Tweets, likes, and needles: Study finds social media messaging spurs vaccine intentions

Despite the demographic limitations, the study’s use of authentic social media content strengthens its practical implications. The authors argue that health communicators should prioritize message strategies that promote self-efficacy and prompt cues to action, particularly when deploying campaigns through influencers or institutional accounts.

A new study has found that social media posts emphasizing personal empowerment and ease of access, called self-efficacy, significantly increased users’ intentions to receive the COVID-19 vaccine. The research, published in Frontiers in Digital Health, offers some of the clearest evidence yet that specific message strategies and engagement behaviors on platforms like Twitter (now 'X') can shape real-world health decisions.

Led by Xiaofeng Jia of Bowling Green State University, with co-authors Soyeon Ahn and Susan E. Morgan of the University of Miami, the study surveyed 1,141 U.S. adults who were not fully vaccinated. Participants were exposed to actual tweets related to COVID-19 vaccination, authored by individuals from various demographic and professional backgrounds. Researchers then analyzed how the posts’ content, framing, and authorship influenced users’ engagement behaviors, likes, shares, and comments, as well as their intentions to get vaccinated or persuade others to do so.

The study, titled "The role of social media messages and content creators in shaping COVID-19 vaccination intentions", found that tweets highlighting self-efficacy, such as messages suggesting that getting vaccinated is easy or within the user’s control, were positively associated with higher vaccination intentions. Content invoking this message strategy had a measurable impact, even when controlling for demographic factors and prior beliefs.

Perceived barriers, such as concerns over vaccine safety, were negatively associated with vaccination intent. In contrast, perceived cues to action, external prompts like advice from others or social encouragement, were also strong positive predictors.

Beyond message content, the study examined how different forms of user engagement influenced behavior. Liking a post emerged as a statistically significant mediator between health perceptions and vaccination decisions. Users who believed they were susceptible to COVID-19 or who reported exposure to cues to action were more likely to like posts, and this behavior in turn predicted a greater likelihood of vaccine uptake.

The researchers also assessed how content creator characteristics like profession, gender, and race shaped user response. Tweets from celebrities received significantly more favorites than those from other groups, while posts authored by politicians drew more comments. However, these traits did not directly influence users’ willingness to get vaccinated. The study concluded that while public figures increase visibility and engagement, message framing remains the primary driver of behavioral intention.

The analysis was grounded in the Health Belief Model (HBM), a psychological framework commonly used in health communication. The study tested six HBM constructs: perceived susceptibility, severity, benefits, barriers, cues to action, and self-efficacy. Although all constructs played a role in user engagement, only self-efficacy and cues to action showed consistent positive effects on vaccination intention. Perceived susceptibility and benefits were positively associated with sharing and commenting behavior, but did not directly predict vaccine uptake.

Structural equation modeling revealed that engagement behaviors, particularly the number of “favorites” given to posts, fully mediated the relationship between HBM perceptions and both vaccination intent and the intention to persuade others to get vaccinated. This finding points to the value of tracking engagement metrics as proxies for public health sentiment and persuasion outcomes.

The researchers used real tweets from the CoVaxxy dataset, a publicly available repository of COVID-19 vaccine-related social media content. Each participant viewed 10 tweets representing different HBM message strategies and then responded to questions measuring their engagement intentions and health beliefs. The sample was drawn from Amazon Mechanical Turk and skewed heavily toward educated, white, and middle-income respondents, limiting generalizability.

Despite the demographic limitations, the study’s use of authentic social media content strengthens its practical implications. The authors argue that health communicators should prioritize message strategies that promote self-efficacy and prompt cues to action, particularly when deploying campaigns through influencers or institutional accounts.

The study did not find evidence that message strategies emphasizing severity or susceptibility directly influenced vaccination intentions, although they did increase engagement behaviors. Future research, the authors suggest, should explore the role of political identity, as well as the effects of emerging technologies such as AI-generated influencers and immersive media platforms.