Online misinformation negatively impacts COVID-19 vaccination campaign, study shows

Reviewed by Emily Henderson, B.Sc.May 2 2022

Online misinformation has negatively impacted the US COVID-19 vaccination campaign. This emerges from a study by Francesco Pierri, researcher at the Politecnico di Milano, in collaboration with Indiana University as part of the H2020 Periscope project, published in Nature Scientific Reports.

The objective of this study is to demonstrate whether or not there are statistically significant associations between the quality of information consumed online and the "anti-vax" sentiment of the US population, with consequent repercussions on the vaccination campaign.

The study confirmed a statistically significant association between the amount of misinformation shared online and the tendency to refuse or delay vaccination in the United States. In particular, in the states and counties where the most online misinformation is consumed, there is greater vaccination hesitancy and, consequently, lower vaccination coverage.

Since the beginning of 2021, researchers at the Politecnico di Milano and the Observatory on Social Media (OSOME) have collected millions of posts shared on Twitter related to vaccines, with the aim of studying the effects of unreliable and/or inaccurate information on the US vaccination campaign, which began at the end of 2020.

Using a list of news sites tagged by journalists, fact-checkers, and other academics as portals spreading false and unreliable news, the researchers identified millions of posts with potentially harmful content (e.g. articles claiming that vaccines don't work or cause death) shared by millions of Twitter users in early 2021 who geolocated themselves in the various states and counties of the United States.

In order to measure people's willingness or not to get vaccinated, the researchers used millions of responses to daily polls administered on Facebook in which they asked geolocated users whether or not they intended to get vaccinated.

The results of the multiple linear regression model, which includes other socio-economic variables, such as the average wealth and ethnic composition of each state/county, show that the proportion of misinformation shared on average by users in a given area is positively correlated with the proportion of people who declare they have no intention of getting vaccinated and, similarly, negatively correlated with the number of vaccine doses administered."

Francesco Pierri, Researcher, Department of Electronics, Information and Bioengineering (DEIB), Politecnico di Milano

These results and other statistics on the online conversations related to the COVID-19 vaccination campaign can be found on the dashboard associated with the CoVaxxy project https://osome.iu.edu/tools/covaxxy.

Many studies published during the COVID-19 pandemic have highlighted how the "infodemic" of false or misleading news about the virus has slowed the efforts made by governments to reduce the infection, from the refusal to wear masks to the breaching of restrictions.