Pitfalls in Evaluating the Effectiveness of Case Management Programs for Homeless Persons

It is generally believed that homeless individuals—particularly those with substance abuse problems or mental illness—are unable to access the full range of available benefits and com munity services on their own. In recent years, community service providers have increasingly looked toward case management as the intervention of choice for solving this problem. Yet the evaluation findings of the National Institute on Alcohol Abuse and Alcoholism Community Demonstration Program, which included three case management projects, showed few clear indications of case management effectiveness. The results of these studies—which focused on homeless clients with alcohol and other drug problems—are consistent with studies of case management services for the homeless mentally ill. This article identifies several phenom ena that potentially explain the apparent lack of positive effects and suggests that the reasons for negative findings may lie as much with the evaluations as with the interventions. These phenomena include bias due to differential attrition from measurement, lack of intervention intensity, lack of distinction between groups, contextual factors, and additional design and measurement issues. Suggestions for improving future evaluations of case management effec tiveness are offered.