Daily numbers, graphs, analysis thread 5

[Barracker]

Welcome to thread 5 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
Google mobility stats

Thank you to all contributors for their factual, data driven, and civil discussions.

We are behind other countries and and have higher death rate - already higher than france overall and poss arcmych as Italy eventually once we peak and inc community deaths

We haven’t see the fall yet do we can’t plan

That’s surely common sense?

They can’t advertise a plan because the public, given a sniff of hope will break lockdown

We’ve shown we are only recently tskingbit seriously

Sorry for typos

Viruses usually have many mutations

The important point with COVID is that the experts so far have said these are not significant wrt severity or treatment,
or in developing a vaccine

Grateful for this thread. So much agenda-pushing on the empty hospital one.

Be good to reconcile what us actually going on with all these mixed messages (111 / empty hospitals).

France, Germany, Belgium and - I think - Ireland includes deaths in care homes,

so we probably need to add an extra 6,000 - according to the Financial Times - to the UK total for hospitals, if we want to compare to those countries

The US, Italy & Spain don't include care homes, so we can compare the UK directly with those

Italy had one of the most advanced health services in the world,
but their hospitals became overwhelmed because they tried to save everybody

UK hospitals are often refusing to admit patients until they are turning blue

• many grim accounts from MN COVID sufferers, inlcuding on these stats threads - .... which is a ruthless / cynical / incompetent way of avoiding the NHS becoming overwhekmed

This of course is one of the major factors in the UK death rate being higher than in some other countries, despite the NHS never actually being full up.

Interesting video re: asymptomatic cases in population.

On the non-COVID wards, many operations and treatments are being cancelled, because it is too dangerous with COVID infections rife

• risk of hospital infections and then of vulnerable people being infected after being discharged

Hence those wards will be empty

Similarly with A&E - many people are too frightened of being infected to go there unless absolutely desperate

When Merkel announced measures would be relaxed, she

• and later politicians of all parties around Germany -

warned that we must go very gradually, because we are walking a tightrope

She said the govt would rely on scientific advice at every stage

Exponential growth is frightening, hence the massive program ongoing of testing and contact tracing.

If R0 increases quickly to 1.1, then the German health system - which currently has about 50% free capacity - would be overwhelmed by October

Germany has a much lower death rate,
3-4 x the number of doctors, ICU beds etc and hospitals say they have enough PPE
Germany also has massive testing capability - currently 500,000 per week, to rise to 1 million later -
and a huge number of public health teams for contact tracing

BUT
despite this, everyone knows we are on a tightrope

Some parents & teachers are up in arms about schools startinfgfrom mid-May,
so the govt has to try to gently bring them along, to be able to restart the economy properly.

Some other employees are very nervous too, about returning later to work / commute where the 2m distancing is not feeasible

China can force people at gunpoint to return;
Western countries cannot.

asymptomatic cases in population
Also interesting about the plasma infusion I thought, and I find his videos quite calming to listen to 😊

Yes some of my friends who have tested positive are donating. This has been done in NY too.

John Burn-Murdoch@jburnmurdoch (FT stats geek)

Mon 20 April update of coronavirus trajectories

Daily deaths
• Still too early to say if US has peaked
• But beginning to look like UK has

Why do I say UK daily deaths may have peaked?

New chart: week-on-week change in daily deaths.

This gets rid of weekly reporting patterns and asks, are more people dying than at same point last week?

In UK, blue bars mean we’re now seeing fewer deaths than same day last week.
....
Now back to cumulative deaths:
• US death is highest worldwide and still rising fast
• UK curve still matching Italy’s

I wonder where Turkey has gone? At one point they were showing a faster rate of increase in cases than the US on John Burn-Murdoch's graphs.

Sorry, I just copied the UK bits of his v long thread

John Burn-Murdoch@jburnmurdoch

Cases in cumulative form:
• US curve beginning to taper?
• Turkey still battling a severe outbreak
• Japan has now passed Korea’s total, Singapore has passed Japan’s curve: both show the danger of thinking a country has dealt with covid

I tend not to post the Uk cases, because with so little testing, they are far more under-estimating the real number than many other countries

Unless countries test every single inhabitant - which would be like painting the Forth bridge - they miss those who are infected but don't yet have symptoms, or will never notice symptoms

So the cases for all countries are far below the real number, but the UK is extreme in its lack of testing
which is shown by the much higher than average rate of infection in those tests
and the much higher % death rate from those cases

Something on test accuracy here. 25-30% false negatives for current tests.

Possible reasons why explained here.

So even if we up testing more people we may still underestimate both symptomatic and asymptomatic cases.

We assume positive if symptomatic currently and isolate, but without testing a swathe of community samples in a region it will be hard to get figures for potential asymptomatic spreaders and they also may be more likely to register false negatives due to reduced viral load (which fluctuates).

Canada is including all the long term care deaths. Which might give people the impression our health care is shits.
At one point and maybe more now over 600 care homes affected.

@BigChocFrenzy Yes, what I would really like to see from countries is the number of people infected per number of tests and per capita. It's really meaningless to give bald numbers of infected people - of course countries like Korea and Singapore will show high numbers if they are testing thousands of people daily.

It makes me really scared for the UK when there are such high numbers as well as a lack of testing.

Maryland bought 500,000 tests from Korea today so it's not like there are no tests to be had. The UK seriously needs to invest in this and start controlling the situation.

it is sad to see professor Ferguson (if you believe the Guardian) comparing our Covid unlocking plans with our Brexit plans. Because, once scientists become political, they lose their scientific authority.

For the great majority of our civil servants , Covid is now the only game in town.

Scientists should stick to the science, keeping their political views private, especially if they are advising government. And, although it would be great to have scientifically literate politicians (Boris trying to explain an exponential curve was quite cringe inducing), politicians are entrusted to weigh up the ethical vs the economic.

The good thing about politicians is that if, as a society, we want to get rid of them, we can.

The dictator scientist is a very scary figure.

larrygrylls

It was a perfectly valid comparison. Saying there had been lots of brexit planning but no pandemic planning. How is that political? He was simply stating a fact.

Fromp,

Where is he getting his data from?

Unless he knows the amount of government time spent on both, which I suspect he does not. Surely, a statistician, of all people, should be wary of making suppositions without good supporting data.

In addition, he has to be careful of his role, which is to advise the government on what various measures would do to the epidemic, and not to try to influence their response via the media.

There was a huge amount of planning put into brexit. A whole department was set up specifically for that.

It is commonly accepted knowledge that our pandemic planning has been lax.

and he isn't a statistician.

As we all know very well by now, modelling and responding to an epidemic is not ‘clean’ science. The data we have is spotty and the confounding variables are many.

For this reason, it is particularly important for a scientist to try to stay apolitical. Once he/she starts to become egotistical and make a name for themselves (sorry about mixed pronouns), it is very easy to be tempted to slightly alter an input variable so that the model makes the point that you want it to make. Here, it is interesting to note the disagreements between the Imperial and Oxford teams, all talented statisticians and all working with the same data set.

BigChoc - I really would like an explanation as to why Sweden is not seeing exponential growth despite a more relaxed approach. The population mostly live in cities and by not having a lockdown they should be seeing exponential growth in deaths and an overwhelmed health service. But the primary schools are open as are restaurants and bars. Why aren’t they facing disaster?