SourceStudy designAims of studyOutcome measuresResults
Chen et al. (1991)

Country: UK

Evidence level: 2-
Case control.

Setting: Nottingham.

Population: consecutive series of N=40 Afro-Caribbean and N=40 non-Caribbean service users matched for age sex and diagnosis – schizophrenia.
To investigate whether there are any differences between Afro-Caribbean and non-Caribbean service users with regard to:

prescribed dosages of neuroleptic medication during the acute phase of the illness


the proportion commenced on depot maintenance therapy after a first episode of psychosis


the proportion admitted to hospital, the duration of admission and the use of compulsory procedures


the degree of behavioural disturbance in the acute stage of illness.

Information obtained from medical records.
  • Differences in receiving neuroleptic medication.
  • Relationship of medication to behavioural disturbance.
  • Episodes of disturbance (violence and bizarre behaviour).
  • Compulsory detention.
  • More Afro-Caribbeans than non-Caribbeans were treated with dosages above 2000mg of chloropromazine equivalent {χ2=4.98 p<0.03] (not adjusted for confounding e.g. diagnosis).
  • 33% (N=24) disturbed Afro-Caribbean service users received a maximum dosage of above 2000mg whereas none of the 12 disturbed non-Caribbean service users received this dosage [p<0.03, Fishers's exact probability test].
  • 24 Afro-Caribbeans and 12 non-Caribbeans had one or more disturbed episodes [χ2=6.1 p<0.01].
  • 15 Afro-Caribbeans and five non-Caribbeans had two or more episodes [χ2=5.4 p=<0.02]. Time period for episodes not stated.
  • 20 Afro-Caribbean service users were detained under the Mental Health Act during admission compared with nine non-Caribbeans [χ2=5.4. p<0.02].
Reviewer's comments
  • This appeared to be a matched case control study, although it was not described as such.
  • Data was obtained from chart records.
  • Prescribing practice is not directly affected by simple ethnic stereotypes. However, some findings provide limited evidence to support concern about over prescribing for some Afro-Caribbean service users.
Choca et al. 1990

Country: US

Evidence level: 2-
Retrospective chart review.

Setting: psychiatric ward, veterans' administration centre.

Population: black N=235 and white N=471 male service users admitted over seven years.
To examine how successful the Millon clinical multiaxial inventory (MCMI) test was in producing a culturally fair test.Clinical fairness of diagnostic tool MCMI to predict psychopathology. (This personality instrument has weighted scores to provide different norms for black, white and Hispanic individuals to address potential racial bias).

Items on MCMI were compared before and after grouping into matched pairs (N=209) of diagnostic categories according to DSM-111(The diagnostic and statistical manual of mental disorders (3rd ed.) of the American Psychiatric Association) on discharge.
(Seven groupings: substance abuse, anxiety, affective, character, psychotic, organic and ‘other’ disorders).
  • Operating characteristics of white and black subjects for the different diagnostic categories (see table).
  • Results of matched pairs, 45 items on the MCMI showed significant difference as opposed to an expected nine items, if the difference between the two groups was due to chance.
  • Multivarate analysis of the 20 scales including antisocial (<.01) and passive-aggressive/explosive (ns) showed blacks scoring significantly differently from whites on nine scales (p<.05).
  • To evaluate the structure of the test factor analysis was conducted - three factor structure measuring a) maladjustment b) extroverted acting out c) psychosis.
  • The factor congruence was .98, .98, .97 respectively indicating high similarity. The factor analysis indicates that the test is measuring similar factors in both groups. Therefore support is maintained for continued use of the test for blacks, with some adjustment at item and scale level. Confounding and potential limitations are discussed thoroughly.
Reviewer's comments
  • Forms completed over seven-year period and may vary greatly in completion of data.
  • Veterans are a specific group of individuals with military experience and therefore are a unique population.

From: Appendix 1, Ethnicity review evidence tables - included studies

Cover of The Experiences of Black Service Users in UK In-Patient Settings in Relation to the Short-Term Management of Disturbed/Violent Behaviour (Supplement)
The Experiences of Black Service Users in UK In-Patient Settings in Relation to the Short-Term Management of Disturbed/Violent Behaviour (Supplement).
NICE Clinical Guidelines, No. 25S.
Nelstrop L, Chandler-Oatts J, Cowl J, et al.
Copyright © 2006, Royal College of Nursing.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means electronic,mechanical, photocopying, recording or otherwise,without prior permission of the Publishers or a licence permitting restricted copying issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1T 4LP. This publication may not be lent, resold, hired out or otherwise disposed of by ways of trade in any form of binding or cover other than that in which it is published,without the prior consent of the Publishers.

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.