Résumé
Purpose: Increased risk of falls and fractures has been reported in elderly users of selective serotonin reuptake inhibitors (SSRIs). However, biases were insufficiently addressed notably temporality between exposure and outcome and confounding by residual depression. Our objective was to examine the associations between SSRIs and fall or fracture incidence focusing on their chronic use and different types of SSRIs.Methods: The population-based cohort included participants aged 65 years and above, who had not fallen before inclusion (n=6,599) or free of recent fracture (n=6823) and followed-up twice over 4 years. New fall and fracture events were self-reported and defined as at least two falls and one fracture, respectively, during the previous 2-years. SSRI users were compared with those taking no antidepressants. Hazard ratios (HR) were estimated using Cox models with delayed entry and adjusted for many confounders including residual depressive symptoms.Results: Incidence of falls was 19.3% over 4 years and that of fractures 9.5%. After multi-adjustment, SSRI intake was significantly associated with a higher risk of falls (HR, 95% CI = 1.58, 1.23-2.03) and fractures (HR, 95% CI = 1.61, 1.16-2.24). The risks were significantly increased by 80% in those continuing the treatment over 4 years. Citalopram intake only was at significant risk for falls and fluoxetine for fractures. Conclusions: In this large community-dwelling elderly sample, SSRI users were at higher risk of falls and fractures. This association was not due to reverse causality or residual depressive symptoms. Different SSRI drugs may have specific adverse effects on falls and fractures.