Daily numbers, graphs, analysis thread 5

[Barracker]

Welcome to thread 5 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
Google mobility stats

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

@Baaaahhhhh where have NHS England said we have peaked?

@abitoflight each day's release of deaths is based on reports received in the 24 hours to yesterday at 5pm.

Each day the vast majority of the reports will be from the previous 3 days, but there will always be some up to about 10 days ago and even older, released each day.

The most number of deaths is announced for each day 2 days later (so the death was reported the next day, and announced in the following day's release). For the last week or more every day we have had 100+ deaths reported on the day of death and 250-350 on the following day. Today's release shows 151 deaths (cell Q37) reported yesterday who died yesterday. If you follow that up diagonally (P36, O35, etc.), you'll see that is the largest ever number of deaths reported on Day 1.

This is explained either by more people dying yesterday than ever before, OR improved reporting efficiency. After 2 to 3 days of reports, we will generally have a very good idea what the total will end up as.... After day 1 only it's hard to say.

Not as bad as I feared.

I had expected something truly awful today because of delays in reporting over the break, so to see it below the previous highest total is something of a relief.

Though I have cross posted with Shoots' analysis, which suggests that we may still be in for some terrible news over the next couple of days.

Shoots, so fewer on the preceding days ? Eg: over the Easter weekend or am I missing something ?

I have to admit, I was dreading todays figures. I thought someone said in last weeks briefing to expect 1000+ a day? Maybe I imagined it. Still a terrible number but I hope we are finally seeing the flattening of the curve. I think next Tuesday/Weds will be telling as that is normally when our number seems to jump following a lag over the weekend

news.images.itv.com/image/file/2185735/stream_2.53423650.jpg

Sorry can't copy the image, also reported in Daily Fail, but thought that might be "suspect".......

We need @Barracker’s volcano graph. If you get some time, Barracker, that is.

Hospital admissions have continued to plateau there is very little change from yesterday so that is a good thing.

Something that I've only just twigged in the DOH data is that the number of tests is greater than the number of people tested.

Concluded Tests 417,649
People Tested 327,608
People testing Positive 103,093

So the number of people tested is 25% lower than the number of tests done. Is this people who are initially testing negative, who then go on to test positive, and or staff groups or are negative, then positive following further exposure?

The predictions of 1000+ death for weeks hadn't separated the death total into how many would die in hospital vs own home vs care home

It looks like the UK has plateaued, but Chief Scientific Advisor Valance and PHE bods were warning that the plateau could stay a couple of weeks before dropping

Let's wait and see - very difficult for them to assess curves & peaks,
when the total is like an iceberg, with the non-hospital deaths invisible

Al1 - that seems entirely as expected. A member of NHS staff who has Cough 1, is tested and is negative may well then be tested again when they get Temperature 2, and found to be positive.

Tests of all kinds usually have some sort of failure rate, including for COVID
so they may choose to test some people twice, to get an idea of this,
or because the 1st result for a particular patient isn't what the doctors expected

Some people may be tested both when they arrive at hospital and before they are discharged

Some staff may need to be tested every few weeks

Yes, that all makes sense. Just something to be aware of when the Politicians are trumpeting about how many thousands of tests they have done

No new confirmed cases in my area of Scotland

Bigchoc,

We have been cautioned about the length of the plateau but it does not see logical to expect the plateau to last a couple of weeks. We all locked down simultaneously so, although I can see a plateau due to some having longer incubation periods and longer time to be admitted to hospital, and then a second (smaller peak) due to transmission within locked down households, two weeks seems very drawn out.

I would expect it to come down more quickly but I guess we will see what transpires.

Thank you shootsfruit I see now
I had been optimistic about today's figures. I reckoned that fewer than 900 with the inevitable inclusion of many from the long weekend would be a good sign that we had indeed peaked last week and now on a plateau

Sorry haven't RTFT but why is the death rate in Wales 50% lower than in England? Is it because of less pressure on beds/ventilators?
Also interesting that those people saying Cheltenham Festival was fine because there wasn't a spike in cases in Cheltenham afterwards were wrong. Gloucestershire (relatively rural with only a couple of larger urban areas) has double the rate of any other county in the South-West (including the City of Bristol, much more built-up with people in closer proximity). So yes, the Festival did lead to several hundred (as of 14 April) excess cases over, for example, Devon (which had far more cases at the start of the outbreak).
Total cases by region of South West (April 14)
Gloucestershire – 701
Bath and North East Somerset – 131
Bristol, City of – 354
North Somerset – 120
South Gloucestershire – 219
Plymouth – 150
Swindon – 187
Cornwall and Isles of Scilly – 321
Wiltshire – 245
Dorset – 203
Devon – 396
Somerset – 212

I think the death numbers may still plateau over several weeks because the time from infection to death/recovery is much more variable than the time from infection to showing symptoms/ needing hospital admission.

We need to recognise that if Witty, Vallance and others over-estimate how bad things will get then everyone breathes a sigh of relief that it's not as bad as we expected, and if they under-estimate then everyone gets in a panic than it's worse than thought. So far PR reasons they are always going to err on the side of pessimism.

If we want an accurate idea of where the numbers are going it's better to look at what's happening in countries that are ahead of us.

By the way regarding testing, I think the repeat tests are for patients. The front-line critical care NHS nurse I know who was off for 2-3 weeks with obvious coronavirus symptoms and only back at work this week was never tested and has been told they won't be tested. Their trust has told staff they can test 5 members of staff every 2 weeks.

regional figures for cases cant be relied on as there are many cases where the region is not available to the websites giving regional figures.

I am happy to see now they have added a new line to the graphs on fhe briefing, it includes the ons stats for all deaths. It is a few weeks behind but at least it gives a true respresenation.

One of the issues they're going to be taking into the equations are the numbers of nhs front line staff that are ill at any one time.

The majority of staff will catch it at some point, or other illnesses which cause them to self isolate, and their absences will impact nhs capacity in some way shape or form. And sadly some are dying.