Daily numbers, graphs, analysis thread 12

[BigChocFrenzy]

Welcome to thread 12 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 📈📶👍

Just seen on worldometer that Florida have recorded over 15,000 new cases today

@FurForksSake

650 cases, 21 deaths. It was 516 cases last Sunday and 22 deaths so not great. Hopefully it is due to more testing in areas with outbreaks.

Yes, I hope so. Those cases come largely from the previous 4 days (7-10); rolling average on cases per day of sample is now quite flat (have excluded last 3 days as more will be added over next few days)

[quote MarcelineMissouri]@FurForksSake where is the first place these figures come out now? I think I’m looking at the wrong thing. Thanks![/quote]
coronavirus-staging.data.gov.uk/ seemed quickest today

There were about 1,500 workers crammed together in that meat plant in the recent German outbreak

and their ignorant, arrogant bosses abandoned their legal responsibility to provide a safe workplace,

because "it's just flu"

The owners are expected to face severe financial penalties for breaking the SD rules that all employers must follow

Reports also that the virus filters in the a/c were aoverwhelmed by COVID,
so either filter changes were not sufficiently frequent,
or the filters used were not adequate for the purpose

Meat plants have been running throughout the pandemic in most countries, as it is essential for the food supply

• Trump even made an executive order in the USA compelling this

Outbreaks are more noticeable now that there are so few elsewhere,
but the cold & humid conditions, noisy so people keep shouting to be heard, are ideal for COVID transmission

So even with v low community infection, having many meat plants with hundreds or thousands of workers each means that
the cumulative risk of outbreaks remains high,
especially if the rules are ignored by fuckwit owners

The Uk figures continue a downward trend in infections, hospital admissions, deaths.

The BLM demonstrations and beaches did not produce a spike
Within the next week we'll see if pubs etc did - but it would be v surprising, as there was no spike on the continent after reopening

These figures do not justify the current restrictions on childcare over the summer, which is causing serious logistical problems - for mothers especially

There is a big difference in transmission risk between childcare vs care homes & meat plants.

Across Europe, we've seen repeated outbreaks at the latter,
but schools and childcare have only been closed down as a precaution, or because of a few infections from outbreaks esewhere

Schools & childcare have not been epicentres of infection as was feared
They are the lowest risk

Where do you see hospital admissions @bigchoc? I used to look at the daily briefing slides and haven't been able to find them since they ended.

Some of the new cases will be caused by very localised outbreaks e.g. the farm in Hereford where there are 73 cases. At least now, those outbreaks are being dealt with quite quickly.

Even the so called '20 areas of concern' are mostly heading downwards. Even where they are not, in most the number of new cases is actually quite small.

I had a bit of panic yesterday when I saw the data but now I understand better that it's not just the last 24 hours. The PHE graph showing cases by specimen date is quite useful for that.

Everything else also seems to continue downward
e.g. deaths/patients in hospital etc.

@Firefliess

Where do you see hospital admissions *@bigchoc*? I used to look at the daily briefing slides and haven't been able to find them since they ended.

coronavirus-staging.data.gov.uk/healthcare?areaType=nation&areaName=England

To see the up to date data you need to look at country level.

England hospital admissions steadily down (last day or so likely to adjust)

Thanks @patricia

73 infected out of 200 farmworkers on a Herefordshire farm.

Tough manual work, but again the pre-COVID setup looks basically legal

With crop-picking, SD would be very difficult and the manual effort & heat probably make masks impractical

Then there are also communal living conditions to spread infection.
However, unlike meat plants where workers work longterm and live in blocks of flats,
accommodation for casual farm workers is typically communal, mixing several "households"
So creative thinking - and some expense - required for other farms to avoid outbreaks

Hey all! I did ask a few comments up but don’t think anyone saw it. Sorry to ask on here as I know it’s more to talk about statistics etc! I just wondered how the Zoe app gets their estimated active cases? Is it people logging symptoms or is it people testing positive? Thanks!

Astonishing, Florida with population about 21 million, had over 15,000 new cases yesterday

@Cusano34

Hey all! I did ask a few comments up but don’t think anyone saw it. Sorry to ask on here as I know it’s more to talk about statistics etc! I just wondered how the Zoe app gets their estimated active cases? Is it people logging symptoms or is it people testing positive? Thanks!

Ah! Sorry.

It's a modelled number, based on a mixed of things. Basically, they look at the users they have who have reported symptoms and then got tested, and use those symptoms to look at all the other users they have to estimate how many of them might also have CV. They then extrapolate that out to the population at large.

It's one approach, but by it's very nature the app will attract a self-selecting group of users, and it will be hard to adjust for that. They have had to adjust the model twice already (once to remove the effects counting antibody tests (tests for having had covid) as positive current tests, and once for including a lot of historical positives as current positives.

I use the Zoe app and I think its a statistical projection from symptoms. They had some blurb last week about the figures falling significantly as they’d changed the methodology to take long term symptomatic people out (eg people still reporting loss of taste and smell in the third week).

I’m new to the thread so apols if this has been thoroughly covered already - I have been really struggling to understand the difference between the hospital and UK deaths. Are there really around 100 deaths a day still in care homes? And if so why aren’t care homes residents being cared for in hospital now? I know many people prefer to die at home but given how the virus spreads and the difference in PPE, care etc outside a hospital it doesn’t seem to make sense to me.

cusano I think Zoe are modelling UK-wide prevalence via their user test data
Their estimates agree pretty well with ONS:

https://covid.joinzoe.com/post/data-update-prevalence-covid

"For the last two months, we have been running the testing programme with the Department of Health and Social Care in England,
which has seen more than 350,000 users being invited to take swab tests when they begin to feel unwell.

Users then log results of the test into the app,
helping us to build a better picture of what symptoms are related to a positive test.

Our new prevalence figures show that 23,459 people in the UK currently have symptomatic COVID and highlight the big regional differences across the UK.
........
This estimate is in line with the latest ONS Infection surveyy_ in which 25,000 people in England were estimated to be infected with COVID-19 during the two week period that goes from the 14th to the 27th of June.
....
Yesterday’s active cases + today’s new cases - today’s recoveries = Number of active cases

@PatriciaHolm thank you! That’s makes sense! So it could be less or could be more? The gov site shows only 2 more cases since June and the the Zoe app added 14 in a day!

@Whentheshipgoesdown

I use the Zoe app and I think its a statistical projection from symptoms. They had some blurb last week about the figures falling significantly as they’d changed the methodology to take long term symptomatic people out (eg people still reporting loss of taste and smell in the third week).

I’m new to the thread so apols if this has been thoroughly covered already - I have been really struggling to understand the difference between the hospital and UK deaths. Are there really around 100 deaths a day still in care homes? And if so why aren’t care homes residents being cared for in hospital now? I know many people prefer to die at home but given how the virus spreads and the difference in PPE, care etc outside a hospital it doesn’t seem to make sense to me.

Are there really around 100 deaths a day still in care homes?

No. The PHE non-hospital data experiences a lot of lag - as in, deaths that happen days ago but we have only just registered them - and very big swings in reporting from day to day.

This graph is from the ever reliable @rp131 on Twitter, showing non-hospital deaths by day of death, not reporting. Below 50 a day for the last 10 days or so.

It's also worth flagging that the non-hospital deaths are people who pass away having had a positive test, not that they necessarily died on CV. In non-hospital settings, this may well include a number of people who died of other things and did not develop symptomatic covid.

Deaths by day of death, all settings England have been under 100 since 27 June.

Thank you @PatriciaHolm

Good to see numbers still reducing

I have to ask .....
why are you still placing faith in numbers which have been proven to be lies?

If you look at the patients admitted section on the staging data

coronavirus-staging.data.gov.uk/healthcare?areaType=nation&areaName=England

just 36 admissions on friday compared with 100+ the previous few days. Not sure I believe that.

...which have been proven to be lies...

No, the UK numbers have never been "lies".

It was very obvious that until the end of April the UK testing system was an absolute mess. The Sky article that was mentioned here several times did not reveal anything particularly new. It confirmed that the daily numbers were calculated at the back of the envelope, collated manually and at best presented a directional view of numbers of COVID patients admitted and diagnosed in hospitals.

However the reality is no-one following COVID seriously paid any attention to daily case numbers up until very recently. Previously all the analysis was based on numbers of deaths. Hospital reporting and ONS records have always been reliable (although available with a time lag) and provided a reasonable view of the state of the epidemic.

@PatriciaHolm

Appendix to The Lancet study says - "We enrolled HCWs at University College London Hospitals between 26th March and 8th April 2020. Eligibility criteria were being asymptomatic at time of enrolment and working in one of five clinical areas: Accident and Emergency (A&E), acute medical admissions (AMU), COVID-19 cohort wards, Intensive Care Unit (ICU) and haematology ward"

So it would appear it was a self selecting sample and not weighted/representative.

Overall, 44% at some point either tested positive for current infection or had a positive antibody result indicating past infection.

Interestingly, 25% already tested positive for antibodies at the start of the study - March 26.

Sorry, this is going back a few pages but - the 25% having antibodies at the start of lockdown seems shockingly high to me! I know it's a small sample size but still. Also 44% of the asymptomatic sample testing positive over the two weeks!

I also found this article a bit depressing - what is the Track and Trace service for, if not this? www.theguardian.com/world/2020/jul/11/rumours-and-threats-what-happened-when-covid-19-shut-our-pubs