LAUSR

PURPOSE To examine hospital-level variation in the timing of asymptomatic umbilical hernia repair in children. METHODS Retrospective analysis of children undergoing umbilical hernia repair at 38 children's hospitals using the Pediatric Health Information System database (01/2013-12/2017). Early repair was defined as surgery performed at 3 years of age or younger. Repairs were categorized as emergent/urgent if associated with diagnostic or procedural codes indicating obstruction or strangulation, or if they occurred within 2 weeks of an emergency department encounter. Multivariable regression was used to calculate hospital-level observed-to-expected (O/E) ratios for early repair adjusting for emergent/urgent repair and patient characteristics. RESULTS 23,144 children were included, of which 30% underwent early repair (hospital range: 6.9%-54.3%, p ≪ 0.001). Overall, 3.8% of all repairs were emergent/urgent, and higher rates of early repair did not correlate with higher rates of emergent/urgent repair across hospitals (r = -0.10). Following adjustment, hospital-level O/E ratios for early repair varied 8.9-fold (0.19-1.70, p ≪ 0.001). CONCLUSION Timing of asymptomatic umbilical hernia repair varies widely across children's hospitals, and the magnitude of this variation cannot be explained by differences in patient characteristics or the acuity of repair. Many children may be undergoing repair of hernias that may spontaneously close with further observation. LEVEL OF EVIDENCE Level III (retrospective comparative study).