DESCRIPTION (provided by applicant): A recent NIDA Health Services Research Program Announcement (PA) called for studies of the ways in which payment mechanisms "affect the degree of access to, and delivery, utilization, and quality of prevention and treatment services for drug abuse or alcohol abuse". The PA specifically called for evaluations of performance contracting models, as proposed in this study. Despite recommendations from the Institute of Medicine, little published exists on the impact of performance contracting on quality of general medical or behavioral health care. One of the first performance based contracts (PBC) for substance abuse (SA) treatment was implemented in Delaware in 2001. The goals of this project are to improve our understanding of how PBC works and how it impacts quality of SA treatment. Results of this study will provide valuable information which will inform current and future use of performance based contracting in SA treatment and in other areas. This study will use administrative data and a facility survey to: 1) Analyze whether the PBC program met its' goals of improving facility occupancy, client active participation in treatment and completion of treatment by analyzing the change in these contracted performance measures overtime, controlling for client characteristics. 2) Analyze the impact of facility characteristics such as the proportion of revenue from public clients, size of the organization and target of incentives, on the change in the probability of clients meeting the active participation criteria over time, controlling for client severity. 3) Determine whether and how SA treatment organizations responded to the change from a cost based contract to a performance based payment system. This study employs a quasi- experimental pre-post design. While all hypotheses will be tested using quantitative methods, a facility survey will be conducted to provide additional context and detail necessary to understand the impact of the PBC and facilities' responses to the PBC. Information obtained through the survey will allow for a more complete description of the PBC. All regression analyses will be conducted using multilevel modeling techniques to account for clustering of clients within facilities. These techniques are an increasingly popular method for analysis of clustered data. The proposed research will provide valuable information for policy makers and SA treatment providers as performance based contracts continue to develop. This study will improve public health by exploring the impact of quality improvement initiatives for substance abuse treatment.