Daily numbers, graphs, analysis thread 21

[BigChocFrenzy]

Welcome to thread 21 of the daily updates

Resource links:

Uk dashboard deaths, cases, hospitals, tests - 4 nations, English regions & LAs
R estimates UK & English regions
Imperial UK weekly LAs, cases / 100k, table, map, hotspots
UK School statistics Attendance
Modelling real number of UK infections February to date
NHS England Hospital activity
MSAO Map of English cases
Cases Tracker England Local Government
ONS MSAO Map English deaths
CovidMessenger live update by council district in England
Scot gov Daily data
Scotland TravellingTabby LAs, care homes, hospitals, tests, t&t
PH Wales LAs, tests, ONS deaths
NI Dashboard
Zoe Uk data
UK govt pressers Slides & data
ICNRC Intensive Care National Audit & Research reports
NHS t&t England & UK testing Weekly stats
PHE Surveillance reports & LA Local Watchlist Maps by LSOA
ONS England infection surveillance report each Friday
Datasets for ONS surveillance reports
ONS Roundup deaths, infections & economic reports
ECDC rolling 14-day incidence EEA & UK
Worldometer UK page
Our World in Data GB test positivity etc, DIY country graphs
FT DIY graphs compare deaths, cases, raw / million pop
Alama Personal COVID risk assessment
Local Mobility Reports for countries
UK Highstreet Tracker for cities & large towns Footfall, spend index, workers, visitors, economic recovery

⏭ Our STUDIES Corner ⏮

We welcome factual, data driven and analytical contributions
Please try to keep discussion focused on these
📈 📉 📊 👍

[quote Frazzled2207]@BigChocFrenzy
is that black line a sign that there was massive demand for tests when schools went back and demand has since decreased? Good news I suppose if so and we would see a similar pattern in england albeit later. Or does it mean tests have become more scarce?
Even if number of positives is increasing if there is less overall demand for tests then surely that is good?[/quote]
....
The black line just indicates tests performed and I'm trying to establish why the dropoff

It is worth doing, even though fomite transmission may only be 10-20% of cases - especially as it may be a higher % for children

Yes, in theory and for other reasons than coronavirus. However, as I was just picking up DD I saw them all crammed together (literally touching) and talking and chatting as they waited for the two sinks. The whole class was waiting to wash hands at the end of the day. I was waiting outside for 10 mins for DD to come out so she was in that crush for at least that long. I suspect that for sars-cov-2 the increased risks of everyone in a huge huddle chatting and laughing might outweigh the benefit of the handwashing.

ok - looking at the timing just after schools went back to school I would hope it was demand, in keeping with kids all having normal coughs and colds shortly after going back to school.

% positivity increasing is always concerning (above 3%, certainly) and usually indicates that indections are increasing and more tests are needed, if possible, for track & trace

Otherwise, too many new cases are being missed and not going into isolation, hence transmitting further infections

I simply don't believe Scotland's test numbers now, see here NS is very clear to say "newly tested": www.gov.scot/publications/coronavirus-covid-19-update-first-ministers-speech-29-september-2020/. It's a "pile of pish". They shouldn't include confirmation tests in the stats, but the same worker getting a test in July and now should certainly be counted as a test.

Actually she's reporting half the.number of tests assigned to Scotland on the PHE coronavirus board. 11.5% would mean 7008 tests. Now look: www.gov.scot/publications/coronavirus-covid-19-update-first-ministers-speech-29-september-2020/.

Bloody nonsense.

Sorry, I meant to put the dashboard link: coronavirus.data.gov.uk/testing?areaType=nation&areaName=Scotland

John Burn-Murdoch@jburnmurdoch (FT stats geek)

.... how to interpret figures from different sources, and what caveats each source does and does not come with:

• Pillar 2 community testing:

these are the bulk of cases picked up at the moment.
Case and positivity rates here could be influenced by where and who is being tested,
so e.g patterns in this data with age, deprivation etc could be skewed by who is getting tested

• @ONSS^ infection survey:

these tests are random, and designed to be representative of the overall population.

Therefore trends and patterns in this data are not due to e.g certain locations or groups of people being more likely to get tested.

Samples taken for *@ONSNS^ tests are re-tested multiple times to make false positives extremely unlikely (one in tens of thousands),
so any false positive chatter is completely absurd for the ONS survey (and also hugely exaggerated re the Pillar 2 tests)

Rupert Beale@bealelab (Infection lab @TheCrick)

We run everything twice to get false positives to ~1x10-4

For ONS, I believe it’s 3x and likely false positive ~1x10-6

@MRex

Actually she's reporting half the.number of tests assigned to Scotland on the PHE coronavirus board. 11.5% would mean 7008 tests. Now look: www.gov.scot/publications/coronavirus-covid-19-update-first-ministers-speech-29-september-2020/.

Bloody nonsense.

.... Nonsense indeed I'm ignoring actual test numbers from Scotland atm and just looking at % positive, cautiously

Very interesting research on antibodies from previous coronavirus on R4

Dr George Kassiotis talking about high prevalence in under 18 then drops precipitately for adults. (I won’t paraphrase although 60% v 5 ish came up). Inside Health is the programme.

‘A possible modification of COVID-19 severity by prior HCoV infection might account for the age distribution of COVID-19 susceptibility, where higher HCoV infection rates in children than in adults5,34,35, correlates with relative protection from COVID-1936.’ From

paper here

Not conclusive, and he said research work needed, I haven’t fully looked at it as busy here but will try later

Patrick Charles - ID Doctor@PCharlessIDDr (Oz)

Specificity for the assays we are using (performed at MDU and VIDRL) appears to be in the range of 99.9%.

Several times (when case numbers were low) we have done many thousands of tests without a single positive.

@BigChocFrenzy - the percentage positivity is based on the smaller test number, so it's potentially much lower.

ONS: Coronavirus (COVID-19) Infection Survey: characteristics of people testing positive for COVID-19 in England, September 2020

https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/articles/coronaviruscovid19infectionsinthecommunityinengland/characteristicsofpeopletestingpositiveforcovid19inenglandseptember2020

A surprising trend reported by ONS:

Unlike in the 1st wave, where infection growth was highest in the most deprived areas,
ONS data shows this current wave / ripple tends to have highest growth in areas of least deprivation.

However, examination in detail reveals this is specifically among the white under 35 age group,
not so much the 35+ or ethnic minorities also in the least deprived quintile

Possibly related to the above trends for age and being least deprived, there is a much sharper rise of infections in those recently returned from abroad:

The Covid-19 Map for Schools has been created by the National Education Union to show the Covid infection rate around individual schools and has links to the NEU’s guidance.

There are now (30 September 2020) nearly 6,000 schools are in areas with coronavirus interventions according to the union.

Currently, the website is updated weekly.

schoolcovidmap.org.uk/

We run everything twice to get false positives to ~1x10-4

For ONS, I believe it’s 3x and likely false positive ~1x10-6*

Erm, are false positives actually independent? Surely a false positive due to something in the source which isn't covid, but does react with the test re-agent will do so the second time it's tested too?

[quote MRex]@BigChocFrenzy - the percentage positivity is based on the smaller test number, so it's potentially much lower.[/quote]
.....
The change would depend on how well they are targeting the tests and what on earth those blasted test numbers in Scotland mean
Was there really such a dramatic drop in tests ?

The % would certainly fall if there were a lot more more tests,
e.g. the extreme case of testing everyone in the country

The question is whether for some reason Scotland really are testing far fewer than before

@sirfredfredgeorge

We run everything twice to get false positives to ~1x10-4

For ONS, I believe it’s 3x and likely false positive ~1x10-6*

Erm, are false positives actually independent? Surely a false positive due to something in the source which isn't covid, but does react with the test re-agent will do so the second time it's tested too?

... I'd assume the Crick Lab know what they are talking about

All reports I've read are that it is a statistical matter whether a test is false or not;
that there is the same chance for each test of detecting something and wrongly identifying it
hence repeated tests for greater certainty

No scientist in the field has yet tweeted or published any article suggesting otherwise

There was a detailed report for Sage that explained you can run a different test, thereby dramatically reducing the risk of false positives. (Also a "false positive" sometimes means the person had covid but doesn't have enough virus to be infectious any more.)
www.gov.uk/government/publications/tfms-consensus-statement-on-mass-testing-27-august-2020

Does this help explain the Scottish figures?

Bbc journalist asked to talk about non-compliance is Dominic something.

@MarshaBradyo

Very interesting research on antibodies from previous coronavirus on R4

Dr George Kassiotis talking about high prevalence in under 18 then drops precipitately for adults. (I won’t paraphrase although 60% v 5 ish came up). Inside Health is the programme.

‘A possible modification of COVID-19 severity by prior HCoV infection might account for the age distribution of COVID-19 susceptibility, where higher HCoV infection rates in children than in adults5,34,35, correlates with relative protection from COVID-1936.’ From

paper here

Not conclusive, and he said research work needed, I haven’t fully looked at it as busy here but will try later

...... I vaguely remember this paper from May and the scientific consensus at the time was that more data and work needed - no updates since, that I can find.

They found cross-reactive antibodies in about 10% of patients in this study

Cross-reactivity between coronaviruses seems to have a solid basis, but little information about cross-neutralisation of live virus,
i.e. how much protection this provides, if any

Sample review from Uni Oxford Immunology:

www.immunology.ox.ac.uk/covid-19/covid-19-immunology-literature-reviews/pre-existing-and-de-novo-humoral-immunity-to-sars-cov-2-in-humans

"Main limitations
· Unable to access extended data, so cannot comment on many of the study’s findings especially regarding the FACS versus ELISA sensitivity experiments (we are limited to what is shown).
· Generally, the figures we do have are missing essential data points to perform statistical analysis and interpretations (ie. Figure 1 c/d, regarding soluble S1 displacement of HCoV IgGs)
· Link to clinical outcomes for IgA/IgG/IgM responses in SARS-CoV-2 necessary to comment on protection from S-binding IgG responses.
· Cross-neutralisation experiments need to be repeated with live virus to better assess protection.
· In the ACE2 overexpression experiments (in HEK293 cells), the authors do not confirm whether the protein is expressed on the surface of the HEK293 cells (accessible to pseudoparticles) or not (remain cytoplasmic), so we cannot then conclude that it has no effect on the SARS2 Spike pseudoviral particle alternative entry (via CD145).
· Multiple sera samples (longitudinal studies) from SARS-CoV-2 patients would be needed to assess the Ig response."

[quote MRex]There was a detailed report for Sage that explained you can run a different test, thereby dramatically reducing the risk of false positives. (Also a "false positive" sometimes means the person had covid but doesn't have enough virus to be infectious any more.)
www.gov.uk/government/publications/tfms-consensus-statement-on-mass-testing-27-august-2020[/quote]
...
and we know that "false negatives" can occur when the person is tested too early and the virus has not yet sufficiently propogated to be detectable.
Hence e.g. a 2nd test some days later for airport arrivals

The bloke running the Crick lab explaining what can cause the problem of "wasted / dud" tests that produce no result - and sounding indignant / exasperated:

Rupert Beale@bealelab

We care about operational sensitivity and specificity.

If the samples are left for a week before processing;
or the swabs aren't taken properly;
or there's a barcoding error,

don't say "the tests are bad!". Rather, improve your operational efficiency.

There was a detailed report for Sage that explained you can run a different test

Yes, so it has to be a different test - I'm not convinced how a PCR test can randomly report wrong, something had to bind to the DNA for it to multiply, so if the something is in the sample (as opposed to contamination) then it would bind again. As noted a false positive due to remains of the covid RNA hanging around, but it could be any RNA similar enough that the DNA binds to it. The assumption being a different test will bind in a different way, in any case I certainly can't imagine they are truly independent enough that running the same test again rules out a false positive.

Not that I think it matters an awful lot given the number of false positives, but I think they're over-egging the chance removal.