Just days after adding an expression of concern to a paper published last week claiming that two people died from COVID-19 vaccinations for every three cases the vaccines prevented, the journal Vaccines has retracted the paper.
[See an update on this post, with more fallout from this case.]
As we have previously noted:
The authors, including Harald Walach, who was also co-author of a just-published paper in JAMA Pediatrics questioning the safety of masks in children, had used data from the Dutch national registry of side effects. That registry carries a warning label about its use. The editors of Vaccines, which published the study last month, wrote that there were concerns over “misrepresentation of the COVID-19 vaccination efforts and misrepresentation of the data.”
At least two members of the editorial board of Vaccines have resigned over the paper’s publication.
Yesterday, we published the authors’ response to a critique from Eugène Van Puijenbroek, of Lareb (The Netherlands Pharmacovigilance Centre), where the data originated. In it, the authors wrote:
We are happy to concede that the data we used – the large Israeli field study to gauge the number needed to vaccinate and the LAREB data to estimate side-effects and harms – are far from perfect, and we said so in our paper. But we did not use them incorrectly. We used imperfect data correctly. We are not responsible for the validity and correctness of the data, but for the correctness of the analysis. We contend that our analysis was correct. We agree with LAREB that their data is not good enough. But this is not our fault, nor can one deduce incorrect use of data or incorrect analysis.
Apparently, the journal found that response wanting. Here’s the retraction notice, published today:
The journal retracts the article, The Safety of COVID-19 Vaccinations—We Should Rethink the Policy [1], cited above.
Serious concerns were brought to the attention of the publisher regarding misinterpretation of data, leading to incorrect and distorted conclusions.
The article was evaluated by the Editor-in-Chief with the support of several Editorial Board Members. They found that the article contained several errors that fundamentally affect the interpretation of the findings.
These include, but are not limited to:
The data from the Lareb report (https://www.lareb.nl/coronameldingen) in The Netherlands were used to calculate the number of severe and fatal side effects per 100,000 vaccinations. Unfortunately, in the manuscript by Harald Walach et al. these data were incorrectly interpreted which led to erroneous conclusions. The data was presented as being causally related to adverse events by the authors. This is inaccurate. In The Netherlands, healthcare professionals and patients are invited to report suspicions of adverse events that may be associated with vaccination. For this type of reporting a causal relation between the event and the vaccine is not needed, therefore a reported event that occurred after vaccination is not necessarily attributable to vaccination. Thus, reporting of a death following vaccination does not imply that this is a vaccine-related event. There are several other inaccuracies in the paper by Harald Walach et al. one of which is that fatal cases were certified by medical specialists. It should be known that even this false claim does not imply causation, which the authors imply. Further, the authors have called the events ‘effects’ and ‘reactions’ when this is not established, and until causality is established they are ‘events’ that may or may not be caused by exposure to a vaccine. It does not matter what statistics one may apply, this is incorrect and misleading.
The authors were asked to respond to the claims, but were not able to do so satisfactorily. The authors were notified of the retraction and did not agree.
Update, 2330 UTC, 7/4/21: Walach and a co-author, Rainer Klement, sent us a statement they had submitted to Vaccines. In it, they write, “The true reason seems to have been pressure on part of some editors of the journal” and “The timeline suggests that the Journal was not really interested in our response and that our response was irrelevant to the retraction.” Read the entire response here.
Like Retraction Watch? You can make a one-time tax-deductible contribution or a monthly tax-deductible donation to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at team@retractionwatch.com.
A number of questions arise in my mind about this incident.
1. The lead author as you note has also released another research paper on CO2 content of air breathed in by children wearing masks. How should this retraction affect the author’s work elsewhere? What about co-authors having no official role in the retracted work?
2. The lead author has not accepted the retraction and expresses no interest in refining their work. What if anything happens next?
3. The focus so far is entirely on the authors, particularly the lead author. What about the peer reviewers who approved the paper? Why did editors feel the need to resign? What sorts of conflicts are going on behind the curtain? Should demands for transparency extend beyond authors to reviewers and editors?
I’ll repeat here the email I sent to the editors, as I can’t find a more succinct way to put it:
Dear editors,
Regarding the above retraction, while I acknowledge that the authors could have been more precise in clarifying that the vaccine adverse reaction data they were using was associative and not causal, I must point out the enormous double standards involved here.
If your journal wishes to establish the requirement that vaccine derived injury can only be counted after case-by-case evaluation, then fine. Unless you wish for your publication to practice willful inconsistency and bias therefore, I request you please commence retraction of ALL papers quoting SARS-CoV-2 “deaths”. The reasons are twofold:
a) The method for counting SARS-CoV-2 deaths is an absolute abomination and an insult to basic measurement science, and
b) To be consistent with the causal evidence standard you just imposed on Wallach et al, all the papers quoting SARS-CoV-2 deaths are in error because a causal link has not been established in all but a fraction of those incidents.
Regards,
John W Clark
Engineer and Principal Measurement Scientist, BE (Hons 1A), Legal Metrologist
Melbourne Australia
@John W Clark: I would like to learn more about your views on how and why attempts to estimate COVID deaths is an “absolute abomination and an insult to basic measurement science”. This sounds existentially serious!
This appears to be the argument, common amongst anti-vaxxers, that Covid deaths are counted as deaths from any cause within 28 days of a positive test and therefore vaccination deaths ought to be counted as deaths from any cause within 28 days of a vaccination.
I don’t know about the statistics in Australia, but in England, there are two measures in use: NHS and ONS. NHS figures are deaths in hospital from any cause within 28 days of a positive test. That’s essentially a management statistic, and it’s available daily, so it’s popular with media and government: regrettably so, in my view. Presumably that’s the sort of count that is being referred to here. ONS figures are based on death certificates, in which the cause of death has been certified by a medical practitioner with Covid as underlying or contributory cause of death. These figures take time to compile and are usually released weekly.
As it happens, the two sets of figures, NHS and ONS, are quite similar, and are in turn in broad agreement with the 2020 figures for excess deaths week by week over the five-year average.
Keith!
John is referring to the well established fact that the statistics of covid deaths releashed from John Hopkins University database (and from the health authorities in presumably all states and nations) are not establishing any causal relations, but are solely ASSOCIATIVE (such as “died up to 30 days after testing positive”).
Or as Dr. Ngozi Ezike put it:
“If you were in hospice and had already been given a few weeks to live, and then you also were found to have COVID, that would be counted as a COVID death. It means technically even if you died of a clear alternate cause, but you had COVID at the same time, it’s still listed as a COVID death. So, everyone who’s listed as a COVID death doesn’t mean that that was the cause of the death, but they had COVID at the time of the death.” Dr. Ngozi Ezike, Director, Department of Public Health, Illinois
https://week.com/2020/04/20/idph-director-explains-how-covid-deaths-are-classified/
“and from the health authorities in presumably all states and nations”. Not at all, as I explained above.
“ONS figures are based on death certificates”,
Which is completely irrelevant, as long as these death certificates are mere assessments that can be based solely on the presence of a positive test up to 30 days before death of the deceased. The UK death certificates referred to by ONS are NOT necessarily based on autopsy and therefore the death certificates are in no way proof of causation anymore than a death certificate claiming vaccine as cause of death would be if it was not based on autopsy. When death certificates are not based solely on autopsy it means that excess death rates can have other undetected causes, such as N1H1, etc.
Richard Pinch!
When Dr. Ngozi Ezike insists: “If you were in hospice and had already been given a few weeks to live, and then you also were found to have COVID, that would be counted as a COVID death.” This of course means that COVID will be regarded (counted) as the cause of death in death certificates as well, even in cases were COVID was not the cause of the death. That this happens regularly have been proved time and time again by subsequent autopsies. Taking a death certificate as proof of causation is like eating the restaurant´s menu-cart because it has “soup” written on it.
Dr Ezike is, I believe, Director of Public Health for the state of Illinois, in the USA. I was referring to the statistical practice of the Office for National Statistics in England, a constituent nation of the UK. These are different countries, with different health authorities and different ways of defining and recording the notion of a “Covid death”. Hence my comment that the “presumably all states and nations” claim was incorrect. What you say about death certificates may or may not be true about the state of Illinois: I don’t know. It is not true about death certificates in England.
How you count deaths from Covid, and indeed how many die, is interesting but makes no difference to the number of deaths.
Claiming that two people died from COVID-19 vaccinations for every three cases the vaccines prevented is quite a different kettle of fish, it most certainly will make a difference to the number dying.
An additional factor not included in this now retracted paper is around vaccinated individuals who have gone on to die from SARS-CoV-2.
I see that as a worthwhile factor for inclusion when drawing comparisons, and if we are to begin playing top trumps of vaccine VS SARS-CoV-2 deaths it should be an important data point.
The sad scenario here is that measurement of SARS-CoV-2 as a primary factor of death is opaque. Even if we use the slightly more accurate UK ONS measurement (inclusion of COVID-19 on the death certificate) it does not imply that COVID-19 was the sole or primary factor – and in the same breath we cannot give certainty to deaths primarily caused by COVID-19 vaccination.
“Even if we use the slightly more accurate UK ONS measurement (inclusion of COVID-19 on the death certificate) it does not imply that COVID-19 was the sole or primary factor”
The ONS writes “We use the term “due to COVID-19” […] when referring only to deaths where that illness was recorded as the underlying cause of death. We use the term “involving COVID-19″ […] when referring to deaths that had that illness mentioned anywhere on the death certificate, whether as an underlying cause or not.”
During 2020, roughly 90% of deaths “involving” Covid were deaths “due to” Covid.
I must be missing something extremely obvious. If the paper were correct, and there were two deaths from vaccines for every three cases prevented, does that mean the authors were claiming that the vaccines were the direct cause of these deaths? If so, does it imply that for every five doses, three cases were prevented, and two people were killed? As I say, I must be missing something obvious, because if that were the case, the vaccines would have killed millions — dozens of millions — and I would be shocked that the paper ever got published, because such a claim is clearly absurd.
Would someone mind explaining the missing piece of this puzzle to me?
From Comment #11 in the PubPeer thread (h/t Philobota Futilis) we learn that
This article was recently republished in the journal “Science, Public Health Policy, and the Law”, Volume 3:87-99, August, 2021, apparently without a DOI, see https://www.publichealthpolicyjournal.com/ This journal is related to IPAK, the Institute for Pure and Applied Knowledge, see https://ipaknowledge.org/
There may be a broader problem at MDPI, extending to more than one journal, and inviting more questions. The publishers were recently soliciting submissions for a Special Issue of COVID, to be devoted to “SARS-CoV Spike Glycoprotein: Structure and Function”.
https://web.archive.org/web/20210901033252/https://www.mdpi.com/journal/covid/special_issues/cov_spike
To be edited by Drs Konstantin Poulas and Konstantinos Farsalinos, tobacco advocates and opponents of vaccination, best-known for their views that the best way to avoid and treat COVID-19 is with nicotine.
At some point the Special Issue was wished into the cornfield and the link is now 404, but someone at MDPI really needs to explain why the editors’ proposal was accepted in the first place.
A responsible individuals who do not want their loved ones injected with hastily prepared, woefully tested, and poorly monitored experimental genetic therapy of dubious efficacy does not make someone an antivaxxer. It makes them a reasoning rational human being.
Now on to the standards applied in the study…All deaths with SARS-CoV-2 were counted as SARS-CoV-2 deaths regardless of comorbidities or whether the death occurred on the first or fortieth day of infection. Were they not? That’s about as arbitrary a standard as one can imagine, but the rationale was “it was better to err on the side of safety.”
This rational of erring on the side of safety should also apply to the hastily prepared, woefully tested, and poorly monitored experimental genetic therapy of dubious efficacy.
You reasoning has turned American into a waiting room filled with 230 million patients forbidden by law and corporate charter from getting a second opinion.
It is so sad that the flawed reasoning of otherwise rational human beings like Michael Gentile is shared by so many others. Equally sad is the fact that, like the delusions of dementia patients, their views are soo impervious to scientific reality. What a way to watch Darwinian principles at work!