Daily numbers, graphs, analysis thread 18

[BigChocFrenzy]

Welcome to thread 18 of the daily updates

Resource links:

Uk dashboard deaths, cases, hospitals, tests - 4 nations, English regions & LAs
Imperial UK weekly LAs, cases / 100k, table, map, hotspots
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
R estimates UK & English regions
PHE Surveillance report infections & watchlists each Thursday
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 test positivity etc, DIY graphs
FT DIY graphs compare deaths, cases, raw / million pop
Covidly.com world summary & graphs
Alama Personal COVID risk assessment

==> Our STUDIES Corner

There was talk of something being developed that models for long haul patients, but that's only at the beginning stage.

I'd guess in Scotland that number will be very small as a percentage of 48 overall.

@Piggywaspushed

Hmmm.... not sure how Michal Rosen, or anyone else who had a long hospital stay (is Kate Garraway's OH still in?) would view being 'not counted'?

Is this just a pure political move? Or is there something I am missing?

Yeah, that part didn't make sense to me. I guess they are trying to distinguish between people with complications solely due to CV, and very elderly people for whom a visit to hospital for any reason is likely to result in a long term stay.

Who would have thought it would be so complicated to count people.

@Piggywaspushed I imagine/hope it’s just based on the most accurate option. If the current method of reporting means that eg 100 people are counted who should not be and the new method means eg that 10 people are not counted who should be then clearly the second way is more accurate. Judging by how much it has made their hospitalised figure drop I’m assuming that there were a lot of people being included who shouldn’t have been.

When/ where was that UK alert level chart from? This was the May 11th version used through June and July, quite different: images.app.goo.gl/7LsHR5kFpDCyuVJt7.

I don't think figures would yet be described as having high community transmission. The alert level dropped from 4 to 3 on 19th June, when we were averaging 1147 cases per day and dropping. Hospitalisations were still much higher at the time. I'd guess we'll return to level 4, but average of 3000/day might be low for it. Next week perhaps. We do have some areas already with increased restrictions, I suppose that's recognition that those areas are at 4.

@Piggywaspushed

Hmmm.... not sure how Michal Rosen, or anyone else who had a long hospital stay (is Kate Garraway's OH still in?) would view being 'not counted'?

Is this just a pure political move? Or is there something I am missing?

Quite.

The FM did mention that this is not a perfect way of doing it and they will be looking into ways of estimating the cases that would be missed.

I understand the need not to overcount the number of hospitalised cases, but I'm not sure that switching to a system where we knowingly undercount is much better.

The most accurate way would surely just be for hospitals to declare to the health board how many COVID patients they are treating each day. Then if someone has ceased to be a COVID patient because they have recovered and are now being treated for something unrelated, they could just be removed from the count. I may be overestimating the ability to share detailed, timely data though!

The inflated Scottish number was just obviously disproportionately high; we were sitting at 200 inpatients even when we were getting just a handful of cases a day. It needed to be rectified for clarity.

@Ecosse where did u get your Scotland figures from? The current figure on the dashboard for patients in hospital with COVID is 269

@PrayingandHoping the most up to date Scottish figures are here: www.gov.scot/publications/coronavirus-covid-19-daily-data-for-scotland/

The number is lower because Scotland has changed the way is counts hospitalised cases.

@conkersarebonkers 🙄🙄 why on earth these countries can't get their act together! Means that actually there's no way to gauge how and what the inpatient numbers are actually doing for weeks.... because now we have no comparable history data

[quote PrayingandHoping]@conkersarebonkers 🙄🙄 why on earth these countries can't get their act together! Means that actually there's no way to gauge how and what the inpatient numbers are actually doing for weeks.... because now we have no comparable history data [/quote]
The cynic in me says that is exactly why they keep changing it. Exactly the same when England changed to the 28 days rule. Nothing to do with "being more accurate" more camouflaging the true figures from the public.

[quote PrayingandHoping]@conkersarebonkers 🙄🙄 why on earth these countries can't get their act together! Means that actually there's no way to gauge how and what the inpatient numbers are actually doing for weeks.... because now we have no comparable history data [/quote]
And in a few months time when everyone has worked out how the new system works...they will change it again to make sure you can't work out the true picture again

@Witchend

The new system in Scotland is not about hiding figures- it is a much more accurate reflection of the number of people actually being treated for COVID.

It makes no sense to include an individual who is being treated for a stroke.

This is why most public health experts, epidemiologists & statisticians say that excess deaths - deaths over the historical average for that timeframe -

are the best metric to assess the seriousness of an epidemic and to compare countries

The standard definition around the world of a Covid death is death from any cause within 28 days of a positive test.
There are a few exceptions, e.g. Belgium who count deaths where the doctor merely suspects Covid, but has no test or lung Xray that confirm this

Belgium's figures for Covid deaths are very high, but match their excess deaths quite well

For the UK, ONS figures show that the real number of deaths is between the 28 and 60 day cutoffs
In the earlier months in particular, official death figures were much lower than ONS ones

[quote BigBeanBag]I shared this yesterday but going to again today as think it will give lots of useful data about schools and transmission, no doubt will take time but looks very interesting

www.bristol.ac.uk/news/2020/september/commins-study.html[/quote]
Interesting? That's not the word I'd use - machiavellian is more what I thought

[quote Ecosse]@Witchend

The new system in Scotland is not about hiding figures- it is a much more accurate reflection of the number of people actually being treated for COVID.

It makes no sense to include an individual who is being treated for a stroke.[/quote]
Covid-19 has been shown to have effect on blood clotting so can cause strokes and heart attacks, so if they've tested positive then there's a good chance that they are being treated for a covid-related condition.

I doubt it's a more accurate reflection, I suspect it's under reading rather than over reading.

If they'd done it in the middle of the low period, I would think they might have reassessed and decided to do it that way however to me the timing is suspicious-just as things are getting worse.
It most definitely is not the time to change the way they measure when we really need accurate figures. Putting head in the sand and massaging figures is not the way we're going to get out of the issue.
By changing it means that it's harder to compare like with like and understand if there is flattening of the peak, worsening, or severely worsening.

This is the link for the school attendance data: explore-education-statistics.service.gov.uk/find-statistics/attendance-in-education-and-early-years-settings-during-the-coronavirus-covid-19-outbreak/2020-week-37

Do people think the threshold for full bubble closure is too sensitive? Or about right?

Just reading posts about 2 positive cases sending 200 students home for two weeks.

I am reminded of early 111 days triage and how a system can need reassessment.

It's worth noting that link says it is Beta. Anyone who wants to add ideas for extending and improving it, please note some of us did that when the main dashboards were being set up and ideas did seem to lead to changes.
I'm considering putting in a request for simple charts and tables that summarise the key data relating to attendance and absences due to covid. I need to not have a toddler shouting to be able to formulate my thoughts.

(I mean putting them on the main page, instead of needing download from separate files.)

I've just found a gem from that data: "Of all schools that responded to the form, 1% said they were not fully open due to suspected or confirmed cases of COVID-19. Schools that reported they were not fully open were asked why.Non-COVID-19 related reasons include staggering entry for children in nursery and reception in the first few weeks of term."
Oh FFS! Come on! This is on top of 10% not having responded at all. This says to me that we shouldn't underestimate the scale of task in getting effective covid monitoring for schools.
One clear fact, whether this is high, low or as expected:'"Of all schools that responded to the form, 1% said they were not fully open due to suspected or confirmed cases of COVID-19."

This might have been covered so apologies. With schools-we have a number of students who aren’t attending due to low grade illness but are now homeschooling when they are well enough-do you know whether they are recorded as absent?

But @witchend that's assuming the figures were accurate in the first place? Given that Scotland has shown consistently over 200 inpatients for months on end now, I think it has to be questioned. The number was extremely stable, which made me wonder if we were counting lots of long term elderly patients who had contracted Covid at some point in their stay which was primarily for other reasons.

It doesn't seem logical that we had more inpatients than the whole of England at points really.

@MarshaBradyo

Do people think the threshold for full bubble closure is too sensitive? Or about right?

Just reading posts about 2 positive cases sending 200 students home for two weeks.

I am reminded of early 111 days triage and how a system can need reassessment.

I fell like I am saying the same sort of thing all the time (because I am). But it depends on how you see your level of risk. If I were in my late 20s, with a mortgage to pay, both parents working and no health conditions,I would be livid if I could not send my kids in, for what seemed a low risk.

As someone who is erm about twice that age, sort of semi retired and ECV, I would be very keen for my kids to stay off if covid is in the school.

So it would be good if parents are allowed some intelligent flexibility.

@Witchend well currently they haven't changed the way they are reporting on the dashboard so for now we still have comparison