PROJECT SUMMARY/ABSTRACT Individuals dually enrolled in Medicare and Medicaid are a socially and clinically vulnerable population who account for a disproportionate amount of healthcare expenditures paid by both programs. Dual eligibles have a higher burden of disease compared to Medicare only beneficiaries, with more than 20% of individuals having multiple chronic conditions. Dually-enrolled beneficiaries also have worse cancer outcomes. They are less likely to receive recommended screening, are less likely to receive definitive treatment after a cancer diagnosis, and have a higher risk of cancer-specific mortality. The complex social and healthcare needs of dual eligibles present a challenge in effectively delivering high-quality cancer care to these patients. Accountable care organizations (ACOs) strive to provide coordinated, high-quality care to patients while containing costs for payers. As Medicare’s predominant ACO model, the Medicare Shared Savings Program (MSSP) has shown promise in improving quality of care, generating savings per beneficiary, and caring for clinically vulnerable patients. Whether the MSSP increases the quality of care across the cancer continuum for dual eligibles remains relatively unknown. The proposed research investigates how much the MSSP ACO model improves quality of cancer care for a clinically complex, vulnerable patient population. Aim 1 will estimate the effect of MSSP ACOs on breast and colorectal cancer screening in dual eligibles ages 65-75. Aim 1 will involve a quasi-experimental approach using an intention-to-treat difference-in-difference analysis with propensity score matching. Aim 2 will employ a time- to-event survival analysis that will compare the time to treatment initiation after a primary breast or colorectal cancer diagnosis for MSSP ACO-attributed dual eligibles ages 18 and older to non-ACO dual eligibles ages 18 and older. Aim 3 will assess Medicare spending on cancer treatment for dual eligibles ages 18 and older with primary breast or colorectal cancer. Analysis of costs will use generalized linear models with a log-linked, gamma distribution. Taken together, the results of these three aims will inform policymakers and providers on whether value-based care reforms deliver better quality of care for dually enrolled beneficiaries. Through this proposed research, the outlined training plan, and the guidance of a highly experienced mentorship team, the applicant will be trained in various quantitative methods using large secondary datasets, strengthen her clinical skills and thinking for providing care to a diverse patient population, and enhance her professional skills that will make her a successful independent academic physician-scientist.