Friday, 22 August 2003
This presentation is part of : Risk Factors in Late Life Suicide: A Comparison of East and West

S102-001 Suicide in the Second Half of Life: A Psychological Autopsy Study

Yeates Conwell1, Paul Duberstein2, Kenneth R. Conner3, Shirley Eberley4, Holly Wadkins5, and Eric Caine1. (1) Psychiatry, University of Rochester Medical Center, Rochester, NY, USA, (2) Laboratory of Personality and Development, Department of Psychiatry, University of Rochester, Rochester, NY, USA, (3) Psychiatry, University of Rochester School of Medicine, Rochester, NY, USA, (4) Biostatistics, University of Rochester School of Medicine, Rochester, NY, USA, (5) Psychiatry, University ofRochester School of Medicine, Rochester, NY, USA

Objectives:

1) To describe the mental and physical health status, functioning, and social circumstances of suicides age 50 years and over, and

2) to determine risk factors for suicide by the comparison of cases with demographically matched community dwelling controls.

Background: Uncontrolled psychological autopsy studies have indicated associations between suicide in later adulthood and male gender, psychiatric illnesses, history of suicidal behavior, physical health and functional status, and social circumstances. To determine whether variables in each domain represent risk factors for suicide, information must be obtained regarding their prevalence in carefully matched controls.

Methods: This presentation reports findings of a study comparing completed suicides ages 50 years and over with controls matched for age, gender, race, and community of residence. Information from each hypothesized risk domain was obtained using standardized instruments administered to knowledgeable informants for both groups. Univariate and multivariate logistic regression examined the influence of risk factors on case status while controlling for pertinent demographic and clinical variables.

Findings: Comparison of 86 suicides and 86 controls revealed that cases were significantly more likely to have psychiatric illness, most often an affective disorder. Anxiety disorders, benzodiazepine prescription, sleep disturbance, and a history of suicide attempts also distinguished the groups. Suicides had significantly greater physical illness burden and functional limitations and experienced more pain than controls. While a majority of both groups had been seen by a primary care provider in the last month, the proportions did not differ. Suicides experienced more family discord and had fewer social supports than controls.

Conclusions: Suicide in older adults is characterized by a distinct set of risk and protective factors that have implications for the design of high-risk as well as population-based prevention strategies.

Back to S102 Risk Factors in Late Life Suicide: A Comparison of East and West
Back to The Eleventh International Congress