A. D. Murray, R. T. Staff, C. J. McNeil, S. Salarirad, L. H. Phillips, J. Starr, I. J. Deary, L. J. Whalley


1520-6394 (Electronic)1091-4269 (Linking)

Publication year



Depress Anxiety

Periodical Number






Author Address

Aberdeen Biomedical Imaging Centre, University of Aberdeen, Aberdeen, Scotland, UK.

Full version

BACKGROUND: The influence of white matter lesions on depressive symptoms in healthy ageing populations remains unclear. In this study, we examined the relationship between depressive symptoms and magnetic resonance imaging (MRI) detected cerebrovascular disease in a normal population living independently in the community, and measured the influence of location of brain abnormalities, fluid intelligence, living alone, and sex. METHODS: Prospective cohort: 497 community dwelling individuals all born in 1936, who took part in the Scottish Mental Survey of 1947, were followed up in 2000 and at biannual intervals in a longitudinal study of health and cognitive aging. Two hundred forty-four volunteered for brain MRI in 2004-2006. Suitable data were available in 219/244, of whom 115 were men. Brain hyperintensities in lobar white matter, basal ganglia , periventricular, and infratentorial regions were measured using Scheltens’ scale. Depressed mood was assessed using the Hospital Anxiety and Depression Scale (HADS) on three biannual intervals. Relationships between Scheltens’ scores, HADS-D scores, fluid intelligence, living alone, and sex were assessed using general linear modeling. RESULTS: The main predictor of depressive symptom scores was poorer fluid intelligence (partial eta(2) =0.023-0.028, P < .05). Ischemic change in the brainstem (partial eta(2) = 0.026, P