But we’re not just fighting an epidemic; we’re fighting an infodemic. Fake news spreads faster and more easily than this virus, and is just as dangerous.” WHO Director-General, Dr Tedros Adhanom Ghebreyesus, Munich 15 February 2020

Vaccine hesitancy fueled by the spread of mis- and disinformation online has led many countries to struggle vaccinating sufficient parts of their population against COVID-19. The ECCMID 2021 seminar “Ethics and conspiracy regarding vaccines” focused on the issue of the current infodemic and approaches for managing it. This series takes a look at three topical questions answered in the seminar. Previously, the focus was on WHO in Question #1: What is an infodemic and how is the WHO fighting it? This time, the focus is on social media:

Hva er situasjonen i Norge? 

Før vi går videre, så ønsker vi å presisere at i Norge er det liten vaksineskepsis i forhold til i mange andre europeiske land. Vi har i Norge høy vaksinasjonsgrad av koronavaksinering og man ser også at vaksinedekningen i barnevaksinasjonsprogrammet har vært godt over 90% i løpet av koronapandemien. Dette viser tallene fra SYSVAK i 2019 og 2020. Vi har likevel valgt å ta med innholdet fra foredraget på ECCMID, da dette temaet er en av de 10 største utfordringene til WHO.

Question #2 Who and what are the drivers behind the success of the anti-vaccine rhetoric?     

In his presentation "The anti-vaccination infodemic on social media”, Federico Germani summarized research comparing the behaviour of pro-vaccine and anti-vaccine supporters in social media and revealed who the “superspreaders” of the anti-vaccination message really are.  

In order to shed light on the polarization of the social media discussion around vaccines, Germani and Biller-Andorno (2021) studied the behaviour of pro-vaccination and anti-vaccination supporters on the social media platform Twitter between September and November 2020. Their aim was to gain a better understanding of the reasons behind vaccine hesitancy and the effectiveness of the anti-vaccination rhetoric.

Anti-vaccination and pro-vaccination profiles on Twitter were identified based on their use of hashtags, with the hashtags #vaccinesharm and #vaccineskill used to detect anti-vaccination supporters, and the hashtag #vaccineswork to identify pro-vaccination supporters. Control individuals were identified using random control hashtags. A total of 50 user profiles were selected for each group.     

The overall Twitter activity of these profiles was analysed by calculating the total Twitter actions, i.e., number of tweets, replies, and retweets published within a month. Using this metric, the anti-vaccination supporters were most active on the platform; however, when looking at tweets alone, the anti-vaccination group was the least active. This suggests that the success of the anti-vaccine supporters’ communication is not due to producing novel content but rather users in the community actively participating in the conversation by replying and retweeting.

Next, the analysis focused on the types of content shared by control, anti-vaccination, and pro-vaccination profiles. Pro-vaccination supporters were most actively sharing science- and vaccine-related contents, whereas anti-vaccination support was found to be associated with belief in conspiracy theories and increased use of emotional content. Furthermore, the analysis of language used by each group in their content showed that anti-vaccination supporters used more political language as well as phrasal expressions. The group producing the most engaging original content was identified to be the pro-vaccine group, with pro-vaccine supporters leading in both average engagement per tweet as well as in the average number of followers per individual profile. Interestingly, the use of emotional language was shown to increase the engagement on a tweet in the pro-vaccine group, but not in the anti-vaccine group.

Finally, the researchers hypothesised that the success of the anti-vaccination rhetoric was largely due to an increased sense of community among the anti-vaccine supporters. This hypothesis was supported by an analysis of the virtual network of pro-vaccine and anti-vaccine profiles: the virtual community of anti-vaccine supporters was better interconnected, with more major influencers. At the time of the analysis, the anti-vaccine group included 15 major influencers, led by some of the most prominent political figures of the world. Using the same criteria, the pro-vaccination network included the World Health Organisation (WHO) as the only major influencer.

In conclusion, the findings of this study showed that 1) emotional language could improve pro-vaccine supporters’ communication, and that 2) anti-vaccination supporters’ activity is mainly based on sharing content from key influencers in their community, as led by political figures and high-profile anti-vaccination supporters. 

The final part of this series will dive into the ethics of infodemic response with: Question #3: Infoveillance, censorship and behavioral manipulation — what are the ethical implications of infodemic management approaches?

ECCMID 2021 references:

The anti-vaccination infodemic on social media. Federico Germani (Zurich, Switzerland). Presented online at ECCMID 2021 on 10  July 2021. 

Additional references:

Folkehelseinstituttet, Barnevaksinasjonsprogrammet i Norge - rapport 2019 og 2020 (fhi.no), rapport 2021.
Folkehelseintituttet, Koronavaksinasjons-statistikk - FHI, oppdatert 5.10.2021, lest 06.10.2021

Germani F, Biller-Andorno N (2021) The anti-vaccination infodemic on social media: A behavioral analysis. PLOS ONE 16(3): e0247642. https://doi.org/10.1371/journal.pone.0247642