Background: Depressive symptoms in the elderly are associated with an increased Alzheimer's disease (AD) risk. We sought to determine whether the association between depressive symptoms and AD is explained by a history of vascular risk factors and stroke.
Methods: Five hundred and twenty-six elderly persons from New York City without dementia at baseline were followed for a mean of 5 years. Depressive symptoms were assessed using the 17-item Hamilton Depression Rating Scale (HAM). Incident AD was ascertained using standard criteria. Diabetes, hypertension, heart disease, current smoking and stroke were ascertained by self-report. Proportional hazards regression was used to relate HAM scores to incident AD.
Results: HAM scores were higher in persons with hypertension, heart disease, and stroke, which in turn were related to higher AD risk. AD risk increased with increasing HAM scores as a continuous logarithmically transformed variable (HR for one point increase=1.4; 95% CI=1.1,1.8) and as a categorical variable (HR for HAM >or= 10=3.4; 95% CI=1.5,8.1; p for trend=0.004 with HAM=0 as the reference). These results were virtually unchanged after adjustment for vascular risk factors and stroke, individually (HR for HAM >or= 10=3.4; 95% CI=1.5,8.1; p for trend = 0.004), and in a composite measure (HR for HAM >or= 10=3.0; 95% CI=1.2,7.8; p for trend=0.02).
Conclusion: The prospective relation between depressive symptoms and AD is not explained by a history of vascular risk factors and stroke, suggesting that other mechanisms may account for this association.