Daily numbers, graphs, analysis thread 16

[BigChocFrenzy]

Welcome to thread 16 of the daily updates

Resource links:

Uk dashboard deaths, cases, hospitals, tests - 4 nations, English regions & LAs
MSAO Map of English cases
[[https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/909430/Contain_framework_lower_tier_local_authority__14_August_2020.pdf
Slides & data UK govt pressers
UK added daily by PHE & DHSC
R estimates UK & English regions
PHE Surveillance report infections & watchlists every Thursday
ONS England infection surveillance reports
ONS UK death stats released each Tuesday
ECDC rolling 14-day incidence EEA & UK
Daily ECDC country detail UK
WHO dashboard
Worldometer UK page
Plot FT graphs compare countries deaths, cases, raw / million pop
Covidly.com world summary & graphs
Plot COVID Graphs Our World in Data test positivity etc

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

It sounds delightful, where do we sign?

This s a data thread. Unfortunately I can not get the research to load so will have to just quote "Summary statistics showed average for-profit hospital efficiency levels at 80.1%, not-for-profit at 82.5%, and public at 88.1%."

Returning to test positivity:

From the data tables Mrex kindly found,
UK positivity as of 25 August was about 0.05% and hasn't changed much since mid-July

@alreadytaken

This s a data thread. Unfortunately I can not get the research to load so will have to just quote "Summary statistics showed average for-profit hospital efficiency levels at 80.1%, not-for-profit at 82.5%, and public at 88.1%."

... Depends on the metric: financial efficiency at minimum cost - and often minimum service - or delivering a prompt, flexible service of the standard that patients want

That's assuming that the random people they ask too take part are all equally likely to do so of course. In reality more anti-authority or chaotic people may be more likely to fail to take part.

I’d imagine that those who are anti-authority or live chaotic lives are significantly more likely to behave in ways that will lead to them being infected, so my hunch is that the ONS figures are an understatement of reality unless they correct for this, though I’m not clear how they would do this.

Superspreader events ....

Interesting article about the Austrian ski resort that was an early epicentre of infection, sending it spreading across N Europe
t&t here found many of the early cases came from there, hence the initial round of younger fitter infected

https://www.theguardian.com/world/2020/sep/05/everyone-was-drenched-in-the-virus-was-this-austrian-ski-resort-a-covid-19-ground-zero

Outbreaks in northern Germany, Denmark, Sweden, Norway and Iceland have all been traced back to skiers returning home from the Paznaun valley,
and the devastating reach of the Ischgl cluster is likely to be considerably wider

and a German study ideas.repec.org/a/wly/hlthec/v17y2008i9p1057-1071.html

"The empirical results for the years from 2001 to 2003 indicate that private and non‐profit hospitals are on average less cost efficient and less technically efficient than publicly owned hospitals. "

Our governments's political ideology favours the private sector, it has no place in a data thread.

From the data tables Mrex kindly found,
UK positivity as of 25 August was about 0.05% and hasn't changed much since mid-July

0.05%! That’s extremely low. My understanding was that anything below 5% was deemed acceptable, with many US states being well above that - in Arizona it was around 20% back in July.

Or rather, it was only when positivity was over 5%
that there was significant cause for concern.

From bits of reading, I've become aware that the UK has mapped over 30,000 covid genome sequences, out of just over 80,000 mapped globally. Does anyone have related links they could share? I've seen the initial spread tables (primarily Spain), that one care homes report and remarks about vaccines, but very little else that's directly referencing genome testing. Perhaps interesting research showing how the information is being used in practice (e.g. NZ use it to track infection clusters, does UK or any other country do this?) - and what conclusions have actually been found as a result? Thanks in advance.

@Derbygerbil

Or rather, it was only when positivity was over 5% that there was significant cause for concern.

I think you are talking about different things - the 0.05 is % of the entire population estimated to be positive; I think you are referring to the positivity rates of actual tests, which will always be higher, and which now I think In the UK is running a little under 1%. Which is still nothing much to worry about, but a different metric. I think.

@Derbygerbil

That's assuming that the random people they ask too take part are all equally likely to do so of course. In reality more anti-authority or chaotic people may be more likely to fail to take part.

I’d imagine that those who are anti-authority or live chaotic lives are significantly more likely to behave in ways that will lead to them being infected, so my hunch is that the ONS figures are an understatement of reality unless they correct for this, though I’m not clear how they would do this.

.... They may simply take their raw data and only weight it for known factors like age

• and they already exclude care home residents, who are the most vulnerable group, but not in the community

or they may try to compensate for those who refuse to participate
As a longterm career modeller (in a totally different field):

I'd first "correlate & validate" the model, by applying it to a situation with known results and then tuning parameters;
then simulating future scenarios, all the time assessing predictive capability and further refining the model

The results data to tune for would be - wrt at least age & sex - confirmed cases, but in particular hospitalisations, ICU, deaths - since the latter 3 would be much less affected by non-cooperation

ourworldindata.org/grapher/positive-rate-daily-smoothed?tab=table&year=latest&time=2020-05-10..latest

Short term test positivity rate at about end August:

UK 0.6%
Germany 0.8%
Sweden 1.4%
Italy 1.5-2.2% (unclear which to take)
France 3.6%
Spain 9,1 %

The WHO states that within 3% test positivity is acceptable
Higher rates indicate insufficient testing and / or cases rising too much
Confirms that situation in France and especially Spain is concerning

USA 5.7 - 6.2% positivity

@MRex

From bits of reading, I've become aware that the UK has mapped over 30,000 covid genome sequences, out of just over 80,000 mapped globally. Does anyone have related links they could share? I've seen the initial spread tables (primarily Spain), that one care homes report and remarks about vaccines, but very little else that's directly referencing genome testing. Perhaps interesting research showing how the information is being used in practice (e.g. NZ use it to track infection clusters, does UK or any other country do this?) - and what conclusions have actually been found as a result? Thanks in advance.

.... This is the German dataset of sequences:

ourworldindata.org/grapher/positive-rate-daily-smoothed?tab=table&year=latest&time=2020-05-10..latest

Oops, should be

https://civnb.info/sequences/

@BigChocFrenzy

https://ourworldindata.org/grapher/positive-rate-daily-smoothed?tab=table&year=latest&time=2020-05-10..latest

Short term test positivity rate at about end August:

UK 0.6%
Germany 0.8%
Sweden 1.4%
Italy 1.5-2.2% (unclear which to take)
France 3.6%
Spain 9,1 %

The WHO states that within 3% test positivity is acceptable
Higher rates indicate insufficient testing and / or cases rising too much
Confirms that situation in France and especially Spain is concerning

Travelling tabby has a daily test positivity rate for Scotland, which was 1.6% yesterday. That's up from about 0.8% around a week ago, from memory

@BigChocFrenzy - would it be valid to scale all of them to 1% to identify expected actual infection rates? Against current testing, if the total number of tests is X then the number needed to get 1% is X/100. If positivity rate is Y% then the "true infections" number is (X*Y)/100. Despite obvious flaws, it would at least be closer as an estimate.

Firstly can I say thank you to everyone who takes the time to delve into the numbers and discuss them here. I’ve been following these threads since the first one and I find them hugely reassuring and fascinating.

I do have a question though, forgive me if it’s been covered already but sometimes the sheer amount of data makes my head spin!

For some perspective to compare March to where we are now, the case numbers are very similar which on the face of it sets my alarm bells ringing, BUT, actually it’s not a true comparison since we were only testing hospital admissions at that point weren’t we? So my head thinks a better comparison would be to compare current hospital admissions with the case numbers from March since they’re essentially the same thing given the testing criteria at the time.

If that’s a reasonable assumption where would I find/does anyone know how the two compare. I know people are keeping an eye on admissions but I’m having a complete block on finding it!

Thank you @BigChocFrenzy. I'm almost scared to comment given the annoyance in the written brief about people taking data out of context ("the priority for doing so will certainly only decrease if unjustified and illogical nationalistic conclusions are being jumped to based on such incomplete data"). They got very cross!!
Great maps, but even better from the link, www.gisaid.org has useful global maps. Slightly concerned by what I consider interesting these days, but will spend some time reading!

@Cornettoninja - you can find all the hospital data here, let us know what conclusions you come to!
coronavirus.data.gov.uk/healthcare
(Worth excluding Wales admissions info by the way, they count anyone with respiratory symptoms as a case even though many turn out to be other pneumonia etc).

[quote MRex]**@BigChocFrenzy* - would it be valid to scale all of them to 1% to identify expected actual infection rates? Against current testing, if the total number of tests is X then the number needed to get 1% is X/100. If positivity rate is Y% then the "true infections" number is (XY)/100. Despite obvious flaws, it would at least be closer as an estimate.[/quote]
....
If we scale different countries to compare to each other, that would bring in the complication of different testing regimes and choices of who to test,
but it would give a very rough ballpark figure

Where it works best is to compare within the same country, cases now vs calculating "real" numbers of cases in the past, say at peak

When I was assessing # cases in Germany compared to peak, to help me assess the curent situation:

I had that handy RKI table of weekly confirmed cases and % positive tests
I saw that positivity at peak was roughly 10 x higher than recently
So I multiplied the # peak cases x 10, to get my rough ballpark estimate that cases are about 3% of peak

Checking back and considering the lower infection age & improved treatment, that correlated roughly with current v low hospitalisations and deaths here

However, it is only a ballpark figure, not something that could be used much for other calculations, just the best I could do with the available data

" Slightly concerned by what I consider interesting these days"

Yes, I never before had the slightest interest in sewage, but now I'm waiting eagerly for sewage reports

[quote alreadytaken]The message I actually came to post - for those planning to take vitamin D this winter there is nothing wrong for most people in taking tablets. I had lowish vitamin D, took tablets and had a healthy level afterwards. OH tested too, also fine.

If you want to read up on bioavailability of different forms the detail is here www.ncbi.nlm.nih.gov/pmc/articles/PMC5643801/
but I know that we are OK with tablets.

For those who have severe digestive issues or difficulty swallowing you can get oral sprays www.sciencedaily.com/releases/2019/10/191015131440.htm[/quote]
Warning - personal experience alert!
I have IBS-D with total lack of voluntary control of my rectum etc. Say no more! Many supplements are just impossible to take as they are hard to balance with Imodium. However, I can take the Vitamin D spray without any difficulty (It pretty much bypasses the gut this way). I just thought I would mention it in case it is useful for anyone with a ‘excitable’ gut who wants to supplement their Vit D. I also go for copious amounts of mushrooms! Yum.

Thanks @MRex - that’s my evening sorted out once the small person is unconscious and I can hear my own thoughts