JMIR Publications

ABSTRACT

Background:

Social media are important channels for health and risk communication, but platforms may impact messaging and engagement differently. Current studies often focus on a single platform, and the media effects on message design and audience engagement have been scarcely explored in the context of Covid-19 or health and risk communication more broadly.

Objective:

This study measures platform effects of Twitter and Facebook by comparing message designs and audience engagement during the Covid-19 pandemic to advance theories of health messaging on social media, and provide recommendations for risk and health communication policies.

Methods:

We adopted a framework of social media message features to analyze posts of United States public health agencies on both Facebook and Twitter. We retrieved all Covid-19 related posts from major federal agencies related to health and infectious disease, all major state public health agencies and selected local public health departments. A total of 100,785 posts related to Covid-19, from 179 different accounts of 96 agencies, were retrieved in 2020. For the analysis, we subsampled 1,677 posts from 95 unique agencies, from Twitter and Facebook, which were coded and assessed via Cohen’s Kappa inter-rater reliability measure. We calculated the prevalence of message elements across the platforms and assessed the statistical significance of differences. We also calculated normalized measures of shares and likes and evaluated the association between message elements with the two measures of user engagement.

Results:

Results show that the distribution of message elements and audience engagement are largely similar across both sites. However, political figures, experts and personalities were significantly more present on Facebook posts compared to Twitter. Moreover, posts on Facebook with photographs and emotionally-related rhetorical strategies received more audience engagement, what we term the local and personal orientation of Facebook. Infographics, surveillance information and certain multi-media elements (e.g., hyperlinks) were more prevalent on Twitter. Moreover, posts related to policy information were more engaged with on Twitter, which we term the data and policy orientation of Twitter. We also observed overall higher engagement with content on Facebook compared to Twitter. Surprisingly, scientific information and correctives were not largely represented in the messages, and different types of images were associated with distinct levels of audience engagement.

Conclusions:

This study found potential platform effects of Twitter and Facebook on message design and audience engagement, although also a large degree of similarities across platforms. This study provides novel evidence on cross-platform characteristics of health messaging, the role of several message features for engagement, and provides evidence-based policy recommendations for health and risk communication by public health agencies.