OBJECTIVES

Individuals with late-life depression (LLD) may have shorter survival, but there is a lack of findings in population-based settings about health-related outcomes of LLD and its subtypes: early-onset depression (EOD) and late-onset depression (LOD). We aimed to evaluate the risk of all-cause mortality of individuals with LLD and its subtypes in an older population-based cohort. Moreover, we investigated whether inflammatory, cognitive, genetic features and multimorbidity could modify the effect of this association.

DESIGN

Longitudinal population-based study with 8-year follow-up.

SETTING AND PARTICIPANTS

We analyzed data on a sample of 1479 participants, all aged >65 years, in the Salus in Apulia Study.

METHODS

LLD was diagnosed through DSM-IV-TR criteria and LOD and EOD according to the age of onset. Multimorbidity status was defined as the copresence of 2 or more chronic diseases.

RESULTS

The overall prevalence of LLD in this older sample from Southern Italy was 10.2%, subdivided into 3.4% EOD and 6.8% LOD. In multivariable Cox models adjusted for age, gender, education, global cognition, apolipoprotein E ε4 allele, physical frailty, interleukin-6, and multimorbidity, LLD showed a greater risk of all-cause mortality. LOD differed from EOD regarding gender, education, cognitive dysfunctions, and diabetes mellitus. There was a significantly increased risk of all-cause mortality for participants with LOD (hazard ratio:1.99; 95% CI 1.33-2.97) in the time of observation between enrollment date and death date (7.31 ± 2.17 months).

CONCLUSIONS AND IMPLICATION

In older age, individuals with LOD but not with EOD had a significantly decreased survival, probably related to increased inflammation, multimorbidity, and cognitive impairments.