See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/6702549

The Psychotic Symptom Rating Scales (PSYRATS): Their usefulness and

properties in first episode psychosis

Article in Schizophrenia Research · February 2007

DOI: 10.1016/j.schres.2006.04.024 · Source: PubMed

CITATIONS READS

182 26,850

5 authors, including:

Richard James Drake Gillian Haddock

The University of Manchester The University of Manchester
225 PUBLICATIONS 9,666 CITATIONS 233 PUBLICATIONS 11,948 CITATIONS

SEE PROFILE SEE PROFILE

Richard Bentall Shôn Lewis

The University of Sheffield The University of Manchester
513 PUBLICATIONS 37,271 CITATIONS 473 PUBLICATIONS 29,220 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Richard James Drake on 15 June 2018.

The user has requested enhancement of the downloaded file.

 Schizophrenia Research 89 (2007) 119 – 122
www.elsevier.com/locate/schres

The Psychotic Symptom Rating Scales (PSYRATS): Their usefulness

and properties in first episode psychosis
Richard Drake a,⁎, Gillian Haddock b , Nicholas Tarrier c , Richard Bentall c , Shôn Lewis a
a
Division of Psychiatry, University of Manchester, 2nd Floor Education and Research Centre, Wythenshawe Hospital,
Manchester, M23 9PL, United Kingdom
b
School of Psychological Sciences, University of Manchester, Rutherford House, Manchester Science Park, Lloyd Street North,
Manchester, M15 6SZ, United Kingdom
c
School of Psychological Sciences, University of Manchester, Oxford Road, Manchester, M15 9PL, United Kingdom

Received 8 April 2006; received in revised form 14 April 2006; accepted 25 April 2006
Available online 13 November 2006

Abstract

The aim of this study was to investigate the reliability, validity and structure of the Psychotic Symptom Rating Scales
(PSYRATS) in 257 subjects presenting with acute first episodes of schizophrenia or related disorders. The PSYRATS have been
shown to assess dimensions of hallucination and delusions reliably and validly in chronically psychotic patients but not in first
episode patients. Item reliability was investigated and subscale performance compared to the PANSS. The PSYRATS had good
inter-rater and retest reliability. Validity was good, as assessed by internal consistency, sensitivity to change, and in relation to the
PANSS. There was evidence of two delusion factors and three for hallucinations. The scales are useful complements to existing
measures of symptom severity.
© 2006 Elsevier B.V. All rights reserved.

Keywords: First episode; Schizophrenia; Delusions; PSYRATS

1. Introduction sis and little of this work has been done with recent onset
psychosis. Watching the development of symptoms
Only a few structured assessments and interviews from first presentation is likely to be particularly
which assess dimensions of psychotic symptoms have informative since these symptoms are close to their
been reported (e.g. Oulis et al., 1995; Wessely et al., origin and some will resolve never to return.
1993). Often little attention was paid to their psycho- The Psychotic Symptom Rating Scales (Haddock et
metric properties and the investigations have mainly al., 1999) are semi-structured interviews designed to
been focused on patients experiencing chronic psycho- assess the subjective characteristics of hallucinations
and delusions (see Appendix). The auditory hallucina-
⁎ Corresponding author. Tel.: +44 161 291 5888; fax: +44 161 292 tions subscale (AHS) has 11 items: for frequency,
5882. duration, controllability, loudness, location; severity and
E-mail addresses: rdrake@manchester.ac.uk (R. Drake), intensity of distress; amount and degree of negative
gillian.haddock@manchester.ac.uk (G. Haddock),
N.Tarrier@manchester.ac.uk (N. Tarrier),
content; beliefs about the origin of voices; and
richard.p.bentall@manchester.ac.uk (R. Bentall), disruption. The delusions subscale (DS) has six items:
shon.lewis@manchester.ac.uk (S. Lewis). duration and frequency of preoccupation; intensity of
0920-9964/$ - see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.schres.2006.04.024
120 R. Drake et al. / Schizophrenia Research 89 (2007) 119–122

distress; amount of distressing content; conviction and affective psychoses and scored 4 or more on the PANSS
disruption. The scales had excellent inter-rater reliability hallucinations or delusions items. Nine were male and 4
and good validity in sufferers from chronic schizophre- female. The psychiatrists rated each of the 13 patients
nia (Haddock et al., 1999). The aim of the current study either simultaneously or on watching an interview
was to examine the psychometric properties of the video-taped by one of them.
PSYRATS in first episode, acutely psychotic patients.
3. Results
2. Method
3.1. Sample
2.1. Sample
Two hundred fifty-seven first episode sufferers
Patients were recruited from consecutive admissions were recruited. One hundred nineteen (77%) were
to acute inpatient and day-patient facilities from followed up at six weeks (Lewis et al., 2002). The first
geographically defined areas in the north of England 103 first episode patients recruited to the study who
over 26 months. They were recruited as part of a trial to had full PSYRATS data at five and six weeks were
assess the effectiveness of CBT for recent onset used in the test–retest analysis. Differences from the
psychosis (Lewis et al., 2002; Tarrier et al., 2004). overall sample in demographics or symptoms were
Inclusion criteria were: DSM-IV (APA, 1994) schizo- small and non-significant but fewer were substance
phreniform disorder, schizophrenia, schizoaffective dependent at baseline (10% compared to 17%, chi sq.
disorder, delusional disorder or psychosis not otherwise p < 0.01).
specified; and age 16–65. Exclusion criteria were:
organic brain disorder; psychosis purely due to substance 3.2. Inter-rater reliability
use; and little fluency in English. A sub-sample, (see
Section 3.1), was used to assess the retest reliability. Average Intra-Class Correlations (ICCs; two way
random effects models; Bartko and Carpenter, 1976)
2.2. Procedure between raters for subscales and totals were excellent
(DS 0.99 to 1.00, AH 0.99 to 1.00, total PSYRATS
2.2.1. Validity 0.99 to 1.00). ICCs for DS items ranged from 0.58 to
Patients were recruited and assessed within 14 1.00, apart from ‘conviction in delusions' which had an
working days of and re-interviewed 6 weeks later ICC of 0.24. For AHS items ICCs ranged from 0.74 to
(Lewis et al., 2002) using the PANSS (Kay et al., 1989) 1.00 apart from ‘location of voices’ which had an ICC
and PSYRATS (Haddock et al., 1999) by three of 0.42 between 2 of the raters. Bland–Altman plots
psychiatrists trained by GH and SL. The PANSS was (Bland and Altman, 1986) were good for all pairings of
chosen to assess validity of the PSYRATS even though raters on each subscale and the total: the largest
the scales differ in structure and content. However, it discrepancy was 13% for 2 raters on the PSYRATS
was thought that the PANSS was the closest and most total.
widely used scale with which to make comparisons to
judge validity. 3.3. Test–retest reliability

2.2.2. Test–retest reliability Data for weeks 5 and 6 were compared. ICCs were:
A sub-group of trial participants (see Section 3.1) DS 0.70; AH 0.70. Individual items had ICCs of 0.50 to
were interviewed weekly for the first six weeks. 0.74 (DS items) and 0.55 to 0.74 (AH items) over the
PSYRATS scores for the 5th and 6th week interviews same period. This is despite changes in AH (median
were compared to establish test–retest reliability, since − 25%) and DS (median − 14%) over the time,
symptoms between these points were relatively stable. presumably due to treatment.

2.2.3. Inter-rater reliability 3.4. Internal consistency

Thirteen patients were selected opportunistically and
interviewed using the PANSS and PSYRATS. Four For the DS items, each correlated between 0.17 and
were from the main trial and nine other patients were 0.41 with the subscale score minus that item (Kendall's
attending for treatment at South Manchester University tau-b). For the AHS, each item correlated between 0.63
Hospitals NHS Trust. All met DSM IV criteria for non- and 0.76 with the total minus that item (Kendall's tau-b)
 R. Drake et al. / Schizophrenia Research 89 (2007) 119–122 121

apart from ‘control over hallucinations’ (Kendall's tau-b consistent, perhaps as the sample was small. The test–
0.16). retest reliability was high despite the fact that the
sample improved between the two time points
3.5. Concurrent validity assessed (probably due to their progression through
treatment), deflating the correlations between item
The DS significantly correlated (Spearman's) with scores.
the PANSS delusion item (0.43), positive subscale The validation and scale item analysis used a large,
(0.20) and total score (0.18). The AH significantly relatively representative first admission sample, fol-
correlated (Spearman's) with the PANSS hallucination lowed-up longitudinally with limited attrition. The
item (0.81), positive subscale (0.31) and PANSS total PSYRATS was compared to the PANSS. Change scores
(0.26). for both PSYRATS subscales correlated well with those
for corresponding PANSS items, suggesting that the
3.6. Sensitivity to change in relation to the PANSS PSYRATS assess psychotic symptoms validly. The
AHS correlated particularly strongly at baseline with
Sensitivity to change was examined in the DS and PANSS hallucinations.
AH over the 6 weeks from inclusion in the trial. Again Both MDS and factor analysis were used to explore
these resembled the overall sample at baseline (Lewis the structure of the PSYRATS subscales. MDS is a more
et al., 2002). For each parameter six-week change the valid method than factor analysis since the items are
score was calculated as: change divided by the baseline ordinal. However, although both produce similar results,
score minus the minimum possible score. Change in the criteria for selecting solutions on PAF are better
the DS significantly correlated (Spearman's) with the recognised and factor analysis enhances comparison
change in the PANSS delusions item (0.80), the with the smaller, chronic sample in Haddock et al.
positive subscale score (0.75) and in the PANSS total (1999). The factor structure observed in both samples
(0.69). Only the 124 patients experiencing auditory was similar.
hallucinations at baseline were included in the AH The PSYRATS have good reliability and validity in
analysis. Change in the AH significantly correlated first episode samples and complement existing measures
(Spearman's) with change in the PANSS hallucination of outcome like the PANSS. They offer a more detailed
item (0.88), the positive subscale (0.63) and in PANSS assessment of symptom dimensions, offering research-
total (0.54). ers and clinicians a better understanding of these key
psychotic symptoms and their changes in response to
3.7. Structure of the subscales treatment. Exploration of the structure of these symp-
toms will aid development of models of aetiology and
The DS and AHS at baseline were examined with intervention.
SPSS 10.0 (SPSS, 2001) using factor analysis (Principal
Axis Factoring, oblique rotation; Tabachnick and Fidell, Acknowledgements
1996) and multi-dimensional scaling (MDS; Euclidean
matrices, ALSCAL subroutine). The DS frequency, The authors acknowledge the work of the members
duration, conviction and disruption items loaded onto of the SOCRATES team and the support of the Medical
one factor and amount and intensity of distress onto Research Council and Stanley Medical Research
another. The AHS distress and negative content items Institute (for Dr. Drake). We also thank Professor
loaded onto one factor; frequency and duration onto a Graham Dunn for his assistance.
second; and belief about voices' reality, location, control
and disruption onto a third. MDS of each subscale gave Appendix A. The PSYRATS
2 dimensions as the most parsimonious solutions of
acceptable fit (Everritt and Dunn, 2001). There were Scale detailed in Haddock et al. (1999).
distinct groupings of items matching the factors found All items are scored 0–4, according to general criteria:
on PAF.
0 No problem
4. Discussion 1 Minimal or occasional
2 Minor to moderate
The inter-rater reliability of the PSYRATS sub- 3 Major
scales was very good but that of the items was less 4 Maximum severity
122 R. Drake et al. / Schizophrenia Research 89 (2007) 119–122

Each item is scored according to detailed anchor References

points, e.g.:
Amount Of Preoccupation With Delusions. American Psychiatric Association, 1994. The Diagnostic and
Statistical Manual of Mental disorders, 4th edition revised.
American Psychiatric Association, Washington.
0 No delusions, or delusions which the subject thinks Bartko, J.J., Carpenter, W.T., 1976. On the methods and theory of
about less than once a week reliability. J. of Nerv. Ment. Dis. 163, 307–317.
1 Subject thinks about beliefs at least once a week. Bland, J.M., Altman, D.G., 1986. Statistical methods for assessing
2 Subject thinks about beliefs at least once a day. agreement between two methods of clinical measurement. Lancet i,
3 Subject thinks about beliefs at least once an hour. 307–310.
Everritt, B.S., Dunn, G., 2001. Applied Multivariate Data Analysis,
4 Subject thinks about delusions continuously or 2nd ed. Arnold, London.
almost continuously. Haddock, G., McCarron, J., Tarrier, N., Faragher, E.B., 1999. Scales to
Subject can only think about other things for a few measure dimensions of hallucinations and delusions: the Psychotic
seconds or minutes. Symptom Rating Scales (PSYRATS). Psychol. Med. 29, 879–889.
Kay, S.R., Opler, L.A., Lindenmayer, J.P., 1989. The positive and
negative syndrome scale (PANSS): Rationale and standardisation.
Scale Items: Delusion Subscale Br. J. Psychiatry 155 (suppl 7) 59–65.
Lewis, S.W., Tarrier, N., Haddock, G., Bentall, R.P., Kinderman, P.K.,
Item 1: Amount Of Preoccupation With Delusions Kingdon, D., Siddle, R., Drake, R.J., Everritt, J., Benn, A.,
Item 2: Duration of Preoccupation with Delusions Leadley, K., Grazebrook, K., Haley, C.J., Ahktar, S., Faragher,
Item 3: Conviction E.B., Dunn, G., 2002. A randomised, controlled trial of
cognitive-behaviour therapy in acute, early schizophrenia: the
Item 4: Amount of Distress SOCRATES trial. Br. J. Psychiatry 181 (Suppl. 43) 91–98.
Item 5: Intensity of Distress Oulis, P.G., Mavreas, V.G., Mamounas, J.M., Stefanis, C.N., 1995.
Item 6: Disruption to Life Caused by Beliefs The clinical characteristics of auditory hallucinations. Acta
Psychiatr. Scand. 92, 97–102.
SPSS for Windows, Rel. 10.0.1: 2001. Chicago, SPSS Inc.
Auditory Hallucinations Subscale
Tabachnick, B.G., Fidell, L.S., 1996. Using Multivariate Statistics, 4th
ed. Harper and Row, New York.
Item 1: Frequency Tarrier, N., Lewis, S.W., Haddock, G., Bentall, R.P., Kinderman, P.,
Item 2: Duration Kingdon, D., Siddle, R., Drake, R.J., Everritt, J., Leadley, K.,
Item 3: Location Benn, A., Grazebrook, K., Haley, C.J., Akhtar, S., Dunn, G., 2004.
Item 4: Loudness 18 month follow-up of a randomised, controlled trial of cognitive-
behaviour therapy in first episode and early schizophrenia. Br. J.
Item 5: Beliefs Re Origin of Voices Psychiatry 184, 231–239.
Item 6: Amount of Negative Content of Voices Wessely, S., Buchanan, A., Reed, A., Cutting, J., Everitt, P., Garety, P.,
Item 7: Degree of Negative Content Taylor, P.J., 1993. Acting on delusions: I prevalence. Br. J.
Item 8: Amount of Distress Psychiatry 163, 69–76.
Item 9: Intensity of Distress
Item 10: Disruption to Life Caused by Voices
Item 11: Controllability of Voices

View publication stats