COVID19

[kukushkin88]

We’re back to this being a daily occurrence for now:

https://mobile.twitter.com/DHSCgovuk/status/1244976749940703233

[bookiebasher]

Do you geta ventilator?

https://thehill.com/opinion/healthcare/490034-how-hospitals-will-decide-who-lives-and-who-dies-in-our-coronavirus-crisis

Do you want one?

https://www.facebook.com/DrKathrynMannix/posts/2949195348436749?__tn__=K-R

I can't really answer that one, but I do think it is unduly negative about ventilators.  You can survive intensive care and ventilators, and get back to a near normal life; well you could until covid-19 happened.  I wouldn't want a ventilator if it meant I was going to be a vegetable, but I'd also want one if there was a good chance of a reasonable life.   

Cheers Doobs , good articles

[StuartHopkin]

Update for my work is next bank meeting on Friday PM and they cannot bring it forward.....

[RickBFA]

Don't have a link for this but listening to LBC earlier on my daily walk, I heard a reporter quoting research from Bristol University.

They reckon a drop of 6.4% or more in our GDP will result in more deaths in the UK than if the corona virus was left to run its course unchecked.

[Doobs]

We’re back to this being a daily occurrence for now:

https://mobile.twitter.com/DHSCgovuk/status/1244976749940703233

A BBC reporter has it on his twitter, so it should be accurate?

https://mobile.twitter.com/lewis_goodall/status/1244985323660152832

And some of them had high blood pressure so were going to die soon anyway :/

I assume the big leap is probably just down to delayed reporting as we have just gone through a weekend.
 

[kukushkin88]

We’re back to this being a daily occurrence for now:

https://mobile.twitter.com/DHSCgovuk/status/1244976749940703233

A BBC reporter has it on his twitter, so it should be accurate?

https://mobile.twitter.com/lewis_goodall/status/1244985323660152832

And some of them had high blood pressure so were going to die soon anyway :/

I assume the big leap is probably just down to delayed reporting as we have just gone through a weekend.
 

It is a big leap from the day before but in terms of expected exponential growth it is still on the low side of what is expected. I guess the working assumption until the next ONS update will be that we are ~20% under the true number.

[kukushkin88]

Don't have a link for this but listening to LBC earlier on my daily walk, I heard a reporter quoting research from Bristol University.

They reckon a drop of 6.4% or more in our GDP will result in more deaths in the UK than if the corona virus was left to run its course unchecked.

https://www.google.co.uk/amp/s/www.theweek.co.uk/106338/why-economic-crash-could-cost-more-lives-than-coronavirus%3famp

[Doobs]

We’re back to this being a daily occurrence for now:

https://mobile.twitter.com/DHSCgovuk/status/1244976749940703233

A BBC reporter has it on his twitter, so it should be accurate?

https://mobile.twitter.com/lewis_goodall/status/1244985323660152832

And some of them had high blood pressure so were going to die soon anyway :/

I assume the big leap is probably just down to delayed reporting as we have just gone through a weekend.
 

It is a big leap from the day before but in terms of expected exponential growth it is still on the low side of what is expected. I guess the working assumption until the next ONS update will be that we are ~20% under the true number.

I am not sure what you are on about, but there is no correct expected value.  If you assume x% per day the current graph will be below exspected, if you assume y%, it will be above expected. 

Some reduction in the growth rate will just happen as infected people will be surrounded by more infected people over time (for example if you have a few people in an office, once most are infected, those infected later in the outbreak will have less people they can spread the virus to.   If your model doesn't factor this in, you should overshoot a bit.  Eventually the graph stops rising once big chunks of people in the area/country have already been exposed to the virus.

Also if you start from a small number, as you do with deaths, you can just get significantly different results depending on what day you start. 

[kukushkin88]

We’re back to this being a daily occurrence for now:

https://mobile.twitter.com/DHSCgovuk/status/1244976749940703233

A BBC reporter has it on his twitter, so it should be accurate?

https://mobile.twitter.com/lewis_goodall/status/1244985323660152832

And some of them had high blood pressure so were going to die soon anyway :/

I assume the big leap is probably just down to delayed reporting as we have just gone through a weekend.
 

It is a big leap from the day before but in terms of expected exponential growth it is still on the low side of what is expected. I guess the working assumption until the next ONS update will be that we are ~20% under the true number.

I am not sure what you are on about, but there is no correct expected value.  If you assume x% per day the current graph will be below exspected, if you assume y%, it will be above expected. 

Some reduction in the growth rate will just happen as infected people will be surrounded by more infected people over time (for example if you have a few people in an office, once most are infected, those infected later in the outbreak will have less people they can spread the virus to.   If your model doesn't factor this in, you should overshoot a bit.  Eventually the graph stops rising once big chunks of people in the area/country have already been exposed to the virus.

Also if you start from a small number, as you do with deaths, you can just get significantly different results depending on what day you start. 

We were told yesterday there’s a ~5 day lag in the numbers, so for a considerable time yet we will see numbers from the period when there were no control measures in place , if we look at all the data from the other countries that are further progressed in the epidemic spread, then doubling every ~3 days is the expected pattern. We are still tracking Italy from the 10 death mark.

The 20% that I mentioned is just the additional deaths the ONS published earlier today.

[Jamier-Host]

This is what i have been trying to find data for the past few weeks. I had found the ~1700 daily deaths number for UK (normally) but been hard to get a sensible estimate on what the actual uptick is / is predicted to be.

https://news.sky.com/video/covid-19-how-do-figures-of-virus-deaths-compare-to-a-normal-year-11964813

[kukushkin88]

Headline so far seems to be that we are, according to Jenny Harries now testing ~12,700 a day, it’s progress:

The latest Murta spreadsheet:

https://docs.google.com/spreadsheets/d/1eTKeK9vRxgw0KhvKxPCaDrfaHnxQP-n9TsLzsEymviY/htmlview?sle=true

[ripple11]

This is what i have been trying to find data for the past few weeks. I had found the ~1700 daily deaths number for UK (normally) but been hard to get a sensible estimate on what the actual uptick is / is predicted to be.

https://news.sky.com/video/covid-19-how-do-figures-of-virus-deaths-compare-to-a-normal-year-11964813

Interesting, saying 1/2 to 2/3 would have died this year anyway. So an approx excess of 8,000... similar to Flu.

[kukushkin88]

This is what i have been trying to find data for the past few weeks. I had found the ~1700 daily deaths number for UK (normally) but been hard to get a sensible estimate on what the actual uptick is / is predicted to be.

https://news.sky.com/video/covid-19-how-do-figures-of-virus-deaths-compare-to-a-normal-year-11964813

Interesting, saying 1/2 to 2/3 would have died this year anyway. So an approx excess of 8,000... similar to Flu.

For the Sky clip, they are working to best case with all the control measures in place, without them the IC model says deaths of ~500,000, so there isn’t a meaningful comparison to be made.

It’s for the US but it’s a pretty good explainer for why the impact it will have on healthcare provision means it nothing like the flu. They also die in a much more condensed period of time. Early evidence also suggests that flu has more co-morbidity type deaths than C19, the data for comparison purposes won’t be available for quite a while.

https://www.google.co.uk/amp/s/www.independent.co.uk/news/world/americas/coronavirus-map-hospital-beds-covid-19-ventilators-a9412026.html%3famp

[Doobs]

This is what i have been trying to find data for the past few weeks. I had found the ~1700 daily deaths number for UK (normally) but been hard to get a sensible estimate on what the actual uptick is / is predicted to be.

https://news.sky.com/video/covid-19-how-do-figures-of-virus-deaths-compare-to-a-normal-year-11964813

Interesting, saying 1/2 to 2/3 would have died this year anyway. So an approx excess of 8,000... similar to Flu.

20,000 isn't the deaths expected, it is a possible scenario, and deaths are only kept that low in that scenario if the UK's measures are really effective and some other optimistic assumptions are met.

This article is good.

https://www.wired.com/story/the-mathematics-of-predicting-the-course-of-the-coronavirus/

The problem is that if you produce a range of possibilities, the doom mongers will emphasise the one where 2 million die; and those that don't want the close down will emphasise the 20,000.  Of course the 20,000 is dependent on ongoing restrictions for a long time/until a succesful vaccine is available.  

The "revised" forecast was covered here by the man who said both 20,000 and 500,000. https://twitter.com/neil_ferguson/status/1243294815200124928

FWIW 20,000 seems pretty optiistic given we have just had nearly 400 deaths in a day, and the lag in the reporting means that is likely to be understated.  I hope it happens, but events are making it increasingly unlikely.

[Pokerpops]

This is what i have been trying to find data for the past few weeks. I had found the ~1700 daily deaths number for UK (normally) but been hard to get a sensible estimate on what the actual uptick is / is predicted to be.

https://news.sky.com/video/covid-19-how-do-figures-of-virus-deaths-compare-to-a-normal-year-11964813

Interesting, saying 1/2 to 2/3 would have died this year anyway. So an approx excess of 8,000... similar to Flu.

20,000 isn't the deaths expected, it is a possible scenario, and deaths are only kept that low in that scenario if the UK's measures are really effective and some other optimistic assumptions are met.

This article is good.

https://www.wired.com/story/the-mathematics-of-predicting-the-course-of-the-coronavirus/

The problem is that if you produce a range of possibilities, the doom mongers will emphasise the one where 2 million die; and those that don't want the close down will emphasise the 20,000.  Of course the 20,000 is dependent on ongoing restrictions for a long time/until a succesful vaccine is available.  

The "revised" forecast was covered here by the man who said both 20,000 and 500,000. https://twitter.com/neil_ferguson/status/1243294815200124928

FWIW 20,000 seems pretty optiistic given we have just had nearly 400 deaths in a day, and the lag in the reporting means that is likely to be understated.  I hope it happens, but events are making it increasingly unlikely.

I think we’d all do well to ignore all posts in this thread that aren’t authored by Doobs, or JonMW.

Including this one probably.