A Proxy-Based Adaptation of the Nottingham Trauma Frailty Index for Older Saudi Trauma Patients

Introduction

Frailty is increasingly recognised as a stronger predictor of adverse outcomes following major trauma than chronological age alone. The Nottingham Trauma Frailty Index (NTFI) conceptualises frailty as a multidimensional construct incorporating functional dependence, cognitive impairment, and physiological vulnerability. However, several original NTFI variables are not routinely available at emergency department (ED) presentation. We aimed to develop and evaluate a proxy-based adaptation of the NTFI using routinely collected trauma registry data in Saudi Arabia.

Methods

We analysed prospectively collected registry data from 1,905 patients aged ≥55 years admitted following trauma to a single Level I major trauma centre between January 2018 and December 2024. Proxy frailty variables were selected a priori according to the NTFI conceptual framework, prioritising clinical relevance, biological plausibility, and availability at ED presentation. Univariate and multivariable logistic regression analyses were performed. The primary outcome was ED disposition requiring higher-level care (ward or ICU admission). Statistical significance was set at p<0.005. Sensitivity analysis excluded non-significant variables. Ethical approval was obtained from the institutional review board (H1RI-23-Jan 22–01); anonymised data were analysed.

Results

The multivariable model was statistically significant (χ²=754.40, df=5, p<0.001), explaining 58.5% of variance (Nagelkerke R²=0.585) and correctly classifying 93.0% of cases. Lower Glasgow Coma Scale score (OR=0.415, p<0.001), ED blood transfusion (OR=12.39, p<0.001), ambulance arrival (OR=1.77, p=0.022), and mechanism of injury (falls; OR=0.144, p<0.001) were independently associated with higher-level care requirement. Chronological age was not statistically significant (OR=0.998, p=0.842). Sensitivity analysis excluding age yielded identical explanatory power and classification accuracy.

Conclusions

A proxy-based frailty construct derived from routinely collected trauma data demonstrates strong predictive performance for early care escalation among older trauma patients in Saudi Arabia. Chronological age alone did not predict outcome. Early multidimensional frailty identification using pragmatic proxy variables may enhance risk stratification and support geriatric-informed trauma pathways.