Risk of type 2 diabetes mellitus associated with plasma lipid levels: The rural Chinese cohort study

Abstract Aim To investigate the association of type 2 diabetes mellitus (T2DM) risk and plasma lipid levels in rural Chinese. Methods Each lipid variable was divided into quartiles and dichotomized by clinical cutoff points. Cox proportional-hazards model was used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for the association of T2DM risk and plasma lipid levels and explore the interaction between plasma lipid levels and other risk factors. Results 11,929 participants were included in the analysis. We documented 720 incident cases of T2DM over 70,720.84 person-years of follow-up, for an incidence of 10.18/1,000 person-years. In the multivariable-adjusted model, risk of T2DM was increased with the highest versus lowest quartiles of total cholesterol (TC) and triglycerides (TG) levels and TC/high-density lipoprotein-cholesterol (HDL-C) and TG/HDL-C ratios. The HRs (95% CIs) for the fourth quartiles, for example, were 1.34 (1.03–1.74), 2.32 (1.73–3.13), 1.66 (1.23–2.25), and 1.84 (1.38–2.45), respectively. In addition, risk of T2DM was increased with high TG level and TC/HDL-C and TG/HDL-C ratios by clinical cutoffs. The HRs (95% CIs) were 1.50 (1.25–1.80), 1.24 (1.03–1.48), and 1.44 (1.18–1.75), respectively. Risk of T2DM was associated with interactions between all lipid variables and age and BMI. TG level and TG/HDL-C ratio additionally interacted with gender (all P interaction Conclusions Risk of T2DM was increased with elevated serum levels of TC and TG and TC/HDL-C and TG/HDL-C ratios and also with interactions between high TC and TG levels and TC/HDL-C and TG/HDL-C ratios and age and BMI in a rural Chinese population.