A study published last week that quickly became another flashpoint for those arguing that COVID-19 vaccines are unsafe has earned an expression of concern.
[Please see an update on this post; the paper has been retracted.]
The original paper, published in the MDPI title Vaccines, claimed that:
The number of cases experiencing adverse reactions has been reported to be 700 per 100,000 vaccinations. Currently, we see 16 serious side effects per 100,000 vaccinations, and the number of fatal side effects is at 4.11/100,000 vaccinations. For three deaths prevented by vaccination we have to accept two inflicted by vaccination.
However, the study’s methods quickly drew scrutiny, and at least two members of Vaccines’ editorial board, Mount Sinai virologist Florian Krammer and Oxford immunologist Katie Ewer, said they have stepped down to protest the publication of the paper.
Here is the expression of concern, published yesterday:
The journal is issuing this expression of concern to alert readers to significant concerns regarding the paper cited above [1].
Serious concerns have been raised about misinterpretation of the data and the conclusions. The major concern is the misrepresentation of the COVID-19 vaccination efforts and misrepresentation of the data, e.g., Abstract: “For three deaths prevented by vaccination we have to accept two inflicted by vaccination”. Stating that these deaths linked to vaccination efforts is incorrect and distorted.
We will provide an update following the conclusion of our investigation. The authors have been notified about this Expression of Concern.
Corresponding author Harald Walach told Retraction Watch that he does not agree with the expression of concern. He said he and his co-authors are meeting to formulate a detailed response to the notice [see an update on this post]:
The short version is: we have used and analyzed the data correctly, and not incorrectly. But that the data are less than optimal is clear to everyone and we said so in our paper. The purpose is to generate enough momentum for governments and researchers to finally create the good data that are long overdue.
Two of the paper’s authors were among those who posted a call to retract a January 2020 study of PCR testing for COVID-19 by virologist Christian Drosten. That call was rejected by the journal.
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 [email protected].
Color me unimpressed by Florian Krammer’s and Katie Ewer’s resignations from the editorial board. Like too many, they apparently accepted the appointment for the perceived prestige, but now are running away when things go south.
I think that is a bit unfair since the whole editorial board do not see every manuscript submitted to a journal. If the submission was handled by another editor, then all of the other editors (perhaps with the exception of the EIC) would only see the paper after publication.
I don’t think that it is unfair in the least. Surely you don’t believe that they didn’t know what they were lending their names to when they agreed to join the editorial board of an MDPI journal trading on the name of a much more established, society-sponsored publication?
“at least two members of Vaccine’s editorial board, Mount Sinai virologist Florian Krammer and Oxford immunologist Katie Ewer”
Should be _Vaccines_’s or “the editorial board of _Vaccines_”, so as not to be confused with the Elsevier journal _Vaccine_. (I know you mention the publication of concern is _Vaccines_ earlier in the article.)
Fixed — thanks.
I do not see the problem with this article. They have some incomplete data and ended up with an original conclusion. I do not see the difference with many many publications in Earth System Sciences which do the same (using small dataset, making gross assumptions,…), on non political topics. These types of studies are usually useless since the data used are not representative of real world. but in ESS, we like global studies, not applied research 😀
The problem with this article is that it misrepresents the data being used. The authors “decided to use the data of the Dutch national register” and the website they link states, at the top, in bold “Belangrijk! Lees eerst deze toelichting
Een gemelde bijwerking hoeft niet altijd door het vaccin veroorzaakt te zijn. Klachten of aandoeningen kunnen ook door een andere oorzaak na de vaccinatie zijn ontstaan.”
which translates to something like
“Important! Read this explanation first
A reported side effect may not always be due to the vaccine . Complaints or disorders can also have arisen from another cause after the vaccination.”
The paper then proceeds as if every death reported after vaccination was due to vaccination: “approximately four people will die from the consequences of being vaccinated per 100,000 vaccinations delivered.” This is significantly different from what the data source states in the quote I gave. Just to spell it out: the numbers reported by the regulator are for deaths that occurred after, not neccessarily as a consequence of, vaccination. This discrepancy is of course sufficient to completely invalid the subsequent analysis completely.
In other words, people should ask if people died “with the vaccine” or “because of the vaccine”?
In the camp of people who didn’t believe that COVID-19 was dangerous or very lethal this question was always asked “it wasn’t proven that all these deaths are caused by COVID-19!” Now I wonder if people from that camp will use the same scrutiny for the “deaths caused by vaccine”.
Your post was very helpful, thank you!
It’s even worse than that.
This abysmal paper cherry-picks two data sets. One is used to estimate the number of lives saved by vaccination. They picked Israel, a country with a very low Covid death rate. They also underestimate the number of lives saved by vaccination by only counting deaths during the couple of months duration of the clinical trial, so assuming the vaccine then stops working.
This is then compared to a Dutch data set of adverse events after vaccination. Why Holland? Because they report a lot of events. As has been said, this assumes that all deaths after vaccination were due to the vaccine. To those who say “Aha, but that’s what YOU do with deaths after a Covid test!” I would say no, we do not. We use proper statistics, not made-up methods nobody else uses, to compare the death rate after a Covid test to the underlying death rate expected in the population. This was not done here. If vaccination was killing 67% as many people as Covid, which this woeful paper claims, we’d see that in excess death stats, and we simply don’t.
These choices can only have been made deliberately so as to exaggerate the risk of the vaccine and minimise the risk of Covid. Incidentally this author has published before on subjects ranging from homeopathy through telepathy and telekinesis so he has considerable form…
Umm that is exactly what happens after a death of someone with a positive covid test- they are (rightfully) counted as a covid death despite potentially dying of something else. In fact people have been counted as a covid death despite negative covid test if it is determined that symptoms were consistent with covid.
I agree with you regarding this paper however we should have discussions regarding the safety of these vaccines along with who really needs them. Clearly not everyone needs to be vaccinated.
I don’t disagree with you Justin, but I was pointing out one important difference. When people die after a positive Covid test, they (as individuals) are counted as a Covid death. However in judging the statistical significance of these figures, they are compared to the number of deaths expected in a “normal” population (hence excess deaths).
By now, there have been large-scale autopsy studies to determine the fraction of people dying “because of Covid” of these who died “with Covid”. To sum up: 80%-90% of these dying “with Covid” died “because of Covid”.
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. ONS figures are based on death certificates, in which the cause of death has been diagnosed by a medical practitioner with knowledge of the specific case (possibly including but not determined by a Covid test) to have Covid as underlying or contributory cause of death. The ONS figures take time to compile and are usually released weekly.
These two sets of figures are in broad agreement, as it happens.
This is a non-symmetrical comparison. Earth Sciences deal with much different scales (time, volume, number of samples) than medical Sciences. As we have our own standards (i.e. minimum number of concordant zircon in provenance studies), medical Sciences have their own for dataset analysis.
Seems like good old fashion censorship, because they don’t like the conclusions.
What has society become that they can’t investigate the serious concerns raised in this robust study, but instead just dismiss it out of hand.
I don’t think that it is unfair in the least. Surely you don’t believe that they didn’t know what they were lending their names to when they agreed to join the editorial board of an MDPI journal trading on the name of a much more established, society-sponsored publication?
Of course it does, Steve. The problem is when people don’t understand basic science they can end up supporting rubbish like this ‘paper’. It’s not hard to see the problem in both their methodology and analysis, that is, if you care to look.
There is enough good data to conclude that the death rate from covid infection is roughly 2%. This is conservative given that numerous studies have shown that the death rate is likely twice that but many covid linked deaths are not recorded as due to covid. Using the conservative estimate this means that there are 2000 deaths for every 100,000 persons who contract covid. This is the number that should be compared to vaccinations. This number is 1000 times larger than the estimations in the paper attributable to vaccination. The reviewers of this paper are the individuals who failed science. It is clear that they did not critically evaluate the manuscript.