key: cord-0933335-wo0pfpq9
authors: Rufai, Sohaib R; Bunce, Catey
title: World leaders’ usage of Twitter in response to the COVID-19 pandemic: a content analysis
date: 2020-04-20
journal: J Public Health (Oxf)
DOI: 10.1093/pubmed/fdaa049
sha: 92fbcf06543d672159672fb123afa017f176cf3c
doc_id: 933335
cord_uid: wo0pfpq9

BACKGROUND: It is crucial that world leaders mount effective public health measures in response to COVID-19. Twitter may represent a powerful tool to help achieve this. Here, we explore the role of Twitter as used by Group of Seven (G7) world leaders in response to COVID-19. METHODS: This was a qualitative study with content analysis. Inclusion criteria were as follows: viral tweets from G7 world leaders, attracting a minimum of 500 ‘likes’; keywords ‘COVID-19’ or ‘coronavirus’; search dates 17 November 2019 to 17 March 2020. We performed content analysis to categorize tweets into appropriate themes and analyzed associated Twitter data. RESULTS: Eight out of nine (88.9%) G7 world leaders had verified and active Twitter accounts, with a total following of 85.7 million users. Out of a total 203 viral tweets, 166 (82.8%) were classified as ‘Informative’, of which 48 (28.6%) had weblinks to government-based sources, while 19 (9.4%) were ‘Morale-boosting’ and 14 (6.9%) were ‘Political’. Numbers of followers and viral tweets were not strictly related. CONCLUSIONS: Twitter may represent a powerful tool for world leaders to rapidly communicate public health information with citizens. We would urge general caution when using Twitter for health information, with a preference for tweets containing official government-based information sources.

The COVID-19 outbreak has been declared a pandemic by the World Health Organization, because of 'alarming levels' of spread, severity and inaction. 1 As of 25th March 2020, there are over 414 000 cases globally and over 18 400 deaths. 2 The causative novel human coronavirus, severe acute respiratory syndrome (SARS)-CoV-2, is primarily transmitted by respiratory droplet spread and direct contact; moreover, it can also survive in air. 3, 4 Transmission is also possible in asymptomatic carriers. 5 It is therefore crucial that world leaders mount effective public health measures to limit human-tohuman transmission, including hand hygiene, social distancing and self-isolation. 1, 6, 7 Twitter may represent a powerful public health tool for world leaders to rapidly and directly communicate information on COVID-19 to citizens, in addition to more conventional media such as television, radio and newspapers. Twitter is a free microblogging social media website with 152 million registered daily users. 8 Furthermore, over 500 million people visit Twitter per month without logging into an account. 9 The Group of Seven (G7) is an international intergovernmental organization, which comprises the seven most advanced economies in the world according to the International Monetary Fund: Canada, France, Germany, Italy, Japan, the UK and the USA. 10 The European Union (EU) is also an invitee to the G7, represented by the Presidents of the EU Council and EU Commission. The language used by world leaders, such as the G7 heads of state, can influence the opinions and behaviour of citizens. 11 Furthermore, the nature of communication by political leaders affects public policy outcomes. 12 Thus, political leaders can use their influence over public opinion to address public health issues, ideally by disseminating evidence-based public health information. 13 A MEDLINE search for the keywords 'COVID-19' and 'Twitter' delivered no results. Thus, to our knowledge, this qualitative study is the first to evaluate the role of Twitter as used by G7 world leaders in response to COVID-19. 10, 14 Content analysis is performed exclusively for 'viral' tweets, meaning those which rapidly spread across a high number of users. 15 Primary research objective: To explore how G7 world leaders are using Twitter in response to the COVID-19 pandemic.

Secondary research objectives:

• To determine what proportion of G7 world leaders use Twitter and assess their reach (followers); • To perform content analysis for all G7 world leaders' viral tweets and classify into appropriate themes; • To evaluate associated characteristics of these tweets, including weblinks to official government-based information sources and use of video; and • To determine what proportion of these tweets are in non-English languages.

Verified Twitter accounts 16 were sought for the following G7 members:

• The advanced Twitter search function was used to filter by the following inclusion criteria: keywords 'COVID-19' or 'coronavirus'; search dates 17th November 2019-17th March 2020 (inclusive); and viral tweets. There is no strict definition for what constitutes a 'viral' tweet-we stipulated a minimum of 500 'likes'. Primary tweets meeting these criteria, plus any secondary tweets as part of a viral thread, were included in our analysis. No language restrictions were stipulated; Twitter's built-in Google translate function was used for non-English tweets. 17 Exclusion criteria were as follows: non-verified Twitter accounts and non-viral tweets.

Data were analysed on 18th March 2020 from a UK IP address. The following outcome measures were recorded: name, number of followers, number of viral tweets, number of associated weblinks to official government-based information sources, top tweet characteristics, languages used and number of viral videos with views. Top tweet characteristics included 'likes', 'retweets' and 'comments'. A 'like' indicates that a registered Twitter user appreciates the tweet, while a 'retweet' indicates that they have shared the tweet on their own Twitter page and a 'comment' indicates that they have added written commentary for the tweet. All of these actions increase the reach of the viral tweet in question. Content analysis was performed by the lead author (SRR) to identify and categorize viral tweets into key themes.

This study has been reported as per the Standards for Reporting Qualitative Research. 18 No ethical approval was required as all data (tweets) were retrieved from the public domain.

Verified Twitter accounts were identified for eight out of nine (88.9%) G7 members; no active account was identified for Angela Merkel, Chancellor of Germany. Results for all outcome measures are displayed in Table 1 . The included G7 members had a total following of 85.7 million Twitter users, and 203 viral tweets on COVID-19. 56 (27.6%) viral tweets were in non-English languages, including French, German, Italian and Japanese. All non-English tweets were successfully translated into English using Twitter's built-in Google translate function. 17 Content analysis revealed three key themes for viral tweets:

(i) Informative-seeking to share information or updates. (ii) Morale-boosting-seeking to boost morale or galvanize. (iii) Political-seeking to raise a point of political debate.

In any cases of overlap, the majority of the tweet content dictated its thematic classification. Out of 203 viral tweets, 166 (82.8%) were 'Informative', of which 48 (28.6%) had associated weblinks to official government resources, while 19 (9.4%) were 'Morale-boosting' and 14 (6.9%) were 'Political'. All 'Political' tweets were sent from the Twitter account of Donald Trump, President of the US. Additionally, 19 viral videos were embedded within Informative viral tweets, with a total of 12.4 million views. Figure 1 displays the number of followers and breakdown of viral tweet themes per G7 world leader for comparison. Table 2 displays examples of viral tweets per theme.

To our knowledge, this is the first study of world leaders' Twitter usage in response to COVID-19. Our findings demonstrate that Twitter can be a powerful tool to rapidly 

The communication and language used by world leaders can influence the opinions and behaviour of the public. 11 World leaders can use their influence to address crises, especially by disseminating evidence-based information. 13 On the bicentennial of her birth, consider the example of Florence Nightingale's extraordinary use of publicity and statistics to promote public health, reform hospitals, inspire and educate. 20 It is essential however to consider the impact of misinformation and propaganda during times of crisis and conflict. For example, the atrocious use of propaganda by the Nazis led to the demonization and mass murder of Jewish people. 21 Misinformation may also negatively impact views of the public on health matters. A recent content analysis of data from The Washington Post revealed a plurality of misleading statements made by the US President Donald Trump on the subject of healthcare, particularly relating to the Affordable Care Act. 22 However, one would hope that in instances where official government-based sources are referenced or signposted, the risk of misinformation should be reduced.

Although no studies of Twitter during COVID-19 were revealed by our MEDLINE search, several previous studies have suggested a role for Twitter data during other public health outbreaks including Middle East respiratory syndrome, 23 SARS, 24 Ebola virus disease, 15 Zika virus, 25 H1N1 ('swine flu') 26 and H7N9 ('avian flu'). 27 Suggested roles for Twitter include infectious disease surveillance, predicting spread of disease, dissemination of public health information and assessing public views toward public health outbreaks. 15,

To our knowledge, this study is the first to analyze the usage of Twitter by G7 world leaders in response to COVID-19. Our search strategy replicates a typical search by a member of the public seeking tweets from on the subject of COVID-19, using the keywords 'COVID-19' and 'coronavirus'. This study has determined that the majority of tweets fell within the 'Informative' category. We feel that these 'Informative' tweets, followed by 'Morale-boosting' tweets, are generally more helpful than 'Political' tweets from a public health perspective. Our study has also determined the proportion of non-English tweets by G7 world leaders and highlighted that this language barrier was easily overcome through Twitter's inbuilt Google translate function. 17 We have evaluated the characteristics of viral tweets, including the use of government weblinks and viral videos to share information. Conducting this study highlighted the user-friendly nature of Twitter in searching information on COVID-19, due to concise content (limited to 280 characters per tweet) and the ability to quickly translate non-English tweets where necessary.

Our study found that the G7 leaders have a large following on Twitter-a combined total of 85.7 million followers. A disproportionately high number of followers were attributed to the Twitter account of US President Donald Trump. This may in part be due to his previous celebrity status through reality television and popular culture. Indeed, musician Justin Bieber has 111 million followers as of 26th March 2020considerably more than the G7 leaders combined. However, our study demonstrated that the number of followers does not necessarily translate into more viral tweets-indeed, highest number of COVID-19 viral tweets were sent from the account of Justin Trudeau, President of Canada. This is because it is not necessary to follow a Twitter account in order to view, like, retweet or comment on a tweet sent from said account. Furthermore, the 19 viral videos embedded within 'Informative' tweets attracted a total of 12.4 million views, indicating that the number of 'likes' alone underestimates the true reach of viral tweets.

Interestingly, this study highlighted that Angela Merkel, Chancellor of Germany, is the only G7 world leader who does not appear to have a verified and active Twitter account. Thus, the use of Twitter is not crucial for world leaders due to alternative forms of communication including television, radio, newspapers and indirect inclusion in social media by other users.

Our methodology could also be applied to a range of other Twitter user groups in response to COVID-19 or other health issues. These might include other politicians, celebrities, organisations, charities and other entities. Interestingly, Twitter is censored in the People's Republic of China, so this study's methodology cannot be applied. 28

This study has a number of limitations. First, this study likely underestimates the true reach of Twitter as these G7 members' tweets could have been viewed without 'liking' them or even needing an account. Furthermore, in-depth analysis of the reach of these tweets is beyond the remit of this study. We also acknowledge that the resulting themes were decided by the first author, albeit he is trained in thematic analysis. Others might apply different terms to certain content. Our study only included tweets with the keywords 'COVID- 19' or 'coronavirus' to maintain a focused search. It is however possible that some relevant tweets were missed if they did not specifically include these words. Our study only included G7 members, on the basis that they represent prominent world leaders across several nations. Further research could evaluate the usage of Twitter by other politicians from different countries. Finally, the threshold of 500 likes to qualify as a viral tweet was an arbitrary one determined a priori in the absence of a strict definition from the literature, albeit it resulted in a good range of data across all included G7 member nations.

This study has demonstrated the widespread use of Twitter by the majority of G7 world leaders with many viral tweets on the subject of COVID-19. The majority of viral tweets were 'Informative', and many had weblinks to official governmentbased sources. Twitter may represent a powerful tool for world leaders to rapidly communicate with citizens during public health crises. We would urge general caution when using Twitter as a public health information source and give preference to tweets containing official government-based health information and advice.

No conflicting relationship exists for any author.

The authors have no proprietary or commercial interest in any materials discussed in this article.

World Health Organization. WHO Director-General's Opening Remarks at the Media Briefing on COVID-19 -11

COVID-19): Situation Report -65

Clinical characteristics of coronavirus disease 2019 in China

Aerosol and surface stability of SARS-CoV-2 as compared with SARS-CoV-1 (published online ahead of print

Presumed asymptomatic carrier transmission of COVID-19

Responding to Covid-19 -a once-in-a-century pandemic?

Estimating the reproductive number and the outbreak size of novel coronavirus disease (COVID-19) using mathematical model in

Fourth Quarter

Testing Promoted Tweets on Our Logged-Out Experience

International Monetery Fund. Major Advanced Economies (G7). 2020

Communication at the core of effective public health

The Nature and Origins of Mass Opinion

Political will: a bridge between public health knowledge and action

Statement on COVID-19

How did Ebola information spread on twitter: broadcasting or viral spreading?

About Verified Accounts

About Tweet translation

Standards for reporting qualitative research: a synthesis of recommendations

Advice for Everyone: Coronavirus (COVID-19)

Feminism and hospital reform

Hitler's Army: Soldiers, Nazis, and War in the Third Reich

President trump and health care: a content analysis of misleading statements

High correlation of Middle East respiratory syndrome spread with Google search and twitter trends in Korea

ProMED-mail: 22 years of digital surveillance of emerging infectious diseases

Use of twitter data to improve Zika virus surveillance in the United States during the 2016 epidemic

Novel insights into views towards H1N1 during the 2009 pandemic: a thematic analysis of twitter data

Social media and outbreaks of emerging infectious diseases: a systematic review of literature

Information Operations Directed at Hong Kong

The authors would like to thank all healthcare professionals worldwide facing the COVID-19 pandemic. SRR's post is