# Association of HDL‐C Levels with Stroke and All‐cause Mortality: A Prospective Cohort Study from the CHARLS

## TL;DR

Higher HDL-C levels are linked to lower stroke risk in older Chinese adults, but not to overall mortality.

## Contribution

This study identifies HDL-C as a protective factor for stroke in specific subgroups and reveals BMI's mediating role.

## Key findings

- HDL-C remained a protective factor for stroke after adjusting for confounders.
- BMI mediated 25% of the effect of HDL-C on stroke risk.
- No significant association was found between HDL-C and all-cause mortality.

## Abstract

Stroke has always been a huge medical challenge faced by China. Previous studies have found a negative correlation between high‐density lipoprotein cholesterol (HDL‐C) and stroke, but in recent years, with unsatisfactory drug experiments, researchers have begun to re‐examine this relationship. Our research is based on the large population of middle‐aged and elderly people in China, exploring the relationship between HDL‐C and stroke and all‐cause mortality.

The participants of this study were from the China Health and Retirement Longitudinal Study (CHARLS) database. We utilized univariate/multivariate logistic regression to screen the influencing factors of stroke. Using Cox regression, we calculated the hazard ratio (HR) for stroke and mortality, presenting our findings through Kaplan–Meier (KM) curves. We also presented the relationship between HDL‐C and stroke and mortality by restricted cubic spline (RCS) regression analysis. A time‐varying Cox analysis was conducted to evaluate the impact of changes in HDL‐C on various outcomes. To strengthen the robustness and causal interpretation, we performed subgroup analysis, sensitivity analysis, and mediation pathway, and established a competing‐risk model.

HDL‐C was negatively correlated with stroke. After adjusting for confounders, HDL‐C remained a protective factor for stroke (HR 0.99, 95% CI 0.99–1.00, P < 0.05), especially in the male group, the normal BMI group, the group without dyslipidemia, and the group with hypertension in the subgroup analysis. However, no significant association was observed between HDL‐C and all‐cause mortality after adjustment (HR 1.00, 95% CI 1.00–1.01, p = 0.482). The results remained robust in sensitivity analysis and competitive risk models. The impact of HDL‐C on the risk of stroke (HR 1.00, 95% CI 0.999–1.001, p = 0.887) and all‐cause mortality (HR 1.00, 95% CI 0.9995–1.001, p = 0.539) did not show a significant trend over time. Mediation analysis revealed that BMI mediated 25% of the effect of HDL‐C on stroke.

In Chinese people over 45 years old, HDL‐C was a protective factor for stroke, while it had no significant association with all‐cause mortality. BMI mediated the effect of HDL‐C on stroke. This research highlights the critical role of HDL‐C in stroke risk assessment.

In Chinese people over 45 years old, HDL‐C was a protective factor for stroke, while it had no significant association with all‐cause mortality.