Daily numbers, graphs, analysis thread 6

[Barracker]

Welcome to thread 6 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

Thank you to all contributors for their factual, data driven, and civil discussions.

Reminds us of the many reasons for the delay in deaths coming into the official figures:

Some information may not be easy to obtain for very sick patients without close family

e.g. UK deaths are recorded against place of residence
DOB would be required, possibly place of birth too

and of course, there could be delay in some cases in order to carry out post-mortem COVID tests

Some data collected to enable COVID studies & analyses might also be delayed
e.g. ethnicity, previous medical history, BMI

larry I understand the mathematical reasons for a long slow decline after peak,
but it is a factor that every country has to consider:

both the anxiety & impatience this can cause among the general public,

as well as the fact that if relaxation causes exponential growth again, then even going into lockdown means a high number of COVID deaths continuing for many weeks
So the caution - that infuriates many people - is understandable and wise

Apparently guardian now reporting an inflammatory syndrome in children related to covid 19 very worrying

there are not many children actually affected by this inflammatory system

*syndrome

whatever this means

These are only preliminary reports, on so far just a tiny number of children

imo we should wait for more data and some expert medical analyses

If you were going to reopen areas you'd do it in geographically logical ways, so that for example London commuter towns even if not severely affected might stay in lockdown, but you would allow for instance Lincolnshire or Devon to reopen as there are few daily transactions by residents there with areas of high infection.

At the moment there may not be a need for instance to stop people in, say, Ceredigion for going to the pub and socialising locally. But you would need mass testing to be sure that of that.

Data on children in hospital with COVID-19
www.rcpch.ac.uk/resources/covid-19-service-evaluation-audit-care-needs-children-admitted-hospital-england#results

Russia has confirmed 6,198 new coronavirus cases, bringing the total number to 87,147
Now Russia has more cases than China
(or what China admitted to)

"But you would need mass testing to be sure that of that."

Yes, again we need more data, or rather the govt do.
Mass testing must be ongoing, to reflect how the infection distribution and scale change in future with the measures

Reports are that sampling will be carried out across the country, with new (functioning) antibody tests
This will help give a more accurate picture

Anyone know the timeline for this ?

Interim results of German antibody tests around the country are expected next month, exact date not known.
This study was delayed until an acceptably accurate test was fully developed and then produced in sufficient quantity
I read that additional studies also likely in hotspots and among selected groups

No I don’t know timing but sounds good they are doing it. Have been waiting on these antibody tests with interest. As have many I’m sure.

Also on keeping say London under restriction and opening up businesses in quieter areas. I do think more of an exodus would occur. Second homers may have gone already to some extent, but the prospect of open pubs / cafes elsewhere will be more of a draw than it is now.

Of course. If they allow for example, North Wales or the Peak District to open up cafes etc but don't restrict travel, masses of people from Greater Manchester will go for days out!

350 England, Scotland, Wales

I predict late 600's tommorow rising to 750 on Friday. It's slooow they did warn us.

Even with the proviso that yesterday was Sunday ...

v encouraging that Germany got back into double figures for that death total
although only just ! - 99 deaths yesterday

but expecting the usual post-weekend jump today & tomorrow

Looks like we are all continuing to drop slowly

• but 2 weeks from now would be when the effects of relaxing measures in some countries may start to show in the figures

So maybe Johnson/Raab are waiting for that....

My suspicion has been for a while that they can’t announce anything until they have a sufficient test and trace infrastructure in place (and at present they very clearly don’t). It’s clear you can’t really release anything without test and trace taking over to keep the R0 down so that has to form part of the plan.

No point announcing that’s what you’re going to do unless you’re at least in striking distance of actually being able to do it!

Without mass testing and contact tracing, there would be much higher risk of realising too late that exponential infection growth had started again

So yes, unless PHE and the govt are totally irresponsible, they'll want that in place before relaxing lockdown

Also, they should hopefully learn from results of mass testing from other countries,
who have already slightly relaxed certain lockdown measures

Here's a graph to show cumulative excess deaths by region. It shows

London first
North East 2 weeks behind but growing more quickly than London
East, North West 1 week behind
South East, East Midlands, South East, Wales 6 days behind and growing more slowly than London
Yorkshire 10 days behind and growing more slowly

Sorry the NE is probably only a week and a half behind, there are excess growths in the third week there. We do generally see a faster spread in London than other regions because London is more cohesive whereas other regions are effectively comprised of sub-regions which may grow largely independently of each other.

If holidays in devon and cornwall go ahead they will be badly hit. As thry have low levels now.

The data from thevpillars testing is interesting as about 1/3 is key workers.

That may well be showing th difference between uk and countries with masks.

West Midlands just a few days behind London.

So what we had was 500 (to 20th March), 2700 (27th March), 6300 (3 April), 8700 (10 April) weekly extra deaths. Probably another week of 8000+ and then subsequent further weeks of thousands of extra deaths.

Lockdown was not until the 23rd, effecting a peak on the 8 April. Lockdown a week earlier (the 16th) and you would have had a peak maybe 1 April, so no effect on week of 27th (still 2700 dead), but reduced dead in week of 3rd and subsequent weeks. 10,000+ lives saved.

Lockdown 2 weeks earlier and this place would be full of even more idiots than now complaining about how the lockdown was unnecessary, but with a peak week of just 2,000 or so, and perhaps 5,000 total extra deaths rather than what will now be around 40,000.

I remember watching Crufts and thinking everyone there was insane, but actually that could still have gone ahead.

The current lockdown is a big failure in that it was too late and still destroyed the economy, but we won't know how many extra deaths we would have had if we did nothing at all.

Allowing those mass public events was irresponsible and cowardly, but that can't be undone

Lockdown was at least one week later than it should have been, if the govt had not dithered
2 weeks if they had been really alert to what was happening elsewhere

However, we were then on an exponential growth curve and without lockdown there is no convincing evidence that the curve would have flattened anyway

Without lockdown, the UK could have been in an unenviable class of its own,
with something like the Imperial scenario of ¼ million deaths

The deaths in this 1st wave will exceed the original Imperial lockdown scenario of 20,000 deaths, especially when we add in care homes and domestic residences.
However, it's in the rough ballpark
and anyway we haven't seen what start date Imperial used for lockdown

Again, I really would like a 2nd model that has been built independently of Imperial and specifically for COVID, to see how the predictions vary,
but until then, we have to use what we have
and of course SAGE would also collate the advice of many experienced epidemiologists and virologists