Association between conicity index and frailty in older Americans: the NHANES cross-sectional study, 2007-2018.

This study utilized NHANES data from 2007 to 2018 to investigate the correlation between frailty and the Conicity Index (CI) in individuals aged 60 and above in the United States. The study used NHANES data from 2007 to 2018.CI was calculated as CI = wc / [0.109 × sqrt(bw / Height)]. Frailty was assessed by the frailty index (≥ 0.25). Weighted multivariate logistic regression analysis, subgroup analyses, and interaction tests were used to investigate the connection between CI and the prevalence of frailty. Generalized additive modeling (GAM) was employed to address any non-linear patterns, and the predictive capability of CI for frailty was evaluated by receiver operating characteristic (ROC) analysis. With a 69% rise in the prevalence of frailty for every 0.1 unit increase in the fully adjusted model, the results demonstrated a strong and positive relationship between CI and frailty prevalence (OR: 1.69, 95% CI: 1.53,1.86; P < 0.001). When CI was categorized, the group with the highest CI had a significantly higher prevalence of frailty than the group with the lowest CI (OR = 2.79, 95% CI: 2.22,3.51; P < 0.001). The association between CI and prevalence of frailty was significant in all subgroups. In addition, statistically significant interactions were present in most subgroups. When the CI > reached 1.35, the GAM model demonstrated a threshold effect and a significant nonlinear connection, with a 105% rise in the prevalence of frailty for every 0.1 unit increase in CI. In the male group, CI was a significantly greater indicator of the prevalence of frailty than both BMI and WC. According to this study, frailty in older persons is substantially correlated with a higher CI. Although greater confirmation in large-scale prospective research is required, this study indicates that increased CI is a more reliable predictor of the prevalence of frailty in older men and is significantly linked with its occurrence.