1080 Does Day of Admission Impact Mortality, Length of Stay, and Endoscopy Performance in Patients With Upper Gastrointestinal Hemorrhage?

Abstract

Introduction: Multiple studies worldwide have previously investigated whether outcomes in upper gastrointestinal hemorrhage (UGIH) are different for patients admitted on a weekday compared with a weekend, with variable results. The latest report (Am J Gastroenterol. 2011;106:1621-8) did not find such a difference in the United Kingdom. We sought to answer this question in the United States using the most recent release of the largest national inpatient database.Methods: This was a retrospective cohort study using the 2009Nationwide Inpatient Sample (NIS) database. NIS is nationally representative and, in 2009, contained data from 39 million discharges. Discharges were included in the study if they were urgent/ emergent admissions and had a primary ICD-9 CM code indicating a diagnosis of UGIH. Discharges from patients under the age of 18 years were excluded. The primary outcome was inpatient mortality. Secondary outcomes were length of stay (LOS) and inpatient endoscopy performance. The exposure of interest was the day of admission. Crude and adjusted odds ratios were calculated using univariable and multivariable regression analyses. Potential confounders included in the multivariable regression models were: age, sex, race, admission type, median income in the patient's zip code, patient's comorbidities (Charlson score), hospital location (rural vs. urban), region (Northeast, Midwest, West, or South), teaching status, and hospital size. Subgroup analyses were performed for patients with variceal and non-variceal UGIH. Results: 179,609 discharges with UGIH were eligible for the study, 3,251 of which were for variceal UGIH. 75% of discharges were admitted on a weekday, and 25% were admitted on a Saturday or Sunday. The overall inpatient mortality rate was 2.2% (2.2% for non-variceal and 6.0% for variceal UGIH). On univariable analysis, the odds ratios (ORs) of mortality were similar for patients admitted during a weekend (OR 1.07; 95% confidence interval (CI) 0.90-1.27) or on a weekday (OR 1.00; reference). Similar results were found on multivariable analysis and subgroup analyses (table). The adjusted LOS was also similar between the two groups of patients and on subgroup analyses (table). However, patients admitted over a weekend were less likely to receive an inpatient endoscopy. On subgroup analysis, this finding was limited to patients with non-variceal UGIH. For patients with variceal UGIH, there was no difference in endoscopy rate between patients admitted on a weekday or over the weekend. Conclusion: Admission day is not an independent predictor of inpatientmortality or LOS for patients with upper gastrointestinal hemorrhage in the United States. However, weekday admission was an independent predictor for inpatient endoscopy performance in the total study population and in patients with non-variceal upper gastrointestinal hemorrhage. Multivariable Analysis Results for Upper Gastrointestinal Hemorrhage Outcomes Based Upon Day of Admission