What's in the PEI?

Jonah 4:5

The Paul Ehrlich Institut (PEI, the German government agency responsible for drug monitoring) is a strange beast. They do not publish the data they had promised to publish, they do not keep protocols, and yet sometimes things are moving in the darker corners of their web page. One year ago, I evaluated their raw data on adverse events after Covid vaccination. This week, I don’t know why, I happened to drop by again, and behold, the files have changed.

Since the files now contain data reported throughout 2024 instead of the first half of 2024 only, we might expect the total numer of events to have increased. That is in fact the case: the number of events has gone from 1,001,539 to 1,160,091. However, there are all sorts of strange effects (I didn’t bother to translate the labels):

• While it is true that the relative increase is largest for reporting year (“Meldejahr”) 2024, there are many more events for 2021 and 2022 as well:

  

• Although the number of events has gone up (by 16%), the number of cases (i.e., combinations of individuals with events) has gone down by 5%, from 348,987 to 331,921. There are therefore more events per case now. This might be explained by the matching of case reports after different vaccine doses for the same person.

• It is very hard to map cases because they completely replaced the case IDs.

• There are disproportionately many more events related to the Moderna vaccine, and few related to AstraZeneca. I guess that this is not surprising considering that vaccination with AZ was stopped on 19.03.2021.

• Now there are fewer events for children, and toddlers are missing completely:

  

• These are the top 30 event types by absolute difference, with negative differences (fewer events now than in the old data set) marked in red:

  

  Influenza-like illnesses (Grippeaehnliche Erkrankung) have been discarded, mostly to be replaced by influenza (Grippe). COVID-19 (and similar events like “positive Covid test” or “vaccination failure” further down the line) have been eliminated almost completely.

• These are event types filtered by “Blutdruck” (blood pressure):

  

  Something happened to high blood pressure (Blutdruck erhoeht) but not to the other conditions.

• These are event types filtered by “karditis”:

  

  OK, they might have replaced some combinations of myocarditis and pericarditis into myopericarditis.

• These are event types filtered by “tod” (death), minus “lupus”:

  

  Many deaths (of which I overlooked some in my previous report) are missing now…

My hypothesis is thus. What we are seeing is the result of two concurring processes:

1. New data have been incorporated, either due to slow reduction of a backlog, or maybe even due to clandestine integration of the SafeVac 2.0 data the release of which we have all been waiting for.

2. Efforts have been made to scrap events that might worry the public, such as deaths, myocarditis and pericarditis, events concerning children, and Covid infection despite vaccination.

What can I do? I will wait for another year and look again.