Educational and homeownership inequalities in stroke incidence: a population-based longitudinal study of mid-aged women

Background: We aimed to determine which socioeconomic status measures are associated with stroke risk in mid-aged women and assess the contribution of lifestyle, biological and psychosocial factors to observed associations. Methods: We included women born in 1946-51 from the Australian Longitudinal Study on Women's Health, who were surveyed every 3 years. Using generalized estimating equation analysis, we determined the association between socioeconomic status and stroke at the subsequent survey, adjusting for time-varying covariates. For significant associations, we calculated the contribution of individual mediating factors in explaining these associations. Results: Among 11 468 women aged 47-52 years, 177 strokes occurred during a 12-year follow-up. Education (odds ratio lowest vs. highest 2.45, 95% confidence interval: 1.40-4.30) and homeownership, but not occupation or managing on income, were significantly associated with stroke. After full adjustment, the overall association between education and stroke was non-significant. Lifestyle (smoking, exercise, alcohol and body mass index), biological (hypertension, diabetes, heart disease and hysterectomy/oophorectomy) and psychosocial (depression and marital status) factors explained 38% of the association in the lowest versus highest education groups. Lifestyle and biological factors together accounted for 34%. Mediators accounted for 29% of the association between homeownership and stroke, with lifestyle and psychosocial factors responsible for most of this attenuation. However, a significant association remained in fully adjusted models (odds ratio non-homeowner vs. homeowner 1.63, 95% confidence interval: 1.12-2.38). Conclusions: Lower education level is associated with increased stroke risk in mid-aged women, and is partially mediated by known risk factors, particularly lifestyle and biological factors. Non-homeownership is associated with increased stroke risk, but the underlying mechanism is unclear.