Association Between the Cholesterol-High-Density Lipoprotein-Glucose Index and Urinary Incontinence: A Prospective...

Objective: This study examined whether the cholesterol-high-density lipoprotein-glucose (CHG) index was associated with subsequent urinary incontinence (UI) among middle-aged and older adults and evaluated its contribution to 4-year UI risk prediction. Methods: A total of 2059 participants without UI at baseline were included and followed from Wave 2 to Wave 7 in this prospective cohort study. The relationship between the CHG index and subsequent UI was examined using Cox proportional hazards models, supplemented by restricted cubic spline and subgroup analyses. A 4-year prediction model for UI was developed using LASSO and Cox regression. Results: UI occurred in 308 participants during follow-up, yielding an incidence of 14.96%. An independent inverse association was observed between CHG levels and the risk of subsequent UI. After full adjustment, each 1-standard deviation increase in CHG corresponded to an 18% lower risk of UI (HR of 0.82, 95% CI from 0.72 to 0.93). Relative to the lowest CHG tertile, the hazard ratios were 0.68 (95% CI from 0.51 to 0.90) for T2 and 0.63 (95% CI from 0.47 to 0.85) for T3. The relationship was predominantly linear, and no significant effect modification was detected in subgroup analyses. The 4-year prediction model showed moderate performance (AUC 0.663), while incorporation of CHG led to modest improvements in discrimination and reclassification. Conclusions: Among middle-aged and older adults, higher CHG levels were independently related to a reduced risk of UI. The CHG index may provide complementary information for early risk assessment in this population, but should not be regarded as a stand-alone predictor.