Background: Poor social health is associated with increased risk of cardiovascular disease (CVD). Recent research suggests that different social health domains should be considered separately as the implications for health and possible interventions may differ.

Aim: To assess social isolation, low social support and loneliness as predictors of CVD.

Methods: Secondary analysis of 11,486 community-dwelling, Australians, aged 70 years and over, free of CVD, dementia, or significant physical disability, from the ASPirin in Reducing Events in the Elderly (ASPREE) trial. Social isolation, social support (Revised Lubben Social Network Scale), and loneliness were assessed as predictors of CVD using Cox proportional-hazard regression. CVD events included fatal CVD, heart failure hospitalization, myocardial infarction and stroke. Analyses were adjusted for established CVD risk factors.

Results: Individuals with poor social health were 42 % more likely to develop CVD (p = 0.01) and twice as likely to die from CVD (p = 0.02) over a median 4.5 years follow-up. Interaction effects indicated that poorer social health more strongly predicted CVD in smokers (HR 4.83, p = 0.001, p-interaction = 0.01), major city dwellers (HR 1.94, p < 0.001, p-interaction=0.03), and younger older adults (70-75 years; HR 2.12, p < 0.001, p-interaction = 0.01). Social isolation (HR 1.66, p = 0.04) and low social support (HR 2.05, p = 0.002), but not loneliness (HR 1.4, p = 0.1), predicted incident CVD. All measures of poor social health predicted ischemic stroke (HR 1.73 to 3.16).

Conclusions: Among healthy older adults, social isolation and low social support may be more important than loneliness as cardiovascular risk factors. Social health domains should be considered in future CVD risk prediction models.