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 💐

[Baaaahhhhh]

....

[Puzzledandpissedoff]

Placemarking for later

[Barracker]

Thanks BigChoc

[itsgettingweird]

Do you think we'll expect to see lower than average deaths in care homes now for a period of time?

We've had such a high number over the average but one would assume these were people very frail. Plus care homes aren't an infinite number iyswim? They have a capacity and once someone has died there can be empty bed and space for a while. They aren't a revolving door type institution.

[GreyGardens88]

Death figures for today will be the real test

[SellFridges]

I’ve been lurking but not posting. A question that has come to my mind though is what statistics should we continue reporting regularly moving forwards? I don’t feel that the daily death reporting is actually very helpful anymore, and I’d be more comfortable with things rolling back to weekly (perhaps via ONS).

[whatsnext2]

Thanks for new thread BCF

[ShootsFruitAndLeaves]

I’ve been lurking but not posting. A question that has come to my mind though is what statistics should we continue reporting regularly moving forwards? I don’t feel that the daily death reporting is actually very helpful anymore, and I’d be more comfortable with things rolling back to weekly (perhaps via ONS).

Deaths are falling at a steady rate and will continue to do so in fairly predictable mathematical decay. Daily deaths haven't been interesting to me for some weeks because what we're worried about now is infections growing exponentially at any point, which would then cause thousands of deaths a few weeks later.

Infection rates are therefore the useful stat now, particularly R rates, but also ONS weekly infection surveys, which have showed week-on-week drops. If they show rises going forward that would be a worry!

[PatriciaHolm]

I think infections combined with hospitalisations are the interesting ones going forward. It's conceivable that we could have a low but maintained/very slowly dropping infection rate but hospital rate decline more, if we are now identifying more mild cases in the community as a ratio of the positive tests, now we are testing a much wider range of people. So someone tests positive, but never needs medical care.

It's also possible, I think, that more of those infected now are those less likely to be made very ill, as the proportion of positives found from outside care homes/hospitals increases.

[PatriciaHolm]

@GreyGardens88

Death figures for today will be the real test

Modelling suggest we might be in for 250+ today, but that means very little. We had super low counts over the weekend and at some point PHE will catch up, having been negative over the weekend, and we know that some of the bigger hospital trusts basically don't report at weekends.

Today would have to be over 320 to even keep the 7 day rolling average static, anything under that then it's still coming down.

[SistemaAddict]

Please can I join? I've been lurking for ages but get lost in all the numbers.

[usernotfound0000]

129 hospital deaths in today's figures.

[ShootsFruitAndLeaves]

Yes hospitalisations also useful.

I don't think we can have a second wave at the moment. I mean I think it is not possible.

Unfortunately the ONS declines to publish the raw numbers being tested and found positive/negative for covid-19

The data are here

www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/bulletins/coronaviruscovid19infectionsurveypilot/5june2020/relateddata

There are no data there on test counts and test dates!

However the ONS say the number being found newly infected is 'too small to publish' (!), and the number currently infected (not necessarily within the last week) is falling sharply.

So R must have been very close to zero for considerable periods of time.... Because we know that at LEAST 2.9 million people have been infected in England (based on antibodies), and now only around 60,000 are positive. As this has been going on for only 10 weeks, it's clear that the number infected was at one point MUCH higher. I.e. late March. The earliest estimate published was 150,000 positive, between 27 April and 10 May.

So we would have had say 100 people arriving from China or Spain or Italy or wherever infected with covid-19, and at one point that was displaying perfect exponential growth doubling every 3 days, until we reached a peak of current infections that was far over half a million. Some would have already recovered at the peak of infections of course, so don't forget that, and they all add into our 2.9 million, but clearly in the exponential growth stage, infectious people are infecting others before many have recovered.

Anyway, we locked the country down, and whereas previously the growth might have gone something like:

1 March - 10,000
4 March - 20,000
7 March - 40,000
10 March - 80,000
13 March - 160,000
16 March - 320,000
19 March - 640,000 (possible peak?)

So if we all went back to our dogging and rock concerts and what have you, our brains having been wiped by some sinister force, then in a week the infections could quadruple from their current level. And in 11 days or so you could get back to the peak.

But clearly that's not possible.

We now have data that were in March not available.

We are not going back to full on sweaty orgies tomorrow.

We are exercising caution.

And if it turned out that infections went from 60k this week to 65k next week that wouldn't necessarily be disastrous. Because clearly there are 'excess deaths' now with covid-19 afflicting 0.1% of the population. And if that were to go up to 0.15% there would be more deaths (not 50% more than now necessarily because of the way that current deaths may have been in hospital for months, and were infected when infection rates were higher).

But there isn't a number written in stone where we have to shut the country down again. We 'unlock' certain activities. We wait, we watch the numbers. Then if that's ok, we continue and maybe unlock more. And if the numbers start going up again then we re-evaluate.

But given major behavioural changes full exponential growth is very unlikely and we'd catch it before it could get back to normal levels. Note that we are seeing signs of boredom with this shit, in mass protests, which wouldnt have taken place in April under the same circumstances, and some people realising that covid-19 is not at all a threat to them (assuming they are young and healthy).

So we could at some point see infections going up again, but for now we have had continually declining infection counts (where that is active infections, not necessarily in the last 7 days) and very low new infections (except they are not publishing the raw data), but it doesn't seem like it's something that could just explode.

[ShootsFruitAndLeaves]

I see this study

www.mrc-bsu.cam.ac.uk/now-casting/

higher figures. In particular they're claiming 5.6 million total cumulative infections which is well above the highest estimate from the ONS.

They claim 362k people infected DAILY as a peak.

Also some absolute arsehole journalist who can't even do English let alone maths claimed

www.cornwalllive.com/news/cornwall-news/south-west-r-rate-coronavirus-4205790

"South West R rate to see coronavirus death toll rise in mid-June according to study"

in fact

We believe it is probable that Rt is below 1 in all regions of England with the exception of the North West and the South West
In the South West, although Rt is around 1, the numbers of new infections occurring in the region on a daily basis is relatively low
There is some evidence that Rt has risen in all regions and we believe that this is probably due to increasing mobility and mixing between households and in public and workplace settings
An increase in Rt will lead to a slowdown in the decrease in new infections and deaths
There is evidence, from the forecast of deaths for the whole of England, that the increases in the regional reproductive numbers may result in the decline in the national death rate being arrested by mid-June


i.e. they are saying 'daily deaths might stop falling by mid-June'. Which is not the same as a rise!!!!

Absolute arsehole journalists.


Anyways whichever methodologies and numbers you look at you can see that we can't possibly go from 10k daily infections back to hundreds of k daily infections without noticing. So second wave on the scale of the first is simply impossible, but you could have infections rise over time.

[whatsnext2]

And in addition R is becoming more accurate as we now know about asymptomatic transmission, superspreaders, have more accurate tests and a more accurate baseline.

It was all a bit stick your finger in the wind to see which way it was blowing before.

[LivinLaVidaLoki]

@whatsnext2

And in addition R is becoming more accurate as we now know about asymptomatic transmission, superspreaders, have more accurate tests and a more accurate baseline.

It was all a bit stick your finger in the wind to see which way it was blowing before.

I was under the impression that R was less accurate as the case numbers lowered which is why, where it was vital a while ago, it is now "a part of the wider picture"

[SistemaAddict]

Is there a new time for the full figures or are they still supposed to be at 2pm?

[PatriciaHolm]

@Bercows

Is there a new time for the full figures or are they still supposed to be at 2pm?

Full figures haven't been coming out much before the daily briefing for a while - 2pm is still (ish) when the NHS England come out, PHE is always later. I'm beginning to think Govt only release them just before/in the briefing to give themselves something to say!

[SistemaAddict]

You're probably right! I've not been as attentive to the briefings lately. I'm shielded with no end in sight, my dc can't go back to school yet, I don't work at the moment, so the briefings have little to offer except frustration at the fudging and that Groundhog Day feeling. It's great the see the curve coming down. I do worry about the potential for a second wave now that the world is opening up again slowly but surely. I might tune in tonight although it's always at bloody dinner time and the children moan about me having it on. No tv in the kitchen or dining room so I watch on my phone much to their disapproval.

[PatriciaHolm]

I'd be worried i would throw the phone in disgust tbh ;-).

I don't watch. Can easily look at the numbers and key points afterwards. Way too rage inducing to watch.

[ShootsFruitAndLeaves]

@LivinLaVidaLoki that doesn't matter that much in that that means the infection numbers are low so deaths cannot increase.

Clearly as infection numbers fall it gets harder to measure increases, especially when they could be very localised. But mass community spread isn't possible without us noticing

[LivinLaVidaLoki]

@shootsfruitsandleaves
That's what I mean, sorry lack of sleep means Im not explaining myself very well. That its only a small part now as infection and hospitalisations are falling which is a better idea of community transmission.

[itsgettingweird]

Shitty journalist now on daily briefing.

What was outcome for review for distance.

2m.

You promised a review what was the outcome.

Sticking with 2m.

but you promised a review. What's the outcome.

We are sticking with 2m.

Guess he didn't like the answer.

But as this is factual thread I know there is different distances in various countries. Has anyone read the WHO report that says 1m? I understood the risk was double at 1m than 2m so can see why currently we are keeping our distance higher.

[itsgettingweird]

The scientist is explaining now. Pretty much confirming what I thought was fact from what I'd read.

[SanityDecreasing]

@itsgettingweird, yes, I didn't like that journalists approach either. Very aggressive.