Association between the weight-adjusted waist index and incident falls among middle-aged and older adults: a longitudinal study based on CHARLS

This study aimed to investigate the longitudinal association between the weight-adjusted waist index (WWI) and the risk of incident falls among middle-aged and older adults. Data were obtained from the China Health and Retirement Longitudinal Study (CHARLS). Participants aged ≥ 45 years who reported no falls at the 2011 baseline and had valid WWI and follow-up fall information were included. The first reported fall during follow-up in 2013, 2015, 2018, or 2020 was defined as an incident fall; participants who remained fall-free were followed until the last survey wave with valid fall information. Participants were categorized into Q1-Q4 according to quartiles of baseline WWI. Missing baseline covariates were handled using multiple imputation by chained equations, generating 50 imputed datasets. Discrete-time survival models with a complementary log-log link were used to examine the association between baseline WWI and incident falls, with hazard ratios (HRs) and 95% confidence intervals (CIs) reported. Trend tests were performed. Restricted cubic splines were used to assess the dose-response relationship between continuous WWI and incident falls. Subgroup analyses with interaction tests were conducted, and complete-case analysis was used to assess the robustness of the findings. A total of 9,492 participants were included, of whom 3,502 (36.9%) first reported a fall during the observed follow-up. In the fully adjusted discrete-time survival model, participants in Q4 had a 13.3% higher risk of incident falls than those in Q1 (HR = 1.133, 95% CI: 1.019–1.260, P = 0.021), whereas the associations for Q2 (HR = 0.982, 95% CI: 0.889–1.086, P = 0.729) and Q3 (HR = 1.099, 95% CI: 0.995–1.213, P = 0.064) were not statistically significant. The trend test was significant (P for trend = 0.006). Restricted cubic spline analysis showed a significant overall association between WWI and incident fall risk (P for overall = 0.009), with no evidence of significant nonlinearity (P for nonlinearity = 0.882). No significant interactions were observed for age, sex, smoking, drinking, marital status, educational level, residence, or BMI (all P for interaction > 0.05). The direction and magnitude of the association in the complete-case analysis were broadly consistent with those in the primary multiple-imputation analysis, supporting the robustness of the findings. Higher baseline WWI was associated with a higher risk of incident falls during follow-up, with the association most evident in the highest WWI quartile. This study provides further epidemiological evidence for a longitudinal association between WWI and incident fall risk among middle-aged and older adults.