PROJECT SUMMARY/ABSTRACT Safety-net hospitals (SNHs) play an essential role in the U.S. health care system by providing care for low income, medically vulnerable populations. Many observers believe that because of their relatively weaker financial position, SNHs have been unfairly penalized under value-based purchasing. This is unsettling for SNHs, as the Centers for Medicare and Medicaid Services (CMS) actively pursues the goal of tying a greater portion of Medicare payments to value through alternative payment mechanisms. Bundled payment is an important model among current value-based payment programs. The Medicare Bundled Payments for Care Improvement Initiative (BPCI), implemented in 2013 for a defined set of conditions, has prompted voluntary participation by about 500 hospitals to date. In April 2016, CMS launched the Comprehensive Care for Joint Replacement (CJR) model for total hip replacement and total knee replacement surgery. Unlike the BPCI, CJR is mandatory for all non-exempt hospitals located in designated MSAs, selected by randomization. Hence analyzing CJR presents a unique opportunity for research because comparing CJR hospitals with other hospitals avoids confounding due to unobservable selection effects associated with voluntary participation. CMS is overseeing an external evaluation of the CJR. Focus is on savings to the Medicare program while upholding quality standards and comparisons are between CJR and control group hospitals. While spending control is the primary purpose of the program, impacts on SNHs are of significant consequence. Yet the evaluation does not consider the unique challenges facing SNHs. This project will address this gap by examining the impact of CJR on SNHs relative to other CJR hospitals by focusing in particular on improvement over the first five years of the program. Using descriptive and statistical techniques our aims are 1) to measure performance improvement under CJR by hospital characteristics focusing on SNH status and 2) to determine the relative contribution of patient/population versus hospital characteristics to driving performance improvement under CJR. Better understanding of what types of hospitals and which populations are achieving the greatest success under CJR and where it has failed to improve performance will be valuable to CMS as it considers future alternative payment mechanisms.