What Social Media Told Us in the Time of COVID-19: A Scoping Review

"Our adopted framework can serve as a fundamental and flexible guideline when studying social media and epidemiology."

Emerging infectious diseases tend to result in increased usage and consumption of media of all forms by the general public in their quest for information. During the COVID-19 pandemic, social media has been a pivotal communication tool for governments, organisations, and universities to disseminate information to the public. In addition, user-generated information, which can be subjective or inaccurate, has flowed on these platforms. In light of the fact that social media plays a role in people's perception of disease exposure, resultant decision making, and risk behaviours, the researchers conducted a global scoping review of peer-reviewed empirical studies relating to COVID-19 and social media from November 2019 to November 2020.

Among the studies included in the analysis, data from Twitter (45 articles) and Sina Weibo (16 articles) were the most frequently studied. From the total of 81 articles analysed, all of which were published in 2020 (see Figure 1 in the paper for details), the researchers identified six themes and organised them under three categories using a modified version of the Social Media and Public Health Epidemic and Response (SPHERE) framework (see Figure 2 in the paper). Here is a brief summary:

• 1. Social media as contagion and vector: identifying infodemics - Sample finding: Islam and colleagues analysed 2,311 infodemic reports that were related to COVID-19 from Dec 31 2019 to April 5 2020 and showed that misinformation was mainly driven by rumours, stigma, and conspiracy theories that were circulating on various social media and other online platforms. Government responses that were distributed via social media have been increasingly crucial in combating infodemics and promoting accurate and reliable information for the public. However, little has been studied about how efficient and effective these official responses are at leading to public belief or behavioural changes. Associations between infodemic and bot activities on social media are another possible future research direction.
• Social media for surveillance and monitoring:
  • 2. Surveying public attitudes - 48 selected articles gauged the attitudes and emotions that were expressed by social media users regarding the COVID-19 pandemic, mainly by use of content and sentiment analysis. Public attitude can be further divided into the following sub-themes: public sentiment towards the COVID-19 pandemic and interventions, stigma and racism, and ageism. Overall, articles regarding public attitudes towards the COVID-19 pandemic have shown sentiments that shifted over time. This theme can be a useful indicator when evaluating interventions, such as physical distancing and wearing masks, that aim to reduce the risk of COVID-19 infection. However, public sentiments had not been incorporated into many intervention studies at the time of this review. "When a disease, such as COVID-19, starts spreading and causing negative sentiments, timely, proper, and effective risk communication is needed to help ease people's anxiety or negative attitudes regarding the COVID-19 pandemic, especially through social media."
  • 3. Assessing mental health - Studies showed that social media data were useful to detect mental health issues at the population level. Due to the early outbreak of COVID-19 and the prevalence of social media use (e.g., Sina Weibo and WeChat) in China, two studies reported increased issues of mental health among the Chinese population. A similar trend of deteriorating mental health could happen in other regions.
  • 4. Detecting or predicting COVID-19 cases - Unlike other infectious diseases, COVID-19 has not had real-time monitoring surveillance developed with social media data. It is possible that the pandemic has evolved so rapidly that finding COVID-19 vaccinations or therapies has been prioritised over real-time monitoring surveillance with social media. In addition, scarcity of accurate and reliable data sources might discourage the development of the COVID-19 real-time surveillance. Moreover, whether COVID-19 is a one-time event or will become seasonal, like influenza, is unknown. "If COVID-19 becomes seasonal, then it might be meaningful and useful to establish a real-time model to monitor the disease by use of social media data."
• Social media as disease control:
  • 5. Analysing government responses to the pandemic - Six of 81 articles examined how government messages and health education material were generated and consumed on social media platforms. Sample finding: In a Canadian study, Merkley and colleagues found that, regardless of party affiliation, members of parliament emphasised the importance of measures for physical distancing and proper hand-hygiene practices to cope with the COVID-19 pandemic, without tweets exaggerating concerns or misinformation about COVID-19. "To inoculate the public against misinformation, public health organisations and governments should create and spread accurate information on social media because social media has had an increasingly important role in policy announcement and health education."
  • 6. Evaluating quality of health information in prevention education videos - Eight chosen studies investigated the quality of YouTube videos with COVID-19 prevention information. YouTube has served as one of the major platforms to spread information concerning the control of COVID-19. Nonetheless, the chosen studies showed that most YouTube videos contained few recommended prevention messages from governments or public health organisations. "[V]ideos, especially from public health authorities, should include accurate and reliable medical and scientific information and use relevant hashtags to reach a large audience, generate a high number of views, and increase responses."

One takeaway message from the review: "For COVID-19, social media can have a crucial role in disseminating health information and tackling infodemics and misinformation."