Evaluating Mental Health Capitation Treatment

The paper evaluates a capitation-financed system of mental health services delivery developed in Rochester, New York. Cost/benefit analysis of the treatment program is implemented on three years of data using program evaluation techniques. Patient outcomes are compared across randomly assigned study groups as well as across enrollment status. The analysis implements difference-in-difference econometric techniques recently developed in the labor economics literature to control for potentially non-random attrition as well as selective non-compliance. We find that patients enrolled in the capitation program do experience significantly lower costs without becoming sicker, even after controlling for attrition and sample selection.