A rigorous study finds that surgical masks are highly protective, but cloth masks fall short. A rigorous study finds that surgical masks are highly protective, but cloth masks fall short.
Stefano Bertozzi, UC Berkley School of Public Health Dean Emeritus and Editor-in-Chief of Rapid Reviews: COVID-19, pointed out something odd about a news article on the (apparently high!) effectiveness of masks in preventing COVID spread making the rounds on Nature’s news section. Normally, this wouldn’t be notable. Nature publishes tons of news summaries about notable studies — from both their own portfolio of journals and others’s — all the time.
What’s interesting about this one is that the source material it’s writing up isn’t a published study, but an unreviewed preprint. It’s also not the only recent Nature news article citing preprints. The piece doesn’t hide the fact that it’s reporting on a preprint, but it doesn’t exactly call it out, either. Here’s the lede of the story (the fact that it’s a preprint is mentioned in paragraph 3):
Face masks protect against COVID-19. That’s the conclusion of a gold-standard clinical trial in Bangladesh, which backs up the findings of hundreds of previous observational and laboratory studies.
This may seem like a trivial observation, or even a bit of a gotcha, but I think it’s a pretty significant development. Nature’s flagship journal, after all, is one of the most-cited and prestigious scientific publications in the world by most measures, and many of its downstream titles feature highly in rankings as well. The brand’s value proposition rests almost entirely on the exclusivity produced by its peer-review selection process (despite scant evidence that it’s effective), which it makes abundantly clear on its own about page:
Over 50 thousand manuscripts are read and assessed for quality and integrity each year by Nature and the Nature research journals, but less than 10% of these papers will be selected by our highly qualified professional editors for publication. The exceptional research in these journals has a profound effect on the disciplines to which it relates and the wider world.
Of course, we’re not talking here about a journal article. But even the published editorial criteria for Nature’s News & Views section makes it clear what the venerable institution considers valuable:
These articles inform nonspecialist readers about new scientific advances, as reported in recently published papers (in Nature and elsewhere). This is a commission-only section.
So, what’s going on here? Has Nature decided that preprints are published works? Have their editorial standards changed a bit due to the pandemic? Are they getting into the preprint curation business?
If it were a one-off, you might be able to argue that in this case, because it’s citing a much-anticipated and apparently methodologically sound RCT from a respected group of researchers about COVID, it’s perhaps an omission that proves the rule (though that argument would beg some of its own questions about the value of review). But as the examples I linked to above show, it’s not a one-off, and it’s not limited to the pandemic.
Whatever’s going on, the fact that the company’s own news team is willing to cite unreviewed work signals that that work has value, and calls into question what value the journal itself provides. To be clear: I think there is value to Nature’s News team engaging in this type of curation, and I’m not arguing that italicized-Nature provides no value. I also don’t think that Nature is trying to pull a fast one on us or anything like that.
But I do think we should pay more attention to developments like these, and ask the questions that they beg.