Does the analysis of Schmeling et al. suggest a batch-dependent safety signal for the BNT162b2 mRNA COVID-19 vaccine?

And my email exchange with the authors

I published this short article on my website on 1 July 2023:

https://denisrancourt.ca/entries.php?id=131

Jessica Rose referred to it recently: Here.

My 4-part email exchange with the authors, which followed, is posted below.

My style is to be direct rather than diplomatic, to draw out the best arguments and authentic reactions. Maybe not the best approach?

First, the short article:

Does the analysis of Schmeling et al. suggest a batch-dependent safety signal for the BNT162b2 mRNA COVID-19 vaccine?

By Denis G. Rancourt, PhD

1 July 2023

The short answer is no.

The conclusion of Schmeling et al. [1]

“In conclusion, the results suggest the existence of a batch-dependent safety signal for the BNT162b2 vaccine, …”

does not follow from their analysis.

Let me explain why.

They assign reported adverse effects (AEs) to vaccine batches, and make and interpret graphs of number of batch-assigned AEs versus number of vaccine doses per batch, delivered in Denmark between 27 December 2020 and 11 January 2022. Each point on such a graph (for a given AE type or severity) is for one of the 52 batches used in Denmark in this period.

They show the resulting graph only for all-AEs, irrespective of type or severity (their Figure 1). They admit “Compared to the rates of all SAEs, serious SAEs and SAE-related deaths per 1000 doses were much less frequent and numbers of these SAEs per 1000 doses displayed considerably greater variability between batches, with lesser separation between the three trendlines,” but they do not show those graphs.

That is the full extent of their analysis. In doing this, they exclude all the information about the subjects experiencing the AEs, such as age and sex; and all the information about the injections, such as the date of the injection. They are blind to virtually everything except what is needed to make the batch assignment for each AE.

This blindness by design is acknowledged as: “in the present study, … , demographics of SAE cases, … , were not examined.”

This is a recipe for studying artifacts arising from invisible factors.

Specifically, here is a significant potential problem:

1. The risk of death following injection may be highly dependent on the age of the subject

2. Elderly people were structurally prioritised for injection

3. The priority injections of elderly subjects, administered early in the vaccine rollouts, involved fewer doses than those given to the general population later in the rollout, and would therefore be drawn in smaller numbers, for the elderly in Denmark, from the manufactured batches used at that time.

In fact, it has been shown ― both from a study of reported AEs in the USA (VAERS) [2] and from excess all-cause mortality synchronous with vaccine rollouts [3] ― that the risk of death following injection rises exponentially with age of the subject, doubling every 5-10 years in age.

This means that one must expect age to be a dominant factor, which causes large (order of magnitude) variability in deaths per dose, by age. This is crucial when injections are prioritised for the elderly population, and batches are used at different dates.

Hickey and Rancourt found (their Figure S6, [2]) that the age factor could explain all the batch differences in deaths, and that increased variability in deaths per dose was likewise consistent with increasing median age, in the USA (VAERS).

It was incorrect for Schmeling et al. to conclude as they did knowing that they had excluded the age consideration.

Of course, I do not mean that there are no manufacturing impurities in the vaccines, or that there cannot be “bad batches”, but I am not aware of any evidence that “bad batches” are a significant problem on population scales. The vaccines are toxic, as designed, and that is the main impact.

References

[1] Schmeling et al. (2023) “Batch-dependent safety of the BNT162b2 mRNA COVID-19 vaccine”. Eur J Clin Invest. 2023; https://doi.org/10.1111/eci.13998

[2] Hickey and Rancourt (2022) “Nature of the toxicity of the COVID-19 vaccines in the USA”. ResearchGate [Preprint] (9 February 2022); https://www.researchgate.net/publication/358489777_Nature_of_the_toxicity_of_the_COVID-19_vaccines_in_the_USA  /  Archived at: https://archive.ph/LZpRj

[3] Rancourt et al. (2023) “Age-stratified COVID-19 vaccine-dose fatality rate for Israel and Australia”. Correlation Research in the Public Interest, 9 February 2023; https://correlation-canada.org/report-age-stratified-covid-19-vaccine-dose-fatality-rate-for-israel-and-australia/  

---

And here is the 4-part email exchange:

1. July 1st

Hello Professor Hansen,

I wrote this pithy critique of your paper:

https://denisrancourt.ca/entries.php?id=131&name=2023_07_01_does_the_analysis_of_schmeling_et_al_suggest_a_batch_dependent_safety_signal_for_the_bnt162b2_mrna_covid_19_vaccine

Please let me know if I made any factual errors.

My style is to be clear and direct, but my goal is to help.

Best regards, DR

2. July 2nd

Dear Dennis,

Thank you for your interest in our study.

Just one comment:

Your state that the conclusion of our report was ‘incorrect’ since the authors were ‘knowing that they had excluded the age consideration’.

In fact, we concluded that our study ‘suggests the existence of a batch-dependent safety signal’ - not that it ‘shows’, ‘demonstrates’ or ‘documents’ such thing.

In academic terms, providing a ‘suggestion’ is, of course, not pretending to give the final word - as was also clearly indicated in the rest of the text.

Indeed, the paper listed lots of limitations to the study, for example, the demographics of subjects (e.g., ‘the age consideration’) were not included in the analyses.

The influence of these confounders was not examined for the obvious reason we did not have access to such data (this was denied to us on several occasions).

Therefore, the take that we were ‘incorrect’ by pretending to give a final word on the matter by ways of a premeditated exclusion of data, seems like a straw man argument.

Moreover, the presumption that the age factor accounted for the results may not be true. We have now gained access to data about the age of vaccinated persons with reported SAEs, and in the group of high SAE batches (blue trendline in Figure 1) only 21% of SAEs were reported in people aged 70 years or above, i.e., 79% were younger than 70 years (younger front line health care workers were, of course, also among the first to be vaccinated). For the ‘medium SAE batches’ (green trendline), 22% were 70 years or older, and for the ‘low SAE batches’ (yellow trendline) this ratio was 27%. Therefore, it seems less likely that selective administration to old and frail persons during the first phase of the pandemic was a main determinant for the ‘emergence’ of the high SAE batches. Furthermore, to just accept the premise that a vaccine inherently produces more SAEs in old and frail persons is debatable since these individuals, in fact, are those that are more vulnerable to SAEs, and the SmPCs for the vaccine does not detail a higher risk of SAEs in these subjects.

Generally, it would appear wise for any of us to avoid stating what is ‘correct’ or not in a field where so much clearly remains to be investigated.

Kind regards, Peter Riis Hansen, Professor MD DMSc PhD

3. July 2nd

Dear Peter,

Thank you for your willingness to discuss your paper and my critique.

• https://onlinelibrary.wiley.com/doi/10.1111/eci.13998#pane-pcw-related

• https://denisrancourt.ca/entries.php?id=131&name=2023_07_01_does_the_analysis_of_schmeling_et_al_suggest_a_batch_dependent_safety_signal_for_the_bnt162b2_mrna_covid_19_vaccine

I was explicit that your conclusion was “In conclusion, the results suggest the existence of a batch-dependent safety signal for the BNT162b2 vaccine, …”, which I stated was incorrect. 

In my view, unconsidered potential significant factors, which reasonably should be expected, make it incorrect to conclude that the data shown suggests that there is a batch-dependent safety signal. In the same way, it would be incorrect to conclude that high life expectancies in Western countries "suggest the existence of a latitude-dependent life-expectancy signal", while not knowing about living conditions.

That age is a dominant effect (exponential, in fact, regarding death) is shown in my references "[2]" and "[3]".

Furthermore, Montano (2021), which you did not reference,

https://www.frontiersin.org/articles/10.3389/fpubh.2021.756633/full

found:

"Individuals age 65 and older were associated with a higher frequency of death, hospitalisations, and life-threatening reactions than younger individuals (relative risk estimates between 1.49 99% CI [1.44–1.55] and 8.61 99% CI [8.02–9.23])."

and pointed out:

"The datasets from the EudraVigilance system are publicly available and can be obtained by querying the line listings view of adverse reactions for each vaccine type and reporting year."

Given the extensive scientific literature on the large age-dependence of toxicity in general, when challenged by exposure, your apparent or in-effect avoidance of considering age as a major factor is surprising, I am sorry to say.

Can you show me your Figure-1-like figures for serious SAEs and SAE-related deaths, not shown in your paper?

Kind regards, DR

4. July 3rd

Dear DE,

I beg to differ: As mentioned before, we could simply not gain access to the age data and this was mentioned as one of many limitations of the study.

Despite these limitations, we felt that our preliminary signal was worth reporting, to spark interest in others doing more definitive research on potential batch-dependent differences in safety (and efficacy) of the vaccines.

Indeed, this issue has not been examined previously and might (for obvious reasons) be a more plausible phenomenon to occur in the setting of the pandemic.

Luckily, the expert peer reviewers of the manuscript and the EJCI editorship felt the same way - and as you know, academics are well aware that ‘suggests’ is not ‘demonstrates’, that a statistical link does not infer causality, that study limitations are declared as appropriate etc.

Against this background, I disagree that our study conclusion was ‘incorrect’ and believe that such wording has unfortunate inferences, e.g., that 1) we/the paper’s peer reviewers/your colleagues don’t know what is ‘correct’, and 2) readers should disregard these ‘incorrect’ results.

I rest my case, thank you.

Kind regards, Peter

---