The purpose of this study was to test the hypothesis that educational attainment and poverty index are inversely associated with incidence of stroke in blacks and in whites. The NHANES I Epidemiologic Follow-up Study data were analyzed. We analyzed 2953 women and 2661 men with no history of stroke before baseline (1971-1975), using the incidence of stroke through 1992, years of education, and poverty index at baseline. In white men aged 45 to 74, Cox regression models showed an inverse age-adjusted association with education that did not attain statistical significance. In white women, those with 12 or more years of education had significantly lower age-adjusted risk of stroke compared with those with less than 8 years. A test for linear trend was significant when adjusting for age (P = 0.0005). In blacks, stroke risk was significantly lower in those with 8 or more years of education than in those with <8 years in adjusted models. In each group, Cox regression models showed an inverse, graded, age-adjusted association with poverty index that attained statistical significance. After controlling for multiple confounders and risk mediators, the association was diminished and nonsignificant.
Published by Elsevier Science Inc.