Predicting all-cause and cardiovascular mortality in hypertensive patients by combining the cholesterol-HDL-glucose (CHG) index and its obesity indicators: a prospective cohort study with survival prediction model
Authors' conclusion
The CHG index and its obesity indicators are significant predictors of all-cause and cardiovascular mortality in hypertensive patients, and CHG-WHtR exhibits better prognostic performance in terms of mortality risk.
Study context
- Population
- Hypertensive participants in the China Health and Retirement Longitudinal Study (CHARLS)
- Participants
- 4,661 participants
- Exposure
- CHG index and its obesity indicators as predictors of mortality
- Outcome
- All-cause and cardiovascular mortality
Methodological detailssurvival machine learning models (Cox, Cox-LASSO, Survival XGBoost), competing-risk (Fine-Gray) regression for cardiovascular mortality
- Survival machine learning models (Cox, Cox-LASSO, Survival XGBoost)
- Competing-risk (Fine-Gray) regression for cardiovascular mortality
Main finding
As reported in the abstract
CHG-WHtR per 1-SD increase, all-cause mortality: HR
2.060
95% CI 1.852–2.292P < 0.001
CHG-WHtR per 1-SD increase, cardiovascular mortality: HR
2.020
95% CI 1.816–2.247P < 0.001
Observational data: associations, not effects.
Further findings
Discrimination of Cox-LASSO model · 3, 5 and 7 years
- Test C-index
- 0.920 (95% CI 0.878-0.962)
- tdAUC range across 3, 5 and 7 years
- 0.851-0.916
Reclassification after adding CHG indices
- Continuous net reclassification index (NRI) range
- 0.349-0.496
- Integrated discrimination improvement (IDI) range
- 0.024-0.040
Dose-response shape
- L-shaped relationship with all-cause mortality (non-linear P < 0.001)
- L-shaped relationship with cardiovascular mortality (non-linear P < 0.001)
Original publication
Predicting all-cause and cardiovascular mortality in hypertensive patients by combining the CHG index and its obesity indicators: a prospective cohort study with survival prediction model. Cardiovasc Diabetol. Published 5 August 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →