Associations between nutritional risk indices and fall history in older adults: a retrospective cross-sectional study

Nutrition plays a vital role in musculoskeletal health, bone density, fall risk, and postural stability, alongside being a reflection of overall health status. Nutritional assessment indices are rapid and noninvasive tools for several indications, yet their association with falls remains uncertain. We aimed to investigate four commonly used nutritional indices: GNRI (Geriatric Nutritional Index), PNI (Prognostic Nutritional Index), CONUT (Controlling Nutritional Status), and MNA-SF (Mini Nutritional Assessment Short Form) to establish their relationship with falls in older adults with potential implications for clinical practice.

Patients aged ≥ 65 years who underwent comprehensive geriatric assessment between 1 January 2022 and 31 January 2024 were included in the analysis. After application of the exclusion criteria, the final study population comprised 952 individuals. Fall history and fall frequency during the preceding 12 months were recorded. Associations between nutritional indices and falls were evaluated using receiver operating characteristic (ROC) analysis and logistic regression models, with additional post-hoc group comparisons based on fall frequency.

MNA-SF, GNRI, and PNI were significantly associated with both fall risk and fall frequency (all p < 0.001), whereas CONUT showed no significant association (p = 0.171). Post-hoc analyses demonstrated that MNA-SF was the only index that showed significant differences between single and recurrent fallers in post-hoc group comparisons (p = 0.010). MNA-SF (OR:0.67, 95% CI:0.62-0.74), GNRI (OR:0.85, 95% CI:0.82-0.88), and PNI (OR:0.92, 95% CI:0.86-0.97) continued to have an independent association with falls in regression analysis (all p < 0.005). ROC analysis showed the highest discrimination for MNA-SF (AUC: 0.697, 95% CI: 0.660-0.733), followed by GNRI (AUC: 0.619, 95% CI: 0.581-0.656) and PNI (AUC: 0.615, 95% CI: 0.578-0.653). Fallers also exhibited greater frailty, poorer physical performance, and higher medication burden (all p < 0.001).

In older adults, poorer nutritional status assessed by MNA-SF, GNRI, and PNI was independently associated with falls, supporting the potential role of nutritional assessment within a broader fall-risk evaluation framework. However, the discriminative performance of all indices was low, indicating that these tools should not be solely considered predictors of falls, but rather complementary markers. In exploratory analyses, MNA-SF showed differences between single and recurrent fallers, which suggests a potentially relevant association that warrants prospective validation. Variability in cut-off definitions across nutritional indices remains an important limitation, underscoring the need for prospective studies to refine clinically relevant thresholds.