The long-term goal of our work is to develop and test suicide prevention measures that are efficient and effective for use in the elderly. Knowledge of the factors that place elders at risk for suicide is necessary for us to reach that goal. A number of medical, psychological and social influences have been shown to be correlates of suicide in late-life. New evidence is emerging to substantiate the effect of a specific domain of medical factors, i.e. cerebrovascular risk factors (CVRFs), on late-life psychiatric disorders, which in turn have already been shown to be robust risk factors for late-life suicide. The primary objective of our study is to examine using a case-control design whether CVRFs at clinical and radiopathological levels independently increase risk of late-life suicide. In our secondary objectives, we will examine the correlation between clinical CVRFs and radiopathological CVRFs. Also, we Will explore whether CVRFs (both clinical and radiopathological) render some at-risk older adults even more vulnerable to self-destruction by way of impaired executive cognitive function (ECF). Cross-sectional case-control design will be employed. 100 cases will be recruited from suicide attempters, ages 65 or over presenting consecutively to a tertiary care setting over a 30-month period; while the same number of controls, of comparable age and sex distribution, without life-time history of suicide, will be recruited from community dwelling elders who attend local elderly recreational centres. Measures of sociodemographic background, suicide behavior profile, psychopathology, personality, physical health profile including clinical cerebrovascular risk factor (CVRFs) and vascular pathologies in MRI of Brain, executive cognitve function and global cognitive function will be administered to all subjects. Logistic regression model will be used to test if there is greater burden of CVRFs at clinical/ radiopathological levels in suicide attempters over controls, after controlling for other biopsychosocial risk factors of late-life suicidal behavior. The model will also test if there is any significant interaction between personality trait of impulsiveness and impaired ECF in increasing risk of suicidal behavior. The correlation between clinical and radiopathological CVRFs as well as the relationship between burden of CVRFs and impaired ECFs will be worked out by correlation statistics.
