Mental health services outcome evaluation

There is an urgent need for pertinent outcome information. Relevance for decision makers must take priority over scientific rigor. However, a review of computer-identified outcome evaluation reports from community service settings, during the past 5 years, suggests that much more has been said than has been done. Although relatively heterogeneous in scope, these studies focused on the effects of community support services for adults with persistent and severe mental illness; traditional outpatient services have been neglected. Studies are characterized by multidimensional, standardized outcome assessment, and nonequivalent comparison group and single cohort designs. Randomized designs, with usual services as the control condition, were feasible in some situations. Inadequate sample size and attrition continue to be method problems.