To characterize the scope and magnitude of practice variation associated with the diagnosis and treatment of appendicitis at freestanding children's hospitals.Background:
Variation in care has been associated with poor outcomes and is believed to be a key driver of excess health care spending.Methods:
Retrospective cohort study of 13,328 patients treated with appendicitis at 34 children's hospitals (9/2010–9/2011). Patients were divided into complicated and uncomplicated cohorts and examined for interhospital variation in the use of diagnostic imaging (computed tomography or ultrasonography), laboratory tests, parenteral nutrition (PN), peripherally inserted central catheters (PICC), and hospital cost. The number and distribution of statistical outliers were calculated for all measures.Results:
Significant variation was found for all measures, including a 3.5-fold difference in preoperative imaging (aggregate rate: 49.0%, range across hospitals: 21.2%–73.5%, P < 0.001) and a 5-fold difference in preoperative laboratory utilization (aggregate median: 2 tests/encounter, range: 1–5 tests/encounter, P < 0.001). For patients with complicated appendicitis, we characterized a 12-fold difference in postoperative imaging (aggregate rate: 19.4%, range: 4.9%–61.6%, P < 0.001), a 48-fold difference in PICC lines (aggregate rate: 18.9%, range: 1.7%–81.8%, P < 0.001), and a 100-fold difference in PN utilization (aggregate rate: 9.3%, range: 0.4%–42.0%, P < 0.001). Median hospital cost differed 4-fold for patients with uncomplicated disease (aggregate median: $6804, range: $4200–$16,796, P < 0.001) and 4.6-fold for patients with complicated disease (aggregate median: $13,138, range: $5419–$24,779, P < 0.001). Statistical outliers on the basis of high and low utilization were identified for all measures.Conclusions:
Significant variation exists in practice, resource utilization, and treatment-related cost associated with the management of appendicitis at freestanding children's hospitals. Value-based measures are needed for benchmarking and to prioritize collaborative quality improvement efforts.