Hospital-level Variation in the Management and Outcomes o Patients With Adhesive Small Bowel Obstruction A Population-Based Analysis

Objective: Determine the association between the rate of early operative management for adhesive small bowel obstruction (aSBO) at the hospitallevel and the incidence of morbidity and mortality. Background: Mounting evidence of the benefits of early operation in patients with aSBO has translated to both an increase in the proportion of patients treated operatively, and the proportion of patients who undergo early operative management. However, variation in practice remains. Methods: We identified a population-based cohort of patients (18-80 years) who were admitted with their first episode of aSBO (2005-2014). The exposures of interest were hospital characteristics and the primary outcome measure was 30-day mortality. Hierarchical logistic regression models were used to evaluate hospital-level variation on 30-day mortality, serious complications, and bowel resection. Results: A total of 27.026 patients were admitted to 122 hospitals, 23% (n = 6090) were managed operatively, 7% (n = 1845) had a serious complication, and 30-day mortality was 4.2% (n = 1146). The proportion of patients managed with early operation ranged from 0% to 33% [median 10% (interquartile range: 5%-14%)]. There was a 17% lower likelihood of 30-day mortality for every 10% increase in proportion of patients managed with an early operation at the hospital-level (odds ratio: 0.83, 95% confidence interval: 0.70-0.99). Conclusions: Hospitals with a higher proportion of aSBO patients treated with an early operation had a lower likelihood of serious complications, bowel resection, and death, independent of hospital type and volume of aSBO admissions. Early operative intervention rates likely are a proxy for additional structures and processes of care focused on aSBO patients that may facilitate patient selection.