UK coronavirus death toll could be 15% higher than previously shown: new data

[snoop1130]

UK coronavirus death toll could be 15% higher than previously shown: new data

By Andy Bruce

 

A medical glove lies discarded in a park as the spread of the coronavirus disease (COVID-19) continues in Manchester, Britain, April 13, 2020. REUTERS/Phil Noble

 

LONDON (Reuters) - Britain’s death toll from the coronavirus could run some 15% higher than official figures have indicated so far, according to broader data published on Tuesday that include deaths in the community such as in nursing homes.

 

The Office for National Statistics said 6,235 people in England and Wales had died by April 3 with mentions of COVID-19 on their death certificates.

 

“When looking at data for England, this is 15% higher than the NHS numbers as they include all mentions of COVID-19 on the death certificate, including suspected COVID-19, as well as deaths in the community,” ONS statistician Nick Stripe said.

 

Unlike the daily data published by the government that show only deaths in hospitals, Tuesday’s figures include deaths in the community, such as at nursing homes.

 

In London, nearly half (46.6%) of deaths registered in Week 14 involved COVID-19, the ONS said.

 

During the week to April 3, deaths mentioning COVID-19 accounted for 21.2% of all deaths, compared with 4.8% in the previous week.

 

The latest daily death toll for the United Kingdom published showed a total of 11,329 people had died in hospitals as of Sunday at 1600 GMT across after testing positive for coronavirus.

 

-- © Copyright Reuters 2020-04-14

Follow Thaivisa on LINE for breaking COVID-19 updates

[Rookiescot]

Figures for Scotland and Wales do include those who died at home or in care homes where corona was a factor.

I do not know what the situation is in Northern Ireland.

[Forethat]

This is a statistical conundrum currently being discussed in more countries than UK. The question is how to statistically record a fatality in an accurate way when the patient is infected by SARS-CoV-2 but has yet to develop COVID-19 or when the patient HAS developed COVID-19 but dies from other comorbidities. Just because COVID-19 is 'mentioned' doesn't mean it was the cause of death.

Edited April 14, 2020 by Forethat

[stevenl]

2 minutes ago, Forethat said:

This is a statistical conundrum currently being discussed in more countries than UK. The question is how to statistically record a fatality in an accurate way when the patient is infected by SARS-CoV-2 but has yet to develop COVID-19 or when the patient HAS developed COVID-19 but dies from other comorbidities. Just because COVID-19 is 'mentioned' doesn't mean it was the cause of death.

Yes, that I an issue.

 

But the issue at hand is deaths in eg nursery homes where testing for covid 19 was not done.

[Rookiescot]

2 minutes ago, stevenl said:

Yes, that I an issue.

 

But the issue at hand is deaths in eg nursery homes where testing for covid 19 was not done.

Yes this is a serious issue. Its because of lack of testing kits available.

However given the contagiousness of corona I imagine single deaths from it in places like care homes will be rare.

More likely you will see a cluster of deaths happening at the same time and from that it is reasonable to conclude the actual cause of death given the symptoms are quite well defined.

[Forethat]

1 minute ago, stevenl said:

Yes, that I an issue.

 

But the issue at hand is deaths in eg nursery homes where testing for covid 19 was not done.

Totally.

But suspicions of COVID-19 or a confirmed SARS-CoV-2 infection doesn't mean it was the cause of death,

 

I guess the real issue is why they weren't tested and brought to hospital.

[Rookiescot]

8 minutes ago, Forethat said:

Totally.

But suspicions of COVID-19 or a confirmed SARS-CoV-2 infection doesn't mean it was the cause of death,

 

I guess the real issue is why they weren't tested and brought to hospital.

Indeed.

Throwing a hand grenade into the care home does not mean they all died because of it.

Some of them might have had a heart attack before it went off.

[Forethat]

7 minutes ago, Rookiescot said:

Indeed.

Throwing a hand grenade into the care home does not mean they all died because of it.

Some of them might have had a heart attack before it went off.

Luckily, the hand grenade attacks are scarce. The statistical issue however, which was the one I addressed, is that cause of death is a medical term that doesn't take into account a media hype around a virus - when someone dies their cause of death is based on medical terms, nothing else. A suspicion is exactly that - and not statistically valid.

 

Having said that, if a nursing home reports a fatality caused by COVID-19 the fatality should be included in the stats. If it is a 'suspicion' it shouldn't. 

[Chomper Higgot]

37 minutes ago, Forethat said:

This is a statistical conundrum currently being discussed in more countries than UK. The question is how to statistically record a fatality in an accurate way when the patient is infected by SARS-CoV-2 but has yet to develop COVID-19 or when the patient HAS developed COVID-19 but dies from other comorbidities. Just because COVID-19 is 'mentioned' doesn't mean it was the cause of death.

It’s not a statistical conundrum, it’s precisely the kind of problem statistics can solve.

 

A statistical analysis of the data will reveal the number of deaths over and above the statistical norm and assign a probability to COVID-19 being the precipitating cause of death.

 

The conundrum is for those who wish to misrepresent the data, the statistical analysis uncovers the underlying causes.

 

An ability historically demonstrated in the solving the cause cholera and in Florence Nightingales success in identifying and controlling the part played by cross infection during the Crimean war.

[Chomper Higgot]

7 minutes ago, Forethat said:

Luckily, the hand grenade attacks are scarce. The statistical issue however, which was the one I addressed, is that cause of death is a medical term that doesn't take into account a media hype around a virus - when someone dies their cause of death is based on medical terms, nothing else. A suspicion is exactly that - and not statistically valid.

 

Having said that, if a nursing home reports a fatality caused by COVID-19 the fatality should be included in the stats. If it is a 'suspicion' it shouldn't. 

You are completely misunderstanding what statistics does.

[Forethat]

12 minutes ago, Chomper Higgot said:

It’s not a statistical conundrum, it’s precisely the kind of problem statistics can solve.

 

A statistical analysis of the data will reveal the number of deaths over and above the statistical norm and assign a probability to COVID-19 being the precipitating cause of death.

 

I think you make the classical mistake of mixing up stats and analysis of data. And as I pointed out, confirmed fatalities and suspected fatalities are two different things. But otherwise I agree. 

 

In this case, the fatality reports from the hospitals do not contain an ounce of analysis. If someone wants to analyse the fatality reports from hospitals AND nursing homes it'd be a different result. A suspected cause of death still shouldn't be included in a report covering confirmed deaths. For obvious reasons.

Edited April 14, 2020 by Forethat

[Chomper Higgot]

14 minutes ago, Forethat said:

I think you make the classical mistake of mixing up stats and analysis of data. And as I pointed out, confirmed fatalities and suspected fatalities are two different things. But otherwise I agree. 

 

In this case, the fatality reports from the hospitals do not contain an ounce of analysis. If someone wants to analyse the fatality reports from hospitals AND nursing homes it'd be a different result. A suspected cause of death still shouldn't be included in a report covering confirmed deaths. For obvious reasons.

Be assured, hospitals report exactly the data statisticians need to make detailed and accurate analysis of causes of mortality.

 

Statistics plays a very significant part in disease control and public health planning.

[Forethat]

1 minute ago, Chomper Higgot said:

Be assured, hospitals report exactly the data statisticians need to make detailed and accurate analysis of causes of mortality.

 

Statistics plays a very significant part in disease control and public health planning.

Be assured, death certificates (including cause of death) are issued by doctors and no one else...at least in the UK.

 

I should also point out that we have issued clear instructions to doctors that COVID-19 is an accepted direct or underlying cause when certifying death. With that in mind, I'm intrigued as to the notion that COVID-19 is 'mentioned' on the death certificate as oppose to stated as cause of death. 

 

[Crazy Alex]

2 hours ago, Forethat said:

This is a statistical conundrum currently being discussed in more countries than UK. The question is how to statistically record a fatality in an accurate way when the patient is infected by SARS-CoV-2 but has yet to develop COVID-19 or when the patient HAS developed COVID-19 but dies from other comorbidities. Just because COVID-19 is 'mentioned' doesn't mean it was the cause of death.

I like how another poster put it in another thread, something to the effect of "China has miraculously decreased deaths in other countries due to diabetes, heart disease and other ailments to near zero!"

 

Sadly, what it will come down to is how to sensationalize the virus as much as possible, with actual facts and science being pretty much a lost priority at this point

[Forethat]

5 minutes ago, Crazy Alex said:

I like how another poster put it in another thread, something to the effect of "China has miraculously decreased deaths in other countries due to diabetes, heart disease and other ailments to near zero!"

 

Sadly, what it will come down to is how to sensationalize the virus as much as possible, with actual facts and science being pretty much a lost priority at this point

Valid point. 

 

I believe there could be a statistical quality issue in both China and the UK - but for two completely different reasons. If I may speculate; It is in the interest of China to reduce the official fatality numbers. In the case of the UK I think it's the lack of testing that creates a situation where doctors have NO CHOICE but to state that COVID-19 is suspected to have contributed to death. Had the person been tested and found SARS-CoV-2 positive it would have been a completely different matter. This is what creates the statistical conundrum (although not understood by all posters on here).

 

Stating COVID-19 as a suspected contributing factor is probably what the data analyst in the OP refers to when stating the COVID-19 is 'mentioned'.

 

 

 

[elliss]

2 hours ago, Rookiescot said:

Indeed.

Throwing a hand grenade into the care home does not mean they all died because of it.

Some of them might have had a heart attack before it went off.

 

       Thats a lot of pensions , UK goverment no longer pay out ..

 

[Chomper Higgot]

23 minutes ago, Forethat said:

Valid point. 

 

I believe there could be a statistical quality issue in both China and the UK - but for two completely different reasons. If I may speculate; It is in the interest of China to reduce the official fatality numbers. In the case of the UK I think it's the lack of testing that creates a situation where doctors have NO CHOICE but to state that COVID-19 is suspected to have contributed to death. Had the person been tested and found SARS-CoV-2 positive it would have been a completely different matter. This is what creates the statistical conundrum (although not understood by all posters on here).

 

Stating COVID-19 as a suspected contributing factor is probably what the data analyst in the OP refers to when stating the COVID-19 is 'mentioned'.

 

 

 

I suspect when you refer to ‘Statistical Quality’ you mean quality of the data set.

 

But it’s not such an insurmountable problem, statistical methods are commonly used remove the ‘noise’ from incorrectly recorded data.

 

 

 

[Chomper Higgot]

1 hour ago, Crazy Alex said:

I like how another poster put it in another thread, something to the effect of "China has miraculously decreased deaths in other countries due to diabetes, heart disease and other ailments to near zero!"

 

Sadly, what it will come down to is how to sensationalize the virus as much as possible, with actual facts and science being pretty much a lost priority at this point

‘China’ has done no such thing.

 

Now what where you saying about facts and science?

[Forethat]

 

6 minutes ago, Chomper Higgot said:

I suspect when you refer to ‘Statistical Quality’ you mean quality of the data set.

No. I'm referring to the quality of the statistical product.

 

https://www.ons.gov.uk/methodology/methodologytopicsandstatisticalconcepts/qualityinofficialstatistics/qualitydefined

 

 

 

Edited April 14, 2020 by Forethat

[Crazy Alex]

4 minutes ago, Chomper Higgot said:

‘China’ has done no such thing.

 

Now what where you saying about facts and science?

Well yes, obviously the Chinese virus came from China. Or is there some other type of denial you're trying to engage in?

 

And besides, it appears you're missing the point anyway. The comment was made in context of the issuance of cause of death in the environment of the virus. Please try and stay on point.

[bkk_mike]

The issue is that we know people in care homes in the UK are literally dying from Covid-19 without being tested for the virus.

The problem is that it isn't "proven" Covid-19, just "suspected" Covid-19, because there are insufficient tests available.

 

 

The grandfather of one of my daughter's flatmates died last Friday night from it (at home - not in a care home). But as he died at home he won't be included in the hospital deaths figures (he got oxygen at home through the NHS rather than go into hospital). The family knows it's what killed him because several other family members have had it as well, and I believe one of them is in hospital (the uncle of my daughter's flatmate).

[BobBKK]

12 hours ago, Chomper Higgot said:

Be assured, hospitals report exactly the data statisticians need to make detailed and accurate analysis of causes of mortality.

 

Statistics plays a very significant part in disease control and public health planning.

Not true. Absolutely not true. Doctors are forced to put something on the death cert and cannot put 'old age' which is what a lot of people die of and there could be many individual things that deliver the last blow but it's irrelevant (read 'How We Die' by Dr. Nuland) Back to C19 I'm sure there are millions who have had it, recovered and are not in the stats ergo the death percentage is much lower.

[stevenl]

29 minutes ago, BobBKK said:

Not true. Absolutely not true. Doctors are forced to put something on the death cert and cannot put 'old age' which is what a lot of people die of and there could be many individual things that deliver the last blow but it's irrelevant (read 'How We Die' by Dr. Nuland) Back to C19 I'm sure there are millions who have had it, recovered and are not in the stats ergo the death percentage is much lower.

Medically people don't die of old age.

[Rookiescot]

I do not understand the problem with including the deaths of people at home or in care in the total number of deaths. Other countries are able to do it. Including Scotland and Wales.

Is there some special reason it can not be done in England?

[BobBKK]

18 minutes ago, stevenl said:

Medically people don't die of old age.

They can't put that so you can be 110 but you die of a heart attack. It's true that you had one, of course, but you died because you were 110 is all I'm pointing out.

[Chomper Higgot]

1 minute ago, BobBKK said:

They can't put that so you can be 110 but you die of a heart attack. It's true that you had one, of course, but you died because you were 110 is all I'm pointing out.

No, the heart attack killed you.

[BobBKK]

4 minutes ago, Chomper Higgot said:

No, the heart attack killed you.

Because you were 110.

[Chomper Higgot]

14 minutes ago, Rookiescot said:

I do not understand the problem with including the deaths of people at home or in care in the total number of deaths. Other countries are able to do it. Including Scotland and Wales.

Is there some special reason it can not be done in England?

I don’t understand the objection some clearly have to an assessment of data by the UK’s Office of National Statistics that indicates the UK has been under reporting deaths from COVID-19.

 

 

 

 

[stevenl]

33 minutes ago, BobBKK said:

Because you were 110.

No, you would have lived longer if the heart hadn't stopped beating.

[BobBKK]

6 minutes ago, stevenl said:

No, you would have lived longer if the heart hadn't stopped beating.

If you were younger you might have survived it. Read Dr. Nulands book and educate yourself.