Thread by @mimifontaine

As an add-on to my thread explaining why regulators are unlikely to detect an increase in the risk of blood clots generally due to the #AstraZenecaVaccine (annotated 100% with #JurassicPark GIFs!) I present:

Hang on a minute, *does* AZ increase overall clot risk??

[thread]

The previous thread outlined how the analysis done by #EMA and #MHRA to detect an increase in the incidence of #BloodClots after AZ was crap because they forgot they were dealing with vaccine reporting (reports filed *if* someone suspects vax link) and…

https://twitter.com/mimifontaine/status/1392252103876304896

*not* a clinical trial where participants are monitored and all health issues are reported

So they ended up comparing a huge baseline rate to the tiny number of clots that actually got reported to them and now they think the AZ vaccine reduces the risk of blood clots by 98%

So I checked for myself using the UK Yellow Card data if there was any sign of blood clots happening more frequently after the #AstraZenecaVaccine

The results are very interesting

We know comparisons to baseline are meaningless for conditions like DVTs and pulmonary embolisms because they are fairly common so doctors aren’t surprised to see them and are therefore less likely to suspect a link to the vaccine and file a Yellow Card report

If only we had a control group...

We do (kind of) - the Pfizer side effect reporting!

Of course there are caveats - Pfizer rolled out earlier, differences in the kind of person who got Pfizer instead of AZ (age, gender etc) but hey - it’s better using the baseline

To avoid potential over-reporting of clots after #AstraZenecaVaccine due to the media attention on AZ and blood clots from mid March I’ll compare the rates of clots reported between the two vaccines as of 15 March

Ready for the results?

So... it looks like DVTs were 3.5 times more likely to be reported after the #AstraZenecaVaccine than after the #PfizerVaccine...

Reports of pulmonary embolism were reported at 1.6x the rate...

And the trend continues for non-site specific thromboses

I dunno how to test for significance, is O/E with Pfizer rate as E the right way? DVT post #AstraZenecaVaccine v significant (99%) if so

Or we could just take MHRA’s word for it that there’s no increased clot risk from AZ

Although.. that’s what they said about the other clots…