Daily numbers, graphs, analysis thread 20

[BigChocFrenzy]

Welcome to thread 20 of the daily updates

Resource links:

Uk dashboard deaths, cases, hospitals, tests - 4 nations, English regions & LAs
Imperial UK weekly LAs, cases / 100k, table, map, hotspots
Modelling real number of infections February to date
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
R estimates UK & English regions
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 civil discussions from all contributors 📈 📉 📊 👍

Request to posters giving a link:
Please do so in full, so people can see in advance what they are clicking
Also at least a brief title so we know what the link is about

6k

That's the third highest single day figure we've had (2nd highest only slightly higher too)

I know there were far more in March we didn't get, but I also wonder with asymptomatic cases how many more cases won't be picked up too. Zoe app is saying 12k symptomatic.
But how many cases are asymptomatic?
I've heard between 50% and 80%.
50% would mean 24k cases, which is still far below, but if it's 80% then 12k symptomatic could mean 60k in total, which is getting towards March levels.

@littleowl1 had it been clear to you why all of Norfolk had been on the watch list? They all seemed to have been fairly low to me, but I might be wrong.

Hi all. Just been reading this pre print and wondered what your thoughts were? Appreciate it is just a pre print but I thought the results were staggering:

"Seroprevalence among Asymptomatic Population in Tokyo during the Second Wave.Seroprevalence increased from 5.8 % to 46.8 % over the course of the summer. The most dramatic increase in SPR occurred in late June and early July" www.medrxiv.org/content/10.1101/2020.09.21.20198796v1uly.

Hang on, there were 36 deaths on 19th March (43 on the 18th, 34 on the 17th) - why should we be less worried now? From there deaths went up very quickly.....

I get the argument that testing is better now...... but deaths are now at the level right before lockdown in March.

Zoe app seems to have gone back to counting anybody ill with anything as having covid, that simply won't work in Autumn/ Winter months. I'm not wanting to minimise the gravity of where we're at, it's an ugly situation, but I really think their approach needs to be looked at.

@sirfredfredgeorge

I totally understand we can’t compare number of positive tests now with for eg March but can we compare daily death numbers

No, not really, since daily deaths are a proportion of the total number of people infected over the previous few weeks, which is highly impacted by the doubling rate - if cases are doubling every week now, compared to every 2-3 days then, then forecasting days ahead needs to take that in to account.

Right, so deaths might not grow as quickly as they did then but we can extrapolate that today there are roughly the same number of cases as early March?

@wintertravel1980

Seems early for Uni

Looks like the Freshers week at the University of Glasgow ran from Sep 12 to 18. Some other universities might follow similar timelines. Assuming 5-7 days for virus incubation / display of symptoms and 2-3 days for testing (Scotland has got shorter turnaround times than the rest of the UK), we can expect to start seeing first cases now. The bad news is there will be more positives coming - the question is whether T&T can track and contain the spread.

I know people might be skeptical of T&T but the local authorities seem to be relatively efficient at tracking "complex" cases. I assume outbreaks at universities would fall into this category.

I'm just musing here speaking from my own experience on unis in Glasgow.

More students stay at home when they attend unis in Scotland. So whilst there will be a lot staying in halls, a lot will also return home at the end of the day.

I don't know what the set up is this year in Glasgow and if students are attending in person lectures.

If they are them large numbers of students going home is a concern

If not then it may be relatively well contained within Halls.

I may be wrong here but weren't the actual deaths in March actually a lot higher because those are only hospital numbers and don't include care homes?

I think watching hospital admissions and in patients is a better comparison. Remember we also mainly locked down to protect the nhs so they weren't overrun....

Here is an article about Scottish students.

I didn't actually consider the fees which is another reason to study close to home.

www.heraldscotland.com/news/16050357.stay-home-scottish-university-students-likely-study-rest-uk/

But how many cases are asymptomatic?
I've heard between 50% and 80%.
50% would mean 24k cases, which is still far below, but if it's 80% then 12k symptomatic could mean 60k in total, which is getting towards March levels

The asymptomatic number will heavily influence the march numbers, so if it's 80% asymptomatic, it will also mean the march estimates are at the higher end.

The Japanese paper @Jrobhatch29 mentioned is also increasingly making me optimistic that actually cases higher than the estimates have previously had and that more healthy people are currently immune, so the longer term outlook is more optimistic as immunity levels are stronger than suggested. But really that is nothing but hints, so we cannot rely on it.

I’ve been thinking about the comments from the 1st page about interconnectedness of the NW and I think there’s a lot in that. Looking at the queue at school today, it was about 50% parents and the other 50% were grandparents. Generalising now, but in my patch of the NW most people live near their parents and grandparents and are set up for informal childcare. I was seen as odd for saying I wouldn’t be using my (vulnerable) parents for childcare. In London I suspect there are higher rates of formalised childcare and less dependency on relatives. Grandparents looking after multiple grandchildren are creating big links between households and work settings and schools. Not blaming anyone, just that for many many people round here that is how life was set up (in a pre-covid world) and they are now stuck for alternatives

I had to remove "uly" to make that link work: www.medrxiv.org/content/10.1101/2020.09.21.20198796v1

@Jrobhatch29 - it's interesting, but it's employees from 11 factories in Japan, all of whom work for the same company. Only 350 of them did both tests. The total number of additional new infections was only 75 people (21.4% of 350). They also had a lot of reversion, 12.2%. And used fingerprick blood tests, but many of those have severe accuracy issues. However unlikely, let's say most of the 75 infections are true; the chance of infection in factories is higher than average, the chance of a group of them having infected each other due to an outbreak is also high, and factories often have younger workers. There need to be other studies, or I don't think we can make too much of the data.

another reason that cases must be much lower than the actual....earlier this week we all had the same virus but DS got it first and was coughing the worst so he got tested - negative. It's very clear to me that we all had the same thing. If he had tested positive I don't think we would have seen the point of getting the rest of us tested especially as so difficult to get tests. So that would be 3 positive cases missed. I wonder how often this happens.

there are roughly the same number of cases as early March?

There were roughly the same cases among at risk individuals in the couple of weeks over end of feb/beginning of march. But we would need to know very much the average time from infection to death to say more, as it cases could easily have been doubling close to daily at this period. Very hard to say exactly other than the deaths will not be going up at the same rate from today as they did in march, perhaps half the rate at worst.

I'm not really surprised that cases are continuing to rise, I would expect them to keep doing so for several weeks yet.

Hopefully the rate of doubling will slow and cases will stabilise at a manageable level before (hopefully) falling a bit. I suspect we may not get back down to the very low levels we saw in a July this side of spring 2021.

Some catch up from the testing issues over the last week or so also seems likely.

@Whydoyouthinkthatthen

I had heard anecdotally that the Norfolk areas were added as precaution after the Banham Poultry outbreak. I guess the risk was that it would spread into the community. But it looks like the local health teams contained it well. Great to have a little good news in the sea of worrying headlines.

And of course remember the deaths reported today didn't actually happen today or yesterday. They may have happened several days ago or even weeks ago in some cases. If you look at the data dashboard so far only 12 people have been reported to have died of COVID yesterday, and 12 the day before. We would expect that to go up over the next few days as reporting catches up. The dashboard is here if you want to see the actual deaths by date. coronavirus.data.gov.uk/deaths While deaths are clearly rising, you need to wait 3- 4 days at least to know more accurate statistics on today.

6k!!! Came to post so I can read up this thread.

I think more or less was on the covid numbers today(or yesterday!?) if anyone heard that?

Re: comparisons of today’s death rates to March. I’m alarmed but not surprised at today’s case numbers (and suspect this is closer to the figure they should have been at for the past few days given the testing problems) but even though the comparison today’s death figure to mid-March is alarming from that perspective looking back at the figures from August up to now it’s not massively out of step, although still the highest for a number of weeks.

A longer trend is needed to see whether that’s what we should be worried about right now

@MRex

I had to remove "uly" to make that link work: www.medrxiv.org/content/10.1101/2020.09.21.20198796v1

@Jrobhatch29 - it's interesting, but it's employees from 11 factories in Japan, all of whom work for the same company. Only 350 of them did both tests. The total number of additional new infections was only 75 people (21.4% of 350). They also had a lot of reversion, 12.2%. And used fingerprick blood tests, but many of those have severe accuracy issues. However unlikely, let's say most of the 75 infections are true; the chance of infection in factories is higher than average, the chance of a group of them having infected each other due to an outbreak is also high, and factories often have younger workers. There need to be other studies, or I don't think we can make too much of the data.

Yes you're right about the sample bias with the factories. They were samples from across several factories across tokyo so would that suggest wide community transmission? I'm not sure. I find the part where it says seropositive participants were negative again within a month which implies it might be quite hard to get a true representation with these surveys. I wonder if it has a lot to do with the type of test too. There is a meta analysis study being discussed alot today on twitter suggesting true asymptomatic covid is only about 20%

Thanks ontop, that's really useful to see the date of death vs date reported graphs.

think watching hospital admissions and in patients is a better comparison

Agree. Does anyone know if there's recent local data on this - I can get a report for my hospital trusts published monthly, but can't see anything more up to date at a local level rather than nationally. Sorry if I'm missing the obvious.

Does anyone know where you can find the cases per 100,000 the govt use to decide who goes on the travel corridor?

T and T for schools has been handed over to the DfE. Does anyone think they will do a good job of this???

Slight correction : not T and T but the advice on what actions to take. General Q still applies..