Risk factors for depressive disorders in very old age: a population-based cohort study with a 5-year follow-up

Soc Psychiatry Psychiatr Epidemiol. 2014 May;49(5):831-9. doi: 10.1007/s00127-013-0771-2. Epub 2013 Oct 8.

Abstract

Purpose: Depressive disorders are common among the very old, but insufficiently studied. The present study aims to identify risk factors for depressive disorders in very old age.

Methods: The present study is based on the GERDA project, a population-based cohort study of people aged ≥85 years (n = 567), with 5 years between baseline and follow-up. Factors associated with the development of depressive disorders according to DSM-IV criteria at follow-up were analysed by means of a multivariate logistic regression.

Results: At baseline, depressive disorders were present in 32.3 % of the participants. At follow-up, 69 % of those with baseline depressive disorders had died. Of the 49 survivors, 38 still had depressive disorders. Of the participants without depressive disorders at baseline, 25.5 % had developed depressive disorders at follow-up. Baseline factors independently associated with new cases of depressive disorders after 5 years were hypertension, a history of stroke and 15-item Geriatric Depression Scale score at baseline.

Conclusions: The present study supports the earlier findings that depressive disorders among the very old are common, chronic and malignant. Mild depressive symptoms as indicated by GDS-15 score and history of stroke or hypertension seem to be important risk factors for incident depressive disorders in very old age.

MeSH terms

  • Aged
  • Depression / diagnosis
  • Depression / epidemiology*
  • Depressive Disorder / diagnosis
  • Depressive Disorder / epidemiology*
  • Diagnostic and Statistical Manual of Mental Disorders
  • Female
  • Follow-Up Studies
  • Geriatric Assessment
  • Humans
  • Hypertension / epidemiology
  • Incidence
  • Logistic Models
  • Male
  • Middle Aged
  • Multivariate Analysis
  • Quality of Life
  • Risk Factors
  • Socioeconomic Factors
  • Stroke / epidemiology