And by "case" you mean detected matter with PCR cycle counts undocumented at aggregate levels. But don't let science get in the way of emotional appeals.
That said, I have some doubts about the 2% number. Estimates vary widely from about 2% as the lowest plausible number I've seen to about 10% at the highest plausible number. If it's really 10% of infected people reporting long-term symptoms, that changes things a bit. Also note that pretty much all of the sources are preprints or otherwise not peer reviewed articles -- an actual rigorous investigation is very welcome here, and hopefully that is what the $1B from NIH will buy.
Uh, what? The WHO updated their guidelines a month and a half ago because PCR tests were being misused: Too many cycles and not taking symptoms into account was resulting in too many false positives.
https://www.reuters.com/article/uk-factcheck-who-instruction...
This is right-wing conspiracy theory.
>the tests were not being used in compliance with instructions
This is science.
The part that primarily caught my eye, which isn't what Reuters is attempting to debunk:
> Where test results do not correspond with the clinical presentation, a new specimen should be taken and retested using the same or different NAT technology.
"clinical presentation" - in other words, the symptoms. This advice runs contrary to the entire past year, where a single positive test in an asymptomatic person has counted as a case. This alone is enough to inflate the numbers, without even taking PCR cycles into account.
And Reuters ends up just barely short of confirming the part about PCR cycles anyway:
> ”The WHO are saying that if the manufacturer has defined a value but you could if you chose to adjust that setting, please don’t - stick to what the manufacturer has stated because they have done the earlier work to determine the best value for that threshold,” Mackay said.
Labs using too high a threshold has been a concern for about half a year: https://www.nytimes.com/2020/08/29/health/coronavirus-testin... (And note the numbers in this article, almost all in the range of 33-40 cycles - I recall on rare occasion seeing higher reported elsewhere, around 45 and 50 cycles)