Comparison of overall survival in gallbladder carcinoma at academic versus community cancer centers: An analysis of the National Cancer Data Base.

Abstract

Gallbladder carcinoma (GBC) has a poor prognosis. Studies demonstrated that teaching facilities may provide a lower risk of mortality in patients undergoing pancreatic and colon resection vs nonteaching facilities. We hypothesized that survival rates are higher in academic cancer centers (ACCs) vs community cancer centers (CCCs). Patients with all stages of GBC were identified from the National Cancer Database (2007-2012). Propensity score matching adjusted for selection bias. Descriptive statistics were calculated for all variables. Overall survival (OS) was compared by facility type (ACC vs CCC) and case volume (low vs high) via multivariable Cox proportional hazards regression. A total of 7967 patients met the inclusion criteria. Following propensity matching, 2801 patients were analyzed from each facility type. Median OS following surgery was higher for ACC (20.99 months, 95% confidence interval [CI], 19.61-22.64, P = .002) than CCC (17.68 months, 95% CI, 16.46-19.25). Following Cox modeling, GBC treatment at ACCs was a protective factor for OS (adjusted hazard ratio 0.876, 95% CI, 0.801-0.958, P = .004). GBC treatment at ACCs is an independent predictor of OS. High volume ACCs are associated with improved OS compared with low volume ACCs. The site of care and case volume in ACCs may contribute to improved survival outcomes.