Predictive value of the combined cholesterol, high-density lipoprotein, glucose and frailty indices for cardiometabolic multimorbidity incidence: evidence from a national prospective cohort study

BACKGROUND: The cholesterol, high-density lipoprotein, and glucose (CHG) is a surrogate of insulin resistance, while frailty reflects cumulative physiological decline, yet their combined utility for cardiometabolic multimorbidity (CMM) is underexplored. We evaluated a combined CHG-frailty index (CHG-FI) for incident heart disease, stroke, diabetes, and CMM. METHODS: We conducted a retrospective cohort study of 6812 Chinese adults aged ≥ 45 years enrolled in the 2011-2020 waves of the China Health and Retirement Longitudinal Study (CHARLS). Participants with baseline lipid and fasting glucose measurements were included and followed prospectively for incident heart disease, stroke, diabetes, and cardiometabolic multimorbidity (≥ 2 conditions). FI was calculated using the cumulative deficit approach, and CHG was incorporated according to established procedures. Multivariable Cox regression estimated associations; CHG-FI interaction was assessed on multiplicative and additive scales. Incremental predictive utility of CHG, FI, TYG‑FI, and CHG‑FI was compared using receiver operating characteristic (ROC) curves, net reclassification improvement (NRI), and decision curve analysis (DCA). Stratified and sensitivity analyses evaluated robustness. RESULTS: Over a median follow‑up of 9.0 years, we observed 1,304 heart disease events, 554 strokes, 932 diabetes cases, and 467 CMM cases. Each 1-unit increase in CHG-FI was associated with higher risk: heart disease (HR 1.34, 95% CI 1.23-1.45), stroke (HR 1.85, 95% CI 1.65-2.06), diabetes (HR 1.29, 95% CI 1.17-1.42), and CMM (HR 1.79, 95% CI 1.58-2.04). Associations showed dose-response patterns and nonlinearity. The multiplicative interaction between CHG and FI for CMM was 0.54 (95% CI 0.35-0.83), reflecting a "risk saturation" or ceiling effect. Despite this, the combined CHG -FI index offered superior predictive performance for CMM relative to individual components: AUC 0.652 (95% CI 0.627-0.678), with significant improvement in reclassification (NRI 0.329, 95% CI 0.236-0.423) and discrimination (IDI 0.011, 95% CI 0.008-0.015). CHG -FI and TyG -FI showed broadly comparable performance across outcomes. CONCLUSION: The combined CHG-frailty index was significantly associated with incident cardiometabolic multimorbidity and its individual components, demonstrating predictive performance comparable to established combined indices. As an accessible tool integrating routine metabolic and physiologic reserve measures, CHG-FI offers a practical alternative approach for risk stratification.