Data and analysis thread, started 12 November

[NoGoodPunsLeft]

Previous thread here:

Data and Analysis Thread, started Oct 29 www.mumsnet.com/Talk/coronavirus/4064113-Data-and-Analysis-Thread-started-Oct-29

Regular lurker but I frequent poster, didn't want to lose the threads.

[Castiel07]

I was watching Jeremy vine this morning, dr Sarah Garvis is on there and she just said that from the pilot test who tests 360,000 people that 1 in 80 has covid now.

[MRex]

Well it's the Guardian, so they have to find an angle to criticise. If it was managed centrally they'd say there's a lack of local control, local authorities are best placed to decide on the ground based on local factors etc etc. The requirements for storage, performing tests and reporting results are all provided to local authorities, who can interact with healthcare providers in the local area.

[TheSunIsStillShining]

sewage in schools project

termproject.org/

[GetAMoveOnTroodon]

Interesting numbers from Liverpool today, in 7 days 430 people across the city tested positive, 200 with symptoms, 230 without.

[herecomesthsun]

ONS survey here www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/bulletins/coronaviruscovid19infectionsurveypilot/13november2020

[herecomesthsun]

1 in 85 with covid. Primary % now stable. Highest numbers in 17-25 years & secondary but falling.(great!!)

Older ages rising though, which is seriously grim.

[Baaaahhhhh]

1 in 85 with covid

I think this is terribly misleading though. As we know across the country the picture is very different. In the first wave I knew many, many, people with Covid, including DD1 and all her friends and co-workers. This time I don't know of anyone, including at DD's secondary school, where they have still only had one case early in Sept. Perhaps they would be better to give regional figures, it might help to know that in the North West it might be 1 in 20 (made up figure), but it might make people more aware of their own risk or meeting someone with Covid.

[cathyandclare]

Positivity rates in Wales have increased in recent weeks; during the most recent week (31 October to 6 November 2020), we estimate that 35,300 people in Wales had COVID-19 (95% credible interval: 25,500 to 46,600), equating to 1 in 85 people (95% credible interval: 1 in 120 to 1 in 65)

The most recent week is the second half of the firebreak lockdown, so maybe a little soon to see the effects. It will be interesting to see next week's figures.

[CoffeeandCroissant]

You can do an approximate calculation from the regional ONS figures (attached).

Twitter thread summary here: mobile.twitter.com/ONS/status/1327220998312222725

[lurker101]

Glad I found you all! I typed a post on the old thread to find it closed!

The latest report is out for Northern Ireland from the Public Health Agency. Some interesting changes on schools data. The last time I shared this report there was a very clear difference between the staff/pupil split in primary vs post primary, however this is somewhat closing now, with 61.5% of primary school cases from 24 August to end of week 45 being pupils, with 76.6% at post primary. Interestingly special schools have a much higher staff incidence - only 26.2% cases were pupils.

However, there is still a difference evident between primary and post primary as 0.31% of post primary school aged pupils have had a positive test as opposed to 0.15% of primary.

Some good data on schools and care homes. I like how they split the data and given NI has had some of the UKs highest infection rates is interesting for U.K. as a whole given some of the patchy data being collected

<a class="break-all" href="https://pdf.browsealoud.com/PDFViewer/_Desktop/viewer.aspx?file=pdf.browsealoud.com/StreamingProxy.ashx?url=www.publichealth.hscni.net/sites/default/files/2020-11/Weekly%20COVID-19%20Bulletin%20week%2045_0.pdf&opts=www.publichealth.hscni.net#langidsrc=en-gb&locale=en-gb&dom=www.publichealth.hscni.net" rel="nofollow noindex" target="_blank">pdf.browsealoud.com/PDFViewer/_Desktop/viewer.aspx?file=pdf.browsealoud.com/StreamingProxy.ashx?url=www.publichealth.hscni.net/sites/default/files/2020-11/Weekly%20COVID-19%20Bulletin%20week%2045_0.pdf&opts=www.publichealth.hscni.net#langidsrc=en-gb&locale=en-gb&dom=www.publichealth.hscni.net

[herecomesthsun]

I'm sorry to see the negative comments on the Guardian. It provides a much-needed balance to most of our other newspapers which appear to be controlled in a concerning way by a handful of tycoons, mostly not living in the UK. Even if I don't always agree with everything it says.

[herecomesthsun]

Age stats up to 6th November

www.ons.gov.uk/visualisations/dvc1056/age/datadownload.xlsx

[RishiMcRichface]

Yes I try to look at several different online papers for news. You get a good up to date news blog in the Guardian and the Sun so I feel like their biases cancel each other out if I read both.

[herecomesthsun]

R apparently has fallen- Telegraph

www.telegraph.co.uk/global-health/science-and-disease/coronavirus-news-lockdown-pfizer-vaccine-cases-deaths/

[NeurotrashWarrior]

I'm not saying I agree with the paper, and I don't always like the guardian; I'm just puzzled as to why they're commenting on that particular aspect of those tests. And then stopping short of the reasons why to use them.

I wonder if that's why it's going to be used for asymptomatic cases. It's 50-70 % better than nothing and also quicker. So less intrusive and will be a good screen. I was told that in one area here hcps are going to be tested 2x weekly with it.

[NeurotrashWarrior]

Interestingly special schools have a much higher staff incidence - only 26.2% cases were pupils

Different ratio of pupils to staff (anywhere between 2+ staff to 1 pupil and 1 staff to 5) and more staff working closely together.

[feelingverylazytoday]

@NeurotrashWarrior

Interestingly special schools have a much higher staff incidence - only 26.2% cases were pupils

Different ratio of pupils to staff (anywhere between 2+ staff to 1 pupil and 1 staff to 5) and more staff working closely together.

Also more likely to be in close physical contact with learners.

[MRex]

Given the clustering of infections around superspreaders, I might even be able to be convinced that less accurate testing is the way forward for the general population (not health/care staff), at least until cases are low enough to adjust restrictions. Mostly because it might help with the tracing of contacts by pinpointing cluster sources who are more likely to have spread it and that Japanese report suggest cluster focus worked for them (whether actually reproducible here is perhaps another matter).

As for newspapers / journalists - there are good and bad of all stripes. Also sometimes people just aren't at the top of their game or get a strange bee in their bonnet, leading to something head-scratching. We're all humans muddling along. Read a few sources, be aware of ownership and other editorial bias, if you're disagreeing then it's all good because your critical thinking cap is on. It's nice to guess who will use "The Goings of Cummings" as their headline.
Totally unrelated to covid BTW, but the attacking Orcas story is really really worth a read if anyone missed it, fascinating: www.bbc.co.uk/news/extra/buqvasp1rr/orcas-spain-portugal

[herecomesthsun]

re lateral tests

Prof Bell from Oxford was quoted as enthusiastic

However, the other professor cited made some fair points

• “I am really concerned that people are not given information to understand what the results mean.

so that could be a real problem with a lot of false negatives

A negative test indicates your risk is reduced to between a quarter and one half of the average, but it does not rule out Covid. It would be tragic if people are misled into thinking that they are safe to visit their elderly relatives or take other risks.

So, you have to take a negative result with a pinch of salt and still continue being very careful.

[TheSunIsStillShining]

I don't agree with the "better than nothing" line in this case.
If it's only 50% -70% accurate then the false sense of being safe will potentially have a huge role in behaviour.

I do agree with the better inform people part

[Tyzz]

Mrex I like to balance out the Daily Telegraph with the Guardian. They annoy me in equal measures. The Orca storey is fascinating.
A negative test indicates your risk is reduced to between a quarter and one half of the average, but it does not rule out Covid. It would be tragic if people are misled into thinking that they are safe to visit their elderly relatives or take other risks
I am so disappointed to read this. Was hoping it might be a way for me to hug DS who is a teacher at a school which is badly hit.

[wintertravel1980]

SAGE has now released the most recent set of their papers. They are finally acknowledging that the partial population immunity (i) does in fact exist and (ii) combined with social distancing measures, might cause the epidemic to plateau as seen in other places (e.g. Florida).

assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/935238/potential-trajectories-covid-190next-6-months-s0848-291020.pdf

Of course, it is only a scenario and not a prediction but it is a useful alternative to the usual exponential graphs presented so many times at government press conferences.

[wintertravel1980]

27,301 cases
376 deaths
379,955 tests processed (very high for Friday reporting)

[NeurotrashWarrior]

Love the orca thing; I've read it twice now. Completely obvious they're playing from the start though!

Obviously learnt they can spin the propeller.

[NeurotrashWarrior]

If they're using the lateral tests for regular screening because they're quick, for example a health care worker, as I've been told, twice a week, that is probably better than the other version of the test and frees up more of those tests for cases. Some will get 'through the net' but twice a week hopefully reduces that. Potentially 3 times in 10 days.

Screening patients coming into a hospital in that situation isn't going to be safe.

It sound like they need to be used in certain ways carefully.