Objective: To assess whether antidepressant treatment is associated with a temporary increase in the risk of a motor vehicle crash among older adults.
Design: Population-based case-only time-to-event analysis.
Setting and subjects: Data from transportation and healthcare databases for adults age 65 and older in Ontario, Canada, between January 1, 2000, and October 31, 2007. Consecutive adults who had a motor vehicle crash anytime following their 66th birthday.
Measurements: The primary exposure variable was treatment with antidepressant medication, and the primary outcome measure was a motor vehicle crash.
Results: A total of 159,678 individuals had a crash during the study, of whom 7,393 (5%) received an antidepressant in the month prior to the crash. The hazard ratio (HR) of crash associated with second-generation antidepressants was 1.10 (95% confidence interval [CI]: 1.07-1.13, χ² = 41.77, df = 1, p <0.0001), adjusted for gender, license suspensions, and other medications, but the risk for first-generation antidepressants was not significant. The increased risk was restricted to those who were also concurrently prescribed a benzodiazepine (adjusted HR: 1.23, 95% CI: 1.17-1.28, χ² = 85.28, df = 1, p <0.0001) or a strong anticholinergic medication (adjusted HR: 1.63, 95% CI: 1.57-1.69, χ² = 627.31, df = 1, p <0.0001), and was confined to crashes where the patient was at fault. The increased risk was apparent for the first 3-4 months following initiation of an antidepressant and returned to baseline thereafter.
Conclusions: Prescriptions for second-generation antidepressants in older adults are associated with a modest increased risk of motor vehicle crashes, when combined with other medications that can impair cognition.