Here’s a very interesting article from the New York Times in 2007
It’s the story of what happens when you believe unquestioningly in the results of a test which has a high false positive rate – in this case, using a PCR test (which, when correctly designed and implemented, are very good for distinguishing between similar diseases) for diagnosis (which is a very different task). I’m sure bells are starting to ring, but in case not, note that the UK COVID-19 response has so far regarded PCR tests (along the lines of those specified by the WHO) as the “gold standard” test for SARS-COV2.
I’ve commented in the past to anyone who would listen about the document from the WHO website which seems to recommend 50 amplification (doubling) cycles, leading to a potential amplification factor of more than 10^15 (1 followed by 15 zeroes). This truly fantastic number is widely understood to lead to very large false positive rates which render the test completely useless for most purposes, and certainly useless for diagnosing individuals in populations with a low probability of occurrence. If the false positive rate of your test is much larger than the incidence rate, the majority of positive test results will be false ones. You cannot determine the incidence rate from that data. It’s like trying to detect very quiet noises through a hearing aid which hisses. The more you turn up the volume (amplification cycles) the more hiss you hear, and the less chance you have of knowing what’s underneath. That failing is almost certainly the case for the current range of SARS-COV2 PCR tests. Here’s another view (from Canada):
Even the estimable Dr. Fauci says so in this interview clip.
So a lot of informed people seem to put the sensible limit somewhere around 30-35 cycles.
PHE, on the other hand, puts a limit at 40 cycles in their SOP document – and then says if you needed more than 40 cycles to get a positive result, you should test again! The reluctance to accept the validity of a negative test result from their “gold standard” test is – erm, a bit peculiar, don’t you think?
Anyway, it turns out there’s much more that’s wrong with the WHO’s PCR test. Here’s a recent article about it.
All in all, it does seem that the SARS-COV2 PCR test is a very flawed thing. However, it is still, undeniably, the best thing available on a wide scale – and thus, for politicians (and other people who have to decide on policy), that’s what they will use. It doesn’t really matter if it’s not fit for purpose – it’s the best they’ve got. And then they believe the statistics that flood forth, and imagine that every positive test has detected a super-spreader in waiting, and impose regional and national lockdowns, to avoid a catastrophe they won’t be able to explain their way out of. In all probability, they actually think they are doing the right thing. After all, the UK government started out trying to let the whatever-it-was just “run through the population” and that was a dismal failure, so what else can they do now?
The same sort of thinking probably explains the WHO’s continued support of their “test, test, test” policy. It’s just the safest option from their point of view. As long as they aren’t responsible for any of the collateral damage.
Incidentally, the much-vaunted new lateral flow tests, while not suffering from exactly the same problems, have their own issues. They are neither accurate enough nor repeatable enough, and thus don’t really have much to say about the likelihood of any particular individual developing Covid-19, which is what we actually care about.
Anyway, I think the take-away from all this is that in all probability, there’s actually a lot less SARS-COV2 about than we are being led to believe. So that’s good, isn’t it?