JMIR Publications

ABSTRACT

Background:

The increasing frequency and diversity of human disease outbreaks due to a plethora of ecological, environmental, and socio-economic factors thereby stressing healthcare systems worldwide. The most recent of these outbreaks is the novel coronavirus (COVID-19). For public health systems, understanding the temporal and spatial distribution of COVID-19 is a top priority. The wild spread of COVID-19 was also mirrored with intense attention on social media platforms, including Twitter. Information shared by individuals on social media concerning the virus can help decision-makers and healthcare professionals identify main issues involving COVID-19 and interventions to address them.

Objective:

This study aimed to analyze posts on Twitter to identify the main thoughts, attitudes, feelings, and topics that are discussed concerning the COVID-19.

Methods:

Leveraging a set of tools (Twitter’s search Application Programming Interface, Tweepy Python library, and Postgress database) and using a set of pre-defined search terms (coronavirus, 2019-nCov, and COVID-19), we extracted the text and metadata (number of likes and retweets, and user profile information including the number of followers) for public tweets in the English language for the period from 2 February 2020 and 3 March 2020. Tweets were analyzed using word frequencies of single words and double-word combinations. We leveraged Latent Dirichlet Allocation for topic modelling to identify the topics in the tweets. Additionally, sentiment analysis, extracting the mean number of retweets, likes, and followers for each topic and calculated interaction rate for each topic.

Results:

Out of approximately 2.8 million tweets for the study period, 167,073 unique tweets from 160,829 unique users met the inclusion criteria and were analyzed. Twelve topics grouped into four main themes: (i) Origin of the virus, (ii) its sources, (iii) its impact on people, countries, and the economy, as well as (iv) ways of mitigating the risk of infection. The mean of sentiment was positive in all topics except two topics: deaths caused by COVID-19 and increased racism. The mean followers of the account posting tweet topics ranged from 2,722 (increased racisms) to 13,413 (economic losses). The highest mean of likes for the tweets was 15.4 (economic losses) while the lowest was 3.94 (travel bans and warnings).

Conclusions:

Public health crisis response activities on the ground and online are becoming increasingly ‘simultaneous and intertwined’ and social media provides a lucrative opportunity to spread public health knowledge directly to the public. Health systems should work on building national and international detection and surveillance systems of “digital” diseases listening and monitoring social media. However, there is also a need for a more proactive and agile public health presence on social media to combat the spread of “fake news”.