Exploring mental health literacy on twitter: A machine learning approach

Area of Science:

• Social Sciences
• Psychology
• Public Health

Background:

• Mental health literacy (MHL) is crucial for addressing mental health challenges.
• Understanding the pathways through which MHL components influence help-seeking attitudes is essential for effective interventions.

Purpose of the Study:

• To investigate the mediating roles of stigma reduction, help-seeking efficacy, and positive mental health maintenance in the relationship between recognizing mental disorders and help-seeking attitudes.
• To examine the temporal relationships among MHL components using social media data.

Main Methods:

• Analysis of 4,471,951 tweets from 941 users collected between April and August 2022, with data organized into three time points.
• Utilized a Sentence-BERT model for tweet classification and structural equation modeling to analyze mediation pathways.
• Keywords aligned with five MHL facets: maintaining positive mental health (M), recognizing mental disorders (R), reducing mental illness stigma (S), help-seeking attitude (HA), and help-seeking efficacy (HE).

Main Results:

• Recognition of mental disorders positively influenced positive mental health maintenance, help-seeking efficacy, and stigma reduction, all linked to better help-seeking attitudes.
• In multiple mediation models, stigma reduction did not significantly mediate the link between recognizing mental disorders and help-seeking attitudes.

Conclusions:

• Recognizing mental disorders impacts help-seeking attitudes primarily through enhanced help-seeking efficacy and positive mental health maintenance.
• Findings offer insights for developing targeted interventions and policies to promote mental health literacy and encourage help-seeking behaviors.