BackgroundDepression is associated with excess mortality, but it is not known how treatment-resistance influences life expectancy. We estimated cause-specific excess mortality and Life Years Lost (LYL) in patients with treatment-resistant depression (TRD). MethodsThe population included all individuals born and living in Denmark who redeemed their first prescription for an antidepressant at age 18–69 years between 2005 and 2012, identified in the Danish National Prescription Registry. TRD was defined as at least two additional and different antidepressant trials within two years. Mortality rate ratios (MRRs) were estimated with Cox regression adjusted for age at first prescription, calendar year and comorbidity. Differences in life expectancy were estimated by the Life Years Lost (LYL) method. ResultsThe cohort included 154,513 first-time pharmacologically treated patients with depression, of whom 8,294 (5.4%) were identified as having TRD. Patients were followed for 1,032,245 person-years during which 9,795 deaths occurred. Men and women with TRD had significantly higher mortality than non-TRD (aMRR: 1.34, 95% CI 1.18–1.52 and aMRR: 1.39, 95% CI 1.19–1.63, respectively). Life expectancy for men and women with TRD was 1.21 (95% CI 0.36–2.44) and 1.24 (95% CI 0.35–2.34) years shorter than in all patients with depression. Suicide accounted for the majority of excess LYL, with 1.10 (95% CI 0.46–1.61) years in men and 0.82 (95% CI 0.44–1.27) years in women with TRD. LimitationsUsing redeemed prescriptions to define TRD may increase the risk of misclassification. ConclusionsPatients not responding adequately to several treatment trials are at increased risk for premature death, particularly suicide.