[throwaway406382]

It is possible to take a finite duration and look at the unvaccinated+covid, unvaccinated+no-covid, vaccinated+covid, vaccinated+no-covid and see differences. And the signals a so large between unvaccinated and vaccinated you don't need much in the way of fancy stats.

Also consider that it wasn't until the late 90's that the flu vaccine was approved for pregnant women and they were still worried about it. So we have the original adverse event studies post flu vaccine and compare them to post covid vaccine.

You are right, there are lots of confounds and interactions, we are careful to handle that properly. We're a bunch of nerds who don't have a dog in the political fight, we just want to show the data.

[manwe150]

Good, that is what I was hoping to hear. Looking forward to seeing it published. Though unvaccinated+no-covid does not exist in steady state, so any signal that includes them cannot be extrapolated into the future. But it is seeming like vaccinated+no-covid won’t exist either in the long term steady-state. So we might eventually be having to deal statistically with considering frequency of infection instead of binary yes/no. Lots of math to consider. Best wishes to you in this.

[jomjomv]

Which vaccine does your research refer to? Can you share?