Building and Evaluating Suicide Attempt Prediction Models using Risk Factors.

Abstract

This study identified risk factors of suicide attempts for the purpose of building its prediction model and thereafter evaluated its performance. The participants of this secondary data analysis study were 11,671 adults aged 19\u2009years old. The prediction model consisted of risk factors identified through multiple logistic regression analysis, and its performance was analyzed in terms of calibration, discrimination, and clinical usefulness. The risk factors of suicide attempts were suicide plan and suicidal ideation for males, and suicide plan and depression diagnosis for females. The prediction models constructed with these risk factors showed good calibration and discrimination, with over .90 of the area under the curves. At the cutoff point of .5%, the sensitivity of the full model was 90.9% for males and 82.4% for females. The net benefit was positive at threshold probability under 30% for males and 40% for females. Given the acceptable performance of such suicide attempt prediction models, it can be used to assess suicide attempt risk and detect high-risk population of suicide attempt in the community at an early stage, with a limited human resources.