Marital status and indices of social support are associated with mortality due to coronary heart disease and stroke. This association seems not entirely due to differences in recognised cardiovascular disease risk factors. The Western Sydney Stroke risk in the Elderly Study examined the relationship between marital status, living arrangements, widowhood and extent of social support, and risk factors for cardiovascular disease in men and women aged over 65 years. Unmarried men had the lowest mean HDL-C levels. Men living alone had the highest mean systolic blood pressures. The lower mean HDL-C levels and higher DBP levels seen among widows were not statistically significant after adjustment for differences in past medical history and education levels. The extent of social support was not associated with any significant differences in cardiovascular risk factor levels among men or women. We conclude that some of the increased risk of cardiovascular disease associated with socio-demographic factors among men in this age-group may be due to differences in primary cardiovascular disease risk factors. However, some of the mechanisms underlying risk of cardiovascular disease in this age-group remains obscure.