In this paper we evaluate the application of CART (Classification and Regression Tree) models for depicting the role of protective and risk factors on suicide risk. The data are from a stratified random sample of over 1000 high school aged youth over-sampling potential school dropouts. All youth received an extensive questionnaire capturing a range of behaviors/attitudes related to school, drugs, mood, family, peers, and self. In addition, each youth receives a 10-15 minute interview (Screen for Youth Suicide Risk) measuring suicide risk behaviors. The interview provides a summary score of risk and includes brief clinical assessments by the interviewer. We use CART to establish relations between risk and protective factors and suicide behaviors (e.g. attempts, ideation) and to evaluate the predictive value of a subset of questionnaire items. The analyses provides insight into the role of specific protective and related risk factors, as well as, identifies a small select set as indicators of suicide risk.
Learning Objectives: The presentation should lead to an understanding of 1) simple predictors of suicide risk among adolescents, 2) Classification and Regression Tree (CART) approaches to analyzing data, and 3) the utility of a short screen for suicide risk.
Keywords: Adolescents, Suicide
Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
Disclosure not received
Relationship: Not Received.