Association of the cholesterol-high-density-lipoprotein-cholesterol-glucose (CHG) with incident hip fracture risk among middle-aged and older Chinese adults: a prospective cohort study from CHARLS

Background Hip fracture imposes a substantial public health burden among older adults. However, integrated metabolic biomarkers remain underexplored as risk stratification tools. The cholesterol-high-density-lipoprotein-cholesterol-glucose (CHG) index consolidates lipid and glucose parameters into a single score. Objective To prospectively examine the association between the CHG index and incident hip fracture in Chinese adults aged 45 years and older. Methods This prospective cohort study used data from the China Health and Retirement Longitudinal Study (CHARLS, 2011–2020). The CHG index was calculated as Ln[total cholesterol × fasting blood glucose / (2 × high-density-lipoprotein cholesterol)]. Incident hip fractures were ascertained through biennial follow-up assessments. Cox proportional-hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). Fine-Gray competing-risk models adjusting for all-cause death were used to confirm the robustness of findings; the CHG index was treated both as a continuous variable (per 1-unit increment) and in quartiles. Time-dependent receiver-operating-characteristic (ROC) analysis was performed to assess predictive performance. Results Over a median follow-up of 8.2 years (range: 0.5–9.0 years), complete-case analysis (7,886 participants, 321 events) showed that each 1-unit increase in the CHG index was associated with a 33.3% lower hip fracture risk (HR = 0.667, 95% CI: 0.492–0.906, p = 0.009) in the fully adjusted model. The highest quartile (Q4) exhibited a 37.0% risk reduction (HR = 0.64, 95% CI: 0.46–0.89, p = 0.009). Fine-Gray competing-risk analyses (accounting for all-cause death) yielded consistent results with the primary Cox model (fully adjusted SHR = 0.664, p = 0.009), confirming the robustness of the findings. Kaplan-Meier curves showed a significant dose-response separation (log-rank p = 0.001). The full model achieved a time-dependent area under the curve (AUC) of 0.647, with a modest improvement over the base model without the CHG index. This association was independent of demographic, lifestyle, and clinical covariates. Age, lower education, alcohol use, and disability were independent predictors. Conclusion A higher CHG index is independently associated with lower hip fracture risk in older adults, but exhibits limited incremental predictive performance beyond conventional risk factors, and these findings warrant validation in diverse populations. This low-cost, routinely available metric may support community-based fracture risk stratification in resource-limited settings, contributing to healthier ageing and more equitable access to preventive care for older populations.