Tracing open data in emergencies: The case of the COVID-19 pandemic.

Abstract

BackgroundThe coronavirus disease 2019 (COVID‐19) pandemic constitutes an ongoing, burning Public Health Emergency of International Concern (PHEIC). In 2015, the World Health Organization adopted an open data policy recommendation in such situations.ObjectivesThe present cross‐sectional meta‐research study aimed to assess the availability of open data and metrics of articles pertaining to the COVID‐19 outbreak in five high‐impact journals.MethodsAll articles regarding the severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2), published in five high‐impact journals (Ann Intern Med, BMJ, JAMA, NEJM and Lancet) until March 14, 2020 were retrieved. Metadata (namely the type of article, number of authors, number of patients, citations, errata, news and social media mentions) were extracted for each article in each journal in a systematic way. Google Scholar and Scopus were used for citations and author metrics respectively, and Altmetrics and PlumX were used for news and social media mentions retrieval. The degree of adherence to the PHEIC open data call was also evaluated.ResultsA total of 140 articles were published until March 14, 2020, mostly opinion papers. Sixteen errata followed these publications. The number of authors in each article ranged from 1 to 63, whereas the number of patients with a laboratory‐confirmed SARS‐CoV‐2 infection reached 2645. Extensive hyperauthorship was evident among case studies. The impact of these publications reached a total of 4210 cumulative crude citations and 342 790 news and social media mentions. Only one publication (0.7%) provided complete open data, while 32 (22.9%) included patient data.ConclusionsEven though a large number of manuscripts was produced since the pandemic, availability of open data remains restricted.