On most days, scientific discoveries are not cover stories or even top headlines. That hasn’t been the case since COVID-19 hit our collective consciousness. We are being inundated with science news, which would be great under different circumstances. But communicating new science during a crisis appears to have its own challenges.
The scientific community has banded together incredibly well to discover everything it can about COVID-19. Researchers are working tirelessly to get information out and make it freely accessible. A recent PubMed search of scholarly work on COVID-19 research yielded 2,075 hits of peer-reviewed work.
And that’s not including the 1,037 articles on COVID-19 that have been posted to the popular “preprint” servers medRxiv & bioRxiv (pronounced “med-archive” and “bio-archive”).
While all of this new research is fantastic because it brings us closer to understanding and controlling this pandemic, and finding treatments, our communication of it has at times fed into fear-mongering and led to confusion.
Part of the issue may be the number of individual studies that have received media attention while they were still preprints. Preprints are posted before the traditional peer-review process most scientific studies go through.
For example, misleading headlines claiming “coronavirus can remain in air for 3 hours,” thereby suggesting there’s widespread airborne transmission of the virus causing COVID-19, made rounds when the study on which they were based was still a preprint.
Popular server medRxiv has a clear disclaimer on its site. In red letters it says: “Caution: Preprints are preliminary reports of work that have not been certified by peer review. They should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.”
So I wouldn’t lay fault on the preprint servers; in fact, I applaud them for helping make important information freely available much faster than the traditional peer review process. Many argue the broad distribution of preprints means they get broader, albeit less structured, peer feedback. Indeed, the paper mentioned above had drastic changes made to it within just a few days. The first version was posted to medRxiv March 10th and the second version was posted just three days later before being published the following week in NEJM.
The challenge is that the lack of peer review before posting means fewer perspectives on experimental limitations, and no external rounds of editing to clarify language. The “publish or perish” culture of academia leaves a lot of researchers accustomed to linking the broad importance of their work, without necessarily including limitations in the first draft.
One of my recent tweets appears to suggest many others agree that plain language summaries should be mandatory for COVID-19 preprints.
For example, in the quick revision process of the preprint on aerosolization of droplets, less substantiated claims got removed, such as this one: “therefore, virus transmission via respiratory secretions in the form of droplets (>5 microns) or aerosols (<5 microns) appears to be likely.”
But even the final paper published in NEJM got misinterpreted in the broader media, indicating the problem of misinformation does not get solved with peer review alone.
The time pressures of a pandemic and justified urgency to disseminate potentially helpful information means clarity is not prioritized when researchers on the front lines are writing and reviewing manuscripts.
Even at the best of times, primary scientific writing can be the source of misinformation. In a 2014 study by Sumner et al, they found press releases created in partnership with scientists were a major source of misinformation. They analyzed 462 press releases and found that 40 per cent exaggerated advice, 33 per cent made inappropriate causal claims, and 36 per cent exaggerated inferences to humans from animals, resulting in a 6.5x and 20x higher likelihood for exaggerations in the broader news, respectively.
The lack of clarity in new papers and interviews from researchers themselves is an important area for immediate improvement, and may be remedied by something as simple as mandatory plain language summaries that clearly state limitations. This has become the norm at some, but not all, publishing houses, and is still missing from the preprint servers.
But while key findings and limitations being identified in plain language at the top of every study would help curb misinformation, I think the problem extends a bit further.
There are too many media reports on single studies compared to balanced reports reviewing multiple, reproducible studies. In general, the media put much higher faith in “new data” from individual studies than the scientific community does.
Science is about reproducible, incremental tests to rule things out until you’ve narrowed things down to a truth. When we only ever report on “a new study,” we deny people from learning the true amount of consideration that goes into finding out even the simplest of answers.
For example, Signals recently reviewed new studies reporting using mesenchymal stem cells to help with COVID-19 patients. What was missing in those papers, but elegantly outlined by blogger Jovana Drinjakovic in her post, is that while these are valuable studies that may guide future larger trials that can be life-saving, they are only one step along the longer path to finding a cure for this pandemic.
When it comes to public health, contradictory messages on how the infection spreads or suitable treatment protocols can lead to mistrust in public health authorities when we most need to heed their message. Hype in the media around single new studies leads to people panic-buying important medications and protective equipment, resulting in shortages for others. So the inaccuracies don’t just misrepresent science, they could potentially be fatal.
The journey to more information may take longer than we’d like, but we cannot discount the meticulous process of science that allows us to proceed safely with treatments. And, in the meantime, it’s important we are transparent about that scientific process so people understand why it’s worth the wait.
Latest posts by Samantha Yammine (see all)
- Communicating new science in a crisis - April 2, 2020
- Warming up to better public relations for scientists - November 22, 2018
- Bioprinting tissues of the future with Dr. Stephanie Willerth - November 16, 2018
Comments