Predictive factors of improved postoperative outcomes in pancreatic resection include patient insurance status and facility surgical volume

Background: Resection is the only potentially curative treatment for pancreatic cancer. While previous studies have demonstrated outcome disparities at low volume facilities and in underinsured patients, few have evaluated institutional variables in a large sample using recent data reflective of our rapidly changing healthcare system. We investigated the impact of facility academic status, case volume, and insurance status on length of stay, and 30- and 90-day mortality. Methods: Data were retrieved from the National Cancer Database for 34,718 pancreatic cancer patients who underwent pancreatectomy between 01/01/2010-12/31/2015. Facilities were classified as "Very Low Volume" (≤5 surgical cases/year), "Low Volume" (6-16 cases/year), or "High Volume" (>16 cases/year). Multivariable logistic regression was used to investigate associations between facility or insurance factors and 30- and 90-day mortality. Results: Insurance status was the strongest predictor of positive outcomes, with privately insured patients demonstrating the shortest length of stay (9.6 days) and lower 30-day mortality [OR (95% CI) 0.61 (0.50, 0.74)] and 90-day mortality [OR (95% CI) 0.68 (0.60, 0.78)] than publicly insured patients (P<0.001). High-volume facilities displayed lower 30-day mortality [OR (95% CI) 0.60 (0.49, 0.72)] and 90-day mortality [OR (95% CI) 0.68 (0.60, 0.78)] than very low volume facilities (P<0.001). Compared to non-academic programs, academic programs displayed lower 90-day mortality [OR (95% CI) 0.84 (0.75, 0.94), P=0.002], but equivalent 30-day mortality. Conclusions: These data illustrate the persistent outcome gap affecting underserved or underinsured patients with pancreatic cancer despite efforts in quality improvement and healthcare reform.