I think @ASPphysician and I experienced a mind-meld today. He is absolutely right that risk of VITT is higher than the upper bound in previous NACI risk analyses. I also thought that VITT vs. ICU was not a fair comparison, as ICU survivability is much higher than VITT survival

This is my version of the table Andrew shared, but here I've used Ontario's ICU admission and survival numbers so that I can compare VITT death to ICU death.

As Andrew notes, with this analysis, AZ only makes sense in older people in hotspots.

Because in all other scenarios and age groups VITT death risk either equals or exceeds COVID death risk. I am *not* advocating use of AZ vax in older people in hotspots, but rather that hotspots be prioritized for mRNA vax.

This analysis also overestimates risk of COVID-19 death while waiting for mRNA vaccine because time has passed since the analysis was done...wait times are shorter now.

As @ASPphysician notes, we are probably going to land close to 1/26,000 for VITT risk by the time this is over (that's the Norwegian estimate) and that would make my analysis less favorable to AZ.

I'll also echo what he's said: AZ vaccine is protecting those who have received it. It's a good vaccine; this is about safety rather than efficacy. I am very confident that you will be boosted with mRNA vaccine in the weeks ahead and have strong protection against SARS-2.