Read on Twitter
Thanks to @LeonardLeibovi1 for sharing his concerns and to @CMIJournal for publishing this editorial. I do sympathize with these"difficult editorial decisions" and appreciate to be given this information. https://twitter.com/CMIJournal/status/1322969763182743554
Though I acknowledge the meta-analysis from Fiolet et al. may raise some commentaries and criticisms, it was difficult for me to understand why CMI published a letter (Million et al.) claiming that Fiolet's conclusions were doubtful as they did not confirm their local findings
(the study by Lagier et al, already well known for its methodological flaws and for the inability to draw any conclusions from them). Suggesting that this meta-analysis was questionable as the authors lacked experience in the treatment of infectious diseases
(but still conducting a so-called alternative "meta-analysis" violating almost all best-practice recommendations for such studies with no methodologist involved).
I will not list all the reasons why this letter was an insult to clinical research - a field that this team apparently discovered a few months ago - as Fiolet et al. already did so in their subsequent response.
I wondered why @CMIJournal accepted to publish such a nonsensical letter and the only reason I could find was that it would make Fiolet et al. some room to explain why these nonsensical criticisms had to be disregarded, as well as most clinical research published by this team.
Which they did, and their response was probably the most didactic I could hope.
However, I was really upset to read, shortly afterwards, that CMI decided to publish this commentary from Prof Raoult about a "struggle between practicing physicians and methodologists."
However, I was really upset to read, shortly afterwards, that CMI decided to publish this commentary from Prof Raoult about a "struggle between practicing physicians and methodologists."
If this comment was a call for physicians and methodologists to unite and work together as clinical research needs both their skills, I would have happily supported him since, as a (nonclinician) physician and methodologist,
it is exactly the way I consider clinical research should be carried out (as certainly do all my methodologist fellows, poke #epitwitter). On the other hand, Prof Raoult's team does not seem to find necessary to request help from methodologists for their work.
This lead by the end of March to one of the most famous (not for good reasons) COVID-19 papers (Gautret et al.), that was extensively commented for its impressive aggregation of methodological flaws. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7357524/
Doubts were then still allowed regarding the benefits of hydroxychloroquine to treat COVID-19 patients despite (not "thanks to") this paper, and the brilliant communication plan from Prof Raoult's team helped perpetuate this doubt as scientific evidence accumulated to show that
the benefit, if any, was not clinically relevant. In the meantime, clinical trials, likely to evaluate the potential benefits of several drugs - in France, at least - struggled to include patients as a famous Professor on TV let them think
that they would miss the opportunity to be given the magical pills. The next joke from this team was the study from Lagier et al., still suggesting that major avoidable biases such as immortal time bias or baseline risk imbalance were not noteworthy as long as they could use
a fancy name such as "propensity score" (though the "Methods" section raises serious concerns about their understanding of this approach). Obviously, these "practicing physicians" still didn't find necessary to request the help of a skilled methodologist.
Fortunately, a consensus seemed to emerge in the scientific community about this team's untrustworthiness and their message's irrelevancy regarding hydroxychloroquine. The article by Fox et al. in @AmJEpi regarding a controversial opinion piece
by a member of the journal's Editorial Board sounded to be the final act of this insane drama: the famous professor was still talking nonsense on TV (unfortunately, never agreeing to debate with a real scientist likely to prove him wrong) but at least,
it was clear among scientists that the controversy was over. Or, should I say, that there had never been any scientific controversy regarding this drug. At least a doubt for a few weeks or months, as for probably dozens of other potential treatments.
The context was as such when CMI published the recent piece by Prof Raoult. The idea of "practicing physicians vs. methodologists" or "Western world against the rest of the world" struggles gives the idea that methodology is optional to evaluate public health interventions.
That there is a controversy between several schools of thought in clinical research. However, if struggle there is regarding this topic, it is between (1) one team of microbiologists (few of them are clinicians but does it really matter?) with no skill in methodology
and (2) the rest of the world: physicians, methodolologists, healthcare and research professionals, trying to follow and to improve research standards and constantly questioning their own practice.
Some among the latter, for instance, were those who, soon after the meta-analysis by Fiolet et al., published several other meta-analyses, all with similar results.
As a conclusion though I am very grateful to Dr Leibovici for sharing with us the process that lead to this controversial publication, I have to admit that I still respectfully disagree with his latest "difficult editorial decision."
Giving to Prof Raoult the opportunity to write a commentary in which he advocates that sound methodological approaches are optional in clinical research, even with an insightful contrasting piece by Prof Paul, gives the false idea that there is still a widespread controversy
in the scientific community, which appears to me as counterproductive in a perspective of scientific knowledge dissemination.
And if, in a future commentary, CMI's Editorial Board is looking for a "dedicated and brilliant clinician," I would be more than happy to introduce them to many wonderful ones I know, all of whom knowing that there is no relevant clinical research without sound methodology.
