Psychotropic drug use and cognitive decline among older men and women.

This epidemiological study focuses on cognitive change related to psychotropic drug use in a population-based sample of subjects aged 65 and over. Cognitive functioning was assessed in 1982 and 1988 by the Short Portable Mental Status by Pfeiffer, and cognitive decline was defined as an increase of more than 2 errors in 1988 relative to the 1982 assessment. Psychotropic drugs were classified into benzodiazepines and non-benzodiazepines. For both medications, four patterns of intake were considered: no use reported at both interviews, continuous use at both interviews, temporary use at the 1982 interview and new use at the 1988 interview. Depressive symptomatology was assessed by the Center for Epidemiologic Studies Depression Scale. The analysis was performed on 1200 subjects with no or minimal impairment at baseline for whom complete data were available. Univariate analyses showed cognitive decline associated with gender, level of education and new medical condition; it was also related to depressive symptomatology and psychotropic drug use. These two factors were the most strongly associated with decrease in performance in multivariate analyses. Cognitive decline differed according to class of psychotropic drugs and pattern of use: benzodiazepine temporary users exhibited a lower risk compared with never users (OR = 0.23, p = 0.056), non-benzodiazepine new users a higher risk (OR = 5.02, p < 0.001). Despite the simple measures of cognitive functioning and psychopathology, and the approximation in pattern of psychotropic drug use, these results emphasize the importance of considering psychotropic drugs in studies of cognitive decline in elderly subjects.