[unreadable] DESCRIPTION (provided by applicant): Depression is a common condition with unusually high patient disability, health care costs, and lost productivity. Despite an unusually strong evidence base for a best practice care model that would improve care, current care is well below that standard and not improving very much. This proposal's overall goal is to evaluate the effects of an impending statewide change in reimbursement combined with facilitated implementation of that best practice model. By conducting this study in close cooperation with all the health plans and a local quality improvement collaborative, the project will be able to achieve the following aims: #1. To use a multiple baseline design with staggered implementation to test the effect of these changes on the frequency and sustainability of evidence-based care processes for patients with depression #2. To test the effect of the care process changes on change in depression symptoms, healthcare costs, and work productivity #3. To identify the organizational factors within a recent conceptual model that affect the implementation and effects of the care changes #4. To describe the reach, adverse outcomes, adoption, implementation, maintenance, and spread of the system changes and new care processes in order to evaluate their potential for broad replication RELEVANCE: This unusual opportunity will provide information that is essential to knowledge of the dissemination and implementation questions that interfere with closing the many large gaps between what is known and what is done in usual medical practice. In addition, if the intervention being tried here is successful, it will serve as a generalizable model for a major improvement in depression care, with all that portends for the human problems and costs that are caused by this common problem. Finally, by conducting the study in cooperation with those who pay for and deliver the care, the project will demonstrate a new approach to research-practice partnerships. [unreadable] [unreadable] [unreadable]