[acqq]

Based on what are you so certain when you say: "I am pretty certain that reinfections are very rare"? I read the AstraZeneca/Oxford vaccine study, which was as serious as it can be (even the errors in the procedures were discovered and honestly reported):

https://www.thelancet.com/journals/lancet/article/PIIS0140-6...

From all people who were tested in the study (a lot, order of 40000) around 2 percent were infected during the study. Remember that as the base infection rate during the study.

At the same time, around 370 were PCR positive at the start of the study. But during the study, of these 370, 3 were reinfected. It's not far from 1% of those. Now, as the infection rate of those not positive at the start (which we can consider "control") was around 2%, it seems that the chance to be reinfected could be as high as 50% during just the months the study was ongoing!

Please don't consider the estimated numbers as too exact, I don't believe there are much significant digits for conclusion, but I believe they represent the right order of magnitude when based on that many subjects and the randomness of the process, and the only conclusion that could follow is: naturally infected could be only weakly protected from the reinfection only months later. The quote from the study:

"A small proportion of participants were seropositive at baseline (138 [1·3%] of 10 673 in the UK and 235 [2·3%] of 10 002 in Brazil). Three participants seropositive at baseline had subsequent NAAT-positive swabs. One participant had an asymptomatic infection 3 weeks after a first dose of ChAdOx1 nCoV-19. Two other participants in the control group had symptomatic infections 8 weeks and 21 weeks after their baseline sample was taken."

That's for me a kind-of proof that the reinfections aren't rare at all. Even more interesting, 2 of 3 were symptomatic, and the reinfections happened quite soon.

The reason it's not more known is only because most of reinfections just aren't tracked in careful studies like this one. Initially, the cases of reinfections weren't dismissed for lacking the "definitive" proof. An sure, most of the tests are just PCR, but there were a few studies where the sequencing was performed (much rarer and more expensive and time consuming procedure) and where it was possible to prove that the detected virus in the reinfection didn't have the same genetic fingerprint as the initial one.

https://www.thelancet.com/journals/laninf/article/PIIS1473-3...

[jsnell]

You claimed: ... around 370 were PCR positive ...

Then you quoted from the paper: ... small proportion of participants were seropositive ...

Which is in direct conflict. The latter is saying they had positive antibody tests. Not PCR tests. In fact a lot of those were most likely just false positives, from a antibody test that "only" had a 99% specificity. Which is most of them.

[acqq]

No conflict: The infection rate is 2% from those starting not infected. The infection rate is 1% for those who started as infected and later tested positive again, after more weeks. That's only twice less common than normal infection, which is a lot.

The quote you requoted only partially lacked the end: "at baseline": "A small proportion of participants were seropositive at baseline" which doesn't say anything about the reinfection.

It's also surely not antibody test, NAAT mentioned is "nucleic acid amplification test" a test that tests for the presence of virus' RNA, not antibodies. Also: false positive while symptomatic is significantly less probable than just the probability of false positive.

If your point is that you don't believe they were seropositive at all at baseline, fine, that's can be said, it would be possible, and that is surely not 100% sure.

The point is still, there's no proof that the reinfections are "rare." And from what is known about another human coronaviruses reinfections do occur each year.

Edit, answer to below: yes, I tried to understand what you say, the last two paragraphs I wrote after I wrote the start. The NAAT is the only test procedure mentioned in the same quote where you took only the first few words. I do admit wrongly remembering the nature of the test at the baseline. It has sense as it would be less probable to have somebody participating in the study at the time when he should be in isolation.

[jsnell]

I don't think you understand. You claimed 370 people were PCR positive at the start of the test. Then you quoted from the paper results for seropositivity. That quote is clearly the source for your claim, since the numbers add up to 371. But seropositivity is not the same thing as PCR positivity.

And I don't understand why you're pulling in the NAAT test from that quote. It is how the three people tested positive later, after being infected. It is not about the 370 people who were seropositive at the baseline.

[lumberingjack]

This is why we need experts that are truthful everyone reading all of these reports just gets their little heads confused. Only problem is everyone's got an ax to grind and everyone's biased. I'd rather talk about facts on the ground. I go to the hospital multiple times a week the whole place is shut down with lights turned off most of the nurses are laid off that I used to know that worked there. They keep saying that the hospital is full because of covid patients it's not true the covid wing is full 30 beds and a hospital designed for an entire city.