Contracting for Outpatient Mental Health Services: Effects on Services Use, Quality, and Costs

Background: During the fiscal year 2004-2005, the County of San Diego engaged in a contracting process whereby non-profit mental health (MH) providers were allowed to bid on the outpatient services provided by five County owned and operated mental health clinics. As a result of this process, the services of two outpatient clinics were contracted to non-profit providers; three clinics remained operated by the County, although existing as reengineered organizations under revised budgets. This study describes changes in service utilization and costs among five outpatient MH programs involved in the contracting process. Methods: Analyses were performed at the person level, using generalized linear models embedded in a quasi-experimental difference-in-difference design. Twelve non-profit providers not participating in the contracting process served as a comparison group. Service utilization data was obtained from an encounter-based management information system; cost data were determined from a detailed examination of cost reports. Service hours, costs, and the probability of inpatient or emergency service admission were compared pre and post contracting. Multivariate models were used to adjust for a number of socio-demographic and clinical characteristics expected to affect service use. Results: Difference-in-difference estimates showed a 10% increase in outpatient mental health service hours and a 31% decline in outpatient service costs among the five clinics participating in the contracting process, in comparison to clients in twelve non-contracting clinics. Use of inpatient / emergency MU services increased among clients who intermittently used services. Discussion: Contracting for outpatient mental health services achieved a primary objective, reducing outpatient costs by 31%. There was some indication of reduction in service quality among contracting clinics, as evidenced by an increase in the probability of using inpatient / emergency services. Staff reductions may have led to longer wait times and fewer walk-in appointments; clients with immediate needs may have discovered easier access to care in the emergency psychiatric unit. Limitations of this study include a lack of detailed information on illness severity and outcomes. Implications for Health Care Provision and Use: The explicit consideration of quality measurement coupled with structured evaluation of implemented programs may inform the adoption of the most cost-effective programs.