Coronavirus outbreak

[Edwardoka]

They're not, any more than anyone is able to forecast future cases by drawing a line on a graph in Paint.

Spoiler: Facetious post warning

I was going to say "How rude" in response to you savaging my carefully crafted rigorous research without even tagging me.
However, as with all things, data allows for far better analysis than an open question.

WRT the original graph, I have extended the axes and am willing to concede that my crude graph does not paint a likely picture, as this would involve the UK hitting 400,000 cases on June 16th.

I am sorry, in my haste to do a Science I have accidentally done a Trump predicting Hurricane paths.

Normal service may now resume

 

[roubaixtuesday]

What do you think of https://epiforecasts.io/covid/posts/national/united-kingdom/ ?

Thanks, I'd not seen anything like that before.

Looks reputable (LSHTM group) and to my eye very, very worrying - in all regions R is only just below one.

I can't see how relaxing restrictions is compatible with this, unless something else replaces those restrictions. And the track and trace doesn't seem remotely ready to cope with the numbers of cases.

 

[Adam4868]

I'd be interested in a link for that?

I was talking to our local councilor,I'm not that clever on graphs,numbers etc.But I'm pretty sure from a family member working in hospital in Blackpool it's bad enough.
Link to local rag,best I could do.
https://www.blackpoolgazette.co.uk/...s-rocket-and-resorts-r-number-hits-16-2870734

 

[mjr]

I think the R number has become a bit of a distraction, like many of the other figures put out by government it is useless (also I'd include testing capacity, no of tests, stage 1-5 of the pandemic).

I think it's quite useful and could be a far better thing to communicate to the public than the covid defcon piri piri number. After all that's been reported, I think enough people would react to a trustworthy rising R number by being more cautious to make a difference.

What really counts is the number of hospital admissions (and even this can misleading as people are dying at home) and the number of people dying of Covid. I honestly think we, as a country (England), aren't doing well on these.

Deaths count, but they tell you about the situation a few weeks ago - I agree we're not doing well, though. I think hospital admissions don't count because there's too many ways to fiddle it. What counts is current cases, but I've yet to see good estimates of that, with gov.uk publishing only confirmed cases, which is farked by the testing failures.

 

[roubaixtuesday]

Iran now suffering 2nd wave.

https://www.theguardian.com/world/2...in-covid-19-cases-stokes-fears-of-second-wave

Of course, being British, and better than everyone else, this isn't relevant to us, which is why copying the precise same mistakes they have made doesn't matter. In exactly the same way we had nothing to learn from China or Italy in the first wave.

 

[Rocky]

I think it's quite useful and could be a far better thing to communicate to the public than the covid defcon piri piri number. After all that's been reported, I think enough people would react to a trustworthy rising R number by being more cautious to make a difference.


Deaths count, but they tell you about the situation a few weeks ago - I agree we're not doing well, though. I think hospital admissions don't count because there's too many ways to fiddle it. What counts is current cases, but I've yet to see good estimates of that, with gov.uk publishing only confirmed cases, which is farked by the testing failures.

I think the issue is that R is only an estimate and a not particularly reliable estimate. And as we've seen, the public doesn't cope well with uncertainty - particularly where there's a debate between different scientists/doctors who each have their own value for R.

If we could get something that was reliable and trustworthy, then it would give the public a valuable steer as to success or otherwise of the public health measures (including lockdown). However PHE suggests we treat R with caution:

It is important to note however that the R number is only one component that determines the rate of growth of the epidemic, and does not say anything about the rate of occurrence of new infections, the prevalence of infection across communities and the current burden faced by the healthcare system. It should, therefore, not be used as the sole indicator of the current threat posed by an epidemic.

But, I agree, in the absence of any other headline measure, it's probably the best we've got.

 

[Joey Shabadoo]

Localised and anecdotal but the main hospital in Dundee has had no Covid cases in ICU for the last 3 days

 

[srw]

What really counts is the number of hospital admissions (and even this can misleading as people are dying at home) and the number of people dying of Covid

Ultimately the only thing that counts is all-cause excess mortality - the number of people dying this year over and above the number who we might expect to die. The number we might expect to die is pretty predictable by simple extrapolation from previous years (allowing for changes in the age structure of the country). The number actually dying from all causes is easy to count, and we're good at it. Subtract one from the other and you get the horrific picture the FT keeps reporting. It's freely available - bookmark it.

https://www.ft.com/coronavirus-latest

But that's a lagging indicator - hence the attempt to find something more realtime. All such things are estimates, and subject to uncertainty. They're not helped by the cack-handed way the UK government is going about testing, reporting and planning.

 

[tom73]

Has the "stay alert" message been dropped ?
For days now it's never been talked about and the daily briefings no longer end with a reminder of what the "Official message" is.
Old Matt today was throwing round the stats like a 10p mix up. The only message he now wants to get over is how rosey the garden is.
Which is fine but it won't tell you what hidden in the soil which may well turn out to be nasty or the health of the bit's that aren't looking good.

 

[tom73]

I think the issue is that R is only an estimate and a not particularly reliable estimate. And as we've seen, the public doesn't cope well with uncertainty - particularly where there's a debate between different scientists/doctors who each have their own value for R.

If we could get something that was reliable and trustworthy, then it would give the public a valuable steer as to success or otherwise of the public health measures (including lockdown). However PHE suggests we treat R with caution:

It is important to note however that the R number is only one component that determines the rate of growth of the epidemic, and does not say anything about the rate of occurrence of new infections, the prevalence of infection across communities and the current burden faced by the healthcare system. It should, therefore, not be used as the sole indicator of the current threat posed by an epidemic.

But, I agree, in the absence of any other headline measure, it's probably the best we've got.

As with most stats the R number is meaningless without context. Outside health and science worlds it's pretty meaningless it's not meant for general mass use my a fixated government and press. But like you say it all we sadly have.

The made up 1-5 virus level thing may have been a good idea if it had been ready to go long before a pandemic came along. But we never had a plan in the 1st place to put it in. The public are use to hearing about terror threat level. So a virus threat level system is not a hard sell to public But it just got added later with a load of other rubbish bolted on like what can open ect. Kept simple with what it means in health terms and how to keep safe at each point. Used from day one case one and kept as a clear consent message. But it just appeared and was just added to the list of other unclear , mixed messages and instantly became meaningless.

 

[midlife]

Did Hancock sort his numbers out in the end? 400 odd discrepancy in the deaths for the last 24 hours between what he said and what the total is.

 

[marinyork]

Did Hancock sort his numbers out in the end? 400 odd discrepancy in the deaths for the last 24 hours between what he said and what the total is.

He settled up with the bookies in the end.

 

[mjr]

Has the "stay alert" message been dropped ?
For days now it's never been talked about and the daily briefings no longer end with a reminder of what the "Official message" is.
Old Matt today was throwing round the stats like a 10p mix up. The only message he now wants to get over is how rosey the garden is.
Which is fine but it won't tell you what hidden in the soil which may well turn out to be nasty or the health of the bit's that aren't looking good.

Yes, it seems that today's message is "Look after your sleep"?!?

View: https://twitter.com/GOVUK/status/1267453731555205121

 

[Slick]

Localised and anecdotal but the main hospital in Dundee has had no Covid cases in ICU for the last 3 days

1 death in the last 24 hours.

No reason to celebrate as Mrs Slick reminded me but there is hope.

 

[Ming the Merciless]

Yes, it seems that today's message is "Look after your sleep"?!?

View: https://twitter.com/GOVUK/status/1267453731555205121

That’s not staying alert though is it. Maybe trying to stay alert is reason for lack of sleep?