Prediction and Prevalence of Depression and Anxiety in Young Adults With Specific Learning Disorders: A Machine Learning Approach in Saudi Arabia

The current study aimed to estimate the prevalence of clinically significant depression and anxiety symptoms in young adults with DSM-5-confirmed specific learning disorders (SLDs) in Saudi Arabia and to develop machine learning (ML) models to identify individuals at elevated risk. In this cross-sectional study, 439 young adults were recruited via online advertisements. They underwent structured clinical interviews administered by licensed psychologists or psychiatrists to confirm their SLD diagnosis in accordance with DSM-5 criteria. Depression and anxiety were assessed using validated self-report measures (BDI-II ≥ 20 for depression; BAI ≥ 16 for anxiety). Clinically significant depression was observed in 175 of 439 participants (39.9%), and clinically significant anxiety in 151 of 439 participants (34.4%). Recursive feature elimination identified key predictors for depression (SLD severity, female gender, burnout and attention-deficit/hyperactivity disorder [ADHD]) and anxiety (insomnia, age, SLD subtype, ADHD, and SLD severity). Four supervised ML algorithms (logistic regression, random forests, extreme gradient boosting [XGBoost], and support vector machine [SVM]) were tuned using 10-fold cross-validation within the training set and evaluated on a held-out test set (30% stratified split). These results indicate that depression and anxiety are highly prevalent among Saudi young adults with clinically confirmed SLD, and ML models can accurately identify those at the highest risk. Findings highlight SLD severity, insomnia and comorbid ADHD as key factors for early intervention in educational and clinical settings.