Re: Coronavirus - Numbers and Statistics

[Hallucigenia]

But effectively they do as if a virus kills all its hosts then it is gone, so the successful virus must not kill all its hosts.

A virus that infects more than one species could wipe one out completely while still being able to spread, however.

Almost all the "bad" introductions of diseases from another country take this form, the hosts in the native range have some immunity but their cousins in other countries don't - potato late blight (Irish potato famine), Dutch elm disease (a great example of a disease getting much worse over time incidentally), the current laurel wilt in North America that must be up there in terms of total biomass killed - 100's of millions of trees, citrus tristeza if you insist on a virus.

[9873210]

I think we need to look at the data again in another week. There could be an increasing number of non-Covid deaths within 28 days of a first positive test (or, for Wales, a first positive test within a six-week testing period).

My graph. Contains public sector information licensed under the Open Government Licence v3.0. : https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/


Julian F. G. W.

What that graph shows is that there is a lag in the numbers from death certificates. That's why the grey area is grey.
The graphs of the most up-to-date numbers has been showing this pattern for over a year, yet as time passes it becomes clear that there is a reporting lag so that for the most recent dates death certificate data was incomplete.

I predict, with confidence, that in two weeks a similar graph will be available showing the two lines tracking closely up until Jan 1 then diverging. I also predict that people will still use this as evidence that "this time it's different".

Updated Julian graph to show the three measures correlating nicely for another two weeks.

[jfgw]

Updated Julian graph to show the three measures correlating nicely for another two weeks.

Deaths with Covid-19 mentioned on the death certificate (which I have identified on the graph as "Certified Deaths"), after being slightly above the deaths within 28 days figures, do seem to have fallen in comparison. This is more easily seen if you take 7-day averages. (Averages are centred around the dates shown.)


My graph. Contains public sector information licensed under the Open Government Licence v3.0. : https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/

I have coloured the areas of incomplete data to correspond with the plot colours although, you will observe, data are updated beyond this 11-day period. That red line will get closer to the dotted black one but I don't think it will reach it.

"Deaths within 28 days" data are today's. These data are subject to revision but they do not change by a great amount, especially after a few days.

I think it likely that deaths with Covid on the certificate will overtake deaths within 28 days after a few weeks:

I see something similar in the South Africa data. For the first three waves deaths followed cases delayed by 14 days fairly closely.
For the omicron wave cases are 6 weeks past the peak and deaths are still increasing (or have just peaked but it's still too soon to see the downslope) although still at a far lower level than prior waves.

https://www.lemonfool.co.uk/viewtopic.php?f=98&t=22737&start=1260#p476999


Julian F. G. W.

[9873210]

Deaths with Covid-19 mentioned on the death certificate (which I have identified on the graph as "Certified Deaths"), after being slightly above the deaths within 28 days figures, do seem to have fallen in comparison. This is more easily seen if you take 7-day averages. (Averages are centred around the dates shown.)

I think it likely that deaths with Covid on the certificate will overtake deaths within 28 days after a few weeks:

I see something similar in the South Africa data. For the first three waves deaths followed cases delayed by 14 days fairly closely.
For the omicron wave cases are 6 weeks past the peak and deaths are still increasing (or have just peaked but it's still too soon to see the downslope) although still at a far lower level than prior waves.

https://www.lemonfool.co.uk/viewtopic.php?f=98&t=22737&start=1260#p476999


Julian F. G. W.

All reasonable. Small (or over the long term large) systematic differences between the various number could occur and could have meaningful explanations. They just don't mean people are fudging data or that it shouldn't be used to manage the pandemic in real time. (A general comment not aimed at you, but there's been many a mole to whack)

Another thing I noted a while ago is that if you plot the ratio of "Covid on death certificate" to "death within 28 days of test" by date of death over the pandemic it is inversely correlated to the number deaths (either measure, since they are 99% correlated). This is consistent with there being some small number of long covid deaths. It is also consistent with doctors being less careful with death certificates when they are overworked*.

I didn't pursue this further since it depended on days with handfuls of deaths and did not seem large enough to be useful. even if it had been statistically significant (which I didn't check).


* Not a slur on medics, paperwork should get triaged.

[jfgw]

After another week's data, the discrepancy between the "deaths within 28 days" data and the eventual "deaths with Covid on the death certificate" data appears to be increasing.
The shaded areas indicate the most recent 11 days' data for the respective plot colour (for example, the most recent 11 days' data for the red plot are within and beyond the pink area).

My graph. Contains public sector information licensed under the Open Government Licence v3.0. : https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/


Julian F. G. W.

[jfgw]

Another week's data. For the most recent reliable(ish) data, deaths within 28 days seem to be approximately 18% higher than certified deaths.

The shaded areas indicate the most recent 11 days' data for the respective plot colour for certified deaths (7 days for deaths within 28 days), for example, the most recent 11 days' data for the red plot are within and beyond the pink area. These data are identified by the UK government as incomplete, although data for prior dates are still subject to revision.


My graph. Contains public sector information licensed under the Open Government Licence v3.0. : https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/


Julian F. G. W.

[jfgw]

This month's monthly NHS publication has been published, https://www.england.nhs.uk/statistics/statistical-work-areas/covid-19-hospital-activity/.

The overall England admissions trend is down, and this can be seen in the largest age groups: 18—64 and 65—84 (60.2% and 15.9% of the population respectively).

Admissions do not appear to be changing much for the 0—5 age group, and 6—17 admissions are possibly falling slowly (7.1% and 14.3% of the population respectively).

The 85+ age group constitutes 2.5% of the population and, while the admission rate is highest for this age group, they are less than 20% of the total admissions.


My graph. Contains public sector information licensed under the Open Government Licence v3.0. : https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/


My graph. Contains public sector information licensed under the Open Government Licence v3.0. : https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/


My graph. Contains public sector information licensed under the Open Government Licence v3.0. : https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/


Julian F. G. W.

[jfgw]

Another week's death certificate data have been published today. This time, I have omitted the most recent data identified by the government as incomplete. Although older data are still subject to change, you will see that this is minimal. Deaths with Covid on the death certificate are now about 20% lower than deaths within 28 days of a positive test.


My graph. Contains public sector information licensed under the Open Government Licence v3.0. : https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/


Julian F. G. W.

[jfgw]

Another month's admissions by age data have been published,


My graph. Contains public sector information licensed under the Open Government Licence v3.0. : https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/

Data from here, https://www.england.nhs.uk/statistics/wp-content/uploads/sites/2/2022/03/Covid-Publication-10-03-2022.xlsx.


Julian F. G. W.

[jfgw]

The "Deaths within 28 Days" and "Deaths with Covid on the Certificate" statistics are converging, although there is still some difference. I have not included data that the government indicate are incomplete.


My graph. Contains public sector information licensed under the Open Government Licence v3.0. : https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/

Here plotted are the ratios of these two sets of data,


My graph. Contains public sector information licensed under the Open Government Licence v3.0. : https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/


Julian F. G. W.

[pje16]

UK case up 43% last week
while Ukriane is dominating the news (quite rightly)
the pandemic seems to have almost been forgotten

https://www.worldometers.info/coronavir ... ekly_table

[Mike4]

UK case up 43% last week
while Ukriane is dominating the news (quite rightly)
the pandemic seems to have almost been forgotten

https://www.worldometers.info/coronavir ... ekly_table

I think deaths are what make both the media and the general public sit up and pay attention. Low deaths = no longer interested. We have been ticking along at 100 per day for a while now, which seems to be an order of magnitude the Great British Public considers acceptable.

[TUK020]

UK case up 43% last week
while Ukriane is dominating the news (quite rightly)
the pandemic seems to have almost been forgotten

https://www.worldometers.info/coronavir ... ekly_table

I think deaths are what make both the media and the general public sit up and pay attention. Low deaths = no longer interested. We have been ticking along at 100 per day for a while now, which seems to be an order of magnitude the Great British Public considers acceptable.

A couple of days ago, there was an interesting analysis in the FT, explaining that the IFR (Infection Fatality Rate) for COVID in the UK has now dropped below that of 'flu. This is attributed to Vaccinations, Prior infection and relative mildness of Omicron.

More people are dying of COVID than 'flu currently because of the widespread prevalence of Omicron. If I recall correctly deaths from respiratory disease in 2022 are running double that of a typical pre-pandemic year (primarily from 'flu).

I have not seen any current stats on the percentage of serious COVID illness who have not been vaccinated, but it is starting to feel like the urgency has gone out of this for those who are able/choosing to take appropriate precautions. I appreciate that this last statement may strike some as oversimplified and judgemental.

[jfgw]

"Deaths within 28 days" were 35% above "Deaths with Covid on the death certificate" on 27th January 2022 (based upon 7-day averages) but the plots have now converged.

Values cited by the UK government as being incomplete have not been plotted. Values plotted (especially the most recent) are still subject to revision, however.


My graph. Contains public sector information licensed under the Open Government Licence v3.0. : https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/


Julian F. G. W.

[Sorcery]

Does anyone look at the Worldometer statistics anymore? I maintain an interest just to see if anything strange is going on. There does seem to something unusual going on. Looking at the Worldometer statistics for yesterday (yesterday provides complete country coverage, today does not) we get new cases for the top 7 country new cases :
South Korea 362328,
Germany 225387,
Vietnam 175480,
France 116118,
Italy 86867
Netherlands 52897,
UK 52698.

Without even trying to do a new cases per million population comparison, it seems obvious that the places where "track and trace" once worked (SE asia), it no longer does. Must admit I never thought it would, note that our version of that was "Test, Track and Trace", where I was slightly more hopeful about the Test part. It's almost as if we are seeing hubris finding it's inevitable nemesis. Another possibility is that the countries that saw low case numbers as a source of national pride and perhaps minimised the test results cannot be bothered to do so anymore.

[jfgw]

One factor: Natural immunity gives better, longer lasting protection against infection than vaccines. Countries that have suppressed the virus more effectively have populations that are more vulnerable to the current, highly transmissible variant.


Julian F. G. W.

[pje16]

Does anyone look at the Worldometer statistics anymore? I maintain an interest just to see if anything strange is going on.

I still look at them, those "missing" numbers are there now
I do find that recently it takes a little longer for the numbers appear
I keep an eye on the new cases in the UK as there is a growing feeling that Covid is over (Ukraine, quite rightly, has pushed Covid news to the back)
It is NOT over
I was argued agalnst on here a couple of week as as new cases in the UK were "only around 30,000", yesterday it was over 90,000
OK not as many as at the peak but tripling is not good
Worldometer have recently added a weekly trend option, Germany, France and the UK all show increases of over 30%

[redsturgeon]

The Zoe app that has been consistently more accurate than the government daily figure and in line with the ONS survey currently reports an all time high of 258,155 cases.

All regions and all age groups show increases.



John

[daveh]

Looks like the latest wave of covid infections (due to the highly infectious Ba.2 strain) in Scotland has peaked looking at the data from:
https://public.tableau.com/app/profile/ ... 0/Overview

It looks as if it is still rising in England.

[Itsallaguess]

It looks like the recent rise in cases following the lifting of restrictions has peaked -



Source - https://coronavirus.data.gov.uk/

With the number of deaths remaining relatively low throughout this recent up-surge, then it looks like England policy-makers haven't got too much wrong since the New-Year wave -



Source - https://coronavirus.data.gov.uk/details/deaths

Cheers,

Itsallaguess