Identifying how COVID-19-related misinformation reacts to the announcement of the UK national lockdown: An interrupted time-series study

Author

Mark Green; Elena Musi; Francisco Rowe; Darren Charles; Frances Darlington Pollock; Chris Kypridemos; Andrew Morse; Patricia Rossini; John Tulloch; Andrew Davies; Emily Dearden; Henrdramoorthy Maheswaran; Alex Singleton; Roberto Vivancos; Sally Sheard

Published

May 9, 2021

Abstract

COVID-19 is unique in that it is the first global pandemic occurring amidst a crowded information environment that has facilitated the proliferation of misinformation on social media. Dangerous misleading narratives have the potential to disrupt ‘official’ information sharing at major government announcements. Using an interrupted time-series design, we test the impact of the announcement of the first UK lockdown (8–8.30 p.m. 23 March 2020) on short-term trends of misinformation on Twitter. We utilise a novel dataset of all COVID-19-related social media posts on Twitter from the UK 48 hours before and 48 hours after the announcement (n = 2,531,888). We find that while the number of tweets increased immediately post announcement, there was no evidence of an increase in misinformation-related tweets. We found an increase in COVID-19-related bot activity post-announcement. Topic modelling of misinformation tweets revealed four distinct clusters: ‘government and policy’, ‘symptoms’, ‘pushing back against misinformation’ and ‘cures and treatments’.

Extended Summary

This research investigates whether the UK’s first national lockdown announcement on 23 March 2020 triggered an increase in COVID-19 misinformation on Twitter. The study analysed a comprehensive dataset of 2,531,888 COVID-19-related tweets from the UK, collected 48 hours before and after the prime ministerial announcement at 20:00-20:30 on 23 March 2020. Using interrupted time-series analysis, the research employed three different statistical models to examine changes in misinformation patterns following this critical policy moment. The methodology involved identifying misinformation through fact-checker websites, detecting automated bot accounts, and conducting topic modelling to categorise different types of misleading content. Contrary to expectations that major government announcements might trigger a surge in false information, the study found no consistent evidence of increased misinformation following the lockdown announcement. While overall COVID-19-related Twitter activity increased by 42% immediately after the broadcast, misinformation tweets actually decreased or remained stable across most analytical models. However, the research identified a notable increase in bot activity post-announcement, with COVID-19-related tweets from automated accounts rising from 30.8% before the announcement to 36.4% on the day and 37.0% the following day. The topic analysis revealed four distinct categories of misinformation circulating during this period: government and policy discussions (22.5%), symptom-related content (31.5%), posts pushing back against misinformation (17.4%), and dangerous cures and treatments (28.6%). Significantly, while most misinformation categories remained stable or declined, content about false cures and treatments increased substantially following the announcement. This finding is particularly concerning given the potential health risks associated with unproven remedies. The research provides valuable insights for public health communication strategies during crises. It suggests that clear, authoritative government announcements may actually help counter misinformation rather than amplify it, at least in the immediate term. However, the concurrent rise in bot activity and dangerous cure-related content highlights ongoing challenges. The study’s innovative approach combines big data analysis with rigorous statistical methods to understand information dynamics during critical policy moments. These findings have important implications for pandemic preparedness, suggesting that coordinated official communication can be an effective tool against misinformation, whilst highlighting the need for targeted monitoring of specific harmful content types, particularly around medical misinformation.

Key Findings

No overall increase in COVID-19 misinformation detected following UK’s first national lockdown announcement on 23 March 2020
Bot activity increased significantly post-announcement, rising from 30.8% to 37.0% of COVID-19-related tweets within 48 hours
Four distinct misinformation categories identified: government policy, symptoms, anti-misinformation content, and dangerous cures and treatments
Misinformation about false cures and treatments increased substantially following the lockdown announcement across all analytical models
Clear government policy announcements may help counter misinformation rather than amplify it in immediate aftermath

Citation

PDF Download BibTeX

@article{green2021identifying,
  author = {Mark Green; Elena Musi; Francisco Rowe; Darren Charles; Frances Darlington Pollock; Chris Kypridemos; Andrew Morse; Patricia Rossini; John Tulloch; Andrew Davies; Emily Dearden; Henrdramoorthy Maheswaran; Alex Singleton; Roberto Vivancos; Sally Sheard},
  title = {Identifying how COVID-19-related misinformation reacts to the announcement of the UK national lockdown: An interrupted time-series study},
  journal = {Big Data \&amp; Society},
  year = {2021},
  volume = {8(1)},
  doi = {10.1177/20539517211013869}
}