A menu with prices: Annual per person costs of programs addressing community integration.

Information on costs of programs addressing community integration for persons with serious mental illness in the United States, essential for program planning and evaluation, is largely lacking. To address this knowledge gap, community integration programs identified through directories and snowball sampling were sent an online survey addressing program costs and organizational attributes. 64 Responses were received for which annual per person costs (APPC) could be computed. Programs were categorized by type of services provided. Program types differed in median APPCs, though median APPCs identified were consistent with the ranges identified in the limited literature available. Multiple regression was used to identify organizational variables underlying APPCs such as psychosocial rehabilitation program type, provision of EBPs, number of volunteers, and percentage of budget spent on direct care staff, though effects sizes were moderate at best. This study adds tentative prices to the menu of community integration programs, and the implications of these findings for choosing, designing and evaluating programs addressing community integration are discussed.

The alliance in psychiatric rehabilitation: Client characteristics associated with the initial alliance in a supported employment program.

BACKGROUND: The therapeutic alliance is associated with outcomes for individuals in psychiatric rehabilitation programs. However, there are no known studies examining factors associated with the alliance in these programs. OBJECTIVE: To determine whether client characteristics associated with the alliance across multiple populations and program types were associated with the alliance for individuals enrolling in a supported employment program. METHODS: Previously collected data were available for 70 individuals served by the single employment specialist in the program. Potential independent variables and the strength of their relationship to the alliance were identified, then further explored using a forward stepwise logistic regression model. RESULTS: Seventy-three percent of the sample had the highest possible alliance rating. Being older and having a smaller social network size were significantly correlated with a high alliance, but neither remained a predictor of the alliance, controlling for the effects of the other. CONCLUSIONS: Client age and social network size were associated with the initial alliance, though not in a model containing both variables. Low power may have hindered the ability to show additional relationships, and the uniqueness of the program structure limits the generalizability to other programs. Future research should investigate the variables examined further.