Social connections and risk of cardiovascular disease: a 25-year population-based cohort study

Abstract Background Although the associations of social isolation and loneliness with cardiovascular disease (CVD) risk have been extensively studied, evidence that simultaneously evaluates multiple dimensions of social connection remains limited. We aimed to examine the associations of measures within the quality, functional, and structural components of social connection, as well as their interactions, with CVD risk. Methods This study included 24,863 participants (mean age: 57.7 years; 60.9% women) from the Malmö Diet and Cancer cohort. Twenty-five measures of social connection were assessed at baseline using a self-administered questionnaire. Incident CVD was identified through national registers. Cox proportional hazards models estimated hazard ratios (HRs) and 95% CIs. Results During 25 years of follow-up, 5,825 incident CVD cases were documented. Within the quality component, loneliness (often vs. never, HR:1.19; 95% CI: 1.04–1.37) and lower solidarity with relatives (no vs. very strong, HR: 1.18; 95% CI: 1.05–1.34) were associated with higher CVD risk, whereas no associations were observed for solidarity with colleagues, a society/club, and neighbourhood. Measures within the functional component were not associated with CVD risk. For structural measures, attending the theatre/cinema (HR: 0.87; 95% CI: 0.83–0.92), art exhibitions (HR: 0.92; 95% CI: 0.87–0.97), nightclubs/dance halls (HR: 0.93; 95% CI: 0.88–0.98), and family gatherings (HR: 0.93; 95% CI: 0.89–0.98) were associated with lower CVD risk, while living alone was associated with higher risk (HR: 1.16; 95% CI: 1.09–1.23). The increased CVD risk associated with loneliness was observed only among individuals who did not attend family gatherings (HR: 1.25; 95% CI: 1.16–1.35; false discovery rate [FDR]-adjusted P interaction = 0.015). Conclusions Loneliness and low solidarity with relatives were associated with higher CVD risk, whereas engagement in specific social activities was associated with lower risk. Lonely individuals who concurrently did not attend family gatherings may represent a high‑risk population.