Meet the Other Phone. Only the apps you allow.

Meet the Other Phone.
Only the apps you allow.

Buy now

Please or to access all these features

Covid

Mumsnet doesn't verify the qualifications of users. If you have medical concerns, please consult a healthcare professional.

Daily numbers, graphs, analysis thread 7

981 replies

Barracker · 28/04/2020 12:53

Welcome to thread 7 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.Flowers

OP posts:
Thread gallery
127
ChazsBrilliantAttitude · 05/05/2020 08:06

Coronavirus: France's first known case 'was in December' www.bbc.co.uk/news/world-europe-52526554

Eyewhisker · 05/05/2020 09:27

The most interesting thing to me about the Bonn study was the relatively low rate of transmission within households - way less than 50% chance. This is surprising as clearly the virus is contagious if it has spread from China all across the world.

One possibility they mention in the study is that some people may recover from the virus by T cells rather than producing antibodies. This would make the total infected/immune population higher than the 15% they report.

sleepwhenidie · 05/05/2020 09:30

Oh that’s interesting Eyewhisker are you relating the two things...so you mean that you think some family members would actually have had it but fought it with T cells?

StrawberryJam200 · 05/05/2020 10:04

I've just googled T cells but don't really understand them. What would be the implications of this were true for individual and herd immunity?

NewAccountForCorona · 05/05/2020 10:13

If that is the case, Eyewhisker, will those people show up as "immune" on antibody tests? Presumably not, so while they may be immune they won't count in any figures.

Immunity via T cells would also explain why many young, otherwise healthy, healthcare staff dealing with Covid patients have not, seemingly, been infected at all.

ChazsBrilliantAttitude · 05/05/2020 10:19

This is the comment on T cells in the study

“ Given the high contagiousness of SARS-CoV-2, one would expect high rates of transmission. However, in our study we found a relatively moderate increase of the secondary infection risk which depended on the household cluster size (increase from 15.5% baseline risk by 28% for two people, 20% for three people, 3% for four people). This finding is consistent with recent
33 34 observations of secondary infection risk of 16.3% in Chinese and 7.56% in South Korea .
The reason for the comparably low secondary infection risk despite the high rate of
transmission is currently unknown, but it is seen with other respiratory infections such as
35 36
influenza (H1N1) 14.5% or SARS 14.9% . Secondary household members may have
acquired a level of immunity (e.g. T cell immunity) that is not detected as positive by our ELISA,
26,37 but still could protect those household members from a manifest infection.”

ChazsBrilliantAttitude · 05/05/2020 10:20

The random numbers are references.

wintertravel1980 · 05/05/2020 10:22

Yes, looks like traditional antibody tests may not give the full picture of historic infections.

www.bbc.com/future/article/20200421-will-we-ever-be-immune-to-covid-19

The hope is that a recovered patient has developed enough Covid-19-specific antibodies to fight off a secondary infection. However, in one study on convalescent patients in China, 30% of those studied had very little or no detectable antibodies in their blood plasma. It would seem that those people were able to neutralise the infection without the need to develop antibodies, either because their innate immune response or the T cells in their adaptive immune response, or a combination of both, were sufficient. Those with the lowest counts of antibodies were most likely to be the youngest patients.

We may need to look at results of antibody testing by age bucket to quantify the true scale of historic infections.

sleepwhenidie · 05/05/2020 10:23

So interesting. Is there any way of (ethically) verifying immunity via T cells?

NettleTea · 05/05/2020 10:40

thats was what Id been wondering - if your body can deal with it quickly, by using the non specific immune response, then it would not have got to the point of building any antibodies, so there wouldnt be any in the system.
It takes a while for the specific immune response to kick in if this is a new exposure (it takes a little while for it to kick in even if you have previously been exposed and have the antibody in your system, as you need to produce an awful lot of them to fight it off) But once the antibody has been created there will remain a small number later

when I was taught immunology we were given a simplistic overview initially.

this is a good overview

NewAccountForCorona · 05/05/2020 10:49

If this virus can be resisted by T-cell immunity, that would explain why relatively few children and younger people are becoming infected.

It's the opposite of the Spanish flu, which affected mostly the young.

But it will completely screw up any idea of "immunity passports" as some people, no matter how many times they have been in contact with the virus, and even if they have actually had it, won't show as immune.

sleepwhenidie · 05/05/2020 10:57

newaccountforcorona why is that, do children/young people tend to have more/more active T cells? Apologies if stupid question (I am no scientist!)

Eyewhisker · 05/05/2020 10:59

That’s true. If there is significant T cell immunity among younger people, the R will be much lower than implied by models that assume that all those without antibodies are vulnerable.

I am still struck by what is happening in Sweden, with falling infections despite much fewer restrictions and very few deaths in the under 70s. The Imperial college model estimated that their lighter touch approach would result in an R of 3, but the latest data from Sweden is that despite less social distancing their R is below 1. It is likely that some level of immunity is slowing the spread.

ShootsFruitAndLeaves · 05/05/2020 11:14

Here's the death date to 24 April. Ignore reports by week of registration as it's the wrong way to do this. Note that the data are lagged.

Note also that week 15 deaths are now 11.5% higher at home than reported +8 days after w15 end, but hospital deaths only 1.5% higher.

Also care home deaths up 6.4% since w15 end.

The week 15 data and week 16 & 17 data are not directly comparable in that a number of 'other communal establishment' deaths were re-coded as hospitals, etc.

Hence it may be more illuminating to note:

Week 16 +8 -> Week 16 + 15

Home +10%
Care homes +4.6%
Hospital +3.2%

Thus deaths at home in fact fell by only 55 (1.5%) vs last week, and are still around 50% above normal.

Care home deaths fell by 500 looking at data in both cases +8 days, which is quite good although we should consider that this may not be a fall as a % of the surviving population.

I.e. we have around 16,000 excess care home deaths in 5 weeks, out of a population of around 300,000, plus the 12,000 'background' deaths, so it follows firstly that covid-19 has killed well over 5% of the population of care homes, and probably will kill somewhere around 7.5% of the total population of care homes. AND we should consider that families if they have any choice are probably NOT sending granny off to be possibly killed by covid-19 in a care home, so hence the population of are homes is likely to be down by almost 10% by now due to natural death and fewer new patients, as well as covid-19 killing off much of the population.

So e.g., if there were 290k residents pre-week 16, and 7500 died, that leaves 282,500 residents in week 17 then that's 2.59% and if 7000 died in week 16, that's 2.41%

So there is a good chance that care homes will develop a measure of herd immunity, in that if we started with 300k and with 260k or whatever post covid-19, then those survivors will not catch covid-19. The problem is of course is that when the new residents start arriving then they won't have immunity and by January, we can expect 80,000 to have died of natural causes, perhaps 90,000 new residents arriving post covid-19 (entry to care homes delayed by covid-19), and half the immune herd will be dead anyway.

As regards total death toll, we can see we have

14,400 hospital deaths (but here we can just look at NHS data)
Around 7,200 home deaths
Around 18,000 care home deaths (but more not yet counted, more dead since 24 April)

Thus there were certainly 40,000 extra dead to 24 April.

TOTAL excess dead was 10,000+ in week 15, 9000 in week 16, and 6500 currently for week 17, but the latter figure is missing maybe 1,000 not registered yet, and previous weeks smaller numbers.

If we look at NHS stats, we had around 3250 deaths of week 17 to a few days later and now that's down to 2400ish. So it looks like we will have a LONG period of excess deaths, i.e. there is no possible way whatsoever that there will be fewer than 50,000 excess deaths in England & Wales.

Daily numbers, graphs, analysis thread 7
ShootsFruitAndLeaves · 05/05/2020 11:17

Sorry just to clarify above, care home deaths are down slightly as a % of surviving residents. As observed elsewhere probably 85-90% survive, in this group, so to the extent that many people have died in a particular care home, it could be a good thing for the survivors, cf. somewhere where no-one has died.

wintertravel1980 · 05/05/2020 11:19

Yes, I really wish more scientists challenged the Imperial model.

I am also very interested in what is happening in London (I know I sound like a broken record but still...). London deaths plateaued earlier than expected which means R0 was below/close to 1 for several days prior to the actual lockdown. The numbers have been falling rapidly since then. On April 26th London had 46 hospital deaths while most other regions in England reported higher numbers. The only exception was South West (who had 23 cases).

I keep wondering whether C19 transmission rates in London are impacted by a meaningful proportion of immune individuals in the overall population.

wintertravel1980 · 05/05/2020 11:20

Sorry - wrong thread.

BigChocFrenzy · 05/05/2020 11:40

Scientists have to have evidence to be credible when they challenge the work of other scientists,
whether from finding specific flaws in the work, or from a fresh study of their own

Of course rl is always the final proof;
hence any model must be continually updated with rl data,
which to be fair the Imperial model was after plugging in data from Italy, instead of just the (fake) data from China

No model can cope with deaths being included from 0-4 weeks in the past - they would predict deaths on a particular day, whether current or in the past.

So, modelling will always show smooth curves and smooth corridors, rather than the very noisy oscillating rl data

Looking at Imperial's prediction for Sweden, it's not that bad, within the limits that any maths model will have

Daily numbers, graphs, analysis thread 7
Daily numbers, graphs, analysis thread 7
RedToothBrush · 05/05/2020 11:48

I don't think I've seen much on this but there is a possibility that blood group might be playing a role in immunity too. We don't really understand why we have blood groups but it's thought to be something to do with viruses and protection. They think that blood group O evolved due to the presence of Malaria and its known that certain blood groups offer a certain degree of extra protection against certain diseases. I believe there was an early suggestion that blood group O was having more resistance to Covid-19 than other groups and there appeared to be a higher number of serious cases in people with blood group A (which would beg the question is there a difference between people who were AA or AO).

I have also seen quite a lot about 'happy hypoxia' and patients turning up at A and E with low levels of oxygen in their blood and being like people who had been acclimatised at high altitude. This got me thinking whether any research around altitude sickness might offer some clues as to what's going on.

Altitude sickness can cause high altitude pulmonary oedema (HAPE) - a build-up of fluid in the lungs and High altitude cerebral oedema (HACE) (swelling of the brain caused by a lack of oxygen). It can affect young people who are otherwise fit and healthy.

It was thought to be random, but research has established that by look at (I think its 6) genes they can predict who is at high risk of altitude sickness with 95% accuracy. Certain populations have genetically adapted to living at altitude - so it makes me wonder if they would be more immune to Covid-19)

The similarities between the covid-19 and altitude sickness have got me wondering. It might help to explain the sudden increase in apparent strokes in younger populations if covid-19 is behaving like altitude sickness.

The treatment for altitude sickness is to descend from altitude as quickly as possible to increase oxygen in the blood (and may include supplementary oxygen).

I find it a fascinating subject and it just highlights how little we understand how the human body actually works in an era where we have this idea of being so knowledgable but when it comes down to a crisis of global magnitude we know next to fuck all.

Eyewhisker · 05/05/2020 11:55

RTB - that’s interesting on the genetic element. I am struck by the instances of twins, siblings, family members dying which seems more than would be expected if the rate was equal for everyone.

Of course, living here I have absolutely no idea what blood group I am. What are white British generally?

titsbumfannythelot · 05/05/2020 12:02

You can look at blood donor services and find some info on blood types etc there @Eyewhisker.

I know I'm B+ and that around 15% of the population is this blood type.

NewAccountForCorona · 05/05/2020 12:07

I wonder could blood groups have any effect on the proportion of BAME people being seriously affected by Covid.

A quick google suggests higher rates of rarer blood types in the BAME population, though I don't know enough to even try to find definitive scientific data on this.

MillicentMartha · 05/05/2020 12:24

‘44% of people in the UK have O group blood; 42% have A group blood.

The least common groups are B (10%) and AB (4%).

85% of the population are Rh D positive and 15% Rh D negative.‘

The B blood group is more common in BAME, more like 20-25%

RedToothBrush · 05/05/2020 12:33

www.bournemouthecho.co.uk/news/18289782.ethnic-minority-blood-donors-needed-says-nhs/

Mr Stredder added: “We would welcome more black blood donors.

"There is a difference in red blood cell groups between different ethnicities. If we give blood from white donors to black recipients, even though we match for the major blood groups, it is more likely minor mismatches can occur and cause reactions.”

If different ethic groups have differences in red blood cells (which carry oxygen) then this might have an impact. It fits with the blood type research and the altitude sickness similarities.

I'm surprised there hasn't been more discussion of the role of blood itself tbh with the focus on antibodies only with the suggestion that this isn't merely a respiratory disease but a vascular one.

BigChocFrenzy · 05/05/2020 12:44

The earlier study on blood types in China is regarded with suspicion,
as blood type theories on diseases, or on best weight loss diet have a dubious history

However, if the different blood groups carry / release oxygen slightly differently, then that could be important in lung and circulatory diseases

  • but tat isn't the aspect of blood ype that seems to have been studied / published so ar

The Chinese study suggested that Blood type O had a lower death rate

However, in Scandinavian countries, which have the lowest death rates, whatever their aocial distancing policies,
Blood type A is the most common

The wiki list of blood types for countries & ethnic groups is controversial & disputed,
maybe because of over-sensitivity about ethnic differences and because .... politics

https://en.wikipedia.org/wiki/Bloodtypeedistributionbyycountry

However, the maps of blood type distribution are interesting:

Daily numbers, graphs, analysis thread 7
Daily numbers, graphs, analysis thread 7
Daily numbers, graphs, analysis thread 7
Swipe left for the next trending thread