M. D. Giglia1, A. A. Gullick1, P. L. Bosarge2, J. D. Kerby2, D. I. Chu1 1University Of Alabama At Birmingham,Gastrointestinal Surgery,Birmingham, AL, USA 2University Of Alabama At Birmingham,Acute Care Surgery,Birmingham, AL, USA
Introduction: Racial disparities have been well-studied in elective specialties. Less is understood about disparities in emergency general surgery. We aimed to determine the contribution of race to readmission rate, length of stay (LOS), and mortality in patients who required emergency general surgery in a contemporary surgical population.
Methods: We queried the 2011-2014 American College of Surgeons National Surgical Quality Improvement Program database for all patients who underwent emergency general surgery and stratified by race. Primary outcomes were readmission rate, LOS and mortality. Predictors of readmission rate, LOS, and mortality were identified with multivariate logistic regression and negative binomial models, from which Odds Ratios [OR] and Incident Rate Ratios [IRR] were obtained.
Results: Of 114,411 patients who underwent emergency general surgery, 69.6% were white, 13.3% were Hispanic, 12.1% were black and 4.9% were Asian. Compared to white, Hispanic and Asian patients, black patients had higher rates of heart failure (2.5% vs 1.8%,1.0% and 0.7%), end-stage renal disease on dialysis (4.9% vs 1.6%,1.8% and 2.3%), total functional dependence (2.6% vs 1.8%, 1.0% and 1.0%), hypertension (49.3% vs 40.3%, 23.9%, and 31.6%), smoking (26.5% vs 21.7%, 15.2%, and 10.9%), diabetes (18.2% vs 12.4%, 12.5%, and 12.8%), presentation with an open wound/wound infection (6.8% vs 4.4%, 3.0% and 1.9%), and ASA Class 4-5 (20.3% vs 18.0%, 7.6% and 9.0%) (p<0.001), respectively. On unadjusted comparison, black patients had higher readmission rates (9.3% vs 8.5%, 5.9% and 5.5%, p<0.001), mean LOS (7.54 vs 6.43, 3.97, and 5.03 days, p<0.001) and median LOS (5 vs 4, 2, 2 days, p<0.001) than white, Hispanic and Asian patients, respectively. White patients had a higher unadjusted mortality rate compared to black, Hispanic and Asian patients (6.5% vs 5.5%, 2.5% and 3.3%, p<0.001). On adjusted comparison, black race was not independently associated with higher readmission rates compared to white patients (OR 1.01, 95%-Confidence Interval [CI] 0.95-1.08); however, Hispanic (OR 0.88, CI 0.81-0.95) and Asian (OR 0.78, CI 0.69-0.88) race was associated with lower risk for readmission. Black race remained independently associated with longer mean LOS (7.75 days vs 6.60, 4.09 and 5.12 days, p<0.001) and had longer LOS of 11%, 14% and 6% compared to white, Hispanic and Asian patients, respectively (IRR 1.11, 1.14 and 1.06, p<0.0001). Compared to white patients, black (OR 0.80, CI 0.72-0.88), Hispanic (OR 0.99, CI 0.86-1.13) and Asian patients (OR 0.90, CI 0.74-1.09) had a lower risk of mortality.
Conclusion: Black patients who underwent emergency general surgery had significantly longer LOS than white, Hispanic and Asian patients. Mortality, however, was highest for white patients compared to all other groups. Further studies are needed to better understand these observations and to identify actionable opportunities to reduce these disparities.