Probable depression is prevalent among parents and primary caregivers of children with specific learning disorders in Saudi Arabia, and machine-learning models can predict it with high accuracy.
The study found that 35.6% of 612 caregivers met criteria for probable depression (PHQ-9 ≥ 10). XGBoost achieved an ROC AUC of 0.916, sensitivity of 0.858, and specificity of 0.830 at a screening threshold of 0.425, with key predictors including parenting stress, anxiety, sleep quality, and child difficulties.
Evidence Score
Study Evidence
Study 1. Exploring the Prevalence and Key Predictors of Probable Depression Among Parents and Primary Caregivers of Children and Adolescents With Specific Learning Disorders in the Kingdom of Saudi Arabia: A Cross-Sectional Explainable Machine-Learning Study.
observationalAlmulla AA, Khasawneh MAS · Clinical psychology & psychotherapy (2026)
Result: The study found that 35.6% of 612 caregivers met criteria for probable depression (PHQ-9 ≥ 10). XGBoost achieved an ROC AUC of 0.916, sensitivity of 0.858, and specificity of 0.830 at a screening threshold of 0.425, with key predictors including parenting stress, anxiety, sleep quality, and child difficulties.
Mechanism Graph
Limitations
- ⚠Cross-sectional design prevents causal inference
- ⚠Sample recruited through service and community pathways may not be representative of all caregivers in Saudi Arabia
- ⚠External validation of machine-learning models is required before clinical implementation
Frequently Asked Questions
What is the prevalence of probable depression among caregivers of children with SLDs in Saudi Arabia?▼
The study found a prevalence of 35.6% (95% CI 31.8-39.6) among 612 caregivers, based on a PHQ-9 score of 10 or higher.
Which machine-learning model performed best for predicting caregiver depression?▼
XGBoost and the soft-voting ensemble achieved the highest ROC AUC of 0.916. XGBoost had an average precision of 0.867 and a Brier score of 0.115.
What are the top predictors of probable depression in these caregivers?▼
Key predictors identified by SHAP analysis include parenting stress, parental anxiety symptoms, sleep quality, SLD severity, child emotional and behavioral difficulties, perceived social support, and coping capacity.
Can these models be used in clinical practice now?▼
No, the authors state that external evaluation is required before implementation, as the models were developed on a specific sample and need validation in other populations.
Products
Affiliate links coming soon. We only recommend products that match the doses and forms used in the cited research.
References
- 1.Almulla AA, Khasawneh MAS. "Exploring the Prevalence and Key Predictors of Probable Depression Among Parents and Primary Caregivers of Children and Adolescents With Specific Learning Disorders in the Kingdom of Saudi Arabia: A Cross-Sectional Explainable Machine-Learning Study.." Clinical psychology & psychotherapy, 2026. PMID: 42490731 DOI: 10.1002/cpp.70310