Stillbirth Discourse on Instagram and X (Formerly Twitter): Content Analysis

Area of Science:

• Computational Social Science
• Public Health Communication
• Digital Health

Background:

• Stillbirth affects 1 in 160 US deliveries and 1 in 70 globally, causing profound parental grief and psychological distress.
• Societal stigma and lack of awareness exacerbate the impact of stillbirth on affected families.
• No prior comprehensive analysis has explored social media discourse surrounding stillbirth.

Purpose of the Study:

• To analyze stillbirth-related content on Instagram and X (formerly Twitter) using computational methods.
• To identify dominant themes, influential hashtags, sentiments, emotions, visual representations, and misinformation.
• To understand platform-specific communication patterns related to stillbirth.

Main Methods:

• Collected 27,395 stillbirth-related posts from Instagram and X.
• Employed topic modeling (LDA, NMF, BERTopic), network analysis, sentiment/emotion analysis (RoBERTa), and manual image analysis.
• Screened X posts for medical misinformation, particularly concerning COVID-19 vaccines.

Main Results:

• Non-negative matrix factorization (NMF) identified key themes with high coherence scores.
• 27.8% of analyzed tweets linked COVID-19 vaccines to stillbirth, indicating significant medical misinformation.
• Instagram content featured remembrance visuals and text-based posts, while X showed more health-related hashtags; positive sentiments and fear/sadness emotions were prevalent.

Conclusions:

• Instagram facilitates emotional expression and commemoration, while X serves as a platform for public health information.
• Evidence-based communication is vital to counteract fear-based narratives and misinformation on X.
• Platform-specific strategies and content moderation can improve health discourse credibility and support bereaved parents.