Journal of Healthcare Leadership | 2026
Authors: Huabbangyang T.; Sangketchon C.; Sawetwong K.; Buathong K.; Dontri C.; Lakhonwong Y.; Nopphakhun T.; Rojsaengroeng R.; Bunta K.
DOI: 10.2147/JHL.S614449
Journal: Journal of Healthcare Leadership
Year: 2026
Publisher: Dove Medical Press Ltd
Document Type: Article
Open Access: All Open Access; Gold Open Access; Green Open Access
Cited by: 0
Introduction: Early identification of critically ill patients in prehospital emergency settings is essential for timely triage, resource allocation, and destination planning in emergency medical services (EMS). Although several early warning scores have demonstrated utility in emergency department and in-hospital settings, their applicability in prehospital EMS environments remains limited because of operational complexity, variable resource availability, and differences in EMS systems across healthcare settings. This study aimed to develop and internally validate a multivariable prehospital prediction score for critical illness progression and 30-day mortality among non-traumatic EMS patients. Methods: This retrospective single-center cohort study included patients aged ≥18 years with non-traumatic illnesses transported by the Vajira Emergency Medical Service to the Faculty of Medicine Vajira Hospital between January 1, 2019 and December 31, 2024. The primary outcome was illness progression during hospitalization, defined as the requirement for mechanical ventilation, vasopressor and/or inotrope administration, intensive care unit admission, or 30-day mortality. Multivariable logistic regression was used for model development. Internal validation was performed using receiver operating characteristic (ROC) analysis and area under the curve (AuROC), with optimal cut-off determined by Youden’s index. Survival analysis using parametric Weibull regression was conducted to assess prediction of 30-day mortality. Results: A total of 700 patients were included, of whom 272 (38.9%) developed critical illness. Independent predictors included respiratory rate ≥25 breaths/min, shock index ≥0.9, Glasgow Coma Scale ≤9, endotracheal intubation, and oxygen supplementation. The model demonstrated good discrimination (AuROC 0.765). A cut-off score of 7 stratified patients into low- and high-risk groups, with sensitivity of 63.2% and specificity of 79.7%. Patients in the high-risk group had a 3.44-fold higher risk of 30-day mortality compared with the low-risk group. The model showed strong prognostic performance for mortality (Harrell’s C = 0.855). Conclusion: This multivariable prehospital prediction score may support risk stratification, EMS triage prioritization, destination planning, and resource allocation, potentially improving operational efficiency and patient outcomes in high-acuity prehospital settings. © 2026 Huabbangyang et al.
clinical; clinical decision support systems; critical illnesses; critically ill; decision supports; emergency medical services