Daily numbers, graphs, analysis thread 4

[Barracker]

Welcome to thread 4 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 reason I worry about obesity levels being quoted a lot is because it's the one underlying condition that society deems to be self-inflicted. For example, the number of comments about one of the nurses who died saying that she was overweight was terrible. As if "she's fat, it's her own fault".

Useful clarification of the differences in counting deaths - and the difference it makes:

1. Comparisons of COVID-19 death counts

www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/bulletins/deathsregisteredweeklyinenglandandwalesprovisional/weekending27march2020#comparisons-of-covid-19-death-counts

Sorry if this is a dumb question, but the stats that say 78991 people have tested positive and out of those 9875 have sadly died, would that represent a 12.5% mortality rate for anyone sick enough to need testing?

Yes, DuLang. UK PPE requirements aren't the same as other countries.

dd is working with Covid patients but because she's on a "normal" ward her PPE is surgical mask (one per shift, unless it gets wet or stained) and a plastic apron over her own uniform that she has to wash herself, and that is only in the last week. Before that, no mask until a positive test had come back, even if they knew the patient was more than likely positive.

The guidelines (sent by email, and which she has kept) have jumped around so much; they make no sense at all.

She worked two full shifts on a full Covid ward at the beginning of this with no PPE (even a mask) at all.

UK death toll 917

US now has over ½ million confirmed cases
also the first country in the world to have > 2,000 deaths in one day: 2,108 deaths

103,000 deaths total worldwide

Pertella, the figures I saw for the UK was about 12% mortality rate, more or less the same as Italy, but the UK has fewer (at the moment) tested and positive cases and a lot fewer tested overall, so until it's over its difficult to categorically state the mortality rate. But, as of yesterday, on the figures given, yes.

Newaccount That's what worries me. And it's not just hospitals, all care settings now are very likely to have covid +ve patients, even if they are unconfirmed, yet staff have gloves, aprons and masks at best.

I can't see how we can come down the other side of the mountain without protecting clinicians and carers.

Sorry if this is a dumb question, but the stats that say 78991 people have tested positive and out of those 9875 have sadly died, would that represent a 12.5% mortality rate for anyone sick enough to need testing?

I strongly believe there are no dumb questions - always better to ask. Not all those tested were sick (but I think the majority were). As long as those deaths were only for those who were tested, then the mortality rate for that particular group is at least 12.5%. (Could be more as some of them may yet still die from virus). Of course, it's possible that some died from something other than Covid-19, but I assume that's rare.

Roughly 10% are sick enough to go to hospital, of those 12.5% die, so, simply speaking around 1.25% mortality rate.

It is more complicated than that as you should lag the testing numbers, which would give a higher CFR. However, others would say far fewer than 10% are sick enough for hospital and estimate the mortality rate somewhere between 0.5 and 1%.

I think it will be hard to get a really accurate handle on it, at least for a while. but nothing like 1/8. Were that the case we would have a Chinese style lockdown with a curfew and the army on the streets.

It's 12.5% of those tested positive who die, as Dad says.
Mortality rate for the whole population is of course way lower, but essentially useless when we're comparing countries because of population differences.

@Bimbleboo thats right

The deaths announced for the last few days, by date of death (today is announced on the 11th, reported on the 10th, died any day up to the 10th, but mostly 8th and 9th

April 10 115
April 9 117,325
April 8 140,356,161
April 7 135,284,154,61
April 6 81,306,100,70,49
April 5 69,272,162,57,43,32
April 4 97,188,179,93,47,49,28

Those are the deaths announced on each succeeding day (i.e April 5,6,7,8,9,10,11 for the 4th). April 5 was Sunday and you can see that fewer people who died were announced that day, and hence more than usual a day after

Anyway the 8th had the highest number of first and second day deaths of any day, whereas the 9th was down. Probably now reporting is slow because of the bank holiday weekend, so it could be a false dawn still.....

Comparing mortality rate between countries is pointless atm, because of the different rates of testing,
iirc Uk 100,000 - 200,000 total tests but Germany already in the millions

Also, we don't know if tests of different accuracy are used, hence requiring more tests per person

• I think headline figures are # tests recorded, rather than # people tested

Once the crisis is over, we could look at deaths per million population, to see the overall impact

However, at this stage of the crisis, according to the FT, the rate of spread is not useful to look at by population,
because it misleads by making small countries look worse and larger ones better

Place mark

This RKI (German public health) table shows how Germany too has deaths occuring on previous days
which are then added to the current day's total
(my translation added)

The weekend/bank hol shouldn't make too much of a difference as the data has to be uploaded daily, so each trust will have someone responsible for uploading the data each day.

This happens for certain national datasets, even on Christmas Day

Doesn't the UK have a bit of a jump in deaths reported on Tuesdays ?

I have just heard about someone local to me who I used to spend time with but haven't seen for a few years - she has died and the post mortem revealed Covid19. But apparently she died suddenly and with no obvious illness beforehand (hence the post mortem) so she will go into the statistics but wouldn't it be more likely something like heart failure or brain aneurism that actually caused her to die? How can you be on the point of death with CV but not know it? She was in her mid 50s.

In almost all countries, "with" COVID like that is counted in the COVID death stats

We don't know what can cause sudden death in ostensibly healthy adults
but maybe COVID was the catalyst for a disastrous chain reaction, or maybe just bad luck

John Burn-Murdoch@jburnmurdoch (FT stats geek)

UK deaths down slightly today.

Remember key caveats:
• Reported numbers always drop the day after a weekend day (or public holiday) regardless of "true" number

• So best to track the total (or average) over multiple days

• In this case, fuller picture will emerge next Wed/Thu

Sunny Hundall@sunny*_hundal

Interesting article on Singapore:

• Cases are now rocketing up
• 'Frontline' migrant workers now main victims (it'll be Middle East next)

• "The system worked until mid-March"
• Now under partial lockdown (worse than UK)

https://www.bbc.com/news/world-asia-52232147

The system worked until mid-March,
says Prof Yik-Ying Teo, dean of the Saw Swee Hock School of Public Health in Singapore.

That's when, as the gravity of the situation became clear around the world, countries started urging their citizens to get home.

Thousands returned to Singapore from countries which had not been as proactive - among them more than 500 people who unwittingly brought the virus back with them.

It was by then mandatory for returnees to stay at home for two weeks.
But other people in their household were told they could carry on with their lives, as long as no-one showed any symptoms..

I put this on another thread today, as my DH and I were musing the terrible news we get every day, and needed to get a slightly better perspective. I know, I know, that the daily figures are awful, but hear me out. So usually 1,500 people die every day, and 1,000 are dying of Covid. So, if you assume some of those will overlap, then say we are getting about 2,000 deaths per day. Of the 66.6 million people in the UK, that is only 0.00003%. Also, 2,000 babies will have been born every day too, so life does go on.

Sorry if this is a dumb question, but the stats that say 78991 people have tested positive and out of those 9875 have sadly died, would that represent a 12.5% mortality rate for anyone sick enough to need testing?

It would be higher I’m afraid, as a fair proportion of the 78,991 will die but haven’t yet, or have died but aren’t yet in the death stats.... very sobering.

With regards PPE there was a dr on bbc news saying that the issue was that they changed the guidelines to needing less PPE rather than increasing the deliveries to required standard.

I thought today's befriending was an absolute shit show. The media are trying to score political points, seem to run out of questions, seem to be trying to pit the ministers against each other and now are questioning how to get out of the lockdown they so forcefully reported the government should be doing just a mere 3 weeks ago.

That's why I love this thread. Basic statistics and facts. Calm and sensible discussion with some really useful links that cut through the crap and point scoring.

Thanks for all contributing.

Roughly 10% are sick enough to go to hospital, of those 12.5% die, so, simply speaking around 1.25% mortality rate.

12.5% of those admitted to hospital (assuming they make up vast majority of tests) have currently died. This % is going up not all people who have tested positive who will die, have died yet.