A Los Angeles Times story reporting the appearance of a “dominant” new strain of COVID-19 has prompted criticism for relying on a document that had not been vetted through a traditional peer-review process.
The May 5 2020 story covers the findings of a group of researchers from Los Alamos National Laboratory led by Dr. Bette Korber, saying that they had discovered the emergence of the new, more contagious, strain of the disease; the URL for the story describes it as a “mutant coronavirus.” The findings of Korber and her team were originally presented in the journal BioRxiv (pronounced “bio-archive.”)
According to the newspaper’s story:
The new strain appeared in February in Europe, migrated quickly to the East Coast of the United States and has been the dominant strain across the world since mid-March, the scientists wrote.
In addition to spreading faster, it may make people vulnerable to a second infection after a first bout with the disease, the report warned.
However, while the Los Angeles Times noted that researchers post on the journal to “share their work before it is peer-reviewed,” it did not mention in full a disclaimer provided by BioRxiv:
BioRxiv is receiving many new papers on coronavirus SARS-CoV-2. These are preliminary reports that have not been peer-reviewed. They should not be regarded as conclusive, guide clinical practice/health-related behavior, or be reported in news media as established information.
Both researchers and journalists who regularly cover scientific matters have pointed out the disclaimer and raised other concerns about the Times’ piece.
“This type of reporting on [the virus] makes my blood boil,” Columbia University virologist Dr. Angela Rasmussen wrote in a Twitter thread. “There is no evidence that the dominant strain is such because it is ‘more contagious.'”
Bill Hanage, an associate professor at Harvard University’s T.H. Chan School of Public Health, also threw what he called “cold water” on the story in his own thread.
“We need to distinguish between selection, in which a variant becomes more common because it leaves more descendants, and founder effects in which a variant becomes more common because it was fortunate rolling the dice,” he wrote. “By that, I mean this variant might have been lucky and got introduced to places outside Wuhan and different approaches to social distancing early on. It’s not about the virus, it’s the environment and the opportunities for transmission.”
Hanage and Rasmussen also expressed their skepticism on the paper’s findings in a separate story published by Gizmodo.
“They didn’t do a single experiment, and this is all conjecture,” Rasmussen told the tech news site. “There’s no indication that this mutation makes the virus more transmissible, and they’ve done nothing to show that this mutation is functionally significant.”
And journalist Wudan Yan, who has reported on BioRxiv and what she called the “growing audience” for studies that have not been peer-reviewed for the New York Times, called the story “frankly incredibly terrible coverage of a preprint.”
“It does not state clearly the limitations of the preprint,” she wrote.”Author notes, ‘While the Los Alamos report is highly technical and dispassionate.’ Cool, pls make sense of this for us. Including the limitations.”
In a separate interview, Yan told us:
Journalists should not cover pre-prints unless they have thoroughly vetted them. Pre-prints exist on those group servers and are published with a lot of academic jargon and that’s because they are meant for other scientists. I cannot stress this enough: pre-prints are for other researchers to verify, validate, [or] study further. They’re not for journalists to cover. Journalists don’t have the expertise, and if we wish to cover them, we need to get the expertise from outside critics.
The Los Angeles Times story did, however, include a statement from Texas Children’s Hospital Center for Vaccine Development co-director Peter Hotez saying that the Los Alamos study contained “a lot of speculation,” and it noted a lack of experimental verification for their findings.
The newspaper’s tweet promoting the story has only been shared 44 times on that platform, while a Facebook post doing the same has only amassed 54 shares since May 7 2020. But the story’s reach was amplified further after it was aggregated by both individual news outlets like The Hill and services like Nextar Media Wire, which supplies stories for broadcast television stations’ news sites.
When we contacted the Los Angeles Times for comment, a staff member defended the newspaper’s reporting and referred to the medical experts who had offered up online critiques of the piece as “Twitter researchers.”
“We made clear in our story the Los Alamos findings had not been peer reviewed, and that researchers put it out there as an early warning for other vaccine researchers,” they added. “I suspect that Bette Korber wanted to make these preliminary finding[s] public, just as Dr. Fauci went public about the Gilead drug before it had been fully studied.” (The staff member then questioned whether we had actually done our due diligence on the story.)
Both Rasmussen and Hanage told us that though they have not published as many findings as Korber, their respective scientific careers have also spanned less time than hers. Hanage, an associate professor of epidemiology at his university, told us that he admires Korber’s work on HIV and that he was “entertained” by being described as a “Twitter researcher” by the newspaper. He added:
I had a peer reviewed paper accepted yesterday by Nature Microbiology urging caution in overinterpretation of sequence data in the early stages of a pandemic when little sequence diversity has yet accumulated.
In my view the major problem with the preprint is the title. The work itself deserves honest robust peer review and will surely be published somewhere decent. There’s no sense in making a fight over it.
Preprints should be handled with caution. Approach other scientists for comment. And they are best reported by science journalists with a good set of contacts.
He also suggested that the Times could read resource material from both Harvard and Scientific American advising journalists on how to best report on pre-prints. One Harvard guide also lists journalists whose work Hanage recommends; that list does not mention any Los Angeles Times reporters.
Rasmussen called it “disingenuous” of the Times to suggest that this invalidated critiques of the newspaper’s reporting, saying:
The number of publications a scientist has is irrelevant to critiquing the findings of the study as well as the way the LA Times chose to present those findings. In my case, my career is not as long as Dr. Korber’s and thus I have published less, but that does not mean I am unqualified to criticize the findings or the reporting of those findings. My issue with the LA Times report is that they did not consult anyone qualified to put the central finding of the paper into proper context—whether or not the mutation observed actually makes the virus more transmissible.
To be clear, I feel that Dr. Korber and colleagues made an important observation, but the data they present is insufficient to show that the virus is more transmissible or contagious. They did nothing to demonstrate that the D614G mutation increases viral fitness, transmissibility, or pathogenicity. While the Times did acknowledge in the text of the article that these conclusions were from the report, they didn’t consult any outside virologists who could speak specifically to the effect of this mutation on the virus itself.
A spokesperson for Los Alamos National Laboratory declined to comment. Dr. John Inglis, the co-founder of BioRxiv, referred us to his comments in one of Harvard’s resource guides; in it he says that journalists should clearly state when the basis of their story is a pre-print but that “it would help even more to have solicited and quoted the opinion of at least one independent expert and include any caveats they may have.”
“Anyone reading a preprint is seeing how its authors describe what they did and how they interpret the results,” he is also quoted as saying. “But peer review may do many things to the authors’ account: catch errors, argue with interpretation, dismantle or scale back claims, cut bits out, request more experiments and data, and more.”
Update 8:58 a.m. May 13 2020: Updated with responses from Los Alamos National Laboratory and Dr. John Inglis, co-founder of BioRxiv.