PROJECT SUMMARY/ABSTRACT Behavioral health disorders affect a large share of the U.S. population, yet the healthcare system fails to provide adequate access to behavioral health care services. Consequently, many patients seek the emergency department (ED) for lack of convenient or accessible alternatives or because they are uninsured or underinsured. Behavioral health visits to the ED increased 44% in the U.S. between 2006 and 2014, despite hospital and payer efforts to steer patients toward lower cost alternatives. Previous research shows that a substantial portion of ED visits could be avoided if patients had improved access to primary care. A key channel is Community Health Centers (CHCs), which deliver primary care and some behavioral health care to a medically underserved population of nearly 30 million individuals. Three-quarters of CHC patients are either Medicaid recipients or uninsured. While CHCs are widely considered to be a primary care alternative to the ED, whether they are effective in preventing behavioral health ED visits is not known. Integration of primary care and behavioral health care is increasingly viewed as an effective strategy for care coordination. While nearly 90% of CHCs provide some behavioral health services, the level of service for behavioral health varies substantially across facilities and over time. We propose to exploit this variation in order to understand whether and to what extent EDs function as a behavioral health care substitute for patients who lack or have poor access to behavioral health care in CHCs. Our aims are to: 1. Measure and describe the distribution of CHC behavioral health services across markets and over time. 2. Determine the extent to which greater integration of behavioral health services in CHCs leads to reduction in ED behavioral health visits. 3. Explore whether the impact of CHC behavioral health integration on ED behavioral health visits differs across payer groups. Descriptive analyses will demonstrate trends in the provision of behavioral health services in CHCs using national level data from the Health Resources and Services Administration Uniform Dataset across Primary Care Service Areas (PCSAs) over the period 2012-2018. Using econometric analyses, we will test hypotheses addressing the relationship between ED and CHC behavioral health visits by all patients and by medically vulnerable patients at the service area level. Results will inform efforts to reduce ED utilization through greater provision of behavioral health services in CHCs and also provide evidence on a potentially important benefit of integration of behavioral health and primary care services.