Background: The triglyceride–glucose (TyG) index has been proposed as a reliable surrogate marker of insulin resistance. Correlation with mortality has produced inconsistent results in studies involving different populations. The present study used data from a large prospective cohort study to evaluate the association of TyG index with all-cause and cause-specific mortality and potential influencing factors, particularly exploring associations and correlations with mortality of thresholds in different subgroups of the population.

Methods: A total of 240,258 participants from the UK Biobank were included for analysis. Kaplan–Meier curves and Cox proportional hazards models were applied to estimate the associations between the TyG index and all-cause and cause-specific mortality. Subgroup analyses examined the closeness of associations in different populations. Dose-response relationships between TyG index and different risks of death in different subgroups were analyzed using restricted cubic spline (RCS), and risk thresholds were identified. Interaction tests and subgroup analyses explored potential influences.

Results: The TyG index exhibited a significant positive association with all-cause, cardiovascular, oncologic, diabetic and kidney-related mortality. A J-shaped relationship was observed between TyG index and mortality in diabetes and kidney disease. The relationship was stronger in females, subjects aged <65 years, with abdominal obesity, smokers, with alcohol abuse, HbA1c ≥7%, prediabetes/diabetes, no hypertension and no hyperlipidemia. In addition, there were different patterns of dose-response relationships between TyG index and mortality in different subgroups of the population. We identified thresholds for each association.

Conclusion: Higher TyG index was associated with higher all-cause, oncologic, cardiovascular, diabetic and kidney-related mortality. There were differences in TyG threshold for different causes of death in different subgroups. The study identified thresholds for each association, providing more precise evidence for risk stratification and clinical decision-making.