Daily numbers, graphs, analysis thread 12

[BigChocFrenzy]

Welcome to thread 12 of the daily updates

Resource links:

Slides & data UK govt pressers
UK dashboard sub-national data, local authorities
Beta Uk dashboard deaths, cases, hospitals, tests, partially sub-national
UK stats updated daily by PHE & DHSC
ONS UK statistics for CV related deaths, released weekly each Tuesday
PHE surveillance report infections & deaths released every Thursday with sep. infographic
NHS England stats including breakdown by Hospital Trust
FT Daily updates
HSJ Healthcare updates
Worldometer UK page
Plot FT graphs compare countries deaths, cases / million pop. / log / linear
Covidly.com filter graphs compare countries
Plot COVID Graphs Our World in Data

We welcome factual, data driven, and civil discussions from all contributors 📈📶👍

@itsgettingweird

Yeah I know the issue. It it's that 114 deaths are Ben v reported yet gov.uk have actually said they aren't reporting due to issues.

We've known about people being hit by a bus in June who had a positive test in March being reported as a CV death for ages.

Big choc has reported same for Germany - any positive test is reported as CV death regardless of actual cause.

Germany has the standard 28 day cutoff after a positive test,

except for any deaths within the currently 6,250 ongoing "active" cases,
where a few of those may die after 28 days, but would be counted.

@StatisticalSense

It's fine to say it's only a small number of deaths up until this point, but clearly as more time progresses a larger and larger proportion of deaths will be old, so while I have no reason to doubt that the 10% is a genuine percentage for the entirety I am willing to bet that the figure for deaths between July 1st and July 15th is much higher. What is a good enough methodology in the early days of a pandemic is not necessarily good enough a number of months into it.

We should look at ONS to obtain reliable figures for July

Latest report is only to 3 July, but more recent data should be published shortly

https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/bulletins/deathsregisteredweeklyinenglandandwalesprovisional/weekending3july2020

"• Of the deaths registered in Week 27,
532 mentioned “novel coronavirus (COVID-19)”,
the lowest number of deaths involving COVID-19 in the last 15 weeks, accounting for 5.8% of all deaths in England and Wales.

• In Week 27, the proportion of deaths occurring in care homes decreased to 20.1%
while deaths involving COVID-19 as a percentage of all deaths in care homes decreased to 9.2%."

given that hospital deaths are now very low - below 20 - I could well believe that some of the extra up to 100 or so other deaths reported daily could be care home deaths of people who did indeed have covid, but perhaps tested positive quite some time ago and now have died of unrelated, or perhaps indirect reasons. Sounds like a complicated exercise to fix the historic deaths number this way, hopefully easier to do so going forward.

"while I have no reason to doubt that the 10% is a genuine percentage for the entirety"

By my calculation, the over estimation was 2,199 above the true total of 38,329
so 5.7% overall

Hopefully PHE will redo their charts from June, which will reveal the more recent %, which would obviously be increasing.

Until then, we can only look a bit in the rear view mirror with ONS

PHE weekly surveillance report & infographic to 14 July
was issued a few hours ago, but still has the old death figures:

https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachmentdata/file/901803/WeeklyyCOVID19SurveillanceeReportweekk29FINAL.pdff_
_
https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachmentt_data/file/901148/COVID19WeeklyyReport144July.pdf

USA

John Hopkins shows a record 77,255 new cases for the USA yesterday (Thursday)

States across the south and west are being badly hit

Where on earth is the analysis of cases broken down by settings? Eg care home etc?

I thought it was on the government pages but now can't find it.

Looks like the data from US is about to be politicised as Trump orders hospitals to bypass the CDC and send data to a central database instead
www.nytimes.com/2020/07/14/us/politics/trump-cdc-coronavirus.html

I’ve seen this theory a few times. Could there be anything in it??

www.medrxiv.org/content/10.1101/2020.07.15.20154294v1

[quote MarcelineMissouri]I’ve seen this theory a few times. Could there be anything in it??

www.medrxiv.org/content/10.1101/2020.07.15.20154294v1[/quote]
.......
I'm not impressed
There has been a lot of wishful thinking and speculation, sometimes irresponsible,
Few epidemiologists, virologists, public health experts around the world share this rosy view, so I'll be guided by the expert consensus

No convincing evidence so far of a v low % herd immunity being sufficient to stop COVID,
or of particular regions having high inbuilt immunity

A lot of wishful thinking here is based on a cruise ship, where passengers were separated and quarantined early on,
which is an effective way to combat superspreaders responsible for 80-90% of COVID infections.

Factors proven to affect infection rates:

. Timing of lockdown
. Population density
. Testing, Track & trace systems with effective isolation
. Temperature & humiidity, season

Factors affecting death rates:
. % of elderly population
. Health services that treat patients early
. Seasonal changes in immune system
. Keeping infection rates down via !) and 3) above

well what you do with death figures depends on whether you want to look good in international comparisons or you want the truth about what is happening in your own country. Up until recently excluding deaths after 28 days would have excluded more deaths that should be counted than including deaths that shouldnt be counted. Now we've gone past 50% where it probably wasnt the main cause of death time to rework the figures.

Whether someone is in hospital or a care home they can be ill from the virus for a long time before death. Elderly care home residents were not being admitted to hospital at the peak of the epidemic, even if really ill. We dont know how long people take to recover - or if the virus does permanent long term damage that will contribute to deaths recorded as being from pneumonia or some other cause. Elderly care home residents rarely get hit by buses.

No-one wants to consider how long people take to recover or if the virus does permanent damage because that conflicts with restarting the economy. Death figures - in all countries - should be based on the best estimate of the cause of death, not some arbitrary cut off point adopted for convenience. Adopting 28 days for convenience and international comparisons would be understandable, dont pretend there is data to justify it.

I think there is some logic in that particular report about herd immunity, unlike most of the wishful thinking. The problem, however, is that herd immunity may not exist at all if antibodies wain after 3 months. And the conclusion is also wishful thinking because regional variations are more likely to be down to things like housing density, employment practises, vitamin D deficiency levels, than any ideas about innate resistance.

I agree with all comments already posted above with regards to over/under counting, but I do think there will some, probably very few, which have not been counted correctly. I would think however, as noted above, it will all come out in the wash over the year.

I can't remember if I have posted this before but, for example, FIL died in June of a heart attack. He was 86, and had had a embolism 40 years ago, a stroke 30 years ago, a couple of heart attacks, not diagnosed, until this final one. He was in hospital for two weeks, and had 5 Covid tests, all negative. He was discharged home for palliative care, and died within 10 days. If any of those tests had been positive, he would have been included in the figures, which would have been misleading. He most definitely did not die of anything Covid related, even if one of those tests had proved positive.

@MarcelineMissouri
If you are on Twitter good argument against sadly

threadreaderapp.com/thread/1283868410363777024.html

[quote MarcelineMissouri]I’ve seen this theory a few times. Could there be anything in it??

www.medrxiv.org/content/10.1101/2020.07.15.20154294v1[/quote]
It is based on studies that have found that half of people unexposed to covid have T cells that react to it from infection with other coronaviruses
. At the minute nobody knows what they mean though - could mean you get covid asympomatically, mildly, maybe immune to it... Or they could do nothing at all. This theory is that if a big % of people are naturally immune, once 10-20%, of suseptible people get it, the cases seen to drop off quickly. Its just a theory though. A french doctor has been talking about how this may explain why children make up such a low % of cases in all countries:

The IHU today published a new scientific information bulletin in which Professor Didier Raoult explains, among other things, why children are very little affected by the coronavirus thanks to the creation of an antibody. He goes further by claiming that 40 to 70% of the population would have been immunized before the start of the epidemic

In a video published on June 2 on the IHU Méditerranée website in Marseille, Didier Raoult explains why, according to him, children are less affected by Covid-19. “In children, there are few or no cases, it's a mystery, begins the professor from the IHU , so we looked and found something very interesting because we were interested in the frequency of coronavirus before this crisis”.

According to the data collected during this first analysis: “ It is the children who make the epidemic coronaviruses every year. It is likely that all children who live in a community have a coronavirus infection at least once a year , ”he explains in the video.

“Until now, this coronavirus infection was considered to have little to do with“ Chinese coronavirus ”. But that's not true, he adds, presenting a second ongoing study, if you look at people who have had an infection, a significant number of them already have antibodies. So they cannot be affected by the coronavirus because they have an immunity created before this epidemic ” , develops Didier Raoult.

According to him, the children would therefore have developed, long before the epidemic, antibodies fighting against Covid-19 because they would be more often subject to so-called “classic” coronaviruses .

Between 40% and 70% of the population immunized before the epidemic?
The Marseille infectious disease specialist goes further by evoking results external to the IHU, based on an “ ultra-sensitive antibody detection technique” , which he himself confirms, demonstrating that “ between 40% to 70% of the population were immune before the epidemic started . "

28 days is a reasonable figure - from the WHO guidelines - that captures most cases and doesn't include many that shouldn't be included

It is also useful for public health experts & epidemiologists to have a common basis for comparison, to help develop advice and measures
(it's odd that PHE ignored this !)

In the minority of cases that officially remain active after 28 days, deaths should be counted, or at least closely checked for COVID

WHO guidance on counting death as COVID

https://www.who.int/classifications/icd/GuidelinesCauseeofDeathhCOVID-19.pdf?

"A death due to COVID-19 is defined for surveillance purposes as a death resulting from a clinically compatible illness, in a probable or confirmed COVID-19 case,
unless there is a clear alternative cause of death that cannot be related to COVID disease (e.g. trauma).

There should be no period of complete recovery from COVID-19 between illness and death.

A death due to COVID-19 may not be attributed to another disease (e.g. cancer) and
should be counted independently of preexisting conditions that are suspected of triggering a severe course of COVID-19."

"A french doctor has been talking about how this may explain why children make up such a low % of cases in all countries"

Most reports I've read - from mainstream scientists, respected institutions - suggest that this is because children, especially before puberty, have far fewer ACE receptors than adults:
e.g.

https://edition.cnn.com/world/live-news/coronavirus-pandemic-05-21-20-intl/h_045ff0a5390267f878680c220f539695

“Lower ACE2 expression in children relative to adults may help explain why Covid-19 is less prevalent in children,”

wrote researchers from the Icahn School of Medicine at Mount Sinai in the Journal of the American Medical Association.

Catching up on this thread a little late, and still on the issue of late/overreporting, if BigChoc’s analysis is correct about 2,199 incorrectly counted deaths, that is actually huge. Most will have occurred more recently, but even if evenly spread over the last 2 months say, that still equates to an additional 37 deaths a day, very significant in the context of reported average daily deaths of around 80 a day.

There were around 2,200 deaths after the 28 day time limit, using figures in that tweet from PHE
Those figures may change after PHE carry out their review
e.g. we don't know how many of those deaths were assessed by doctors as being at least COVID-related

Going by PHE data quality so far - I wouldn't be surprised by more changes in either direction !
We'll see when / if they update their historical tables etc

The last 2 months are after the bulk of COVID deaths, so the discrepancy during that period would become of increasing importance,
both as statistically the chance of deaths increases with elapsed time
and because the genuine COVid deaths have been decreasing steadily.

I have not read any explanation from PHE why they chose to ignore the WHO cutoff at 28 days,
that almost every other country - including Scotland & NI - followed

It adds to the increasing mistrust over official figures

• which is due to several reasons: to human error, to system bugs, to the very reasonable updates as delayed data is added, to some clumsy attempts at political spin -

and makes the opposing camps of the chronically over-anxious and the "it's just flu" cool kids dig their heels in.

Public health experts in the WHO, SAGE etc have stressed the importance of transparency during epidemics and of the public trusting the authorities.
So this matters, not just to stats wonks like us

Many of us rely on ONS data & reports, which are high quality and give total deaths, excess deaths etc
but we have to accept that most of the public will only read the headline official daily figures

Just trying to get this straight in my own head and I have a question; does the way we’re counting deaths muddy the waters in terms of what may be longer (even short long term i.e a month after recovery from an active infection) term complications from covid? There’s been plenty reported about clotting and I’m wondering if sweeping everything into one count misses important data that could indicate preventable deaths with the right treatment/surveillance.

I’ve seen plenty of discussions that the U.K. figures are particularly bad compared to Italy, Spain and France because we report more than hospital deaths but I don’t know enough about other countries reporting criteria to put much weight behind that tbh. I suspect that may be the case but I don’t have anything concrete to back that up.

@AprilLady

Catching up on this thread a little late, and still on the issue of late/overreporting, if BigChoc’s analysis is correct about 2,199 incorrectly counted deaths, that is actually huge. Most will have occurred more recently, but even if evenly spread over the last 2 months say, that still equates to an additional 37 deaths a day, very significant in the context of reported average daily deaths of around 80 a day.

Indeed, especially as PHE non-hospital deaths are now running at an average of 44 per day in July (day of death not reporting and ignoring 16th) It could make significant difference.

Cornertoninja, the waters are indeed very muddy, and, to cut the authorities a little slack, the full impacts of Covid on mortality and morbidity will only be known over time when full data and statistics are available to study trends before and after (I anticipate there will be many future academic studies on this!).

In the meantime, there is public need and pressure for up to date statistics, and bodies like PHE are having to make it up as they go along to some extent (unlike the ONS, which has BigChoc says, has always had good, solid data collection practices.)

The long term impacts will actually be quite a tangle to sort out - those directly due to Covid, those indirectly due to Covid because of lasting impacts on health, those indirectly due to Covid because NHS resources were focussed away from other diseases (such as cancer) and because people delayed seeking treatment for emergencies such as heart attacks.

@BigChocFrenzy

"A french doctor has been talking about how this may explain why children make up such a low % of cases in all countries"

Most reports I've read - from mainstream scientists, respected institutions - suggest that this is because children, especially before puberty, have far fewer ACE receptors than adults:
e.g.

[[https://edition.cnn.com/world/live-news/coronavirus-pandemic-05-21-20-intl/h_045ff0a5390267f878680c220f539695]]

“Lower ACE2 expression in children relative to adults may help explain why Covid-19 is less prevalent in children,”

wrote researchers from the Icahn School of Medicine at Mount Sinai in the Journal of the American Medical Association.

That could well be the case. There is an article out today that states that as many studies have now found these cross reactive t cells, it is now really important to find out how protective they are, if at all.

immunology.sciencemag.org/content/5/49/eabd6160

A contentious area in SARS-CoV-2 immunity has been the extent to which there is cross-reactive immune memory from past infections by distantly related HCoVs. After the initial premise that the present outbreak has had such great impact partly due to being a new introduction with no prior immunity in the population, antibody studies indeed indicated cross-reactivity, at least for some antigens; individuals never exposed to SARS-CoV-2 can demonstrate cross-reactive antibodies to the nucleocapsid (17). At the level of TCR recognition, it might be predicted that epitopes representing stretches of conserved sequence might confer cross-reactive recognition, even between viral sequences that have low overall sequence conservation. Studies have produced somewhat divergent answers, sometimes attributable to focusing on different antigens or using different measures of T cell recognition. A significant proportion of pre-COVID-19 blood donor samples show cross-reactive immunity to SARS-CoV-2 S and M peptide pools (4, 9). Studies have highlighted the presence of cross-reactive epitopes within the ORF1 region that can elicit responses in non-SARS-CoV-2 immune people who presumably have been exposed to other HCoVs (16). Less surprisingly, people with immune memory for SARS-CoV-1 mount good, cross-reactive responses to SARS-CoV-2 (16). On balance, the evidence that a subset of people has a cross-reactive T cell repertoire through exposure to related coronaviruses is strong. The key point that remains to be determined is to what extent this could impact protection from disease. For example, is this a factor that could underpin decreased susceptibility in school-age children, presumed to be regularly boosted by exposure to common cold HCoVs

Thanks @AprilLady.

I think it’s a pretty common suspicion that we’re over reporting but then I remember Germany getting criticism for not counting covid deaths where it didn’t appear as the primary cause on a death certificate. It’s a tricky balance particularly given the short time span all of this has occurred and the sudden interest in these particular sets of data.

I can’t decide which is more harmful - over or under reporting, but either way excess deaths will add another facet but like you say - this all takes considerable time to iron out.

Baaah Covid-19 does nasty things to the body that increase clotting of the blood. Therefore if he had had a positive test and then died it would be quite likely that Covid-19 had accelerated his death. Whether his doctor would have put that as the main cause of death only they could say.

This is just nonsense “ between 40% to 70% of the population were immune before the epidemic started . " All the information we have on areas that have been badly hit shows this is nonsense.

Even the harshest criticism of the herd immunity paper recognises that they have a point about the level needed for immunity probably being overstated - if herd immunity was possible (but we dont know if it is).

As for T cells - maybe they will give some protection from infection or severe infection, maybe not. They dont activate unless your vitamin D levels are sufficient and they are not reactive to the most common cold viruses. T cells also change with aging so not going to be much use to those most at risk. www.ncbi.nlm.nih.gov/pmc/articles/PMC3800142/