Daily numbers, graphs, analysis thread 2

[Barracker]

A follow on thread from here

Please try to keep it data driven, factual and civil.

www.worldometers.info/coronavirus/country/uk/

After seeing the statistics from Great Ormond Street where 73/181 tested staff were positive (unclear to me how the 181, a fraction of the total number of staff, were selected), I've started to wonder:

Is the government not testing HCPs because the virus is clearly so rampant among them (perhaps with many mild/ asymptomatic cases) that the NHS would collapse immediately if all of the CV positive people had to stay off work?

The Government briefings suggest they have ask the NHS to prioritise critical staff who may have the illness or have a family member who may have the illness. So you would expect more positive results.

DuLANG Looks like the UK may be heading for the same sort of grim situation in care homes

BBC Newsnight: "as care homes face staffing and equipment shortages, how can they protect their vulnerable residents from coronavirus?"

and Lewis Goodall's long sad thread on it:

https://mobile.twitter.com/lewis_goodall/status/1245830505146322944

Wow. That mumsnetter account is scary esp if we aren't near the worst. :( why arent all thise deaths included?

Barracker tell her to take vit C and D and try r
to get as much sleep as sge can to keep her immunity up. On tv yesterday Sky news a person in charge of a hospital trust said they had 250 either off sick or isolating as someone in their house was unwell. They gof tested and only 17 were positive so he was ordering thecrest backnto work. I missed beginning of interview so don't know which trust or how they got them all tested as but he said something about a cap being lifted.

Gosh poor people at Kings, sounds so hard. Worrying about what’s not being reported though. Is it an overrun system causing the difference. I feel for all the people there.

Interesting thread everyone, thanks.

At the trust I work for there are currently over 900 members of staff sick/self isolating/special leave.

Also, for info our death figures had not yet been fed into national totals due to a misunderstanding about official reporting lines for this data. These will have all gone in at once over the last couple of days, or will do over the next couple (not sure exactly when as I've been off for a few days).

In my NHS trust, very few are being tested - only critical workers in small clinical teams which are deemed essential. And only in the first 72hrs of presenting symptoms of covid.

So your average nurse, doctors or care worker in a hospital setting is not being tested.

It's not being used as a screening process.

Not good enough!

Morning Barracker, just catching up on the thread this morning and saw your 2359 post. I may be overthinking this but if I were you I would ask MNHQ to remove it.
The person who PMed you PMed rather than post on the thread to avoid being identified and getting her daughter disciplined at work.
But by putting the full PM on there it makes her poor DD quite identifiable.

" Presumably the shortfall of 67 missing cases will be swept into tomorrow's figures."

I think there was a specific issue with 40 cases in Scotland not having been reported previously - there was an explanation in the Guardian which I now can't find, but it sounded as though it related to some earlier cases. I didn't get the impression that the intention was to add those 40 to today's figures, but, once the facts had been established, to 'allocate' them the correct day in previous figures ie to revise previous figures.

Agree that as the basis of testing and reporting deaths changes it will be difficult to know what the trends are, unless a 'like for like' report is also done, continuing to report on the current basis as well.

How accurate are the test? Just asking as my elderly aunt died from CV two weeks ago ( in hospital, tested positive) Three of my cousins (who cared for her) came down with symptoms. The one who was her main career ended up in hospital for 2 nights with breathing difficulties, they tested her and it came back negative! She was the one getting her out of bed, doing all personal care etc and living with her 4 days a week. I really can’t believe she didn’t have it.

ow accurate are the test

That's a good question.
I used to do RT-PCR tests - a long time ago - but I haven't seen any data looking at the sensitivity of the tests.

These are the words to look for:

Sensitivity - say it was 99% - then 99 people out of 100 with the disease would test positive.

Specificity - 99% - 99 people out of 100 who were negative would test negative and 1 would be positive.

I am looking for accurate data on these.

Don't know if it's been linked already but this is a really useful resource from Google for stat lovers. Google uses their maps and phone data to show how much people's movements have reduced across a range of settings including retail, parks, transport stations, groceries and pharmacies etc. Data is broken down by country and then by council area for the UK.

You can see which parts of the country are obeying the lockdown and which seem to be flaunting it

www.google.com/covid19/mobility/

I posted before re the tests that scientist in France have said that positive results are accurate, i.e. if you test positive you definitely have it, but that negative results are only 60-70% accurate, so you could test negative and still be infected. I’ll try to find the French report again.

@coronade

There was a mention, perhaps upthread or poss a different thread, by a pp who had seen Drs tweeting that there can be a discrepancy where nasal and throat swabs are negative but sputum tests positive. Perhaps your cousin didn't have sputum test? If this is a widespread issue though it throws up all sorts of difficulties around the testing.
I am so sorry for your loss

There was a mention, perhaps upthread or poss a different thread, by a pp who had seen Drs tweeting that there can be a discrepancy where nasal and throat swabs are negative but sputum tests positive

Yes - the gold standard is the sputum test. I have seen just one set of figures on a small sample - I think out of 20 positive sputum tests, then 19 were swab positive and 1 was swab negative.

Because RNA is a very delicate molecule

There’s one here, in French though sorry www.ouest-france.fr/sante/virus/coronavirus/entretien-la-plupart-des-tests-covid-19-actuels-ne-sont-fiables-qu-70-6794022

And in this one a different virologist explains the accuracy with a hypothetical group of symptomatic patients: nasal swab would give a 40% positive finding, but deeper swabs (trachea or bronchus) would give 80% positive. So nasal swabs would miss half.
www.franceinter.fr/sciences/coronavirus-les-tests-sont-ils-fiables

Ha, cross post with Ink

the NHS would collapse immediately if all of the CV positive people had to stay off work?

I have to admit the same thought has crossed my mind recently.

Barracker, I agree with LivinLaVida that the post about the nurse at King's could identify her.

d in this one a different virologist explains the accuracy with a hypothetical group of symptomatic patients: nasal swab would give a 40% positive finding, but deeper swabs (trachea or bronchus) would give 80% positive. So nasal swabs would miss half

Symptomatic isn't the same as having the disease.

When I did lots of routine blood tests, it was not unusual for many people with a suspected disease to come back as 'within the reference range'

The key is knowing how many confirmed cases (ie sputum positive) does the swab test pick up?

You always compare a test's accuracy to the gold standard.

Hammond ,in yesterdays briefing ,said

Prioritisation ( for tests )has to be the patients for whom the result of a test could be the difference between life and death

does anyone else wonder about this statement? Is there special life saving treatment for a Covid patient ?Earlier intervention with intubation/ventilation ?

Or is the importance in detecting and trying to prevent that patient from infecting non Covid patients ? Though presumably patients with any serious Covid symptoms are treated seperately ?

Peridito - there is no special treatment for covid patients (other than isolation, which they would have anyway if they were a suspected case).

The management is all supportive - oxygen, ventilation if that isn't enough, fluids. Same would go for any bad pneumonia

Ive posted this elsewhere. Interesting article on how many Excess deaths we are saving in each scenario, balanced against effects on the population of lockdown. Something we will only know for sure in the final analysis.

www.bbc.co.uk/news/health-51979654

Pocketem ,that's what I thought .So it seems a rather odd justification for prioritising that group to have the test .

Hasn't Scotland decided to report deaths suspected of involving Covid ? Maybe to free up tests ?