Cook, J., Leff, H., Blyler, C., Gold, P., Goldberg, R., Mueser, K., Toprac, M., McFarlane, W., Shafer, M., Blankertz, L., Dudek, K., Razzano, L., Grey, D., & Burke-Miller, J. (2005). Results of a multisite randomized trial of supported employment interventions for individuals with severe mental illness. Archives of General Psychiatry, 62, 505-512.
- The study’s objective was to estimate the 24-month employment effects of supported employment on people with severe mental illness. Supported employment interventions use a combination of employment and health services and supports to improve employment and other outcomes.
- The authors analyzed data from seven sites, each of which implemented a distinct supported employment intervention using a randomized controlled trial. At each site, researchers interviewed participants in person twice yearly and collected weekly employment data.
- The supported employment interventions had a significant and positive impact on employment, working for 40 or more hours in a single month, and monthly earnings from paid employment.
- The quality of causal evidence provided in this study is moderate. This means we are somewhat confident that the estimated effects are attributable to the supported employment interventions, but other factors might also have contributed.
Supported Employment Programs
Features of the Intervention
Supported employment interventions use a combination of employment and health services and supports to improve employment and other outcomes for people with psychiatric conditions. The supported employment programs in this study shared five core features:
- Integrated services delivered by a multidisciplinary team that met at least three times per week to plan and coordinate employment interventions, psychiatric treatment, and case management
- Placement into competitive employment
- Jobs tailored to participants’ career preferences
- Immediate job search beginning at program entry
- Ongoing vocational support
Features of the Study
The study team recruited volunteers for the study. Candidates who responded to outreach efforts—self-referral, family referral, word of mouth, and advertisements—were considered eligible if they had a qualifying diagnosis, had exhibited symptoms for a sufficient period, were age 18 or older, had expressed willingness to work, and had provided written informed consent. A total of 1,648 eligible participants were randomly assigned to the treatment group, which received supported employment services, or the control group, which received a comparison condition that varied by site (exact breakdowns by treatment group are unknown). The analysis sample consisted of 1,273 study participants (648 treatment and 625 control).
Researchers interviewed all participants at baseline and twice yearly thereafter, in addition to collecting weekly employment records. They computed impacts on competitive employment—which the authors defined as a position in a privately held, socially integrated setting not reserved for people with disabilities that paid at least minimum wage—working 40 or more hours a month, and monthly earnings at 24 months after random assignment.
The authors analyzed supported employment interventions in seven states:
- South Carolina
- The study found that the supported employment interventions had a significant and positive effect on the probability of being competitively employed, working for 40 or more hours in a single month, and monthly earnings from paid employment.
- Specifically, higher proportions of treatment group members secured competitive employment (55 versus 34 percent) and worked 40 or more hours in a given month during the 24-month study period (51 versus 39 percent). Treatment group members also had higher earnings than control group members over the follow-up period ($122 per month versus $99 per month).
Considerations for Interpreting the Findings
The study had high attrition. Hence, it was not eligible for a high evidence rating. However, the authors controlled for many relevant observable characteristics of the two groups.
The authors noted that the study sample was not nationally representative, which limits the results’ generalizability. In addition, the supported employment programs being evaluated differed across participating study sites, as did the services available to control group members.
Causal Evidence Rating
The quality of causal evidence provided in this study is moderate because it is a randomized controlled trial with high attrition but controls for many relevant baseline characteristics of the treatment and control groups. This means we are somewhat confident that the estimated effects are attributable to supported employment, but other factors might also have contributed.