Self-directed care allows individuals with disabilities and elderly persons to control public funds to purchase goods and services that help them remain outside institutional settings. This study examined effects on outcomes, service costs, and user satisfaction among adults with serious mental illness. Methods: Public mental health system clients were randomly assigned to self-directed care (N=114) versus services as usual (N=102) and assessed at baseline and 12 and 24 months. The primary outcome was self-perceived recovery. Secondary outcomes included psychosocial status, psychiatric symptom severity, and behavioral rehabilitation indicators. Mixedeffects random-regression analysis tested for longitudinal changes in outcomes between the two conditions. Differences in service costs were analyzed with negative binomial and zero-inflated negative binomial regression models. Results: Compared with the control group, self-directed care participants had significantly greater improvement over time in recovery, self-esteem, coping mastery, autonomy support, somatic symptoms, employment, and education. No betweengroup differences were found in total per-person service costs in years 1 and 2 or both years combined. However, selfdirected care participants were more likely than control group participants to have zero costs for six of 12 individual services and to have lower costs for four. The most frequent nontraditional purchases were for transportation (21%), communication (17%), medical care (15%), residential (14%), and health and wellness needs (11%). Client satisfaction with mental health services was significantly higher among intervention participants, compared with control participants, at both follow-ups. Conclusions: The budget-neutral self-directed care model achieved superior client outcomes and greater satisfaction with mental health care, compared with services as usual.

Lessons learned about the importance of community collaboration are discussed in light of the current emphasis on public mental health system transformation through alternative financing mechanisms.

This article describes the use of evidence-based practice along with a multi-stakeholder consensus process to design the psychosocial rehabilitation components in a benefit package of publicly funded mental health services in Texas. The Texas Benefit Design initiative demonstrates how the combination of science and consensus can be used as a powerful tool for change. It applies the findings of rigorous research regarding psychosocial rehabilitation service delivery approaches that achieve positive outcomes in real world, community settings. At the same time, it makes use of the unique knowledge and experience that mental health service consumers, providers and other advocates can bring to service system design and planning.