JMIR Publications

ABSTRACT

Background:

Vaccine hesitancy is a global public health threat. Vaccine opinion pieces published on microblogging sites have affected individuals’ decisions on whether to receive a vaccine. Understanding the distribution of such information is critical; however, related studies are limited. This study examined the dissemination of provaccine and antivaccine messages on Twitter.

Objective:

To analyze the determinants of provaccine and antivaccine information dissemination by using the elaboration likelihood model (ELM) and to identify the relative value of the central and peripheral routes in decisions to share information.

Methods:

English-language tweets from the United States that contained provaccine and antivaccine hashtags (n = 164,109) were collected between April 26 and August 26, 2021. Logistic and negative binomial regressions were conducted to predict retweet outcomes. The content-related central route predictors were measured by the number of hashtags and mentions, emotional valence and intensity, and concreteness. The source-related peripheral route predictors were measured by the number of favorites and followers and whether the source was verified.

Results:

The central route has a prominent role in shaping the retweet decisions of antivaxxers. For antivaccine messages, positive valence and concreteness are strongly associated with an increase in retweeting (valence: OR=1.31, IRR = 1.26, P=.004; concreteness: OR=1.15, IRR = 1.16, P <.001). However, these 2 factors had either no or only small effects on sharing provaccine tweets. For provaccine messages, emotional intensity marginally decreased retweeting (OR=0.95, P=.01; IRR = 0.96, P=.02). In addition, the number of “likes” and followers increased the likelihood of dissemination for both sets of tweets.

Conclusions:

This study indicates that antivaxxers process vaccine information more cognitively than provaxxers. This finding raised the concern that the widespread antivaccine information in the US may continue to prevail, because decisions formed through central route processing tend to last. These findings signify the value of message framing and leveraging the characteristics of source information to devise effective communication strategies to promote vaccination. Based on these results, this study proposes recommendations for COVID-19 vaccination campaigns targeting provaccine and antivaccine groups. Clinical Trial: Not applicable.