COVID-19

Covid-19 discussion, bring your own statistics
User avatar
shpalman
Light of Blast
Posts: 4912
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond

Re: COVID-19

Post by shpalman » Thu Jul 22, 2021 6:33 pm

Seems like the UK is peaking this week - is that the effect of schools closing for the summer? Deaths still going up so expect a peak of about 120 per day if this is the peak in cases. Hospitalizations are still going up but maybe with a longer doubling time.

Of course we'll see if it goes up again because of "freedom day".

Italy has decided to base its Yellow, Orange, and Red restriction levels not on case rates but on hospital occupancy: 10%, 20%, and 30% occupancy of intensive care respectively (or 20%, 30%, and 40% of all beds I think).

Some regions have already gone over 50/100,000/week which was the previous case threshold for Yellow but Lombardy only has 28 patients in intensive care (peaked over 1300 during the first wave, more than 900 in November and another peak a bit less than 900 in March this year). The total in hospital is 163.

Fair enough with covid being less serious because of vaccination but if ~10 times* fewer of the cases end up in hospital then set your case thresholds ten times higher; basing it on hospitalizations just adds another ~two weeks' lag.

* - may not even be such a strong factor, I mean, the recent UK numbers suggests it's more like a factor of 2-3 than of 10 and it only looks lower when cases are going up faster than people deteriorate.
molto tricky

User avatar
lpm
Stummy Beige
Posts: 3263
Joined: Mon Nov 11, 2019 1:05 pm

Re: COVID-19

Post by lpm » Thu Jul 22, 2021 8:00 pm

shpalman wrote:
Thu Jul 22, 2021 6:33 pm
Seems like the UK is peaking this week - is that the effect of schools closing for the summer? Deaths still going up so expect a peak of about 120 per day if this is the peak in cases. Hospitalizations are still going up but maybe with a longer doubling time.
It's a bit weird. 10 days after Wembley football and, as everyone expected, cases have... er... stalled.

State schools closed today/yesterday.
What ever happened to that Trump guy, you know, the one who was president for a bit?

User avatar
shpalman
Light of Blast
Posts: 4912
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond

Re: COVID-19

Post by shpalman » Thu Jul 22, 2021 8:10 pm

lpm wrote:
Thu Jul 22, 2021 8:00 pm
shpalman wrote:
Thu Jul 22, 2021 6:33 pm
Seems like the UK is peaking this week - is that the effect of schools closing for the summer? Deaths still going up so expect a peak of about 120 per day if this is the peak in cases. Hospitalizations are still going up but maybe with a longer doubling time.
It's a bit weird. 10 days after Wembley football and, as everyone expected, cases have... er... stalled.

State schools closed today/yesterday.
The football seems to have caused cases to shoot up in Lazio and... Veneto? ¯\_(ツ)_/¯
molto tricky

OffTheRock
Clardic Fug
Posts: 177
Joined: Mon Dec 02, 2019 6:52 pm

Re: COVID-19

Post by OffTheRock » Thu Jul 22, 2021 8:53 pm

lpm wrote:
Thu Jul 22, 2021 8:00 pm
shpalman wrote:
Thu Jul 22, 2021 6:33 pm
Seems like the UK is peaking this week - is that the effect of schools closing for the summer? Deaths still going up so expect a peak of about 120 per day if this is the peak in cases. Hospitalizations are still going up but maybe with a longer doubling time.
It's a bit weird. 10 days after Wembley football and, as everyone expected, cases have... er... stalled.

State schools closed today/yesterday.
Attendance last Thursday was 67.3% in secondary and 82.8% in primary. 14% of secondary absences were non-covid related. Most of that is probably parents taking kids out of school so they didn't end up isolating over booked holidays. Some state schools broke up last week and some of the Midlands ones even earlier. Although they'll be going back 23rd August. That might account for the drop. I'd imagine it might lead to fewer LFTs and confirmatory PCRs when they come back positive as well.

User avatar
lpm
Stummy Beige
Posts: 3263
Joined: Mon Nov 11, 2019 1:05 pm

Re: COVID-19

Post by lpm » Thu Jul 22, 2021 9:54 pm

That implies the schools effect is even stronger than we maybe thought. If there's a specific Midlands area with early summer hols then we could easily look at local case data.

Scotland has R well below 1, although their school hols combine with higher lockdownicity elsewhere such as no nightclubs.
What ever happened to that Trump guy, you know, the one who was president for a bit?

OffTheRock
Clardic Fug
Posts: 177
Joined: Mon Dec 02, 2019 6:52 pm

Re: COVID-19

Post by OffTheRock » Thu Jul 22, 2021 10:05 pm

Mostly East Midlands I think.

Leicestershire broke up on 14th July.
Leicester City on 9th July.

I thought Nottinghamshire/Derbyshire did too, but looks like they are still in until next week. Or end of this week assuming they've chucked a couple of INSETS in for Mon/Tues next week. Last day is 27th down here, but I think most schools have Mon/Tues as INSET.

User avatar
lpm
Stummy Beige
Posts: 3263
Joined: Mon Nov 11, 2019 1:05 pm

Re: COVID-19

Post by lpm » Thu Jul 22, 2021 10:19 pm

A quick last night check shows Leicester cases racing upwards.
What ever happened to that Trump guy, you know, the one who was president for a bit?

OffTheRock
Clardic Fug
Posts: 177
Joined: Mon Dec 02, 2019 6:52 pm

Re: COVID-19

Post by OffTheRock » Fri Jul 23, 2021 10:37 am

School LFT instead of isolation study is out. Not peer reviewed yet. It’s not going to matter much since they are abandoning isolation and LFTs from the start of the Autumn Term.

User avatar
lpm
Stummy Beige
Posts: 3263
Joined: Mon Nov 11, 2019 1:05 pm

Re: COVID-19

Post by lpm » Fri Jul 23, 2021 5:06 pm

WTF is wrong with the stats? Down from 52,000 a week ago to 36,000. Not plausible.

1) Collapse in testing rates, due to shortage of tests? Or people can't be bothered to test.

2) Something weird in the specimen date vs reporting date?

3) Spreadsheet error.

4) A freak convergence of every possible favourable factor: good weather so everyone outside, some schools closing, loads of pupils off school, everyone being voluntarily careful because we disagree with "let it rip", pingdemic nonsense communicating effectively how high cases are?

5) Just randomness.
What ever happened to that Trump guy, you know, the one who was president for a bit?

OffTheRock
Clardic Fug
Posts: 177
Joined: Mon Dec 02, 2019 6:52 pm

Re: COVID-19

Post by OffTheRock » Fri Jul 23, 2021 7:07 pm

6) Everybody’s been at home since the middle of last week because they’ve been caught by the pingdemic. So nobody’s been going out to be able to catch anything.

As implausible as that is, it might be fun trying to watch the people arguing that cases are going down also argue that too many people are isolating.

I suspect it’s more likely that the testing system has started to creak again. There’s a drop in tests, but there seem to be a number of respiratory viruses doing the rounds at the moment, which you would think would increase the testing rate while lowering the positivity rate.

User avatar
Bird on a Fire
Light of Blast
Posts: 6380
Joined: Fri Oct 11, 2019 5:05 pm
Location: Iceland

Re: COVID-19

Post by Bird on a Fire » Fri Jul 23, 2021 7:30 pm

If schools have stopped testing (because they're shut) and workplaces have stopped testing (because Freedom Day) and people have stopped lateral flowing themselves generally (ditto) you might see way fewer people taking tests.
He has the grace of a swan, the wisdom of an owl, and the eye of an eagle—ladies and gentlemen, this man is for the birds!

User avatar
shpalman
Light of Blast
Posts: 4912
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond

Re: COVID-19

Post by shpalman » Fri Jul 23, 2021 9:39 pm

molto tricky

User avatar
Woodchopper
Stummy Beige
Posts: 3670
Joined: Sat Oct 12, 2019 9:05 am

Re: COVID-19

Post by Woodchopper » Sat Jul 24, 2021 7:57 am

Estimates of 3.4 - 4.9 million excess deaths in India.
https://cgdev.org/sites/default/files/t ... ndemic.pdf

Herainestold
Dorkwood
Posts: 1159
Joined: Mon Nov 25, 2019 1:23 pm

Re: COVID-19

Post by Herainestold » Sat Jul 24, 2021 6:59 pm

Bird on a Fire wrote:
Fri Jul 23, 2021 7:30 pm
If schools have stopped testing (because they're shut) and workplaces have stopped testing (because Freedom Day) and people have stopped lateral flowing themselves generally (ditto) you might see way fewer people taking tests.
It seems unlikely that it is a real drop, considering circumstances around Freedom Day, rather more likely to be some kind of testing artifact.
Delta changes everything.

User avatar
Woodchopper
Stummy Beige
Posts: 3670
Joined: Sat Oct 12, 2019 9:05 am

Re: COVID-19

Post by Woodchopper » Sun Jul 25, 2021 2:48 am

lpm wrote:
Fri Jul 23, 2021 5:06 pm
WTF is wrong with the stats? Down from 52,000 a week ago to 36,000. Not plausible.

1) Collapse in testing rates, due to shortage of tests? Or people can't be bothered to test.

2) Something weird in the specimen date vs reporting date?

3) Spreadsheet error.

4) A freak convergence of every possible favourable factor: good weather so everyone outside, some schools closing, loads of pupils off school, everyone being voluntarily careful because we disagree with "let it rip", pingdemic nonsense communicating effectively how high cases are?

5) Just randomness.
1. Too early to talk about a trend. Previous waves had minor bumps on the way up and down.

2. Wait for hospitalisation data to see whether the numbers are due to fewer cases or lower reporting.

3. If it does turn out to be a trend:

- perhaps there was a football effect (with messy data).
- people have noticed the case numbers and are reacting by being more careful.
- there are now very few adults left without antibodies. A combination of antibodies plus some social distancing and mask wearing is enough to keep r down.
- wait for the influence of freedom day.

User avatar
snoozeofreason
Fuzzable
Posts: 303
Joined: Fri Nov 15, 2019 1:22 pm

Re: COVID-19

Post by snoozeofreason » Sun Jul 25, 2021 10:17 am

Woodchopper wrote:
Sun Jul 25, 2021 2:48 am
lpm wrote:
Fri Jul 23, 2021 5:06 pm
WTF is wrong with the stats? Down from 52,000 a week ago to 36,000. Not plausible.

1) Collapse in testing rates, due to shortage of tests? Or people can't be bothered to test.

2) Something weird in the specimen date vs reporting date?

3) Spreadsheet error.

4) A freak convergence of every possible favourable factor: good weather so everyone outside, some schools closing, loads of pupils off school, everyone being voluntarily careful because we disagree with "let it rip", pingdemic nonsense communicating effectively how high cases are?

5) Just randomness.
1. Too early to talk about a trend. Previous waves had minor bumps on the way up and down.

2. Wait for hospitalisation data to see whether the numbers are due to fewer cases or lower reporting.

3. If it does turn out to be a trend:

- perhaps there was a football effect (with messy data).
- people have noticed the case numbers and are reacting by being more careful.
- there are now very few adults left without antibodies. A combination of antibodies plus some social distancing and mask wearing is enough to keep r down.
- wait for the influence of freedom day.
It's been in free-fall for a week now, and I don't think we've seen bumps as bumpy as this before, so I think I'd rule out randomness. I'd guess possible remaining options include:

(a) Our plucky government has timed this with the calm self-possession of a rifleman waiting to see the whites of his enemy's eyes before opening fire, and scheduled freedom day so that it would occur exactly two days after an unprecedented hairpin turn in rates of infection.

(b) People aren't getting tested because their physical presence is now required at work, or invited at night clubs and other forms of entertainment, and a positive test would interfere with that.

I'd be more inclined to go with (b), but if that was the explanation then I would expect to see a similarly sharp fall in the number of tests carried out, and I can't see that in the data.
In six days the Lord made the heavens and the earth, the sea, and all that is in them. The human body was knocked up pretty late on the Friday afternoon, with a deadline looming. How well do you expect it to work?

User avatar
shpalman
Light of Blast
Posts: 4912
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond

Re: COVID-19

Post by shpalman » Sun Jul 25, 2021 10:34 am

Only a person's first positive test is counted so if someone who had it last year gets reinfected it doesn't show up.
molto tricky

User avatar
bob sterman
Snowbonk
Posts: 580
Joined: Mon Nov 11, 2019 10:25 pm

Re: COVID-19

Post by bob sterman » Sun Jul 25, 2021 10:47 am

shpalman wrote:
Sun Jul 25, 2021 10:34 am
Only a person's first positive test is counted so if someone who had it last year gets reinfected it doesn't show up.
I wonder if this applies to the hospitalisation and death data too?

Dashboard says on Deaths...

"Number of deaths of people who had had a positive test result for COVID-19 and died within 28 days of the first positive test."

So if you have COVID in 2020 you can't die of it this year!

(Although worth perhaps nothing that the magnitude of this effect is limited by the fact that prior to this wave of infections there had only been about 4.5 million officially recorded cases - so there are only 4.5 million people who cannot be recorded as a case if they get reinfected).

User avatar
shpalman
Light of Blast
Posts: 4912
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond

Re: COVID-19

Post by shpalman » Sun Jul 25, 2021 2:54 pm

molto tricky

Herainestold
Dorkwood
Posts: 1159
Joined: Mon Nov 25, 2019 1:23 pm

Re: COVID-19

Post by Herainestold » Sun Jul 25, 2021 5:33 pm

Delta changes everything.

User avatar
bob sterman
Snowbonk
Posts: 580
Joined: Mon Nov 11, 2019 10:25 pm

Re: COVID-19

Post by bob sterman » Sun Jul 25, 2021 8:57 pm

So for the past 3 days new reported cases have been down 30-40% on the same day 7 days previously.

Surely English schools closing must be the strongest candidate to explain such rapid fall? Soon to be counteracted by nightclubs opening??

User avatar
lpm
Stummy Beige
Posts: 3263
Joined: Mon Nov 11, 2019 1:05 pm

Re: COVID-19

Post by lpm » Sun Jul 25, 2021 10:00 pm

The lag between cause and case stats is 10 days? Thereabouts? A few days to catch, a couple more for testing, collation of data? Which means looking at two weeks ago to explain the fall that started a few days ago.

Football is a dud, it doesn't fit the timings.

Partial school closures could be a factor, but it'll be another week till the main end of term effect.

We are falling out the bottom of models. There's something about Delta we don't understand. For the early countries hit by Delta:

- It set India ablaze and then suddenly fizzled, with partial lockdown not being a very satisfactory explanation. This in a low vaccinated country.

- Scotland was in a terrible place, but cases dived (although that fits with school holidays). High vaccinated country.

- Now England is falling for no obvious reason.

One solution to the puzzle would be if Delta has a dual nature - very contagious for a portion of the population, with high R, but same as other strains for the rest of the population. So it rips through the susceptible portion then fizzles when trying to spread wider.
What ever happened to that Trump guy, you know, the one who was president for a bit?

User avatar
Woodchopper
Stummy Beige
Posts: 3670
Joined: Sat Oct 12, 2019 9:05 am

Re: COVID-19

Post by Woodchopper » Mon Jul 26, 2021 2:24 am

Scottish hospital admissions now falling in line with earlier case decline: https://twitter.com/birdyword/status/14 ... 86880?s=21

So it’s reasonable to assume that there the case decline was real and not due to changes in testing.

Post Reply