COVID-19 death toll in England 41% higher than early data suggested: ONS

 

A medical worker wearing personal protective equipment (PPE) at the back of an ambulance outside Lewisham hospital as the spread of the coronavirus disease (COVID-19) continues, London, Britain, April 20, 2020. REUTERS/Hannah McKay

 

LONDON (Reuters) - The true extent of the death toll in England and Wales from COVID-19 was 41% higher than the daily figures from the government indicated by April 10, according to data on Tuesday that includes deaths in the community.

 

The Office for National Statistics said it recorded 13,121 deaths by April 10, compared with 9,288 in the government’s daily toll for those who died in hospital.

 

COVID-19, the respiratory disease caused by the novel coronavirus, was mentioned in a third of all death certificates issued in England and Wales in the week to April 19.

 

In London, more than half of the death certificates issued that week mentioned COVID-19.

 

The latest hospital deaths data published on Monday show 16,509 people had died across the United Kingdom.

 

-- © Copyright Reuters 2020-04-21

 

- Whatever you're going through, the Samaritans are here for you

- Follow Thaivisa on LINE for breaking COVID-19 updates

One can only dream of such data being released to the public in Thailand.

Not a rounding error then?!

Getting a handle on deaths outside of hospitals is a challenge everywhere. Especially in hard hit areas where normal procedures and investigations aren't always being followed by an overwhelmed health and emergency response system. I know that is the case in New York City, There has been a 10 fol increase in deaths and they are struggling to figure out how to quantify that since most will never have been tested and info on cause (other than not foul play suspected) is sketchy.

 

They will problem en up just assuming the excess numbers are COVID related for lack of a better method.

 

Another place deaths get missed is hospital deaths early in an epidemic curve, before doctors were sensitize to the possibility of COVID.

 

These stats are going to be refined many times in fitire as these sorts of issues get looked at.

 

And we aren't going to have a reliable denominator until there is reliable, statistically representative antibody testing one. Which presupposes reliable, quality-assurred tests.

17 minutes ago, Sheryl said:

Getting a handle on deaths outside of hospitals is a challenge everywhere. Especially in hard hit areas where normal procedures and investigations aren't always being followed by an overwhelmed health and emergency response system.

This is pretty much the case in the UK. Here, the procedure for registering the death accurately is damaged by a series of flaws.

 

First of all, no testing outside NHS hospitals takes place. Guidance is issued that allows the death certificate issuing doctor to identify the cause of death OR the condition leading to the direct cause of death as COVID-19. No one can validate the accuracy from a clinical perspective since clinical judgement is in effect. What is happening now is that statisticians are trying to identify a discrepancy in death rates against the established statistical norm.

 

In addition, the reporting of deaths from non-NHS hospitals (care homes, hospices and private homes) where deaths can be registered electronically directly to Public Health England, lags behind by an alleged 6 days. The latest ONS report which included analysis of death certificates lagged by 16 days. ONS have announced today that they will make the required changes to their processing in order to deliver the analysis in 11 days.

 

In my opinion, to even suggest that someone should be able to provide an accurate report on the daily fatalities under these conditions only proves that the challenge is not properly comprehended.

4 hours ago, DrTuner said:

One can only dream of such data being released to the public in Thailand.

"Face"

Coronavirus: How big is the problem in care homes?

 

Ring of steel needed to support care homes as deaths double in a week

 

Is it probable that Thailand doesn’t include deceased ‘with’ corona virus?

 

9 hours ago, Forethat said:

This is pretty much the case in the UK. Here, the procedure for registering the death accurately is damaged by a series of flaws.

 

First of all, no testing outside NHS hospitals takes place. Guidance is issued that allows the death certificate issuing doctor to identify the cause of death OR the condition leading to the direct cause of death as COVID-19. No one can validate the accuracy from a clinical perspective since clinical judgement is in effect. What is happening now is that statisticians are trying to identify a discrepancy in death rates against the established statistical norm.

 

In addition, the reporting of deaths from non-NHS hospitals (care homes, hospices and private homes) where deaths can be registered electronically directly to Public Health England, lags behind by an alleged 6 days. The latest ONS report which included analysis of death certificates lagged by 16 days. ONS have announced today that they will make the required changes to their processing in order to deliver the analysis in 11 days.

 

In my opinion, to even suggest that someone should be able to provide an accurate report on the daily fatalities under these conditions only proves that the challenge is not properly comprehended.

Fortunately a statistical analysis of deaths will determine the number of COVID-19 related deaths to a stated degree of accuracy, the larger the sample the greater the accuracy.

 

Meanwhile somebody in Government needs to explain* the decisions process behind excluding 41% of COVID-19 related deaths.

 

——

 

*Providing the ‘reporting criteria’ the ‘decision process’ chose to adopt is not an explanation of the ‘decision process’.

10 hours ago, Sheryl said:

Getting a handle on deaths outside of hospitals is a challenge everywhere. Especially in hard hit areas where normal procedures and investigations aren't always being followed by an overwhelmed health and emergency response system. I know that is the case in New York City, There has been a 10 fol increase in deaths and they are struggling to figure out how to quantify that since most will never have been tested and info on cause (other than not foul play suspected) is sketchy.

 

They will problem en up just assuming the excess numbers are COVID related for lack of a better method.

 

Another place deaths get missed is hospital deaths early in an epidemic curve, before doctors were sensitize to the possibility of COVID.

 

These stats are going to be refined many times in fitire as these sorts of issues get looked at.

 

And we aren't going to have a reliable denominator until there is reliable, statistically representative antibody testing one. Which presupposes reliable, quality-assurred tests.

Excess deaths is likely to be the only method to give a reasonable estimate of all Covid-related deaths, since there are certainly non-Covid deaths that occur because of the unavailability of medical care due to the overwhelming of facilities.  Someone did such an estimate for a period in April for New York City and found an additional 30% of such excess deaths.  Since wherever someone dies a death certificate is filed, this is only method that is not likely to undercount.

10 hours ago, Forethat said:

<snip>

First of all, no testing outside NHS hospitals takes place

 

Incorrect. 

 

Whilst not very much, certainly not enough, testing takes place, some does. 

 

From the first link in my previous post:-

Until recently only the first few people with symptoms in each care home were tested to confirm the virus

 

The phrase 'until recently' surely means that it is safe to assume that more testing in care homes is now taking place: as it should. 

 

 

 

 

12 hours ago, Sheryl said:

And we aren't going to have a reliable denominator until there is reliable, statistically representative antibody testing one. Which presupposes reliable, quality-assurred tests.

The French estimate that by May 11th 5.7% will have been infected. By that time 30000 to 40000 will have died from the disease (Currently about 20k fatalities). This gives us an IFR (Infection Fatality Rate) of roughly 1%, hence 300 000 to 400 000 fatalites by the time 70% of the French population have developed antibodies. 

Let's hope that new treatments and/or vaccines come up before we rich the end of this journey So far the only thing we can do is slowing down the progress of the disease while destroying our lifestyle. 

China did exactly the same thing. They recently revised the numbers upwards by about 50% to account for deaths outside hospitals.

You also have to account for deaths that took place early in the pandemic before coronavirus became a household word. Every medical statistic is the same....when you look for a disease you find it. 

2 hours ago, cmarshall said:

Excess deaths is likely to be the only method to give a reasonable estimate of all Covid-related deaths, since there are certainly non-Covid deaths that occur because of the unavailability of medical care due to the overwhelming of facilities.  Someone did such an estimate for a period in April for New York City and found an additional 30% of such excess deaths.  Since wherever someone dies a death certificate is filed, this is only method that is not likely to undercount.

This was going to be my take on this too. Different countries are counting in different ways with some over, and some under, attributing.

Comparing to average deaths over a similar period in each city will show the increase that might be COVID related.

4 hours ago, 7by7 said:

 

Incorrect. 

 

Whilst not very much, certainly not enough, testing takes place, some does. 

 

From the first link in my previous post:-

Until recently only the first few people with symptoms in each care home were tested to confirm the virus

 

The phrase 'until recently' surely means that it is safe to assume that more testing in care homes is now taking place: as it should. 

 

 

 

 

Correct. What I meant is that NHS is the testing regime. It's their labs. 

 

Though I seriously question the decision to "as laboratory capacity increases, all care home residents who develop symptoms will be tested." That's not good enough. They need to test ALL residents and ALL care home workers. IN particular, care home workers need to be tested for Immunoglobulin G.

 

6 hours ago, cmarshall said:

Excess deaths is likely to be the only method to give a reasonable estimate of all Covid-related deaths

You might need to wait until 2024 to get that data for Thailand. MoPH has last given data in pneumonia deaths to WHO in 2016. Makes you wonder what the status of their own records is, as this is likely an automated process.

 

54 minutes ago, Forethat said:

IN particular, care home workers need to be tested for Immunoglobulin G

Why? I thought this is a common antibody, not specific to any disease.

8 minutes ago, rickudon said:

Why? I thought this is a common antibody, not specific to any disease.

Given the context of this discussion, do you think the Immunoglobulin G in question would be SARS-CoV-2 Immunoglobulin G...?

In case you had doubts, let me assure you that it is.

Financial Times published an article earlier this morning. It suggests that the COVID-19 death too is far higher than what is reported. Their figure is even higher than the ONS report published the other day. Clearly, there is need for drastic change in policies and procedures.

 

https://www.ft.com/content/67e6a4ee-3d05-43bc-ba03-e239799fa6ab

 

It appears what's happening is:

1. People in care homes aren't being tested (at some point they introduced a scheme where the first five suspected cases are being tested to identify an outbreak)

2. Death certificates sometimes mentions COVID-19 as the direct cause of death, sometimes as a condition leading to the direct cause of death. In 1% of the death certificates analysed by ONS, "suspected COVID-19" was mentioned as a condition contributing to death.

3. There is a significant lag between issuing the death certificate until the death certificate has been posted to be registered

4. There's an additional lag between the time the death certificate has been registered and the ONS report has been published after analysis

 

I mentioned in another thread that an even bigger concern is how many cases are there where there is NO mention of COVID-19 on the death certificate. If FT are to be believed I was spot on.

 

Either way, I think it's safe to say that no one knows the number of COVID-19 deaths in the UK.

4 hours ago, Pedrogaz said:

China did exactly the same thing. They recently revised the numbers upwards by about 50% to account for deaths outside hospitals.

You also have to account for deaths that took place early in the pandemic before coronavirus became a household word. Every medical statistic is the same....when you look for a disease you find it. 

 

When China did the same thing by revising the numbers , the whole world said China was covering up but when it happened to UK, they were silent.

 

We must understand that in this chaotic time, there will be mistakes as it is more important to save lives first and not have accurate count. Accurate count can come later when they are less busy saving lives.

 

16 minutes ago, rhyddid said:

BS BJ as a good clown is hiding real data, just compare di numbers at present time 
1 - total tests       535,352
2 - total infected  129,044
3 - total deaths     17,337 
show a 13.5% deaths on infected which is very high already, but the real scary data is on 
average of tested vs infected a good 25% of infected vs tested this means in UK 25% of population is infected!

now compare the bad and unlucky Italy 
183,000 infected vs 1,450,000 tested a much lower percentage , just  12% if infected 

The joker, is achieving his goal "herd immunization" and ready to say goodbye to your dear one !

Thanking his voters and there will be always hope he will be sent to court and tried for manslaughter !
 

Ehh, no.

 

The number of tested people mostly represent the people that were hospitalised and therefore tested. There are obviously a small number of NHS staff tested, but expect that a vast majority of the 535,352 tests to have been performed on people who showed serious symptoms. No one knows the number of infected citizens. I wouldn't even speculate in this matter.

 

Germany performs wide-scale testing and in theory, should display a reasonably good picture of how widespread the epidemic is. They have an infection rate of 8%.

Clearly they’re doing something wrong in the UK! They should do it the Thai way! Seems to work sooooo much better! 

I think it's important to point out that there is no such thing as a manual. Personally, I think characters like Piers Morgan who's using the pandemic as a way to attack people and blame them for being "incompetent", acting as if they are trying to take advantage of the pandemic, should be shut down. Of course, there will be mistakes. Lots of them. The entire world is in uncharted territory here. He's not contributing ONE single bit to any type of solution. In this case, people are dying all over the place and he's upset no one knows the exact number?

 

Every country is struggling. The UK is doing plenty wrong. So are others. But, to behave like Piers Morgan and act as if there was a known solution and a "best practice" only proves he's nothing but a sensationalist <deleted>. If he knows the solution to this he should step up and be heard. Otherwise, I suggest he take a slightly more humble approach and simply shut his trap. But I guess it's easier to complain and act like a tosser while others attempt to solve the problem...

 

If Piers Morgan worked in a creative environment and behaved this way he'd be sacked faster than anyone on this forum can say "Tom Yum Gong". Actually, make that "Tom"...

4 hours ago, Forethat said:

Correct. What I meant is that NHS is the testing regime. It's their labs. 

Nice save ????

 

4 hours ago, Forethat said:

Though I seriously question the decision to "as laboratory capacity increases, all care home residents who develop symptoms will be tested." That's not good enough. They need to test ALL residents and ALL care home workers. IN particular, care home workers need to be tested for Immunoglobulin G.

 

 

It seems we are in agreement on this point. 

1 minute ago, 7by7 said:

Nice save ????

I decided to omit the fact there is currently an unknown number of care home residents who COUNTS as tested in care homes while they are in fact tested at an NHS hospital after being admitted to hospital with severe symptoms. 

 

There's another statistical conundrum for you right there: a care home resident, being brought to hospital and tested positive for SARS-CoV-2, eventually dies. Is this patient counted as a fatality reported in a Daily Briefing? Most likely. Is the patient counted again as a care home resident who died in from COVID-19? Possibly. And an even more important question: Does anyone know the exact details of the fatalities? Evidently not.

23 hours ago, DrTuner said:

One can only dream of such data being released to the public in Thailand.

https://www.moph.go.th

58 minutes ago, Yinn said:

https://www.moph.go.th

http://bps.moph.go.th/new_bps/index.php actually.

 

But they blocked access to it: https://deathbirthrepo.dcs.moph.go.th/login

 

So, not available.

 

Also, the latest reports to WHO by MoPH are from 2016: 

 

12 hours ago, Forethat said:

I think it's important to point out that there is no such thing as a manual. Personally, I think characters like Piers Morgan who's using the pandemic as a way to attack people and blame them for being "incompetent", acting as if they are trying to take advantage of the pandemic, should be shut down. Of course, there will be mistakes. Lots of them. The entire world is in uncharted territory here. He's not contributing ONE single bit to any type of solution. In this case, people are dying all over the place and he's upset no one knows the exact number?

 

Every country is struggling. The UK is doing plenty wrong. So are others. But, to behave like Piers Morgan and act as if there was a known solution and a "best practice" only proves he's nothing but a sensationalist <deleted>. If he knows the solution to this he should step up and be heard. Otherwise, I suggest he take a slightly more humble approach and simply shut his trap. But I guess it's easier to complain and act like a tosser while others attempt to solve the problem...

 

If Piers Morgan worked in a creative environment and behaved this way he'd be sacked faster than anyone on this forum can say "Tom Yum Gong". Actually, make that "Tom"...

Don't we know that methods used in South Korea and Taiwan are the best practices?  And didn't we know that before the pandemic?

44 minutes ago, cmarshall said:

Don't we know that methods used in South Korea and Taiwan are the best practices?  And didn't we know that before the pandemic?

Correct, and this is one of the many reason why the Prime Minister and his Government need to answer for the appallingly poor decisions they made:

 

https://www.independent.co.uk/news/health/coronavirus-update-testing-news-herd-immunity-who-uk-cases-map-a9402051.html

 

 

14 hours ago, Yinn said:

https://www.moph.go.th

Nope, the COVID figures shown here will become credible only when the the number and the nature of the tests carried out in each province will be given. Remember: "test, test and test". 

 

Having less than half a million of tests a week in Thailand is like filtering water with cobble stones. Successful containment is based on "Test and Trace", tracing seems to be effective here but becomes meaningless with inadequate and/or insufficient testing.