Chat about Novel Coronavirus - 2019-nCoV - COVID-19

Nice one. The only place I could be exposed would be a pub or restaurant which will have my details anyway.

1 Attachment

Are they releasing the source code for this one? I thought that was for the government driven version, whereas this is the Apple/Google-data-handled backend version.

Yes. The Bluetooth beacon bit is the Apple/Google framework but there is still plenty of code in the app and a backend system to support it. Some bits of the app have nothing to do with proximity, eg the venue checkin.

System & architecture https://github.com/nhsx/covid19-app-system-public
Android https://github.com/nhsx/covid-19-app-android-ag-public
iOS https://github.com/nhsx/covid-19-app-ios-ag-public

There was a commit about an hour ago, maybe it is up to date now? Probably impossible to tell.

That's good - I stopped paying attention once it went private and assumed the codebase would follow. Glad they've kept some level of it open.

I'm a school governor, and we've just had some guidance come through from the council: if someone has the 'classic symptoms' of Covid (high temp, continuous cough, lack of smell), they should self-isolate for 10 days (and household for 14 days), even if they get a negative test result. Their rationale seems to be that there is a 1 in 5 chance you're still positive even if the test comes back negative. I thought the false negative rate was much lower than that? Or is this one of those counter-intuitive Bayesian result things?

I think if there's a decent amount of virus in the sample the false negative rate is pretty low. But that assumes the sample was taken correctly, at the right time in the virus life cycle, you didn't wee on the swab instead of sticking up your nose, etc.

I'd view 4 out of 5 working out as being pretty optimistic tbh.

(I was quite surprised the text I got from the NHS said "You tested negative. Get back to work." with zero caveats)

The problem is that anosmia/parosmia isn't exclusive to Covid-19, just google post viral olfactory loss.

I'd challenge the idea that you should continue to self-isolate even with a negative test result on those grounds alone. (Edited for clarity, see responses below.)

even if they get a negative test result.

That echo government advice on being exposed to someone who’s positive.

Their rationale seems to be that there is a 1 in 5 chance you're still positive even if the test comes back negative. I thought the false negative rate was much lower than that? Or is this one of those counter-intuitive Bayesian result things?

The figures look off anyway, and the counter-intuitive bayesian result thing concerns positive test results which is where I think they're getting confused.

The tests are supposed to have roughly 98% sensitivity and 98% selectivity.

Prevalence is currently estimated at 8%.

So the chances of a negative test result are:-

True negative:- 92% * 98% = 90.16%
False negative:- 8% * 2% = 0.16%

So with these figures a negative test is likely to be correct > 99.8% of the time.

The problem is with positive results, doing those sums:-
True positive:- 8% * 98% = 7.84%
False positive:- 92% * 2% = 1.84%

So a positive result is likely only to be correct 81% of the time. There's your 1 in 5, they've just applied it to the wrong thing, or assumed that the 1 in 5 change of an incorrect positive test also applies to negative tests.

[EDIT] The above figures will be wrong as the 8% prevalence assumption is way out. That's the ONS's figure for how many people have HAD the virus, which is not the same as the percentage of the population who currently have it. But it doesn't change my argument much as using a lower prevalence value only increases the odds that the negative test is actually correct.

Indeed, the guidance ( https://www.nhs.uk/conditions/coronavirus-covid-19/testing-and-tracing/what-your-test-result-means/ ) says:-

"
Negative test result

A negative result means the test did not find coronavirus.

You do not need to self-isolate if your test is negative, as long as:

• everyone you live with who has symptoms tests negative
  
• everyone in your support bubble who has symptoms tests negative
  
• you were not told to self-isolate for 14 days by NHS Test and Trace – if you were, see what to do if you've been told you've been in contact with someone who has coronavirus: https://www.nhs.uk/conditions/coronavirus-covid-19/testing-and-tracing/nhs-test-and-trace-if-youve-been-in-contact-with-a-person-who-has-coronavirus/
  
• you feel well – if you feel unwell, stay at home until you’re feeling better
  
  

If you have diarrhoea or you’re being sick, stay at home until 48 hours after they've stopped.
"

And that second link has:-

"
What to do when you get your test result

If you test negative (the test did not find coronavirus):

• keep self-isolating for the rest of the 14 days from when you were last in contact with the person who has coronavirus – as you could get symptoms after being tested
  
• anyone you live with can stop self-isolating if they do not have symptoms
  
• anyone in your support bubble can stop self-isolating if they do not have symptoms
  "
  
  

But the justification for doing it because of a supposed 1 in 5 chance of the negative result being wrong is just bollocks.

Had an email from school last night having a rant because kids had been sent to school despite having persistent cough, temperature and anosmia. Beggars belief. Confirmed cases in years 8, 10 and 12 so far.
Child 1 is a freshers rep at uni - the freshers start arriving this weekend. At least he's living off-campus so slightly away from the petri dish, but given the noises coming out of the unis already back it's going to be bad.
Feel very sorry for this year's intake - bugger all socialising and having to do lectures online means being cooped up in your room when you should be going out and meeting people. Will be a really shit experience.

Where there's a will(y) theres a way

1 Attachment

Hmm, thanks, I may push back on this then. Bit worrying that I seem to have a better grasp of this (at least inasmuch as I can say 'that doesn't look quite right...') than our council's public health team.

Time to unleash the hounds?
https://www.theguardian.com/world/2020/sep/24/close-to-100-accuracy-airport-enlists-sniffer-dogs-to-test-for-covid-19

How many sniffer dogs will £100bn buy?

https://www.bbc.co.uk/news/uk-scotland-54268780

The total number of reported cases since the start of term is 124 but the actual number is likely to be higher.

Less than two weeks since the start of freshers week...

From Serco and Deloitte, about 20.

It seems like the app doesn’t work very well either, The Times reporting one in every three positives are false

https://www.thetimes.co.uk/article/download-coronavirus-tracing-app-to-protect-your-families-public-told-sjqr5wqk8

Freshers week here in Manchester now, I expect it will be the same here in 2 weeks.

Yeah. Not going to be good. Seriously wonder if we'll see him for Christmas.

Yep. Nephew is has just started at MMU and not happy with his freshers experience.

In what sense? Too much bodily contact or not enough?

Must be really grim. No chance to get out and meet all the people who can become lifelong friends. Stuck in a room watching lectures on a laptop.
Not what your first weeks of uni are meant to be like.

Must be really grim. No chance to get out and meet all the people who can become lifelong friends. Stuck in a room watching lectures on a laptop.
Not what your first weeks of uni are meant to be like.

Have the people studying Comp Sci noticed?

https://twitter.com/SkyNews/status/1308655561081225217

Rev Thomas Bayes to a white courtesy telephone please...

Is it too cynical to think that what will end up happening with the new jobs scheme is that workers (who are kept) will end up working at/over the 65% of their hours that is paid by their employer, rather than the 33% that is the government share?