Intended for healthcare professionals


Helping people with severe mental illness to obtain work: systematic review

BMJ 2001; 322 doi: (Published 27 January 2001) Cite this as: BMJ 2001;322:204
  1. Ruth E Crowther, research fellowa,
  2. Max Marshall (mmarshall{at}, reader in community psychiatrya,
  3. Gary R Bond, Chancellor's professorb,
  4. Peter Huxley, professor of social workc
  1. a School of Psychiatry and Behavioural Sciences, University of Manchester, Manchester PR2 9HT
  2. b Department of Psychology, Indiana University-Purdue University Indianapolis, IN 46202-3275, USA
  3. c Health Services Research, King's College Institute of Psychiatry, London SE5 8AF
  1. Correspondence to: M Marshall, Academic Unit, Royal Preston Hospital, Preston PR2 9HT
  • Accepted 20 October 2000


Objective: To determine the most effective way of helping people with severe mental illness to obtain competitive employment—that is, a job paid at the market rate, and for which anyone can apply.

Design: Systematic review.

Participants: Eligible studies were randomised controlled trials comparing prevocational training or supported employment (for people with severe mental illness) with each other or with standard community care.

Outcome measures: The primary outcome was number of subjects in competitive employment. Secondary outcomes were other employment outcomes, clinical outcomes, and costs.

Results: Eleven trials met the inclusion criteria. Five (1204 subjects) compared prevocational training with standard community care, one (256 subjects) compared supported employment with standard community care, and five (484 subjects) compared supported employment with prevocational training. Subjects in supported employment were more likely to be in competitive employment than those who received prevocational training at 4, 6, 9, 12, 15, and 18 months (for example, 34% v 12% at 12 months; number needed to treat 4.45, 95% confidence interval 3.37 to 6.59). This effect was still present, although at a reduced level, after a sensitivity analysis that retained only the highest quality trials (31% v 12%; 5.3, 3.6 to 10.4). People in supported employment earned more and worked more hours per month than those who had had prevocational training.

Conclusion: Supported employment is more effective than prevocational training at helping people with severe mental illness obtain competitive employment.


  • Funding The research was supported by the NHS Health Technology Assessment Program (grant number 96/41/3). The views expressed in this paper are not necessarily those of this programme.

  • Competing interests GRB has a close collaborative relationship with Bob Drake and Debbie Becker, developers of the individual placement and support model.

  • Embedded ImageDetails of the trials are available on the BMJ's website

  • Accepted 20 October 2000
View Full Text