The website is used by a general audience, including patients, researchers, and trial administrators, and searches on a query like “Covid-19” produce results that may not relate directly to the disease — for instance, a study about an unrelated condition that mentions Covid-19 in passing in the description of the trial or in an auxiliary field in the database.
In order to mitigate the risk of including such results, a more detailed analysis was conducted on the complete dataset of nearly 350,000 current and historical records. That analysis helped weed out studies tangentially related to the disease, such as those only dealing with the consequences of the outbreak.
The investigation also found that many studies were simply labeled “coronavirus,” a family of RNA viruses to which SARS-CoV-2, the virus that causes Covid-19, belongs. The coronavirus family also includes both SARS-CoV-1, commonly referred to as SARS after the 2002 outbreak in Guangdong Province, China, and the Middle East respiratory syndrome (MERS), a coronavirus that still is the subject of ongoing vaccine research. Coronavirus studies unrelated to SARS-CoV-2 and Covid-19 were also excluded from the analysis.
Observational studies were also removed, resulting in a final collection of interventional clinical trials.
Patient enrollment is categorized by confirmed enrollment for completed and active trials, and projected enrollment for scheduled trials. Projected enrollment accounts for 96% of all enrollment, meaning that most Covid-19 trials are either still enrolling or have yet to begin enrolling patients.
The analysis reflects data as of June 24. In the days before this article was published, approximately 20 Covid-19 studies a day were added to the database.
Joanna Lin Su contributed reporting.