Look, errors happen. We're all rushing to publish, and given the emergency that's necessary. But there is a robust preprint *and* post peer-review criticism process that's been great. Reporters have to embed within the scientific community to follow this and report accordingly.
-
-
Show this thread
-
That now-retracted study was underpowered to the point of no use *and* it had math and interpretability problems. The opposite of don't publish quackery isn't parrot peer-reviewed studies without reflection. There's a process. It's messy. But it bends towards empirical validity.
Show this thread -
That single paper which was already worth very little before the retraction was cited very very widely as evidence that cloth masks don't work. Hopefully, the retraction will be just as big news. *waits*
Show this thread -
(This one my followers may not like to hear). Similarly, evidence is not *yet* conclusive on hydroxychloroquine harms or lack of benefit. One Lancet study everyone cites has real issues and is observational only. One unblinded RCT found no harms. Let randomized studies conclude.
Show this thread -
Here's the letter my co-authors had sent on that now retracted "cloth masks don't work" study (42 citations on Google Scholar already and countless news articles). Note: it was always so underpowered that it was fairly useless even without these errors.https://twitter.com/jeremyphoward/status/1267832478502391813 …
Show this thread -
Again, folks, errors happen. But there really is a robust meta-review process among the scientific community right now—from preprint to peer-review to post peer-review. It's a real community working hard, debating, checking. Reporting should include this.https://twitter.com/CT_Bergstrom/status/1267659926811230209 …
Show this thread -
Here's our own preprint on masks for reducing *transmission* of COVID, now in version two because we got a lot of useful feedback to version one and systematically went through and updated our paper with clarifications, additions and fixed a few errors! https://www.preprints.org/manuscript/202004.0203/v2 …pic.twitter.com/m9HyorltZp
Show this thread -
Media: please plug into the scientific community and its current pre- *and* post- peer review process before running with the headlines. Remember my warning about the Lancet study on HCQ harms? Even more questions about it now. Wait. For. RCTs. To. End.https://twitter.com/zeynep/status/1264671101587804172 …
Show this thread -
Article with many questions about that observational Lancet study alleging harms from hydroxychloroquine. This doesn't mean there are no harms, but a randomized trial was stopped and checked and none was found. It also doesn't mean it's beneficial. Wait!https://www.sciencemag.org/news/2020/06/mysterious-company-s-coronavirus-papers-top-medical-journals-may-be-unraveling …
Show this thread -
Despite many alarmist headlines and the polarized social media chatter about hydroxychloroquine, many scientists from top institutions questioned that Lancet study because.. well, there were questions. There's a process. It is messy but it does bend toward empirical validity.
Show this thread -
Yes, it's hard to stop him from hijacking people's minds. But important to try. Recent reporting and social media reaction on hydroxychloroquine has been less than stellar, and has actually harmed our ability to understand potential harms or benefits.https://twitter.com/trengarajan/status/1268162690570051584 …
Show this thread -
Here we go. That Lancet study on hydroxychloroquine may well be retracted. Folks, many scientists *warned* about this study as soon as it was published. Peer review catches most crap, and we also catch what gets through. Wait. Listen. Don't sensationalize. https://www.theguardian.com/world/2020/jun/03/covid-19-surgisphere-who-world-health-organization-hydroxychloroquine …
Show this thread -
Besides the fact that the Lancet HCQ study was observational with glaring problems of controlling for severity, a randomized trial on HCQ was unblinded as a result and found no harms. That's a very loud signal. I tried to warn and got a lot of push-back.https://twitter.com/zeynep/status/1264671101587804172 …
Show this thread -
Just like the non evidence-based beach/park shaming, the sensationalized & rushed reporting of and social media reaction to potential HCQ harms will do enormous harm to what we do need: more science, but one that is seen through a messy but now strikingly robust and fast process.
Show this thread -
Find the people who didn't; find the people who try to fight their own instincts and ideological priors; find the people who admit errors. The last one might be the most important. Errors are unavoidable. It's the process that works, not any single stance.https://twitter.com/LJAkaar67/status/1268185055546900480 …
Show this thread -
Aaaand, that Lancet HCQ harms study is retracted. Again, please do note that the study had quickly raised concerns among the scientific community. It contradicted existing knowledge. There was an open letter by prominent scientists from top universities.https://twitter.com/TheLancet/status/1268613313702891523 …
Show this thread
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
That severely under-powered study (n=4) that allegedly showed cloth masks were ineffective in blocking SARS-CoV-2? My co-authors argued that it had errors in it (and took a lot of grief for that). It has been *retracted* because of errors.