that learning new things takes more time
and effort.
1.15 With my increasing age, I realise that I am more forgetful.
1.16 With my increasing age, I realise that I have become wiser.
1.17 With my increasing age, I realise that I think things through more carefully.
1.18 With my increasing age, I realise that I gather more information before I make
decisions.
(Answers’ option: 1= not at all; 2= a little bit; 3= moderately, 4= quite a bit; 5= very
much).

2.1 “Many people feel older or younger than they actually are. Fill in the age (in
years) that you feel most of the time: ___”
3.1 Things keep getting worse as I get older. (answer yes or no)
3.2 I have as much pep (vim, vigor, “get-up-and-go”) as I had last year. (answer yes
or no)
3.3 As I get older, I am less useful. (answer yes or no)
3.4 As I get older, things are (better/worse) than I thought they would be. (answer
better or worse)
3.5 I am as happy now as I was when I was younger. (answer yes or no)

Assessment: Thinking
People think and do many different things when they feel depressed. Please read
each of the items below and indicate whether you almost never, sometimes, often, or
almost always think or do each one when you feel down, sad, or depressed. Please
indicate what you generally do, not what you think you should do.

324
Appendices

How often do you:


1.1 Think about how alone you feel
1.2 Think “I won’t be able to do my job if I don’t snap out of this”
1.3 Think about your feelings of fatigue and achiness
1.4 Think about how hard it is to concentrate
1.5 Think “what am I doing to deserve this?”
1.6 Think about how passive and unmotivated you feel
1.7 Analyse recent events to try to understand why you are depressed
1.8 Think about how you don’t seem to feel anything anymore
1.9 Think “Why can’t I get going?
1.10 Think “Why do I always react this way?”
1.11 Go away by yourself and think about why you feel this way
1.12 Write down what you are thinking about and analyse it
1.13 Think about a recent situation, wishing it had gone better
1.14 Think “I won’t be able to concentrate if I keep feeling this way”
1.15 Think “why do I have problems other people don’t have?”
1.16 Think “why can’t I handle things better?”
1.17 Think about how sad you feel
1.18 Think about all your shortcomings, failings, faults, mistakes
1.19 Think about how you don’t feel up to doing anything
1.20 Analyse your personality to try to understand why you are depressed
1.21 Go someplace alone to think about your feelings
1.22 Think about how angry you are with yourself
(Answers’ option: 1= almost never; 2= sometimes; 3= often; 4= almost always)

325
Appendices

Appendix B: Questions used from the main PROTECT data set


Demographic information

1.1 What is your date of birth?


1.2 Are you male or female? (Male; Female)
1.3 What is your ethnic origin? Select your ethnic origin:
a) White: English / Welsh / Scottish / Northern Irish / British
b) White: Irish
c) White: Gypsy or Irish Traveller
d) White: European
e) White: Non-European
f) Mixed: White and Black Caribbean
g) Mixed: White and Black African
h) Mixed: White and Asian
i) Mixed: Any other Mixed / Multiple ethnic background
j) Asian / Asian British: Indian
k) Asian / Asian British: Pakistani
l) Asian / Asian British: Bangladeshi
m) Asian / Asian British: Chinese
n) Asian / Asian British: Any other Asian background
o) Black / African / Caribbean / Black British: African
p) Black / African / Caribbean / Black British: Caribbean
q) Any other Black / African / Caribbean background
r) Other ethnic groups: Arab
s) Any other ethnic group
1.4 What is your marital status? Select your marital status
a) Married
b) Widowed
c) Separated
d) Divorced
e) Civil Partnership
f) Co-habiting
g) Single
1.5 What is the highest level of education you have completed?

326
Appendices

a) Secondary Education (GCSE/O-Levels)


b) Post-Secondary Education (College, A-Levels, NVQ3 or below, or
similar)
c) Vocational Qualification (Diploma, Certificate, BTEC, NVQ 4 and
above, or similar)
d) Undergraduate Degree (BA, BSc etc.)
e) Post-graduate Degree (MA, MSc etc.)
f) Doctorate (PhD)
1.6 What is your current employment status? Select your current employment
status
a) Employed (full-time)
b) Employed (part-time)
c) Self-employed
d) Retired
e) Unemployed

Lawton’s Instrumental Activities of Daily Living Scale


2.1 Preparing meals
2.2 Ordinary housework (e.g. doing the dishes, dusting, making the bed,
doing laundry)
2.3 Managing household finances (e.g. paying bills, balancing household
expenses)
2.4 Managing medications (e.g. remembering to take medications)
2.5 Using the telephone (i.e. making and receiving calls)
2.6 Shopping (e.g. completing the weekly shop, managing the money)
2.7 Transportation (e.g. driving, taking the bus)

(Answers’ option: No difficulty; Some difficulty; Great difficulty)

The Patient Health Questionnaire-9

3.1 Over the last 2 weeks, how often have you felt little interest or pleasure in
doing things?
3.2 Over the last 2 weeks, how often have you felt down, depressed, or hopeless?

327
Appendices

3.3 Over the last 2 weeks, how often have you had trouble falling or staying asleep,
or sleeping too much?
3.4 Over the last 2 weeks, how often have felt tired or had little energy?
3.5 Over the last 2 weeks, how often have you been bothered by a poor appetite or
overeating?
3.6 Over the last 2 weeks, how often have you been bothered about feeling bad
about yourself or that you are a failure or have let yourself or your family down?
3.7 Over the last 2 weeks, how often have you had trouble concentrating on things,
such as reading the newspaper or watching television?
3.8 Over the last 2 weeks, how often have you been bothered by moving or
speaking so slowly that other people could have noticed? Or the opposite —
being so fidgety or restless that you have been moving around a lot more than
usual?
3.9 Over the last 2 weeks, how often have you had thoughts that you would be
better off dead or of hurting yourself in some way?

(Answers’ option: Not at all; Several days; More than half the days; Nearly every day;
Prefer not to answer)

The Composite International Diagnostic Interview-Short Form


Lifetime depression

4.1 Have you ever had a time in your life when you felt sad, blue, or depressed
fortwo weeks or more in a row?
4.2 Have you ever had a time in your life lasting two weeks or more when you lost
interest in most things like hobbies, work, or activities that usually give you
pleasure?
4.3 Thinking of the two-week period in your life when your feelings of depression or
loss of interest were worst...did you feel more tired out or low on energy than is
usual for you?
4.4 Thinking of the two-week period in your life when your feelings of depression or
loss of interest were worst...did you gain or lose weight without trying, or did you
stay about the same weight?
4.5 Thinking of the two-week period in your life when your feelings of depression or
loss of interest were worst...did your sleep change?

328
Appendices

4.6 Thinking of the two-week period in your life when your feelings of depression or
loss of interest were worst...did you have a lot more trouble concentrating than
usual?
4.7 Thinking of the two-week period in your life when your feelings of depression or
loss of interest were worst...people sometimes feel down on themselves, no
good, worthless. Did you feel this way?
4.8 Thinking of the two-week period in your life when your feelings of depression or
loss of interest were worst...did you think a lot about death – either your own,
someone else's or death in general?

(Answers’ option: Yes; No)

Lifetime anxiety
5.1 When you were worried or anxious, were you also...restless?
5.2 ...keyed up or on edge?
5.3 ...easily tired?
5.4 ...having difficulty keeping your mind on what you were doing?
5.5 ...more irritable than usual?
5.6 ...having tense, sore, or aching muscles?
5.7 ...often having trouble falling or staying asleep?
(Answers’ option: Yes; No)

The Generalised Anxiety Disorder Assessment 7-item GAD-7


6.1 Over the last 2 weeks, how often have you been bothered by feeling nervous,
anxious or on edge?
6.2 Over the last 2 weeks, how often have you been bothered by not being able to
stop or control worrying?
6.3 Over the last 2 weeks, how often have you been bothered by worrying too much
about different things?
6.4 Over the last 2 weeks, how often have you had trouble relaxing?
6.5 Over the last 2 weeks, how often have you been bothered by being so restless
that it is hard to sit still?
6.6 Over the last 2 weeks, how often have you been bothered about becoming easily
annoyed or irritable?

329
Appendices

6.7 Over the last 2 weeks, how often have you been bothered about feeling afraid as
if something awful might happen?

(Answers’ option: Not at all; Several days; More than half the days; Nearly every day;
Prefer not to answer)

The Informant Questionnaire on Cognitive Decline in the Elderly- Self Version


and informant
7.1 Remembering things about family and friends, e.g. occupations, birthdays,
addresses?
7.2 Remembering things that have happened recently?
7.3 Recalling conversations a few days later?
7.4 Remembering your own address and telephone number?
7.5 Remembering what day and month it is?
7.6 Remembering where things are usually kept?
7.7 Remembering where to find things which have been put in a different place
from usual?
7.8 Knowing how to work familiar machines around the house?
7.9 Learning to use a new gadget or machine around the house?
7.10 Learning new things in general?
7.11 Following a story in a book or on TV?
7.12 Making decisions on everyday matters?
7.13 Handling money for shopping?
7.14 Handling financial matters, e.g. the pension, dealing with the bank?
7.15 Handling other everyday arithmetic problems, e.g. knowing how much food to
buy, knowing how long between visits from family or friends?
7.16 Using your own intelligence to understand what's going on and to reason
things through?

(Answers’ option: Much improved; A bit improved; Not much change; A bit worse;
Much worse)

330
Appendices

Appendix C: Application for access to PROTECT study data and / or samples


Scientific title: “Awareness of age-related change and its association with mental and physical well-being in cognitively
Title of project
healthy adults living in the UK”

Principal Investigator Serena Sabatini

Centre for Research in Ageing and Cognitive Health (REACH)


Organization and
University of Exeter Medical School, South Cloisters, St Luke’s Campus, Exeter EX1 2LU, UK
Department (full contact
Tel: +44 (0) 1392 726 754
details)
Email: ss956@exeter.ac.uk
Prof. Linda Clare, Prof. Clive Ballard (PROTECT investigator), Prof. Kaarin Anstey, Dr Barbora Silarova, Dr Rachel
Co-investigators
Collins, and Dr Sarang Kim

Name: Serena Sabatini


Primary contact for study
Address: Centre for Research in Ageing and Cognitive Health (REACH), University of Exeter Medical School, South
Cloisters, St Luke’s Campus, Exeter EX1 2LU, UK
Tel: 07757029595
Email: ss956@exeter.ac.uk

Start and completion dates Project start Date: 24 September 2017

Project completion Date: 24 March 2021

Participant characteristics
What is being requested?

(Place a X in all that apply) Sample Size This will be determined by the number of participants who will fill in the AARC questionnaire.
Everyone invited to take part in the annual assessment in November 2018 and 2019 will
have an opportunity to fill in the AARC questionnaire. We are requesting available 2018 and

331
Appendices

2019 data, as outlined below, for all participants who complete the AARC questionnaire at
the 2018 assessment.

Participant As part of the annual assessment in November 2018 and 2019 participants will be asked to
characteristics fill in the AARC questionnaire (please see further details in Serena Sabatini Study
summary). Therefore we request data for all participants who will fill in the AARC
questionnaire as part of their annual assessments starting in November 2018.

Anonymised Datasets

Two time-points: Nov. 2018 and Nov.


2019

Full PROTECT Data set

Partial PROTECT Demographics Items 1.1, 1.2, 1.3, 1.5, only for Nov.
Dataset 2018 while items 1.4, 1.6, 1.7 both for
Nov. 2018 and 2019

Medical History Selected items: 1.3, 1.5, 1.7, 1.8,

Lifestyle Item 9.1

Cognitive Test Package 1 (PROTECT) Yes

Cognitive Test Package 2 (CogTrack)* Yes

IQCode Yes

IADL Yes

332
Appendices

Mental Health Questionnaire (full)

Mental Health Questionnaire (partial – please PHQ-9; CIDI-SF; GAD-7; Items from 1.1
specify which sections): to 1.4

Mild Behaviour impairment Scale

Pain short-form Yes

Sleep Yes

Fertility / menopause

Genetic Data / Samples

Genotype / GWAS Data Not requested

Access to extracted Not requested


DNA samples

Personal details and re-contact

Personal details data We do not request any personal details. All communication with participants will be managed
by the PROTECT team.

Re-contact of As part of the PROTECT study we plan to add new assessments for all participants who
participants for nested agree to take part in their annual assessment starting in November 2018. Additional
PROTECT study questionnaires are as follows: 1. Self-rated health; 2. AARC-10 SF; 3. AARC-cognitive
subscale from the AARC-50 item questionnaire; 4. ATOA measure; 5. Subjective age: one
item taken from the National Survey of Midlife development in the United States; and 6.

333
Appendices

e.g. addition of new Ruminative Response Scale. Details are provided as part of the Serena Sabatini study
assessment or email summary.
distribution of survey etc.

Re-contact of Not requested


participants for
separate study

e.g. identification of
participants for clinical
trial

The aims of this study are:


Study Aims & Objectives

Please outline the aims of your 1) to validate a measure of Awareness of age-related change (AARC): the AARC-10 SF (Kaspar, Gabrian, Brothers,

study Wahl, Diehl, 2017) and the cognitive functioning subscale of the AARC-50 (Brothers et al, 2018) in people aged at
least 50 years old and living in the UK;
2) to investigate the association between AARC (gains and losses) and physical health by exploring indicators of
physical limitations including current and past physical health conditions, hearing and visual impairments, physical
pain, subjective health, instrumental activities of daily living and quality of sleep;
3) to investigate the association between AARC (gains and losses) and symptoms of depression and anxiety and to
understand the role of rumination in the association between AARC and symptoms of depression and anxiety;
4) to understand the role of cognition in relation to AARC by exploring the possible association between the AARC-
cognitive subscale (gains and losses) and objective cognitive performance. As part of this sub-study the possible
association between the AARC-cognitive subscale score and cognitive abilities rated by an informant will be explored.

334
Appendices

The study “Awareness of age-related change and its association with mental and physical well-being in cognitively healthy
Brief Description of Project
adults living in the UK” will be a longitudinal study exploring the concept of awareness of age-related change (AARC) and
(If access to personal details
how it is related to mental well-being, physical health and cognitive abilities. Data will be collected at two time points since
or genetic data is required
the study aims to explore the direction of the associations.
please include study protocol)
Since assessments of mental well-being, physical health and cognition are part of PROTECT, we request access to
PROTECT data and to add extra questionnaires to assess variables such as rumination and AARC that are not already
assessed in PROTECT.

Tick relevant box


Ethical approval status

(Some data requests may fall No additional approval required (please justify)

within the existing PROTECT Ethics submission in process Ethics will be managed as an amendment to the
approvals. If in doubt please PROTECT Ethics application. The process will be
contact a PROTECT Anticipated date of approval:
managed by the PROTECT team.
administrator to discuss) REC:

R&D dept:

Reference number:

(Please attach confirmation of sponsorship)

Ethical approval confirmed

Date of approval:

REC:

R&D dept:

335
Appendices

Reference number:

HRA approval required:

(Please attach letter of approval from REC and HRA if


appropriate)

1. Sample size and participant characteristics: We are requesting available 2018 and 2019 data, as outlined below,
Justification for request
for all participants who complete the AARC questionnaire at the 2018 assessment.
Please provide rationale for 2. Data type: Quantitative – Partial PROTECT data set as outlined above and new measures added as part of Serena
the data or samples requested
Sabatini study.
above, including:
3. New assessments. In order to collect data on constructs that are not covered in the PROTECT dataset, new
1. Sample size questionnaires will be added as part of the PROTECT study.
2. Participant
characteristics
3. Data type
4. Sample type
5. New study or
assessment

If relevant provide a clear


justification for requests that
include both access
participants with specific
genotypes and personal
details, including any plans to

336
Appendices

disclose genetic information


and how this will be managed.

- Managing participant contact through PROTECT


Involvement of PROTECT
- Technical support to include new questionnaires
Investigators
- Data management
If you expect to require
- Help to identify participants that fulfil criteria for mild cognitive impairment and age-associated cognitive decline
support from PROTECT
based on standard deviation from age-matched norms across the series of cognitive tests.
investigators please give
details. This may include:

- Managing participant
contact through
PROTECT
- Technical support for
PROTECT website,
incl new assessments
- Statistical support
- Data management

Budget

Do you have budget available


to support this request?
Please provide details

This is particularly relevant to


requests that require active
involvement of PROTECT

337
Appendices

investigators or amendments
to PROTECT study
assessments

Further Information

Please provide any further


relevant information

Key information for applicants

- Use of PROTECT data is subject to full ethical approval. Some requests may fall under existing PROTECT approvals.
Please contact the study administrator to discuss further
- This application will be reviewed by the PROTECT Strategy Group to ensure the request is appropriate to the data, and does
not conflict with existing research and analyses within the PROTECT portfolio
- GWAS data will be released as summary data only (list of markers with effect size / overall frequency of requested alleles
per marker) and will not include individual-level genetic data
- The PROTECT investigators ask that any publications or outputs arising from the use of PROTECT data include authorship
for PROTECT investigators and acknowledgement of PROTECT support and funders.
- CogTrackTM is provided by a third party. Use of this data may be subject to additional cost.
Document to be submitted to maria.megalogeni@kcl.ac.uk.
Questions regarding this process should be directed to maria.megalogeni@kcl.ac.uk and zunera.2.khan@kcl.ac.uk in the first
instance

338
Appendices

Appendix D: Approval for access to PROTECT data

Ms Serena Sabatini
REACH Research Group
South Cloisters, St Luke’s Campus
University of Exeter
Exeter EX1 2LU

RE: Data Access Request

Dear Serena
I am pleased to confirm that your Data Access Request to the PROTECT Study has been approved. This letter
is a formal acknowledgement that the following activities will go forward:
1. Addition of the AARC scale to the PROTECT platform
2. Ongoing data collection through the AARC in the full platform
3. Creation of bespoke dataset for AARC analysis (interim and full dataset at one year)

This approval is also subject to any additional agreements, contracts or authorship arrangements as
negotiated with the PROTECT investigators. Any amendments to your request should be submitted to the
Committee for review.

On behalf of the PROETCT Steering Committee I wish you the best of luck with your research. We look forward
to working with you.

Yours sincerely

Dr Anne Corbett

PROTECT Academic Lead, University of Exeter

On Behalf of the PROTECT Study Steering Committee

CC by email: Ellie Pickering, (PROTECT Coordinator, University of Exeter); Maria Megalogeni (PROTECT
Coordinator, King’s College London)

339
Appendices

Appendix E: Ethical approval for the PROTECT study

HRA APPROVAL for the Amendment

Dear Dr Corbett,

IRAS Project ID: 136118


Short Study Title: A longitudinal study of cognition in people over 50
Amendment No./Sponsor Ref: 11
Amendment Date: 01 October 2018
Amendment Type: Substantial Non-CTIMP

I am pleased to confirm HRA and HCRW Approval for the above referenced
amendment.
You should implement this amendment at NHS organisations in England and Wales,
in line with the conditions outlined in your categorisation email.
User Feedback
The Health Research Authority is continually striving to provide a high quality service
to all applicants and sponsors. You are invited to give your view of the service you
have received and the application procedure. If you wish to make your views known
please use the feedback form available on the HRA website:
http://www.hra.nhs.uk/about-the-hra/governance/quality-assurance/.
Please contact hra.amendments@nhs.net for any queries relating to the assessment
of this amendment.
Kind regards

340
Appendices

341
Appendices

Appendix F: Ethical approval for the secondary data-analyses

Title Scientific title: “Awareness of age-related change and its association with
mental and physical well-being in cognitively healthy adults living in the UK”
Lay title of survey: “How we change as we get older”
Application type Student. This application aims to seek ethical approval exclusively for the
conduct of analyses as this study was approved by NHS REC: London -
London Bridge Research Ethics Committee (13/LO/1578) on 15th
November 2018. The HRA Approval for the Amendment was granted on
15th November 2018.
Project supervisor Linda Clare, Barbora Silarova and Rachel Collins
Campus St. Luke’s Campus
Applicant Serena Sabatini
Applicant I confirm I have read the University of Exeter 'Good Practice in the Conduct
declaration of Research' code of practice.
Has your project Yes
been externally
ethically reviewed?
Estimated start Data collection started on 10/01/2019
date
Is the project This PhD study is linked to the externally-funded collaboration in the Centre
externally funded? for Research Excellence in Cognitive Health, UNSW Sydney, and is funded
jointly by CLES and UEMS.
Lay summary Awareness of age-related change (AARC) is a construct (Wahl and Diehl,
(max 250 words) 2010) that refers to those events, situations and circumstances that make
individuals aware that their behaviour, level of performance, or ways of
experiencing their lives have changed due to having grown older. AARC is
associated with measures of physical and emotional well-being. However
evidence on AARC is limited to U.S. and German contexts. Therefore, the
overall aim of this project is to explore associations between AARC and
mental well-being and cognitive abilities in the UK population.
As part of the project we will conduct four empirical sub-studies aiming: (1)
to validate two measures of AARC: the AARC-10 SF (Kaspar, Gabrian,
Brothers, Wahl, Diehl, 2017) and the AARC-cognitive functioning subscale
from the AARC-50 item questionnaire (Brothers, Gabrian, Diehl, Wahl,
2018) in people aged at least 50 years old and living in the UK and to
explore how the profile of AARC gains and losses changes with age; (2) to
investigate the association between AARC and physical health; (3) to
investigate the association between AARC and symptoms of depression

342
Appendices

and anxiety and to understand the potential role of rumination in this


association; (4) to understand the role of cognition in relation to AARC by
looking at the strength and direction of the association between the AARC-
cognitive domain and objective cognitive performance and cognitive
abilities rated by an informant.
This study will be embedded within the larger ongoing PROTECT study. As
part of the PROTECT annual assessments in January 2018 and January
2019 participants will be asked to fill in additional questionnaires specific to
this study.

Location

Where will your Within the United Kingdom


research take place
Please provide The PROTECT study is an online study coordinated by researchers from
details of your the University of Exeter Medical School. University of Exeter is the Sponsor
research location(s) of the PROTECT project and King’s college in London is the PROTECT
study research site. PROTECT administrative teams are present both at
the University of Exeter Medical School and at King’s College in London.

Details

Is the application linked to a previous No


application?
Has your project been peer reviewed by Research team / group members: YES
any of the following? External peer review: YES
Does your project involve: Humans: YES
Animals: NO
Does this study involve human samples? NO
Track type B

Questions

Human aspects
I am familiar with the BPS Guidelines for ethical practices in psychological research (and YES
have discussed them with other researchers involved in the project.)
Communication and consent
Will you describe the main research procedures to participants in advance, so that they YES
are informed in advance about what to expect?

343
Appendices

Will you tell participants that their participation is voluntary? YES


Will you obtain written consent for participation? YES
If the research is observational, will you ask participants for their consent to being N/A
observed?
Will you tell participants that they may withdraw from the research at any time and for YES
any reason?
With questionnaires, will you give participants the option of omitting questions they do YES
not want to answer?
Will you tell participants that their data will be treated with full confidentiality and that, if YES
published, it will not be identifiable as theirs?
Will you debrief participants at the end of their participation (i.e. give them a brief NO
explanation of the study)?
Possible harms
Will your project involve deliberately misleading participants in any way? NO
Is there a realistic risk of any participants experiencing either physical or psychological NO
distress or discomfort?
Vulnerable groups
Does your project involve:
Children under 18 years of age NO
People with learning or communication difficulties NO
Participants who are unable to give informed consent NO
Any member of the research team or participants are members of the Armed Forces or NO
their dependents
Those at risk of psychological distress or otherwise vulnerable NO
People in custody or on probation NO
People engaged in illegal activities (e.g. drug taking) NO
NHS patients or social care service users YES
Data protection and storage
Will you ensure participant data is kept confidential YES
Will data be stored securely YES
Will you inform participants about how the data collected from them will be used YES
Will you retain collected data for a specified period (e.g. five years) YES
Will you inform participants of this data retention period YES
I am satisfied that the research study is compliant with the Data Protection Act 1998, and YES
that necessary arrangements have been made with regard to the storage and
processing of participants’ personal information and generally, to ensure confidentiality
of such data supplied and generated in the course of the research.

344
Appendices

Statement

Research The study aims will be explored by conducting a 1-year cohort longitudinal
methodology study that will be implemented through PROTECT (Platform for Research
Online to investigate Genetics and Cognition in Ageing;
http://www.protectstudy.org.uk/about.aspx). Protect is an ongoing 10-year
longitudinal cohort study that started in November 2015 and is conducted
online. PROTECT aims to explore the role of genetic, lifestyle, and medical
factors on cognition in individuals aged 50 years and over and living in the UK.
In PROTECT data are collected using an online platform in which participants
can log in to the site and complete questionnaires and assessments.
PROTECT participants initially provided consent to complete a baseline
assessment in 2015 and they are invited to complete a follow-up assessment
each year. As part of the PROTECT annual assessments in January 2018 and
January 2019 participants will be asked to fill in additional questionnaires
specific to this study. The measures will be: AARC-10 SF, eight additional
items taken from the AARC-50- cognitive functioning subscale, ATOA 5-item
subscale from the Philadelphia Geriatric Center (PGC) Morale Scale, single-
item question assessing age, Ruminative Response Scale (RRS), single-item
question taken from the SF-36 to assess self-rated health. Moreover, to
conduct this study we will also use data collected as part of the main
PROTECT study during the annual assessments in January 2018 and January
2019. Requested questionnaires from PROTECT are: demographic
information; bespoken items on lifetime physical health, 5-item Pain Scale,
Lawton’s Instrumental Activities of Daily Living Scale, 9-item St Mary’s Sleep
scale, bespoke items from PROTECT assessing lifetime mental health, the
Patient Health Questionnaire-9 (PHQ-9), the Composite International
diagnostic interview-short form (CIDI-SF), the GAD-7, and the Informant
questionnaire on Cognitive Decline in the Elderly- Self version (IQCODE-Self).
Cognitive functions are measured with Cognitive test package 1 from
PROTECT, Cognitive test package 2/ CogTrackTM (www.wesnes.com), Trail
Making Tests A and B and the Stroop test.
Permission to conduct this study has been granted by the PROTECT steering
Committee and PROTECT Academic Lead, Dr Anne Corbett, Senior Lecturer,
UEMS (Appendix 1). This covers 1) adding the AARC-10 SF and the AARC-
cognitive functioning subscale to the PROTECT platform; 2) collecting data
through the PROTECT online platform; 3) creation by PROTECT team
members of a bespoken data set for the analyses of the present study (cross-
sectionally and longitudinally over one year).

345
Appendices

In order to obtain permission from the PROTECT steering Committee, the


PROTECT Academic Lead was provided with: a) a data access form
(Appendix 2 ); b) an outline of the study (Appendix 3 ) including research
questions, variables requested from PROTECT, questionnaires to add to the
PROTECT platform, and an analysis plan; c) a Memorandum of Understanding
that was set in place with the authors of the AARC-measures (Appendix 4);
and d) a data agreement form (Appendix 5).
Full details of the PROTECT protocol and Participant Information Sheet can be
found in the supporting documents (Appendix 6 and 7).
Anonymised data will be sent using encrypted files from the administrative
centres at King’s College London and University of Exeter Medical School.
Research For this study we do not define a maximum limit to the sample size as this will
participants be determined by the number of participants who will complete the additional
questionnaires specific to this study alongside the rest of the PROTECT
assessments. We aim for a minimum sample size of 2500 participants as this
is estimated to be sufficient to provide enough power to reject the null
hypothesis when it should be rejected in all our sub-studies. When calculating
a minimum sample size we took into account the effect size, p value, design of
the study, power and recommendations regarding sample size for validation
studies. Therefore if on 1st March 2019 we have at least 2500 participants, we
will close the survey given the time restrictions for the conduct of this study
that is part of a PhD.
What follows is an overview of the PROTECT participants’ characteristics and
an explanation of how these participants were recruited at the beginning of the
study.
Individuals were potentially eligible to participate in the PROTECT Study if they
were UK residents, English speakers, aged 50 years and older, had access to
a computer and internet and did not have a clinical diagnosis of dementia or
mild cognitive impairment at the time of recruitment. For the present study we
will exclude from analyses those individuals who we identified as having mild
cognitive impairment at the January 2019 round of assessment.
In PROTECT participants were recruited through several channels: (1)
advertisement at King’s College London, (2) invitation of participants in existing
cohorts and completed trials who had indicated they would like to be contacted
about future research, (3) advertisement on the Join Dementia Research
online portal, and (4) information leaflets placed in GP surgeries and memory
clinics. Recruitment began in 2013 and baseline assessment was completed in
November 2015. From 2017 additional PROTECT participants have been
recruited through the University of Exeter and the Royal Devon and Exeter

346
Appendices

NHS Trust. Up to now a total of approximately 22,000 participants have been


recruited to the study.
We decided to use PROTECT as a platform to implement this research for
pragmatic reasons. However, PROTECT is an appropriate platform to conduct
this study as we are interested in a population of middle-aged and older
individuals and PROTECT includes participants aged 50 years old and over.
We are interested in individuals aged 50 years old and over for several
reasons: 1) we aim to validate the AARC-10 SF in individuals aged 50 and
over as psychometric properties for this scale have only been explored in
individuals aged 70 years and older; 2) 50 years is an appropriate cut-off that
should allow us to include individuals who are old enough to be likely to
experience AARC. Indeed research shows that individuals aged 50 are likely
to experience many age-related changes. 3) Individuals in their 50s report
being concerned with their physical health and hence they may be willing to
participate in future interventions aiming to promote healthy ageing.
The use of PROTECT also allows us to conduct a longitudinal study as
PROTECT includes follow-ups. Moreover, PROTECT excludes individuals with
dementia, which is an important exclusion criterion in the present research.
Ethical Sensitive questions about awareness of age-related change, attitudes toward
considerations ageing and ruminative thoughts:
Participants may feel some discomfort when answering questions about
awareness of age-related changes, attitudes toward ageing and ruminative
thinking. However, participants do not have to complete the survey if they feel
distressed in doing so. In addition, the questions are not too probing and will
not cause significant psychological distress or discomfort. They are
comparable with questions asked in relation to other domains in the existing
PROTECT survey.
Data protection procedures:
PROTECT is available through University of Exeter and King’s College
London; baseline data of the present study will be collected via the King’s
College and Exeter PROTECT platform. PROTECT platforms will merge in
June 2018 and therefore 1-year follow-up data will be collected through the
merged platform. Personal data will be collected on a separate database from
all other data. This will be matched up with a unique ID number for each
participant. Personal data will be held separately by King’s College London
and Exeter University, dictated by the recruitment route taken by the
participant. Databases will be password protected and located on University
shared drives which only the study team at that University have access to. The
data will also be on the Universities’ servers which are not accessible to
anyone outside the organisations.

347
Appendices

All non-identifiable data (i.e. data that are not personal) will be stored in
independent databases at the two sites in Exeter University and King’s College
London. These data will be accessible for import to either site to enable
creation of a full database for analyses purposes. Data generated through the
Wesnes Online Cognitive Test will be stored in a secure server and transferred
to the host site via a regular encrypted transfer.
Data will also be stored online on Google Cloud to provide a back-up database
(https://cloud.google.com/terms/data-processing-terms ). Raw data (not
including personal details) will be stored in encrypted form, subject to ISO/IEC
27001:2005 Certification. Only core study team members will have access to
personal data through the electronic database. This will include the Chief
Investigator, study coordinators and IT specialist. Once data have been
collected, colleagues from the PROTECT team are able to extract relevant
data requested by researchers and prepare this to be sent anonymously.
In order to facilitate analyses for the present study, PROTECT team
members will create bespoke datasets (cross-sectional and longitudinal over
one year). The bespoke datasets for this study will not include any personal
data. Quantitative data will be managed according to the data protection and
governance requirements set out in the standard procedures for the ethical
approval of research by the University of Exeter. Access to data will be limited
to the minimum number of individuals necessary for quality control, audit, and
analyses. Access to the databases will be password protected and limited to
Serena Sabatini, Professor Linda Clare, Dr Barbora Silarova, Dr Rachel
Collins and the research team when necessary.
The study database will be held on University servers in restricted access
buildings (restricted staff only swipe card entry). All university computers are
password protected and access to electronic drives and folders containing
study data are restricted to the research team.
Data that will be stored on the study databases will include:
• Data from surveys
Responsibilities and Resources: Serena sabatini will be the custodian of the
generated data, including the implementation of the data management plan,
storage and initial archive. University of Exeter will be responsible for the
storage of the data, storage backup and recovery data during the project.
Benefits of participation:
The main benefit of participation is that the study will provide new knowledge
about individuals’ awareness of age-related changes as they get older and its
association with emotional and physical well-being and cognitive functioning.
This knowledge may be useful in developing interventions to promote healthy

348
Appendices

ageing and prevent cognitive decline and tackle low levels of physical and
emotional well-being.
There are no financial inducements, and participants will take part in the study
voluntarily. Participants will receive feedback about results in the form of a
newsletter following the completion of the study. This newsletter will contain a
lay summary of the findings from the study. This information will also be
available on the PROTECT website.
Researcher I have previously collected, stored and analysed data during my
experience undergraduate and master’s degree studies (at the University of Padua, Italy)
and during a postgraduate research internship (at the University of St.
Andrews, Scotland).

349
Appendices

Appendix G: Summary of the PhD project outlined as part of the application to


request PROTECT data

SURVEY SUMMARY
Scientific title of project: “Awareness of age-related change and its association
with mental and physical well-being in cognitively healthy adults living in the UK”
Title for participants: “How we see ourselves: changes over time”
Principal Investigator: Serena Sabatini
Organization and Department: Centre for Research in Ageing and Cognitive Health
(REACH), University of Exeter Medical School, South Cloisters, St Luke’s Campus,
Exeter EX1 2LU, UK
Co-investigators: Linda Clare, L.Clare@exeter.ac.uk
Clive Ballard, C.Ballard@exeter.ac.uk
Kaarin Anstey, Kaarin.Anstey@anu.edu.au
Barbora Silarova, B.Silarova@exeter.ac.uk
Rachel Collins, R.A.Collins@exeter.ac.uk
Sarang Kim, Sarang.kim@anu.edu.au

Primary contact for study: Serena Sabatini


Address: Centre for Research in Ageing and Cognitive Health (REACH), University
of Exeter Medical School; South Cloisters; St Luke’s Campus; Exeter EX1 2LU; UK
Email: ss956@exeter.ac.uk
Tel: +44 (0) 1392 726 754
Start and completion dates: Project start Date: 24 September 2017
Project completion Date: 24 March 2021

350
Appendices

Background
The positive association between the psychological construct of subjective ageing
(SA) and good health and well-being has been well-documented (Westerhof et al.,
2014). Previous studies have found associations between SA and physical and
cognitive functioning (Levy, Slade, & Kasl, 2002; Robertson, King-Kallimanis &
Kenny, 2016; Robertson, Savva, King-Kallimanis, & Kenny, 2015), risk of medical
problems such as cardiovascular events, falls and hospitalisations (Levy, Slade,
Chung, & Gill, 2015; Levy, Zonderman, Slade, & Ferrucci, 2009; Moser, Spagnoli, &
Santos-Eggimann, 2011), as well as Alzheimer’s-like neuropathology (Levy et al.,
2016). SA (Diehl, Wahl et al., 2014) is used in theories and empirical evidence as an
umbrella term and covers a range of concepts that try to understand the ways in
which individuals experience the ageing process. Examples of such concepts are
subjective age and age identity (Kastenbaum, 1972), self-perception of ageing
(Kleinsphen-Ammerlahn et al., 2008; Kötter-Grühn & Hess, 2012; Kötter-Grühn et
al., 2009), attitudes toward ageing (Bennet & Ekman, 1973) and age stereotypes
(Levy, 2009).
To overcome limitations of previous SA concepts, Diehl and Wahl (2010)
developed the theoretical concept of awareness of age-related change (AARC).
AARC refers to all those events, situations and circumstances that make individuals
aware that their behaviour, level of performance, or ways of experiencing their lives
have changed due to having grown older (Diehl & Wahl, 2010, page 340). It is a
multidirectional concept, assessing both positive and negative subjective
experiences (AARC gains and AARC losses) of awareness of ageing. It is also a
multidimensional concept as it assesses awareness in five hypothesised life domains
- health and physical functioning, cognitive functioning, interpersonal relationships,
socio-cognitive and socio-emotional functioning, and lifestyle/engagement.
Empirical evidence on AARC indicates that AARC explains greater
proportions of variance in measures of well-being and health-related outcomes than
the variance accounted by well-established subjective ageing constructs such as
subjective age and attitudes toward own ageing (Brothers, Gabrian et al., 2017, from
Kaspar et al., 2017, p. 5). AARC predicted depressive symptoms over 2.5 years in a
sample including middle-aged and older adults (Dutt, Gabrian, & Wahl, 2016) and it
is also related to negative affect (Miche et al., 2014; Neupert et al., 2017), emotional
well-being and health (Brothers, Miche, Wahl and Diehl, 2015). A feasibility
351
Appendices

intervention aiming at modifying individuals’ negative views of ageing (Brothers and


Diehl, 2016) resulted in an increase in perceived age-related gains and a decrease
in perceived age-related losses.
For instance such knowledge may be vital in the design of future interventions
addressing maintenance of cognitive health.
This study will provide data for four sub-studies that aim:

5) to validate a measure of Awareness of age-related change: the AARC-10 SF


(Kaspar, Gabrian, Brothers, Wahl, Diehl, 2017) and the AARC-cognitive
functioning subscale from the AARC-50 item questionnaire (Brothers, Gabrian,
Diehl, Wahl, 2018) in people aged at least 50 years old and living in the UK.
As part of this sub-study convergent validity of the AARC measures and other
two measures of subjective age, subjective age (Barret, 2003) and ATOA
(Lawton, 1975), will be explored. Finally this sub-study aims to explore how the
profile of AARC gains and losses changes with age, taking into account possible
confounding factors such as current and past physical health conditions,
perceived pain, hearing and visual impairments.
A qualitative study on adults aged 70-88 years old (Miche et al., 2014) suggested
that older participants report both more negative AARC and more positive AARC;
however this needs to be further explored in a wider and more heterogeneous
sample.
6) to investigate the association between AARC (gains and losses) and physical
health by exploring indicators of physical limitations including current and past
physical health conditions, hearing and visual impairments, physical pain, self-
rated health, instrumental activities of daily living and quality of sleep.
Participants’ ability to carry out instrumental activities of daily living provides
information on physical limitations arising from either chronic illness or
impairment, or acute illness. Similarly, quality of sleep may be an indicator of
physical health since self-perceived sleep problems have found to be associated
with poorer general health (Kupper, 1995).
By exploring the association between AARC (gains and losses) and aspects of
self-rated health, this sub-study will make it possible to explore which aspects are
most strongly associated with AARC (gains and losses).

352
Appendices

7) to investigate the association between AARC (gains and losses) and symptoms
of depression and anxiety and to understand the role of rumination in the
association between AARC and symptoms of depression and anxiety.
8) to understand the role of cognition in relation to AARC by looking at the possible
association between the AARC-cognitive subscale (gains and losses) and
objective cognitive performance.
As part of this sub-study the possible association between the AARC-cognitive
subscale score and cognitive abilities rated by an informant will be explored.

Summary of study design


The present study uses a longitudinal cohort design amd is implemented through
PROTECT (Platform for Research Online to investigate Genetics and Cognition in
Ageing; http://www.protectstudy.org.uk/about.aspx). Participants who are part of
PROTECT complete an annual assessment every year in November. As part of the
annual assessments in November 2018 and November 2019, participants will be
asked to fill in additional questionnaires specific to this study. In addition to these
questionnaires, we will use data collected as part of the main PROTECT study
during the annual assessments in November 2018 and November 2019.

Participants’ eligibility
Individuals were potentially eligible to participate in the PROTECT Study if they were
UK residents, English speakers, aged 50 years and older, had access to a computer
and internet and did not have a clinical diagnosis of dementia or mild cognitive
decline at the time of recruitment. Participating in the PROTECT study does not
preclude participation in intervention trials or other observational studies.
For the present study we have no additional inclusion or exclusion criteria.

Data collection
As part of PROTECT participants have to give their informed consent at every
annual assessment. After giving their consent, participants will be able to complete
the new questionnaires as well as all the measures already included in PROTECT at
each time point, and will be able to save their answers and access the survey
multiple times until they have fully completed all measures.

353
Appendices

Measures
An overview of the measures that we plan to use for this study, including the time
point of data collection and whether the measure is new or already part of PROTECT
is provided in Table 1. See the Appendix for a list of the additional questions for this
study which will be included with the PROTECT assessment in 2018 and 2019.

Table 1. Overview of measures

Constructs Description Source Time of data


collection
NEW MEASURES
Awareness of age-related Awareness of Age Related New November 2018
change (AARC) Change-10 SF (AARC-10 SF) and 2019
(Kaspar, Gabrian, Brothers,
Wahl, & Diehl, 2017) - 10 items.
AARC-cognitive subscale from
the AARC-50 item questionnaire
(Brothers, Gabrian, Diehl, Wahl,
2018)- 8 items
Attitudes toward own Attitudes Toward Own Ageing 5- New November 2018
ageing item measure (ATOA-5) (Lawton, and 2019
1975)
Subjective age Single item question adapted New November 2018
from the National Survey of and 2019
Midlife development in the United
States (MIDUS, Barrett, 2003, p.
S104)
Rumination Ruminative Response Scale New November 2018
(RRS) (Nolen-Hoeksema & and 2019
Morrow, 1993) - 22 items
Self-rated health One item from the SF-36 (Ware, New November 2018
1992) and 2019
PROTECT MEASURES
Objective cognitive Cognitive Test Package 1 and PROTECT November 2018
abilities Cognitive Test Package 2 and 2019
(CogTrack)*
Cognitive abilities rated by Informant questionnaire on PROTECT November 2018
an informant Cognitive Decline in the Elderly- and 2019

354
Appendices

Constructs Description Source Time of data


collection
informant version (IQCode-
informant) (Jorm, & Jacomb,
1989)
Depressive symptoms Patient Health Questionnaire-9 PROTECT November 2018
(PHQ-9) (Kroenke, Spitzer, & and 2019
Williams, 2001) to assess
depressive symptoms over the
last two weeks and The
Composite International
Diagnostic Interview- Short form
(CIDI-SF) (Kessler, Andrews,
Mroczek, Ustun, & Wittchen,
1998) to assess lifetime
depression
Anxiety symptoms The Generalised Anxiety Disorder PROTECT November 2018
questionnaire (GAD-7) (Spitzer, and 2019
Kroenke, Williams, & Löwe 2006)
to assess anxiety symptoms over
the last two weeks and CIDI-SF
(Kessler, Andrews, Mroczek,
Ustun, & Wittchen, 1998) to
assess lifetime anxiety

355
Appendices

Demographic variables: PROTECT November 2018


1. age (all items are in Items that may have
2. sex PROTECT data set changed will be
3. ethnicity demographics collected again in
4. being a carer section except being November 2019: 1.
5. marital status a carer which is in Being a carer; 2.
6. Indicators of the lifestyle section) Marital status; 3.
socioeconomic Employment status; 4.
status: current Voluntary work
employment and
education level
7. Employment
status
8. Voluntary work
Instrumental Activities of Instrumental PROTECT November 2018 and
Daily Living Activities of Daily 2019
Living (IADL)
(Lawton and Brody,
1969)
Lifetime physical health Medical health PROTECT November 2018 and
history 2019
Past history of mental Mental health PROTECT November 2018 and
health 2019
Physical pain Pain scale from PROTECT November 2018 and
PROTECT main data 2019
set
Hearing and visual Medical health PROTECT November 2018 and
impairments 2019
Sleep St Mary’s Sleep PROTECT November 2018 and
Scale (Ellis, Johns, 2019
Lancaster,
Raptopoulos,
Angelopoulos, Priest,
1981) from
PROTECT lifestyle
assessment

New measures will be completed in the following order:


1. Self-rated health.

356
Appendices

2. AARC-10 SF.
3. AARC-50 cognitive functioning subscale.
4. ATOA measure.
5. Subjective age: single item taken from the National Survey of Midlife
development in the United States.
6. Ruminative Response Scale.

357
Appendices

Appendix H: Memorandum of understanding set in place with the authors of


the AARC measures

358
Appendices

Appendix I: Data agreement form

359
Appendices

Appendix J: PROTECT protocol

Platform for Research Online to investigate Genetics and Cognition in Ageing


(PROTECT)
Protocol: Version 15
07/04/2017

Background
Cognitive function is known to naturally decline with age, and this is an established
factor in the ageing brain. This results in a 'slowing down' of cognitive abilities such
as memory, reasoning or attention. However, there is still only limited evidence for
the specific mechanisms that underlie these subtle changes.
There is a growing body of evidence indicating the role for genetic
factors in cognition. Studies using cohorts of twins, such as the Twins Early
Developmental Study (TEDS) in the UK, have demonstrated the important
role of genetics as a risk factor in the development of cognitive disabilities.
Furthermore, molecular studies have also begun to identify some of the genes
responsible for this genetic influence. Studies have identified a number of
genes that appear to govern cognition in older adults. These include 20 single
nucleotide polymorphisms (SNPs) identified in a large genome-wide
association study that appear to determine performance in specific cognitive
domains (Docherty et al 2010). A number of genetic variants are also linked to
cognitive function and the risk of development of cognitive impairment in later
life. These include ApoE4, BIN1, COMT and TOMMO40 (www.Alzgene.com).
Furthermore, recent work has indicated that whilst individual genes have been
shown to predict cognitive decline, it is likely that combinations of genetic
factors have a cumulative effect on cognition. To date there have been no
large studies examining the polygenic impact of these key genetic factors on
the progression of cognition in older adults over the long term or how they
might predict an individual's future cognitive decline.
There is also some evidence indicating the role of specific lifestyle and
medical factors, such as exercise and smoking status which may influence
cognition. However, to date this potential association remains unclear.

360
Appendices

In addition to the need to better understand the processes of cognition and the
underlying mechanisms for differential decline between individuals, this knowledge
could provide key data to improve our understanding of cognitive impairment and
dementia. There are 800,000 people with dementia in the UK, and this number will
continue to rise as the population ages. In order to develop better approaches to the
prevention and treatment of this devastating condition, it is essential to understand
the processes that govern cognitive decline in older adults and the factors that
influence it.

Objective
This ten-year study aims to determine the role of defined genetic factors on cognition
in people over 50 over a period of eight years. It will additionally explore the role of
lifestyle and medical factors on cognition to understand how they contribute to
cognitive decline.
Additional objectives of this study are to develop a minimum data set and consent-
for-consent cohort to support future research in cognition in people over 50, and to
investigate the short-and long-term sensitivity of a series of cognitive tests when
completed by this group.

Research Questions
Primary research question: How do defined genetic factors affect cognition in people
over 50 over the long term?
Secondary research questions:
1. Which genetic factors (SNPs and candidate genes) predict cognition and cognitive
decline in people over 50?
2. Do individual genetic factors (SNPs and candidate genes) exert a cumulative
effect on cognition and cognitive decline in people over 50 in the long term?
3. How do lifestyle factors affect cognition in people over 50 over the long term?
4. What is the feasibility of an online cohort study approach for a larger future
Genome Wide Association Study?
5. How do mental health factors such as depression and anxiety associate with
cognition in people over 50?
6. What are the short- and long-term learning effects and test re-test sensitivities of
the cognitive test batteries employed in the PROTECT study?
361
Appendices

Study design
Design: Longitudinal study
Study length: Ten years (Two years recruitment + Eight years follow-up)
Target population: Healthy adults over 50
Setting: Living in the community, cohort to be primarily supported online
Inclusion criteria: Aged over 50, reside in the UK, have access to a computer and the
internet
Exclusion criteria: An established diagnosis of dementia or cognitive impairment
Sample size: 5000 people over 50

Recruitment
All potential participants will be directed to the study website for detailed information
and to register and consent to take part. Recruitment will be achieved through the
following channels:
1. Promotion of the study through communication channels at King’s College
London (KCL), the Biomedical Research Centre at KCL, the University of Exeter and
the Royal Devon and Exeter NHS Trust. This will include publicity through the media
and University press partners, and online content
2. Invitation of people over 50 who are participants in existing cohorts and
completed trials hosted by KCL, SLaM and Exeter. Research staff will contact these
individuals to provide information about the study, send out information through the
post and invite them to register online. This will include the Exeter 10,000 (EXTEND)
cohort hosted through the Royal Devon & Exeter NHS Trust.
3. Participants will be signposted to the PROTECT site by the Department of
Health ‘Join Dementia Research’ website through the standard procedures for the
site.
4. Leaflets will be made available for placement in GP surgeries and memory
clinics and by completed trial staff at end-of-study visits.
5. Depending on the success of the recruitment channels described above,
purchasing of public advertising and invitation letters for GP surgeries will be
considered to support recruitment.

362
Appendices

Data collection protocol


Baseline data collection
All data will be collected through the online study platform, with the exception of DNA
samples (described below). Participants will provide the following baseline data:
1. Demographic characteristics: Age, gender, ethnicity, marital status, education
level, employment status, NHS number
2. Medical history (amended): Participants will be asked to select existing
medical conditions and prescriptions through a tick-box selection with optional free
text. They will also be asked to provide their weight and height.
3. Lifestyle (amended): Participants will use tick-box selection or enter numerals
to indicate their current amount and type of exercise, smoking status, diet, use of
dietary supplements, use of technology, current use of cognitive training and number
of languages spoken
4. Mental Health history: Participants will complete a series of validated mental
health scales to provide information about their history and current status with regard
to stress, depression, anxiety, mood and psychosis. The test battery also includes
items relating to childhood stress and alcohol and drug use.
5. Mild Behaviour Impairment scale – a validated scale to detect early changes
in behaviour in older adults
6. Pain Interference Scale (4-item short form) – a validated scale to capture
regular experience of pain in adults
7. St Mary’s Sleep Questionnaire – a nine-item questionnaire that captures sleep
quality in the last month.
8. Menopause and fertility assessment (females only) – a 17-item questionnaire
to capture past and current information about female fertility and menopause, to
enable investigation of links with cognitive health
Participants will complete the following validated measures of cognition, three times
in one week at baseline:
1. Baddeley Grammatical Reasoning test. This test has been shown to correlate with
measures of general intelligence and involves determining the accuracy of a series
of grammatical statements about a presented picture.
2. Simple and Choice Reaction time - brief computerised test of reaction time

363
Appendices

3. Paired Associate Learning – a validated measure of verbal learning which is highly


sensitive to change in cognition and has been used to predict conversion to
Alzheimer’s Disease in people with cognitive impairment
4. Spatial Working Memory – a simple memory test involving recalling the position of
a hidden object behind a series of panels
5. Instrumental Activities of Daily Living – a well validated scale used to determine
overall function in daily live.
6. Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE): An
informant-rated measure of cognition validated to detect change in cognitive
function. A self-rated version of this measure will be included as an alternative if an
informant is not available.
7. The Wesnes Online Cognitive Testing System: A cognitive test battery that
includes interactive assessment of reaction time, grammatical reasoning, picture
recognition and digit vigilance. This battery complements the core test battery and
will provide data regarding learning effects and more sensitive detection of changes
in cognition. An expanded version of this test battery, containing tasks relating to
verbal recall, recognition and working memory, will be an optional additional outcome
measure. Participants will have the option of completing this extended version each
time they complete the test battery.
Before each assessment point participants will be asked a brief mood questionnaire,
using the Bond-Lader VAS of Mood and Alertness (Bond and Lader, 1974), to give a
real-time assessment of mood on the day of assessment.
Participants will also be asked for permission for the research team to request their
medical notes from their primary care practitioner to enable analyses of study
outcomes against medical status.
Participants will be contacted throughout the study and offered the opportunity to
provide feedback on aspects of the PROTECT study. This will be in the form of
simple online surveys and will be optional. Participants will have the chance to opt in
or out of this process.

Genetic Sampling
Participants will also be asked to provide a DNA sample. Two saliva sample kits will
be posted to each participant to self-administer at home. This is a very swift, simple
and painless procedure.
364
Appendices

Participants recruited through the London (King’s College London) site will provide
DNA samples through the NIHR Bioresource (Biobank) A freepost envelope will be
provided for participants to return their sample to the Bioresource. In the unusual
circumstance where someone is unable to give a saliva sample the Bioresource will
offer a cheek swab or blood sample as an alternative option. DNA will be extracted
and stored according to Bioresource protocols. Genome-wide association studies will
also be conducted. These will be carried out using commercially available chips
(Pyshc-plus) which allow the analyses of over 750,000 genetic changes. The
samples will be stored, processed and analysed within the NIHR Bioresource,
enabling these participants to be part of this additional shared research
infrastructure. Specific wording relating to involvement in the Bioresource will be
included in the consent form and Participant Information Sheet.
Participants recruited through the Exeter (University of Exeter and Royal Devon &
Exeter NHS Foundation Trust) will provide DNA samples through the Exeter Clinical
Research Facility, in partnership with the international life sciences company, LGC.
Saliva kits will be posted to participants from the Exeter site and returned to the
Clinical Research Facility. These will be sent in batches to LGC under a dedicated
agreement for DNA extraction, storage and genotyping. If participants have already
provided a DNA sample as part of their involvement in a linked cohort (for example,
Exeter 10,000) they will not be required to provide a second sample. Anonymised
genetic data will be shared through a sharing agreement at the University of Exeter,
utilising unlinked participant IDs so ensure data protection.

Follow-up data collection


From the date of baseline data collection to the end of the data collection period,
contact will be maintained with participants through the study website and a bi-
annual newsletter with updates from the study. Each participant will be asked to
complete the following outcome measures taken at baseline on an annual basis:
1. Medical status
2. Lifestyle factors
3. Use of Technology Questionnaire – participants will be asked about their use
and frequency of use of technology, devices, the internet and social media.
3. Baddeley Grammatical Reasoning test (three times in one week)
4. Simple and Choice Reaction time (three times in one week)
365
Appendices

5. Paired Associate Learning (three times in one week)


6. Spatial Working Memory (three times in one week)
7. Instrumental Activities of Daily Living (three times in one week)
8. IQCODE
9. Wesnes Online Cognitive Test (three times in one week), with optional
expanded version
10. Mental health questionnaire – an abridged version of the questionnaire set
completed at baseline, which measures recent change and does not include
duplications of mental health history.
11. Mild Behaviour Impairment scale – a validated scale to detect early changes
in behaviour in older adults
12. Menopause and fertility questionnaire (females only) – a scale tailored to capture
changes in fertility

Consent for Contact


Participants will give consent for contact regarding future related research studies
into ageing, mental health and dementia. This will include targeted recruitment of
participants based on the data they provide at baseline and annual follow-up,
including cognitive scores, mental health assessments, lifestyle information, medical
history and demographics. In cases where participants are potentially eligible for a
new study contact will be made through the PROTECT email system. The email will
contain brief information about the referring study, a link to further information and
instructions on how to register their interest. No personal details will be passed to
external researchers or organisations without consent of the participant.
This consent for contact mechanism will enable PROTECT to support future
research related to ageing and mental health that has direct relevance to the stated
objectives for the study.

Sample size
Previous work to identify risk polymorphisms has indicated that the best individual
SNPs explain 0.5% of the variance and that cumulatively the best 10 SNPs explain
3% of the variance (Butcher et al 2008, Meaburn et al 2008). Based upon standard
power calculation tables, a sample of 4200 individuals would give 99% power to
detect genetic effects explaining 0.5 of the total variance of cognitive performance,
366
Appendices

uncorrected for multiple testing (P = 0.05, one-tailed). A sample of 5000 people will
allow for incomplete data and people dropping out over follow-up. This represents
the minimum number of participants required for genetic analyses. A larger, open
recruitment target will be implemented to provide additional power to all analyses
within the cohort.

Data analyses
Conventional parametric statistical methods (Student’s t-test) will be used to test for
differences in cognition between people with the different candidate alleles. Because
quantitative genetic research strongly suggests that the majority of genetic effects
are additive, we will also test SNPs for their additive effect. Therefore, genotypes of
SNPs passing quality control will be tested for additive genetic effects using a
Pearson correlation (r) and coding the three observed genotypes such that 0 ¼ AA,
1 ¼ AB and 2 ¼ BB. In addition, we followed a procedure recommended by Balding
(2006) to test whether a non-additive model predicted significantly better than an
additive model.

Data storage
Personal data will be collected on a separate database from all other data. This will
be matched up by a unique ID number for each participant. Personal data databases
will be held separately by King’s College London and Exeter University, dictated by
the recruitment route taken by the participant. Databases will be password protected
and located on University shared drives which only the study team at that University
have access to. The data will also be on the Universities’ servers which are not
accessible by anyone outside the organisations.
All non-identifiable data (i.e. not personal data) will be stored in two independent
databases at the two sites in Exeter University and King’s College London. This data
will be accessible for import to either site to enable creation of a full database for
analyses purposes.
Data generated through the Wesnes Online Cognitive Test will be stored in a
secure server and transferred to the host site via a regular encrypted transfer. Data
will also be stored online on Google Cloud to provide a back-up database
(https://cloud.google.com/terms/data-processing-terms ). Raw data (not including
personal details) will be stored in encrypted form, subject ISO/IEC 27001:2005
367
Appendices

Certification. Access to DNA analyses data will be coordinated through the


NIHR Bioresource and LGC Ltd according to the approved regulations and
security arrangements for the resource. Investigators will have full access to
all data relating to this study. Personal data will be transferred between King’s
College London and the NIHR Bioresource (name, contact details and DNA /
RNA sample ID number). Personal data and cognitive outcome data will be
transferred between investigators within each institution (King’s College
London / Exeter University) but not across institutions to enable long-term
follow-up for collaborating studies. Only core study team members will have
access to personal data through the electronic database. This will include the
Chief Investigator, study coordinators and IT specialist.

Participant Withdrawals
In the event that a participant chooses to withdraw from the study they will have the
option to retain or destroy any identifying data stored within the study (email address,
home address, GP details). This option will be given at the point of withdrawal. A
name will be retained to ensure a record of consent is kept, according to upcoming
legislature. All anonymised data will also be retained. This will include all
anonymised assessment data, genetic data and the anonymised extracted DNA
sample. Participant withdrawal will lead to automatic destruction of any remaining
saliva sample or un-extracted DNA and withdrawal from the NIHR Biorsource,
according to the Bioresource protocols, if relevant.

Patient and Public Involvement


The Patient and Public Involvement group at King's College London and separate
Lay Advisory Group at the University of Exeter will act as consultants to support the
study as it progresses, including the design of the website, recruitment strategy and
dissemination activities.

368
Appendices

Appendix K: Participant information sheet for the PROTECT study

Participant Information Sheet


Version 16 Date: 14/02/2017

PROTECT Study: Platform for Research Online to investigate Genetics and Cognition
in Ageing

Invitation to take part in a research study

We would like to invite you to take part in our research study. Before you decide, we would
like you to understand why the research is being done and what it would involve for you.

Please take time to read the following information carefully and discuss it with family or
friends if you wish. We recognise that there is a lot of information contained within this
document. If you have any further questions, please contact a member of the study team
(details are on the last page of this information sheet).

It is important that you understand that you do not have to take part in the study and that if
you do take part you are free to withdraw at any time. If you decide to take part we will ask
you to read and sign the declaration on the next page of the website.

What is the purpose of the study?

This study aims to understand how the functioning of the brain changes as we age. In
particular the study will look at how certain genes and lifestyle factors (such as exercise or
education) affect the way our brain ages. This will provide valuable information about the
brain and could inform future research to prevent conditions such as dementia. The study is
being led by King’s College London.

Why have I been invited?

We are inviting adults over 50 from across the UK to take part in this study. We are looking
for 5000 people to join the study for the next ten years.

In order to participate, you will also need to

! Have a good working understanding of the English language


! have the ability to use a computer with internet access.

If you have an established diagnosis of dementia from your doctor then unfortunately you will
not be eligible for this study.

Do I have to take part?

It is up to you whether or not to join the study. The purpose of this information sheet is to
describe the study in detail to help you make your decision. If you agree to take part, you will
then need to read and sign a consent form on the website. You are free to withdraw at any
time, without giving a reason. This would not affect the standard of care you receive through
your own General Practitioner or local NHS services. This study does not replace those

369
Appendices

Appendix L: Application for access to PROTECT Study Data (Follow-up)


Scientific title: “Awareness of age-related change and its association with
Title of project
mental and physical well-being in cognitively healthy adults living in the
UK”

Principal Investigator Serena Sabatini

Centre for Research in Ageing and Cognitive Health (REACH), University


Organization and
of Exeter Medical School, South Cloisters, St Luke’s Campus, Exeter
Department (full
EX1 2LU, UK
contact details)
Tel: +44 (0) 1392 726 754
ss956@exeter.ac.uk
Prof. Linda Clare, Prof. Clive Ballard (PROTECT investigator), Prof.
Co-investigators
Obioha Ukoumunne, Prof. Kaarin Anstey, Dr Rachel Collins, Dr Sarang
Kim

Name: Serena Sabatini


Primary contact for
Address:Centre for Research in Ageing and Cognitive Health (REACH),
study
University of Exeter Medical School, South Cloisters, St Luke’s Campus,
Exeter EX1 2LU, UK
Email: ss956@exeter.ac.uk
Tel: 07757029595

Start and completion Project start Date: 24 September 2017


dates
Project completion Date: 24 March 2021

Participant characteristics
What is being
requested? Sample Size We are requesting available 2020 data, as outlined

(Place a X in all that for all participants who complete the AARC

apply) questionnaire at the 2019 assessment.

Participant As part of the annual assessment in 2020


characteristics participants will be asked to fill in the AARC
questionnaire (please see further details in Serena
(Provide as much
Sabatini Study summary). Therefore, we request
detail on required
data for all participants who will fill in the AARC
participants as
questionnaire as part of their annual assessments
possible)
starting in January 2020.

Anonymised Datasets

370
Appendices

Two time-points: Nov.


2018 and Nov. 2019

Full PROTECT Data set

Partial PROTECT Demographics All items


Dataset
Medical History Selected items: 1.1, 1.2,
1.3, 1.4, 1.5, 1.6, 1.7, 1.8,
1.9, 1.10

Lifestyle Items: 1.1., 1.2, 1.3, 2.1,


3.2, 3.2, 3.3, 4.1, 4.2, 4.3,
4.5, 5.1, 5.2, 5.3, 5.4, 5.5,
6, 7.1, 9.1

Cognitive Test Yes


Package 1 (PROTECT)

Cognitive Test
Package 2 (CogTrack)*

IQCode Yes

IADL Yes

Mental Health
Questionnaire (full)

Mental Health PHQ-9; CIDI-SF; GAD-7;


Questionnaire (partial – Items from 1.1 to 1.4
please specify which
sections):

Mild Behaviour 1.2, 1.2, 1.4, 1.6


impairment Scale

Pain short-form Yes

Sleep Yes

Fertility / menopause Yes

Genetic Data / Samples

Genotype / Not requested


GWAS Data

371
Appendices

Access to Not requested


extracted DNA
samples

Personal details and re-contact

Personal details We do not request any personal details. All


data communication with participants will be managed by
the PROTECT team.

Re-contact of As part of the PROTECT study added new


participants for assessments for all participants who agreed to take
nested PROTECT part in their annual assessment starting in January
study 2019. Additional questionnaires were: 1. Self-rated
health; 2. AARC-10 SF; 3. AARC-cognitive subscale
e.g. addition of
from the AARC-50 item questionnaire; 4. ATOA
new assessment
measure; 5. Subjective age: one item taken from
or email
the National Survey of Midlife development in the
distribution of
United States; and 6. Ruminative Response Scale.
survey etc.
Details are provided as part of the Serena Sabatini
study summary.

Re-contact of Not requested


participants for
separate study

e.g. identification
of participants for
clinical trial

The aims of this study are:


Study Aims &
Objectives 9) to validate a measure of Awareness of age-related change (AARC):

Please outline the aims the AARC-10 SF (Kaspar, Gabrian, Brothers, Wahl, Diehl, 2017) and

of your study the cognitive functioning subscale of the AARC-50 (Brothers et al,
2018) in people aged at least 50 years old and living in the UK;
10) to investigate the association between AARC (gains and losses) and
physical health by exploring indicators of physical limitations
including current and past physical health conditions, hearing and
visual impairments, physical pain, subjective health, instrumental
activities of daily living and quality of sleep;

372
Appendices

11) to investigate the association between AARC (gains and losses) and
symptoms of depression and anxiety and to understand the role of
rumination in the association between AARC and symptoms of
depression and anxiety;
12) to understand the role of cognition in relation to AARC by exploring
the possible association between the AARC- cognitive subscale
(gains and losses) and objective cognitive performance. As part of
this sub-study the possible association between the AARC-cognitive
subscale score and cognitive abilities rated by an informant will be
explored;
13) to explore whether age predicts changes in AARC (gains and losses)
over one year;
14) to explore the association between attitudes toward own ageing
(ATOA) and cognition and to compare the associations found in
PROTECT participants with those found in a sample of people living
with dementia;
15) to explore the associations between AARC and lifestyle (e.g. items
on smoking, alcohol, exercise, cognitive activities);
16) to explore the association between menopause and AARC.

The study “Awareness of age-related change and its association with


Brief Description of
mental and physical well-being in cognitively healthy adults living in the
Project
UK” will be a longitudinal study exploring the concept of awareness of
(If access to personal
age-related change (AARC) and how it is related to mental well-being,
details or genetic data
physical health and cognitive abilities. Data will be collected at two time
is required please
points since the study aims to explore the direction of the associations.
include study protocol)
Since assessments of mental well-being, physical health and cognition
are part of PROTECT, we request access to PROTECT data and to add
extra questionnaires to assess variables such as rumination and AARC
that are not already assessed in PROTECT.

Tick relevant
Ethical approval
box
status

(Some data requests No additional approval required (please justify)

may fall within the


Ethics submission in process Ethics will be
existing PROTECT managed as an
approvals. If in doubt, Anticipated date of approval:
amendment to
please contact a REC: the PROTECT
PROTECT Ethics
R&D dept:
application. The

373
Appendices

administrator to Reference number: process will be


discuss) managed by the
(Please attach confirmation of sponsorship)
PROTECT team.

Ethical approval confirmed

Date of approval:

REC:

R&D dept:

Reference number:

HRA approval required:

(Please attach letter of approval from REC and HRA


if appropriate)

4. Sample size and participant characteristics: We are requesting


Justification for
available 2020 data for all participants who complete the AARC
request
questionnaire at the 2019 assessment.
Please provide
5. Data type: Quantitative – Partial PROTECT data set as outlined
rationale for the data or
above and new measures added as part of Serena Sabatini
samples requested
study.
above, including:
6. New assessments. In order to collect data on constructs that are
6. Sample size not covered in the PROTECT dataset, new questionnaires will be
7. Participant added as part of the PROTECT study.
characteristics
8. Data type
9. Sample type
10. New study or
assessment

If relevant provide a
clear justification for
requests that include
both access
participants with
specific genotypes and
personal details,
including any plans to
disclose genetic

374
Appendices

information and how


this will be managed.

- Managing participant contact through PROTECT


Involvement of
- Technical support to include new questionnaires
PROTECT
- Data management
Investigators
- Help to identify participants that fulfil criteria for mild cognitive
If you expect to require
impairment and age-associated cognitive decline based on
support from
standard deviation from age-matched norms across the series of
PROTECT
cognitive tests.
investigators, please
give details. This may
include:

- Managing
participant
contact
through
PROTECT
- Technical
support for
PROTECT
website, incl
new
assessments
- Statistical
support
- Data
management

Budget

Do you have budget


available to support
this request? Please
provide details

This is particularly
relevant to requests
that require active
involvement of
PROTECT
investigators or

375
Appendices

amendments to
PROTECT study
assessments

Further Information

Please provide any


further relevant
information

Key information for applicants

- Use of PROTECT data is subject to full ethical approval. Some requests may
fall under existing PROTECT approvals. Please contact the study
administrator to discuss further
- This application will be reviewed by the PROTECT Strategy Group to ensure
the request is appropriate to the data, and does not conflict with existing
research and analyses within the PROTECT portfolio
- GWAS data will be released as summary data only (list of markers with effect
size / overall frequency of requested alleles per marker) and will not include
individual-level genetic data
- The PROTECT investigators ask that any publications or outputs arising from
the use of PROTECT data include authorship for PROTECT investigators and
acknowledgement of PROTECT support and funders.
- CogTrackTM is provided by a third party. Use of this data may be subject to
additional cost.
Document to be submitted to maria.megalogeni@kcl.ac.uk
Questions regarding this process should be directed to maria.megalogeni@kcl.ac.uk
and zunera.2.khan@kcl.ac.uk in the first instance

376
Appendices

Appendix M: Approval for access to PROTECT data (Follow up)

Serena Sabatini 07/04/2020


South Cloisters, University of Exeter
St. Luke’s Campus, Heavitree Road
Exeter, EX1 2LU

RE: Data Access Request


Ref: PROT_UK_DA_037

Dear Serena,

I am pleased to confirm that your Data Access Request to the PROTECT Study for your project
Awareness of age-related change and its association with mental and physical well-being in
cognitively healthy adults living in the UK has been approved. This letter is a formal
acknowledgement that the following activities will go forward:

1. Creation of dataset for analysis (Baseline and Follow-Up data) to include the following:
a. Lifestyle questionnaire
b. Fertility and menopause questionnaire

This approval is also subject to any additional agreements, contracts or authorship arrangements as
negotiated with the PROTECT investigators. Any amendments to your request should be submitted to
the Committee for review.

On behalf of the PROTECT Steering Committee I wish you the best of luck with your research. We look
forward to working with you.

Yours sincerely,

Dr Byron Creese

Interim PROTECT Academic Lead, University of Exeter

On Behalf of the PROTECT Study Steering Committee

CC by email: Ellie Pickering, (PROTECT Research Centre Coordinator, University of Exeter); Zunera
Khan (PROTECT Trial Manager, King’s College London)

377
Appendices

Appendix N: PROTECT data agreement form 2020

PROTECT data usage agreement – Internal

This data usage agreement governs access to the PROTECT data as requested below. You
agree to be bound by these terms. You may only use PROTECT data for non-commercial
research.

Principal Investigator Name: Prof Linda Clare


(PI) / Supervisor name
Position: Professor of Clinical Psychology of Ageing and Dementia
and position:
Email: L.Clare@exeter.ac.uk

Other individuals who Serena Sabatini,


need access to the
PhD student,
PROTECT data
ss956@exeter.ac.uk
(including any students)
(“Users”):

Institution: University of Exeter

Project Title: Awareness of age-related change and its association with mental and
physical well-being in cognitively healthy adults living in the UK

PROTECT data: Anonymised datasets from the main PROTECT study which has
involved the collection of information from participants by the
PROTECT study team, limited specifically to the datasets set out in
Schedule 1 to this agreement.

PROTECT study team: The University of Exeter, King’s College London and South London
and Maudsley NHS Foundation Trust. The PROTECT study team is
being led by the University of Exeter and this agreement is with the
University of Exeter.

Permitted usage: Non-commercial research for the Project only.

Date of request: 21/01/2020

As PI / Supervisor for the Project detailed above, I agree to comply, and ensure the
Users comply, with the following terms covering the PROTECT data:

378
Appendices

1. The PROTECT data has been anonymised, but it must still be stored securely
and only transferred in an encrypted form. The security procedures in place at
the Institution must be complied with.
2. The PROTECT data is only to be used for non-commercial research purposes
as specified in paragraph 4.
3. The PROTECT data is not to be transferred to, or used by, any other person
or organisation other than the Users listed above. The PROTECT data must
be kept confidential.
4. Usage of the PROTECT data is strictly limited to addressing the
hypotheses/research questions defined in the Project, as attached as
Schedule 2 to this agreement. Any broadening of scope of research must be
discussed with the PROTECT study team and cannot proceed without their
written authorisation.
5. The Project must have any required ethical approval and the PROTECT study
team have the right to check this and to check if you are complying with this
agreement.
6. Any outputs of research involving PROTECT data including but not limited to
conference presentations, journal publications and posters must include
acknowledgement to funders of PROTECT and the PROTECT study team,
and make use of the PROTECT logo where fitting as detailed in Appendix 1:
Use of PROTECT-UK Data: Publication and Dissemination Guidelines.
7. Any output must adhere to any authorship agreements made between you
and the PROTECT study team (Appendix 1).
8. Once your Project has finished you must permanently delete the PROTECT
data.
9. The PROTECT data must not be used for any purpose contrary to any law or
regulation.
10. The PROTECT data must not be linked (wholly or partly) with any other data
and you must not attempt to identify any PROTECT participants.
11. Please note that the PROTECT data is provided “as is” with no guarantee that
it will be fit for your Project or that it will be accurate, complete or reliable.
12. If you notice any errors in the PROTECT data or need to notify the team of
anything, please inform the PROTECT study team at
support.protect@exeter.ac.uk.
379
Appendices

Please sign below to confirm your agreement to the above terms and to
confirm that you have overall responsibility for the Project as Principal
Investigator / Supervisor:

Name Professor Linda Clare Signed

_______________________ _________________________

Date 30 April 2020 Position Professor of Clinical Psychology

________________________ ________________________

380
Appendices

Use of PROTECT-UK Data: Publication and Dissemination Guidelines


This document outlines the information and author who should be included in any
publication, poster, presentation or other dissemination activity arising from the use
of PROTECT-UK data.

Authorship
Middle author placement is anticipated for key PROTECT investigators unless
otherwise agreed.

Please include the following authors:


Name: Dr Anne Corbett
Position: Senior Lecturer in Dementia Research
Institution: University of Exeter
Full Address: South Cloisters, College of Medicine & Health, St Luke’s Campus,
University of Exeter, Exeter EX1 2LU
Email address: a.m.j.corbett@exeter.ac.uk

Name: Professor Clive Ballard


Position: Pro-Vice Chancellor / Professor of Age-Related Diseases
Institution: University of Exeter
Full Address: Medical School Building, College of Medicine & Health, St Luke’s
Campus, University of Exeter, Exeter EX1 2LU
Email: c.ballard@exeter.ac.uk

Name: Dr Byron Creese


Position: Research Fellow
Institution: University of Exeter
Full Address: RILD Building Level 4, University of Exeter Medical School, RILD
Building, RD&E Hospital Wonford, Barrack Road, Exeter, EX2 5DW, UK
Email: b.creese@exeter.ac.uk

Name: Professor Dag Aarsland


Position: Head of Old Age Psychiatry
Institution: Institute of Psychiatry, Psychology and Neuroscience

381
Appendices

Full Address: Institute of Psychiatry, Psychology & Neuroscience, King's College


London, 16 De Crespigny Park, London SE5 8AF
Email: dag.aarsland@kcl.ac.uk

Dr Adam Hampshire
Senior Lecturer in Restorative Neurosciences
Faculty of Medicine
Department of Medicine
Imperial College London
London SW7 2AZ
Email: a.hampshire@imperial.ac.uk

Any analysis using polygenic scores or other genetic data must also include:

Dr Ryan Arathimos
Social, Genetic and Developmental Psychiatry Centre
Institute of Psychiatry, Psychology and Neuroscience
King’s College London
UK
Email: ryan.arathimos@kcl.ac.uk

Acknowledgement
Please use the following text in full in any publications, and where possible in other
dissemination materials:
“This paper represents independent research coordinated by the University of Exeter
and King’s College London and is funded in part by the National Institute for Health
Research (NIHR) Biomedical Research Centre at South London and Maudsley NHS
Foundation Trust and King’s College London. This research was also supported by
the National Institute for Health Research (NIHR) Collaboration for Leadership in
Applied Health Research and Care South West Peninsula and the National Institute
for Health Research (NIHR) Exeter Clinical Research Facility. The views expressed
are those of the author(s) and not necessarily those of the NHS, the NIHR or the
Department of Health.”

382
Appendices

Use of Logos
The following logos are available for use on posters, presentations and other
dissemination materials for your Project. We request that where possible all of these
are included:
• PROTECT study Logo
• University of Exeter (see here:
http://www.exeter.ac.uk/departments/communication/mark-
ops/design/downloads/)
• King’s College London

Genetic Data Use Acknowledgement


Please use the following text in any publications and where possible in other
dissemination materials if you have used PROTECT genetic data:
“This work was funded in part by the University of Exeter through the MRC Proximity
to Discovery: Industry Engagement Fund (External Collaboration, Innovation and
Entrepreneurism: Translational Medicine in Exeter 2 (EXCITEME2) ref.
MC_PC_17189). Genotyping was performed at deCODE Genetics.”

Family History of Neurological Disease Questionnaire Acknowledgement


Please use the following text in any publications and where possible in other
dissemination materials if you have used PROTECT Family History of Neurological
Disease Questionnaire data:
“This work was funded in part by the Alzheimer’s Research UK South West
Network.”

Schedule 1 Details of Data Set(s) Requested


1. Creation of data set for analyses (Baseline and Follow-Up data) to include the
following:
a. Lifestyle questionnaire
b. Fertility and menopause questionnaire

383
Appendices

Schedule 2: Project Description


This request is a follow-up request to PROT_UK_AM_002 for additional data
from PROTECT on lifestyle and menopause.
The study “Awareness of age-related change and its association with mental and
physical well-being in cognitively healthy adults living in the UK” will be a longitudinal
study exploring the concept of awareness of age-related change (AARC) and how it
is related to mental well-being, physical health and cognitive abilities. Data will be
collected at two time points since the study aims to explore the direction of the
associations.
Since assessments of mental well-being, physical health and cognition are part of
PROTECT, we request access to PROTECT data and to add extra questionnaires to
assess variables such as rumination and AARC that are not already assessed in
PROTECT. (These were added in application: PROT_UK_AM_002)
This request is a follow-up request to PROT_UK_AM_002 for additional data
from PROTECT on lifestyle and menopause.
§ to validate a measure of Awareness of age-related change (AARC): the
AARC-10 SF (Kaspar, Gabrian, Brothers, Wahl, Diehl, 2017) and the
cognitive functioning subscale of the AARC-50 (Brothers et al, 2018) in people
aged at least 50 years old and living in the UK;
§ to investigate the association between AARC (gains and losses) and physical
health by exploring indicators of physical limitations including current and past
physical health conditions, hearing and visual impairments, physical pain,
subjective health, instrumental activities of daily living and quality of sleep;
§ to investigate the association between AARC (gains and losses) and
symptoms of depression and anxiety and to understand the role of rumination
in the association between AARC and symptoms of depression and anxiety;
§ to understand the role of cognition in relation to AARC by exploring the
possible association between the AARC- cognitive subscale (gains and
losses) and objective cognitive performance. As part of this sub-study the
possible association between the AARC-cognitive subscale score and
cognitive abilities rated by an informant will be explored;
§ to explore whether age predicts changes in AARC (gains and losses) over
one year;

384
Appendices

§ to explore the association between attitudes toward own ageing (ATOA) and
cognition and to compare the associations found in PROTECT participants
with those found in a sample of people living with dementia;
§ to explore the associations between AARC and lifestyle (e.g. items on
smoking, alcohol, exercise, cognitive activities);
§ to explore the association between menopause and AARC.

385
Appendices

Appendix O: PRISMA 2009 Checklist


Reported
Section/topic # Checklist item on page
#
TITLE
Title 1 Identify the report as a systematic review, meta-analysis, or 1
both.
ABSTRACT
Structured 2 Provide a structured summary including, as applicable: 3
summary background; objectives; data sources; study eligibility criteria,
participants, and interventions; study appraisal and synthesis
methods; results; limitations; conclusions and implications of key
findings; systematic review registration number.
INTRODUCTION
Rationale 3 Describe the rationale for the review in the context of what is 5, 6, 7
already known.
Objectives 4 Provide an explicit statement of questions being addressed with 7
reference to participants, interventions, comparisons, outcomes,
and study design (PICOS).
METHODS
Protocol and 5 Indicate if a review protocol exists, if and where it can be 7
registration accessed (e.g. Web address), and, if available, provide
registration information including registration number.
Eligibility 6 Specify study characteristics (e.g. PICOS, length of follow-up) 8
criteria and report characteristics (e.g. years considered, language,
publication status) used as criteria for eligibility, giving rationale.
Information 7 Describe all information sources (e.g. databases with dates of 7, 8
sources coverage, contact with study authors to identify additional
studies) in the search and date last searched.
Search 8 Present full electronic search strategy for at least one database, 46
including any limits used, such that it could be repeated.
Study 9 State the process for selecting studies (i.e. screening, eligibility, 8, 9, 10
selection included in systematic review, and, if applicable, included in the
meta-analysis).
Data 10 Describe method of data extraction from reports (e.g. piloted 9, 10
collection forms, independently, in duplicate) and any processes for
process obtaining and confirming data from investigators.
Data items 11 List and define all variables for which data were sought (e.g. 9

386
Appendices

Reported
Section/topic # Checklist item on page
#
PICOS, funding sources) and any assumptions and
simplifications made.
Risk of bias in 12 Describe methods used for assessing risk of bias of individual 10
individual studies (including specification of whether this was done at the
studies study or outcome level), and how this information is to be used
in any data synthesis.
Summary 13 State the principal summary measures (e.g. risk ratio, difference 8
measures in means).
Synthesis of 14 Describe the methods of handling data and combining results of 10, 11
2
results studies, if done, including measures of consistency (e.g. I ) for
each meta-analysis.

Risk of bias 15 Specify any assessment of risk of bias that may affect the 11
across studies cumulative evidence (e.g. publication bias, selective reporting
within studies).
Additional 16 Describe methods of additional analyses (e.g. sensitivity or Not
analyses subgroup analyses, meta-regression), if done, indicating which applicable
were pre-specified.
RESULTS
Study 17 Give numbers of studies screened, assessed for eligibility, and 35
selection included in the review, with reasons for exclusions at each
stage, ideally with a flow diagram.
Study 18 For each study, present characteristics for which data were 9, 12, 13
characteristics extracted (e.g. study size, PICOS, follow-up period) and provide
the citations.
Risk of bias 19 Present data on risk of bias of each study and, if available, any 14-17, 47
within studies outcome level assessment (see item 12).
Results of 20 For all outcomes considered (benefits or harms), present, for 36-39
individual each study: (a) simple summary data for each intervention group
studies (b) effect estimates and confidence intervals, ideally with a forest
plot.
Synthesis of 21 Present results of each meta-analysis done, including 36-39
results confidence intervals and measures of consistency.
Risk of bias 22 Present results of any assessment of risk of bias across studies 47
across studies (see Item 15).

387
Appendices

Reported
Section/topic # Checklist item on page
#
Additional 23 Give results of additional analyses, if done (e.g. sensitivity or Not
analysis subgroup analyses, meta-regression [see Item 16]). applicable

DISCUSSION
Summary of 24 Summarize the main findings including the strength of evidence 17-24
evidence for each main outcome; consider their relevance to key groups
(e.g. healthcare providers, users, and policy makers).
Limitations 25 Discuss limitations at study and outcome level (e.g. risk of bias), 22-24
and at review-level (e.g. incomplete retrieval of identified
research, reporting bias).
Conclusions 26 Provide a general interpretation of the results in the context of 24
other evidence, and implications for future research.
FUNDING
Funding 27 Describe sources of funding for the systematic review and other 2
support (e.g. supply of data); role of funders for the systematic
review.

388

You might also like