Suicide in the oldest old: an observational study and cluster analysis

Int J Geriatr Psychiatry. 2016 Jan;31(1):33-40. doi: 10.1002/gps.4286. Epub 2015 Mar 24.


Objectives: The older population are at a high risk for suicide. This study sought to learn more about the characteristics of suicide in the oldest-old and to use a cluster analysis to determine if oldest-old suicide victims assort into clinically meaningful subgroups.

Methods: Data were collected from a coroner's chart review of suicide victims in Toronto from 1998 to 2011. We compared two age groups (65-79 year olds, n = 335, and 80+ year olds, n = 191) and then conducted a hierarchical agglomerative cluster analysis using Ward's method to identify distinct clusters in the 80+ group.

Results: The younger and older age groups differed according to marital status, living circumstances and pattern of stressors. The cluster analysis identified three distinct clusters in the 80+ group. Cluster 1 was the largest (n = 124) and included people who were either married or widowed who had significantly more depression and somewhat more medical health stressors. In contrast, cluster 2 (n = 50) comprised people who were almost all single and living alone with significantly less identified depression and slightly fewer medical health stressors. All members of cluster 3 (n = 17) lived in a retirement residence or nursing home, and this group had the highest rates of depression, dementia, other mental illness and past suicide attempts.

Conclusions: This is the first study to use the cluster analysis technique to identify meaningful subgroups among suicide victims in the oldest-old. The results reveal different patterns of suicide in the older population that may be relevant for clinical care.

Keywords: cluster analysis; oldest old; suicide.

Publication types

  • Observational Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Age Factors
  • Aged
  • Aged, 80 and over
  • Cluster Analysis
  • Depressive Disorder / complications
  • Depressive Disorder / epidemiology
  • Female
  • Humans
  • Life Style
  • Male
  • Marital Status
  • Ontario / epidemiology
  • Residence Characteristics
  • Risk Factors
  • Social Behavior
  • Suicide / statistics & numerical data*