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.

@ClarasZoo - I think you missed a few things:

1. overwhelmed health services mean a higher percentage of people dying
2. new treatments may mean even less dying
3. the trauma of people managing all that death
4. post viral fatigue making people useless economically
5. vaccine would be needed anyway to prevent this coming back in future
6. industries would collapse with everyone off sick, They're better protected with a controlled short-term closure.

Yes fair enough those are all good points. 4 and 6 are ones I had not thought of. I was a little bit comforted by the IDR of less than 1% overall though. It's not the numbers is it - it's the volume in a short space of time...

@NeurotrashWarrior

I saw the t-cell report. If I remember correctly there was a suggestion that younger people might use t-cells to defeat the virus but that this might not show up as antibodies in the blood? It will be interesting to see how the research develops.

It looks that way. I forgot one other thing, as humans we're noisy and it looks like shouting / singling spreads the virus faster and may lead to more severe infections. As long as we aren't back to normal life is quieter, so infections may be less severe.

Clara why are you not counting any over 65s in your calculation as worth saving? Are they not worth it?

Yes I remember something about that too sunshine

@ clara, isnt that calculation though based on figures skewed by a lockdown though?

Please could someone do me a massive favour and tell me whether I can take what the OpenSAFELY factors associate with Covid-19 hospital deaths Preprint on MedRXiv (BMJ/Yale) at face value?

I think it's saying that - age apart - there is only a very small increased risk of death with diagnosed hypertension? This would change everything for us so I will be hugely grateful to anyone who can look at the study and advise whether it's solid. The researchers' platform had access to 17 million NHS records.

"There was no association
between hypertension (defined as a
recorded diagnosis, or high blood pressure
at the last measurement) and outcome (HR
0.95, 0.89-1.01). However, in sensitivity
analyses, diagnosed hypertension was
associated with slightly increased risk (HR
1.07, 1.00-1.15) while high blood pressure
(≥140/90 mmHg) at the most recent
measurement was associated with lower
risk (HR 0.61, 0.56-0.67). "

I think in the early research from China hypertension was a risk factor, in more recent research it was much less of a factor.

Also telling all over 65 year olds to stay in? How long for? One week, two weeks, a year? The seventy year olds i know lead very active lives and would have strong views on this rightly so,

cathyandclare - thanks. The log scale for hazard ratio seems to be saying virtually no increased risk at all - I think...?

Can you see any flaws in the methodology of this research? I think they've managed to untangle all the various comorbidities haven't they, even if one is known to be associated with or contribute to the other?

I'm wishing for about the fiftieth time this week that I was better at stats. I just cannot do it.

@ClarasZoo

Applying your analysis to the rest of the UK, out of 36,000 deaths, with 8,000 in London, 28,000 have occurred elsewhere out of a population of 57 million.

If infections are at 5%, that gives a fatality rate of very close to 1%, almost double London’s number! Even taking account London’s somewhat younger population that’s a big difference.

I’ve no idea what the sample size is though... I’ve heard 1,000, which of true would me quite a margin of error if analysing at a regional level.

@VitreousHumour - the big difference between diagnosed hypertension and high blood pressure readings might be drugs. So it could be that certain blood pressure medications are a risk factor, whereas high blood pressure itself might not be. (Caveat that I've literally only read your extract and am not medically literate.)

It may also be that most people with known heart conditions are shielding, so less of them have been dying but that wouldn't apply to care home deaths so it's less likely to be a reason.

I think the problem with that is younger people are

• Still hospitalised for oxygen or icu maybe 1% of 20-30yo and increasing by age.

Even those just hospitalised could have long term effects we already have 100 kids with new kawasaki type thing of which some will have long-term heart issues. So at 5% immunity over the uk thats 1k to 1750 to get to 75%?

NeurotrashWarrior i also read that bbc article and thought vit d.

clarexbp i'm.not worried he's at greater risk - I really don't think he is. I'm worried that should the worst happen to an autistic person that the government would record that in the "underlying conditions" category. Which would be very disingenuous IF that is the case.

I will try to find where I read it.

Callimanco
I cannot remember where I saw it but Autism was on the list of learning difficulties that could lead to a DNR order
However it was not a blanket rule
more that those who would not cope with invasive medical treatment
such as my cousin who lives in a sheltered residential home and has no awareness of the wider world but is happy and healthy
should not be forced to undergo something that would terrify them and not improve their Quality of Life.

It's in this article - 451 deaths with learning difficulties or autism.

I can see that a person with autism as part of a profound learning need might be at increased risk. Where do they draw the line at when it is described as an underlying condition and when not?

www.independent.co.uk/news/health/coronavirus-deaths-diabetes-england-covid-19-a9514761.html

I also noticed in a paper shared on here the other day with the "case descriptions" for the first few novel-kawasaki cases that one of the children was described as having "autism and adhd" as an underlying condition, which is what started me off wondering.

Callimanco
It will come out in the wash but I suspect the Learning Difficulties / Autism link is due much more to the difficulty of treating those who cannot properly cooperate than with any of their actual medical pre existing situation

My daughter is autistic and she cannot describe pain nor does she display it easily. She has had tonsillitis in the past and I genuinely had no idea she was even unwell until her tonsils were infected. Its a military operation to get her to take a paracetamol

I read somewhere that it is poorly controlled hypertension that is a problem
So when either the meds aren't working, or the person isn't taking them properly

Also the criteria for high BP was lowered some years ago in the USA
to 140 / 80 which used to be within the normal range
and many doctors in other countries have followed

So this level is probably not a risk (and may not be medicated unless you are in the USA ?)

Many people with hypertension are overweight and unfit
so it is one of a suite of factors
if the person with raised blood pressure takes their medication and stays fit and loses weight then the risk should lower

and yes, picking a number and medicating when the overall data is unclear may not have been the best approach.

But what we want to know is Hypertension on its own.

Not all Hypertension is due to bad habits.

It does seem to be a lot quieter now on Hypertension than it was earlier. And silence is probably a good thing. I think that bP drugs do not necessarily increase risk.

Having a spouse with Hypertension it has been worrying. But now due to his age, good health, being only slightly overweight, and reasonable levels of activity especially in his workplace (complete opposite of what Listening describes) i am no longer as concerned.

if infections are at 5%, that gives a fatality rate of very close to 1%, almost double London’s number! Even taking account London’s somewhat younger population that’s a big difference

It's really not. Firstly London is MUCH younger. Remember that the death rate for ethnicities than white British and Irish, just adjusting for age is expected to be just 1/4 of the white British/Irish rate. And London is very much less than 50% white British

Secondly London has fewer care homes, which seem to be killing the most vulnerable at a much higher rate than they would in the community.

Hadenough
My husband is on blood pressure pills. He is not overweight and is fit. Its hereditary.
BUT
my point stands about the vast majority
and
covid is very new so the data is not out yet