Development of a Mortality Prediction Model for Fall Injury Patients: An Analysis of the PATOS Trauma Registry

This retrospective, multicenter, registry-based cohort study used data from the Pan-Asian Trauma Outcomes Study registry between January 2016 and December 2024. Adult patients with fall-related trauma were included if they had available data on age, sex, emergency department vital signs, and 30-day mortality. The final cohort was randomly divided into derivation and validation cohorts in a 2:1 ratio. Candidate predictors were selected based on clinical relevance, field triage criteria, prior trauma literature, and early availability. Multivariable logistic regression was used to develop the model, and regression coefficients were converted into an integer-based score. Model performance was assessed using discrimination, calibration, Brier score, and threshold-based metrics.