Objective: The triage algorithm of the modified Emergency Severity Index (ESI) version 4 has established validity and reliability in adults but has not been adequately evaluated in pediatric triage. The present study aimed to assess the validity and reliability of the modified ESI version 4 in pediatric triage in the emergency department.

Materials and Methods: Between January and December 2015, 1,193 patients were enrolled. The inter-rater reliability was measured with kappa statistics. Validity was evaluated by comparison with the ESI ratings with outcomes including disposition and resource consumption.

Results: One thousand one hundred ninety-three patients, including 684 boys and 509 girls, were enrolled in the present study. Most of the patients were three to eight years old and the most three common presentation at the ED were fever (24.2%), dyspnea (16.0%), and upper respiratory tract infection (2.8%), respectively. Only six (0.5%) patients received life-saving intervention and 572 (81.5%) were discharge home with medications. There were 523 cases (43.84%) in which the triage nurses and researcher assigned a similar triage rating (weighted kappa 0.25 (0.22 to 0.29), p<0.001), while 302 cases (25.31%) were over-triage and 368 cases (30.85%) were under-triage. The correlation coefficient between disposition and triage level was 0.37 (p<0.001). Resource consumption increased significantly in the lower triage level (correlation coefficient, p<0.001).

Conclusion: The present study showed a low validity and reliability using ESI tool for triage pediatric patients in emergency department at Songklanagarind Hospital.

Keywords: Triage, Pediatrics, Emergency department