In introduction, the authors claim that chloroquine is an effective treatment against Covid-19. To support their assumption, they rely on reference 1 (https://doi.org/10.1038/s41422-020-0282-0 ) for in-vitro study, reference 2 (https://doi.org/10.5582/bst.2020.01047 ) and 3 (https://doi.org/10.1016/j.ijantimicag.2020.105949 ).
It should be noted that the last two of these references have important unanswered concerns on PubPeer:
Reference 2 has no record of studies, and also seems to mention cancelled and ongoing trials.
Reference 3 (https://pubpeer.com/publications/E09AC9D25125B0AB077971FBA6DD7B ) deals with use of hydroxychloroquine, not chloroquine, and has been published in one single day (Between submission and acceptation) and has also been questioned regarding inclusion/exclusion factors, ethics concerns and data consistency.
Thus, we should be cautious with the authors’ assumption as showed a few months before the current publication in this article https://doi.org/10.1016/j.antiviral.2020.104762
Further, the authors state that “many countries have adopted this treatment to treat patients with SARS-CoV-2”. This assumption is not accurate, what does “many” mean in this case?
The authors try to be more precise by stating that “60% of humans worldwide live in a country where chloroquine and its derivatives are recommended to treat patients with SARS-CoV-2”, without any evidence supporting it. This assumption requires citations.
They later write that they “determined which countries recommended hydroxychloroquine using the recommendations issued by the authorities of these countries, using data collected by c19study”.
This website is part of the websites having appeared during the pandemic, with proven data unreliability and can therefore not be used as a scientific basis! We can read here about this kind of website:
https://reutersinstitute.politics.ox.ac.uk/types-sources-and-claims-covid-19-misinformation
Or here, for this particular one (In French):
Peut-on faire confiance au site https://t.co/5qtsbfaosG sur les traitements pour la COVID ? La réponse est non : C… https://t.co/uD49eNSu7S
1/21 Problème 1 : Ce site compare des choses qui n'ont rien à voir D'abord il donne l'impression (avec les pource… https://t.co/A4Xs1i29wi
2/21 Ce site mélange tout et n'importe quoi : - Des études in vitro - Des news (dont une de 1896) - Des méta-analy… https://t.co/4SyUDFCECW
3/21 Problème 2 : Des "études" qui n'ont rien à faire là (non cliniques) On y trouve par exemple une "étude" de J… https://t.co/ygPeXnyqZC
4/21 On y trouve aussi un papier de Maxime Izoulet, qui n'a rien à voir avec un essai clinique Il y est fait deux… https://t.co/WKnpR6nFql
5/21 Problème 3 : Des études sans données, comptées comme positives Par exemple, ces 3 études à la suite ne fourn… https://t.co/Nx2TuOr4Sh
6/21 Problème 4 : Beaucoup d'études sont simplement des études de cas d'un traitement HCQ, avec 0 comparaison On… https://t.co/fZrsuevhZH
7/21 Problème 5 : Des études aux conclusions modifiées, sur la base de tweets On y trouve par exemple l'étude de… https://t.co/nif9YVLuP6
8/21 Et ce, sur la base d'un simple tweet, d'un cherry picking, d'une modification du seuil statistique et d'une i… https://t.co/55FSQkrMkT
9/21 Même si on fait du cherry picking comme le tweet le fait (jours 1, 1-2, 1-3, 1-4), on obtient toujours des p-… https://t.co/GqyzpQHqiW
10/21 En faisant cela, on introduit un biais dans la randomisation : Les patients traités plus tard sont différent… https://t.co/UOBrQlkQVP
11/21 Et malgré tout ça, on obtient quand même une p-value > 0.05, ce qui, selon toutes les conventions en médecin… https://t.co/eq4SYc7rHd
12/21 Dans les études PEP (100%) on a : - Une étude "advisory" non comptée - Une étude sans comparaison (problème… https://t.co/s9t2x63bY1
13/21 Problème 6 : Des études manquantes Exemple : - L'étude de Barbosa [1] - Les études de Bessière [2] et Mercu… https://t.co/r1x1dHaCpI
14/21 Problème 7 : Le traitement que l'on regarde ("positive") change à chaque étude - Dans l'étude de l'AP-HP, l… https://t.co/XrwoAIATuu
15/21 Ce problème est très frappant pour une étude en particulier (voir image) Cette étude compare HCQ+AZ+Zinc vs… https://t.co/oCnSVFpu2W
16/21 Problème 8 : Les critères d'évaluation changent aussi selon l'étude - L'étude de l'AP-HP montre par exemple… https://t.co/QEhdejheYV
17/21 Comme critère évalué on a, selon l'étude : - Portage viral à un certain moment - Durée du portage viral - H… https://t.co/BMSByXHxC8
18/21 Problème 9 : La classification "Early vs. Late" est inconsistante, et sert à faire passer le traitement earl… https://t.co/GP0df1Q4gH
19/21 À côté de ça, l'étude randomisée espagnole, avec "early" dans le nom, est considérée comme "late" Justifica… https://t.co/BX0q0RHiRN
20/21 Conclusion - Ce site mélange tout et n'importe quoi : Des news, des sondages, des études in vitro - Certain… https://t.co/Odx3yI9GcV
21/21 - Certaines études "négatives" pour l'HCQ manquent - Le critère d'évaluation change selon l'étude choisie -… https://t.co/jXU0BDbluQ
Un autre thread qui parle de problèmes de ce site https://t.co/3eLdKFM94z
Poke @a_1_0_2 @lonnibesancon @achilledamu @Damkyan_Omega @nathanpsmad @FZores
Plus de décryptage sur les manipulations pendant la pandémie ici https://t.co/bwtAH8HTwS
Le compte @CovidAnalysis a récidivé avec le site https://t.co/Sn8uH3DXmL Explications de pourquoi ce site n'apport… https://t.co/dw3HftwBqL
It is therefore difficult to share the comments made by the authors on the maps they have drawn up from data whose reliability has not been established and have important concerns.
Later, they assert that physicians have been “turning en masse to hydroxychloroquine as a treatment for SARS-CoV-2”, citing a survey to support their statement. It should be noted that no evidence showing that this survey is representative is given. Furthermore, the use of a survey to determine the efficacy of a treatment is absolutely irrelevant.
As a conclusion, the authors rely on a correlation between the maps they draw from questionable data and the mortality rate, to correlate these two factors.
This is clearly abusive.
The demonstration of their correlation is weak, and even if it exists, nothing in the words of the authors proves that there is a causality relationship between these two factors!
https://en.wikipedia.org/wiki/Correlation_does_not_imply_causation
There are many factors differentiating the Western World and other countries. One could cite number of tests, age distribution, diet, mobility, statistics quality, ….
To overlook all of these differences is a major mistake that makes it even more difficult to support the opinion expressed here by the authors.
As a conclusion, relying on questionable bibliography, with questionable data leads to a highly questionable conclusion.
Attach files by dragging & dropping, selecting them, or pasting from the clipboard. Uploading your files… We don’t support that file type. with a PNG, GIF, or JPG. Yowza, that’s a big file. with a file smaller than 1MB. This file is empty. with a file that’s not empty. Something went really wrong, and we can’t process that file.
Comment must be at least 15 characters.