Factors Associated With Pediatric Nontransport in a Large Emergency Medical Services System

Background: Pediatric patients attended to by emergency medical services (EMS) but not transported to the hospital are an at-risk population. We aimed to evaluate risk factors associated with nontransport by EMS in pediatric patients. Methods: We reviewed medical records of 24 agencies in a regional EMS system in Southwestern Pennsylvania between January 1, 2014, and December 31, 2017. We abstracted demographics (age, sex, medical complaint, median household income by zip code, race, ethnicity), clinical characteristics (abnormal vital signs by age, procedures done), and transport characteristics. We excluded patients ≥ 18 years, interfacility transfers, scene assists, cardiac arrest, and those without a patient encounter. We used unadjusted and adjusted logistic regression to identify factors associated with nontransport, reporting adjusted odds ratios (aOR) with 95% confidence intervals (CIs). Results: We included 30,663 pediatric patients (52.9% male, mean ± SD age = 8.5 ± 6.2 years), of whom 5,002 (16.3%) were nontransports. In adjusted analysis (aOR, 95% CI), nontransports were associated with medical categories of trauma (4.32, 3.57–5.23), respiratory (4.03, 3.09–5.26), toxicologic (2.53, 1.66–3.86), and syncope (5.97, 3.78–9.41). Nontransports were less likely for psychiatric (0.52, 0.34–0.79) complaints; for black patients compared to white (0.31, 0.26–0.37); and in patients 6 to <12 years (0.76, 0.65–0.90), 2 to <6 years (0.77, 0.65–0.91), 1 to <2 years (0.53, 0.42–0.66), and 1 month to 1 year (0.52, 0.40–0.66) compared to patients ≥ 12 years of age. Nontransport was associated with longer scene time (1.03, 1.02–1.04) and with fall compared to winter (1.29, 1.08–1.54) and was less likely in those with abnormal mental status (0.45, 0.33–0.62), medication administration (0.16, 0.08–0.31), or monitor application (0.10, 0.06–0.15). Conclusion: Pediatric nontransports are associated with traumatic, respiratory, and toxicologic complaints and older age. These findings can facilitate development of refusal protocols and research on outcomes of these at-risk patients.