The longitudinal bidirectional association between cardiovascular disease and depressive symptoms among middle-aged and elderly adults: evidence from a nationwide cohort study in China.

BACKGROUND

Although cardiovascular disease (CVD) and depressive symptoms frequently co-occur, their temporal directional relationship remains poorly understood. This study aimed to investigate the bidirectional association between depression and CVD in middle-aged and elderly Chinese adults through longitudinal analyses.

METHOD

This longitudinal cohort study included 12792 participants from the China Health Retirement Longitudinal Study (CHARLS), 2015-2020. Depressive symptoms were defined as a score of 10 or higher on the 10-item Center for Epidemiological Studies Depression Scale (CES-D-10). Cardiovascular disease (CVD) was defined as physician-diagnosed coronary heart disease (CHD) and/or stroke. Cox proportional hazards models were applied to investigate the longitudinal association of baseline depressive symptoms with follow-up CVD events, as well as the association of baseline CVD status with follow-up depressive symptoms. Cross-lagged panel models were performed to evaluate bidirectional associations and strength of temporal relationships simultaneously. Effect modification by demographic and lifestyle factors was also examined to identify vulnerable populations for each directional pathway.

RESULTS

Of 12792 participants, the mean (SD) age was 60.8 (8.9) years and 6833 (53.4%) were females. After adjusting for potential confounders, baseline depressive symptoms were associated with higher risks of subsequent CVD (HR=1.55, 95% CI: 1.40-1.72), CHD (HR=1.51, 95% CI: 1.34, 1.70), and stroke (HR=1.71, 95% CI: 1.43-2.04); conversely, baseline conditions of CVD (HR=1.22, 95% CI: 1.10-1.35), CHD (HR=1.20, 95% CI: 1.08-1.33) and stroke (HR=1.43, 95% CI: 1.16-1.77) were associated with increased risk of depressive symptoms at follow-up. The cross-lagged panel analysis further confirmed the bidirectional associations over time (all -values < 0.001), revealing that the standardized effect size of CVD status on depressive symptoms was greater than the effect size in the reverse direction. The directional pathway from depressive symptoms to CVD was modified by body mass index, educational level, residence, and alcohol consumption, while the reverse directional association was modified by education level and alcohol consumption.

CONCLUSION

There is a longitudinal, bidirectional association between CVD status and depressive symptoms in mid-to-late life, with CVD status emerging as a larger driving force in these dynamic interactions. These findings suggest that targeted interventions addressing either CVD or depressive symptoms may yield reciprocal benefits over time.