Adolescents’ Use of the Emergency Department

Abstract

Background : Many of the 18 million emergency department visits by adolescents annually in the United States are for nonurgent problems that might be addressed in a primary care setting. Methods : As part of a larger randomized controlled intervention, 1023 adolescents aged 12 to 21 years registering in an urban pediatric emergency department (PED) were tracked over the subsequent 365 days to record all visits to the PED. Adolescents identifying an adolescent medicine service (AMS) as the primary care source were compared with adolescents receiving primary care elsewhere in an integrated urban medical system (non-AMS) to determine how often after the index PED visit they revisited the PED, returned to primary care (PC), visited a subspecialist (SS), or were hospitalized. Mean values and odds ratios of each type of visit were compared between AMS and non-AMS patients using multivariate logistic and ordinary least squares regressions to control for covariates. Results : AMS patients (n = 124, 12%), compared to non-AMS patients (n = 899, 88%), were more likely female (75% vs 48%, P < .001) and used public insurance (52% vs 40%, P = .017). In unadjusted comparisons, AMS and non-AMS patients did not differ in the probability of any return PED visit (46% vs 37%, P = .052) in the 365 days following the index PED visit but differed in the mean number of return PED visits (1.35 vs 0.93, P = .026). AMS patients were more likely to be hospitalized (15% vs 7%, P = .006) after the index PED visit and also had a greater mean number of hospitalizations (0.41 vs 0.19, P = .048). Multivariate analyses controlling for demographic variables, triage level of initial PED visit, and hospitalizations showed AMS patients returned to primary care after an index PED visit 24.6 days earlier than non-AMS patients ( P = .026). Conclusions : This study demonstrates attending an AMS for primary care predicted earlier return to the primary care provider after an index PED visit. Elements of adolescent specialty care producing such outcomes are worthy of further study.