DO HOSPITAL CHARACTERISTICS PREDICT ADVERSE EVENTS AFTER BARIATRIC SURGERY?
Shrey Patel3, Rohan M. Shah*1, Lakhvir Sandhu2, Shiv Patel1, Bipan Chand2
1Northwestern University Feinberg School of Medicine, Northwestern University Feinberg School of Medicine, Chicago, IL, US, academic/medsch, Chicago, IL; 2Loyola University Chicago Stritch School of Medicine, Maywood, IL; 3University of Miami School of Medicine, Miami, FL
Introduction:
Bariatric surgery is widely performed to manage clinically severe obesity and its associated medical conditions. Though generally safe, complications can be difficult to manage, and rarely, catastrophic. Hospital characteristics may affect outcomes in bariatric surgery patients. The present study intends to evaluate effects of geographic region, hospital teaching status, and number of beds on outcomes after bariatric surgery using data sourced from the Nationwide Inpatient Sample (NIS) 2016 - 2019.
Methods:
Patients undergoing Roux-en-Y gastric bypass (RYGB), vertical sleeve gastrectomy (VSG), or adjustable gastric band (AGB) surgery were identified with ICD-10 coding. Hospitals were grouped by location/teaching status as "urban non-teaching," "urban teaching," and "rural." Hospitals were trichotomized as low-volume, medium-volume, and high-volume. Geographic regions were Northeast, Midwest, South, and West. Outcomes were medical complications (≥1 myocardial infarction, cardiac arrest, venous thromboembolism, pulmonary embolism, respiratory arrest, pneumonia, sepsis, stroke, or urinary/renal complications), in-hospital mortality, and length of stay (LOS). Multivariate logistic regressions evaluated effects on medical complications and mortality, and generalized linear modeling was used for LOS. All models controlled for race, age, sex, household income, insurance, hypertension comorbidity, hospital teaching status, and hospital case volume.
Results:
A total of 713,290 patients were included (average age: 45.0 years, 79.2% female). Patients in high-volume hospitals were more likely to experience a medical complication than those in low-volume hospitals (OR: 1.23; 95% CI: 1.07 - 1.42). Patients treated in the Midwest (OR: 1.29; CI: 1.09 - 1.52) and South (OR: 1.27; CI: 1.07 - 1.51) were more likely to have a medical complication than those in the Northeast. Hospital location/teaching status did not have an effect. There was no effect of geographic region, hospital location/teaching status, and case volume on mortality. Medium-volume (β: 0.12; SE: 0.034; P<0.001) and high-volume (β: 0.38; SE: 0.065; P<0.001) had longer LOS. Hospitals in the Midwest (β: 0.08; SE: 0.038; P=0.033) and West (β: 0.11; SE: 0.040; P=0.006) had longer LOS. There was no effect of hospital location/teaching status.
Conclusions:
Patients treated in high-volume hospitals and in the Midwest were more likely to have a medical complication and longer LOS. However, it is important to note the relatively small effect sizes in this study. Future studies should use national data to better understand and address potential disparities in medical treatment based on hospital characteristics.
Back to 2023 Abstracts