Daily numbers, graphs, analysis thread 8

[Barracker]

Welcome to thread 8 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.

I think london could be as high as NYC as we locked down late and kept public transport and 50% of people working.

I think here like usa a lot of people died at home. I would expect the excess deaths to drop alot now.
But even on some data yesterday we have more in icu than hospitalised which seems very wrong...
So i would expect more antibodies than italy france spain and Germany.

Although i think it is likely uk had infections pre-xmas i think we would have seen fewer deaths of KW now if so. However because of the initial dose it seems likely people infected arkund times when lots of people are shedding it are more likely to die. So maybe many medics had little exposures around xmas.

Maybe they need to look back at deaths on cruise ships and of people after cruising.
And of anyone dying within a month of flying.

Uk probably has a lot of chinese students not sure how it compares to the rest of EU?

That uk one where the choir person had a link to wuhan. They need antibody tests all of them.
It is likely uk dont want the blame from Trump for not noticing it spreading.
Afterall it only seemed to take off in italy at half term. Seems like due to the cold weather it would persist in the environment very well.

Isn't the most reasonable explanation for dementia Alzheimer's death that they were in fact killed by covid-19 . This seems especially true as even as care home deaths have fallen they have become more likely to die of covid-19

Any putative lockdown effect should cause more people to die of heart attacks etc., but dementia seems like something that the lockdown wouldn't aggravate at all.

Has the ONS not yet published a release on individual death causes by week?

I find it slightly ironic that the tinfoil hat types are claiming that people who died of heart attacks are being diagnosed as having been killed by covid, when it's much more likely to be the other way round.

A 'deaths by cause of death and week' would disprove the ridiculous claims by people that there are suddenly thousands of extra suicide or DV deaths

Someone put a link a few threads ago about a study where the air con or heating in a restaurant blowing out air was found to be linked to the people who then came down with covid 19 after eating there. I can’t find the article anywhere, does anyone have the link or anything to help me find it? Would like to send it in to work who would like larger groups of people when required to meet in a room which has the fan heaters that blow warm air straight in you to work and I’d like to express my potential concerns based on the study.

It was a blog post rather than a study but it made interesting reading:

quillette.com/2020/04/23/covid-19-superspreader-events-in-28-countries-critical-patterns-and-lessons/

Reastiei

Is this it?
www.erinbromage.com/post/the-risks-know-them-avoid-them

Thank you, that’s perfect. Is there a reason why the govt isn’t seeming to mention air conditioners/fan heaters in office places as potential routes of transmission ? If it might spread it you’d think they’d look at how to minimise risk.

Saw this on another thread and thought it would be of interest

medium.com/wintoncentre/what-are-the-risks-of-covid-and-what-is-meant-by-the-risks-of-covid-c828695aea69

The London thing is interesting today

Not sure why this thread keeps disappearing for me from TIO

for anyone having trouble finding the thread - bookmark it Then you just need to carry on reading from that page and bookmark again.

Itsgettingweird re your concerns about DV and suicides accounting for some of the excess deaths:
V unlikely, because the figures are normally so much smaller than this number, that there would be screaming headlines if they rose to becomes a significant %

We are looking at several thousand deaths that were not put down to COVID, but:

The UK averages only about 700 murders of all kinds per year, so in 2 months about 120 murders

The UK averages about 1,000 suicides in 2 months

If COVID is only put down as a contributory cause, then afaik it is not included in the death certificate statistics
It may also avoid the longer process of post-death COVID tests - if not done while ill - and the longer process of dealing with the Coroner's Office,
which might be a consideration when dealing with bereaved families

Just click "watch" next to the OP

@AnotherEmma that article misses a whole bunch of points. In particular, the death rate at 90+ is more than double the usual rate, not 61% as it implies there. Almost certainly this is due to covid-19. And it doesn't really matter WHY, if you are dead, the point is now that people aged 90+ have more than DOUBLE the usual risk of dying.

There is in fact negative mortality among younger age groups.

It's faintly ridiculous to describe mortality risk in terms of normal death risk for young people, since young people have essentially no risk of death (for reasonable values of 'no risk'). If we consider a game of Russian Roulette where there are million chambers in the gun and then we halve that to 500,000, then we still should play the game, if offered a decent sum of money.

It does not follow even for a moment that if we consider our lives as Russian Roulette at a certain older age with a 1 in 50 risk of death each year, that it's remotely comparable for that risk to double from 1 in 50 to 1 in 25 as it is from 1 in 1 million to 1 in 500,000.

Also it seems a bit silly to group 40-44 year olds with 15-19 year olds, when the former is 10x higher at risk.

It also claims that BAME people are at higher risk of catching covid according to the ONS. However the ONS did not in fact claim anything of the kind, and only found that BAME people were more likely to have died from Covid-19. Other studies might show that BAME people are not more likely to die having caught the disease, but the ONS did NOT show that, they ONLY proved that they are more likely to have died of it.

It then goes on to claim "ethnicity was not an important risk factor for COVID patients who were hospitalised" except that, er, the study they link to

www.medrxiv.org/content/10.1101/2020.04.23.20076042v1.full.pdf

doesn't mention ethnicity at all!

This is only a blog post, but for someone with such grandiose credentials

"Statistician, communicator about evidence, risk, probability, chance, uncertainty, etc. Chair, Winton Centre for Risk and Evidence Communication, Cambridge."

it's not terribly impressive.

To be fair I didn't read it very closely so I am not going to defend it

It’s been difficult to find out what the excess deaths not attributed to Covid were actually caused by, although as BigChoc says, murder and suicide are orders of magnitude smaller so very unlikely to be significant contributors.

The ONS have just made a preliminary statement that dementia death rates are significantly up, and because dementia is such a big killer it seems likely that that’s driving most of the difference. Whether that’s the result of undiagnosed Covid or lack of proper care due to social distancing is the question that needs answering, and the answers are probably going to be deeply upsetting.

We know that covid also attacks various other sites in body apart from lungs - blood vessel/strokes, kidneys, and digestive system. Not a huge leap to imagine it can exacerbate Alzheimer’s.

@Oakmaiden
I find it so frustrating that there are so many pieces of "research" published by "respected" scientists and shouted from the media headlines when a decent read of the paper by even a layperson (ie me) can see glaring holes/ errors or simply dangerous assumptions.

There was an article in New Scientist about this last week. It was pointing out that everyone's thirst for new info about CV19 means research at an early stage, on pre-publication servers, is being read and taken as fact by people unqualified to scrutinise it properly.

Normally, they would be posted for peer review, and flaws in the research pointed out in a calm way. Now, they are making headlines, and it is undermining public perception of the scientific process.

Totally agree shoots re you post earlier about complete breakdown of deaths. I said the same and also because there will be decreases in some type of death causes (knife crime/RTAs are most obvious) and so that in effect actually makes excess deaths higher iyswim?

There a type of dementia called vascular which I guess if blood vessel are involved is a risk on a risk?

' research at an early stage, on pre-publication servers, is being read and taken as fact by people unqualified to scrutinise it properly.

Normally, they would be posted for peer review, and flaws in the research pointed out in a calm way'

Actually, I think the dichotomy between preprints and peer reviewed is part of the problem because people are taking them as absolute indicators of quality. I keep seeing people dismissing thoroughly decent research on medarxiv because it's not peer reviewed, while that rubbish Manchester paper actually HAS been through peer review and is being taken more seriously than it deserves because of it.

itsgettingweird Yes, I could well understand if vascular dementia were increasing, but I'm not sure why Alzheimers is

To expand on my previous comment, 'days of risk' doesn't really make sense in that for example, if a 90 year old has a carer who comes in daily, then at this time of year, the chance of that person passing on a infection disease such as influenza would normally be very small. Nonetheless, clearly for a given population of say 2000 90 year olds, then perhaps 1 will die on a given day.

We might say that covid-19 has doubled this so that now 2 die on a given day, but the thing we are concerned about is transmission of an infectious disease, and the risk of that has gone from near-zero to perhaps a 10% risk of death following contact with an infected person.

Clearly this is materially very different from saying that for example, of 5000 people aged 70 that 250 will die in a year, of which 50 will be from heart disease caused by lifestyle factors, 50 from lung cancer caused by smoking, (or whatever) and so on, which is roughly 5 deaths per week.

Clearly almost all of those 5 deaths per week are now non-preventable due to things like past lifestyle, genetic factors, etc. It's not particularly helpful to tell a dead man's children that he shouldn't have drunk so much in that it's not going to bring him back to life or fix his liver failure.

As such it makes a lot more sense to consider covid-19 in the wider context of risk assessment. For example we might consider that 'motorcycling is too risky'. This would presumably reflect the 120 deaths per billion passenger miles by motorbike as opposed to the 1.8 deaths per billion passenger miles in a care, an increased risk of 66x. However that doesn't mean that much unless we consider both the absolute as well as relative size of the risk. One death per 8 million miles is not that many - with an average of 4000 miles per year, that's one death in 2000 years of motorcycling, or a small chance that we will die if we ride a motorcycle our whole life (say 50 years = 1 in 40)

If we consider that a person in their late 60s may have a 1% risk of death from covid-19 infection, then it it appears very strongly that individuals in this age group would correctly judge that exposure to covid-19 by for example going out to restaurants is 'too risky', since many will have judged various other things 'too risky'. And that's true even though even if 1% of the same group will die in the next year anyway. It's very clearly the absolute size of the risk, not the relative value of it that's important here.

It does follow in general that we should take larger risks as we get older, since we both have less time left to life, and a risk of constant size will become relatively less serious as we age, however for covid-19 since the risk is proportional to underlying natural death risk, this definitely is not the case - we are taking an already unpleasantly high risk of dying (as old people) and making it even bigger. Sorry, no.

"there is a question whether there is such a thing as T-cell immunity and if people who got C19 mildly and did not produce antibodies may get reinfected."

WIntertravel and it's not whether these (mostly young) people could do the same again if infected - the young have only a tiny risk from COVID -
but whether they could then infect others who are vulnerable

Nonetheless, clearly for a given population of say 2000 90 year olds, then perhaps 1 will die on a given day.

I should say that is 1 death per day from all causes. Not from influenza.

@BigChocFrenzy I believe Alzheimer’s doesn’t cause death per se, but due to difficulties swallowing may cause vulnerability to other infections, dehydration, cachexia, as well as immobility related problems.

I think when people get to a certain point there is a chicken and egg labelling with cause of death.

SHoots That's basically my personal calculation as a 63-yr-old:
COVID is a completely additional risk I didn't have before, not huge, not something I ever lie awake worrying about

However, even with low cases here atm, I reduce my exposure without cutting out activities that are very enjoyable and / or healthy.
So, 1-or 2-hour daily Rhine walk, but only my fav outdoor restaurants and 5 x weekly gym (main risk) without sauna, shower or coffee bar socialising afterwards