Predictors of high-acuity outcomes among 10–17-year-olds utilizing pediatric emergency services in Taiwan: a sex-based comparison of male and female adolescents

Abstract

Background

Adolescents’ emergency department (ED) presentations are multi-faceted. However, patient characteristics and chief complaints associated with severe outcomes, particularly sex differences, remain underexplored. This study aimed to identify predictors of high-acuity outcomes among adolescents utilizing pediatric emergency services in Taiwan and to examine sex differences.

Methods

We analyzed data from 16,910 non-traumatic pediatric ED visits by adolescents aged 10–17 years at a major tertiary-care hospital in northern Taiwan (2009–2014). Trauma-related cases were not included, as they follow distinct clinical pathways. We tracked four severe outcomes longitudinally as high-acuity outcomes and used them in predictive modeling: (1) intensive care unit (ICU) admission or in-ED death; (2) inpatient ward admission; (3) return to the ED within 72 h for the same presenting complaint; and (4) ED length of stay exceeding 6 h. We used chi-square (χ²) and Fisher’s exact tests to analyze bivariate associations. Multivariate logistic regression models, stratified by sex, were constructed with final model selection based on the lowest Akaike Information Criterion (AIC) value to optimize model fit and parsimony. Results are presented as adjusted odds ratios (AORs) and 95% confidence intervals (CIs).

Results

A total of 2,508 adolescents (14.8%) experienced at least one high-acuity outcome. In the final model for all adolescents, the following patient characteristics were significantly associated with high-acuity outcomes: male sex (AOR = 0.90, 95% CI: 0.82–0.98); ages 16–17 (AOR = 1.23, 95% CI: 1.10–1.37); triage levels 1–2 (AORs = 1.98–2.27, 95% CIs: 1.45–3.00), indicating greater urgency for intervention; ≥2 abnormal vital signs (AORs = 1.59–1.91, 95% CIs: 1.08–2.87); and a Glasgow Coma Scale score of 13–14 (AOR = 0.49, 95% CI: 0.26–0.94), indicating mild impairment of consciousness. In this overall model, we also identified 10 chief complaints as significant predictors of high-acuity outcomes, including endocrine-related diseases (AOR = 2.10, 95% CI: 1.52–2.91), skin-related diseases (AOR = 1.95, 95% CI: 1.02–3.73), nervous system diseases (AOR = 1.34, 95% CI: 1.08–1.68), and poisoning (AOR = 1.38, 95% CI: 1.06–1.81). Among male adolescents, the significant chief complaints mirrored those in the overall model, except that eye diseases (AOR = 1.47, 95% CI: 1.01–2.17) emerged as an additional male-only predictor, and headaches were not, but musculoskeletal system diseases (AOR = 1.45, 95% CI: 1.01–2.08) were retained in the male-specific model. By contrast, only two chief complaints remained significant predictors among female adolescents: endocrine-related diseases (AOR = 1.97, 95% CI: 1.31–2.98) and headaches (AOR = 0.72, 95% CI: 0.54–0.96).

Conclusions

This study demonstrated that male and female adolescents with high-acuity outcomes exhibited distinct clinical profiles, underscoring the importance of sex-specific approaches in pediatric emergency care. Our empirical findings highlight the need for heightened clinical attention to adolescents presenting with certain chief complaints. By identifying predictors of high-acuity outcomes, this study contributes to improving clinical decision-making and quality assessment in ED settings. These findings may also inform preventive strategies and early interventions in broader healthcare contexts, including school-based and primary care services.

Clinical trial number

Not applicable.