Triglyceride is an independent predictor of type 2 diabetes among middle-aged and older adults: a prospective study with 8-year follow-ups in two cohorts

Background Although there is abundant evidence indicating the connection between triglyceride and type 2 diabetes mellitus (T2DM), few reports or cohort studies confirm that high TG concentration may predict the incidence of T2DM independently. Thus, we studied the association between triglyceride (TG) and T2DM in a male-dominated, middle and older aged cohort, Tianjin General Hospital Cohort. And we further verified our results in the China Health and Retirement Longitudinal Study (CHARLS). Methods We conducted an 8-year retrospective cohort study (2009–2017) with 7241 participants who were free from T2DM at baseline. Three groups were constructed based on baseline TG levels (normal, borderline-high, and high). We used a Cox proportional hazards model to evaluate the relationship between TG and T2DM after adjusting for possible risk factors. A Kaplan–Meier survival analysis was performed to compare the incidence of T2DM among subjects in each TG group. We also tested the association between TG and T2DM in the CHARLS cohort. Results In Tianjin General Hospital Cohort, 7241 participants (male 75.8%, female 24.2%) were included, mean age was 61.49 ± 13.85 years at baseline. The cumulative incidence of T2DM in our cohort study was 8.6% (9.2% in men and 6.6% in women). Compared with the normal TG group, the hazard ratios in the borderline and high group were 1.30 (95% CI 1.04–1.62) and 1.54 (95% CI 1.24–1.90). The Kaplan–Meier survival analysis indicated that higher TG levels may predict higher onset of T2DM. These results were verified in the CHARLS cohort, the hazard ratio with T2DM (95% CI ) for logTG was 3.94 (2.64–5.87). Conclusions Our findings suggest that the TG level may be an independent risk factor and predictor for T2DM.