Objective: To investigate the reported association between low vitamin E levels and depressive symptoms in a population-based study.
Methods: The study is based on a cohort of 3884 adults aged 60 years and over who participated in the third survey of the Rotterdam Study, were screened for depressive symptoms with the Center of Epidemiological Studies Depression Scale and from whom blood was drawn. All screen-positive subjects had a psychiatric work-up. Blood levels of vitamin E were compared between 262 cases with depressive symptoms and 459 randomly selected reference subjects. All analyses were stratified by sex, and adjusted for age, cholesterol, cognitive score, smoking, dietary supplement use, marital status, living alone, and functional disability score.
Results: Vitamin E levels in men with depressive symptoms were lower than in non-depressed men after adjusting for age, whereas no such difference was found in women. This association in men was substantially weakened after controlling for biological factors, and disappeared with additional adjustment for nutritional behaviour and social factors. No differences were observed when the analyses were restricted to cases with depression as defined in the Diagnostic and Statistical Manual of Mental Disorders IV.
Conclusions: After control for several biological and behavioural factors relating to health we found no association between low vitamin E levels and depressive symptoms or depression in the elderly.