The use of propensity scores to evaluate outcomes for community clinics: identification of an exceptional home-based program.

This article presents a model for mental health programs to estimate causal effects of treatment in community settings, where experimental studies, in which subjects are randomly assigned to treatment and control groups, are not feasible. This article describes an observational study that used a propensity score analysis with stratification and a repeated-measures analysis of covariance model to estimate treatment effects. This article includes results from one example site that was identified as having an exceptional home-based community program. The results include treatment effects for 3 outcomes identified as useful goals for home-based community programs. The study also serves as a model of how local programs can establish credibility where no evidence-based treatments exist for severely impaired youths.