Did recently released internal Pfizer documentation show a much lower efficacy rate for the vaccine than previously claimed?

Question

Background

A search on Twitter for Pfizer currently shows many people talking about a supposed recent release of documentation (tens of thousands of pages) from Pfizer. A simple Internet search shows many personal blogs discussing the release, sites claiming to have the documents in question (I was unable to access them, and at any rate am not inclined to go through such a large document myself), and news articles debunking a claim about adverse effects.

In the Twitter discussion, I have noticed a few common themes: along with the usual anecdotes (that I see every time there is a swell in Twitter discussion of COVID vaccines) about individuals supposedly harmed by the vaccine and protest of Pfizer's profit margins, there are four distinct claims I have noticed:

• A claim that the leak of the Roe v Wade decision was timed to distract from the contents of the Pfizer documents (this is, of course, an unprovable and irrelevant conspiracy theory, so I have not selected it as the topic here)

• A claim that the documentation indicates over a thousand deaths caused by vaccines during the trials (the news coverage I can find all seems to focus on this claim, making the simple argument that these deaths cannot necessarily be attributed to the vaccine; that's good enough for me as it stands, so this is also not the claim I want to highlight. I suppose this is similar to the classic misuse of VAERS seen in many other covid-19 related questions.)

• A claim that the vaccine was not recommended during pregnancy or lactation, as the studies did not determine whether the vaccine would be excreted in human milk. It was established later that the vaccine is passed on this way, but this has been deemed safe and beneficial, so this is also not the claim I find interesting for skeptics.SE purposes.

Claim

The preceding is largely meant to establish that the documents in question indeed exist and that they have been looked over, but that I don't know where to look for a thorough treatment of the topic. There is one claim outstanding in the discussion which I have not seen addressed. I will cite example Tweets here:

https://twitter.com/amerix/status/1521838512663941121

While the system kept you busy with abortion in USA & War in Ukraine,

Pfizer data came out after a court order.

The vaccine was 12% effective & dropped to less than 1%.

No trials were done on pregnant women before giving them!

Pfizer vax had a 0.83% chance to save you.

Scam!

https://twitter.com/stevienut/status/1521553172665413635

Pfizer data just out! We no why they fought to seal data for 75 years! Vaccine was 12% effective. Never trialled in pregnant women, tested on 7 Rats before it was given to them! Actual Pfizer vaccine had a 0.83% chance to save ur life from Covid. SHOCKED

https://twitter.com/CarloGrims/status/1521663669939187714

Now that the "#Science" is out that the #Pfizer vax is actually 12% effective the first 7 days, and then below 1% after that time lapse... How do we fight the mandates legally? Even the kangaroo courts we have in #Canada won't be able to come up with enough bs to cover for that.

The numbers cited are consistent: it is claimed that Pfizer has now admitted their vaccine to be only "12% effective".

My question is: Where does this number come from, and does it have any salience or validity? How are the anti-vax crowd interpreting the document to infer such a result, and how could this happen with the same data that was used to produce headlines of over 90% efficacy at the time the vaccines were released? It seems as though outright fraud is being alleged.

Answer 1

I'm not entirely sure when the documents were released, but Le Monde connects the twin claims (12% and 1%) of the first tweet in the question to the writings of Sonia Elijah, some of which you can read here, as found by Mad Scientist. Le Monde comments that

Pour arriver à ce « 12 % », Mme Elijah s’appuie sur un document, rendu public depuis au moins le 8 décembre 2020 : la notice d’information envoyée par Pfizer aux FDA, dans laquelle l’entreprise se targue des fameux 95 % d’efficacité du vaccin qui allait être autorisé. Une phrase y interpelle l’ancienne journaliste de la BBC : « Parmi les 3 410 cas totaux de Covid-19 suspectés mais non confirmés dans la population globale de l’étude, 1 594 sont survenus dans le groupe vacciné contre 1 816 dans le groupe placebo », écrit Pfizer.

Un calcul réalisé à partir d’individus… négatifs « Si vous calculez l’efficacité vaccinale à partir de ces chiffres, elle est incroyablement basse, 12 % », s’emporte Sonia Elijah. Sauf que contrairement à ce qu’elle présuppose, rien ne permet d’affirmer avec certitude que ces 3 410 personnes symptomatiques ont bien été malades du Covid-19. Certes, il n’est pas possible d’écarter une part de faux négatifs (des personnes contaminées, mais non détectées en raison d’un problème de réglage du test PCR, ou d’une charge virale trop basse, par exemple).

So, the Pfizer documents are from their Dec 2020 FDA submission, and Elijah derived the 12% efficacy figure by including all the participants who had some symptoms related to Covid-19, although they were not confirmed (e.g. PCR) as being infected with SARS-CoV-2.