Concurrent and Prospective Prediction of Suicidal Ideation in Adolescents Using Multimethod Data and Machine Learning: A Pilot Study.

ABSTRACT:

Objective

Despite an increased focus on prevention, rates of suicidal thoughts and behaviors have failed to decline and robust predictors have yet to be identified, highlighting a critical need to integrate multiple risk processes across methodological approaches in predictive models.

Method

The current pilot study leveraged machine learning with multimethod data to predict concurrent and prospective suicidal ideation (SI) in a sample of adolescents 14 to 17 years of age (N = 165) oversampled for depression and depression risk. Predictors included clinical diagnoses and comorbidity load, interviewer-rated chronic stress, self-reported internalizing symptoms, daily experiences of positive and negative affect, and neural measures of emotion processing and reward responsiveness

SUBMITTER: Dickey L 

PROVIDER: S-EPMC12684466 | biostudies-literature | 2025 Dec

REPOSITORIES: biostudies-literature