Three different death counts, one missing case: what Wolfdietrich Burde wrote to the PEI before May 15

On 25 February 2026, Federal Finance Minister Lars Klingbeil presented the German government’s action plan against organised crime. The key sentence, documented on the press release of his own Federal Ministry of Finance:

“The perpetrators defraud the public, endanger lives, and cause billions in damages every year.”

The sentence is meant as a standard against criminals. It only works as a standard if it also holds when the public is being kept uninformed by a federal agency. That is precisely the question that Wolfdietrich Burde put to Germany’s Paul-Ehrlich-Institut (PEI) in two emails on 27 and 28 April 2026, with a deadline for an answer of 15 May 2026.

1. Who is Wolfdietrich Burde

#

Wolfdietrich Burde reports that he fell critically ill in early 2022 with anaplastic large cell lymphoma (ALCL, ALK-), a member of the non-Hodgkin lymphoma family. He had been vaccinated with Comirnaty, batch numbers ET3045 and SCTN4, on 14 April and 11 November 2021. He filed his case as a suspected adverse drug reaction with the PEI on 12 September 2022. The PEI case number for this report is CADR2022306791 — that number anchors the entire investigation that follows.

Burde gave his explicit consent to be named in this article, with the standing caveat “without warranty as to correctness or anything else, except that I have worked and am working to the best of my knowledge and conscience”. He puts his own case number forward for verification. That is not the tone of an anecdote. It is the method of an auditor with his own case.

2. The anchor: the 679,777-row spreadsheet

#

Since 2021 the PEI has published spreadsheets of suspected adverse-reaction reports for COVID-19 vaccines on its website. The current version for the vaccination period 27.12.2020 – 31.12.2021 contains 679,777 rows (own verification on 29 April 2026 against the XLSX downloaded from the PEI website, dataset as of 31 August 2025). Burde sorted the same spreadsheet multiple ways: by vaccination dates 14.04. and 11.11.2021, by batch numbers ET3045 and SCTN4, by reported reactions including lymphoma, non-Hodgkin lymphoma, NHL, anaplastic large cell lymphoma, ALCL.

His own case, whose existence is established by the case number CADR2022306791 in correspondence with the PEI, does not appear in the spreadsheet. Burde wrote on 27 April 2026:

“No case fits, evidently my case is not in the spreadsheet. […] If I have made a mistake, please let me know my Case-ID.”

That is the first finding: a suspected adverse-reaction report submitted to and processed by the PEI is not findable in the spreadsheet of those reports published by the PEI itself.

3. The three different death counts

#

The second finding emerges from Burde’s email of 28 April 2026 and is arithmetically tight. For the same vaccination period 27.12.2020 – 31.12.2021, the PEI has published — at three temporally distinct points in time — three different death counts as suspected adverse drug reactions with fatal outcome:

The 2,060 can be verified verbatim from the PEI’s own Pharmacovigilance Report (PDF p. 22). The 637 can be verified from the PEI’s current XLSX — my own count over all distinct case IDs returns 639 (the difference of 2 is explained by Burde’s not counting the “terminal phase” category and one additional stillbirth). The 1,217 comes from the table published before November 2024, whose breakdown Burde documents: cardiac death (29) + brain death (11) + sudden cardiac death (54) + sudden death (181) + terminal phase (2) + death (930) + death of a premature infant (1) + stillbirth (9) = 1,217. The arithmetic is clean, and every other number Burde cites is exact. The old table itself is no longer directly retrievable from web archives — the PEI added a confirmation form to its download page that blocks archive.org snapshots.

What this sequence means in plain words: between March 2025 and 31 August 2025, in five months, the PEI’s own count of suspected fatal adverse-reaction reports for the vaccination period 27.12.2020 – 31.12.2021 fell from 2,060 to 637 — a reduction of about 69 per cent.

4. The justification gap

#

Cleaning up suspected-case data is not in itself a scandal. Duplicates, nullified entries, mis-classifications are routinely identified in pharmacovigilance databases. The PEI itself documents one such cleanup explicitly in its Pharmacovigilance Report as of 31.12.2024:

“Of the 3,323 suspected fatal-outcome cases reported directly to the Paul-Ehrlich-Institut, 203 were identified as duplicates and consolidated. 25 cases are nullified reports […]. In addition, four cases with fatal outcome were identified as miscarriage reports erroneously classified as deaths. A further five cases were reported in the context of a study and not via spontaneous reporting […]. Thus 3,086 suspected adverse-reaction cases with fatal outcome […] were included in the analysis presented below.”

That is the reduction 3,323 → 3,086 (total, all vaccination periods) — a reduction of 6.1 per cent, methodologically named: 203 duplicates, 25 nullified, 4 miscarriages misclassified as deaths, 5 study reports.

The second, very much larger reduction — 2,060 → 637 for the vaccination period 27.12.2020 – 31.12.2021 alone, that is minus 69 per cent in five months — is not findable in a comparably accessible public methodological document (research as of 29 April 2026). It simply happened in the spreadsheet.

That is the structural question Burde asked. Not “what is the truth behind the numbers”. Rather: when the same federal agency, within five months and for the same period, performs a 69 per cent reduction in a public spreadsheet, what methodology is that reduction following, and why was that methodology not published in the same step in which it was applied — the way it had been a published standard in the March 2025 bulletin?

5. Miscarriages: a second analysis with temporal distribution

#

From the previous PEI table (snapshot 28 November 2024) Burde additionally produced a distribution of miscarriage suspected-reports by month of vaccination:

Burde’s observation, in his own words:

“You can see that the number of miscarriage reports oscillates with the rhythm of vaccination intensity. […] In any case there could also have been a numerically modest push for miscarriages, and thus non-births, through the vaccinations in 2021.”

The data are pulled from the PEI table. The interpretation (“could have been a push”) is Burde’s hypothesis — it is marked as such, it is not adopted in this article as a methodologically established causal finding. What is supportable from the perspective of this article is this: the miscarriage suspected-reports too are present in the May 2025 snapshot of the table, in the same data source where Burde’s lymphoma case is missing.

6. Destatis 2026: the lowest German birth year since 1946

#

On 28 April 2026 the Federal Statistical Office (Destatis) published, in press release 26/146, the lowest value: about 654,300 babies born in Germany in 2025 — the lowest figure of the post-war era. Fourth consecutive year of decline. In 2021 the figure was still around 795,000.

The Federal Office’s own explanation in the press release:

“The low birth figures of recent years arise from two developments: the entry of the numerically small 1990s birth cohorts into the important fertile age range of the early thirties, and the total fertility rate that has been declining since 2022.”

Burde re-stated that sentence:

“I would put it differently: the low birth figures of recent years arise from two developments — the entry of the numerically small 1990s birth cohorts into the important fertile age range of the early thirties, and the low birth figures since 2022. In plain German: something defined by itself. Why are we forced to put up with this kind of language from federal agencies?”

The point is linguistically exact. The “total fertility rate” is the average number of children per woman of fertile age. When a federal statistical office explains the decline of the annual birth count by a decline in a metric that is itself a measure of exactly that decline, this is not a causal explanation. It is a re-phrasing of the observation.

Burde’s own conclusion that follows — “this can only be due to the vaccinations” — is his. As a hypothesis it is argumentatively understandable (the birth count breaks downward in 2022, after the vaccination push, not in 2020/2021 where it was still in the prior years’ range), but it is not methodologically forced: there are several plausible contributing channels, from long-COVID effects through socio-economic shifts to pharmacological mechanisms. A serious investigation of this correlation and possible causation is not a two-sentence statement. It is the research that Destatis does not do and that the public does not get in a press release.

What is supportable: a press release that explains the lowest German birth year since 1946 with a tautology answers no question. It closes a question.

7. The Klingbeil standard, returned to sender

#

“The perpetrators defraud the public, endanger lives, and cause billions in damages every year.” That is the standard which the sitting Vice-Chancellor and Federal Finance Minister set in February 2026 for the federal government’s handling of organised crime. Translated, that standard reads: anyone who keeps from the public information that concerns its life is causing damage that must not remain without consequence.

The question that arises from Burde’s emails to the PEI is exactly that standard, turned against another federal agency. If the Paul-Ehrlich-Institut, the German federal institute for vaccines and biomedicines, in five months publishes three different death counts for the same vaccination period — with a major reduction (-69 %) of the figure that has no publicly linked methodology —, and if a documented suspected-case (CADR2022306791) of a critically ill patient is not findable in the spreadsheet meant for it, then the Klingbeil sentence applies not only to smugglers and money launderers. It applies to the agency itself.

That is not polemic. It is the simple application of a publicly stated standard.

8. The deadline — and what will be visible on 16 May

#

In his email of 27 April 2026, Burde set the PEI a deadline for response of 15 May 2026:

“I request notification by 15.5.2026 of the reasons for this. Should there be criteria according to which certain reported cases are not to be included in the spreadsheet, please state those criteria.”

Three response possibilities are conceivable:

1. The PEI answers publicly and substantively. It states the methodology of the reduction 2,060 → 637, it clarifies the whereabouts of case CADR2022306791 in the spreadsheet. This would be public-information clarification in the proper sense.
2. The PEI answers privately and substantively. It clarifies Burde’s individual case but does not name a public methodology of the reduction. This would be the closure of an individual case while leaving the structural question open.
3. The PEI does not answer, or answers evasively. This would confirm the observation that the public is informed about a large, silently-performed reduction in a suspected-cases spreadsheet neither methodologically nor in personnel terms.

Which of these three responses is given can be documented after 15 May 2026 — on the basis of what is publicly visible on the PEI website. A follow-up post will do that.

Conclusion

#

A federal agency that, in a public spreadsheet within five months, reduces its death count for the same vaccination period by 69 per cent and provides no publicly linked methodology has done precisely what the standard “do not defraud the public” rules out. The question whether this reduction was methodologically justified is open — but it is not the first question to be answered. The first question to be answered is why it was not publicly explained.

Anyone who will not credit Burde — a private citizen who used his own life-threatening cancer diagnosis as the anchor for a spreadsheet-audit method — should consider whether they wish to be quoted, on a different topic, by a federal finance minister presenting action plans against the defrauding of the public. That standard is universal, or it is not a standard.

On 16 May 2026 we will be able to measure whether the standard holds in this case.

Sources and references

• PEI spreadsheet of suspected-case reports 27.12.2020 – 31.12.2021 (XLSX, 32 MB, dataset as of 31.08.2025), retrievable via the PEI download page for COVID-19 vaccine suspected-case reports.
• PEI: Bulletin on Drug Safety Issue 1/2025 (March 2025), article on suspected fatal-outcome adverse reactions after COVID-19 vaccination, pp. 15-28.
• PEI: Pharmacovigilance Report as of 31.12.2024 (PDF, official publication).
• Federal Statistical Office (Destatis): Press release No. 26/146 of 28 April 2026, “Number of births in 2025 falls to the lowest level of the post-war era”. URL: https://www.destatis.de/DE/Presse/Pressemitteilungen/2026/04/PD26_146_126.html.
• Federal Ministry of Finance: Press release of 25 February 2026, “Federal Government adopts action plan against organised crime”, quote Lars Klingbeil.
• Correspondence Wolfdietrich Burde to pei@pei.de and Leitung@pei.de, 27 and 28 April 2026, with notes from 28 and 29 April 2026, with the author’s publication consent of 29 April 2026.
• PEI case number of the suspected-case report: CADR2022306791 (filed 12 September 2022).

Methodological note: This article follows the methodology documented for the structural-analysis series of this blog (see Methodology). Third-party material is quoted verbatim and source-marked. Burde’s evaluative statements are flagged as such and are not adopted as methodologically established causal findings. What counts is the documented inconsistency of the public publications of a federal agency, not the derivation of a biomedical causation from that inconsistency.