The prediction of suicide. Sensitivity, specificity, and predictive value of a multivariate model applied to suicide among 1906 patients with affective disorders

Abstract

Stepwise multiple logistic regression was utilized in an attempt to develop a statistical model that would predict suicide in a group of 1906 Iowans with affective disorders admitted to a tertiary care hospital. The risk factors identified by this approach included the number of prior suicide attempts, suicidal ideation on admission, bipolar affective disorder (manic or mixed type), gender, outcome at discharge, and unipolar depressive disorder in individuals with a family history of mania. However, the model failed to identify any of the patients who committed suicide. The results appear to support the contention that, based on present knowledge, it is not possible to predict suicide, even among a high-risk group of inpatients.