Work

Adverse Childhood Experiences and Social Determinants of Mental Health as Predictors of Adult Depression: A Machine Learning Approach

Year: 2026

Type: article

Abstract: ABSTRACT This study applied machine learning (ML) models to the 2023 Behavioral Risk Factor Surveillance System (BRFSS) dataset, a nationally representative and state‐based survey conducted annually by the Centers for Disease Control and Prevention (CDC), to examine how adverse childhood experiences (ACEs) and social determinants of mental health (SDMH) predict adult depressive disorders. Among ML models, eXtreme Gradient Boosting (XGBoost) achieved the strongest performance, and we identified key predictors, including family history of mental illness, sex, total ACE score, loneliness, unemployment, and healthcare barriers. Subgroup analyses revealed variation across racial groups, showing the need for culturally responsive approaches. We discuss the utility of ML for advancing early identification, trauma‐informed counseling practice, and equity‐focused prevention strategies.

Source: Journal of Counseling & Development

DOI: 10.1002/jcad.70029

Authors: Eunae Han, Jihoon Shin

Institution: The University of Texas at El Paso

Language: English

Topic: Child Abuse and Trauma

Subfield: Clinical Psychology

Field: Psychology

Domain: Social Sciences

SDGs: Good health and well-being, Reduced inequalities

Open Access status: hybrid

Estimated APC paid: $2,900