Daily numbers, graphs, analysis thread 3

[Barracker]

Welcome to thread 3 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.

Not sure if anyone's looked at the covid-19 age stats, but for children they are very good, for people middle aged not so much.

Firstly 2017 respiratory deaths:

influenza - 458
pneumonia - 27181 (often caused by influenza)
chronic obstructive pulmonary disease (nearly all caused by smoking) - 28,597
asthma - 1320

Considering ONLY influenza + pneumonia and comparing with covid-19:

0-19: 65 (0.2%) vs 5 (0.1%)
20-39: 161 (0.6%) vs 43 (0.8%)
40-59: 978 (3.5%) vs 404 (7.1%)
60-79: 5385 (19.5%) vs 2259 (39.9%)
80+: 21050 (76.2%) vs 2944 (52.1%)

So it's relatively much more dangerous than influenza for people below 80, but less relatively dangerous for children.

It isn't clear the breaking point, but clearly, say 20-25 is still at low risk, and presumably say 35-39 have a relatively much higher risk than from influenza.

thank you for this thread and contributions, it is very good.

I am interested in the speed of hitting "peak". As a previous poster said, it seemed that the US were a lot further behind the UK, yet NY particularly looks to be peaking before us. I understand that NY took speedier precautions and it's a city so different than a country.

Question : is there a ratio or speed metric for how the virus is progressing through an area that we can look at as a measurement ? hope that makes sense, rather than the input factors - is there a hard number / progression rate.

thank you and apologies if I have missed it

stormy One metric of CV progression in a country or city is to look at the speed of doubling cases
i.e. whether cases double say every 3 days or every 10 days or whatever
NYC was doubling about every 2 days at one point

Some graphs of cases have dotted lines on 2 days, 3 days and 10 day,
so we can see into which sector each country falls

However, that is not specifically speed at which peak is reached - I haven't seen one for that

A country may have a target rate of doubling before relaxing measures and this target will be n estimate of what the health services of that country can handle with safety margin

thank you, very helpful

Unfortunately it is hard to measure cases without random testing, which is generally not happening.

Deaths are massively lagged, but are the most reliable figures

Do people think that study predicting 66,000 UK deaths is accurate?

The study seems to use some odd assumptions , way too few ICU beds for example

But it's impossible to predict the final total number of deaths as it depends on what type of exit strategy we decide to peruse and how effective we are at perusing it

If we go for suppress it will be way fewer than if we try to keep hospital admissions lowish ( eg run at winter loading over the summer) ( an unstable approach I think )

www.sciencedaily.com/releases/2020/04/200406120130.htm

Good positive article on treatment.

Bigchocfrenzy
China now alleged to have covered up large numbers of deaths
(one reason given why Western govts underestimated the crisis in the crucial early weeks)

Many western governments have constantly suggested that ANY data from Chine should be taken with a pinch of salt.

Here we have a life threatening event that surely those same Governments did not think that Chine was over estimating the numbers.
No.
So they should have been assuming the numbers were being HEAVILY UNDER ESTIMATED.
So it is even more, to me, a mystery why the British Government did not act earlier. Even if the British medical experts were saying one thing, with the WHO saying the opposite and Chine being the measure the Government went against normal logic.
I can not forgive them for that.

Thank you for this thread, I mainly lurk but find it very useful.

Can I ask if anyone can tell me where to find no.s for those in hospital and for no.s in intensive care? I am in france and track the no.s there and they give these figures each day which i find useful.

Thanks

This is from the USA part of worldmeters April 6th.
QUOTE
An estimated additional 180 - 195 deaths per day occurring at home in New York City due to COVID-19 are not being counted in the official figures. "Early on in this crisis we were able to swab people who died at home, and thus got a coronavirus reading. But those days are long gone. We simply don't have the testing capacity for the large numbers dying at home. Now only those few who had a test confirmation before dying are marked as victims of coronavirus on their death certificate. This almost certainly means we are undercounting the total number of victims of this pandemic," said Mark Levine, Chair of New York City Council health committee [source]
END QUOTE

How this effects the statistics who knows!

Sorry just re- read and it is not clear. I mean for cases in the UK not france.

FATE destiny
*What the government is trying to do by making sure the NHS has beds and capacity is to ensure that no one ever has to be turned away from a hospital if a hospital bed is requested.

So as long as the NHS maximum is not breached, there will be beds for those in care homes or at home if they want / need that bed.

I'm not in hospital logistics but I'd imagine such patients may have to travel longer distances to find beds, but the whole point if this policy and data tracking is so that those hospital beds are there for everyone who needs to be admitted.*

That does not take into account the policy of NOT sending older people in care homes to Hospital.

Gfplux
I agree wholeheartedly with your last point. Why did there not seem to be a pandemic plan for such as this? Surely PHE should have been on top of this? e.g a list of approved research labs that would immediately be put on to churning out PCR's? Some are now doing it but it's taken time. Also, why no purchasing of PPE earlier. I have non-medical friends who were just following the news and bought PPE in Jan and Feb at low prices. I would have thought this sort of thing would have been done by PHE /flagged up to NHS months ago

There were extensiive pandemic plans written by govt boffins many years ago,
but for a bad flu and didn't apply well to CV

Those original plans, together with our world-ranked scientists in the field,
made us #2 (behind the USA at #1) in international ranking for being prepared for pandemics !

They led imo to a false sense of superiority and complacency for the first several weeks of the crisis
However, they had to be junked and new plans made on the fly.

The problems with the original plan are the spread of CV before any symptoms and the much greater fatality rate than "bad flu," both of which change everything

Those original plans for disposal of bodies had nearly as many pages as the plans for handling the epidemic

• the "bad flu" plan was pure herd immunity, little to save lives, hardly any testing, so the facilities were never there

It was like being trained and well equipped for a cricket match, then arriving at the ground to find you have to play football instead.

There was a pandemic plan but they were expecting a flu pandemic where testing has less value. With hindsight that was a huge misjudgment. They should have taken a broader approach. However, virus is new and more infectious than than flu and SARS

Spiegelhalter explains his thinking here
www.bbc.co.uk/sounds/play/m000gwy8

Doesn’t at all help that governments everywhere were given false hope by initial reports and false stats from China

Surely the 10% who die of coronavirus are separate to the other 10%? Seems really odd to correlate them like that, almost as if they are suggesting CV isnt an issue. It's very confusing

I think some people are using that data to suggest that CV is not hitting life expectancy much more than it would without it. Which I find rather harsh. I would not want any of my relatives to die in isolation with those symptoms.

The truth is we don't know how many of the 10% of over 80's (for example) who die of CV would have died anyway. There will be some overlap, but people without CV continue to die from other causes and people who would have lived longer will die from CV. So it could be a lot more.

Also note that the chart on the BBC which @QuentinWinters quotes says that 10% of those who get CV will die. So the numbers will move depending on the infection rate. That's massively important.

As I've said, my issue with it is that it's not clear if the coronavirus death risk data v likelihood of dying as a woman or as a man was adjusted to reflect the lower likelihood of dying as a woman from/with coronavirus that the figures so far show.

I haven't yet listened to the radio programme posted earlier so it's possible Spiegelhalter clarifies this.

Some stats from Luxembourg showing how younger people are also at risk.

Rebelwithallthecause
Doesn’t at all help that governments everywhere were given false hope by initial reports and false stats from China

Governments should and did know not to make huge decisions based on statistics from Chine.
Governments (particularly the British and USA) were NOT mislead, they chose to ignore the reality.

According to Spiegelhalter the graph for the risk of dying from CV (if contracted) against age almost exactly tracks the risk of dying from anything against age.
So the pandemic means that you get a whole year’s worth of deaths crammed in to a few weeks, overwhelming hospitals, mortuaries and so on. (That’s the 600 000 deaths figure.) Hence the need for societal measures to mitigate this.

But there appears to be a lesser risk for women (at most ages) of death from coronavirus relative to men if we base it on the intensive care audit figures. I just want to clarify if that has been taken into account or not. Otherwise I think it's misleading to have female and male general risk of death.

I appreciate that there are limitations to that.

Thanks very much to all on this thread, I'm a lurker as I have very basic stats knowledge myself, but I find all your explanations and analysis so useful.
I wanted to link this article which people might find interesting, it seems to be a different way of presenting graphs and of representing numbers of deaths clearly, but I don't know enough to say if it's any good theconversation.com/three-graphs-that-show-a-global-slowdown-in-covid-19-deaths-135756?utm_medium=email&utm_campaign=Latest%20from%20The%20Conversation%20for%20April%207%202020%20-%201586915196&utm_content=Latest%20from%20The%20Conversation%20for%20April%207%202020%20-%201586915196+CID_81a602a233d9c69d7b82c835fb17f8f8&utm_source=campaign_monitor_uk&utm_term=these%20three%20visually%20stunning%20graphs