Racial/ethnic disparities in emergency general surgery: Explained by hospital-level characteristics?

Background: To quantify racial/ethnic differences in outcome after emergency general surgery (EGS). Methods: Patients receiving a representative EGS (colectomy, small bowel resection, or ulcer repair operation) performed within the first 24 hours of hospital admission were identified in the Nationwide Inpatient Sample between 2000 and 2008. Multivariable logistic regression was used to estimate the overall disparity in odds of death between African Americans (AAs) and Caucasians. Hierarchical models were then used to evaluate association of hospital-level factors and death after EGS. Results: A total of 116,344 patients were identified. AA patients had 10% higher odds of dying after EGS than Caucasian patients (adjusted odds ratio 1.10, P = .02). All patients treated at hospitals with greater than 6% AA EGS patients had higher odds of death than those at hospitals with fewer percentage of AA EGS patients (adjusted odds ratio 1.16 to 1.42, P \u3c .002). Conclusion: There is racial/ethnic disparity in outcome after selected EGS; however, this disparity is explained by hospital-level factor