In the elderly, who have higher suicide rates than any other age group, the diagnosis most frequently associated with suicidal behavior is major depression. Yet a great majority of elderly depressives do not kill themselves. The objective of this proposal is explore in elderly suicide attempters a series of factors shown in the literature and preliminary studies to be associated with late life suicidal behavior, and to determine their usefulness in distinguishing individuals with depressive illness who attempt to take their own lives from those who do not.
Specific aims are: (1) to establish those characteristics of the symptomatology and course of affective disorder in late life that are associated with attempted suicide; (2) to describe the stressful life circumstances in the year preceding entry into the study of elderly depressives with suicide attempts, and to investigate the relationship to late life suicidal behavior of two factors that have been most frequently associated with completed suicide in the elderly - physical ill health and social isolation; and to establish (3) the personality traits and (4) neuropsychological function of elderly depressed suicide attempters. In this case-control study, subjects include all patients 50 years of age and over the diagnoses of major depressive disorder admitted to the hospital following a suicide attempt (SAs). Major depressives aged 50 years and over whose admission did not follow a suicide attempt (NAs) constitute the control group. Trained raters will conduct independent interviews with both subjects and family informants, coupling objective and informant-report measures with subjective assessments of symptomatic state, physical health status, social support, and personality traits. Both cases and controls will be reassessed at discharge, 12 and 24 months after admission in order to establish the course of illness and outcome, as well as to provide ratings in the asymptomatic state that reveal state/trait characteristics of each group. The principle analyses will employ multiple logistic regression to predict categorical group membership (SA vs. NA) as a function of independent variables reflecting characteristics of the subjects' affective disorder, medical and psychiatric comorbidity, social network, personality traits, and performance on neuropsychological testing. Secondary analyses will take attempter status as the independent variable. Taking suicidality as a dimensional construct, multiple linear regression analyses will also be employed in which the above independent variables predict a continuous measure of suicidal ideation/intent.
Showing the most recent 10 out of 18 publications
