Objectives: The aim of this study was to evaluate changes in benzodiazepine use over time, and the association between benzodiazepine use and select outcomes.
Design: A five-year longitudinal cohort study in subjects 65 years of age and older.
Setting: Select urban communities and institutions across Canada with senior citizens.
Participants: Subjects who were first seen in 1990 to 1991, recontacted in 1996, and agreed to undergo a second clinical examination. Mortality rates were based on the initial 2914 subjects enrolled.
Measurements: Number and type of medications used. Outcomes (mortality, incident institutionalization, change in cognition, depression, function, self-rated health) associated with benzodiazepine use. Logistic regression to predict outcomes and pattern of benzodiazepine use.
Results: Mean number of medications being taken by senior citizens increased to 5.8 from 3.9. The proportion of subjects using benzodiazepines at time 1 and time 2 was similar (26.4% versus 25.2%). Affect, self-rated health, cognition, function and incident institutionalization were significantly associated with benzodiazepine use. Subjects with a depressed mood were more likely to be prescribed a benzodiazepine (37%) than an antidepressant (26.9%).
Conclusion: Benzodiazepines were associated with a number of adverse outcomes. The relative benefits and risks of benzodiazepine use in an older population should be re-examined.