Abstract
OBJECTIVE: Both triglyceride glucose index(TyG) and weight-adjusted waist circumference index (WWI) are independent predictors of diabetes onset, but the relationship between TyG index and WWI index and diabetes risk is not well understood. The aim of this study was to investigate the interaction between TyG and WWI indices on diabetes risk and to assess whether the combination of TyG and WWI indices is a better predictor of diabetes incidence. METHODS: A total of 5555 participants aged 45 years and older without diabetes were enrolled and followed up for a duration of up to 9 years according to the China Health and Retirement Longitudinal Study (CHARLS). After identifying key predictors using the least absolute shrinkage with selection operator (LASSO) technique, COX proportional risk model assessments were performed to evaluate the association between TyG index and WWI index and diabetes risk, Kaplan-Meier analyses were used for cumulative risk and mediation analyses were used to assess the mediated relationship between TyG index, WWI index and diabetes risk. In addition, subgroup analyses and sensitivity analyses were performed to investigate the predictive performance and robustness of the results in different populations. RESULTS: A total of 880 participants developed diabetes over the course of 9 years of follow-up.COX proportional risk regression modelling revealed that participants with high TyG and high WWI had the highest risk (HR = 1.59, 95% CI: 1.28-1.97), followed by participants with high TyG and low WWI (HR = 1.26, 95% CI: 1.01-1.57), and participants with low TyG and low WWI participants (HR = 1.22, 95% CI: 0.97-1.52). Subgroup analyses and sensitivity analyses further confirmed these findings. CONCLUSION: There is a complex interaction between TyG index, WWI and diabetes risk, and the combined prediction of TyG index and WWI has higher clinical value. These findings highlight the importance of considering both insulin resistance(IR) and visceral adiposity in diabetes risk assessment to improve the identification of at-risk individuals, who can be monitored and intervened on to reduce the overall burden of diabetes. CLINICAL TRIAL NUMBER: Not applicable.