Daily numbers, graphs, analysis thread 13

[PatriciaHolm]

Taking the liberty of starting a new thread as we've just bust the old one, with much thanks to @BigChocfrenzy and I will copy her header..

Welcome to thread 13 of the daily updates

Resource links:

Slides & data UK govt pressers
UK dashboard sub-national data, local authorities
Beta Uk dashboard deaths, cases, hospitals, tests, partially sub-national
UK stats updated daily by PHE & DHSC
ONS UK statistics for CV related deaths, released weekly each Tuesday
PHE surveillance report infections & deaths released every Thursday with sep. infographic
NHS England stats including breakdown by Hospital Trust
FT Daily updates
HSJ Healthcare updates
Worldometer UK page
Plot FT graphs compare countries deaths, cases / million pop. / log / linear
Covidly.com filter graphs compare countries
Plot COVID Graphs Our World in Data

We welcome factual, data driven, and civil discussions from all contributors 📈📶👍

There is one being trialled @Reastie. Fingers crossed something actually comes of it although it’s all gone a bit quiet and the trial would be finishing around now....
www.theguardian.com/world/2020/may/21/uk-coronavirus-test-with-20-minute-wait-being-trialled

Also, the trialling of the saliva tests could also be a massive help, as it sounds like they intend to use that to do large scale surveillance testing....
www.gov.uk/government/news/new-saliva-test-for-coronavirus-piloted-in-southampton

Let’s keep everything crossed something useful comes from these!

@MarcelineMissourinthwt first link is for an antibody test I think which is different I believe from a test to see if you have Covid at that time. I saw saliva tests being trialled which is great news, I’m not sure how Ds (1) would get on with w swab test, but not sure how long the results take for that.

Wondering if the clever posters on this thread can help answer something which has been bothering me and which I saw mentioned in the Telegraph today. Is counting case numbers (positive test numbers) a pointless exercise? I know we need to know who is carrying the virus asymptomatically so they can isolate, but is it not incorrect to be referring to them as a case for statistical purposes? They have no illness, they have only been exposed to the virus and their immune system is fighting it off. So essentially we are counting "immune" people in our case numbers and this is distorting the picture? As time goes on, would those who have antibodies and a 2nd exposure, show a positive PCR test even if they immediately fight thr virus and have no symptoms? It just seems strange to me, in the history of pandemics cases have always been "ill people" have they not? You wouldn't have counted asymptomatic measles or polio. Apologies if I've misunderstood, I just find this area hard to understand.

@Reastie no it’s definitely talking about an actual 20m test for Covid, it’s just also talking about antibody tests as well x

That is my current understanding of the isolation procedure bigchoc. For secondary students I suspect there might be the expectation that they will get themselves to school. For primary, I'm assuming schools will understand if children are off for the full two weeks but this might not always be the case if there is still pressure on attendance figures. In our case we have no family locally and I'm not sure I would be comfortable asking a friend to do drop off/pick up for a whole week (esp given the rest of us might be harbouring covid, although we could kick ds onto the street to wait it's a big ask). I think in reality the 7 day vs 14 day is only useful if your index case is old enough to go out by themselves.

(I have seen threads on here about schools only allowing immediate families to do the school run so presumably those schools will expect all children to be off for 14 days, our school is more flexible)

Mummabeary Pandemic cases or epidemic cases would be any positive test, not just those with symptoms.
Or it could even be just a diagnosis without a test
So COVID is not being over-counted

In fact, when calculating cases of a flu pandemic after it has finished, several countries use maths modelling and the FLUMOMO algorithm to estimate the cases
e.g.
For England:

https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachmentdata/file/839350/SurveillanceeofinfluenzaaandotherrrespiratoryvirusessintheeUK20188to_2019-FINAL.pdf

Page 24 lists flu deaths as a few hundred annually over the last few years
and then
Page 51 lists estimated deaths for those same years as up to nearly 30,000 via the FLUMOMO algorithm

So neither cases nor deaths are being over-counted
Requiring a positive test is being quite strict

Thanks @BigChocFrenzy
That makes sense. I wish there was a way where we could know from the daily case numbers how many are people who feel poorly and how many were tracked/traced and took a test, but otherwise would never have known they had it. It feels like quite an important distinction when we are making such big decisions based on case numbers but I can see why it wouldn't be possible (pre-symptomatics etc).
Thanks for all your posts over this time - I have found you and other posters on here so informative these past few months.

Mumma it gives a more accurate picture of the prevalence of infection within the population.

That's helpful for many reasons. To even find out how many may be immune/ asymptomatic, what their features are, how much risk there is to the local community as asymptomatic people can still pass it on.

On of the biggest mistakes we made was not doing mass testing early on.

Reastie and choc, that's why teachers are concerned.

Especially teachers who have their own children. We really need the incidence level in the population to be very low for school to work.

Staff absence is going to be high if we ourselves are at home with a child for 2 weeks. Not to mention if we then catch it ourselves. If negative, there's still taking a day or so to actually get the test and wait. With normal coughs and colds that process could be quite frequent.

I was told via a union rep that there are discussions and plans around either testing within schools occurring or mobile testing units who can be called to come and test on sight for any pupil or staff member, and obviously a much faster testing process will help enormously. (See, the unions are actually working to support the crisis.)

Whether that can be organised by sept or mid August is another matter of course.

Sorry Reastie, I hadn't noticed you're a teacher. I am too and CV; it's hard thinking about sept. I'm not allowing myself to at the mo as so much can happen between now and then.

One of the biggest issues for me regarding U.K. schools is the class sizes. Plus some school buildings are like sardines cans, built for smaller classes.

Here, Roughly 30, even as high as 34.

Even Australia caps at 25 I think, usually less. Most of the Eu ranges from 15 (!) to 20/25 max for older children.

Our numbers need to be lower for this to work.

Can someone shed some light on the figures on the Zoe App vs the stated rates from PHE. For example Cotswolds is showing as dark red (1,500 per million) whereas PHE figures suggests less than 5/100,000 thats a factor of 10 out. Should we believe the Zoe App which would suggest still quite high levels(and rising) of community transmission or the offical figures ?

@ mumma In France the weekly round up gives the % of asymptomatic tests - last weeks was 54% which is down from previous weeks.This was noted as important as a bad sign.

@Humphriescushion

@ mumma In France the weekly round up gives the % of asymptomatic tests - last weeks was 54% which is down from previous weeks.This was noted as important as a bad sign.

.... Isn't that pre-symptomatic as well as asymptomatic ? Otherwise they'd have to update the numbers for each day for a week or more

Oh will have a look.think it just says asymptomatic.

Germany - Date of onset of symptoms vs date of reporting:

Chart with about 31% of cases having "unknown" date of symptoms,
which would presumably also include those who never had symptoms

Symptoms:
cough (48%)
fever (40%)
rhinorrhoea (21%)
Pneumonia (3.0%)
loss of taste or smell (15%)

Treatment:
Hospitalisation (17%)
ICU (2%)

Hospitalisation levels are now much lower due to much lower age of infection

@Humphriescushion

Oh will have a look.think it just says asymptomatic.

I think it would be "asymptomatic at the time " which would include pre-symptomatic because otherwise with having to wait for symptoms to develop, they would only be reporting the status up to 14 days in the past

It must be so valuable to have published/publicised figures for %asymptomatic, as it drives home the message that you can have Covid-19, be merrily going out and possibly infecting your colleagues, friends, maybe also those who are more vulnerable, and not have a clue!

Yes that what I would assume. I am trying to paste the graph but it is not happening.

Not sure i can understand @ bigchoc. I presumed it meant you had a test, you tested positive but you had no symptons? CHart i cant link has data for asymptomatic, and also those that had symtpons in the run up to the test. I understood this would not be unusal since iI think that younger people are testing positive at the moment.

I too don’t understand the Zoe app. I’ve asked on here before too which is a guy embarrassing as someone did explain it, I just really don’t get how my area goes so up and down daily. The area next to me was 0 then 65 then 0 again the best day...is it literally anyone who reports one of the 19 symptoms is then classed as symptomatic?

@cusano The figures from the Zoe app are modelled. So it's not that's there's 65 people one day and none the next - more likely one or two people one day and none the next, but they estimate the total number based on their small number of reported cases. I'm dubious about their figures at local level as I don't see how your can estimate those from a modelling approach when we know from the testing figures that the total numbers are so low. Plus their model looks at the number of people with any symptoms and multiplies that by the proportion of people with symptoms who test positive - they know this proportion nationally because they ask their app users to get tests. But I can't see how they can be doing that at a local scale - we know that other colds, hay fever, etc will come and go across different areas, they're not constant. So the actual proportion of people with symptoms who in fact have Covid will vary between places from one week to the next.

Thought folks on this thread might find this article interesting:

www.bbc.co.uk/news/world-51235105

Think I understand now i have engaged brain. Of course they have no symptons at the time of the test but they can go onto to get symptons later. Was a fairly interesting statistic to follow though since it came down from 64 percent.

Reastie and choc, that's why teachers are concerned.

Especially teachers who have their own children. We really need the incidence level in the population to be very low for school to work.

Staff absence is going to be high if we ourselves are at home with a child for 2 weeks. Not to mention if we then catch it ourselves. If negative, there's still taking a day or so to actually get the test and wait. With normal coughs and colds that process could be quite frequent.

I was told via a union rep that there are discussions and plans around either testing within schools occurring or mobile testing units who can be called to come and test on sight for any pupil or staff member, and obviously a much faster testing process will help enormously. (See, the unions are actually working to support the crisis.)

I wonder what would be classed as a low level in the community?

Interesting re school plans. Presumably that would only be for people coming down with symptoms whilst in school rather than those staying off as soon as they realise they have symptoms?