Longitudinal Analysis of the Association between Depressive Symptomatology and Cognitive Deterioration

Although many studies have found a cross-sectional relation between depression and dementia or depressive symptomatology and cognitive functioning, the direction of the association is still unknown. The purpose of this analysis was to determine whether high depressive symptomatology is predictive of cognitive deterioration among the elderly 3 years later. Data came from a community-based prospective cohort study of noninstitutionalized and nondemented subjects aged 65 years and over living in the Gironde departement in southwest France (1, 600 subjects were interviewed at both study entry in 1989 and 3-year follow-up). Cognitive functions were assessed with the Mini-Mental State Examination (MMSE), and cognitive deterioration was defined as an MMSE score decrease of at least five points between two assessments. The Center for Epidemiologic Studies Depression (CES-D) Scale was used to evaluate the level of depressive symptomatology. The present study reports that a high level of depressive symptomatology is not predictive of cognitive deterioration 3 years later (relative risk = 0.8, 95% confidence interval 0.3–2.1). The authors observed that the risk of cognitive deterioration was associated with the concomitant level of depressive symptomatology at the 3-year follow-up, independent of depressive symptoms at entry. These results indicate that the association between high depressive symptomatology and poor cognitive functioning is cross-sectional, and they illustrate the importance of adjusting for depressive symptomatology in epidemiologic studies assessing cognitive functions. Am J Epidemiol 1996;144:634-41.