Messy New Pregnancy Preprint
Beware any analysis of Mensah, et al. (until further updates)
A new preprint has arrived which the reader should take care to ignore until it is updated. This post serves as a PSA to save the reader any time and confusion.
The preprint:
Why to avoid:
In processing the results to percentages (to get around the authors’ over-processing and adjusting) it became clear that something weird has gone on at some point in their matching process. I have emailed the corresponding author; we will see if there is any reply or update.
Meanwhile anyone interested can access my reformatting of the data at https://docs.google.com/spreadsheets/d/1qcl5DRGewTn8Lb5sPJK-w1po2iYFo2icckITqbvokHs/edit?usp=sharing - the major discrepancy is in proportion of “vaccinated <90 days before pregnancy” in the control (“no”) groups.
Why I am bothering to tell you to avoid:
If the reader is unaware, there is to-date no good evidence that the Covid vaccines had any measurably negative impact on pregnancy outcomes, anecdotes notwithstanding. This may be because no one is willing to publish such evidence — the CDC/V Safe study all but simply “went dark” after October, 2021. Fine — but the result is still that the evidence hasn’t been published, if it exists.
Nonetheless there is a thriving industry in pretending that there is such evidence, which isn’t hard because of how easy it is to introduce wild outcomes in pregnancy statistics just by defining denominators however one wishes. This is fine1 — everyone needs a hobby — but if the reader encounters such an effort regarding this new preprint, note that all the data may be erroneous and changed in an update.2 (In fact, the preprint mostly shows no safety signals anywhere — except for Neonatal deaths and VTE, but as said, this is possibly only because something went wrong in collecting and processing the raw data.)
Now back to regularly scheduled programming…
If you derived value from this post, please drop a few coins in your fact-barista’s tip jar.
It’s not fine. It’s Toddler Basketball.
At which point any mistaken “take-downs,” if already posted despite the ease of detecting problems in this preprint, will be followed by flustered admonitions that this (failure by critics to detect problems) is why the critics should have the “real data.” Lol
My quest to find undeniable, bulletproof evidence that Covid vaccines negatively affect pregnancy outcomes, mostly was unsuccessful due to denominator issues. (plenty of interesting stuff to discuss about birthrates, but it is correlations, not causations).
However, I had two notable successes finding a solid signal, comparing Pfizer-vaccinated women with Moderna-vaccinated women.
One comparison found 42% increase in miscarriages with Moderna (super low p), and a 93% increase with stillbirths with Moderna (p=0.041). In both cases Moderna (higher dose) was compared with Pfizer.
https://www.igor-chudov.com/p/cdc-data-moderna-causes-42-more-miscarriages
https://www.igor-chudov.com/p/moderna-doubles-the-chance-of-infant
There were 295 total maternal VTEs.
So, 53 women who were jabbed in the 3 months prior to pregnancy had maternal VTE; but only 23 women had maternal VTE if they were vaccinated at any point during the entire 9 month pregnancy?
Am I reading that correctly? It doesn't sound correct.