Research Article: Non-linear relationship between triglyceride glucose-body mass index and risk of diabetes in adults: a general population-based cohort study of Chinese adults using a publicly available DRYAD dataset
Abstract:
Despite triglyceride glucose-body mass index (TyG-BMI) being a validated marker of insulin resistance, its prospective association with diabetes risk remains unclear; we therefore examined the impact of TyG-BMI on incident diabetes.
This was a retrospective cohort study based on 48,288 adults without diabetes at baseline, identified from a publicly available dataset on the DRYAD platform originally derived from 11 cities and 32 locations in China.Using Cox proportional risk regression modeling combined with cubic spline functions and smoothed curve fitting, we assessed the relationship between the baseline TyG-BMI and the risk of developing DM and explored its nonlinear association. We performed subgroup analyses to assess the consistency of the association across different subgroups.
Among 48,288 initially diabetes mellitus (DM)-free participants, 1,230 (2.54%) developed diabetes at follow-up.Adjusted for covariates, TyG-BMI levels(per 10-unit) were positively associated with the risk of diabetes onset (HR = 1.25,95% CI: 1.20-1.29, P < 0.001). The risk of incident diabetes increased progressively across quartiles of TyG-BMI levels (Q1 to Q4), with a significantly higher risk in Q4 compared to Q1 (adjusted HR = 11.24,95% CI = 5.05-24.99).Furthermore, a threshold effect of TyG-BMI on DM risk was found, with a threshold of 163.38. The HR to the right of the inflection point was 1.025 (95% CI: 1.020-1.029). When TyG-BMI was <163.38, the association was not statistically significant(HR = 0.928; 95% CI: 0.843–1.022), suggesting that the relationship may be absent or even slightly inverse.
In Chinese adults, TyG-BMI exhibits a threshold effect on incident diabetes. Individuals above the threshold have a significantly increased risk of developing diabetes. However, the identified threshold (163.38) requires prospective validation in contemporary cohorts. If validated, it may help identify high-risk individuals for targeted prevention in the future.
No summary available.