Low credibility URL sharing on Twitter during reporting linking rare blood clots with the Oxford/AstraZeneca COVID-19 vaccine

Abstract

The COVID-19 pandemic was accompanied by an “infodemic” of misinformation. Misleading narratives around the virus, its origin, and treatments have had serious implications for public health. In March 2021, concerns were raised about links between the Oxford/AstraZeneca (AZ) COVID-19 vaccine and recipients developing blood clots. This paper aims to identify whether this prompted any reaction in the diffusion of low-credibility COVID-19-relate information on Twitter. Twitter’s application programming interface was used to collect data containing COVID-19-related keywords between 4th and 25th March 2021, a period centred on the peak of new coverage linking rare blood clots with the AZ vaccine. We analysed and visualised the data using temporal analysis and social network analysis tools. We subsequently analysed the data to determine the most influential users and domains in the propagation of low-credibility information about COVID-19 and the AZ vaccine. This research presents evidence that the peak of news coverage linking rare blood clots with the AZ vaccine correlated with an increased volume and proportion of low-credibility AZ-related content propagated on Twitter. However, no equivalent changes to the volume, propagation, or network structure for the full dataset of COVID-19-related information or misinformation were observed. The research identified RT.com as the most prolific creator of low-credibility COVID-19-related content. It also highlighted the crucial role of self-promotion in the successful propagation of low-credibility content on Twitter. The findings suggest that the simple approach adopted within the research to identify the most popular and influential sources of low-credibility content presents a valuable opportunity for public health authorities and social media platforms to develop bespoke strategies to counter the propagation of misinformation in the aftermath of a breaking news event.