Chat about Novel Coronavirus - 2019-nCoV - COVID-19

Yeah - seems surprising from my anecdata. Wonder if it really past the peak or people either not testing or moderating behaviour to allow xmas to go ahead. Probs some combo of that I guess.

Took just over 2 days for me to get a PCR result back this week. My other half and her sister both had them back in just over 3 days.
I think they're hitting a bottle neck of the amount of tests they can handle, whether that's because of postage or lab capacity but I expect it's hitting the case figures, rather than it actually being down on the week

Labs also suffering from staff off sick

Struggle to see how it will hit 200k by Wed or even the 1 Million Witty et Al projected by Friday?

ZOE are estimated their are currently 1.4m people with Covid

Sorry that's predicted daily infections in the UK FYI as quoted up thread.

It won't.. Interesting that it takes JP Morgan to point out the obvious in the Imperial and Warwick models.

got my pcr back in 8 hours yesterday, sadly was positive so Crimbo alone but still a quick turn around

Ooof. SAGE gonna get told off?

Symptoms.. Self selected group... Need I go on

Tomorrow's data will be more telling...

Pretty nasty reaction to Pfizer booster here, been knocked out flat since last night. Feel like shite.

Not good.. I had mine yesterday and have been fine (other than wanker's arm).

Sorry that's predicted daily infections in the UK FYI as quoted up thread.

That isn't what was modelled in any of the scenario's

You can see the pre print of the modelling:
https://cmmid.github.io/topics/covid19/reports/omicron_england/report_11_dec_2021.pdf

And doubling time is still sub two days on the latest data
https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1042235/20211219_OS_Daily_Omicron_Overview.pdf

A reliable source tells me 28th or 29th for a 4 week lockdown. If they're wrong I'll rely on them less in future.

Struggle to see how it will hit 200k by Wed or even the 1 Million Witty et Al projected by Friday?

We already hit well over 200k infections last week reportedly.

There are also reporting delays due to labs struggliong to keep up. I suspect the ONS data will show we exceed 400k this week.

Must say, at least half the time I hear people criticising SAGE or Indie SAGE its usually because the person didn't understand what the forecast actually was rather than their models being wildly wrong.

Symptoms.. Self selected group... Need I go on

Zoe don't use self reported symptoms for their prevalance estimates afaik. I think they send out a few thousand random PCR tests and use that instead.

ZOE are estimated their are currently 1.4m people with Covid

A bunch of London boroughs are well above that prevalance rate already. Lewisham was nudging 2,000 cases per 100k 7 days ago, three doubling times at current estimated R.

Yep

London borough of Lambeth has the worst UK Covid19 infection rate per 100k, of 1664.9 - high by all standards in this crisis. Among 24 to 29 years olds it is a staggering 3,962.6.

https://twitter.com/Peston/status/1472944885011472389

Kind of lines up with the earlier discussion on someone saying the yoof at the clubs not knowing how to take a LFT

AIUI they look at how many people who have symptom X also test positive and then make a guess about prevalence based on how many other people reported symptom X.

The whole thing smells distinctly unsciencey to me. But great for generating press releases.

AIUI they look at how many people who have symptom X also test positive and then make a guess about prevalence based on how many other people reported symptom X.

The method they use is intended to estimate asymptomatic infections too, so doesn't just send tests to people who think they might have Covid. But sure, I have no idea on how rigorous their actual methods are.

Edit: I explained that really poorly. They do not look for asymptomatic cases, more that they look for people reporting symptoms who do not have Covid as much as they are interested in the people who do.

The ONS is doing sample testing as well, and they use PCR. They must have an estimate also?

From the preprint..
We assume that the severity of
Omicron is identical to that of the Delta variant

And..
All scenarios considered
assume a 7.5% reduction in transmission following the introduction of limited mask-wearing
measures by the UK government on 30th November 2021, which we assume lasts until 30th
April 2022. This is in keeping with our previous estimates for the impact of increased
mask-wearing on transmission (22)

Based on..
We assume that this
decrease could be completely reversed by the face covering measure, and that this would result
in a 7.5% reduction in R over the contingency period, comparable to the effect size measured in
a regression analysis across 92 regions (Leech et al.) and a cluster randomized trial in
Bangladesh (Abaluck et al.)

Looking at their website, there is no real transparency about the modelling used. In my view the secret sauce for the sale/IPO in the future.

https://covid.joinzoe.com/post/how-we-are-predicting-covid-in-the-uk

I'm perturbed that so many people would share their medical details and location with a private company.