Daily numbers, graphs, analysis thread 6

[Barracker]

Welcome to thread 6 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.

The government are not writing them off, because we have lockdown
and hopefully a science-based strategy to exit it in stages

Those who keep pushing the narrative that the dead would have died soon anyway did not agree with the lockdown in the first place
and want to return to their "normal" pre-COVID life

I think it would be good to keep this thread for discussing data and research.

The final death rate will depend on the strategy followed until we have an effective treatment, eventually a vaccine

If the UK just abandons the science-based strategy and goes for herd immunity, that is likely to make a considerable difference to that death rate.

Possibly COVID will mutate and die out as a serious threat anyway within a few years, but we can't rely on that.

Plbrookes you clearly also have an agenda - to other those who have died and will continue to die when lockdown is lifted. The continued refusal in official briefing to mention the probable role of vitamin D makes it quite obvious that the agenda is actually to let a lot of the vulnerable die.

No-one knows how long an individual would have survived, there have been some surprising recoveries in the very old and some really unexpected deaths in fit and healthy people. Therefore assuming average life expectancy is the only thing that makes sense.

@alreadytaken
Your post is a disgrace. I have pointed out weak esses in a piece of research. Don't project your dishonesty and spitefulness on others.

I posted we should agree to disagree,
since this is a stats thread and I now post papers on the studies thread

Actually, if you look at the caveats on the paper itself, the authors acknowledge that their results may be an overestimate for the reasons PIBrookes and I have stated. Is this because they have an agenda or are they simply reporting the obvious caveats to their analysis?

But let us stop this, as it is ruining the thread for everyone else. We need dispassionate analysis, not accusations.

Plbrookes you pointed out what you see as a flaw because it suits your agenda. You claim others have an agenda while refusing to admit to your own. I simply pointed out the fallacy in your argument - but you resort to insults to defend the indefensible.

I agree!

Oh I didn't agree with @alreadytaken s dishonest post of course, but with Eyewhisker!

This goes to show the difficulty statisticans have in analysing data without preconceptions.

It's difficult for anyone to do.

But we shall all try our best to analyse the data dispassionately.

There are caveats, which is why I said it is as good a study as can be made,
within the obvious limitations of not being able to research each patient in detail
and having to take averages

and that I think "several years" of life expectancy is rather more likely than 10 years

However, the important issue with the paper is to counter the frequent claims that the dead were mostly in the "would have died soon anyway" category

The important issue with the paper is whether it provides evidence on how many years of life have been lost as a result of Covid19. And we still don't know.

It provides reasonable evidence that the lost life is several years;
that the dead were not about to die anyway

If you have another study suggesting only a very short remaining life expectancy - by scientists - it would be very useful to weigh up the 2 papers

Pilbrooks, there is no definite way of concluding that the people with the lower life expectency are the ones who are dying. I know an 85 year old with MS and severe dementia who has tested positive and survived, and a very fit 79 year old with well controlled diabetes and a past history of cancer who died. The 79 year old would most likely have lived another 10 years; the 85 year old was more likely to have been near end of life.

It's not possible to conclude that the sicker old people are the ones more likely to die; I haven't seen any figures that have proved that at all.

Many such scientific papers with estimations have caveats, if the authors are responsible

but that is no reason to reject the conclusions that the people who died probably had several years left on average and would not have died within months

And calling people disgraceful and dishonest, on this thread is childish and unnecessary.

I disagree.

Whatever the lost life years actually is, I don't think that the research methodology, as I understand it, would give a reliable or unbiased (in the statistical sense) estimate of it. If someone can explain how the methods used allow for the issue I've raised I'd be very interested! I have no interest in having a high estimate or a low estimate, I'm interested in the truth.

I'm not aware of any other research on this which is why, as I said at the start, I was so interested to look at it.

Anyway, I repeat:
let us agree to disagree and move on to stats

Research showing average loss of year of life to be 10 years.

twitter.com/FergusJChadwick/status/1253269399404007427

*years of life lost.

Ah, sorry, I see it’s already been posted and is what started the ‘discussion.’

@NewAccountForCorona
Your first post just repeats the problem. We've don't have adequate evidence on the true life expectancy of those who died of Covid19. That's the point of what the research was trying to achieve.

You can read the post I responded to which I described as disgraceful and dishonest. It was disgraceful and dishonest. This thread above all should allow discussion of data and research methods without hysterical accusations of trying to 'other' people if you point out methodological flaws.

Here's separate sex-based death rates by age. The exponent for women is 0.100 age, and for men 0.111 age. This results in a doubled risk of death by age 60+ for men compared with women of the same age.

Note that I have re-normalised the age based on the average age of the population, as the ONS stats are for age bands 0-1, 1-4.99, 5-9.99, 10-14.99, etc. In most cases the average age is exactly the middle of the band, i.e. 0.5, 2.5. In fact because death risk is growing exponentially with age, the average age/death risk will be more like 3.0 than 2.5 years above the bottom of any given 5 year band, but this isn't too critical.

Okay