Our Approach

The COVID-19 pandemic has generated unprecedented global research innovation and collaboration, but the volume of research is challenging for scientists and decision makers to efficiently digest. We wanted to make the job easy for you. A number of dedicated COVID-19 websites serve as “evidence hubs,” to provide information on ongoing research studies and study results; others synthesize the emerging evidence. We created this web directory to consolidate all of them in one place to facilitate access to these online evidence hubs and characterize them in near-real time. Our goal is to efficiently direct clinicians, researchers, public health agencies, journalists, and policy makers to the web sources most appropriate to their needs. We also hope to make the research enterprise more efficient by avoiding duplication of efforts.

Methods

In order to be included in the directory, a website must be specifically dedicated to COVID-19 research and have publicly available at least one of the following:

  • A database of studies (excludes databases primarily for patient recruitment) 
  • A database of study results
  • An appraisal of emerging evidence

Websites/databases that are primary data sources were not included e.g. clinicaltrials.gov for studies and PubMed for study results. Websites are assigned to one or more of the three domains, according to the categories of human studies that they contain. Includes sites that require free registration. 

Websites are identified through referrals from clinical trial experts and through Google searches using the following search terms: “COVID-19 trials,” “COVID-19 trials tracker,” “COVID-19 Studies,” “COVID-19 Research,” and “COVID-19 Evidence.” We review the first 25 results for each search.

For each website, we identify the sponsoring organization, whether the site offers a study database, a study results database, results appraisal or some combination of those. We also determine the types of study designs (e.g., interventional studies, observational studies) included and topic areas (e.g., drugs, vaccines) included. A website is considered to have “more than interventional studies” if they include any of the following: cohort studies, case series, or modeling studies. For websites with a study results database we also note if they include systematic reviews in their database. These website details are determined by reviewing the methods section of each website or by identification of studies with the features described in the database. Websites performing results appraisal are assessed for inclusion of meta-analyses or systematic reviews, narrative commentary, and results quality assessment. We exclude those websites whose appraisals are written primarily in a journalistic style.

For websites with a study database, we note their underlying data sources (Clinicaltrials.gov, World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP), and the Chinese Clinical Trials Register (ChiCTR). Individual study databases that feed into WHO ICTRP are not listed separately as data sources. We similarly identify the underlying databases for websites with study results (PubMed/MEDLINE, MedRxiv/BioRxiv, Clinicaltrials.gov, WHO ICTRP, publisher databases and press releases). The frequency of database and appraisal updates are also determined through review of the website methods or by looking at the posting frequency of studies, results, and appraisals on the website. We contact the websites if we are unclear about any of the above.

All website directory content is updated every 2-3 weeks, by including new websites that have come to our attention and by repeating the search strategy above. We also revisit each existing website to determine if its characteristics have changed. At that time, we identify the number of studies and study results populating each website’s database, and the various forms and number of results appraisal performed. If a website’s study results database includes systematic reviews, we consider each review to represent one database result. In cases where a website does not specify the number of studies or results in their database or combines studies and study results in one database without a filter option, a manual count or search and filter of studies, results, and appraisal postings is performed. For websites who combine study results with non-research articles (e.g., guidelines, commentary) in their database without a filter option, the total number of combined database articles listed on the website is noted as such. All listings are verified by two staff members prior to posting.