An analysis of COVID-19 article dissemination by Twitter compared to citation rates

  1. Nicholas Fabiano
  2. Zachary Hallgrimson
  3. Sakib Kazi
  4. Jean-Paul Salameh
  5. Stanley Wong
  6. Abrar Kazi
  7. Rudy R Unni
  8. Ross Prager
  9. Matthew DF McInnes

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Background

The COVID-19 pandemic has resulted in over 1,000,000 cases across 181 countries worldwide. The global impact of COVID-19 has resulted in a surge of related research. Researchers have turned to social media platforms, namely Twitter, to disseminate their articles. The online database Altmetric is a tool which tracks the social media metrics of articles and is complementary to traditional, citation-based metrics. Citation-based metrics may fail to portray dissemination accurately, due to the lengthy publication process. Altmetrics are not subject to this time-lag, suggesting that they may be an effective marker of research dissemination during the COVID-19 pandemic.

Objectives

To assess the dissemination of COVID-19 articles as measured by Twitter dissemination, compared to traditional citation-based metrics, and determine article characteristics associated with tweet rates.

Methods

COVID-19 articles obtained from LitCovid published between January 1st to March 18th, 2020 were screened for inclusion. The following article characteristics were extracted independently, in single: Topic (General Info, Mechanism, Diagnosis, Transmission, Treatment, Prevention, Case Report, and Epidemic Forecasting), open access status (open access and subscription-based), continent of corresponding author (Asia, Australia, Africa, North America, South America, and Europe), tweets, and citations. A sign test was used to compare the tweet rate and citation rate per day. A negative binomial regression analysis was conducted to evaluate the association between tweet rate and article characteristics of interest.

Results

1328 articles were included in the analysis. Tweet rates were found to be significantly higher than citation rates for COVID-19 articles, with a median tweet rate of 1.09 (IQR 6.83) tweets per day and median citation rate of 0.00 (IQR 0.00) citations per day, resulting in a median of differences of 1.09 (95% CI 0.86-1.33, P < .001). 2018 journal impact factors were positively correlated with tweet rate ( P < .001). The topics Diagnosis ( P = .01), Transmission ( P < .001), Treatment ( P = .01), and Epidemic Forecasting ( P < .001) were positively correlated with tweet rate, relative to Case Report. The following continents of the corresponding author were negatively correlated with tweet rate, Africa ( P < .001), Australia ( P = .03), and South America ( P < .001), relative to Asia. Open access journals were negatively correlated with tweet rate, relative to subscription-based journals ( P < .001).

Conclusions

COVID-19 articles had significantly higher tweets rates compared to citation rates. This study further identified article characteristics that are correlated with the dissemination of articles on Twitter, such as 2018 journal impact factor, continent of the corresponding author, topic, and open access status. This highlights the importance of altmetrics in periods of rapidly expanding research, such as the COVID-19 pandemic to localize highly disseminated articles.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2020.06.22.20137505: (What is this?)

    Please note, not all rigor criteria are appropriate for all manuscripts.

    Table 1: Rigor

    Institutional Review Board Statementnot detected.
    Randomizationnot detected.
    Blindingnot detected.
    Power Analysisnot detected.
    Sex as a biological variablenot detected.

    Table 2: Resources

    No key resources detected.

    Results from OddPub: Thank you for sharing your data.

    Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:
    As altmetrics do not suffer from this limitation, they are better equipped to quantify dissemination of articles and provide an objective metric for healthcare workers or researchers to find impactful research, as evidenced by our finding that tweet rates were significantly higher than citation. A journal’s 2018 impact factor was found to be positively associated with tweet rate. This is fitting as journals with higher impact factors are often exposed to larger audiences, increasing the likelihood that articles published in these journals would be tweeted about. The continent of the corresponding author was found to impact the number of tweets with South America having the largest negative association, followed by Africa, then Australia, relative to articles published from Asia. An analysis of worldwide Twitter usage using tweet geolocation, found that Twitter has a strong prevalence in North America and Europe compared to other continents with African and South American countries having the lowest involvement [23]. This could explain the lower number of tweets associated with Africa and South America. There was also a lower research output, as reflected by the relatively low number of articles from Africa, South America and Australia which could account for the lower number of tweets received. Similarly, Asia had the highest number of articles, followed by Europe and North America which could account for their higher number of tweets. The number of tweets also followed the p...

    Results from TrialIdentifier: No clinical trial numbers were referenced.

    Results from Barzooka: We did not find any issues relating to the usage of bar graphs.

    Results from JetFighter: We did not find any issues relating to colormaps.

    Results from rtransparent:
    • Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
    • Thank you for including a funding statement. Authors are encouraged to include this statement when submitting to a journal.
    • No protocol registration statement was detected.

    About SciScore

    SciScore is an automated tool that is designed to assist expert reviewers by finding and presenting formulaic information scattered throughout a paper in a standard, easy to digest format. SciScore checks for the presence and correctness of RRIDs (research resource identifiers), and for rigor criteria such as sex and investigator blinding. For details on the theoretical underpinning of rigor criteria and the tools shown here, including references cited, please follow this link.

  2. Version published to 10.1101/2020.06.22.20137505v1 on medRxiv