COVID-19

[lpm]

What the ever loving f.ck was that?
I know I’m not the sharpest, but I had no idea what he was on about. I’m going to have to find some analysis somewhere.

It's easy.

The middle classes can continue to work from home, and may now play golf, meet friends (2 metre distancing) and go on day trips to the seaside.

The working classes must now go out to work, not using public transport, but their employers will sort something out to make it safer, and if you don't like it you can quit.

[lpm]

Per the govt:

COVID Alert Level = R + Number of Infections.

The COVID Alert Level is currently 18,000.8.

[purplehaze]

What the ever loving f.ck was that?
I know I’m not the sharpest, but I had no idea what he was on about. I’m going to have to find some analysis somewhere.

It's easy.

The middle classes can continue to work from home, and may now play golf, meet friends (2 metre distancing) and go on day trips to the seaside.

The working classes must now go out to work, not using public transport, but their employers will sort something out to make it safer, and if you don't like it you can quit.

We know that social distancing applies only to the middle class because it's 2 metres. It would be in yards or feet for the working class.

[Little waster]

What the ever loving f.ck was that?
I know I’m not the sharpest, but I had no idea what he was on about. I’m going to have to find some analysis somewhere.

It's easy.

The middle classes can continue to work from home, and may now play golf, meet friends (2 metre distancing) and go on day trips to the seaside.

The working classes must now go out to work, not using public transport, but their employers will sort something out to make it safer, and if you don't like it you can quit.

We know that social distancing applies only to the middle class because it's 2 metres. It would be in yards or feet for the working class.

That's unfair it quite clearly says in the government advice that 2 metres is about the length of a chaise-longue or the equivalent of your butler holding up a copy of the Times.

[lpm]

It's in metres so one's au pair understands it. This enables people to work during school closures.

[Bird on a Fire]

Look forward to the freedumb protesters. They're on their way.

[Little waster]

Look forward to the freedumb protesters. They're on their way.

Presumably they were originally going to meet next to the fibreglass treehouse but some big Year 10s like to drink cider there and the picnic table bit underneath smells of tinkle.

[Woodchopper]

As of 9 May 2020, the Office for National Statistics (ONS) had received the results of swab tests collected from 7,087 individual participants in the coronavirus (COVID-19) Infection Survey in England between 26 April and 8 May 2020.

It is estimated that 0.24% of the population in England tested positive for COVID-19 (95% confidence interval: 0.14% to 0.40%).

It is estimated 136,000 people in England were currently infected with COVID-19 (95% confidence interval: 76,000 to 225,000).

https://www.ons.gov.uk/peoplepopulation ... d10may2020

[AnnaD]

How it reallly happened..

Source

[shpalman]

As of 9 May 2020, the Office for National Statistics (ONS) had received the results of swab tests collected from 7,087 individual participants in the coronavirus (COVID-19) Infection Survey in England between 26 April and 8 May 2020.

It is estimated that 0.24% of the population in England tested positive for COVID-19 (95% confidence interval: 0.14% to 0.40%).

It is estimated 136,000 people in England were currently infected with COVID-19 (95% confidence interval: 76,000 to 225,000).

https://www.ons.gov.uk/peoplepopulation ... d10may2020

Yesterday's number for total official cases was about 220,000, but there have been about 30,000 deaths. There still haven't been any official numbers for recoveries, but in Italy the number of recoveries only just overtook the number of current positives.

So this study on a pathetically small sample seems to have found a number which might be lower than the number of actual official positives?

[JellyandJackson]

Thanks brightonian and lpm, I feel much enlightened now.

Per the govt:

COVID Alert Level = R + Number of Infections.

The COVID Alert Level is currently 18,000.8.

Even my teens spotted this was a bit odd.

[Woodchopper]

As of 9 May 2020, the Office for National Statistics (ONS) had received the results of swab tests collected from 7,087 individual participants in the coronavirus (COVID-19) Infection Survey in England between 26 April and 8 May 2020.

It is estimated that 0.24% of the population in England tested positive for COVID-19 (95% confidence interval: 0.14% to 0.40%).

It is estimated 136,000 people in England were currently infected with COVID-19 (95% confidence interval: 76,000 to 225,000).

https://www.ons.gov.uk/peoplepopulation ... d10may2020

Yesterday's number for total official cases was about 220,000, but there have been about 30,000 deaths. There still haven't been any official numbers for recoveries, but in Italy the number of recoveries only just overtook the number of current positives.

So this study on a pathetically small sample seems to have found a number which might be lower than the number of actual official positives?

I assume it’s 136 000 cases today, whereas the 220 000 is the number of confirmed cases ever. Presumably all of the 136 000 will either be recovered or dead in a few weeks.

[shpalman]

Subtracting the known deaths and assuming half of what's left have recovered so far (roughly true in Italy) would get you to something of the order 100,000.

Which is within the confidence interval given by that small sample.

So there's something interesting in the fact that random sampling did not seem to find an order of magnitude more mild/asymptomatic cases, which similar studies found in Germany and Italy.

Still it's unacceptable that there are no official recovery numbers (the official number is stuck at 135 since ages ago and an unofficial count at https://github.com/datasets/covid-19 says 1001, and either of those would put the number of official UK Covid cases higher than the number just produced by that study).

[bob sterman]

Subtracting the known deaths and assuming half of what's left have recovered so far (roughly true in Italy) would get you to something of the order 100,000.

Which is within the confidence interval given by that small sample.

So there's something interesting in the fact that random sampling did not seem to find an order of magnitude more mild/asymptomatic cases, which similar studies found in Germany and Italy.

Johns Hopkins tracker data shows 62987 new cases diagnosed in the UK between April 26th and May 8th (the period of this random survey).

If the ONS data on population prevalence during this period (0.24%) can be extrapolated to the UK as a whole then about 160,000 would actually test postive during that period. Suggesting that total postive cases in the population during this period was about 2.5x the reported case numbers.

If actual cases numbers were only 2.5x reported numbers for the whole pandemic period in the UK - then this would obviously point to the CFR being really much worse than people are hoping.

BUT it's worth bearing in mind that this random swab study was carried out during a period when testing for people for symptoms was becoming much more widely available (the run up to the 100,000 per day deadline).

So it seems likely if this survey had been carried out earlier in the pandemic - the ratio of cases detected by random survey / cases detected through clinical testing would have been much higher.

[shpalman]

Subtracting the known deaths and assuming half of what's left have recovered so far (roughly true in Italy) would get you to something of the order 100,000.

Which is within the confidence interval given by that small sample.

So there's something interesting in the fact that random sampling did not seem to find an order of magnitude more mild/asymptomatic cases, which similar studies found in Germany and Italy.

Johns Hopkins tracker data shows 62987 new cases diagnosed in the UK between April 26th and May 8th (the period of this random survey).

If the ONS data on population prevalence during this period (0.24%) can be extrapolated to the UK as a whole then about 160,000 would actually test postive during that period. Suggesting that total positive cases in the population during this period was about 2.5x the reported case numbers.

If actual cases numbers were only 2.5x reported numbers for the whole pandemic period in the UK - then this would obviously point to the CFR being really much worse than people are hoping.

BUT it's worth bearing in mind that this random swab study was carried out during a period when testing for people for symptoms was becoming much more widely available (the run up to the 100,000 per day deadline).

So it seems likely if this survey had been carried out earlier in the pandemic - the ratio of cases detected by random survey / cases detected through clinical testing would have been much higher.

Ok thanks that does make a bit more sense. More than 60000 people were being tested per day (if you believe the official figures) during most of that period compared to less than 20000 people per day up until the 24th of April. The true number of cases must have been much higher a month ago and is closer to reality now.

Yeah the CFR is bad but we'll have to compare these numbers to the death rate over the next week or two, not the total who have died so far. If the actual official number of new cases per day is about 5000 but the true number is 2.5x that, 12500, and the death rate is about 500 and (slowly) falling, maybe you could get a CFR of better than 4%. Still difficult to get anywhere near Germany's CFR with these data.

(I'd still argue that the factor should be smaller than 2.5x, because on the 26th of April there were about 125,000 who had officially tested positive and not died and I suppose maybe less than half could have recovered, so the random testing window should have picked up part of the iceberg of mild/asymptomatic cases who hadn't been tested by then but hadn't recovered yet.)

[Little waster]

One from reddit via xkcd. SFW.

https://preview.redd.it/h2t2mj67k1y41.g ... cdf83939f3

[shpalman]

One from reddit via xkcd. SFW.

https://preview.redd.it/h2t2mj67k1y41.g ... cdf83939f3

You can find something similar but interactive here: https://aatishb.com/covidtrends/?locati ... ed+Kingdom

[tom p]

Why are only people arriving by air subject to quarantine? Are people travelling by boat somehow immune?

Because they're all squashed together on planes, whereas on a ferry....er.....

Actually, you're not squashed together on a ferry.
Ferries are f.cking massive and there's loads of space. I could travel from Hull to Hook of Holland for 12 hours and the only time I'd possibly be under 2m from other people would be at the lifts. It would be simple during loading and unloading times to have a staff member on each floor checking that nobody gets into an already-full lift and to change the stairs so that some from the car decks are up only & others down only.

Probably the reason it applies to airports only rather than train stations (well St Pancras international) or ports is because there there is infrastructure to isolate individuals and track them easily, whereas at ports it's mostly freight or groups of people in cars. Isolating them for tracing would be pretty tricky (although it shouldn't be beyond the wit of government to commandeer a space on each ferry and then do what's necessary to/with the passengers during the crossing).
There's not really the space for a holding pen at St Pancras either.
Also, the number of passengers on Eurostar (11 mil/year) is way less than in all the airports combined. Probably way less than in most individual airports (luton is 16 mil, stansted 20, heathrow 80, gatwick 46, manchester 29, birmingham 12, john lennon 5, forfar 4), so it makes more sense to go for the 95% rather than the 5%

[tom p]

What the ever loving f.ck was that?
I know I’m not the sharpest, but I had no idea what he was on about. I’m going to have to find some analysis somewhere.

It's easy.

The middle classes can continue to work from home, and may now play golf, meet friends (2 metre distancing) and go on day trips to the seaside.

The working classes must now go out to work, not using public transport, but their employers will sort something out to make it safer, and if you don't like it you can quit.

We know that social distancing applies only to the middle class because it's 2 metres. It would be in yards or feet for the working class.

Why would it be in yards and feet for the working classes? That doesn't make sense. It would be in yards & feet for the elderly.

[tom p]

Look forward to the freedumb protesters. They're on their way.

Good to see that Debenhmas is still open, but that park is called Hoglands.
Plus, why are only trans parents invited? I smell a rat, this is a blatant attempt to get trans people infected in order to wipe them out.

[Fishnut]

Because they're all squashed together on planes, whereas on a ferry....er.....

Actually, you're not squashed together on a ferry.
Ferries are f.cking massive and there's loads of space. I could travel from Hull to Hook of Holland for 12 hours and the only time I'd possibly be under 2m from other people would be at the lifts. It would be simple during loading and unloading times to have a staff member on each floor checking that nobody gets into an already-full lift and to change the stairs so that some from the car decks are up only & others down only.

Probably the reason it applies to airports only rather than train stations (well St Pancras international) or ports is because there there is infrastructure to isolate individuals and track them easily, whereas at ports it's mostly freight or groups of people in cars. Isolating them for tracing would be pretty tricky (although it shouldn't be beyond the wit of government to commandeer a space on each ferry and then do what's necessary to/with the passengers during the crossing).
There's not really the space for a holding pen at St Pancras either.
Also, the number of passengers on Eurostar (11 mil/year) is way less than in all the airports combined. Probably way less than in most individual airports (luton is 16 mil, stansted 20, heathrow 80, gatwick 46, manchester 29, birmingham 12, john lennon 5, forfar 4), so it makes more sense to go for the 95% rather than the 5%

tbh I wasn't thinking ferries as much as cargo/fishing vessels. Sure, the social distancing on the vessels will probably be easy to achieve (depending on vessel) but when you're in dock I'm not quite so sure. The very first boat I worked on had just finished an outbreak of flu before I boarded, which I then caught. Having flu while also finding your sea legs is not something I recommend.

[Bird on a Fire]

Look forward to the freedumb protesters. They're on their way.

Good to see that Debenhmas is still open, but that park is called Hoglands.
Plus, why are only trans parents invited? I smell a rat, this is a blatant attempt to get trans people infected in order to wipe them out.

I'm not strictly a Southamptonite but my aunts live 5 minutes from that park, and my granny did until recently. I never knew its name (and there are precious few hogs around these days). "Skate park near Debenhams" is how I think of it too. Probably the best bit on that poster, given that they can't even spell "next".

[tom p]

Look forward to the freedumb protesters. They're on their way.

Good to see that Debenhmas is still open, but that park is called Hoglands.
Plus, why are only trans parents invited? I smell a rat, this is a blatant attempt to get trans people infected in order to wipe them out.

I'm not strictly a Southamptonite but my aunts live 5 minutes from that park, and my granny did until recently. I never knew its name (and there are precious few hogs around these days). "Skate park near Debenhams" is how I think of it too. Probably the best bit on that poster, given that they can't even spell "next".

Oh yeah, I would initially think of it as the park next to Debenhams before thinking of it as Hoglands (unless I was playing cricket or football, then the team captain will reference the correct name of the park), but yeah, the "next" typo was priceless

[AMS]

One further reason for differentiating between ferries/trains and air travel for quarantine is where people are arriving from.

Ignoring the variations in specific countries' infection rates, anyone arriving in the UK by train or ferry has arrived from a country that has enacted a stricter lockdown than us and has a more coherent plan for easing it. Whereas air travel also includes travellers from the US...

[lpm]

Yes, it should be like air pressure differences in a BSL4 lab. Low risk can leak into higher risk.

We don't want people coming here from failed states like Russia, Brazil, USA, Iran. Fine from properly run states like EU, South Korea.