Honig et al1 reported an association between stroke and subsequent onset of Alzheimer disease (AD) in the Washington Heights-Inwood Columbia Aging Project (hazard ratio, 1.83; 95% confidence interval, 1.14-2.95). Data from the Cache County Study2 may support these findings.
Among 3167 persons aged 65 years and older who did not have dementia initially and underwent follow-up for 3 years, we evaluated the incidence of AD in relation to the baseline self- or proxy-reported history of stroke. Table 1 presents odds ratios and 95% confidence intervals for AD onset with stroke. Additional models evaluate this relationship, controlling for important covariates and interaction terms considered by Honig and colleagues. Our analysis (using logistic regression) yielded an adjusted odds ratio of 2.03 for AD with self- or proxy-reported stroke controlling for age, sex, education, hypertension, diabetes, and myocardial infarction or coronary artery bypass graft. Whereas this estimate was similar to the estimate from the Washington Heights-Inwood Columbia Aging Project,1 the confidence interval for our estimate was broader (0.80-4.46). Two factors account for the difference in precision between the 2 estimates. Honig and colleagues had twice the number of stroke events (188 vs 98) and AD cases (212 vs 98) in their sample.