PCR false negatives... shocked!

[Warhertisuff]

The mantra in here regarding tests is generally:

"LFTs are a bit shit but I suppose they're better than nothing if testing someone who's assymptomatic, but PCRs are the gold standard and holy grail of testing, and can pretty much be guaranteed to be accurate."

www.bmj.com/content/372/bmj.n287/rr

However, this apparently piece in the BMJ indicates a false negative rate of 30%! This is on a par with LFTs. That feels like a worryingly high figure in terms of mass testing being a way of stopping transmission. However, I'm guessing that a significant chunk of the 30% are those with low viral loads and with no or very low-level symptoms and who therefore aren't especially infectious anyway.

It does seem to mean that the cumulative effect of a week of daily LFTs after exposure is far better at detecting disease than a solitary PCR a few days in.

A key benefit to PCRs over LFTs seems to be that there is an extremely low false positive rate with a PCR.

So a positive LFT may be false - unlikely but possible. A positive PCR is extremely unlikely to give a false positive unless you have cross-contamination.

Oh wow! That is really useful thank you. I know a family who are insistent that because one of their pcrs was negative that person doesn't have it..even though they have symptoms and everyone else has it.

I always knew this but I have a degree in maths & statistics and it was one of the first things I looked up when testing became widespread, I'm entirely unsurprised that the majority don't know.

I’d read this a while back I think the real benefit for the government of the pcr is tracking variants and contact tracing.

Why do they say if you have a negative PCR you can just go and do what you want then? No wonder it's spreading.

Because despite the issues, it's still the best we have and we need to keep the economy and national life going @CheesyFootballsAreEvil

@CheesyFootballsAreEvil

Why do they say if you have a negative PCR you can just go and do what you want then? No wonder it's spreading.

Also, if your Covid infection isn't sufficiently strong to register a positive PCR, you're less likely to be infectious. It's a numbers game... as with vaccines, testing helps considerably but it isn't guaranteed. People understandably don't like that. They want tests that work every time and vaccines that give complete protection.

@ColettesEarrings

I always knew this but I have a degree in maths & statistics and it was one of the first things I looked up when testing became widespread, I'm entirely unsurprised that the majority don't know.

I'm a bit of Covid nerd, have a maths background and went out of my way to research for this... so if I didn't know you can bet the vast majority don't know either.

Thanks for this. It ties in with my testing experience.

@Warhertisuff

A key benefit to PCRs over LFTs seems to be that there is an extremely low false positive rate with a PCR.

So a positive LFT may be false - unlikely but possible. A positive PCR is extremely unlikely to give a false positive unless you have cross-contamination.

Do you have a source for false positive LF? My understanding (read a lot when the immensa thing kicked off) was that false positive LF are very rare assuming test done properly according to instructions. They are tested for cross reactivity with other viruses eg colds.

The other benefit to LF is that they detect someone likely to be contagious. A PCR done on me right now could well be positive, because I had covid not too long ago. However, I am not contagious and my LF have been clearly negative since day 11.
But a PCR would have me pointlessly isolating for 10 days.