Development and External Validation of a Prediction Model to Identify Suicide Attempters in Treatment-Naive Adolescents With Major Depressive Disorder.

Key Points

• To create and validate a model for predicting suicide attempts in untreated adolescents with major depressive disorder.
• Utilized XGBoost algorithm for modeling
• Included 11 clinical variables for identification
• Conducted external validation using independent datasets
• Model effectively identifies suicide attempters among treatment-naive adolescents
• Achieved stable performance in external validation sets
• Supports evidence-based suicide prevention efforts for clinicians