ABSTRACT
BACKGROUND: Racial minorities with gastrointestinal cancer suffer disproportionately poor overall and disease-specific survival. We used a nationally representative sample to examine the relationship between race/ethnicity and mortality and determine whether these disparities were observed in the perioperative period. MATERIALS AND METHODS: The Nationwide Inpatient Sample (NIS) was used to examine patients undergoing surgery for cancers of the esophagus, stomach, pancreas, colon and rectum ("GI cancer") between 2008 and 2012. Logistic regression was used to evaluate whether race/ethnicity was associated with perioperative mortality after adjusting for sociodemographic characteristics, perioperative factors and presentation (ER vs elective). RESULTS: A total of 110,044 subjects were identified, including 75.8% Whites, 10.5% Black patients, 7.2% Hispanic patients, and 3.1% Asian/Pacific Islanders (API). Whites were generally older than minorities. In adjusted multivariable generalized linear mixed logistic models, no increase in perioperative mortality was seen for minorities. Worse outcomes were observed for those with higher Elixhauser comorbidity score (OR 6.90, CI 5.96-7.99), lower income region (OR 1.24, CI 1.10-1.40), males (OR 1.54, CI 1.42-1.68), and those without private insurance (Medicare OR 1.34, CI 1.16-1.55; Medicaid OR 1.27, CI 1.02-1.58; self-pay OR 1.64, CI 1.24-2.17). Differences in mortality were predominantly driven by comorbidities (pseudo %ΔR2 = 38.56%) and only minimally by race (pseudo %ΔR2 = 0.49%). CONCLUSION: Minority groups do not suffer higher rates of perioperative mortality for GI cancer surgeries after controlling for clinical and demographic factors. Future work to address cancer disparities should focus on areas in the cancer care trajectory such as cancer screening, surveillance, socioeconomic factors, and access.