Background: Several studies have identified a "weekend effect" for surgical outcomes, but definitions vary and the cause is unclear. Our aim was to better characterize the weekend effect for emergency general surgery using mortality as a primary endpoint.
Methods: Using data from the University HealthSystem Consortium from 2009 to 2013, we identified urgent/emergent hospital admissions for seven procedures representing 80% of the national burden of emergency general surgery. Patient characteristics and surgical outcomes were compared between cases that were performed on weekdays vs weekends.
Results: Hospitals varied widely in the proportion of procedures performed on the weekend. Of the procedures examined, four had higher mortality for weekend cases (laparotomy, lysis of adhesions, partial colectomy, and small bowel resection; p < 0.01), while three did not (appendectomy, cholecystectomy, and peptic ulcer disease repair). Among the four procedures with increased weekend mortality, patients undergoing weekend procedures also had increased severity of illness and shorter time from admission to surgery (p < 0.01). Multivariate analysis adjusting for patient characteristics demonstrated independently higher mortality on weekends for these same four procedures (p < 0.01).
Conclusions: For the first time, we have identified specific emergency general surgery procedures that incur higher mortality when performed on weekends. This may be due to acute changes in patient status that require weekend surgery or indications for urgent procedures (ischemia, obstruction) compared to those without a weekend mortality difference (infection). Hospitals that perform weekend surgery must acknowledge and identify ways to manage this increased risk.
Keywords: Emergency general surgery; Mortality; Outcomes; Policy; Weekend effect.