Aims
Clinical Frailty Scale (CFS) 4 represents a heterogeneous group of older adults transitioning from robustness to increased vulnerability to adverse outcomes such as falls. This study aims to identify predictors of falls among CFS 4 patients and to develop a clinically applicable risk stratification system.
Methods
A retrospective analysis was performed in community-dwelling CFS 4 adults (≥65 years). Univariate analyses using independent t-tests and Fisher’s exact tests compared fallers and non-fallers across demographic, nutritional, and physical performance variables. Multivariate logistic regression identified independent predictors of falls. Model performance was assessed using the area under the receiver operating characteristic curve (AUC) and the Hosmer–Lemeshow goodness-of-fit test. A weighted risk score was derived from regression coefficients to stratify patients into risk categories.
Results
Our univariate analysis showed fallers had significantly higher Falls efficacy Scale (FES) and lower Mini Nutritional Assessment (MNA) scores. Cognition, grip strength, gait speed, calf circumference, and Short Physical Performance Battery scores showed association with higher fall risk but were not statistically significant. In multivariate analysis, FES (OR 31.62, 95% CI 6.76–147.9) and MNA (OR 5.67, 95% CI 1.61–20.05) remained independent predictors. The model demonstrated good discrimination (AUC 0.82, 95% CI 0.75- 0.88) and calibration (Hosmer–Lemeshow p = 0.77). Using a weighted FES–MNA score, patients were classified into low, moderate, and high-risk groups with fall rates of 1.6%, 22.4%, and 72.7%, respectively.
Conclusions
Our FES-MNA risk stratification system identify subgroups within CFS 4. FES and MNA assessments may support targeted fall-prevention strategies in this transitional frailty group.