Mental health symptoms associated with morbidity, not mortality, in an elderly community sample

Philip J. Batterham*, Helen Christensen, Andrew J. MacKinnon

*Corresponding author for this work

    Research output: Contribution to journalArticlepeer-review

    16 Citations (Scopus)

    Abstract

    Purpose Six previous reviews have found a relationship between depression and mortality. However, many past studies have failed to adequately control for the role of physical health. A proposed mechanism of the depression- mortality relationship suggests that physical health may mediate the relationship. The present study used new methods to examine relationships between mental health symptoms and mortality in an elderly community cohort while accounting for potential mediation of these relationships by physical health. Method 896 community-dwelling participants aged 70-97 were assessed four times over 12 years and vital status was tracked for up to 17 years. Relationships of depression and anxiety with survival time, controlling for physical health, age and gender, were tested using Cox proportional hazards regressions embedded in structural equation models. Results A significant unadjusted relationship between depression symptoms and mortality (HR = 1.09, p < .001) was attenuated to non-significance after controlling for measures of physical health (HR = 1.03, p = .18). No significant relationship was found between anxiety symptoms and mortality. Conclusions The relationship between depression and mortality was accounted for by physical health status in this cohort. This finding casts doubt on studies that report a relationship between depression and mortality without adequately considering the effect of physical health.

    Original languageEnglish
    Pages (from-to)79-85
    Number of pages7
    JournalSocial Psychiatry and Psychiatric Epidemiology
    Volume47
    Issue number1
    DOIs
    Publication statusPublished - Jan 2012

    Fingerprint

    Dive into the research topics of 'Mental health symptoms associated with morbidity, not mortality, in an elderly community sample'. Together they form a unique fingerprint.

    Cite this