Application of Intersectionality Framework andArea-level Indicators in Machine Learning Analysisof Depression...

Area of Science:

• Public Health
• Sociology
• Mental Health Research

Background:

• Depression is a complex mental health disorder influenced by social determinants of health (SDOH) at individual and community levels.
• Area-level factors and intersectionality are crucial for understanding depression disparities.

Purpose of the Study:

• To examine the association between depression diagnosis and various individual, area-level, and intersectional factors.
• To provide a nuanced understanding of depression disparities using an intersectionality framework.

Main Methods:

• Cross-sectional study using electronic health records from the All of Us research network (n=20,042).
• Logistic regression model applied to variables identified by LASSO method, including sociodemographic characteristics, area-level data, and their interactions.
• Depression diagnosis served as the outcome variable.

Main Results:

• Area-level religious adherence was linked to increased depression odds for women and non-binary individuals.
• Interactions between personal identities and area-level factors revealed higher depression odds for specific groups, e.g., younger, unemployed, never-married Middle Eastern and North African participants.
• Childbirth rates at the area level also influenced depression outcomes.

Conclusions:

• Findings highlight the critical importance of considering individual, area-level, and intersectional factors in depression research.
• A comprehensive approach is necessary to address and mitigate depression disparities effectively.