COVID-19

Discussions about serious topics, for serious people
Locked
User avatar
Woodchopper
Princess POW
Posts: 7043
Joined: Sat Oct 12, 2019 9:05 am

Re: COVID-19

Post by Woodchopper » Wed Mar 25, 2020 12:38 pm

sTeamTraen wrote:
Wed Mar 25, 2020 11:56 am
In the meantime, though, it seems to me that it ought to be easy to test empirically: Just pick 1,000 people at random (as for an opinion poll) and see how many positive results you get (assuming there is a test that is sufficiently sensitive to detect asymptomatic/early stage cases). 30 million out of 66 million infected is basically Remain v Leave, so 1,000 or maybe 2,000 people tops ought to be enough.
As far as I'm aware there isn't yet a sufficiently accurate antibody test. So we can't yet reliably test to see whether someone has had the disease in the past. But a lot of work is being put into developing a test, so it may be available soon.

User avatar
Woodchopper
Princess POW
Posts: 7043
Joined: Sat Oct 12, 2019 9:05 am

Re: COVID-19

Post by Woodchopper » Wed Mar 25, 2020 12:56 pm

lpm wrote:
Wed Mar 25, 2020 12:30 pm
lpm wrote:
Wed Mar 25, 2020 12:14 pm
sTeamTraen wrote:
Wed Mar 25, 2020 11:56 am
Apparently a new model from Oxford [Dropbox, PDF] suggests that up to 30 million people could be infected in the UK, with the overwhelming majority having no symptoms.

I note that the people retweeting and otherwise spreading news of this study the most enthusiastically are those of a free-market political persuasion, presumably because if it's true the UK will reach herd immunity levels in about a week and everyone can go back to Greggs. :roll:

I'm trying to work out in my head how we would expect the number of new cases or deaths to evolve if this model were correct. In the meantime, though, it seems to me that it ought to be easy to test empirically: Just pick 1,000 people at random (as for an opinion poll) and see how many positive results you get (assuming there is a test that is sufficiently sensitive to detect asymptomatic/early stage cases). 30 million out of 66 million infected is basically Remain v Leave, so 1,000 or maybe 2,000 people tops ought to be enough.
Like with Italy, this theory would basically imply no hot spots. There would be infections in every single town and village in the UK, with a very small percentage needing hospital treatment and a smaller proportion needing ICUs. Geographically it would be almost perfectly flat. But what we see is overloaded hospitals in north London exporting cases to little used hospitals elsewhere.

And the UK wouldn't be exceptional. You would likewise see flatter geography in the US, not huge peaks in New York vs normal activity in Michigan.
For example, if half the population on average is infected, you wouldn't expect one county to be at 75% and the neighbour at 25%. 50% would surely result in averaging out of peaks and troughs.

For example, Kent, Surrey and Hampshire. Why would they diverge from each other in a 50% scenario? Why would known cases be different?

Here's the current cases per million population:

Kent 45
Surrey 110
Hampshire 207
I agree in general, but there would be hot spots for mortality and people needing hospital treatment. But those hotspots would be correlated with the presence of at risk populations. For example, towns or villages with relatively large populations of retirees. ETA which isn't what we are seeing - eg a hotspot in New York which has a comparatively young population.

User avatar
shpalman
Princess POW
Posts: 8225
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond
Contact:

Re: COVID-19

Post by shpalman » Wed Mar 25, 2020 12:57 pm

headshot wrote:
Wed Mar 25, 2020 12:01 pm
Do any of you clever people have opinions on this Twitter thread that’s doing the rounds: https://twitter.com/mac_puck/status/124 ... 39456?s=21
The model is reasonable. The subtlety is that the infectious period effectively ends when an infection is detected and the person is isolated one way or another, so it doesn't really matter if a person continues to shed viruses after that.

The other subtlety is that the number of cases we have is the number of cases considered serious enough to be worth testing, so we still don't have very good statistics on the number of mild/asymptomatic carriers at large in the general population. This is why everyone needs to simultaneously behave as if they are infectious and everyone else is susceptible, and at the same time that they are susceptible and everyone else is infectious.

If you let a SIR model carry on running with an R0 > 1 it will get to a substantial fraction of the population sooner or later.
having that swing is a necessary but not sufficient condition for it meaning a thing
@shpalman@mastodon.me.uk

User avatar
lpm
Junior Mod
Posts: 5932
Joined: Mon Nov 11, 2019 1:05 pm

Re: COVID-19

Post by lpm » Wed Mar 25, 2020 1:21 pm

Woodchopper wrote:
Wed Mar 25, 2020 12:56 pm
lpm wrote:
Wed Mar 25, 2020 12:30 pm
lpm wrote:
Wed Mar 25, 2020 12:14 pm

Like with Italy, this theory would basically imply no hot spots. There would be infections in every single town and village in the UK, with a very small percentage needing hospital treatment and a smaller proportion needing ICUs. Geographically it would be almost perfectly flat. But what we see is overloaded hospitals in north London exporting cases to little used hospitals elsewhere.

And the UK wouldn't be exceptional. You would likewise see flatter geography in the US, not huge peaks in New York vs normal activity in Michigan.
For example, if half the population on average is infected, you wouldn't expect one county to be at 75% and the neighbour at 25%. 50% would surely result in averaging out of peaks and troughs.

For example, Kent, Surrey and Hampshire. Why would they diverge from each other in a 50% scenario? Why would known cases be different?

Here's the current cases per million population:

Kent 45
Surrey 110
Hampshire 207
I agree in general, but there would be hot spots for mortality and people needing hospital treatment. But those hotspots would be correlated with the presence of at risk populations. For example, towns or villages with relatively large populations of retirees. ETA which isn't what we are seeing - eg a hotspot in New York which has a comparatively young population.
In the UK, testing is now correlated with hospitalisation, so known cases should be biased to infections of vulnerable people rather than infections of the general public.

There's a high proportion of pensioners in Devon, Dorset, East & West Sussex, Norfolk. Extending the above suggests these areas are currently well below average.

East & West Sussex: 52
Dorset 35
Devon 60
Norfolk 49

Overall averages

England 122
Scotland 107
Wales 152
N.Ireland 91

Data from https://www.theguardian.com/world/2020/ ... -your-area
⭐ Awarded gold star 4 November 2021

User avatar
Bird on a Fire
Princess POW
Posts: 10137
Joined: Fri Oct 11, 2019 5:05 pm
Location: Portugal

Re: COVID-19

Post by Bird on a Fire » Wed Mar 25, 2020 1:47 pm

lpm wrote:
Wed Mar 25, 2020 12:14 pm
sTeamTraen wrote:
Wed Mar 25, 2020 11:56 am
Apparently a new model from Oxford [Dropbox, PDF] suggests that up to 30 million people could be infected in the UK, with the overwhelming majority having no symptoms.

I note that the people retweeting and otherwise spreading news of this study the most enthusiastically are those of a free-market political persuasion, presumably because if it's true the UK will reach herd immunity levels in about a week and everyone can go back to Greggs. :roll:

I'm trying to work out in my head how we would expect the number of new cases or deaths to evolve if this model were correct. In the meantime, though, it seems to me that it ought to be easy to test empirically: Just pick 1,000 people at random (as for an opinion poll) and see how many positive results you get (assuming there is a test that is sufficiently sensitive to detect asymptomatic/early stage cases). 30 million out of 66 million infected is basically Remain v Leave, so 1,000 or maybe 2,000 people tops ought to be enough.
Like with Italy, this theory would basically imply no hot spots. There would be infections in every single town and village in the UK, with a very small percentage needing hospital treatment and a smaller proportion needing ICUs. Geographically it would be almost perfectly flat. But what we see is overloaded hospitals in north London exporting cases to little used hospitals elsewhere.

And the UK wouldn't be exceptional. You would likewise see flatter geography in the US, not huge peaks in New York vs normal activity in Michigan.
This is true, but can be accounted for quite straightforwardly in the analysis by stratified sampling. That is to say, divide the population into strata (such as NHS trusts, or counties, or distance bands from known hotspots, or whatever) and sample each unit randomly. Then to get your total estimate you weight the results from each stratum by how many people fall into that category.
We have the right to a clean, healthy, sustainable environment.

User avatar
shpalman
Princess POW
Posts: 8225
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond
Contact:

Re: COVID-19

Post by shpalman » Wed Mar 25, 2020 1:47 pm

lpm wrote:
Wed Mar 25, 2020 12:14 pm
sTeamTraen wrote:
Wed Mar 25, 2020 11:56 am
Apparently a new model from Oxford [Dropbox, PDF] suggests that up to 30 million people could be infected in the UK, with the overwhelming majority having no symptoms.

I note that the people retweeting and otherwise spreading news of this study the most enthusiastically are those of a free-market political persuasion, presumably because if it's true the UK will reach herd immunity levels in about a week and everyone can go back to Greggs. :roll:

I'm trying to work out in my head how we would expect the number of new cases or deaths to evolve if this model were correct. In the meantime, though, it seems to me that it ought to be easy to test empirically: Just pick 1,000 people at random (as for an opinion poll) and see how many positive results you get (assuming there is a test that is sufficiently sensitive to detect asymptomatic/early stage cases). 30 million out of 66 million infected is basically Remain v Leave, so 1,000 or maybe 2,000 people tops ought to be enough.
Like with Italy, this theory would basically imply no hot spots. There would be infections in every single town and village in the UK, with a very small percentage needing hospital treatment and a smaller proportion needing ICUs. Geographically it would be almost perfectly flat. But what we see is overloaded hospitals in north London exporting cases to little used hospitals elsewhere.

And the UK wouldn't be exceptional. You would likewise see flatter geography in the US, not huge peaks in New York vs normal activity in Michigan.
It's not saying that up to 30 million people could be infected now, it's modelling the possibility that the cases we see are only some small fraction of the total cases in the population. The question is not whether 30 million people will end up being infected (because if the social distancing doesn't work, this will happen), the question is how many of them would be serious enough to show up in the statistics and/or need to go to hospital.

You could still expect to see hotspots as long as the true number of infections was a relatively small fraction of the population, since the mild/asymptomatic cases would be in a cloud around the detected hotspots rather than completely evenly spread throughout the country.

For: Italy probably had an undetected epidemic before our official "patient 1" showed up; either that or Codogno was a surprisingly popular international work and tourism destination.

For: We could hope that Italy's CFR of ~10% would translate to a more reasonable IFR of ~1% if only 10% of cases were serious enough to be considered for testing. The UK's CFR seems similar to Hubei's right now, about 4.5% but Hubei (similar population to both the UK and Italy) has has registered about 70,000 infections in total. You could imagine that it wasn't the social distancing and confinement which limited that, but rather that only 0.1% of the cases showed up in testing and in reality just about everybody caught it.

Against: an entire small town near Padua was swabbed twice, and only about 3% of the population were found to have been infected. You would have to argue that half the population were swabbed while they were not producing enough of the virus to show up in the test.

Against: Italy has already overtaken Hubei for the number of deaths and could well reach ~30,000 which is an order of magnitude higher than what Hubei registered. So you'd be looking at ~15,000 deaths anyway in the UK given your lower CFR if the confinement and social distancing works. However, Italy's CFR was similarly low back when the number of cases was similar to what the UK has now, so you may well have a CFR of 10% to look forward to in the next couple of weeks as the number of cases increases by a factor of 100 and the ICUs fill up. Either that, or it's an artefact of only testing the most serious cases, which is something else that Italy and the UK have in common.
having that swing is a necessary but not sufficient condition for it meaning a thing
@shpalman@mastodon.me.uk

greyspoke
Fuzzable
Posts: 252
Joined: Mon Nov 11, 2019 4:36 pm

Re: COVID-19

Post by greyspoke » Wed Mar 25, 2020 1:49 pm

And the Wales figure means basically the South-East Wales conurbation:
https://www.walesonline.co.uk/news/uk-news/coronavirus-map-wales-uk-covid19-17969655?_ga=2.170813984.1228564543.1585143683-1709078515.1585143683
Quite a few unhealthy people in that arearound by here, but it is interesting that it hasn't taken off in Bristol to the same extent, though there will have been plenty of commuting between the two, and to London. Here's the heat map:
Image

User avatar
Woodchopper
Princess POW
Posts: 7043
Joined: Sat Oct 12, 2019 9:05 am

Re: COVID-19

Post by Woodchopper » Wed Mar 25, 2020 1:56 pm

greyspoke wrote:
Wed Mar 25, 2020 1:49 pm
And the Wales figure means basically the South-East Wales conurbation:
https://www.walesonline.co.uk/news/uk-news/coronavirus-map-wales-uk-covid19-17969655?_ga=2.170813984.1228564543.1585143683-1709078515.1585143683
Quite a few unhealthy people in that arearound by here, but it is interesting that it hasn't taken off in Bristol to the same extent, though there will have been plenty of commuting between the two, and to London. Here's the heat map:
Image
Also remarkable how few cases there are in the various bits of Yorkshire.

User avatar
sTeamTraen
After Pie
Posts: 2548
Joined: Mon Nov 11, 2019 4:24 pm
Location: Palma de Mallorca, Spain

Re: COVID-19

Post by sTeamTraen » Wed Mar 25, 2020 1:58 pm

headshot wrote:
Wed Mar 25, 2020 12:01 pm
Do any of you clever people have opinions on this Twitter thread that’s doing the rounds: https://twitter.com/mac_puck/status/124 ... 39456?s=21
Well, as he says, he's not an epidemiologist. But also, his implied claim that Johnson could have locked down the country on 18 February is, I think, difficult to sustain.

Imagine if the government --- any government, even Tony Blair's at the height of its popularity circa 1998 --- had tried to do so. I don't think anyone would have taken them seriously. As a concrete exercise, try (this is aimed at all readers, not just you, headshot!) scrolling back through this very thread to around that time and imagine how many people, even on here, would have gone "Oooh, yes, now you mention it, we should indeed close every pub and restaurant and tell 80% of people to work from home, starting today" on 18 February (or, if you prefer, any of the other dates mentioned in that Twitter thread). For example, on 22 February, shpalman reported that here was a total of 20 cases (not deaths) in the whole of Italy.

As a further exercise, ask yourself when you realised that a lockdown, with all its consequences (e.g., several million people losing their income at least temporarily) was inevitable. (As an aside, Remainers who argued that a 5% hit to the economy from Brexit would be TEOTWAWKI will at least have to acknowledge that there are circumstances in which an almost certainly bigger hit to GDP is necessary. This is not an argument for Brexit, but it might put things in context.)

Now, of course, you could argue that the evidence of looming catastrophe was available from China, and we didn't pay attention because of racist assumptions about poor hygiene or something like that. (Personally I think that racism may have played a part, but the fact that the Chinese government wasn't exactly forthcoming with footage of overwhelmed ICUs didn't help much either.) But in these situations, you can only play the hand you're dealt. I think there is no way on earth that any politician could have taken that decision then. In fact I don't think that any of the lockdowns that we've seen would have been politically possible more than a few days before they happened.

We can regret that we don't have scientifically literate populations who are able to understand exponential functions in such proportions that they can form a popular movement that would allow governments to take necessary action, but pretty much everything we've seen in terms of popular rejection of science (GMOs, anti-vaxx, glyphosate, nuclear power) and/or logic (Brexit, Trump, etc) over the past few years suggests that we have a way to go before we get to that point.
Something something hammer something something nail

User avatar
Woodchopper
Princess POW
Posts: 7043
Joined: Sat Oct 12, 2019 9:05 am

Re: COVID-19

Post by Woodchopper » Wed Mar 25, 2020 2:22 pm

From the latest ECDC report:
The four coronaviruses that are endemic in human populations are responsible for 10–15% of common cold infections and display a marked winter seasonality in temperate climates, with a peak between December and April, but are hardly detected in the summer months[61-64].The seasonality of coronaviruses might be driven, in part, by environmental conditions and host susceptibility, because coronaviruses are more stable under low and midrange relative humidity (20–50%) when the defence mechanisms of the airways are suppressed[65,66].However, based on preliminary analyses of the COVID-19 outbreak in Chinaand other countries, high reproductive numbers were observed not only in dry and cold districts but also in tropical districts with high absolute humidity, such as in Guangxi and Singapore[68].There is no evidence to date that SARS-CoV-2 will display a marked winter seasonality, such as other human coronaviruses inthe northern hemisphere, which emphasises the importance of implementing intervention measures such as isolation ofinfected individuals, workplace distancing, and school closures.
https://www.ecdc.europa.eu/en/publicati ... 9-pandemic

User avatar
Bird on a Fire
Princess POW
Posts: 10137
Joined: Fri Oct 11, 2019 5:05 pm
Location: Portugal

Re: COVID-19

Post by Bird on a Fire » Wed Mar 25, 2020 2:46 pm

Portugal still at 3-day doubling time after 1 week of 'national emergency', no unnecessary travel, businesses shut, working from home, etc.

The map of cases is largely a map of population density. Cases are mostly in Lisbon and Porto, which have higher populations but also are very touristic. Note other clusters in Algarve (tourists) and Coimbra (popular destination for international students, also touristy). The shutdown does seem to have largely avoided transferring the illness to the interior of the country, which is also largely rural with an ageing population (as young people have moved to Lisbon and Porto for work).

I'm hoping that in the next week or so we'll start to see a levelling off of new cases.

Image
We have the right to a clean, healthy, sustainable environment.

greyspoke
Fuzzable
Posts: 252
Joined: Mon Nov 11, 2019 4:36 pm

Re: COVID-19

Post by greyspoke » Wed Mar 25, 2020 3:39 pm

Well that Prof Ferguson has a slightly optimistic prediction.

But is getting some flack on twitter for not open-sourcing his code - "I wrote the code (thousands of lines of undocumented C) 13+ years ago to model flu pandemics..."

I assume that his publications on the topic gave the underlying algorithms but not the actual code.

User avatar
EACLucifer
Stummy Beige
Posts: 4177
Joined: Fri Dec 13, 2019 7:49 am
Location: In Sumerian Haze

Re: COVID-19

Post by EACLucifer » Wed Mar 25, 2020 3:47 pm

This is the same model that was resolutely not updated with real data coming in from China, Korea and Italy until well into March, so I'm less than reassured.

raven
Catbabel
Posts: 644
Joined: Mon Nov 11, 2019 8:58 pm

Re: COVID-19

Post by raven » Wed Mar 25, 2020 5:04 pm

lpm wrote:
Wed Mar 25, 2020 12:30 pm
For example, Kent, Surrey and Hampshire. Why would they diverge from each other in a 50% scenario? Why would known cases be different?

Here's the current cases per million population:

Kent 45
Surrey 110
Hampshire 207
When I was looking at the UK map a week or two ago, back when we hadn't restricted testing to the seriously ill, the home counties seemed to have more cases and I wondered if that was simply because more people there had come back from skiing hols in Italy/Austria.
As a further exercise, ask yourself when you realised that a lockdown, with all its consequences (e.g., several million people losing their income at least temporarily) was inevitable.
When I read about the British guy who triggered the cluster in France and was it Mijorca?, and the cluster in Germany caused by a lady from China who had no symptoms until she was on the plane back, but manage to pass it on to 3 or 4 people at the business she visited. Asymptomatic and infectious makes it very hard to contain, as does the 'mildly ill for a few days, feels well enough to go out again, but then suddenly critically ill' pattern that I've read about.

But I am perhaps a pessimist....

User avatar
jimbob
Light of Blast
Posts: 5266
Joined: Mon Nov 11, 2019 4:04 pm
Location: High Peak/Manchester

Re: COVID-19

Post by jimbob » Wed Mar 25, 2020 6:42 pm

OneOffDave wrote:
Wed Mar 25, 2020 11:13 am
FlammableFlower wrote:
Wed Mar 25, 2020 10:58 am
Really good More or Less "coronavirus special" today

The bit with David Spiegelhalter on the risk and mortality is very interesting - my understanding isn't great on this but getting the virus is essentially the equivalent of hitting you with an entire year's worth of risk of death at once. Hence if you're old or have a health problem you're likely to be at much higher chance of dying. It also underscores the Imperial College report - it's basically a whole year's-worth of deaths in a compressed time period which would overwhelm the system.
This has always been an interesting debate for as long as I've been an emergency planner. How many are "excess deaths" versus "early harvest". Most UK pandemic planning has always been around demand smoothing as opposed to eradication as it's perceived that total containment is effectively impossible. It can be done in small outbreaks like the imported Monkey Pox cases in the last couple of years and animal outbreaks where "stamping out" methods can be used but much harder otherwise
So basically, COVID-19 is The Machine?
Have you considered stupidity as an explanation

User avatar
sTeamTraen
After Pie
Posts: 2548
Joined: Mon Nov 11, 2019 4:24 pm
Location: Palma de Mallorca, Spain

Re: COVID-19

Post by sTeamTraen » Wed Mar 25, 2020 7:04 pm

raven wrote:
Wed Mar 25, 2020 5:04 pm
When I read about the British guy who triggered the cluster in France and was it Mijorca?
He and his family turned up in Mallorca and immediately reported to hospital. He tested positive, the others were negative. I think he probably didn't spread the virus much further.

I'm in Mallorca right now and we have one of the lowest infection rates in Spain. (I guess an island is quite likely to have either a lot more or a lot fewer cases than a large continental area.) The lockdown is pretty ferocious. As I write this, though, the nightly applause for healthcare workers is taking place, followed by 30 minutes of music from the balcony of a nearby apartment --- our building overlooks a valley leading down to the yacht harbour and we can see (and blast music at) several hundred other people's homes. Like many others, we have put our Christmas lights up!
Something something hammer something something nail

User avatar
lpm
Junior Mod
Posts: 5932
Joined: Mon Nov 11, 2019 1:05 pm

Re: COVID-19

Post by lpm » Wed Mar 25, 2020 7:05 pm

I still don't get why anyone is taking the 50% thing remotely seriously.

50% implies about 32 million now. At a couple of doublings a week, that's

8 million 18 March
2 million 11 March
500,000 4 March
125,000 26 Feb

These vast numbers a few weeks ago never showed up as positive in testing, implying testing was a dud. On 26 Feb there were 12 confirmed cases from 125,000 actual infections? 0.01%? Absurd. By 4 March we had tested 16,000 people but only managed to find 87 cases. Out of 500,000 infected people?
⭐ Awarded gold star 4 November 2021

User avatar
shpalman
Princess POW
Posts: 8225
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond
Contact:

Re: COVID-19

Post by shpalman » Wed Mar 25, 2020 7:51 pm

lpm wrote:
Wed Mar 25, 2020 7:05 pm
I still don't get why anyone is taking the 50% thing remotely seriously.

50% implies about 32 million now. At a couple of doublings a week, that's

8 million 18 March
2 million 11 March
500,000 4 March
125,000 26 Feb

These vast numbers a few weeks ago never showed up as positive in testing, implying testing was a dud. On 26 Feb there were 12 confirmed cases from 125,000 actual infections? 0.01%? Absurd. By 4 March we had tested 16,000 people but only managed to find 87 cases. Out of 500,000 infected people?
What they're actually claiming is "thousands" on 5 March when the first death was reported, to take into account a low IFR and time lag between infection and death.

There were actually about 41000 total positives on the 19th of March when they downloaded the Italian data, with no kind of inflection in the graph, so it ought to have been obvious that we weren't anywhere near 40 million Italians actually infected (because that would have put enough of a dent in the susceptible population that the rate of new infections would be slowing down a lot).

If they knew what they were doing with their model they could have reduced the contact parameter to deal with the introduced containment measures and carried on with the simulation, rather than just ignoring everything but the first 12 days of data.
having that swing is a necessary but not sufficient condition for it meaning a thing
@shpalman@mastodon.me.uk

User avatar
jimbob
Light of Blast
Posts: 5266
Joined: Mon Nov 11, 2019 4:04 pm
Location: High Peak/Manchester

Re: COVID-19

Post by jimbob » Wed Mar 25, 2020 8:03 pm

lpm wrote:
Wed Mar 25, 2020 7:05 pm
I still don't get why anyone is taking the 50% thing remotely seriously.

50% implies about 32 million now. At a couple of doublings a week, that's

8 million 18 March
2 million 11 March
500,000 4 March
125,000 26 Feb

These vast numbers a few weeks ago never showed up as positive in testing, implying testing was a dud. On 26 Feb there were 12 confirmed cases from 125,000 actual infections? 0.01%? Absurd. By 4 March we had tested 16,000 people but only managed to find 87 cases. Out of 500,000 infected people?
Yup
HMG Website wrote:As of 9am on 24 March 2020, a total of 90,436 people have been tested, of which 82,359 were confirmed negative and 8,077 were confirmed positive. 422 patients in the UK who tested positive for coronavirus (COVID-19) have died.
That suggests that of those suspected of being most likely to have COVID-19 about 90% have been negative. About 12% of the positive tests were today.

The Oxford figures seem very difficult to fit to any data.
Have you considered stupidity as an explanation

User avatar
shpalman
Princess POW
Posts: 8225
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond
Contact:

Re: COVID-19

Post by shpalman » Wed Mar 25, 2020 9:57 pm

The early phase of the COVID-19 outbreak in Lombardy, Italy https://arxiv.org/abs/2003.09320

It would have arrived here some time in January, I'm not sure if that fits with the timeline of coming from Webasto in Bavaria.
having that swing is a necessary but not sufficient condition for it meaning a thing
@shpalman@mastodon.me.uk

raven
Catbabel
Posts: 644
Joined: Mon Nov 11, 2019 8:58 pm

Re: COVID-19

Post by raven » Wed Mar 25, 2020 10:46 pm

sTeamTraen wrote:
Wed Mar 25, 2020 7:04 pm
He and his family turned up in Mallorca and immediately reported to hospital. He tested positive, the others were negative. I think he probably didn't spread the virus much further.
I think we're talking about different cases. Hang on a mo... goes to google...

The third UK case was a bloke from Brighton who'd been in Singapore & picked it up there unknowingly, then had flown to a family gathering in France, thus giving it to other family members who lived in France, and some who were over from Mallorca and I think had gone home before they realised. (That second family is probably who you mean, so we're talking about different links in the same chain.) The bloke from Brighton didn't know he had it until he returned to the Uk and was tested, but in that time he'd passed it on.

User avatar
shpalman
Princess POW
Posts: 8225
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond
Contact:

Re: COVID-19

Post by shpalman » Wed Mar 25, 2020 11:01 pm

Well, just a couple of days after Codogno "patient 1" was confirmed to be suffering from covid-19 we had about 60 official cases. It's unlikely they all caught it from him;* if you follow the exponential curve back you get the true first case about 12 days before. Minus another week or so for incubation gets us to the first few days of February. That does fit with the Bavarian cases at the end of January. That company (Webasto) in Munich also has various offices in the north of Italy and of course genetic analyses of the Bavarian and Codogno strains indicated they were quite similar.

* bear in mind though that he was symptomatic and worsening for at least a week before his wife persuaded them to suspect covid-19.
having that swing is a necessary but not sufficient condition for it meaning a thing
@shpalman@mastodon.me.uk

User avatar
shpalman
Princess POW
Posts: 8225
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond
Contact:

Re: COVID-19

Post by shpalman » Wed Mar 25, 2020 11:53 pm

https://www.tgcom24.mediaset.it/cronaca ... 002a.shtml

Andrea Crisanti, who ran the study in which the entire town of Vò got swabbed twice, is maybe best placed to know how many cases there actually are. He estimates 250,000 in Lombardy (of which 100,000 are asymptomatic) and 450,000 in the whole of Italy. This compared to the official number of 60,000.
having that swing is a necessary but not sufficient condition for it meaning a thing
@shpalman@mastodon.me.uk

User avatar
dyqik
Princess POW
Posts: 7485
Joined: Wed Sep 25, 2019 4:19 pm
Location: Masshole
Contact:

Re: COVID-19

Post by dyqik » Thu Mar 26, 2020 1:06 am

greyspoke wrote:
Wed Mar 25, 2020 3:39 pm
Well that Prof Ferguson has a slightly optimistic prediction.

But is getting some flack on twitter for not open-sourcing his code - "I wrote the code (thousands of lines of undocumented C) 13+ years ago to model flu pandemics..."

I assume that his publications on the topic gave the underlying algorithms but not the actual code.
Any publications based on undocumented and unreviewed code without unit tests should probably be retracted as unrepeatable and unverifiable. You might as well use instruments with no calibration or traceability to standards.

User avatar
Bird on a Fire
Princess POW
Posts: 10137
Joined: Fri Oct 11, 2019 5:05 pm
Location: Portugal

Re: COVID-19

Post by Bird on a Fire » Thu Mar 26, 2020 1:50 am

dyqik wrote:
Thu Mar 26, 2020 1:06 am
greyspoke wrote:
Wed Mar 25, 2020 3:39 pm
Well that Prof Ferguson has a slightly optimistic prediction.

But is getting some flack on twitter for not open-sourcing his code - "I wrote the code (thousands of lines of undocumented C) 13+ years ago to model flu pandemics..."

I assume that his publications on the topic gave the underlying algorithms but not the actual code.
Any publications based on undocumented and unreviewed code without unit tests should probably be retracted as unrepeatable and unverifiable. You might as well use instruments with no calibration or traceability to standards.
That's most of them, then. Publishing code is really quite a recent phenomenon (and it still isn't standard practice, nor is code necessarily reviewed).
We have the right to a clean, healthy, sustainable environment.

Locked