The differential impact of depression across different periods in life on mortality remains inconclusive. We aimed to examine the association of depression that occurs at different age with all-cause mortality, and to explore the roles of dementia, as well as genetic and early-life environmental factors, in this association.

From the Swedish Twin Registry, 44,919 twin individuals were followed for up to 18 years. Depression was ascertained using the National Patient Registry and categorized as early-life (<45 years), midlife (45–64 years), and late-life (≥65 years) depression according to the age of the first diagnosis. Deaths were identified through the Cause of Death Register. Generalized estimating equation, generalized structural equation, and conditional logistic regression were used for unmatched, mediation, and co-twin matched analyses, respectively.

In unmatched analyses, the multivariate-adjusted odds ratios (ORs) and 95% confidence intervals (CIs) of mortality were 1.71 (1.46–2.00) for depression at any age, 1.72 (1.36–2.17) for early-life, 1.51 (1.19–1.90) for midlife, and 4.10 (2.02–8.34) for late-life depression. Mortality was significantly higher in individuals with late-life depression than those with earlier-life depression (p < 0.05). The mediation analysis showed that 59.83% of the depression-mortality association was mediated by dementia. No significant difference in ORs between the unmatched and co-twin matched analyses was observed (p = 0.09).

Depression is associated with an increased risk of all-cause mortality, and dementia mediates approximately 60% of the impact of depression on mortality in late life. Genetic and early-life environmental factors may not play a significant role in the depression-mortality association.