Mortality and depressive symptoms in inhabitants of residential homes.

It has been hypothesised that there is a relationship between depression and mortality rate. Some earlier studies have confirmed this relationship, but others have not. In the present study the association was examined between depressive symptoms and mortality in the inhabitants of ten residential homes for the elderly in The Netherlands. Four hundred and twenty-four subjects who were not cognitively impaired, and who participated in an intervention study, were included. One year after the initial interview, they were contacted again and it was found that 69 (16.3%) had died. In the initial interview, depressive symptoms and psychological distress were assessed with the Geriatric Depression Scale and the mental health subscale of the MOS-SF-20. The following correlates of depression were assessed: functional impairment, earlier depression, pain, social support, loneliness, and the presence of seven common chronic illnesses. In bivariate analyses no significant relationship was found between depression and mortality, while controlling for living in an experimental or control home. In logistic regression analyses with mortality as the dependent variable and depressive symptoms, demographic variables, and correlates of depression as predictors, no significant relationship between depression and mortality was found either. It is concluded that no evidence was found in this population for a significant relationship between depression and mortality. Mortality was related to measures of social support, to activities of daily living, and to the presence of chronic non-specific lung disease. Copyright 2001 John Wiley & Sons, Ltd.