Background. To help mitigate empirically-identified shortcomings in Supported Employment (SE) outcomes for consumers with severe mental illness in Indiana (e.g., few long-term placements), Vocational Rehabilitation Services (VRS) and the Indiana Division of Mental Health and Addictions (DMHA) proposed the implementation of a results-based funding (RBF) system, in which providers received payment only when clients successfully attained each of 5 employment milestones. The current study presents the results of a multi-site controlled trial of RBF vs. a traditional fee-for-service (FFS) model.Method. The sample included 122 (RBF=81, FFS=41) consumers of SE (52% schizophrenia, 35% mood disorder). Assessments of clinical and life outcomes were conducted quarterly across 12 months.Results. Those served in RBF were more likely to attain all milestones collectively, and to have a completed person-centered plan and attain 9-months of employment, specifically. There were few differences between those in RBF and FFS on non-milestone employment variables (e.g., job match, wages) or clinical measures (e.g., quality of life, functioning).Conclusions. RBF produces better overall vocational outcomes, specifically, higher rates of completion of a person-centered plan and retention of employment for 9 months. However, improvements with RBF were limited to those specified and did not generalize to vocational areas not targeted by the milestones (e.g., wages, benefits) or to changes in other life domains typically valued by psychiatric rehabilitation. [ABSTRACT FROM AUTHOR]