Journal of Obesity · Nov 16, 2026

Oxidative Stress and Inflammation in the Association Between Body Composition Indices and Frailty: A Multinational Population‐Based Cross‐Sectional Analysis

By Li J, Li X, Lin Z et al.

Background: Associations between body composition and frailty have been reported, but the comparative discriminatory performance of traditional and novel anthropometric indices remains uncertain. Inflammation and oxidative stress may be involved in these observed associations, although cross‐sectional data cannot establish temporal or causal mechanisms. Methods: We performed a multinational population‐based cross‐sectional analysis using survey data from CHARLS, ELSA, and NHANES. Frailty was defined using a cumulative deficit frailty index (FI > 0.25). ROC analysis was used to compare the discriminatory performance of BMI, WC, ABSI, BRI, CI, and WWI for frailty. WWI and BRI were carried forward for multivariable logistic regression, restricted cubic spline, subgroup, sensitivity, and exploratory statistical mediation/decomposition analyses. Modified FI sensitivity analyses were conducted in NHANES to reduce potential overlap between anthropometric or inflammation‐related components and the FI outcome. Results: Among 24,906 participants, WWI showed the highest AUC in CHARLS (0.620), whereas BRI showed the highest AUCs in ELSA and NHANES (0.670 and 0.643, respectively). Although WWI and BRI showed relatively higher discriminatory performance than BMI and WC, all AUC values were modest. In fully adjusted models, higher WWI was associated with higher odds of frailty in CHARLS, ELSA, and NHANES, with ORs of 1.19, 1.61, and 1.45, respectively. Higher BRI was also positively associated with frailty, with corresponding ORs of 1.10, 1.31, and 1.58. Modified FI sensitivity analyses in NHANES showed directionally consistent but attenuated associations after excluding overlapping FI components. Exploratory statistical mediation/decomposition analyses in NHANES suggested statistical indirect effects involving SIRI and UHR. Conclusions: WWI and BRI were positively associated with frailty and showed relatively higher but modest discriminatory performance compared with traditional anthropometric indices. The observed indirect effects involving SIRI and UHR should be interpreted as hypothesis‐generating findings rather than evidence of causal mechanisms. Longitudinal studies and external validation are needed to clarify temporal relationships, incremental predictive value, and potential clinical utility.

doi.org/10.1155/jobe/9155742Open accessBiomarkers

AgingResearch.News