COVID-19

[FlammableFlower]

What's the view on the French report that samples from earlier cases (Dec 2nd) of "pneumonia" that tested negative at the time for flu have now tested positive* for COVID? (Particularly as the subjects hadn't left France for a while prior to their illness) Was it sloshing around as an increase in flu/pneumonia before it took off?

*apparently checked multiple times to avoid error.

[lpm]

It's not impossible that the spouse of a French airport worker caught it in December.

It's also not impossible that the positive tests run now are duds.

Out of two very unlikely possibilities, I'd go with testing error.

[lpm]

Neil Ferguson has resigned from the SAGE group. He was caught breaking the lockdown to do a bit of shagging.

Sex and money, always the classic motives for any crime.

[Holylol]

What's the view on the French report that samples from earlier cases (Dec 2nd) of "pneumonia" that tested negative at the time for flu have now tested positive* for COVID? (Particularly as the subjects hadn't left France for a while prior to their illness) Was it sloshing around as an increase in flu/pneumonia before it took off?

*apparently checked multiple times to avoid error.

Does it say explicitly that the case was from a sample collected on Dec 2nd?
Or does it say that they look at samples collected between the 2nd of December et the 16th of January, and found some positive, but possibly from late December?

[dyqik]

Neil Ferguson has resigned from the SAGE group. He was caught breaking the lockdown to do a bit of shagging.

Sex and money, always the classic motives for any crime.

Also an important precedent for when an MP is caught. Or when Farage gets stopped again.

[shpalman]

Oh and it's actually possible that within a few days, the UK will overtake Italy for the number of official covid-19 positives who have died.

And that is indeed what happened.

[shpalman]

What's the view on the French report that samples from earlier cases (Dec 2nd) of "pneumonia" that tested negative at the time for flu have now tested positive* for COVID? (Particularly as the subjects hadn't left France for a while prior to their illness) Was it sloshing around as an increase in flu/pneumonia before it took off?

*apparently checked multiple times to avoid error.

Does it say explicitly that the case was from a sample collected on Dec 2nd?
Or does it say that they look at samples collected between the 2nd of December et the 16th of January, and found some positive, but possibly from late December?

https://www.bbc.com/news/world-europe-52526554 says 27th December.

[FlammableFlower]

Damn phone. I had to make so many bl..dy corrections to that post!

[dyqik]

It would appear so:

https://twitter.com/NateSilver538/statu ... 18470?s=20

https://twitter.com/NateSilver538/statu ... 87682?s=20

EXN0eznWoAAITlP.jpeg

Possibly they meant "cubic splines" so that's OK then.

https://twitter.com/CT_Bergstrom/status ... 60615?s=19

Of course, today's death toll is already pushing the most pessimistic of those models, the IHME (5/4) model, presumably released yesterday.

Yep, we've now had more deaths today than any of the models predict. Even the one released yesterday. Obviously you should expect fluctuations about the running mean, so yesterday's (the most pessimistic model) isn't ruled out yet. But the rest look pretty silly.

And the most pessimistic model assumes continued social distancing and other measures. But (red) states are removing those measures.

[headshot]

A good friend of my dad has just died of Covid-19. That’s the first death of someone I know.

He was admitted to hospital a couple of weeks ago after suffering from uncontrollable hiccups. He tested negative twice, then exhibited symptoms of the virus and tested positive, so it looks like he picked it up whilst in hospital.

f.ck.

[Woodchopper]

Spike mutation pipeline reveals the emergence of a more transmissible form of SARS-CoV-2
https://www.biorxiv.org/content/10.1101 ... 1.full.pdf

[Woodchopper]

Lots on Covid and age
https://docs.google.com/document/d/1Wxa ... zxqFg/edit

[Lydia Gwilt]

Oh and it's actually possible that within a few days, the UK will overtake Italy for the number of official covid-19 positives who have died.

And that is indeed what happened.

Yebbut the BBC is contorting itself to show that it doesn't really mean that, and that Covid in Italy is still much worse in every way than the brilliantly handled situation in the UK. Pisses me off so much!

https://www.bbc.com/news/52530918

[lpm]

How can you criticise the UK government on this?

They deliberately decided to go higher than Italy in this first wave - delaying the lockdown, keeping schools open, keeping the lockdown looser etc - in order to have more immunity and hence an easier time in the second wave. And they succeeded.

It's absolutely typical of you lefty scum on this forum, accusing the government of incompetency even when they achieve their goals.

[Lydia Gwilt]

This time I was actually criticising the BBC - they seem to be hell-bent on demonstrating that you can twist like an eel while simultaneously being totally spineless.

I only don't criticise the government because I'm not sure that the appropriate words actually exist. Willing to give it a go though...

[Gentleman Jim]

Neil Ferguson has resigned from the SAGE group. He was caught breaking the lockdown to do a bit of shagging.

Sex and money, always the classic motives for any crime.

Does that mean a certain poster on this forum is now "Sage" bound?

[shpalman]

Oh and it's actually possible that within a few days, the UK will overtake Italy for the number of official covid-19 positives who have died.

And that is indeed what happened.

Yebbut the BBC is contorting itself to show that it doesn't really mean that, and that Covid in Italy is still much worse in every way than the brilliantly handled situation in the UK. Pisses me off so much!

https://www.bbc.com/news/52530918

Coronavirus: Can you compare the UK with Italy?

A rare counter-example to Betteridge's Law.

1. Population sizes: yes, because the UK is only 10% less populous than Italy and will probably catch up to deaths per population in about a week or so.

2. The two countries may not be counting the same way: yes, up until recently, the UK was counting in a worse way but they now seem to be the same.

3. Testing: yes, the UK currently has more tests per day that Italy because a load were all saves to be posted out at once to meet the arbitrary end-of-April target while Italy has been steadily ramping up capacity and has done 60% more swabs to test 50% more people so far.

4. What is the age and health of the population? Italy has fewer young people and more old people, because the UK tends to kill people before they get that old through sh.t public health and sh.t health provision, and still manages to have the same death toll.

5. What about population density and other factors? Well, what about it? It makes little sense to compare the hot spot in the province of Bergamo (Bergamo itself is not a big city and the high death toll was not even in the city itself) with the whole of London.

If we had numbers for cases in hospital and cases in intensive care which were as easily accessible for the UK as those I can get for Italy then we'd have a better sense about how close to "the shocking scenes of overcrowding that were seen in some Italian hospitals" the London hospitals were. We don't even seem to have official UK numbers for the number of cases who have recovered from the virus.

I have in mind (and may have posted in this thread) an estimate of something like 48,000 excess deaths in Italy compared to the past few years, but the graph (which may be in here somewhere too) ramped up and peaked somewhat ahead of the official covid-positive deaths and then appeared to match it on the way down. The UK peak isn't coming down.

[sTeamTraen]

1. Population sizes: yes, because the UK is only 10% less populous than Italy and will probably catch up to deaths per population in about a week or so.

10% more, shirley?

[shpalman]

Still, better than don't know best wait a few years.

Obviously for the next 5-10 years you're going to have lower death rates in 'flu season, because of all the people who would have lived another 5-10 years except covid killed them.

photo_2020-04-30_18-18-00.jpg (64.42 KiB) Viewed 3544 times

The graph I had in mind was this one, but it's only for Lombardy, and it's only from a Facebook post written by an astrophysicist. I think the estimate of 48000 excess deaths was my own, scaling this result to the whole of Italy.

[shpalman]

1. Population sizes: yes, because the UK is only 10% less populous than Italy and will probably catch up to deaths per population in about a week or so.

10% more, shirley?

Yes, sorry.

[lpm]

The FT tracker of excess deaths in the UK is about 50,000. UK and Italy are pretty similar right now - except, as you say, the current UK daily rates are frustratingly slow to come down. Definitely a fear we'll plateau much too high.

[PeteB]

Institute and Faculty of Actuaries (IFoA) saying 55,000 excess deaths.

[Woodchopper]

Institute and Faculty of Actuaries (IFoA) saying 55,000 excess deaths.

Similar to the FT estimate which is 53,800
https://twitter.com/ChrisGiles_/status/ ... 8958338048

Presumably because they are using the same data and similar methods.

[jimbob]

Institute and Faculty of Actuaries (IFoA) saying 55,000 excess deaths.

Yup, and it's simple maths.

ONS data for England and Wales deaths all causes is up to week 17.

There were 33600 more deaths in 2020 to end of week 17 (24th April) than the 5 year average. 23000 of which were weeks 16 and 17*.

It looks like the excess deaths are currently running above 11000, and would still be above that at the moment, even with a slight decline trop the top of tge plateau.

So for the end of this week (Friday, week 19) we can get a low-end estimate of 33600 plus 22000 for England and Wales alone. And for yesterday, take 4700 off that as only being 4/7 through the week.

Adding Scotland and Northern Ireland gets a simplistic, optimistic estimate to pretty close to that.



*There were 37000 excess deaths in April as a whole (the early part of the year was better than average).

[dyqik]

Does that mean they just fit a third-order polynomial in Excel?

It would appear so:

https://twitter.com/NateSilver538/statu ... 18470?s=20

https://twitter.com/NateSilver538/statu ... 87682?s=20

EXN0eznWoAAITlP.jpeg

Possibly they meant "cubic splines" so that's OK then.

https://twitter.com/CT_Bergstrom/status ... 60615?s=19

This is the most convincing reconstruction I've seen - cubic polynomial fit to log(deaths).

https://twitter.com/potatoffel/status/1 ... 09792?s=20