Daily numbers, graphs, analysis thread 21

[BigChocFrenzy]

Welcome to thread 21 of the daily updates

Resource links:

Uk dashboard deaths, cases, hospitals, tests - 4 nations, English regions & LAs
R estimates UK & English regions
Imperial UK weekly LAs, cases / 100k, table, map, hotspots
UK School statistics Attendance
Modelling real number of UK infections February to date
NHS England Hospital activity
MSAO Map of English cases
Cases Tracker England Local Government
ONS MSAO Map English deaths
CovidMessenger live update by council district in England
Scot gov Daily data
Scotland TravellingTabby LAs, care homes, hospitals, tests, t&t
PH Wales LAs, tests, ONS deaths
NI Dashboard
Zoe Uk data
UK govt pressers Slides & data
ICNRC Intensive Care National Audit & Research reports
NHS t&t England & UK testing Weekly stats
PHE Surveillance reports & LA Local Watchlist Maps by LSOA
ONS England infection surveillance report each Friday
Datasets for ONS surveillance reports
ONS Roundup deaths, infections & economic reports
ECDC rolling 14-day incidence EEA & UK
Worldometer UK page
Our World in Data GB test positivity etc, DIY country graphs
FT DIY graphs compare deaths, cases, raw / million pop
Alama Personal COVID risk assessment
Local Mobility Reports for countries
UK Highstreet Tracker for cities & large towns Footfall, spend index, workers, visitors, economic recovery

⏭ Our STUDIES Corner ⏮

We welcome factual, data driven and analytical contributions
Please try to keep discussion focused on these
📈 📉 📊 👍

@IloveJKRowling

Yes, some of the Oxford colleges can take a hit (they're not all rolling in it!) but at the same time one of the big things about academia in Oxford is the tutorial system. The tutorials are often 1-2-1 or maximum 1-2-3, whereas my dd's uni has 20 in her tutorials, and says some of the tutors probably wouldn't even know her name.

So if they are doing lectures online and tutorials face to face it is on that basis less risky.
But also, looking back when I started uni, I would really have struggled that first term, trying to understand the work, sort out a new way of working, etc without peers to tell me they were also struggling with the same things and work together to sort it.
So it isn't just about what can you get online. It is also about the peer support, which for me in those early days was far more valuable than anything else.
Yes, it isn't the university experience they wanted, by a long factor. But I think staying at home would not help that, and for some of the students they would find it far harder to work and far harder to then flee the nest afterwards.

My dd's 2nd year. She's in a house with 2 others doing the same subject-in a lockdown area, coming from our low area. Everything is online, they're not expecting to leave the house except to go to the supermarket once a week. But she's still going back. Not because she's desperate to get back, she's having a lovely comfortable life at home(!) (her words).
But she needs the peer support to get through her course. I very much doubt that if she stays at home she will go back. She would work hard anyway-she is a worker, but being around the people her own age and having to work out living in this situation.

I'd rather tuck her up here and keep her here, but it isn't the best thing for her long term.

Students and universities are caught between a rock and a hard place. Say it's all online and they don't need to come back-students will say how hard it is stuck at home without peer support, without access to what they need, perhaps poor internet etc.
Say they come back and they get accused of trying to take money for nothing and spreading the virus.

@Augustbreeze - different methodologies and different sample groups in the datasets.

ONS 25th Sept: "An estimated 103,600 people (95% credible interval: 85,600 to 123,400) within the community population in England had the coronavirus (COVID-19) during the most recent week, from 13 to 19 September 2020, equating to around 1 in 500 people (95% credible interval: 1 in 600 to 1 in 400)." www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/bulletins/coronaviruscovid19infectionsurveypilot/englandwalesandnorthernireland25september2020

Covid Zoe estimate today says 230,967 or about 1 in 300: covid.joinzoe.com/data.

Imperial swabs 18th-26th: 1 in 200 - www.imperial.ac.uk/news/205473/latest-react-findings-show-high-number/

UK R number 25 September 2020: 1.2-1.5, Latest growth rate range for the UK+4% to +8%. www.gov.uk/guidance/the-r-number-in-the-uk.

UK test positivity is 2.5%, so still controllable with an increasing but acceptable test positivity rate overall (worse in some regions): ourworldindata.org/coronavirus/country/united-kingdom?country=~GBR.

Graph of all latest estimates (personally I ignore YYG as they go very high or low depending on how much China likes a country): ourworldindata.org/grapher/daily-new-estimated-infections-of-covid-19?tab=chart&stackMode=absolute&country=~GBR&region=World.

These are all ESTIMATES apart from the actual number of cases and test positivity. Nobody knows. Minimise interactions, wait and see the impact of the new restrictions.

I had a very different uni experience. I needed to fund my uni, so worked full time after the first semester, attended only a very small handful of lessons. No online version then either. Got the reading lists, exam dates and phoned in every week for any coursework that came up. Thinking back, that first semester was really useful on setting up a network of ppl who actually attended and then getting their notes. But if I had online options I wouldn't have needed that either. Of course, this was a non-practical course. And I lived at home. (and then did 2 other diplomas the same way)
So I'm a bit torn between going to uni for knowledge and having a uni experience. I see how the latter can be important, but I don't really "get it". We had discussions about this with my husband - he's a civil engineer. He needed to be there, he had to draw stupid amounts, wouldn't have had time to work. And needed his peers more than I ever did. He lived in student accomodation far from his parents.
I would say that for courses that don't absolutely require live attendance getting kids back for the sake of money and then shutting them in halls was a big mistake.
But on the other hand having a set of adults with one experience and robbing the other half of something they perceive valuable isn't fair either.
I guess "is uni non-educational experience overrated" would be a separate topic of its own :)

ONS sample should in theory be more representative.

Imperial is testing volunteers, i.e. people who are more likely to be socially active and therefore more likely to catch disease. What is probably more interesting is the trend rather than the actual number.

Is that true though about Imperial. Both my son and I were contacted by them at random to provide samples and no questions about our lifestyle was asked until after we had agreed to take part. And we don’t go out apart from to work/school at all!

@MollyButton

I agree it is a great article. We have talked a lot about the k-factor and superspreading events but here is what I have found particularly interesting:

• Test and trace operation is much more effective when its primary focus is on identifying superspreading events rather than infected individuals.
• Cheap rapid tests with lower level of sensitivity can be as useful (if not more useful) than “proper” highly sensitive PCR tests. If we accept that we no longer need to identify every single infected person but are instead looking for recent clusters, timeliness becomes more important than accuracy.

I am afraid this approach (cluster busting) is quite different from what is currently being recommended by SAGE but the arguments in the article and the reference to Japan sound very plausible.

@Worriedmum999

The REACT study website mentions that they (i) select people at random and (ii) invite them to voluntarily participate in the project. They are not necessarily looking for a perfectly representative sample because they do not have the same level of information as ONS.

ONS is more focused on representation and they encourage participation from groups who might be reluctant to participate in the project by offering monetary incentives (vouchers). This would remove some of the selection bias.

The recent spike in cases was largely driven by white middle class population under 35. I would expect this group to be over represented in the Imperial sample.

@wintertravel1980 - I agree it's a useful quicker approach, though my thinking is that both strategies should have their place. Running tests for everyone in a care home, everyone in a school year group including teachers, everyone in a hospital area, everyone in a factory, whole steeets if there are suspected clusters, all uni students in a block etc - of course some people opt out, and you don't want everyone getting too comfortable with a whole "we're all fine" when an infection could enter the group tomorrow... But surely that method would highlight the infections.

Money at the moment has been put into the sewage testing, which would tell you everywhere infected people have been to toilet (home or work or pub) and effectively get those results faster. Mass testing those districts could then identify where exactly people are infected, or individual testing dependent on how close to the source the sewage testing gets (spit tests + sewage would be the ultimate game changer). The trouble is that we aren't seeing reports from sewage testing yet. I suspect that means there is uncertainty, because the sewage picture says something different than the confirmed cases and even swab test research, or because they can't make sense of the total infection in sewage to expected cases, and so there are people scurrying to confirm what is correct. Is that good news because many more cases have been found and rates will drop like a stone, bad news of big hidden clusters or a mixed bag?

Ok, thank you. That makes some sense now. So you’re basically saying that it’s unlikely to be as high as 1 in 200 then.

[quote MRex]@BigChocFrenzy - just spotted you said you wanted datasets for the accompanying slides, it's literally just the data shown but you can download it here: www.gov.uk/government/publications/slides-and-datasets-to-accompany-coronavirus-press-conference-7-april-2020.

By the way, did I post this link before? It's quite useful for getting quick comparisons between a local authority and others in the region, plus other regions. lginform.local.gov.uk/reports/view/lga-research/lga-research-report-covid-19-rolling-weekly-tracker?mod-group=AllRegions_England&mod-type=namedComparisonGroup[/quote]
.....
Thanks, Mrex

@Worriedmum999

Is that true though about Imperial. Both my son and I were contacted by them at random to provide samples and no questions about our lifestyle was asked until after we had agreed to take part. And we don’t go out apart from to work/school at all!

Ditto, my very careful DC aged 12 was asked to take part (but we didn't agree at that point as we were certain he would be negative, because not socialising or shopping etc., and the test sounded quite invasive and unpleasant).

@Worriedmum999

The fact that 1 in 200 people currently have active Covid in the latest Imperial study has really scared me this morning. It seems so high that 330,000 currently have it. Does this mean that we are going to see a huge death rate over half term? Can anyone help me unpick the numbers?

.... No serious scientist is predicting large numbers of deaths that early

The number of daily cases atm is about 40% below the Whitty / Vallance predictions of the previous briefing on 21 September
So either measures are working, or their predictions were on the pessimistic side

The most reliable estimates of infection are from the ONS, because they do representative population samples,
whereas other study groups are statistically biased because of being volunteers / blood donors / in a particular town or city

@Augustbreeze

Can someone explain for me, why are Imperial saying the rise is slowing if the incidence is now 1 in 200, thought to be 1 in 500 last week (by ONS iirc)?

I'm sure I'm missing something here?

.... If the rate of increase is slowing, there is still an increase, just not as steep

To see a consistent fall in cases or incidence, we would need to have R

No idea what is happening at Oxford this year but a few points on universities generally and Oxbridge in particular.

1. Research heavy unis (like Oxbridge) have quite a few postgraduate students. They dont move around as much/ act stupidly when drunk as often but they are more likely to move around the community, dont ignore them. They will be around when undergrads are not.

1. Most unis are staggering arrival times, sometimes over weeks. Oxbridge normally have conferences so couldnt do this, no idea what is happening this year re return dates - but they have lost a lot of income from conferences.

1. The tutorial experience is part of the Oxbridge sell, that gives them more of an excuse to call students in.

1. Oxbridge courses are very intense, students probably need to be in and have the peer support and peer pressure. They also can use uni health services.

1. Oxford has its own testing, as does Cambridge. Cambridge will be testing students in uni accommodation (virtually all undergrads) weekly, Oxford's testing sounds more like when they have symptoms.

Oxford's testing sounds more like when they have symptoms

This is correct and I think inadequate given the current levels in Oxford.
They need to be more proactive to keep things under control.

I think Oxford should be giving international students at least the option to defer a year or do a year online. Whilst I agree many students may want the experience of being with their peers, for those that don't and are scared to come to a country with higher infection levels than their own, I think they shouldn't be disadvantaged. It would be utterly miserable for many international students to be so far from your family in lockdown.

International students pay a lot more and they are paying for tutorials/ the uni experience/ making contacts for the future.

If they tested more often - and in particular test all students on arrival/ before leaving for home - it wouldnt be an issue. I believe Cambridge are batch processing the tests. If you combine 10 tests and they are negative you are fine, if one if positive you process each test separately. Oxford could do that.

Thanks to @MRex, @BigChocFrenzy and any others who've addressed my v basic question!

Expert reaction to preprint with the latest interim data from the REACT-1 study on COVID-19 spread across England:
www.sciencemediacentre.org/expert-reaction-to-preprint-with-the-latest-interim-data-from-the-react-1-study-on-covid-19-spread-across-england/

If they tested more often - and in particular test all students on arrival/ before leaving for home - it wouldnt be an issue. I believe Cambridge are batch processing the tests. If you combine 10 tests and they are negative you are fine, if one if positive you process each test separately. Oxford could do that.

Agree, but they're not. It's too late now to test all students on arrival.

Re the rate of increase slowing, here’s an illustration, with made up simple figures.
If case were rising steadily, you would see
10 per 1000
20 per 1000
30 per 1000

If the rate of increase slows, you would then see
39 per 1000
47 per 1000
54 per 1000

So cases would still be increasing, but the rate of increase would be decreasing steadily.

From the guardian
A single alert to public health officials in Cornwall from the national NHS test and trace service led to the identification of 170 Covid-19 cases at a food factory.

The vast majority of workers at Pilgrim’s Pride Ltd in Pool who proved positive did not have coronavirus symptoms and had no idea they had the virus.

Wowsers. Good that this outbreak was found but terrifying that there are 170 cases where they didn’t realise in an area which must be towards the bottom of the watch list.

MRex yes I'm in Edinburgh and will write to my MSP about the % positives and newly tested definitions.
668 new positive tests in Scotland, with 10.8% positive tests.
First Minister has said the R could be as high as 1.7, although as this is worked out from past data, it's too early to see the affects of the restrictions brought in last week.

The number of daily cases atm is about 40% below the Whitty / Vallance predictions of the previous briefing on 21 September
So either measures are working, or their predictions were on the pessimistic side

The other possibility is that the aim of the speech was to scare people into following the rules especially in lockdown areas and that it worked.

I do wonder whether the reason some of the universities that are not doing mass testing is because they don’t want to know about asymptotic cases. Let’s face it if we were doing more testing of School pupils and university students who are much more likely to be asymptotic it could have a massive negative impact on our ability to provide a decent education. In most cases adults are only tested if they have symptoms and lots of us are travelling to work, working with others and coming into contact with children who then go to school and may even be socialising. Why do University students need to be tested on mass if we don’t do the same for the general population?

Sorry to derail such an interesting and wide-randing thread but just to reply to @IloveJKRowling I think Oxford should be giving international students at least the option to defer a year or do a year online. All Oxford students can apply for dispensation from residence requirements and can have remote teaching if that is possible for the course. I have a number of students doing precisely this. Not sure whehter you have a connection to someone who wants to do so but the details are here if you do www.ox.ac.uk/students/life/residency

Fingers crossed but this looks promising, especially given the numbers over the past few days have slowed in how much they are increasing.

^Report published 1st October 2020
Cases data from week 22nd-28th September 2020^
Data extracted covering testing up to 28th September 2020 show that the total number of confirmed cases for the last 7 days is 1387, an increase of 317 cases on the previous week. The latest weekly rate of Covid-19 in Liverpool is 278.5 per 100,000 population and the latest positivity testing rate* is 15.8%.

This is the first decline in the positivity rate in Liverpool for the past few weeks.

It certainly suggests at least a stablising of numbers.