Temel Kayan, Hasan Oktay Emir, Fatih Seğmen, Hayriye Cankar Dal, Firdevs Tuğba Bozkurt Biçer, Esra Yakışık Aktekin, Cihangir Doğu
Background/Objectives: Ground-level falls (GLFs) are the leading cause of injury-related deaths in older adults, yet a validated bedside tool for early mortality risk stratification remains unavailable. The Braden scale, recorded routinely at intensive care unit (ICU) admission, has shown prognostic value across several critical illness populations. This study aimed to determine whether the admission Braden score is associated with in-hospital mortality in ICU patients admitted following a GLF, and to derive a simple bedside clinical risk score. Methods: A single-centre retrospective cohort study was conducted in a 216-bed mixed ICU over a four-year period (January 2020-June 2024). Adult patients admitted to the ICU following a GLF occurring outside the hospital were included. The primary outcome was in-hospital mortality. Multivariable logistic regression was used, and a five-component integer-weighted bedside risk score was derived. Internal validation was performed by 1000-fold bootstrap resampling. Results: Among 500 patients (median age 83 years; 63.3% female), 84 (16.8%) died. The admission Braden score was significantly lower in non-survivors (median 12 vs. 13; p < 0.001) and was an independent predictor of mortality after multivariable adjustment (adjusted OR 0.821 per one-point increase; 95% CI 0.706-0.954; p = 0.010). A five-component bedside risk score-incorporating the Braden score, INR, emergency admission, serum sodium, and procalcitonin-achieved a bootstrap-corrected AUC of 0.833 (95% CI 0.780-0.881) and stratified patients across a 16-fold mortality gradient (5.0% to 79.4%). Conclusions: The admission Braden score is associated with in-hospital mortality. When combined with four routine variables, discriminative performance improved. This score has undergone internal validation only. External validation in multicentre prospective cohorts is warranted before clinical use.