K. L. Anderson, Jr.1, S. Thomas2,3, M. A. Adam4, R. P. Scheri4, M. T. Stang4, S. A. Roman4, J. A. Sosa3,4,5,6 1Duke University Medical Center,School Of Medicine,Durham, NC, USA 2Duke University Medical Center,Department Of Biostatistics,Durham, NC, USA 3Duke Cancer Institute,Durham, NC, USA 4Duke University Medical Center,Department Of Surgery,Durham, NC, USA 5Duke University Medical Center,Department Of Medicine,Durham, NC, USA 6Duke Clinical Research Instiute,Durham, NC, USA
Introduction: Understanding and reducing racial disparities in health care is a high priority nationwide. Disparities in outcomes have been demonstrated for pancreatic cancer, but studies have been limited to single institution experiences or analyses focused only on White vs. Black patient groups. Our aim was to investigate how racial disparities have changed over the last two decades in the management of pancreatic cancer.
Methods: Adult patients undergoing pancreaticoduodenectomy for Stage I and II pancreatic adenocarcinoma were identified from the National Cancer Data Base, 1998-2011. Univariate analysis was used to compare demographic, treatment, and short-term outcomes. Multivariate regression and survival analyses were used to examine differences between races in access to high-volume (H-V) (≥11 cases/year) facilities and overall survival.
Results: Over the study period, 10520 patients underwent pancreaticoduodenectomy for adenocarcinoma at 1044 institutions, of which 8852 (84.1%) were White, 887 (8.4%) Black, 522 (5.0%) Hispanic, and 259 (2.5%) Asian. Compared to White (56.4%) patients, Black (63.1%), Hispanic (66.5%), and Asian (61.4%) patients were more likely to receive care at low-volume institutions. All other races experienced higher rates of 30-day readmission (Black: 9.6%, Hispanic: 10.7%, Asian: 11.2%) compared to Whites (8.0%, p=0.02), and longer length of hospital stay (Black and Hispanic median 10 days, Asian 11 days,) compared to White patients (9 days, p<0.001). Positive margin and 30-day mortality rates did not differ between races. After adjustment, Black [OR 0.74 (0.63-0.86), p<0.001] and Hispanic [OR 0.78 (0.64-0.95), p=0.01] patients had decreased odds of accessing a H-V facility compared to White patients. Over time, access to H-V facilities improved for Hispanic [p=0.03] and Asian [p=0.04] patients, while there was no significant change for Black patients [p=0.81] compared to Whites (Figure). In unadjusted analysis, there were no differences in overall survival based on race (p=0.47). After adjustment, Hispanic patients had decreased mortality compared to Whites (HR=0.81, p=0.05), but no differences were seen for Black or Asian patients.
Conclusion: Over time, access to H-V institutions improved for Asian and Hispanic patients compared to White patients, while the gap in access persisted for Black patients. Improving access to H-V institutions may provide improved outcomes for vulnerable minority patients.
