I don't know off hand how that 90% number compares to other diseases (or again how reliable their test is), so even that is difficult to judge. That's the kind of thing they should be telling us in their discussion.

I suspect there's an answer there? But I also suspect that COVID-19 is simultaneously more infectious and less harmful than we're commonly assuming it is, based on the LANL/DARPA studies. Which calls into question our entire disease model and right now we're in a space where we cling to any new factoid like a rope from a coast guard cutter.

When I was trying to figure out how to kill COVID with UV light I dug deep into what dicks coronaviruses are for the pork industry and how they tend to provoke really lackluster immune responses in general, which makes them a pain in the ass to vaccinate against. Knowing that a case of COVID gives you a one in five chance of getting COVID again two months from now, a one in two chance four months from now and all your immunity has dissipated six months from now is data. It's shitty, dispiriting, expensive, society-changing data but it's data.

I'd like to see that data, or at least people researching that data. I'm sure the studies are being done. I doubt the results will be heartening, and I sincerely hope they're more robust than this.