Abstract
BACKGROUND AND AIM: To determine the United States-based in-hospital gastroparesis mortality rate and independent predictors associated with it. METHODS: A retrospective study was conducted using the deidentified National Inpatient Sample and Healthcare Cost and Utilization Project database between the years 2012 and 2014. The in-hospital gastroparesis mortality rate was calculated. Patients' demographics, including age, gender, race, comorbid conditions, and hospital characteristics, were examined as potential predictors of mortality. RESULTS: The gastroparesis mortality rate was 3.19 per 1000 gastroparesis patients for the years 2012-2014. Caucasians had the highest mortality rate, with odds ratio (OR) = 2.27; 95% confidence interval (CI) 1.52-3.38, and P = 0.0001. Rural hospitals had higher mortality, with OR = 1.51, 95% CI 1.10-2.10, and P = 0.01, whereas urban nonteaching and teaching hospitals showed no statistically significant mortality difference, with OR = 0.83, 95% CI 0.6-1.15, and P = 0.27 and OR = 0.82, 95% CI 0.59-1.15, and P = 0.25, respectively. In hospitals in the south region, mortality was the highest at 65.6%, with OR = 2.05, 95% CI 1.48-2.84, and P < 0.0001. Patients with diabetes mellitus had 39% lower probability in the mortality group. CONCLUSION: Being of advanced age; being White; and being in a rural, southern U.S. hospital were predictors of in-hospital mortality in gastroparesis patients.