Page 142 of 258
Re: COVID-19
Posted: Fri Jun 05, 2020 8:50 pm
by AMS
lpm wrote: ↑Fri Jun 05, 2020 8:45 pm
SW is odd - very low Devon, very high Somerset.
Somerset has a lot more commuting into Bristol? Devon is more of a retirement zone, so probably fewer people who go skiing? Far SW is mostly tourism and agriculture, so few visitors in February?
Re: COVID-19
Posted: Fri Jun 05, 2020 9:02 pm
by Bird on a Fire
It's worth looking at the measures of uncertainty as well, which are reported on the writeup here:
https://epiforecasts.io/covid/posts/nat ... d-kingdom/
The credible intervals for the country's R value are 0.9-1.0, and the uncertainty is likely greater for smaller (regional) samples. They don't seem to report the numbers, but you can see in this plot:
- index.png (169.3 KiB) Viewed 4870 times
that small differences in the centre of the estimate are swamped by the relatively large amount of uncertainty. Five regions are running hot enough that we can't be sure if R is above or below 1.
The sensible precautionary approach would be to make sure error propagation is being done correctly, and then ensure that the confidence/credible intervals
exclude 1, rather than assuming the central estimate is correct and ignoring the uncertainty.
To communicate with the public, I'd report "R estimated to be below
x" where
x is the upper credible limit. Otherwise people hear 0.95 and think "oh that's good" and don't realise it really means "anywhere between 0.9 and 1.05".
Re: COVID-19
Posted: Fri Jun 05, 2020 9:05 pm
by Bird on a Fire
Further to the above, here's what they say in their
paper about the data sources:
We use daily counts of confirmed cases reported by the European Centre for Disease Control for all analyses conducted at the national level12,13. To estimate the delay from symptom onset to reporting (once confirmed with a positive laboratory test), we use all cases from a publicly available linelist for which onset and notification dates are available13,14. This linelist combines all known linelist data from over 100 countries and at the time of writing has 4,132 entries with both an onset date and a notification date. Countries are only included in the reported estimates if at least 60 cases have been reported in a single day. This restriction reduces the likelihood of spurious estimates for countries with limited transmission or case ascertainment.
For sub-national analyses, the source of the data is reported on the respective page on our website. The data are fetched from government departments or from individuals who maintain a data source if no official data are available. Subnational entities within countries are only reported if at least 40 cases have been reported in a single day. A lower limit is possible for sub-national compared to national data due to more consistent case reporting in the source datasets.
Those credible intervals are derived from the statistical analysis only, and don't seem to include the uncertainty from actual sampling effects (or lack thereof). If there are regional or temporal changes in case-report data that will also affect the estimates of R in a way the analytical results won't indicate.
ETA no need to take my word for it - here's what the authors say:
The results presented here are sensitive to changes in COVID-19 testing practices and the level of effort put into detecting COVID-19 cases, e.g. through contact tracing. For example, if numbers of incident infections remain constant but a country begins to find and report a higher proportion of cases, then an increasing value of the reproduction number will be inferred. This is because all changes in the number of cases are attributed to changes in the number of infections resulting from previously reported cases, and are not assumed to be a result of improved testing and surveillance. On the other hand, if a country reports a lower proportion of cases because a lower number of tests are performed (which can happen if reagents required for testing are no longer available, for example) or the surveillance system captures a lower proportion of infections, then the model will attribute this to a drop in the reproduction number that may not be a true reduction. In order for our estimates to be unbiased not all cases have to be reported, but the level of testing effort (and therefore the proportion of detected cases) must be constant27. This means that, whilst a change in testing effort will initially introduce bias, this will be reduced over time as long as the testing effort remains consistent from this point onwards.
Countries may also change the focus of their surveillance over the course of the outbreak. They may initially focus on identifying travellers returning from areas of known COVID-19 transmission and performing contact tracing on the contacts of known cases. As the outbreak evolves this may change to passive surveillance at hospitals. Here, the case definition may also change from tests based on polymerase chain reaction (PCR) to diagnoses based on symptoms and computed tomography (CT) scans. In the future, different kinds of COVID-19 tests may be deployed that could influence results, such as tests that detect both active and past infections.
So I guess the question for the UK context is, has contact tracing improved sufficiently to be inflating the estimates of R, or is it still totally sh.t and missing loads of asymptomatic cases?
Re: COVID-19
Posted: Sat Jun 06, 2020 7:12 am
by Woodchopper
Similarly worrying picture from the latest nowcast from Cambridge/Public Health England
Looks like r is close to 1 in several regions, and slightly over in the North West.
https://www.mrc-bsu.cam.ac.uk/now-casting/
They’re also estimating an IFR of 23% in the over 75s.
Only 10% in England infected so far. So herd immunity is a long way off.
From here:
https://www.mrc-bsu.cam.ac.uk/tackling- ... -covid-19/
Re: COVID-19
Posted: Sat Jun 06, 2020 11:43 am
by Woodchopper
Re: COVID-19
Posted: Sat Jun 06, 2020 6:04 pm
by FairySmall
AMS wrote: ↑Fri Jun 05, 2020 8:50 pm
lpm wrote: ↑Fri Jun 05, 2020 8:45 pm
SW is odd - very low Devon, very high Somerset.
Somerset has a lot more commuting into Bristol? Devon is more of a retirement zone, so probably fewer people who go skiing? Far SW is mostly tourism and agriculture, so few visitors in February?
Yes, we are odd in the SW. Although Bristol has relatively low figures for a city (last time I checked, Bath too). So I doubt it's only that. Somerset is the county that gave us Jacob Rees Mogg tho, so there's extra weird going on there.
Re: COVID-19
Posted: Sat Jun 06, 2020 7:48 pm
by Woodchopper
FairySmall wrote: ↑Sat Jun 06, 2020 6:04 pm
AMS wrote: ↑Fri Jun 05, 2020 8:50 pm
lpm wrote: ↑Fri Jun 05, 2020 8:45 pm
SW is odd - very low Devon, very high Somerset.
Somerset has a lot more commuting into Bristol? Devon is more of a retirement zone, so probably fewer people who go skiing? Far SW is mostly tourism and agriculture, so few visitors in February?
Yes, we are odd in the SW. Although Bristol has relatively low figures for a city (last time I checked, Bath too). So I doubt it's only that. Somerset is the county that gave us Jacob Rees Mogg tho, so there's extra weird going on there.
The
latest surveillance report suggests that the infection is now spreading in Bristol.
That can be combined with the estimate that the South West now has a reproduction rate which is slightly over one (see earlier link).
Re: COVID-19
Posted: Sat Jun 06, 2020 7:50 pm
by Fishnut
FairySmall wrote: ↑Sat Jun 06, 2020 6:04 pm
AMS wrote: ↑Fri Jun 05, 2020 8:50 pm
lpm wrote: ↑Fri Jun 05, 2020 8:45 pm
SW is odd - very low Devon, very high Somerset.
Somerset has a lot more commuting into Bristol? Devon is more of a retirement zone, so probably fewer people who go skiing? Far SW is mostly tourism and agriculture, so few visitors in February?
Yes, we are odd in the SW. Although Bristol has relatively low figures for a city (last time I checked, Bath too). So I doubt it's only that. Somerset is the county that gave us Jacob Rees Mogg tho, so there's extra weird going on there.
Just to nitpick, it was North East Somerset that gave us Jacob Rees Mogg. North Somerset gave us Liam Fox, and Weston-super-Mare gave us Roger Penrose. It's a trifecta of fuckwits, all with over 50% of the vote (50.4% down 3.3%; 52.9% down 2.9% and 57.5%
up 4.4% respectively at the last election).
Re: COVID-19
Posted: Sat Jun 06, 2020 8:57 pm
by FairySmall
Fishnut wrote: ↑Sat Jun 06, 2020 7:50 pm
Just to nitpick, it was North East Somerset that gave us Jacob Rees Mogg. North Somerset gave us Liam Fox, and Weston-super-Mare gave us Roger Penrose. It's a trifecta of fuckwits, all with over 50% of the vote (50.4% down 3.3%; 52.9% down 2.9% and 57.5%
up 4.4% respectively at the last election).
Nitpick away, I was being liberal in my definition of Somerset. Still think you proved my point about it being extra weird!
Oh nuts about cases in Bristol increasing. And I thought we were maintaining our anarchist credentials by not listening to the government and actually being sensible.
Re: COVID-19
Posted: Sat Jun 06, 2020 10:53 pm
by sTeamTraen
Fahhhhhkkkkk, that's huge.
Re: COVID-19
Posted: Sat Jun 06, 2020 10:58 pm
by sTeamTraen
Herainestold wrote: ↑Fri Jun 05, 2020 7:30 pm
Article on the experiment that is being done in Amerika of protesting during a pandemic.
https://thedispatch.com/p/were-about-to ... -about-how
Here’s the good news: Two weeks from now, we’ll know a lot more about how easily COVID-19 spreads outdoors than we do right now.
After a week of mammoth crowds clustering in cities across the nation to protest the death of Minneapolis resident George Floyd at the hands of police, many have asked the obvious question: Hey, aren’t we still in the middle of a pandemic? After two months of shutdown managed to slow the advance of the coronavirus—but not to beat it back—a sudden deluge of crowds have led to understandable fears of a resurgence.
It's widely believed in Spain that the virus got a nice boost from a big women's rights march in Madrid on 8 March, 5 days before the lockdown.
Re: COVID-19
Posted: Sun Jun 07, 2020 12:15 am
by jimbob
sTeamTraen wrote: ↑Sat Jun 06, 2020 10:58 pm
Herainestold wrote: ↑Fri Jun 05, 2020 7:30 pm
Article on the experiment that is being done in Amerika of protesting during a pandemic.
https://thedispatch.com/p/were-about-to ... -about-how
Here’s the good news: Two weeks from now, we’ll know a lot more about how easily COVID-19 spreads outdoors than we do right now.
After a week of mammoth crowds clustering in cities across the nation to protest the death of Minneapolis resident George Floyd at the hands of police, many have asked the obvious question: Hey, aren’t we still in the middle of a pandemic? After two months of shutdown managed to slow the advance of the coronavirus—but not to beat it back—a sudden deluge of crowds have led to understandable fears of a resurgence.
It's widely believed in Spain that the virus got a nice boost from a big women's rights march in Madrid on 8 March, 5 days before the lockdown.
Whilst a 2m separation might be reasonable in normal situations, in a large crowd, with the potential for multiple infected people overlaying their own contribution, especially in if moving, I'd have thought it would need to be bigger
Re: COVID-19
Posted: Sun Jun 07, 2020 9:15 am
by Woodchopper
sTeamTraen wrote: ↑Sat Jun 06, 2020 10:53 pm
Fahhhhhkkkkk, that's huge.
Yes, I just checked to make sure hadn't made a mistake. Would be useful to see how it compares to other countries.
Its probably higher than that. See the earlier post about how the ONS assumes that the most likely explanation for excess deaths where Covid isn't mentioned on the death certificate is undiagnosed Covid infection in the elderly (basically people dying in care homes and at home who may never even see a doctor).
Re: COVID-19
Posted: Sun Jun 07, 2020 9:24 am
by Trinucleus
How to turn a negative to your advantage:
Andrew Marr "Matt Hancock, how many people have you contacted through track and trace?"
"I can't tell you that Andrew, as we don't want to give out figures until we know the ONS will be happy with the information"
Re: COVID-19
Posted: Mon Jun 08, 2020 4:06 am
by basementer
As of midnight tonight, things go back to normal within NZ.
Not really - there will still be strict border controls, quarantine, and everyone is urged to keep close track of every shop, pub or office they visit every day so that we can stamp on any reimported cases. In March I was thinking by now there'd be five figures of cases, three or four figures of deaths, and stringent measures would carry on until mid/late July. We're lucky in our bubble, but any prick could burst it.
Re: COVID-19
Posted: Mon Jun 08, 2020 5:00 am
by Bird on a Fire
basementer wrote: ↑Mon Jun 08, 2020 4:06 am
We're lucky in our bubble, but any prick could burst it.
This is the most Kiwi coronavirus slogan imaginable.
"Don't be a c.nt c.nt, f.ck off home!" is a close contender.
Re: COVID-19
Posted: Mon Jun 08, 2020 10:13 am
by basementer
Bird on a Fire wrote: ↑Mon Jun 08, 2020 5:00 am
basementer wrote: ↑Mon Jun 08, 2020 4:06 am
We're lucky in our bubble, but any prick could burst it.
This is the most Kiwi coronavirus slogan imaginable.
If it's OK with you, I'd like to cite this when I apply for citizenship.
Re: COVID-19
Posted: Mon Jun 08, 2020 10:14 am
by FlammableFlower
FairySmall wrote: ↑Sat Jun 06, 2020 6:04 pm
AMS wrote: ↑Fri Jun 05, 2020 8:50 pm
lpm wrote: ↑Fri Jun 05, 2020 8:45 pm
SW is odd - very low Devon, very high Somerset.
Somerset has a lot more commuting into Bristol? Devon is more of a retirement zone, so probably fewer people who go skiing? Far SW is mostly tourism and agriculture, so few visitors in February?
Yes, we are odd in the SW. Although Bristol has relatively low figures for a city (last time I checked, Bath too). So I doubt it's only that. Somerset is the county that gave us Jacob Rees Mogg tho, so there's extra weird going on there.
My take is that the SW was largely spared early on when the pandemic arrived and whilst they are now coming down it's spreading into other areas that are much less prepared. Much like in the US, thinking about it.
Whilst Bristol may considered a big city (especially for the region), it comes in about 11th in the country. With comparatively low levels of infection a lot of people in the region have a) been lucky and b) got rather relaxed about things. When you add in that now we've got the easing of the lockdown, it means there's ample opportunity for the virus to spread.
Re: COVID-19
Posted: Mon Jun 08, 2020 11:59 am
by lpm
Indications Florida has started secondwaving. Seems pretty straightforward - reduce lockdown without replacing it with something like test & trace, R>1, second wave.
Although Florida's unlockdown is much more unlocked than the UK's unlockdown.
Re: COVID-19
Posted: Mon Jun 08, 2020 1:42 pm
by Woodchopper
lpm wrote: ↑Mon Jun 08, 2020 11:59 am
Indications Florida has started secondwaving. Seems pretty straightforward - reduce lockdown without replacing it with something like test & trace, R>1, second wave.
Although Florida's unlockdown is much more unlocked than the UK's unlockdown.
That's bad. There's a huge number of retirees in Florida.
Re: COVID-19
Posted: Mon Jun 08, 2020 1:44 pm
by FlammableFlower
Re: COVID-19
Posted: Mon Jun 08, 2020 2:38 pm
by Gentleman Jim
My wife, in common with many americans (at least that she knows)*, still think it is a biological warfare virus
*Other US stereotypes are available
Re: COVID-19
Posted: Mon Jun 08, 2020 2:49 pm
by Opti
tbh, if this is meant to be a biological weapon ... it's been pretty underwhelming.
OTOH, if it is, a virus that is a lot less infectious, and a lot less deadly than many viruses ... look at the havoc even a useless bio-weapon has wrought.
The whole world order is exceedingly fragile at the moment.
It don't need a yuge thing ... just a little spanner in the works.
Re: COVID-19
Posted: Mon Jun 08, 2020 3:47 pm
by veravista
If it's a bio weapon then whoever did it learned from the IRA. Want to close down the country? Use the freephone number on Friday and tell the authorities that there is a bomb under a bridge on the M1.
Instant lockdown for free.
Re: COVID-19
Posted: Mon Jun 08, 2020 4:10 pm
by tom p
Woodchopper wrote: ↑Mon Jun 08, 2020 1:42 pm
lpm wrote: ↑Mon Jun 08, 2020 11:59 am
Indications Florida has started secondwaving. Seems pretty straightforward - reduce lockdown without replacing it with something like test & trace, R>1, second wave.
Although Florida's unlockdown is much more unlocked than the UK's unlockdown.
That's bad. There's a huge number of retirees in Florida.
For the world it's not so bad though - many of them are despicable racists and if they die before November, then T'rump might not get in (he only won Florida by 100k votes last time and it would have halved his electoral college victory)