SBM Blog Response
Background.
We note that the Twitter personality here has previously taken a public position that treatment is ineffective. The result of a false negative (incorrectly finding the treatment to be ineffective) here is significantly more death. It is possible that such Twitter personalities may have an incentive to attack signs of efficacy (indeed, they do attack signs of efficacy in general, not just our study - see sciencebasedmedicine.org/hydroxychloroquine..cant-seem-to-kill-it/).
Did not contact us.
Author has never contacted us, questioning their interest in accurate reporting.
Copy-paste of Twitter feeds making false unfounded accusations.
The main premise of the blog here seems to be copy-pasting Twitter feeds with false statements. The first one falsely claims "disinformation" with no evidence (the claim appears to be based on the author feeling the web site looks professional so therefore it must be fake).
Cherry-picking studies.
The blog cherry-picks <10% of studies (which have major flaws including inapplicable study domain), in order to claim a lack of efficacy.
Cherry-picking government actions.
The blog cherry-picks government responses from one country to make claims unsupported by either data or by many other countries.
Incorrect unsupported accusation of "astroturfing".
The blog makes an extremely serious and false accusation, presenting no related information.
Falsely claims this is retrospective study.
Groups are assigned in advance, irrespective of patient medical status.
Claims revising papers is bad.
The blog claims that revising papers based on feedback is somehow bad and seemingly uses this to support their claim of "astroturfing".
Out-of-context snippets.
The blog presents out-of-context snippets to assert that the semantics of language related to the trial is not perfect. However, when reading from the title and first sentence of the abstract, it is not possible to make the misinterpretation claimed.
Incorrect claims on RCTs.
The blog claims the following RCT confirms the treatment is ineffective, for example: an Internet survey RCT with unknown survey bias, an unknown percentage of fake responses, that shows a combined result that is more likely positive than negative, and that shows a statistically significant treatment delay-response relationship.
Incorrectly claims no trial.
The blog claims there is no trial here, but there is a medication being trialled.
Incorrectly claims not prospective.
This blog claims the trial is not prospective, but the group assignments were made in advance.
Incorrectly claims not randomized.
This blog claims the study is not randomized but for a person it is random which group they get, unrelated to their medical status.
Ideological positions.
The blog takes ideological positions about treatment or interventions, like "antimasking" or "hcq believers". We believe there is no place for this in science.
Informed consent.
The blog complains about informed consent, those complaints should be directed to the decision makers implementing the trial.
Irrelevent ecological fallacy discussion.
Talks a lot about ecological fallacy (you cannot directly infer the properties of individuals from the average of a group), however the study does not do this.
Sample size claims.
Copy-pasting another random Twittter account, claims the sample size should be 36 when the actual number of deaths alone is much larger.
Incorrectly claims France is left out.
France was not left out.
Incorrectly claims cherry-picking.
Claims cherry-picking but the criteria are clear and about 200 references support the assignments.
Incorrectly claims lack of control for interventions.
Authors did not read the study which specifically analyzes interventions.
Incorrect claim based on assuming all other authors statements are correct.
Claims a problem where one of ~80 studies in the literature review has a comment mismatching the original author comment. It does mismatch the original author comment, but that doesn't make it incorrect.
Incorrect comment on mask exclusions.
Copy-pasting another random Twitter account that didn't read the paper, which misinterprets the mask/low spread exclusions (masks are used to some degree in a majority of countries).
Random claims about the authors.
Makes seemingly random implications about the authors.