The study, conducted at a tertiary women’s and children’s hospital in China, included 24,479 children treated through the institution’s high-priority pathway during 2025. The pathway was broader than a room reserved for active cardiopulmonary resuscitation: triage levels I and II were assigned to the red zone, and level III was normally assigned to the yellow zone. The analysis retained 123 level-IV encounters recorded in the yellow zone in the primary model and excluded them in a sensitivity analysis.

Researchers ordered visits chronologically for each child and used a rolling 72-hour window to identify returns. A subsequent record represented a new presentation because the hospital generated a new emergency registration and nurse triage only after a child left the hospital; continuous observation did not create another record. The primary analysis treated every visit as a potential index encounter, including visits that were themselves returns, allowing chained returns. It did not capture lower-acuity revisits to other emergency or outpatient areas.

The primary model used mixed-effects logistic regression with a patient-level random intercept to account for repeated visits by the same child. It included emergency zone, sex, age group, focused diagnostic category, abnormal vital-sign count, night-shift presentation and season. Emergency zone and triage level were not included together because they were structurally linked in the institutional workflow. Sensitivity analyses examined alternative adjustment for triage level and excluded level-IV encounters, neonates, high-frequency patients, chained returns, or visits without a complete 72-hour observation period; another analysis used only each child’s first visit.

Yellow-zone encounters had higher adjusted odds of return than red-zone encounters (adjusted odds ratio, 2.28; 95% confidence interval, 1.34–3.87). Oncologic conditions had the strongest adjusted diagnostic association (adjusted odds ratio, 6.48; 95% CI, 3.54–11.87), followed by hematologic diseases (adjusted odds ratio, 2.49; 95% CI, 1.06–5.89). By contrast, the abnormal vital-sign count, night-shift presentation, circulatory diseases and neonatal or perinatal diagnoses were not independently associated with return in the primary model.

The unadjusted subgroup rates differed from the adjusted diagnostic findings. Circulatory diseases had a crude 72-hour return rate of 23.46%, compared with 19.16% for oncologic conditions and 10.81% for hematologic diseases. The circulatory association did not persist after multivariable adjustment, illustrating how age distribution, emergency-zone allocation, repeated visits and clustering of patients can shape crude rates. Abnormal vital-sign count represented fever, hypoxia, tachycardia and tachypnea measured at presentation; it was not an assessment performed at discharge.

An interaction model indicated that the oncologic association differed by emergency zone. The interaction between yellow-zone presentation and oncologic diagnosis was 5.10 (95% CI, 1.25–20.84; p=0.023). Zone-specific comparisons showed an adjusted odds ratio of 6.20 (95% CI, 3.41–11.28) for oncologic conditions versus other diseases in the yellow zone, compared with 1.22 (95% CI, 0.27–5.46) in the red zone. The wide confidence interval for the interaction indicates limited precision, and the authors characterized this finding as exploratory pending corroboration in larger samples.

The oncologic association remained statistically significant across the reported sensitivity analyses. For example, it was 8.57 (95% CI, 6.59–11.04) in the first-visit analysis and 5.76 (95% CI, 2.51–13.24) when chained returns were excluded. The yellow-zone association weakened after excluding neonates and high-frequency patients, suggesting that patient composition may partly explain the pooled zone difference. Hematologic disease remained associated with return in several analyses but weakened when high-frequency patients were removed.

The findings have important limitations. This was a single-center study whose triage and zone-allocation policies may limit generalizability, and the outcome covered only returns to the same high-priority pathway rather than all pediatric emergency revisits. Data were incomplete for final disposition, outpatient follow-up, caregiver decisions and return preventability, so the analysis could not distinguish disease progression, planned reassessment and potentially avoidable returns. Residual confounding by disease severity, treatment stage or comorbidity burden also remained possible. The authors proposed that children with oncologic conditions, particularly in the yellow zone, may warrant strengthened discharge assessment, explicit return precautions and short-interval follow-up, but the observational study did not test whether such measures reduce returns.

Subject of Research: Clinical and diagnostic factors associated with 72-hour unscheduled returns to high-priority pediatric emergency care

Article Title: Clinical and diagnostic factors associated with 72-h unscheduled return to high-priority pediatric emergency care: a health services retrospective cohort study

Article References: Zhao, J., Pan, L., Jin, Q., Gao, J., & Cheng, M. (2026). Clinical and diagnostic factors associated with 72-h unscheduled return to high-priority pediatric emergency care: a health services retrospective cohort study. Frontiers in Public Health, 14, Article 1955991. https://doi.org/10.3389/fpubh.2026.1955991

Image Credits: AI Generated

DOI: 10.3389/fpubh.2026.1955991

Keywords: pediatric emergency care, unscheduled return visits, oncologic conditions, hematologic diseases, care transitions, triage zones, health services research, retrospective cohort, Clinical, diagnostic, factors, associated