A nationwide assessment of the “july effect” and predictors of post-endoscopic retrograde cholangiopancreatography sepsis at urban teaching hospitals in the United States

Abstract

Background/Aims: To analyze the incidence of post-endoscopic retrograde cholangiopancreatography (ERCP) sepsis in the early (July to September) and later (October to June) academic months to assess the “July effect”. Methods: The National Inpatient Sample (2010–2014) was used to identify ERCP-related adult hospitalizations at urban teaching hospitals by applying relevant procedure codes from the International Classification of Diseases, 9th revision, Clinical Modification. Post-ERCP outcomes were compared between the early and later academic months. A multivariate analysis was performed to evaluate the odds of post-ERCP sepsis and its predictors. Results: Of 481,193 ERCP procedures carried out at urban teaching hospitals, 124,934 were performed during the early academic months. The demographics were comparable for ERCP procedures performed during the early and later academic months. A higher incidence (9.4% vs. 8.8%, p<0.001) and odds (odds ratio [OR], 1.07) of post-ERCP sepsis were observed in ERCP performed during the early academic months. The in-hospital mortality rate (7% vs. 7.5%, p=0.072), length of stay, and total hospital charges in patients with post-ERCP sepsis were also equivalent between the 2 time points. Pre-ERCP cholangitis (OR, 3.20) and post-ERCP complications such as cholangitis (OR, 6.27), perforation (OR, 3.93), and hemorrhage (OR, 1.42) were significant predictors of higher post-ERCP sepsis in procedures performed during the early academic months. Conclusions: The July effect was present in the incidence of post-ERCP sepsis, and academic programs should take into consideration the predictors of post-ERCP sepsis to lower health-care burden.