Are higher antidepressant plasma concentrations associated with fall risk in older antidepressant users?

Key summary points Aim To explore the association between antidepressant plasma concentrations and fall risk in older users. Findings An association was found between baseline TCA plasma concentration and fall risk within users. No association was found between the SSRIs plasma concentrations and fall risk. Message The association between TCA plasma concentration and fall risk needs to be replicated before its relevance to clinical practice can be established. Purpose Antidepressants are well-established fall-risk increasing drugs (FRIDs) and therefore falls should be considered an important adverse drug event (ADE) of antidepressants. However, not all antidepressant users experience fall incidents and factors associated with increased fall risk among antidepressant users are incompletely understood. Our objective was to explore whether antidepressant plasma concentrations are associated with falls in older antidepressant users. Methods For this study, we included antidepressant users of the multicenter B-PROOF study. Fall incidents were recorded prospectively using fall calendars. Antidepressant plasma concentrations were analyzed by Liquid chromatography-mass spectrometry (LC–MS) at baseline and at 2 years follow-up. The associations between the observed antidepressant concentration and fall risk were assessed using Cox proportional hazard and logistic regression models and adjusted for potential confounders. Results In total 93 selective serotonin reuptake inhibitor (SSRI) and 41 antidepressant (TCA) users were identified. There was a significant association between baseline TCA plasma concentration and fall risk within users (HR 2.50, 95% CI 1.07–5.87, crude model). In the adjusted model, there were no significant associations between concentrations of SSRIs and fall risk. Conclusion There might be an association between plasma concentrations of TCAs and the risk of falling in older users. However, these results needs to be interpreted with caution considering the small sample size and accompanying limitation of confinement to crude analyses. Therefore, these novel findings need to replicated in a larger cohort, preferably including adjustment for potential confounders and more frequent measures of plasma concentrations is needed.