Background Being married is known to decrease the risk of all cause and cardiovascular mortality for men, but it is not clear if marriage confers a similar benefit for women. In addition, few studies have examined marital status in relation to both mortality and disease incidence to determine whether the beneficial effects of marriage influence the aetiology or prognosis of disease.

Methods A total of 747,532 women (mean age: 59.8, SD 4.9), without previous heart disease, stroke or cancer, were followed prospectively for an average of 6.1 years for hospital admissions and deaths. We used Cox regression models to calculate relative risks for the first ischaemic heart disease (IHD) event and for IHD deaths comparing married women to unmarried women, adjusting for age, region of residence, socio-economic status, health-related behaviours, social support, depression and self-rated health. We also calculated relative risk of subsequent death in participants whose first IHD event was a hospital admission (n = 19,506, mean follow-up 2.7 years).

Results At baseline, 2000-2003, 81% of the sample was married or living with a partner. Over follow-up, there were 20,458 first IHD events (19,572 non-fatal and 886 fatal), and among those whose first event was a hospital admission there were 407 deaths. Married women had lower rates of smoking and physical inactivity, were less deprived, and less likely to rate their health as poor. In women who had no prior hospital admissions for IHD, married women had a 34% lower risk of IHD death as a first event than unmarried women (p < 0.001), but no significantly decreased risk of non-fatal hospital admission for IHD. Among the women whose first IHD event was a (non-fatal) hospital admission, married women did not have a significant reduction in subsequent mortality.

Conclusion Being married or living with a partner appears to reduce the risk of death from IHD for women, which is not fully explained by health-related behaviours or by socio-economic status.