Development and internal validation of a simple EMS-based clinical score for predicting 24-hour mortality among older adult trauma patients: a retrospective single-centre cohort study

Background Early prediction of short-term mortality among older adult trauma patients in prehospital settings remains challenging. This study aimed to develop and internally validate a simple emergency medical services (EMS)-based clinical score for predicting 24-hour mortality among older adult trauma patients using routinely available prehospital variables. Methods This retrospective, registry-based, single-center cohort study included 665 trauma patients aged ≥ 60 years who were transported by EMS to a tertiary trauma center in Bangkok, Thailand, between January 2018 and December 2025. Patients were classified under Thailand Emergency Medical Triage Protocol symptom groups 21–25 with red-level severity. Candidate predictors were screened using univariate analyses and entered into multivariable logistic regression to derive a simplified clinical score. Model discrimination was assessed using the area under the receiver operating characteristic curve (AUC), calibration was evaluated using the Hosmer–Lemeshow goodness-of-fit test, and internal validation was performed using 500 bootstrap resamples. Risk groups were defined according to the optimal cutoff identified using Youden’s index. Results The 24-hour mortality rate was 5.7% (38/665). Injury mechanism (fall versus other mechanisms), facial injury, multiple injuries, and oxygen support were identified as independent predictors associated with early mortality. The derived score demonstrated high sensitivity (97.4%) and negative predictive value (99.7%) at the optimal cutoff point. The model showed good discrimination (AUC 0.856, 95% confidence interval [CI] 0.810–0.901) and remained stable after bootstrap validation (AUC 0.872, 95% CI 0.821–0.922), with good calibration (Hosmer–Lemeshow p = 0.599). Conclusion We developed and internally validated a simple EMS-based clinical score for predicting 24-hour mortality among older adult trauma patients using routinely available prehospital variables. The candidate model demonstrated good discrimination, excellent sensitivity, and stable performance after bootstrap validation. The score may support early prehospital risk stratification; however, external and prospective validation studies are required before routine clinical implementation can be recommended.