Methodological issues in evaluating mental health outcomes of a children's mental health managed care demonstration.

A key question of the Fort Bragg Evaluation was whether the continuum of care model implemented at the Demonstration produced better mental health outcomes for children and adolescents than a traditional system of care. This article describes a few of the key methodological issues that were addressed to help ensure that findings about mental health outcomes were valid. The research design was quasi-experimental and longitudinal. Nearly 1,000 families participated, and attrition was low throughout the study. Multiple informants responded to a comprehensive array of widely used instruments in the area of children's mental health and mental health services. Despite the absence of random assignment, sites were comparable upon entry to the study on numerous factors that might affect outcome. Further, data lost through the course of study did not vary appreciably across sites. Finally, some evidence suggests that the findings of the Outcome Study may generalize to other populations of low-middle to middle-class youth in treatment.