The empirical association between high hospital procedure volume and lower mortality rates has led to recommendations for the centralization of complex surgical procedures. Yet redirecting patients to a select number of high-volume hospitals creates potential negative consequences for market competition. We use patient-level data to estimate the association between hospital procedure volume and patient mortality and costs. We also estimate the association between hospital market concentration and mortality, cost, and prices. We use our estimates to simulate the change in social welfare resulting from redirecting patients at low-volume hospitals to high-volume facilities. We find that a higher procedure volume leads to significant reductions in mortality for patients undergoing surgery for pancreatic cancer, but not colon cancer. Procedure volume also influences costs for both surgeries, but in a nonlinear fashion. Increased market concentration is associated with higher costs and prices for colon cancer, but not pancreatic cancer patients. Simulations indicated that centralizing pancreatic cancer surgery is unambiguously welfare enhancing. In contrast, there is less evidence to suggest that centralizing colon cancer surgery would be welfare improving.
Keywords: Cancer costs; centralization; hospital competition; volume-outcome.