UC Davis

INTRODUCTION: Pancreatic cancer (PC) surgery has been associated with improved outcomes and value when performed at high-volume centers (HVC; ≥20 surgeries annually) compared to low-volume centers (LVC). Some have used these differences to suggest that regionalization of PC surgery would optimize patient outcomes and expenditures. METHODS: A Markov model was created to evaluate 30-day mortality, 30-day complications, and 30-day costs. The differences in these outcome measures between the current and future states were measured to assess the population-level benefits of regionalization. A sensitivity analysis was performed to evaluate the impact of variations of input variables in the model. RESULTS: Among 5958 new cases of pancreatic cancer in California in 2021, a total of 2443 cases (41 %) would be resectable; among patients with resectable PC, a total of 977 (40 %) patients would undergo surgery. In aggregate, HVC and LVC 30-day postoperative complications occurred in 364 patients, 30-day mortality in 35 patients, and healthcare costs expended managing complications were $6,120,660. In the predictive model of complete regionalization to only HVC in California, an estimated 29 fewer complications, 17 fewer deaths, and a cost savings of $487,635 per year would occur. CONCLUSIONS AND RELEVANCE: Pancreatic cancer (PC) surgery has been associated with improved outcomes and value when performed at high-volume centers (HVC; ≥20 surgeries annually) compared to low-volume centers (LVC). Complete regionalization of pancreatic cancer surgery predicted benefits in mortality, complications and cost, though implementing this strategy at a population-level may require investment of resources and redesigning care delivery models.