Related Experiment Videos
Predicting multiple mental health outcomes in adolescents using explainable machine learning models
Xiaojie Dong1, Jing Meng1, Peng Wang1
1Department of Psychiatry, Ward 1, Wuhan Mental Health Center, Wuhan, Hubei, China.
Introduction:
Adolescent mental health problems, including depression, anxiety, and stress, are an increasing public health concern, yet the factors associated with different mental health outcomes may vary across domains. This study investigated shared and outcome-specific predictive features of depression, anxiety, perceived stress, and psychological well-being using explainable machine learning models.
Methods:
A cross-sectional study was conducted among 1,088 adolescents aged 13-18 years recruited from secondary schools in Wuhan, China, using multistage cluster sampling. Participants completed validated measures of emotional dysregulation, loneliness, social media addiction, self-esteem, sleep quality, academic stress, family support, physical activity, and mental health outcomes. Four algorithms-linear regression, support vector regression, Random Forest, and XGBoost-were trained using an 80/20 train-test split with five-fold cross-validation, and model performance was evaluated using test-set R², RMSE, and MAE. SHapley Additive exPlanations (SHAP) were used to examine feature contributions. To minimize target leakage, outcome-specific feature sets were used, with PSQI and RSES excluded from the depression model because of direct or substantial conceptual overlap with PHQ-9 content, and PSQI excluded from the well-being model because of overlap with WHO-5 content.
Results:
XGBoost showed the strongest out-of-sample predictive performance across all four outcomes, explaining 49% of the variance in depression (R² = 0.49, 95% CI: 0.44-0.53; RMSE = 3.52; MAE = 2.81), 55% in anxiety (R² = 0.55, 95% CI: 0.50-0.59; RMSE = 3.04; MAE = 2.47), 60% in perceived stress (R² = 0.60, 95% CI: 0.56-0.64; RMSE = 3.71; MAE = 2.87), and 50% in psychological well-being (R² = 0.50, 95% CI: 0.45-0.54; RMSE = 3.38; MAE = 2.68).
Discussion:
Emotional dysregulation and loneliness were consistently among the most influential features, while academic stress, family support, social media addiction, self-esteem, and sleep quality showed outcome-specific contributions. SHAP rankings for the depression model were stable across five-fold cross-validation, with emotional dysregulation and loneliness consistently occupying the highest ranks.
