Abstract
Background/Objectives: Pediatric trauma systems require accurate metrics for evaluating triage decisions. Undertriage occurs when an injured child requires pediatric trauma center resources but is treated at a center lacking those resources. Current undertriage definitions utilize mortality-based scores, including the Injury Severity Score (ISS) > 15 or the International Classification of Disease (ICD) Injury Severity Score (ICISS). However, resource-based metrics like the ICD Critical Care Severity Score (ICASS) may be preferable in children. This study evaluated the relationship of ISS, ICISS and ICASS to the need for pediatric trauma resources (NFPTCR) to derive a more empiric definition of undertriage. Methods: The American College of Surgeons Trauma Quality Improvement Program database was queried for patients aged ≤ 15 years old. NFPTCR was defined as blood product transfusion within 4 h, invasive procedure for cardiopulmonary stabilization/contamination/bleeding within 72 h, initial admission to intensive care unit (ICU) or ICU stay ≥ 3 days, intubation, mechanical ventilation and general anesthesia ≤ 5 years old, or physical child abuse. ICASS and ICISS were derived from 2014 to 2018 datasets and applied to the 2019 dataset. The ability of ISS, ICISS and ICASS to distinguish NFPTCR patients was assessed using multivariable logistic regression and receiver-operator characteristic (ROC) analysis. Results: Out of 97,773 children, 15,985 (16%) were NFPTCR+. ISS, ICISS and ICASS had areas under the curve of 0.760, 0.701 and 0.812 for NFPTCR+, respectively (all p < 0.001). ISS had 36% sensitivity at 15; whereas ICASS had 95%, 93% and 89% sensitivity at 5, 10 and 15, respectively. Conclusions: ICASS was superior to ISS and ICISS for identifying NFPTCR. Consideration should be given to redefining pediatric trauma undertriage based on resource-based metrics, like ICASS.