• Lisa McLeod, John Flynn, Mark Erickson, Nancy Miller, Ron Keren, and John Dormans.
• *Section of Hospital Medicine, Children's Hospital Colorado ‡Division of Orthopedic Surgery, Children's Hospital Colorado, Aurora, CO †Division of Orthopedic Surgery, Children's Hospital of Philadelphia §Department of Pediatrics, Children's Hospital of Philadelphia, Philadelphia, PA.
• J Pediatr Orthop. 2016 Sep 1; 36 (6): 634-9.

BackgroundReadmission for surgical-site infection (SSIs) following spinal fusion for NMS impacts costs, patient risk, and family burden; however, it may be preventable. The purpose of this study was to examine variation in hospital performance based on risk-standardized 60-day readmission rates for SSI and reoperation across 39 US Children's Hospitals.MethodsRetrospective cohort study using the Pediatric Health Information Systems (PHIS) database involving children aged 10 to 18 years with ICD9 codes indicating spinal fusion, scoliosis, and neuromuscular disease discharged from 39 US children's hospitals between January 1, 2007 and September 1, 2012. Readmissions within 60 days for SSI were identified based on the presence of ICD9 codes for (1) infectious complication of device or procedure, or (2) sepsis or specific bacterial infection with an accompanying reoperation. Logistic regression models accounting for patient-level risk factors for SSI were used to estimate expected (patient-level risk across all hospitals) and predicted (weighted average of hospital-specific and all-hospital estimates) outcomes. Relative performance was determined using the hospital-specific predicted versus expected (pe) ratios.ResultsAverage volume across hospitals ranged from 2 to 23 fusions/quarter and was not associated with readmissions. Of the 7560 children in the cohort, 534 (7%) were readmitted for reoperation and 451 (6%) were readmitted for SSI within 60 days of discharge. Reoperations were associated with an SSI in 70% of cases. Across hospitals, SSI and reoperation rates ranged from 1% to 11% and 1% to 12%, respectively. After adjusting for age, sex, insurance, presence of a gastric tube, ventriculoperitoneal shunt, tracheostomy, prior admissions, number of chronic conditions, procedure type (anterior/posterior), and level (>9 or <9 vertebrae), pe ratios indicating hospital performance varied by 2-fold for each outcome.ConclusionsAfter standardizing outcomes using patient-level factors and relative case mix, several hospitals in this cohort were more successful at preventing readmissions for SSIs and reoperations. Closer examination of the organization and implementation of strategies for SSI prevention at high-performing centers may offer valuable clues for improving care at lower performing institutions.Level Of EvidenceLevel III.

52 keywords...

hide…