BACKGROUND: Primary prevention of hypertension depends on early detection of high-risk individuals. Several useful risk-stratification tools are provided by lipid-glucose-insulin obtained from standard blood testing. This study compared the predictive performance of four indices for incident hypertension: Atherogenic Index of Plasma (AIP), Triglyceride-Glucose (TyG) index, Cholesterol-HDL-Glucose (CHG) index, and Metabolic Score for Insulin Resistance (METS-IR). METHODS: A total of 1079 adults with initially normal blood pressure were enrolled between January 2016 and January 2019 and followed up until January 2024. Participants were classified according to incident hypertension status during follow-up, the participants were divided into the hypertension group and the non-hypertension group. Multivariable Cox proportional hazards models were used to assess the associations between the four insulin resistance indices and the incidence of hypertension, adjusting for a comprehensive set of clinical and biochemical confounders. Time-dependent receiver operating characteristic (ROC) curve analysis was used to assess predictive accuracy at 12, 36, and 60 months. RESULTS: Among the participants, 297 developed hypertension during follow-up. In the fully adjusted model, only the CHG index maintained a statistically independent association with hypertension risk. Each unit increase in the CHG index was associated with a more than twofold higher risk of hypertension (Hazard Ratio [HR] 2.186, 95% Confidence Interval [CI] 1.269-3.766). However, the AIP, TyG index, and METS-IR index showed no statistical significance after final adjustment. According to time-dependent ROC analysis, both the CHG index and METS-IR showed excellent and consistent discriminative capacity across initial, medium, and long-term follow-up (AUCs consistently >0.71). At the follow up time of 36 and 60 months, the CHG index exhibited the highest AUC values. The TyG index showed intermediate predictive performance, whereas AIP consistently exhibited the lowest predictive power at all intervals. CONCLUSION: The CHG index demonstrated the most independent association with incident hypertension, better than the AIP, TyG, and METS-IR indices in predictive accuracy.