Relationship between the cholesterol, high-density lipoprotein, and glucose (CHG) index and incident impaired fasting glucose in normoglycemic Chinese adults: a retrospective cohort analysis

Among Chinese health-screening populations, impaired fasting glucose (IFG) is the most common prediabetes subtype. Conventional fasting plasma glucose (FPG) cutoffs may miss individuals with subtle metabolic dysfunction, particularly younger adults and those with normal body weight. The Cholesterol, High-Density Lipoprotein, Glucose (CHG) index, derived from routine lipid and glucose measurements, may help identify metabolic dysfunction. However, whether its association with incident IFG varies among demographic subgroups remains unclear. This secondary analysis included 108,152 Chinese adults with normoglycemia at baseline from the Rich Healthcare Group cohort (2010–2016). The CHG index was calculated as ln{[total cholesterol (TC) × fasting blood glucose (FBG)]/[2 × high-density lipoprotein (HDL)]}. Incident IFG was defined as an FPG level of 6.1–6.9 mmol/L at any follow-up assessment. We used Cox proportional hazards models to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) and restricted cubic splines to assess dose-response relationships. Multiplicative and additive interactions were assessed according to sex, age, and body mass index (BMI). Over a median follow-up of 3.1 years, 4,021 participants developed IFG (3.72%). Each 1-unit increase in the CHG was associated with a higher incidence of IFG (HR = 2.57, 95%CI, 2.29–2.88; P < 0.001), with a graded association across quartiles (Q4 vs. Q1: HR = 3.05, 95%CI, 2.67–3.50). Restricted cubic spline analysis indicated a nonlinear association ( P < 0.001; inflection point approximately 5.047). The relative association was stronger among women, younger adults, and individuals with a BMI of 18.5–23.99 kg/m². Multiplicative interactions were significant for sex, age, and BMI (all P < 0.001); positive additive interactions were observed for sex (relative excess risk due to interaction [RERI], 0.47, 95%CI: 0.29–0.64) and age (RERI, 0.90, 95%CI, 0.49–1.32) but not for BMI. Adding TC and HDL to FPG improved discrimination (area under the curve, 0.781 vs. 0.772; P < 0.001), and findings were robust in an analysis using the composite outcome (HR = 2.40, 95%CI, 2.18–2.65). The CHG index is associated with incident IFG, with stronger relative association among women, younger adults, and normal-weight individuals. As a measure derived from routine lipid and glucose measurements, the CHG index may complement FPG for early IFG risk stratification.