TThe wild, wild west of Covid-19 preprints is about to get a new sheriff. On Monday, MIT Press announces the launch of an open access journal that will publish preprint reviews related to Covid-19, in an effort to quickly and authoritatively report misinformation, as well as highlight important and credible research.
“Preprints have been a great help to scientific communication, but they carry some dangers, as we’ve seen with some that have been based on flawed methods,” said Nick Lindsay, magazine editor for MIT Press, who will publish Rapid Comments: Covid- 19. “We want to discredit research that is poor and elevate research that is good.”
The Covid-19 pandemic has produced a preprint fire hose (documents published on servers like bioRxiv and medRxiv without peer review), many of questionable validity. The poster for that is a bioRxiv preprint that suggested that the new coronavirus had somehow been engineered from HIV; It was quickly removed. But many other preprints, while not clearly wrong, used weak methodology, such as a small number of patients or inadequate controls, as in an experiment that concluded that a commercially available immunoglobulin could protect against disease.
ad
“There have definitely been some previous bad impressions,” said Richard Sever, co-founder of bioRxiv, in 2013, and medRxiv, whose methodical deployment a year ago was sped up by the pandemic and has now released some 5,000 documents on the matter. . “A lot of people are talking about doing something like [the MIT Press effort]. Your challenge is to get people to do the reviews quickly. It’s a great idea, but it could be easier said than done. “
Rapid Reviews: Covid-19 Editor-in-Chief Stefano Bertozzi of the University of California, Berkeley believes this project has a secret sauce that similar efforts do not. It will use an artificial intelligence system developed at the Lawrence Berkeley National Laboratory to classify new preprints by discipline (such as epidemiology or clinical care) and degree of novelty.
ad
“There is a large volume of material every day, our goal is to do quick reviews of preprints that are most interesting,” said Bertozzi. “Interesting” means studies that could influence public health officials, doctors and the public, he said, “as well as those that need to be validated or discredited, especially if they are receiving a lot of attention in the media or social media.”
That attention can come almost instantly, posing a challenge for a magazine with “fast” in its name. The AI screen should speed up the process at the front. Humans will step in as well, with around 100 graduate student volunteers from around the world scanning preprints for those who need revision most.
Once a preprint has been marked, Bertozzi and its editors will ask up to three experts to review their strengths and limitations, with or without their name attached to the review.
Both medRxiv and bioRxiv would indicate “absolutely” whether a preprint received a quick review, Sever said, just as when a preprint is published in a journal. “One of our missions is to alert readers to relevant conversations,” he said.
The first revisions should be ready by mid-July, with the goal of publishing them within seven to 10 days after a preprint appears.
The closest similar effort took place in April at Johns Hopkins University, where epidemiologist Emily Gurley and pathologist Kate Grabowski launched the 2019 New Coronavirus Research Digest. Its 50 volunteers, mostly Hopkins, include experts. in mathematical models, diagnostics, vaccines, and related fields. Using keyword searches, they select new studies, both preprinted and published in journals, that they believe contain important information for clinicians and policy makers. Postdoctoral fellows and graduate students summarize the findings of the work and point out its strengths and limitations. Two members of the Hopkins faculty review and edit the reviews, which recently approved 220.
“Our goal is to make sure that people, especially frontline doctors, can find the information they need,” said Gurley. “No one has time to review all the documents that are coming out” on Covid-19.
Quick Reviews: Covid-19 plans to tap a group of 1,600 potential reviewers from hundreds of institutions, and will review articles on the economics and anthropology of the pandemic and biomedicine. If it lives up to its founders’ hopes, it would be the largest formal effort to mount herd on Covid-19 preprints. He also plans to publish original research from areas of the world that have been underrepresented in western magazines.
Praising the Hopkins project, Lindsay said: “The reality is that there is so much Covid-19 research out there that there will be plenty for us all to examine.”
