Utilization and Performance Benchmarking for Postoperative Imaging in Children With Complicated Appendicitis: Results From a Multicenter Collaborative Cohort Study.

ABSTRACT

OBJECTIVE: The aim of this study was to characterize hospital-level variation and establish diagnostic performance benchmarks for postoperative imaging in children with complicated appendicitis. SUMMARY BACKGROUND DATA Wide variation in preoperative imaging in children with suspected appendicitis has been previously described. Variation in the use and accuracy of postoperative imaging to diagnose suspected organ space infection (OSI) following appendectomy has not been characterized. METHODS: Multicenter retrospective analysis of children who underwent appendectomy for complicated appendicitis using data from the NSQIP-Pediatric Appendectomy Pilot Collaborative. Resource utilization measures included rates of postoperative imaging [ultrasound (US) and computed tomography (CT)] and imaging-associated diagnostic efficiency ratio (DER; number of OSIs diagnosed/study obtained). Radiation stewardship measures included US utilization process measures (rate of US as the initial diagnostic study and rate of CTs preceded by an attempt at US) and CT-associated DER. Hospital-level observed-to-expected ratios (O/Es) were calculated for each measure after adjusting for demographic characteristics and disease severity using multivariable regression. RESULTS: A total of 1316 patients from 20 hospitals were included. Overall, 18.3% of patients underwent postoperative imaging (hospital range 4.8%-33.3%), and O/Es varied 3.5-fold among hospitals (P < 0.01). The overall imaging-associated DER was 0.56 OSIs/study (hospital range 0-1.00), and O/Es varied 2.7-fold among hospitals (P < 0.01). Significant variation was also observed for US as the initial diagnostic study (overall 41.5%; O/E range 0.40-2.01, P < 0.01), CTs preceded by US (overall 27.3%; O/E range 0-3.66, P < 0.01), and CT-associated DER (overall 0.69 OSI's/CT; O/E range 0-1.80, P < 0.01). Fifty percent of hospitals were a statistical outlier on at least 1 measure. CONCLUSION: Significant variation exists across hospitals in imaging practices to diagnose suspected OSI following appendectomy. Imaging utilization benchmarking may assist hospitals in prioritizing quality improvement efforts to optimize resource utilization and radiation stewardship.