10 Comments

If someone got Pfizer or Moderna after J&J (because J&J was single-dose, and later considered insufficient by out overlords), would that count as 1st dose Pfizer/Moderna?

Expand full comment

These cases are probably listed as two "IDs". The data shows that third vaccinations from Moderna and Pfizer were most likely recorded as first vaccinations. For example, we find Moderna 1: 174k, Moderna 2: 99k, Moderna 3: 56k.

Expand full comment

Well done. Good support for what Steve is trying to achieve. Now let's get the real office of statistics data and join forces to prove the hypothesis once and for all.

Together you may have a chance.

Expand full comment

Thank you, Sherrie... although I have stopped trying to achieve something when looking for correlations in data. I think that's a good way to go. I'm also extremely cautious about formulating hypotheses, especially when it comes to data on the effects of vaccination. In fact, all publicly available data sets (UK, Czech Republic, Netherlands) have been massively corrupted so far.

Expand full comment

If your histogram of survival time by brand, the reason why the first few bars are lower for Moderna than Pfizer is probably because the average date of the first dose is 2021-04-14 for Pfizer but 2021-06-11 for Moderna. Pfizer vaccines were more commonly given in the first two months of 2021 when there was a rapidly decreasing trend in the background mortality rate because the COVID wave in the winter of 2020-2021 was passing by: sars2.net/connecticut.html#Deaths_by_weeks_since_vaccination.

Expand full comment

The percentage of first doses that were given in the last 3 months of 2021 was about 37% for Moderna and 17% for Pfizer. Most of those were probably misclassified third doses. But there was also an increasing trend in background mortality during the last three months of 2021, which is another explanation for why the temporal healthy vaccinee effect appears to be more pronounced for Moderna than Pfizer.

The file Medicare-2-1-23.xlsx which Kirsch published last year had a comment which said "The vax date listed in the record is the date of the FIRST vaccine (that Medicare knows about for the patient) ONLY". And another comment said: "Be careful when looking at first vaccination data after October 2021. It may in fact be a third shot but 'appear' to be a first shot due to record keeping errors." (https://kirschsubstack.com/p/game-over-medicare-data-shows-the)

Uncle John Returns found that in the new Connecticut data the number of first doses in late 2021 is much higher than the total number of new first doses reported in Connecticut over the same period of time: https://x.com/UncleJo46902375/status/1840441446777683997.

Expand full comment

henjin, your calculation includes the misclassified third vaccinations. If you limit the quantities to the actual first vaccinations, e.g. before August 1st, Pfizer and Moderna only differ by 10 days.

Expand full comment

Yeah but even when I only included doses given before August 2021, Pfizer gets a much bigger drop in the baseline during the first 10 weeks after vaccination than Moderna: https://i.ibb.co/tCkCXdF/connecticut-medicare-deaths-by-weeks-after-vaccination-doses-before-august-2021.png. However even then it seems like the temporal healthy vaccinee effect might be stronger for Pfizer than Moderna. But when I only included first doses given in March to July 2021, now conversely the temporal HVE seems to be more strong for Pfizer than Moderna: https://i.ibb.co/m8LMNcb/connecticut-medicare-deaths-by-weeks-after-vaccination-doses-in-march-to-july-2021.png.

Expand full comment

Thank you for your comment and yes, I was aware of the possible seasonal influences and thought about it for a while. The matter did not seem as clear to me as you write and I decided not to go into it any further for the time being. The waves of death are preceded by waves of illness. It could also be the other way round, that an earlier brand was exposed to a stronger HUB.

Expand full comment

I tried plotting deaths by weeks since vaccination and vaccine type, but I calculated a baseline number of deaths by multiplying the number of person-days for each combination of age and month with the mortality rate in Connecticut for the same combination of age and month. During the first 10 weeks from vaccination, Pfizer had a sharp drop in the baseline but the baseline for Moderna remained roughly flat: sars2.net/connecticut.html#Deaths_by_weeks_since_vaccination_and_vaccine_brand.

Expand full comment