Covid : Infection rates v death rates and a possible second wave

[bramling]

I can't imagine another national lockdown going ahead. What we might end up with, which is equally stupid, is local authorities all locking down themselves as they all adopt policies different from the national ones.

Yes I've a nasty feeling something like that could happen, or at least start to happen - as it would of course fall completely apart as people fail to keep up with what restrictions apply in different places. It's already been awkward enough with the four parts of the UK doing different things.

 

[Crossover]

To work properly local lockdowns do need to include travel restrictions. Without them they could be negative - people will go to the next town to the pub and spread it round more.

My understanding is that is what happened with the Leicester continued lockdown (since eased)

 

[Bletchleyite]

My understanding is that is what happened with the Leicester continued lockdown (since eased)

It only sort-of happened in that a load of the Nottingham pubs said they'd check proof of address and wouldn't serve people from Leicester. Other than that even if it was technically the case there was no real way to enforce it.

To do it properly you'd need to actually enforce it, i.e. as a minimum random spot-checks with fines, or even blocking roads and withdrawing stops in train services.

 

[Crossover]

It only sort-of happened in that a load of the Nottingham pubs said they'd check proof of address and wouldn't serve people from Leicester. Other than that even if it was technically the case there was no real way to enforce it.

To do it properly you'd need to actually enforce it, i.e. as a minimum random spot-checks with fines, or even blocking roads and withdrawing stops in train services.

I’d heard of it more in the context of some shops that couldn’t open in Leicester but could elsewhere, not so much the pubs. Either way, it certainly sounds like it could have been more effectually implemented

 

[adc82140]

The ONS have reported a higher than average overall death rate for the first time since June.

However this is due to the heatwave. Best lock down everyone each time it gets above 23 degrees then

https://www.bbc.co.uk/news/live/world-53802218

Deaths in the UK rose above average for the first time since mid-June, according to new figures. But the Office of National Statistics (ONS) said the increase was "likely due to the heatwave", as coronavirus deaths continued to fall.

There were 139 deaths recorded in the week up to 14 August that mentioned Covid-19 on the death certificate, down from 152 the previous week.

But overall, there were 10,580 deaths registered across the UK – 265 more than the five-year average and 370 more than the previous week.

“The increased number of deaths, and the rise above the five-year average, were likely due to the heatwave; the coronavirus did not drive the increase," said the ONS.

 

[scotrail158713]

The ONS have reported a higher than average overall death rate for the first time since June.

However this is due to the heatwave. Best lock down everyone each time it gets above 23 degrees then

https://www.bbc.co.uk/news/live/world-53802218

That article actually highlights an important point that some people really need to know. Using the five year average given there, an average of around 1470 people die every day in the UK. That puts recent Covid daily numbers into perspective.

 

[talldave]

Use the figures for the last couple of years and it's just under 1700. So 4 from Covid is terrifying - not.

 

[30907]

That article actually highlights an important point that some people really need to know. Using the five year average given there, an average of around 1470 people die every day in the UK. That puts recent Covid daily numbers into perspective.

Quite so. The excess death figure for the year as a whole - the least unreliable measure - is likely to be around 50k so ballpark 10%. ATM it's thankfully less than 1% whereas at peak it was at least 50%.

The ONS have reported a higher than average overall death rate for the first time since June. However this is due to the heatwave. Best lock down everyone each time it gets above 23 degrees.

Seriously, when the major heatwave of 2003 contributed to 15k deaths in France, significant mitigations (not lockdown, of course!) were put in place in many countries. The 2019 French heatwave is estimated to have contributed to only 1.5k deaths by comparison. (Source via Googling: morts canicule 2003, 2019)

 

[Freightmaster]

A map showing infection rates in England as of last week:

ArcGIS Web Application

www.arcgis.com

So no sign of any second waves or cases rising uncontrollably in the overwhelming majority of the country,
even areas "overrun" with tourists such as Cornwall and the Lake District, just a few isolated outbreaks in
densely populated, ethnically diverse 'high risk areas' in the Manchester/West Yorkshire areas.



Oh, and in case anyone in wondering what an MSOA is, Wikipedia explains...

A Middle Layer Super Output Area (MSOA) is a geospatial statistical unit used in England and Wales to facilitate the reporting of small area statistics. They consist of contiguous Lower Layer Super Output Areas. They are part of the ONS coding system created by the Office for National Statistics. The mean population of a MSOA is 7,200, with a minimum population of 5,000.

MARK

 

[Meerkat]

But the BBC reports that French R has hit 1.4, and hospital admissions are rising to 800 a week
Are we doing better or just a few weeks behind on reopening?

 

[apk55]

A very high percentage of people who die of covid have serious heath conditions and I have heard suggestions that many (suggested 50%) would probably die shortly anyway.
A person with terminal cancer or failing heart would easily fall victim to covid (as well as other infections). People in nursing or care homes are often in this situation.
As a result we could see below average death rates shortly and this does seem to be the case.
However the effects of lock down could make other deaths more likely due to other factors such as deferred medical treatment, alcoholism and other mental problems etc.

Let us not forget that while rare it is not unknown for supposedly young heathy and fit people fall victim to various illness. A person with a weak immune system could survive on a diet of antibiotics.

 

[adc82140]

But the BBC reports that French R has hit 1.4, and hospital admissions are rising to 800 a week
Are we doing better or just a few weeks behind on reopening?

Ours are about 700 a week. We're just discharging more than we admit, therefore a net reduction. Possibly the same in France.

 

[Chester1]

But the BBC reports that French R has hit 1.4, and hospital admissions are rising to 800 a week
Are we doing better or just a few weeks behind on reopening?

Weekly cases in UK have dropped for the first time in 6 weeks, which with increased testing is very encouraging. French cases have been rising for a while now, I don't think its because they opened up slightly earlier than we did.

 

[Bantamzen]

A very high percentage of people who die of covid have serious heath conditions and I have heard suggestions that many (suggested 50%) would probably die shortly anyway.
A person with terminal cancer or failing heart would easily fall victim to covid (as well as other infections). People in nursing or care homes are often in this situation.
As a result we could see below average death rates shortly and this does seem to be the case.
However the effects of lock down could make other deaths more likely due to other factors such as deferred medical treatment, alcoholism and other mental problems etc.

Let us not forget that while rare it is not unknown for supposedly young heathy and fit people fall victim to various illness. A person with a weak immune system could survive on a diet of antibiotics.

This is also true of pretty much any infection. My step father whilst in a very poorly state during treatment for leukaemia died from what was believed to a a common cold. Its why understanding the true nature of each person that died, and correctly reporting it in the data is key. In years to come many governments around the world will have some very difficult questions to answer around this.

 

[Justin Smith]

I'd just like to clarify something.
I have repeatedly said there will be "no second wave" in any country with a death rate above 500 per million, particularly any European country. I am referring to deaths not infections. TBH, I'm only really bothered about how many people die from this virus, and so should everybody be, infections were only (possibly) useful in predicting future deaths, but the relationship between the two (infections and deaths) seems to have broken down, thankfully. The infection rate in may countries is going up, but not the death rate, certainly not be any where near the same percentage.
What does worry me though is the very broad definition of a Covid death, as far as I'm aware anyone who dies and has tested positive for Covid within the previous 28 days is counted as a Covid death (can anyone confirm this is correct ? ). So, if the infection rate goes up that almost certainly means the death rate will go up simply because there is a greater chance that peopel who would have died anyway will have Covid !

This is also true of pretty much any infection. My step father whilst in a very poorly state during treatment for leukaemia died from what was believed to a a common cold. Its why understanding the true nature of each person that died, and correctly reporting it in the data is key. In years to come many governments around the world will have some very difficult questions to answer around this.

I think you're right.

This graphic might be if interest.
People's adherence to the lockdown actually started easing off within a few weeks but it is not easily quantifiable (other than traffic density, does anyone know where to get those stats ?). I worked right through this Covid episode (so was "out and about") and I can tell you for certain that people started getting less strict about the lockdown inside 3 or 4 weeks, slowly at first, then more widely as time went on.

Graphs of the falling infection rate and death rate as the lockdown is officially eased off, no increase despite the "experts" telling us, at every stage, there could / would be :

View attachment 82957

 

[MikeWM]

I'm trying to make some sense of the government dashboard on hospitalisations (at https://coronavirus.data.gov.uk/) and I'm more confused than when I started.

Wales seems to have much higher daily figures than England, for example, but the *total* number of people in hospital in Wales with Covid is only slightly higher than the daily admission numbers...

Conversely admissions in Scotland are effectively zero, but they still have a lot of people in hospital which doesn't appear to be decreasing at all.

Carl Heneghan has picked up on the Wales issue too. (I'm quite chuffed to have spotted this a few days before he did

https://www.cebm.net/covid-19/the-flaw-in-the-reporting-of-welsh-data-on-covid-hospital-admissions/

In the about section, we are told that “the numbers of admissions are not comparable with other nations.” Data should, therefore, not be combined if the criteria differ. But in the daily figure, it is. Does the problem highlight a lack of epidemiological understanding about creating accurate and consistent data?

The upshot is that the daily hospital admissions numbers are probably 30 or 40 too high, and are therefore averaging probably 50-60 per day rather than the 90-100 reported.

 

[Kingspanner]

I'd just like to clarify something.
I have repeatedly said there will be "no second wave" in any country with a death rate above 500 per million, particularly any European country. I am referring to deaths not infections. TBH, I'm only really bothered about how many people die from this virus, and so should everybody be, infections were only (possibly) useful in predicting future deaths, but the relationship between the two (infections and deaths) seems to have broken down, thankfully. The infection rate in may countries is going up, but not the death rate, certainly not be any where near the same percentage.

Only really bothered about how many people die?

COVID-19 symptoms can sometimes persist for months. The virus can damage the lungs, heart and brain, which increases the risk of long-term health problems.

COVID-19 (coronavirus): Long-term effects

Learn about the long-term effects of COVID-19 on the lungs, brain and nervous system.

www.mayoclinic.org

.

 

[Class 33]

I did think it was a bit odd on the Gov Coronavirus dashboard that the Testing and Healthcare figures seems to have been exactly the same for the past few days or so. Of particular interest are the figures for Patients in Hospital(showing as 764), Patients in Ventilator Beds(showing as 60) and Patients Admitted(showing as 109). Surely these figures can't be the exact same every single day. But when I clicked in on the More in Healthcare, I was pleasantly surprised that these figures are wrong and the true numbers are much lower. Patients in Hosptial now is just 305, Patients in Ventilator Beds is now just 33, and Patients Admitted yesterday was just 34. Along with the daily deaths now, these are very low figures now.

 

[MikeWM]

and Patients Admitted yesterday was just 34

It will consistently be around that number going forward (rather than the 100+ we've regularly seen over the last few weeks) if they fix the issue with the reporting from Wales...

 

[Dent]

I did think it was a bit odd on the Gov Coronavirus dashboard that the Testing and Healthcare figures seems to have been exactly the same for the past few days or so. Of particular interest are the figures for Patients in Hospital(showing as 764), Patients in Ventilator Beds(showing as 60) and Patients Admitted(showing as 109). Surely these figures can't be the exact same every single day. But when I clicked in on the More in Healthcare, I was pleasantly surprised that these figures are wrong and the true numbers are much lower. Patients in Hosptial now is just 305, Patients in Ventilator Beds is now just 33, and Patients Admitted yesterday was just 34. Along with the daily deaths now, these are very low figures now.

As far as I can work out the headline futures on the summary page and at the top of the healthcare page are for the most recent day for which figures are available for all four nations. The 305 and 33/34 figures you saw are not final totals for the UK as the figures from some nations have yet to be counted/added.

This is not made at all clear, and the way the headline figures are presented is highly misleading.

 

[talldave]

As I've pointed out before, they don't list discharges as a complementary figure to admissions - because it doesn't contribute towards maintaining the fear factor. If 100+ are being admitted to hospital each day, then the total number in hospital should in theory be doing up by 100+ a day - but it's not, it's down at plus or minus single digits. So it's best to ignore admissions and look at the delta on "in hospital".

 

[DB]

In response to kingspanner, It's already been pointed out repeatedly that severe long term effects appear to be very rare - if they were common, there would have been loads of cases reported by now and there hasn't been.

It's typical scaremongering.

 

[big_rig]

In response to kingspanner, It's already been pointed out repeatedly that severe long term effects appear to be very rare - if they were common, there would have been loads of cases reported by now and there hasn't been.

It's typical scaremongering.

Yes, it's just the goalposts shifting. It is more difficult to say the virus is 'indiscriminate' now that we know 99% of deaths in England are over 45, a quarter were over 90 and a quarter had dementia etc, so the next thing is 'long covid.' Given that Imperial College says 6% of England/3.4 million people have antibodies as of June surely there should be millions of people with brain damage/heart damage/'brain fog' and other symptoms to interview? It is striking that it seems to be mostly linked to people very active on Twitter.

 

[adc82140]

In response to kingspanner, It's already been pointed out repeatedly that severe long term effects appear to be very rare - if they were common, there would have been loads of cases reported by now and there hasn't been.

It's typical scaremongering.

In addition, if you are sick enough from any respiratory disease to require an ITU stay, you will take quite some time to recover.

 

[BJames]

Two types of steroid have now been found to save lives in severe cases:

Two types of steroid found to save lives of some Covid-19 patients

Analysis of seven trials finds dexamethasone and hydrocortisone should be given in severe cases

www.theguardian.com

Studies around the world have confirmed that steroids can save lives in the Covid-19 pandemic, leading to new recommendations from the World Heath Organization that doctors should give them to severely ill patients.

In June, the Recovery trial run in most NHS hospitals and led by Oxford University found that the lives of one in eight people sick enough from Covid-19 to need a ventilator could be saved by a steroid called dexamethasone.

Now, combined results from that trial and six others have confirmed those findings and established that at least one other equally cheap and widely available steroid, hydrocortisone, also saves lives.

The drugs reduce the risk of death in these seriously ill patients by 20%, according to a meta-analysis of the results of the seven trials covering a total of 1,703 patients, published in the Journal of the American Medical Association. Three of the trials have also been published separately in the journal.

“Steroids are a cheap and readily available medication, and our analysis has confirmed that they are effective in reducing deaths amongst the people most severely affected by Covid-19,” said Jonathan Sterne, a professor of medicine and epidemiology at Bristol University and the lead author of the meta-analysis.

“The results were consistent across the trials and show benefit regardless of age or sex.”

The pooled results add weight to the Recovery trial findings because they are from a diverse group of patients in several countries, including Brazil and France. “We’ve got a consistent message from all these trials, and the effect of hydrocortisone appears consistent with the effect of dexamethasone,” he said.

Some good news? Obviously dexamethasone was already being used, but a second drug is excellent. At least, some progress has been made. I expect even if we do find ourselves with increased cases again the death toll will be much lower.

 

[adc82140]

This is where efforts should be concentrated. It's all well and good to have a vaccine that may work, but surely ideally we should just let the virus circulate amongst the healthy and have an effective treatment option for anyone who becomes seriously unwell.

 

[Crossover]

What does worry me though is the very broad definition of a Covid death, as far as I'm aware anyone who dies and has tested positive for Covid within the previous 28 days is counted as a Covid death (can anyone confirm this is correct ?

That's my understanding. The death could be for any reason (don't know if it extends to walking out of hospital having overcome hospitalisation and then being hit by the bus...wouldn't surprise me though to be honest!)

 

[MikeWM]

Carl Heneghan has picked up on the Wales issue too. (I'm quite chuffed to have spotted this a few days before he did

And now he's got another article on the issues with the number of people still in hospital in Scotland, the other issue I said made no sense I should ask him for a job at this rate!

https://www.cebm.net/covid-19/is-scotland-overcounting-the-number-of-patients-in-hospital-beds/

According to Scottish government’s Coronavirus (COVID-19) definitions, the number of patients in hospital with suspected COVID-19 “may include people who are in hospital for other reasons but have previously tested positive for COVID-19.” While a count of new admissions to hospital only includes those who “tested positive for COVID-19 in the 14 days prior to admission to hospital, on the day of their admission, or during their stay in hospital”, no equivalent cutoff appears to be made for this count of individuals currently in hospital.

The impact of this appears to be potentially substantial

The conclusion is that the 'number of people in hospital in Scotland with Covid' figure is probably far too high.

 

[adc82140]

Is it too much to ask to have:

A death rate where Covid is confirmed as the primary cause

A hospital admissions rate where Covid is the sole reason for that admission

A positive test rate where the test is the first positive test on a particular person only, and is not a follow up to see if they've recovered.

I'm no statistician, but surely this isn't beyond the capabilities of our country?


.

 

[greyman42]

Is it too much to ask to have:

A death rate where Covid is confirmed as the primary cause

A hospital admissions rate where Covid is the sole reason for that admission

A positive test rate where the test is the first positive test on a particular person only, and is not a follow up to see if they've recovered.

I'm no statistician, but surely this isn't beyond the capabilities of our country?


.

It sounds like the most obvious way of compiling useful data to me. The only reason i can think of for not using this is that someone in government does not want to.