A Prospective Comparison of Frailty Scores and Fall Prediction in Acutely Injured Older Adults

Background: Elderly (65 and older) fall-related injuries are a significant cause of morbidity and mortality. Although frailty predicts poor outcomes in geriatric trauma, literature comparing frailty scoring systems remains limited. Herein, we evaluated which frailty scoring system best predicts falls over time in the elderly. Materials and methods: Acute surgical patients 65 y and older were enrolled and prospectively observed. Demographics and frailty, assessed using the FRAIL Scale, Trauma Specific Frailty Index (TSFI), and Canadian Frailty Scale (CSHA-CFS), were collected at enrollment and 3 mo intervals following discharge for 1 y. Surveys queried the total number and timing of falls. Changes in frailty over time were assessed by logistic regression and area under the curve (AUC). Results: Fifty-eight patients were enrolled. FRAIL Scale and CSHA-CFS scores did not change over time, but TSFI scores did (P ≤ 0.01). Worsening frailty was observed using TSFI at 6 (P ≤ 0.01) and 12 mo (P ≤ 0.01) relative to baseline. Mortality did not differ based on frailty using any frailty score. Increasing frailty scores and time postdischarge was associated with increased odds of a fall. AUC estimates with 95% CI were 0.72 [0.64, 0.80], 0.81 [0.74, 0.88], and 0.76 [0.68, 0.84] for the FRAIL Scale, TSFI, and CSHA-CFS, respectively. Conclusions: The risk of falls postdischarge were associated with increased age, time postdischarge, and frailty in our population. No scale appeared to significantly outperform the other by AUC estimation. Further study on the longitudinal effects of frailty is warranted.