The study will evaluate a unique natural experiment that aims to transform the primary care of patients with depression. The experiment's aim is to cause statewide implementation of an evidence-based care model for depression by changing the reimbursement system and by using an established regional collaborative (the Institute for Clinical Systems Improvement) to both facilitate the model's widespread use and certify that medical groups have implemented it. Study hypotheses include the following:

1. Newly treated depressed patients in medical groups that have implemented the new reimbursement and facilitation will report receiving higher rates of best care processes than such patients in these medical groups before implementation.
2. Rates of best care practices reported to be received by newly treated depressed patients two years after each medical group implements changes will be maintained at least at the rate reported by patients one year post implementation.
3. Newly treated depressed patients in medical groups that have participated in the new reimbursement and facilitation will have greater improvement in depression symptoms and work productivity and lower healthcare costs than such patients in groups before participation.
4. Medical group measures of priority for improving depression care, capability to manage change, and practice systems will be predictive of more patient-reported best care processes, both at one point in time and in change over time.