Peer-reviewed article
Machine learning-based early warning model for adolescent mental health risk using the p factor.
Yunjing Li, Wenxuan Bian, Xiaohong Wen, Yuanyuan Fang, Yihan Wang et al. ·
Summary
Written by AI, not yet AICS's view · claude-haiku-4-5-20251001Researchers developed a machine learning model to identify adolescents at high mental health risk using data from over 6,000 Chinese youth. The model combined 23 predictors across individual, school, family, and social domains to predict a general mental health risk factor. Sleep quality emerged as the strongest predictor of mental health risk, followed by repetitive negative thinking and interpersonal stress. The model achieved good accuracy in both internal and external testing, suggesting it could help target early prevention efforts.
Key points
- XGBoost machine learning model achieved macro F1 score of 0.73-0.80 in identifying adolescents at mental health risk
- Sleep quality was the most influential predictor, followed by repetitive negative thinking, interpersonal stress, impulsivity, and emotional intensity
- Model used 23 predictors across individual, school, family, and societal domains based on ecological systems theory
- Study included 5,283 adolescents for model development and 968 for external validation
- Model suggests sleep quality as a key intervention target for early mental health prevention
Limitations
The study uses cross-sectional data from Chinese adolescents only, so findings may not generalise to other populations or contexts. The paper does not discuss how the model would be implemented in real-world service delivery settings or whether early identification actually leads to improved outcomes when interventions are provided.
Abstract
BACKGROUND: Accurate identification of adolescents at high mental health risk is essential for timely intervention and improved service delivery. The general factor of psychopathology (p factor), capturing shared variance across mental disorders, was used as a transdiagnostic outcome to develop a machine learning model for risk identification. METHODS: Data from 5,283 Chinese adolescents were used for model training and testing, with an additional 968 participants for external validation. Guided by ecological systems theory, a multidimensional early warning framework covering 59 indicators across individual, school, family, and society domains was constructed. Seven machine learning algorithms were tested using the p factor as the outcome. Shapley Additive exPlanations (SHAP) values ranked predictor importance and identified an optimal feature subset. RESULTS: The XGBoost model showed the best performance (macro F1 = 0.73 on internal validation, 0.80 on external validation). The final model used 23 predictors. SHAP confirmed sleep quality as the most influential factor, followed by repetitive negative thinking, interpersonal stress, impulsivity, and emotional intensity. CONCLUSIONS: This study suggests that a p-factor-based machine learning model can effectively identify adolescents at mental health risk, with sleep quality emerging as the most influential predictor, suggesting potential intervention targets for early prevention.
Findings the Hub has extracted
Each claim below is labelled by its type of design and links back to this report.
Problem
Adolescent mental health risk conditions require early identification for timely intervention
- Intervention
- Machine learning-based early warning model using the p factor and 23 predictors across individual, school, family, and societal domains to identify at-risk adolescents
- Population
- 5,283 Chinese adolescents for model training and testing; 968 for external validation
Key finding
XGBoost model achieved macro F1 score of 0.73 on internal validation and 0.80 on external validation
Li et al. · PubMed (US National Library of Medicine) · 18 Jun 2026