Daily numbers, graphs, analysis thread 2

[Barracker]

A follow on thread from here

Please try to keep it data driven, factual and civil.

www.worldometers.info/coronavirus/country/uk/

Why has our UK serious critical figure been unchanged at 163 for so long?

Wow, that's interesting @sh0ppingstar, and quite problematic re drive through testing etc

Always- on Worldometer it was stuck at 20 for over a week. Which caused a well known in these parts Corona denier to wet her/his pants over everyone else scaremongering and lying.

Truth is more prosaic I imagine. Different ways of recording, not recording at all etc.

What is true without a doubt is that the figures are higher than being recorded.

Often viruses increase in virulence while spreading - I'm wondering if that's what's happened in Europe vs the original outbreak in china. Too soon to tell I guess
@QuentinWinters I was wondering if this was the case. Or if the virus had mutated. Very worrying, especially if govt is being lead on the figures from China.

Don’t think we’ve ever doubled Every 2 days have we?

27 Feb: 15
28 Feb: 20
29 Feb: 21
01 Mar: 35
02 Mar: 39
03 Mar: 51
04 Mar: 85
05 Mar: 115
06 Mar: 163
7th March. 206
8th March. 273
9th March. 321
10th March. 373.
11th March. 456
12th March. 590
13th March 798
14th March. 1140 cases
15th March. 1372 cases. 35 deaths
16th March 1543 cases. 55 deaths
17th March. 1950 cases. 71 deaths.
18th March. 2626. Cases. Deaths 104
19th March. 3629. 137 deaths
21st March. 5018 220 deaths
22nd March. 5683 281 deaths
24th March 8077. 422 deaths
25th March. 9529
26th March. 11,658 cases. 578 deaths
27th March. 14,579 cases. 759 deaths
31st March. 25,150 cases. 1,789 deaths.
1st April. 2352 deaths.

This graphic of UK tests per day is from an article in the guardian.
www.theguardian.com/world/2020/apr/01/absolutely-wrong-how-uk-coronavirus-test-strategy-unravelled
It does show a shocking LOW numbers of tests

Someone asked about Aneurin Bevan University Health Board a while back - you’ve probably seen this article but in case you haven’t:

www.bbc.co.uk/news/uk-wales-52093126

Graph from the article.

Patients families are alleging that their loved ones caught the virus AT the hospital. That’s what we saw in the Portsmouth story yesterday (Queen Alexandra Hospital). It’s also what I’ve read about Italy:

www.nytimes.com/interactive/2020/03/27/world/europe/coronavirus-italy-bergamo.html

Screenshot from the NYT article

VivaLeBeaver deaths in the early stage, between 10 deaths and 100 deaths, were roughly doubling every two days.

I tend not to reference cases much because our testing strategy is so poor, I'm not sure how much can be inferred from the data.

The Italian health minister said something about this in Marr on Sunday, I was surprised Marr didn’t pick it up. He said that whole hospitals needed to be uniquely
Covid hospitals, to reduce the transmissions within hospitals. I suppose the new hospitals fit this though so they are more than back up, I bet they’ll end up priority areas.

I agree that deaths are probably the most reliable data to track in the UK

Shocking lack of tests, still not up to 10,000 daily, whereas Germany is managing 500,000 weekly now

Testing seems to be only on the more serious UK cases, at hospital

• and a cap of only 15% on NHS staff

Difficult from that to obtain a reliable estimation of how long it takes for cases to double

It's an absolute known fact here that people are becoming infected in hospitals. Our non Covid hospital has been traced as the one place almost every infected person in a radius of 4-5 towns had had some contact with and the paediatrics dept and one other were closed because so many staff were positive.

Concerning, if these persistent reports are true:

If Western modellers have indeed been using false China data,
then it has likely skewed their forecasts - on which some governments will have been basing their decisions

Epidemiology is not my field, but complex mathematical modelling is:

After one has built a model with discretisation, parameters and interactions etc representing the situation,
then a vital step is to validate that model using dat from a real life case and tune the parameters,
before using it for the case you really want to forecast

https://www.bloomberg.com/news/articles/2020-04-01/china-concealed-extent-of-virus-outbreak-u-s-intelligence-says

China has concealed the extent of the coronavirus outbreak in its country, under-reporting both total cases and deaths its suffered from the disease,

the U.S. intelligence community concluded in a classified report to the White House, according to three U.S. officials.
....
the thrust, they said, is that
China’s public reporting on cases and deaths is intentionally incomplete.

Two of the officials said the report concludes that China’s numbers are fake.

@TheCanterburyWhales
I wish one of the journalists would ask for the true figure at the daily briefing.

"chinese numbers are fake"
I laughed when I read that they are considered fake because at some point they didn't include people with mild symptoms. Britain doesn't include people who come to A&E, have pneumonia, are given drugs and sent him because their oxygen levels are good enough! They are not tested, not counted. But OF COURSE Britain is a democracy so no one is shouting that infection numbers we are releasing are complete bollocks.

not him, home

There is an important difference between a country that is incapable of testing adequately - like the UK - and hence underestimates cases

and a country that deliberately hides deaths - if China has indeed done so

I believe the death count in the UK - and Italy, Spain, France etc - is being honestly done, without intention to deceive,
It is unfortunately limited where tests were not done on people ill at home or in care homes

“The medical community made -- interpreted the Chinese data as:
This was serious, but smaller than anyone expected,”
she said at a news conference on Tuesday.

“Because I think probably we were missing a significant amount of the data, now that what we see happened to Italy and see what happened to Spain.”

China isn’t the only country with suspect public reporting.

Western officials have pointed to Iran, Russia, Indonesia and especially North Korea, which has not reported a single case of the disease, as probable under-counts

Others including Saudi Arabia and Egypt may also be playing down their numbers.

That Chinese CV "aid" that they so loudly trumpeted about sending to some European countries is looking dodgy too:

Health services have since been binning the PPE as defective and the tests too - missing 30% of cases

To be fair, pretty much everyone on this thread has said our published infection rates are worthless!

TheCanterbury it makes perfect logical sense that hospitals would work as hubs for the virus, especially now we now more about ‘subclinical’ cases. Absolutely awful to comprehend that the places that we rely on to make us well are doing the opposite.
My DD’s paediatric appointment was cancelled with an hour to go on the 19th. Then I assumed it was because they didn’t want to risk it being brought in but in hindsight, I guess it’s more because they didn’t want us to catch it while we were there.

I just read the phrase ‘pre-traumatic stress disorder* in relation to the anxiety and fear HCP are experiencing, more due to the fear of unknowingly infecting others than fear for themselves.

Not such good news from Italy today - deaths are down but new cases up:

4782 new cases and 727 new deaths in Italy

One of the new graphs from the FT.
Showing Italy and Spain are beginning to curve off and heading towards a plateau.

The UK is still rocketing up with no sign of slowing, yet.

BigChoc Sure, the mechanism is different. Where an authoritarian government would go for repression, a democratic one will dwell in indecision, lie, misinform, then backpedal, then gaslight, etc etc. I am not advocating for authoritarianism, I am just saying that I cannot really at the moment find a little pedestal of moral superiority to stand on in regards to other countries. Not with Boris and Cummings, and what has been going on. It also felt - from reading that link - that it was something extremely useful to Trump.

And the major cities/regions: Madrid and Lombardia beginning to curve.

London and NY not.

From that graph it looks like that's because they are ahead of us.

lifeafter40 I'm not interested in moral superiority, just quality of data.

We can all agree that the number of cases is always going to be a very unreliable number almost anywhere, so don't take too seriously

However, if the Chinese regime has deliberately falsified the number of deaths,
then that data error would be in addition to - and probably much more serious than -
the usual reporting errors that any country in a pandemic would have

Falsification of data is an egregious "crime" that I absolutely despise, as a maths modeller,
but I have dealt before with data that is inaccurate due to incompetence or circumstances beyond control.

No serious commentator has yet suggested that the UK - or France, Spain etc - are falsifying their deaths

I saw an Excel sheet on another forum with figures for Uk and Spain that were pretty comparable with an 8-9 day offset

i.e. they were suggesting that the UK is 8-9 days behind Spain, on a similar curve

It does depend where one takes the starting point