Antidepressant treatment may increase the risk of death. The association between antidepressants and mortality has been evaluated in community-dwelling older people, but not in representative samples of very old people, among whom dementia, multimorbidity, and disability are common.

Umeå 85+/GERDA study participants (n = 992) aged 85, 90, and ≥95 years were followed for up to five years. Cox proportional hazard regression models were used to analyze mortality risk associated with baseline antidepressant treatment, adjusted for potential confounders.

Mean age was 89 years; 27% of participants had dementia, 20% had stroke histories, 29% had heart failure, and 16% used antidepressants. In age- and sex-adjusted analyses, antidepressant use was associated with a 76% increased mortality risk (hazard ratio [HR] = 1.76; 95% confidence interval [CI], 1.41–2.19). Adding adjustment for Geriatric Depression Scale score, HR was 1.62 (95% CI, 1.29–2.03). The association was not significant when adjusting for additional confounding factors (HR = 1.08; 95% CI, 0.85–1.38). Interaction analyses in the fully adjusted model revealed a significant interaction between sex and antidepressant use (HR: 1.76; 95% CI, 1.05–2.94). Among male and female antidepressant users, the HRs for death were 0.76 (95% CI, 0.47–1.24) and 1.28 (95% CI, 0.97–1.70), respectively.

Among very old people, baseline antidepressant treatment does not seem to be independently associated with increased mortality risk. However, the risk may be different in men and women. This difference and the potential risk of initial treatment require further investigation in future cohort studies of very old people.