Daily numbers, graphs, analysis thread 9

[Barracker]

Welcome to thread 9 of the daily updates.

Resource links:
Worldometer UK page
Financial Times Daily updates and graphs
HSJ Coronavirus updates
Johns Hopkins Coronavirus Resource Centre
NHS England stats, including breakdown by Hospital Trust
Covidly.com to filter graphs using selected data filters
ONS statistics for CV related deaths outside hospitals, released weekly each Tuesday

Thank you to all contributors for their factual, data driven, and civil discussions.

@NeurotrashWarrior covers it all really!

Callimanco Several other countries had young skiers returning with infections
Nothing particular wrt the UK there; much of Europe go skiiing - this has been mentioned as one of the initial vectors for much of Europe
e.g. Germany had many early infections from young skiers returning

As for coming off more lightly in a 2nd wave, why ?
I have read many concerns that since the Uk has to relax lockdown with a higher rate of deaths & infections than other countries who did so, any 2nd wave is more likely to be heavier

However, no signs of any 2nd wave on the continent
Hopefully countries now have their systems in place to lockdown local outbreaks before they can spread

today's slides are here -
assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/888823/2020-05-30_COVID-19_Press_Conference_Slides.pdf

Not linked from normal page as yet, but changing the URL to today's date gets them ;-)
Very slight uptick in admissions last 2 days, but way too early to draw any conclusions from that (as previous days at beginning of week saw quite a drop - bank holiday related maybe?)

Edward Morgann@Edward*_Morgan (ONS)

New ONS data out this morning on the numbers of families and households eligible to send primary school aged children back to school next Monday.

We estimate 680,000 families in England will have all their children eligible to return to school next week.

https://ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/articles/coronaviruscovid19roundup/2020-03-26#primarychildren…

No surprise that catering & hospitality vacancies have plummeted, also retail & wholesale.
Jobs / vacancies are a key factor the govt must consider, alongside public health

ONS report:

https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/articles/coronaviruscovid19roundup/2020-03-26#jobads

Prior to this, they had been relatively stable since the beginning of 2019.
......
Total online job vacancies declined more than 50% from the start of March to the start of May 2020

Whats next I work in an area of health where we have multiple close daily contacts with patients. We create aerosols which we know carry many pathogens. Surprisingly we suffer few major infections, partly due to PPE but mostly due to busy immune systems. It’s long been thought that constant low level or sub clinical infections are dealt with by T cell immunity and probably like children our T cells are more active than most people.
As a profession it has been surprising how few of us have had Covid-19 but many have reported Covid -like infections retrospectively from Jan/Feb which was too early to have been Covid. I had Covid-like infection mid Feb along with most of the staff. Fortunately due to work pattern I was only in work for a couple of days with symptoms so may have limited transmission. Also wearing PPE would have reduced spread. There have been a few dentists with severe infection mostly early on in the outbreak.
I developed symptoms 5 days after spending a day in a large secondary school doing a careers fair. At the time there was a virus sweeping through our local schools the symptoms were high temp and cough. Having read the paper re T cell cross reactivity it does make sense.

@BigChocFrenzy
Why getting off lighter in a second phase was (apologies for not being clearer) linked to the Twitter feed on t cell immunity and the hypothesis on that thread that a lower percentage of the population might need infecting to generate immunity than seen from antibody results.

I guess my hypothesis was, that, other than the South West and Northern Ireland, we have had a pretty good dose across the whole country - whereas other countries like France, Italy, USA have had severe infections but heavily localized. So in theory they may have swathes of localities where little exposure has taken place and could grow exponentially if an infection takes hold. Whereas as we have seen transmission across the country already and have few untouched pockets where many susceptible people have not yet been infected - relatively speaking of course.

Has anyone either seen or plotted themselves the curve for London vs curve for rest of UK?

I’ve heard a couple of people say things like ‘London is a couple of weeks ahead of the rest of UK’ and I can see why London was worst hit and sooner, but for the figures to be so different at this point is confusing me given we all locked down at exactly the same time... am I missing something? I’m not as mathematical as you lot! But very interested. Thank you.

CallimancoThe infection rate across the UK does not seem anywhere near the level needed for herd immunity, even in London

I've not read any epidemiologist saying that having let infections spread more widely would be an advantage in a 2nd wave

Coronavirus (COVID-19) Infection Survey pilot: 28 May 2020

Confirmed cases neither increasing nor decreasing, but with tests still in the post being counted - maybe even twice ! -
we probably need to consider the deaths, which do seem to be reducing

www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/bulletins/coronaviruscovid19infectionsurveypilot/28may2020

•	There were an estimated 54,000 new COVID-19 infections per week in England (95% confidence interval: 34,000 to 86,000), 
	a similar level to the previous estimate 
	<strong>indicating that the number of new COVID-19 infections is relatively stable;</strong>
	this equates to an incidence rate per week of 0.10 new cases per 100 people.

•	Of those individuals providing blood samples, 
	<strong>6.78% (unweighted) tested positive for antibodies to COVID-19</strong> 
	(95% confidence interval: 5.21% to 8.64%); this equates to around 1 in 15 people.

btw, Above was England only, so the other 3 Uk nations likely to have a lower % infected

ONS breakdown on who is infected - in England:

https://www.ons.gov.uk/file?uri=%2fpeoplepopulationandcommunity%2fhealthandsocialcare%2fconditionsanddiseases%2fdatasets%2fcoronaviruscovid19infectionsurveydata%2f2020/covid19infectionsurveydatasets20200528.xlsx

Hardly surprising that those who work outside the home are about 3.5 x more likely to be infected

Age distribution of those infected shows 20-49 group with highest positive test rate,
with age 2-11 the next highest, maybe surprisingly.

hospital admissions for last 14 days - last week was lower than the previous week but havent declined in the last 6 days and may be increasing again. Not, in my view, time to relax lockdown considerably, that's a big gamble. Businesses back to work is necessary and places like car showrooms and national trust gardens are not major risks - but a couple more weeks to see how that worked out would have been wise.
17/5 719
18/5 678
19/5 639
20/5 637
21/5 697
22/5 713
23/5 675
24/5 685
25/5 595
26/5 471
27/5 472
28/5 475
29/5 552
30/5 562

Pretty spot on article about how the response cannot be the same across the country.

Where are the traffic lighting systems?

The north it's taking matter into its own hands.

And tbh, from the data in this articles, London isn't hugely different to the north east at the moment.

www.theguardian.com/society/2020/may/30/london-doesnt-always-know-best-how-the-north-is-choosing-its-own-pace-out-of-lockdown

I don't understand stats like you do BCF, but that wouldn't seem to be very good news for the soon to return primary school teachers, all working with 2-11 years olds...

Using the testing data to follow the outbreak is flawed by the massive increase in testing. 4000 positives out of of 10000 tests that we were seeing at the peak of the outbreak is significantly different from 2000 positive tests out of 100000 tests. IMO it suggests that had we had mass testing at the peak we would have been seeing tens of thousands of positives.
A better but more complex study would be to follow each positive test and see how many end up hospitalised. We know that most of the early positive tests were patients already admitted to hospital. If we have around 500 being admitted daily presuming that the average patient is admitted around day 10 this may give a rough idea of the percentage of positive tested people who go on to needing hospital care. I’m sure this data is being studied.

It is unfortunate if a lot of people are being infected - but if they are largely mild infections and few people die it's not a crisis. That's why the figures that interest me are the admission data and the excess death figures. The adissions show whether the NHS is potentially being overwhelmed, increasing the death rate, and the excess deaths show whether people are dying in the community before they get to hospital or dying early from other conditions that are not being treated.

Of those individuals providing blood samples,
6.78% (unweighted) tested positive for antibodies to COVID-19. (95% confidence interval: 5.21% to 8.64%); this equates to around 1 in 15 people.

I wonder how representative the sample is... I imagine that someone would be more likely to respond to a request to provide a sample if they believed they may have been infected.

An interesting view of excess deaths. As always using average excess deaths over a number of years skews the figures.
www.cebm.net/covid-19/covid-10-excess-all-cause-death-comparisons/
This group have been peer reviewing some of the data available and newly submitted papers. Worth a read.

@Piggywaspushed

I don't understand stats like you do BCF, but that wouldn't seem to be very good news for the soon to return primary school teachers, all working with 2-11 years olds...

The ons study tested just short of 19k people. Out of those 36 people tested positive across the 27 households. I'm not sure how much weight should be put on the age range spreading (although it is interesting). One infected family household with small children could easily influence the spread at this point. I think it's more interesting to watch this data over time as the ons continues its survey

It definitely is!

Lots of interesting discussion on T-cell immunity... It seems to be a widely held view by epidemiologists that viral load plays a significant role in how ill you get, and that a certain number of viruses are needed for infection to take hold.

I wonder if those with good t-cell responses simply have a higher viral load threshold.... but that they would become infected, and require antibody production, if viral load reached a certain level. So, whereas t-cells may offer a degree of protection, they don’t give immunity in the way antibodies do.... so it’s misleading to talk about “t-cell immunity”.

However, this is just my theory - I’m no epidemiologist so would be happy to be corrected if there’s anything I’ve misunderstood about immune response.

A lot of interesting views on CEBM site published during height of pandemic and largely ignored by the press, such as that it was clear early on that rather than being a pandemic disease Covid was behaving far more like a late seasonal flu. It was compared to Swine flu which followed more pandemic course. Clinically swine flu was killing 65 yr olds (roughly).
The WHO were reluctant to call it pandemic and if I remember correctly they suggested it be treated as pandemic rather than formally announcing pandemic.

I remember that swine flu was significantly abated by the summer holidays. But probably not comparable.

What does that mean in regards to its course over the next year?

T cell immunity is the bodies short cut, T cell cross reactivity will gives us a quick reaction to pathogens that may share similar biochemical or structural similarities to ones we have been previously infected with. For example the common cold virus.
Children’s T cell immunity is much more active because they have yet to build up antibody immunity which is a “library-like” reserve. Also it takes time for antibody activation so you need T cell response immediately so the body doesn’t become overwhelmed by infection before antibodies kick in.
Our T cell response is impaired by age and disease which is why the elderly are often very ill with what would be mild infections in younger people such as pneumonia, flu, norovirus. These viral diseases probably mutate rapidly rendering antibody resistance useless from year to year which is why they are always a seasonal problem in the elderly and vulnerable.

My apologies, the 27 households were the ones tested between 11-24may while the data set above covers 26th April to 24th may. For the 27th April to 10th may there were an additional 33 people from 30 households. So looks like we are talking in the region of 60 households. Again I think this is interesting but still early to drawing conclusions