How does single stream-funded, capitated payment for mental health services affect provider behavior? This study evaluated Baltimore's experiment to provide capitated mental health services to long term state hospital inpatients being discharged to community care and Medicaid high cost users. The evaluation included two components: 1) a prospective controlled study of Medicaid high cost users, comparing two samples of patients randomly assigned to the capitation program or to traditional service systems; and 2) a strategic analysis of the capitation programs, comparing their methods with those used by traditional service providers. The objective of the study was to develop a good understanding of how mental health provider performance is affected by program incentives and of how these providers adapt to a single funding stream, in terms of flexibility and innovation in service delivery.