4 ms·

As far back as August 2020, this paper was known to be "fully irresponsible": https://retractionwatch.com/2020/07/19/french-hydroxychloroquine-study-has-major-m

by dfxm12 2y ago

As far back as August 2020, this paper was known to be "fully irresponsible": https://retractionwatch.com/2020/07/19/french-hydroxychloroquine-study-has-major-methodological-shortcomings-and-is-fully-irresponsible-says-review-but-is-not-being-retracted/ https://retractionwatch.com/2020/07/19/french-hydroxychloroq... This retraction is better than nothing, but it's about four and a half years too late.

lelandfe 2y ago
I'm a little appalled that it went completely undiscussed why it took 4 days for the paper to be reviewed and published and 4 years for it to be retracted.
- jfengel 2y ago
  Journals do want to publish, especially when information is exciting. The usual process of science is to trust that writers have at least some integrity. Peer reviewers aren't expected to be able to reproduce the work. They're more of a sanity check. Bad papers usually aren't retracted. More often, they're forgotten. And that usually works out well enough. Retractions are needed only when a paper has enough attention that it isn't sufficiently forgotten. So it's very common for a retraction to take a long time. It takes a while even to notice that a retraction is needed, and then you make quintuple-certain that the call for retraction is correct.
ndsipa_pomu 2y ago
The retraction also won't get anywhere near the same publicity. It should not be possible for people to promote snake oil treatments without bearing any responsibility. Unfortunately, our timeline puts the responsibility onto the general public and the quality of scientific education means that a sizeable proportion simply don't have the tools to understand publications/retractions etc.
- throw0101c 2y ago
  > The retraction also won't get anywhere near the same publicity. Even if it did, it may not be helpful (at least for the general public): * https://en.wikipedia.org/wiki/Belief_perseverance https://en.wikipedia.org/wiki/Belief_perseverance * https://en.wikipedia.org/wiki/Anchoring_effect https://en.wikipedia.org/wiki/Anchoring_effect
weebull 2y ago
"Responsible" or "irresponsible" is not a valid criticism of a scientific paper. Those words have no part to play in the search for facts. The biggest way that I think the reputation of the science community is being harmed is that it's not holding true to it's own tenets. The only way to discredit a paper is to falsify it's claims. Not to call it "irresponsible" or "immoral", label it "disinformation" or "pants-on--fire". Bad science gets proved wrong, but nobody is interested in reproducing results. As a result our journals are full of accepted papers documenting unreproducible crap. They have been for a long time. When COVID hit and we needed science to be clean and authoritive, it was already in disarray. Unable to tell us what actually worked, we just got a bunch of folklore from all sides some of which got "blessed" by the high priests and some which was "sacrilegious", but very little of it was backed up with any data beyond modelling (which is just predictions). Nobody should be claiming the high ground w.r.t. COVID IMHO. The science community failed us by acting in response to politics and not science.
- tim333 2y ago
  > The only way to discredit a paper is to falsify it's claims. The issue with the paper in question is stuff like: >six patients who were treated with hydroxychloroquine had been dropped from the study—one of whom had died, and three of whom had transferred to intensive care—which potentially skewed the results in the drug’s favor. So you can point out their research is crap without actually falsifying the claim that hydroxychloroquine helps with covid. It's hard to prove a negative experimentally but easy to show stuff like the above is iffy.

by dfxm12 2y ago

lelandfe 2y ago

I'm a little appalled that it went completely undiscussed why it took 4 days for the paper to be reviewed and published and 4 years for it to be retracted.

jfengel 2y ago
Journals do want to publish, especially when information is exciting. The usual process of science is to trust that writers have at least some integrity. Peer reviewers aren't expected to be able to reproduce the work. They're more of a sanity check. Bad papers usually aren't retracted. More often, they're forgotten. And that usually works out well enough. Retractions are needed only when a paper has enough attention that it isn't sufficiently forgotten. So it's very common for a retraction to take a long time. It takes a while even to notice that a retraction is needed, and then you make quintuple-certain that the call for retraction is correct.

ndsipa_pomu 2y ago

The retraction also won't get anywhere near the same publicity. It should not be possible for people to promote snake oil treatments without bearing any responsibility. Unfortunately, our timeline puts the responsibility onto the general public and the quality of scientific education means that a sizeable proportion simply don't have the tools to understand publications/retractions etc.

throw0101c 2y ago
> The retraction also won't get anywhere near the same publicity. Even if it did, it may not be helpful (at least for the general public): * https://en.wikipedia.org/wiki/Belief_perseverance https://en.wikipedia.org/wiki/Belief_perseverance * https://en.wikipedia.org/wiki/Anchoring_effect https://en.wikipedia.org/wiki/Anchoring_effect

weebull 2y ago

"Responsible" or "irresponsible" is not a valid criticism of a scientific paper. Those words have no part to play in the search for facts. The biggest way that I think the reputation of the science community is being harmed is that it's not holding true to it's own tenets. The only way to discredit a paper is to falsify it's claims. Not to call it "irresponsible" or "immoral", label it "disinformation" or "pants-on--fire". Bad science gets proved wrong, but nobody is interested in reproducing results. As a result our journals are full of accepted papers documenting unreproducible crap. They have been for a long time. When COVID hit and we needed science to be clean and authoritive, it was already in disarray. Unable to tell us what actually worked, we just got a bunch of folklore from all sides some of which got "blessed" by the high priests and some which was "sacrilegious", but very little of it was backed up with any data beyond modelling (which is just predictions). Nobody should be claiming the high ground w.r.t. COVID IMHO. The science community failed us by acting in response to politics and not science.

tim333 2y ago
> The only way to discredit a paper is to falsify it's claims. The issue with the paper in question is stuff like: >six patients who were treated with hydroxychloroquine had been dropped from the study—one of whom had died, and three of whom had transferred to intensive care—which potentially skewed the results in the drug’s favor. So you can point out their research is crap without actually falsifying the claim that hydroxychloroquine helps with covid. It's hard to prove a negative experimentally but easy to show stuff like the above is iffy.