#SARSCOV2: ANALYSIS OF AN INTERNATIONAL COVID-19-RELATED DIGITAL COMMUNITY

ABSTRACT

BACKGROUND: Twitter is playing a pivotal role in rapid communication between healthcare stakeholders, governments, and the public during the COVID-19 pandemic. Communication on Twitter can reach over 152 million registered daily users and over 500 million people visit the site monthly without logging into an account. The purpose of this study was to analyze tweets associated with #SARSCoV2 during the initial phase of the COVID-19 pandemic. METHODS: Symplur Signals, a healthcare social media analytics platform, was used to analyze all publicly available #SARSCoV2 tweets (excluding spam) between 2/11/20-4/12/20. Tweet activity, content, sentiment, associated hashtags, user characteristics/engagement, and network analysis were analyzed. Select metrics were compared before/after the COVID-19 pandemic declaration by the World Health Organization (WHO;3/11/20). Tweet sentiment was analyzed within 3 hours before/after specific timepoints. RESULTS: Exactly 894,983 tweets (73% retweets) by 400,653 users met study inclusion criteria, resulting in 2,998,817,360 impressions. Daily tweet activity peaked on 3/18/20 with 58,666 tweets. There was a 2677% increase in number of tweets from the first to last week of the study period. The average user had 2.2 tweets. There were 292,478 (33%) and 264,198 (30%) tweets containing links and media, respectively. Trending term analysis before/after the WHO pandemic declaration showed a shift in conversation across the study period toward trending terms such as patients, critical care, studies, and antibodies. The top 15 hashtags associated with #SARSCoV2 predominantly included COVID-19 related hashtags. The top 3 countries by users were the United States, Spain, and Indonesia. The top 100 influencers consisted primarily of researchers/academics (30.1%), doctors (15.1%), journalists/media (9.4%), and media organizations (9.4%). Twitter network analysis showed key central hubs of communication were researchers/academics, doctors, and journalists/media. Three hours before/ after sentiment analysis revealed that after the Unites States announcement of European travel restrictions, negative sentiment tweets rose from 41% to 52%. After the Center for Disease Control's public mask use recommendation, negative sentiment decreased from 65% to 55%. CONCLUSIONS: Our study demonstrated that #SARSCoV2 rapidly coalesced into a vital digital international community for the COVID-19 pandemic with tweets from #SARSCoV2 generating nearly 3 billion impressions across a worldwide Twitter community. Notably, the most influential members of this community were researchers/academics and doctors. Quantifiable shifts in community sentiment were observed immediately following key changes in government policy. Changes in trending terms across the study period suggests #SARSCoV2 users engage in real-time discussions concurrent with recent scientific and medical discourse. Future studies should examine changes in digital health communication over the course of the COVID-19 pandemic.