Background: Stressful life events (SLEs) in adulthood are potential risk factors for cardiovascular disease (CVD), yet evidence remains inconsistent. This study examined the association between adulthood SLEs and incident CVD and evaluated mediation by depression, physical inactivity, and smoking.
Methods: We analyzed harmonized data from two nationally representative aging cohorts: the US Health and Retirement Study and the English Longitudinal Study of Ageing. SLEs were assessed at baseline, and incident CVD was physician-diagnosed. We used Cox and restricted mean survival time (RMST) regression to estimate hazard ratios (HRs) and RMST differences, set at the time point when 90% of incident CVD events had accrued (7.8 years). Population attributable fractions (PAFs) based on RMST and mediation analyses via RMST pseudo-value regression quantified contributions of SLEs and behavioral pathways.
Results: Among 18,898 participants without baseline cardiovascular disease (mean age 64.5 years; 39.8% male), 2,782 incident CVD cases were documented (incidence rate: 2.51 per 100 person-years). RMST analysis showed that any SLE exposure was related to significant reductions in event-free survival, including a 2.22-month decrease for CVD (95% CI: -2.90 to -1.53), a 1.57-month decrease for heart disease (95% CI: -2.16 to -0.99), and a 0.79-month decrease for stroke (95% CI: -1.16 to -0.41), with a clear dose-response relationship. Corresponding Cox proportional hazards models showed adjusted HRs of 1.20 (95% CI: 1.11–1.30) for CVD, 1.16 (95% CI: 1.07–1.27) for heart disease, and 1.26 (95% CI: 1.09–1.46) for stroke. Each additional SLE increased CVD, heart disease, and stroke risks by 11%, 10%, and 11%. PAFs attributable to SLEs exposure were 2.6% for CVD, 1.7% for heart disease, and 0.8% for stroke. Mediation analyses revealed that depression, physical inactivity, and current smoking accounted for 7.2%, 4.5%, and 4.5% of the total effect of SLEs on CVD, respectively.
Conclusions: Adulthood SLEs are independently associated with increased CVD risk, with depression, physical inactivity, and smoking explaining a modest proportion of this association.
Supplementary Information: The online version contains supplementary material available at 10.1186/s12916-026-04890-0.
Keywords: cardiovascular disease; heart disease; mediation analysis; population-attributable fraction; stressful life events; stroke.