An implicit expectation of the pioneers of trauma system design was that high clinical volume at select centers could lead to superior outcomes. There has been little study of the regionalization of pediatric craniospinal trauma care, and whether it continues to trend in the direction of regionalization is unknown. The motivating hypothesis for this study was that trauma system design in the United States is proceeding on a rational basis, producing hospital caseloads that are increasing over time and, because of geographic siting appropriate to the needs of catchment areas, in an increasingly uniform manner.
Data were obtained from the Kids’ Inpatient Database (KID) for 1997, 2000, 2003, 2006, 2009, and 2012. Cases of traumatic spinal injury (TSI) and severe traumatic brain injury (sTBI) were identified by ICD-9 diagnostic and procedural codes. Records of patients 18 years of age and older were excluded. Hospital caseloads and descriptive statistics were calculated for each year of the study, and trends were examined. The distributions of hospital caseloads were compared year with year and with simulations of idealized systems.
Caseloads of TSI trended upward and caseloads of sTBI were stable, despite a declining nationwide incidence of these conditions during the study period, so the pool of hospitals providing services for pediatric craniospinal trauma contracted to a degree. The distributions of hospital caseloads did not change, and in every year of the study large numbers of hospitals reported small numbers of discharges. In the last year of the study, a quarter of all children with TSI were discharged from hospitals that treated approximately 1 case or fewer every other month and a quarter of all children with sTBI were discharged from hospitals that treated 1 case or fewer every 3 months.
There has been no previous study of nationwide trends in pediatric craniospinal trauma caseloads. Analysis of hospital caseloads from 1997 through 2012 supports inference of a persisting geographical mismatch between population needs and the availability of services. These observations falsify the study hypothesis. A notable fraction of pediatric craniospinal trauma care continues to be rendered at low-caseload institutions. Novel quality assurance methods tailored to the needs of low-caseload institutions deserve development and study.