JMIR Publications

ABSTRACT

Background:

2021 was marked by vaccinations against COVID-19, and spurred a wider discussion from the general population, with some in favor and some against it. Twitter, a popular social media platform, was instrumental in providing information about the COVID-19 vaccine, and has been effective in observing public reactions. We focus on tweets from Japan and Indonesia, two countries with a large Twitter-using population, where concerns about side effects were consistently stated as a strong reason for vaccine hesitancy.

Objective:

This study aimed to investigate the reporting of vaccine side effects on Twitter, and to compare the side effects from two mRNA vaccines types, Pfizer and Moderna, in Japan and Indonesia.

Methods:

A final data were obtained from Twitter from January 1, 2021, to December 31, 2021, and consisted of N = 433,024 tweets (Japan) and N = 16,282 tweets (Indonesia). Logistic regression models were used to compare the side effects for Pfizer and Moderna for each country. We also compared these results to side effect ratios in public reports (press releases) released by Pfizer and Moderna after clinical trials.

Results:

There were some differences in the ratio between the public reports and tweets. Specifically, fever was mentioned much more frequently in tweets than what would be expected based on the public reports. For logistic regression results, all side effects were significant for Japanese tweets (P value<.001), and all but two were significant in Indonesian tweets (P value <.01). The odds ratio suggested that the terms for side effects were more frequent in Moderna than Pfizer tweets.

Conclusions:

Fever was the most prevalent side effects mentioned in Japanese and Indonesian tweets. Compared to Pfizer, side effects were observed more frequently in Moderna tweets. We note possible applications for COVID-19 vaccine side effect surveillance on Twitter, but caution that there is a need for additional resources to establish validity.