Education, sex and risk of stroke: a prospective cohort study in New South Wales, Australia

OBJECTIVE: To determine whether the association between highest educational attainment and stroke differed by sex and age; and whether potential mediators of observed associations differ by sex. DESIGN: Prospective cohort study. SETTING: Population based, New South Wales, Australia. PARTICIPANTS: 253 657 stroke-free participants from the New South Wales 45 and Up Study. OUTCOME MEASURES: First-ever stroke events, identified through linkage to hospital and mortality records. RESULTS: During mean follow-up of 4.7 years, 2031 and 1528 strokes occurred among men and women, respectively. Age-standardised stroke rate was inversely associated with education level, with the absolute risk difference between the lowest and highest education group greater among women than men. In relative terms, stroke risk was slightly more pronounced in women than men when comparing low versus high education (age-adjusted HRs: 1.41, 95% CI 1.16 to 1.71 and 1.25, 95% CI 1.07 to 1.46, respectively), but there was no clear evidence of statistical interaction. This association persisted into older age, but attenuated. Much of the increased stroke risk was explained by modifiable lifestyle factors, in both men and women. CONCLUSION: Low education is associated with increased stroke risk in men and women, and may be marginally steeper in women than men. This disadvantage attenuates but persists into older age, particularly for women. Modifiable risk factors account for much of the excess risk from low education level. Public health policy and governmental decision-making should reflect the importance of education, for both men and women, for positive health throughout the life course.