Objective: To test the hypothesis that late-life depression is associated with dementia-related pathology. Method: Older participants (n = 1,965) in 3 longitudinal clinical-pathologic cohort studies who had no cognitive impairment at baseline underwent annual clinical evaluations for a mean of 8.0 years (SD = 5.0). The authors defined depression diagnostically, as major depression during the study period, and psychometrically, as elevated depressive symptoms during the study period, and established their relation to cognitive outcomes (incident dementia, rate of cognitive decline). A total of 657 participants died and underwent a uniform neuropathologic examination. The authors estimated the association of depression with 6 dementia-related markers (tau tangles, beta-amyloid plaques, Lewy bodies, hippocampal sclerosis, gross and microscopic infarcts) in logistic regression models. Results: In the full cohort, 9.4% were diagnosed with major depression and 8.6% had chronically elevated depressive symptoms, both of which were related to adverse cognitive outcomes. In the 657 persons who died and had a neuropathologic examination, higher beta-amyloid plaque burden was associated with higher likelihood of major depression (present in 11.0%; OR = 1.392, 95% CI = 1.088, 1.780) but not with elevated depressive symptoms (present in 11.3%; OR = 0.919, 95% CI = 0.726, 1.165). None of the other pathologic markers was related to either of the depression measures. Neither dementia nor antidepressant medication modified the relation of pathology to depression. Conclusion: The results do not support the hypothesis that major depression is associated with dementia-related pathology.
