In order to determine whether expenditures for mental health could be reduced and quality improved, Congress mandated that the Department of Defense conduct a demonstration project utilizing a wraparound mental health service system for child and adolescent military dependents. A longitudinal quasiexperimental design was used to evaluate the cost-effectiveness of the demonstration. The results showed that children in the Wraparound Group received more wraparound services than those in the treatment as usual (TAU) Comparison Group. These services included case management, in-home treatment, and other nontraditional services. The Demonstration also provided better continuity of care. Multiple methods were used to investigate the impact of wraparound. Both groups showed some improvement on some measures but there were no differences between the groups in functioning, symptoms, life satisfaction, positive functioning, or sentinel events. Regardless of which statistical model was used to estimate costs, the Demonstration was also more expensive. The higher level of expenditures for the Wraparound group was a result of some expensive traditional care and the addition of nontra-ditional services. Several possible explanations of these results are provided.