Applying the Health Belief Model to Characterize Racial/Ethnic Differences in Digital Conversations Related to...

Area of Science:

• Digital health
• Mental health research
• Sociology of health

Background:

• Depression prevalence in the US increased over threefold mid-COVID-19 compared to pre-pandemic levels.
• Racial/ethnic differences in depression mindsets and pandemic impact remain under-characterized.
• Understanding these disparities is crucial for targeted mental health interventions.

Purpose of the Study:

• To describe pre- and mid-COVID-19 attitudes, mindsets, drivers, and barriers related to depression across racial/ethnic groups.
• To utilize digital conversations mapped to Health Belief Model (HBM) concepts for analysis.
• To identify specific challenges and facilitators for seeking depression treatment by race/ethnicity.

Main Methods:

• Analysis of 4.2 million open-source digital conversations about depression from US adults (pre-pandemic and mid-pandemic).
• Utilized AI, NLP, and social data mining to categorize conversations by race/ethnicity and map them to HBM constructs.
• Descriptive analysis of conversation themes and sentiment.

Main Results:

• Significant racial/ethnic differences in depression help-seeking drivers and barriers were identified.
• Mid-COVID-19, conversations about depression became more negative, with increased discussion of care barriers.
• Hispanics showed the highest rate of avoidance mindsets; external barriers (stigma, resources) were more prevalent among Hispanics, Black Americans, and Asian Americans.

Conclusions:

• Considerable racial/ethnic disparities exist in depression treatment seeking, amplified by the COVID-19 pandemic.
• Digital conversation analysis, guided by the HBM, effectively organized themes by race/ethnicity.
• Culturally competent, targeted approaches are needed to address barriers and improve mental health help-seeking across diverse populations.