BackgroundThe association between mortality risk and use of antidepressants in people with dementia is unknown.ObjectiveTo describe the use of antidepressants in people with different dementia diagnoses and to explore mortality risk associated with use of antidepressants 3 years before a dementia diagnosis.MethodsStudy population included 20,050 memory clinic patients from Swedish Dementia Registry diagnosed with incident dementia. Data on antidepressants dispensed at the time of dementia diagnosis and during three-year period before dementia diagnosis was obtained from the Swedish Prescribed Drug Register. Cox regression models were used.ResultsDuring a median follow-up of 2 years from dementia diagnosis, 25.8% of dementia patients died. A quarter (25.0%) of patients were on antidepressants at the time of dementia diagnosis while 21.6% used antidepressants at some point during a three-year period before a dementia diagnosis. Use of antidepressant treatment for 3 consecutive years before a dementia diagnosis was associated with a lower mortality risk for all dementia disorders (HR: 0.82, 95% CI: 0.72–0.94) and in Alzheimer's disease (HR: 0.61, 95% CI: 0.45–0.83). There were no significant associations between use of antidepressant treatment and mortality risk in other dementia diagnoses.ConclusionAntidepressant treatment is common among patients with dementia. Use of antidepressants during prodromal stages may reduce mortality in dementia and specifically in Alzheimer's disease.Disclosure of interestThe authors have not supplied their declaration of competing interest.