Engagement Trends in Online Vaccine Content: Longitudinal YouTube Study

Area of Science:

• Social Media Analysis
• Public Health Communication
• Digital Epidemiology

Background:

• YouTube hosts diverse viewpoints on vaccination, including authoritative health information and vaccine-skeptic content.
• Understanding engagement dynamics is crucial for addressing misinformation and promoting public health.

Purpose of the Study:

• To investigate the temporal and textual engagement dynamics of YouTube viewership with vaccination content.
• To analyze content favoring or opposing vaccination, considering channel authority and platform moderation.

Main Methods:

• A 6-month longitudinal analysis of 7213 vaccine-related YouTube videos.
• Zero-shot large language model classification with manual verification for stance detection.
• Bayesian regression modeling for engagement and disagreement dynamics.

Main Results:

• Vaccine-hesitant videos exhibit significantly higher raw and normalized engagement rates compared to pro-vaccine content.
• Vaccine-hesitant content achieves cumulative views faster, indicating rapid early-stage amplification.
• Comment analysis reveals echo chambers around hesitant content and battlegrounds around pro-vaccine content.

Conclusions:

• Vaccine-hesitant content can dominate YouTube's engagement ecosystem through rapid amplification.
• Findings have direct implications for the timing of public health interventions.
• Platform governance policies need to address the rapid spread of vaccine-hesitant viewpoints.