Overview
The Employment Intervention Demonstration Program (EIDP) was a federally-funded multi-site randomized controlled trial (1995–2002 enrollment, funded through early 2000s) examining the effectiveness of supported employment programs for unemployed adults with severe mental illness in the United States. Funded by the Center for Mental Health Services (CMHS) of SAMHSA (Cooperative Agreement No. SM51820), the EIDP enrolled 1,273 participants at seven sites and followed them for two years. The coordinating center was the Department of Psychiatry, University of Illinois at Chicago, in collaboration with the Human Services Research Institute in Cambridge, MA. The program is the largest US RCT of supported employment and has generated more than a dozen published analyses.
Key Features
- Sample: 1,273 unemployed adults with severe and persistent mental illness; schizophrenia/schizoaffective 49%; major depression 21%; bipolar 16%; 53% male; median age 38; 72% receiving SSA disability income support; 34% without high school diploma
- Sites (7): Baltimore MD, Hartford CT, Worcester MA, Sumter SC, Portland ME, Phoenix AZ, San Antonio TX (Pennsylvania site excluded from primary analysis due to different target population)
- SE models tested: Individual Placement and Support (IPS) at Maryland/Connecticut/South Carolina; PACT vocational model at Massachusetts; Family-Aided ACT + Employer Consortium at Maine; Employment Assistance through Reciprocity in Natural Supports (EARNS) at Texas; integrated team model at Arizona
- Control conditions: Services as usual (Arizona, Connecticut, Maryland, South Carolina); "unenhanced" versions of experimental model (Texas, Maine); Clubhouse/transitional employment model (Massachusetts)
- Outcomes: Competitive employment (minimum wage, community setting, not reserved for disabled); total earnings; hours worked; job tenure; workplace disability disclosure
Key Results
- 55% experimental vs. 34% control achieved competitive employment over 24 months; effect increased over time
- Only 4% of SSDI beneficiaries (n=450) earned enough to complete trial work period and leave rolls
- SE remained effective even in high-unemployment-rate sites
- Physical comorbidities substantially reduce competitive employment (OR=0.48); substance abuse has null effect
- Job development is the most powerful individual service component (~5× competitive employment odds)
- Fidelity check: experimental participants received significantly more of all 10 vocational service types (p<0.001)
Related
Sources