Thank you for raising that good point. I'm not a physician so I'm not sure to what extent physicians are encouraged to investigate and report VITT.

Some VITT cases might be missed, leading to underestimated incidence rate. But some vascular events might also be misattributed to VITT, leading to overestimated incidence rate. So, the two conundrums sort of balance each other, in my humble opinion.

That said, vascular events in general are hard to miss. Thus, by comparing overall vascular events (including VITT) between vaccinated and unvaccinated periods/population, we can fairly gauge what the safety profile of the vaccines looks like.

Several large-scale surveillance studies have done this, finding a small increase in excess events of vascular events mainly from the DNA vaccines. Here are a few recent ones:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9794399/

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9524320/

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9425709/

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9198750/

And please note that it may be better to look at excess cases/events per x people, rather than the relative risk increase. Because a rare event relative to a rare event will still be a rare event, even with huge relative risk. For example, an increase from 0.01 to 0.1 is already a 10-fold increase.