Daily numbers, graphs, analysis thread 10

[BigChocFrenzy]

Welcome to thread 10 of the daily updates.

Resource links:

Worldometer UK page
Financial Times Daily updates and graphs
HSJ Coronavirus updates
Johns Hopkins Coronavirus Resource Centre
NHS England stats, including breakdown by Hospital Trust
Covidly.com to filter graphs using selected data filters
ONS statistics for CV related deaths outside hospitals, released weekly each Tuesday

We welcome factual, data driven, and civil discussions from all contributors 💐

@IrenetheQuaint

It's not even just about formal lockdown, is it. In some ways the key date was 16 March, when Boris told everyone who could to work from home and avoid restaurants and theatres etc (though he didn't order closure until a few days later). If he had taken these measures a week earlier it would have made a very significant difference.

Interestingly, this calculation of R would suggest that those measures alone were made a significant difference - R was already falling before we went into Lockdown; raising the awkward issue of how much a wide community lockdown was really needed, if we had protected our care homes properly (they probably represent 50% of deaths). The failure to do so is no less than criminal.

This does suggest, though, that carrying on with simple precautions like hand washing/sensible social distancing such as no crammed indoor environments will keep infections under control. Especially as data from Kings College today suggests the number of people being infected per day in the community has nearly halved in a week (only one source, but encouraging, especially with the easing of lockdown that has already been happening.)

Has anyone got any thoughts about this? (I’m trying to ignore the IDS bit because u just want to understand what was going on with infection rates at & immediately before lockdown). It’s from the guardian blog.

In his article Nelson claims: “Infections peaked about five days before lockdown and were in fast decline by the time it was introduced.” His source for this is a paper (pdf) by the Bristol University statistician Prof Simon Wood. Wood used death figures to work backwards and infer the fatal infection rate in early March.

But Nelson’s argument been contested. Last night, after I posted the Nelson tweet in the blog, Dr David Shaw from the Care and Public Health Research InstituteMaastricht University got in touch to say that Nelson was wrong to refer to the infection rate peaking before lockdown was announced. “It’s not the infection rate, but the DEATH from infection rate, and there are all sorts of reasons why the death rate might start coming down before the infection rate, not least the massive ramping up of ICU capacity,” Shaw pointed out.

Is there any evidence that death rates - CFR - were falling in UK hospitals during the period covering broad peak of infection within the community. Is that what the guardian are suggesting - that we had not reached peak infection rates but that the CFR was lower during the time that the NHS was under most pressure (admissions, beds occupied, critical care bed occupied)?

ah -sorry - that was timeline for fatal infections (showing that they were in decline before proper lockdown) This is R.

Just one source, and it's just one way of looking at the data. But I do think it bodes well for us managing to keep a lid on this going forward .

The deaths peaked long after lockdown. However we'd better look at hospitalisation peak, or rather when most people were in hospital. That takes place a few days after infection and then death follows later.

Snap @Laniakea. The Bristol study is where my graphs come from.

Shaw is correct in observing the graph is modelled from the death from infection rate, it's pretty much impossible to model historical R from anything else because we don't have reliable stats on exactly how many people were infected at any one time over the past 3 months (because, of course, very limited testing). So it's not perfect. I think it does suggest ,though, that R was on a decline before we went into lockdown.

He is mathematically correct in that the death rate could be different over time, I think it's optimistic, though, to suggest we had improved it significantly by early April which is when peak hospital deaths occurred.

R was already falling before we went into Lockdown; raising the awkward issue of how much a wide community lockdown was really needed,

^ which is the opposite of what the guardian are claiming - they are saying R was still increasing at that time but magically fewer people died as a result of infection acquired then than of infection acquired at any other period. I’m dubious!

, I think it's optimistic, though, to suggest we had improved it significantly by early April which is when peak hospital deaths occurred.

^ that’s exactly what I suspect - incredibly optimistic!

England hospital daily admissions peaked on April 2, and total number in hospitals in England on April 11. (hospital deaths peaked on April 8).

www.bbc.co.uk/programmes/m000jw02
'More or Less' on antibody tests
(approximately 20 minutes in)
'there's a lot of spin in the way that manufacturers are presenting results To the press and the public' 'Every trick that can be done seems to have been done'
🙄

Sooooo -if we say 10 days from infection to hospitalisation, that would put peak infections based on peak daily hospitalisations at or about the 23rd, the day lockdown was announced.

Of course that 10 days may not be accurate. Saying 7 would have peak infections a couple of days after lockdown.

Either way, pre-lockdown measures were clearly having a significant affect in the community.

Care homes are a different story. Deaths there seem to have peaked a week or so later.

Either way, pre-lockdown measures were clearly having a significant affect in the community.

^which is optimistic in terms of controlling infection rates in a post-lockdown world? What we all want surely? Why is there spin suggesting that the only way to control infection rates is to maintain lockdown restrictions?

Derby instead of 'party like it 1999' we have 'distance like it's 20 - 19'

It is very interesting seeing how one measure can have such an impact on R compared to another.

I always believed (but had no evidence) infection rates soared the last week in feb and first week of March after all the half term ski trips.

Yesterday the report re 1300+ imported cases/ strains seems to confirm this.

I think they are much closer to knowing what triggers high infections (eg conferences) and what doesn't really have much effect.

I also think bigchoc you are right re the nationwide lockdowns being a one time thing. That's what track and trace is designed for.
So regional lockdowns, towns even etc. But our government needs to develop a good solid way of policing this and have the balls to enforce it.

I'm hopeful though as because BJ seemed to be opening everything up early but now seems to be sticking more to the plan, sticking to 2m for now and explaining (in a roundabout and blustering way but scientists explain it better!) why it's important to do what they are doing.

I'm having to keep switching my channels for MSM reporting. I absolutely am on the side we've had a shit show but I also don't think it helps that 2 weeks ago there were cries we were moving too fast and now it's we aren't moving fast enough.

I think there needs to be better explanation of statistics, scientific and epidemiologist explanations of the virus and a clear understanding given to people about how this hasn't "gone away now" but that it's there and will act the same was as Feb/March if we don't make sacrifices for what could, realistically - be another year.

we have never had a full lockdown, many businesses carried on because you could do so if you could not work from home.

As we never had a full lockdown community transmission was reduced, not eliminated.

Voluntary restrictions on activity slowed the growth rate before lockdown.

Scientists are suggesting antibodies dont last in everyone.

We dont yet know how many deaths were avoided and how many were simply postponed. However we do know that there are better treatments now than there were at the beginning. CFR should therefore drop.

Tim Harford Retweeted
Financial Times
@FinancialTimes
·
2m
What questions do you have about coronavirus data? Here's your chance to put them directly to our data journalist @jburnmurdoch
. He will be doing a Reddit AMA on Friday at 12pm BST/7am EST. Go here to join the discussion tomorrow: reddit.com/r/ukpolitics/

May have missed this but I found it interesting, think it's looking at actual death rates

www.medrxiv.org/content/10.1101/2020.05.13.20101253v2.full.pdf

Another estimation from another modeller !

James Annan has investigated what would happen

if lockdown dates had been just 1 week different:

1. UK 1 week earlier
   ==> deaths 11k instead of 43 k (as of the date of his calculation)

2. Germany 1 week later:
   ==> deaths 34k instead of 9k

https://bskiesresearch.wordpress.com/2020/05/12/the-human-cost-of-delaying-lockdown/
https://bskiesresearch.wordpress.com/2020/05/14/why-cant-the-germans-be-more-like-us/

The Sunday Times Insight report on lockdown delay would have had a lot more attention ... but it was published at the same time as the scandal of Cummings' goings came to light

Sunday Times Insight: 22 days of dither and delay on coronavirus that cost thousands of British lives
[[http://archive.is/qbvY7
archive.is/qbvY7]]
(without paywall)

....After being initially hostile to the idea,
the prime minister put his libertarian instincts to one side and agreed in principle that a lockdown would be necessary.

However, rather than locking down immediately,
there was a further nine-day delay as he deliberated over how and when a lockdown should be introduced.

That prevarication proved, for some, to be fatal.

New back-dated modelling assessing the historic spread of the disease

• which is published for the first time today -

estimates the number of people infected in the UK was indeed doubling every three days during late February and early March,

just as some of the initial reports from China in late January had suggested they might.

The work, produced jointly by an Imperial College London team led by Samir Bhatt and Oxford University,

suggests that on March 3

• the day the government committee gave the warning about the dire consequences of a mitigation approach -

there were about 14,000 infections in the UK.

Such was the speed of the spread of the virus that 200,000 people were estimated to be infected by the time the government began to change its mind about its policy on Saturday, March 14.

The last nine days while Johnson wrestled over the decision on when and how to go for lockdown were particularly brutal.

By the time the lockdown was announced on Monday, March 23,

such large numbers were doubling over such a short period that infections are estimated to have soared to 1.5 million.

According to the data,
no other large European country allowed infections to sky-rocket to such a high level before finally deciding to go into lockdown.

Those 20 days of government delay are the single most important reason why the UK has the second highest number of deaths from the coronavirus in the world.

The point is that back in March it was not known which measures would be the most effective

So with the carnage in N Italy, they had to apply everything all at once,
no time to try one measure after the other to find which are essential and which are unnecessary

Scientists hadn't the evidence then to realise that e.g. COVID has a low k (dispersion factor) and that about 80% of infections are transmitted by only 10% of the infected

Now, we are most unlikely to need a further national lockdown, just v localised lockdowns around new outbreaks in care jomes, meat plants etc

Also, mass testing & contact tracking systems should - well, at least in most European countries - prevent mass infections getting to the same crisis level as in March

Now, we are most unlikely to need a further national lockdown

^let's hope that idiot politicians, self serving journalists, scientists desperate to make a name for themselves & the British public catch up quickly because those of us who are the collateral damage can't manage this much longer.

Novel approach for a vaccine

• and with the hundreds of billions that lockdown costs, it is well worth investing even another hundred million in a 2nd British approach, rather than all British eggs in the Oxford basket

https://news.sky.com/story/coronavirus-scientists-warn-government-is-putting-big-bets-on-vaccines-that-may-not-work-12004889

The prototype is a tiny - and harmless - gold bullet coated in DNA, which is fired under the skin by compressed gas.

First results from trials in mice show it provokes the immune system to make a high number of '"killer T-cells" that clear infections from the body.

Lania Decisions about entering / leaving lockdown are entirely the responsibility of the government
and are always about balancing different interests and advice, often conflicting

e.g. Merkel started relaxation only 3-4 weeks after peak deaths, despite her virologist advisers recommending lockdown stay until R0 was down to 0.2 or 0.3

She weighed up information & advice from them and also from business, education experts etc then came to a decision,
because that's her damn job as leader

Any leader who gets distracted / scared off by journalists, business leaders, special interests or whatever is not going to cope with crisis - and there will always be crises of some sort

This morning DH infomed me that if blood group was the only factor his 0 would give him a better chance against C19, but my A would help me more in case of bubonic plague.
And then he wondered:

"...when the greatest threat to life is the overreaction of the immune system in case of C19 - do people who suffer such a storm have a history of allergies? ...
Do you know anything about this?

Kings are currently asking about hay fever, I dont know if they are looking/ have looked at other allergies. www.kcl.ac.uk/news/new-symptom-tracking-app-aims-to-slow-spread-of-coronavirus

Briefing slides are still not up, anyone seen the hospital admissions today? I managed to find the number of positive tests and that is still 1300 or so. Significant number of contacts not being traced and those giving detail have been in close contact with a bit under 4 people.

Few people have had the virus and antibodies may wear off after a few months so a lot of potential for this to get out of control again quite quickly. It may be a few people who spread the virus but we dont know which ones they are - and possibly those least likely to name their contacts.

I feel very stupid asking this as it's the sort of thing I should know, but when I looked it up originally it was very vague.
How does the contact tracing thing work? When you get in touch/they do when you're infected, the diagram I saw said that you 'share contacts'. Is this a phone app thing, or do you literally just give the names of the people you've been within two meters of? How does this work if you've only been out to Tescos?