Is the UK government misleading the public on COVID tests?

By Neil Lock

So, that’s over 9 million COVID tests done in the UK up to June 27th a.m. Sounds pretty impressive, doesn’t it? As of today (July 1st), that count has moved on to 9,426,631 – fourth in the world in total tests! (The UK is also fourth in the world in COVID deaths per million population, and closing in on Andorra for third place; but that’s another story). Now… is that figure believable?

I recently wrote a paper about understanding the published statistics – deaths, cases, tests – on the effects of this virus around the world. It is very long, and a little bit technical – although it does include lots of pretty (and not so pretty) pictures! Those interested in the detail can find it here: https://wattsupwiththat.com/2020/06/20/covid-19-understanding-the-numbers-coronavirus/. I had a bit of a laugh when one commenter at “the world’s most viewed site on global warming and climate change” mentioned me (though, I must say, not totally approvingly) in the same breath as Judith Curry, who is a true climate-science expert!

In the course of writing it, I compared the two primary sources of world-wide statistics on this virus. One is worldometers.info, https://www.worldometers.info/coronavirus/. This is kept updated daily with data provided by the national health systems. The other, far more comprehensive because it includes historical daily data from the beginning of the epidemic, is Our World in Data, https://ourworldindata.org/coronavirus. I used Our World in Data.

I found some interesting discrepancies between the two. One was with the Swedish cases numbers – a political hot potato, because of the lack of lockdown in Sweden. With the help of a Swedish commenter at WattsUpWithThat, I found that the issue seems to arise because the Swedes allocate each positive test to the date the test was done, whereas Our World in Data (whose data, if I understand right, comes via the World Health Organization) allocates each positive test to the date the test was reported, which is often days or even weeks later.

More concerning, though, was the UK’s data on numbers of tests carried out. Now for the UK, new cases and deaths reported by Worldometers and Our World in Data are in sync, with Worldometers always one day ahead. That’s consistent with the idea that Our World in Data gets its feed via a third party. The UK does report tests on a daily basis, but there’s often a gap of three or four days before a particular day’s tests appear at Our World in Data.

So… the daily Twitter update, shown at the head, gives the numbers of new cases and deaths on the day in question as 890 and 100 respectively. I’d expect those two numbers to appear in Our World in Data against the following day, June 28th. And indeed, they do:

But what about those numbers of tests? 4,852,547 is the cumulative total recorded here, against the 9,067,577 stated in the Twitter feed. This means the total reported on the Twitter feed was 87% greater than – i.e. almost twice – the “official” figure which, if I understand right, must have been reported to the WHO. That’s an awful lot of missing test kits!

Is such a discrepancy normal? To answer that question, I compared the UK with other countries. I took the cumulative total numbers of tests per million population reported at Worldometers up to June 23rd, and compared these with the numbers reported at Our World in Data up to June 26th. I had no expectation that the numbers would match anywhere near exactly. Indeed, what I found is that the Worldometers numbers were consistently above the Our World in Data ones, in most cases by between 1% and 18%. This seems reasonable to me, given that testing is still ramping up in many countries, and the Worldometers count will probably include situations such as test kits sent out but not yet returned.

I then plotted the numbers of tests per million from the two data sources on a scatterplot:

The plot thickens! The UK shows by far the biggest discrepancy in absolute terms among all the countries, and as a ratio it is only surpassed by Peru and France (and the French are not providing any meaningful data on tests at all). Of the three other “bad boys,” two, Belgium and Spain, are also among the countries hardest hit by the virus. Exactly the places, where you would expect there to be most political pressure to make the numbers look good!

Even the BBC seem to think all is not well on the subject of COVID testing in the UK: https://www.bbc.co.uk/news/health-51943612. It looks as if one problem is that antibody tests are being counted along with the swab tests, thus making the ratio of positives to tests lower than it ought to be. Also, they are counting test kits that have been sent out, many of which may never be returned. Moreover, the Chairman of the UK Statistics Authority, Sir David Norgrove, wrote to the government a month ago: https://www.statisticsauthority.gov.uk/correspondence/sir-david-norgrove-response-to-matt-hancock-regarding-the-governments-covid-19-testing-data/. He said, among much else: “The aim seems to be to show the largest possible number of tests, even at the expense of understanding.”

And one thing more. As I discovered while writing my article, the UK’s statistics collectors were recently required to move from their original basis of counting people tested to counting tests performed instead. This, obviously, resulted in increases in the headline numbers of tests right through the course of the epidemic. It also, unfortunately, meant that all the daily numbers of tests done in the UK prior to April 26th got wiped. And, while this move did bring the UK more into line with many other countries’ reporting procedures, countries such as Canada, Japan and the Netherlands are still reporting by people tested. So, my guess is that this move (likely both difficult and expensive), the over-reporting of test numbers, and the poor presentation of the data that Sir David criticizes, have all come about because of political pressure from those who want the numbers to look as good as possible. Sigh.

So, is the UK government misleading the public on COVID tests? Sir David Norgrove obviously thinks so; and I agree with him.

0 0 votes
Article Rating
72 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Patrick MJD
July 2, 2020 2:09 am

Is the title a trick question?

Greg
Reply to  Patrick MJD
July 2, 2020 7:21 am

All govts mislead all the time. Beyond that truism, it’s not really trick.

Seems like a false polemic to me. If someone has covid-19, they need testing several times after an eventual successful treatment.

If someone tests negative one and a week later has possible contact with a known covid positive person they need a new test. Why would all that be counted as one “test”. It is not.

What is subject to scrutiny is the govt’s ability to get new test kits out where needed and get tests done. Again the number of tests , not the number of people involved is what we are interested in.

Geo Rubik
Reply to  Greg
July 2, 2020 11:26 am

Here in Illinois we have contact tracing. If you test positive everyone you contacted is automatically listed as positive (with no test) until they test negative. That sure ups the numbers.

Reply to  Geo Rubik
July 2, 2020 12:30 pm

Here in Illinois, I am not participating in any of Governor Hog Jowls stupid games.

Patrick MJD
Reply to  Greg
July 3, 2020 2:04 am

Here in my state of NSW in Australia, there have been a total of about 7800 (I am not sure of the timeframe as I am bored S!HTLESS with COVI-19 and the panic) cases that tested positive, of those about 780 are still being investigated. Just over 100 have died since the plandemic kicked off.

And yet, people are panicking and wanting to feel safe so they go get tested.

old white guy
Reply to  Patrick MJD
July 3, 2020 6:43 am

Just a brief update, The fatalities in the U K equal 0.00064% of the current population.

Big Al
July 2, 2020 2:45 am

It is not a question the CDC is misleading , distorting the truth. Part of the PLAN. No mistake CDC sent live virus out in test kits. As Paw Paw fruits are known carriers of the Chinese flu, be best to avoid them.

David Guy-Johnson
Reply to  Big Al
July 2, 2020 2:52 am

Big Al. I see you are off your medication again.

klem
Reply to  David Guy-Johnson
July 2, 2020 3:24 am

as are you, Devid.

cedarhill
Reply to  David Guy-Johnson
July 2, 2020 4:56 am

But see this meeting of Texas Collins County gov (close to Austin, TX):
https://collincountytx.new.swagit.com/videos/62477 starting at 15:20 ffd–especially the question/answer starting at 18:35.

The site appears to be a legit gov site. If so, then as the “re-opening” began in many states and separate from the issue of timing of case reporting, the entire method was drastically overhauled to include counting as cases those anyone with “close contact”, two symptoms, etc. Then toss in the acceleration of testing and the various types allowed and …

A reminder – government rules only be consent of the governed and a large part of consent is having faith (yes, faith/belief) that gov is providing unbiased facts. It’s not likely many on any side of an issue have the requisite faith.

Greg
Reply to  cedarhill
July 2, 2020 7:48 am

Wow, this looks so much like trying to create fake “2nd wave” simply by playing with very loose definitions. This WILL lead to an exponential growth in both “probable cases” and covid deaths ( where the “probable” qualifier gets lost ).

Dems see this as means of undermining Trump and using member within the bureaucracy to falsify the numbers and create a false statistical pandemic.

MarkW
Reply to  cedarhill
July 2, 2020 3:39 pm

Government sites usually end in “.gov”.

Adam Gallon
Reply to  Big Al
July 2, 2020 4:52 am

Everybody point and laugh.

MarkW
Reply to  Big Al
July 2, 2020 3:39 pm

What else did the voices in your head tell you?

July 2, 2020 2:49 am

“But what about those numbers of tests? 4,852,547 is the cumulative total recorded here, against the 9,067,577 stated in the Twitter feed. “

There may be fine print issues here. If you look at this page, and scroll down to “Tests by type” and click “cumulative”, it gives a total of about 9M, and a breakdown. One type is by commercial partners. If you remove that, about 4.8M remain. So it looks to me as though Worldometers is reflectying total tests, but OWID is reporting government tests only.

“The other, far more comprehensive because it includes historical daily data from the beginning of the epidemic”
I plot here the historic series for cases and deaths for countries and US states, based on Johns Hopkins data. It is updated daily about 4.30am GMT (which JH post). But I don’t plot tests.

Dodgy Geezer
Reply to  Nick Stokes
July 2, 2020 3:49 am

There are actually so many issues with gathering data in the middle of a pandemic, that it is unreasonble to place much credence on any data gathered from anywhere….

Reply to  Dodgy Geezer
July 2, 2020 8:07 am

Maybe so, but it doesn’t seem fair to slam people for doing the best they can. Do you think they should just not try to collect the data?

Reply to  Dodgy Geezer
July 2, 2020 8:49 am

I agree with Nick here. If you do your utmost to make the data you provide as good quality as you can, that’s all you can do. Let the quality of the final result be judged when the dust has settled.

Patrick MJD
Reply to  Neil Lock
July 3, 2020 2:06 am

I find it rather ironic that you think Nick provides (Or uses) good quality data.

Reply to  Patrick MJD
July 3, 2020 7:21 am

Patrick, please read what Nick and I wrote. I said nothing about the quality of any data Nick uses or provides. I merely agreed with his premise that it is better to do what you can with the best data that is available at the moment, than just to say “I can’t use that.” At the least, if the data does contain errors and contradictions, you have a chance of finding them.

Reply to  Nick Stokes
July 2, 2020 5:16 am

Nick, please see my reply at 4:45 am, which unfortunately went in the main thread rather than here.

Reply to  Nick Stokes
July 2, 2020 5:17 am

This site gives a comprehensive breakdown.
https://coronavirus-staging.data.gov.uk

David Guy-Johnson
July 2, 2020 2:51 am

There is no pressure from the government involved. You must produce evidence when making such a claim. Otherwise you risk straying into conspiracy theory nutjob territory

Reply to  David Guy-Johnson
July 2, 2020 4:50 am

David, my suggestion is merely that the government is trying to paint the rosiest picture it possibly can for its own interests. And that that picture is not representative of “the truth, the whole truth, and nothing but the truth.”

Richard (the cynical one)
Reply to  Neil Lock
July 2, 2020 6:13 am

You are saying that (gasp!) our elected ‘servants’ are concealing the real actual truth from us, deliberately putting us at greater risk, just to look better in our eyes?
But if we vote them out come election time, which greedy mendacious crew should we choose then?

Tonyb
Editor
Reply to  Neil Lock
July 2, 2020 6:39 am

Neil

If the govt are trying to paint a rosy picture then they are terrible artists!

Until all countries report cases and deaths in the same way and from all settings on an agreed basis then we shall remain in the dark

Whilst you talk of cases the ultimate test are the number of deaths. Here we have the very knotty problem that one country reports deaths definitely as a result of covid 19, whilst another says people died with it, which is dfferent to dying of it

I know of only 2 people who died, both elderly and who went into that killing field known as a hospital. There they caught cv but it seems more than likely it was the original health problem that was the reason for their passing.

Deaths are now below normal and it seems likely that deaths of the very elderly have been brought forward by a few months and the overall deaths for the entire year are unlikely to show anything extraordinary has occurred.

Hope you will be doing one of your interesting studies on deaths in order to put your article on cases into overall context

Tonyb

AndyL
Reply to  David Guy-Johnson
July 2, 2020 5:06 am

David
The claim of government pressure comes from the UK’s cheif statistician in a direct quote:
“The aim seems to be to show the largest possible number of tests, even at the expense of understanding.”

July 2, 2020 3:05 am

The UK Government has been promising all sorts of targets during this pandemic., for want of a better word, for testing some of the targets have been achieved with minutes to spare. I don’t believe anything this government, led by a mediocre journalist and his svengali says about anything.

Gerry, England
Reply to  Ben Vorlich
July 3, 2020 8:17 am

Very wise especially when our Prime Minister is a serial liar and you can print that in a national newspaper and nothing will happen. Both he and his svengali are ignorant and not very bright. Whoever thought we would look back at Theresa May and say she wasn’t as bad as Boris and did her best with enormous odds stacked against her.

RB
July 2, 2020 3:06 am

If the test is of the “do you currently have the virus” type, just how useful a test is it anyway? To get full coverage you would need to test the entire population every fortnight say. You would hope that antibody testing would be more useful, but we still seem to be learning a lot about that currently.

icisil
Reply to  RB
July 2, 2020 4:06 am

It’s a worthless test. Pathogens cause disease when they increase to such large numbers that the cellular damage they are causing manifests as symptoms. Without symptoms there is no disease, but the prevailing unscientific paradigm equates one viral particle, detectable or not, to disease.

Samuel C Cogar
Reply to  icisil
July 2, 2020 4:38 am

So, one has to ask, …. how many “Typhoid Mary” Covid-19 carriers have been included in the “infected” totals?

A C Osborn
July 2, 2020 3:15 am

Does anyone know why worldometers have drastically reduced the numbers of daily deaths for the last few weeks in Sweden when worldindata haven’t?

Reply to  A C Osborn
July 2, 2020 5:09 am

As I understand from Mats Bengtsson who commented on the last thread, a couple of weeks ago the Swedes decided to stop all external reporting at week-ends and holidays. And they have just had a big midsummer festival. The numbers of deaths (and cases) at One World in Data for June 20th, 21st, 22nd, 28th and 29th are all zero; so, I presume, the following days will have higher than usual counts to make up.

As to why the Swedish deaths at Worldometers have gone down, a possible hypothesis might be that on that platform they are switching to recording deaths by date of death rather than date of report, as they have done for cases. I did notice that their deaths curve at Worldometers, which last week had a weekly wiggle in it, has now been greatly smoothed! That would bias recent deaths numbers low, until the reporting system has had time to catch up.

Carl Friis-Hansen
Reply to  A C Osborn
July 2, 2020 5:14 am

I cannot answer that question in detail, but it could be a question of distinguishing between death with and death of.

There is really only one figure you can rely on, namely the excess deaths in comparison to previous years.
https://carl-fh.com/images/offsite/sweden-death-week-20200629.pdf

As you will see from the above link, the casualties from this bad virus is now history.

I personally have the feeling that it may also be over in many other countries, but the problem is to get up-to-date raw data of total deaths per week and compare them to an average over the the same moths for other years.

Joe Crawford
Reply to  Carl Friis-Hansen
July 2, 2020 11:29 am

I agree with the validity of ‘excess deaths’ for a country that didn’t implement a lock down. However, for countries that did implement some form of lock down there have been several ways suggested where the lock down itself can potentially cause as many or more deaths that the virus. You might be able be able to get data for the major causes of death and then plot excess deaths for each of those but then figuring out which or what percentage of those to apply to the virus would not be a simple task.

Dodgy Geezer
July 2, 2020 3:47 am

“…Is the UK government misleading the public on COVID tests?…”

EVERYONE is misleading the public on EVERY aspect of COVID. Because NONE of the data is either accurate, or measured from a base which allows reasonable comparisons. We will only know what happened with any degree of accuracy, and be able to learn lessons, in a couple of years.

Incidentally, the rush to set up a testing regime in the UK meant that many initial tests turned out to be not fit for purpose. As one example, difficulties in attaching bar codes to samples meant that a lot of test results had to be discarded. How do you count this? As tests done, or not? There are a lot more issues like that, everywhere….

Reply to  Dodgy Geezer
July 2, 2020 5:28 am

An honest person would count tests which failed to produce any results in a separate category like “tests whose results could not be evaluated,” and would attach a reason code to each test placed in that category.

A politician, on the other hand, would count them as tests done, thus achieving the twin objectives of making the number of tests done look as high as possible, and the positives per test as low as possible.

Dodgy Geezer
Reply to  Neil Lock
July 2, 2020 7:15 am

What you have said may be trivially true – but not at all relevant.

That is how an honest individual would behave when writing a paper. however, governments do not and cannot work with individuals. They work with bureaucracies.

A bureaucracy comprises multiple layers of responsibility – each one looking after itself. A politician may order that tests will be done, and that he is to be given the total numbers, but along the line this instruction will be re-interpreted multiple times for the benefit of the person doing the interpretation. Evan an honest politician (if such a person were to exist) would find it very difficult to provide data in the way that he wanted.

You will rapidly find, in a bureaucracy, that you can only do the things the bureaucracy will let you do, and that there is no obvious person to blame when things go wrong. “Yes, Minister” refers….

icisil
July 2, 2020 3:54 am

The tests are crap. How can anyone place confidence in something that oscillates back and forth between positive and negative test results in the same person(s)? Kary Mullis invented PCR as a manufacturing technology and disagreed with its use for diagnostic purposes. None of the manufacturers claim their tests are to be used for diagnostic purposes; research only.

COVID19 PCR Tests are Scientifically Meaningless
https://off-guardian.org/2020/06/27/covid19-pcr-tests-are-scientifically-meaningless/

icisil
Reply to  icisil
July 2, 2020 4:28 am

Here’s an example of how PCR was used to diagnose a pertussis outbreak that actually didn’t exist.

Faith in Quick Test Leads to Epidemic That Wasn’t
https://www.nytimes.com/2007/01/22/health/22whoop.html

Reply to  icisil
July 6, 2020 8:30 am

PCR was also used to diagnose an HIV outbreak that, as far as I can tell, also doesn’t exist…

Newt2u
July 2, 2020 4:22 am

There are lots of ways in which the data is being presented in a disingenuous way. For example one of the most egregious is the excess deaths reported over the covid spike compared to the 5 year average. This has been presented by the Beeb, one of the worst scare mongers in this scam, over the 11 weeks around the peak in the corona virus, i.e. offset compared to the normal peak in respiratory deaths, which normally happen around December January rather than April May. Of course this will show much more alarming numbers.

Adam Gallon
Reply to  Newt2u
July 2, 2020 4:51 am

You mean like this one, that goes back to December/January & shows mortality within the normal range, in December/January & soaring skywards in April. far above the 5 year average for December & January? https://images.app.goo.gl/CwMtDiCmV1ZukV3y9
Now, what was your objection again?

TRM
Reply to  Adam Gallon
July 2, 2020 8:27 pm

You have to be careful just going by graphs. The Financial Times did one and put the CSV file on GitHub so I downloaded it and looked at it. They only used 11 of the 52 datasets from the CDC. 50 states plus DC & New York City (CDC counts NY state and city separately).

I’ve been doing up my own “excess deaths” from the CDC data for all 52 sets. As of the data on June 29th the year to date excess death rate for the entire USA is 10.1% (124,419) higher than the previous 4 year average for weeks 1 to 22. West Virginia is missing week 22 and North Carolina is missing weeks 20, 21 & 22.

I’ve added a “Monthly” folder with all the data for each month. You can view the rise and fall of excess deaths for the country as a whole or on a state by state basis. This should help see the effects of re-opening. April was the worst month with New York City being more than 4 times their average.

For the entire USA: January -0.3; February 0.7; March 4.6; April 33.3; May 11.6

The script and all related files are here if you want to kick the tires:
https://www.dropbox.com/sh/fh9x5fngmfbeiiu/AAAH-OtOMqiY_R9qqG6YccCRa?dl=0

newt2u
Reply to  Adam Gallon
July 3, 2020 12:57 am

Actually it was a percentage figure given. Comparing the spike with normal deaths over the 11 weeks of lockdown. Given that the normal deaths usually spike over the cold/flu season of December through February, this headline figure is very misleading if you want to compare excess deaths to a normal year. I would also say that the last 5 years have been relatively benign in this respect. A more sensible comparison would be to take the average excess deaths in the last 5 bad flu seasons, where we were not locked down. I would guess the deaths this year are little different, only time shifted by 3 months.

Tom in Florida
July 2, 2020 4:45 am

The problem with testing is that if you test negative today that doesn’t mean you won’t have the virus tomorrow. A negative test result does not mean immunity. The only way to get it right is to test everyone for antibodies. Those that are negative for the antibodies should then have the virus test every two weeks.
Of course that is probably impossible to do.

Reply to  Tom in Florida
July 2, 2020 8:14 am

Actually a negative antibody test does not mean that you haven’t been infected.
PCR Antibody Symptoms Conclusion
+ + + Infection probably been hospitalised – reactionary immune system response

+ – + Infection currently – Innate immune system managing

+ – – May be currently infected or has been – Innate immune system controlling infection

– – – No infection

This is based on my own understanding.

July 2, 2020 4:45 am

As I understand it, Nick, the headline number of tests includes all four of the so-called “pillars” of testing. In fact, the DHSC’s daily message says: “Due to revisions in historical data in Pillars 1, 2, 3 and 4 the cumulative total for tests is 9,022 higher than if you added the daily figure to yesterday’s total.” That very strongly suggests that the headline total is intended to cover all four pillars.

Pillar 1 is swab tests carried out by government labs and NHS hospitals.

Pillar 2 is swab tests carried out by “commercial partners.” That seems to mean, drive-in testing centres and private labs processing home test kits ordered on-line.

Pillar 3 is antibody tests.

Pillar 4 is other blood tests. (I think this is what Sir David Norgrove is referring to when in his letter he talks of “the ONS survey.”)

It seems the local MSM have woken up to this issue at much the same time as I did: https://www.express.co.uk/news/uk/1303877/Pillar-1-and-Pillar-2-coronavirus-cases-explained-UK-test.

There seems to be doubt about whether sub-national testing statistics include Pillar 2 tests, but the data at One World in Data is supposed to be national, so should include both. Your suggestion that the discrepancy may be due to the non-inclusion of Pillar 2 tests in the OWID figures isn’t borne out by looking at, for example, the period May 3rd-10th. In the graph you linked to, NHS tests (Pillar 1) were bumbling along around the 30K per day mark at that time. But OWID data is showing daily numbers in the 60Ks. More recently (June 21st-27th), the OWID numbers look closer to the sum of Pillars 1 and 3 than anything else; but the match isn’t that close.

Furthermore, a recent report from inews.co.uk suggests that Pillar 2 test results aren’t being made available to the general public, or even to local authorities: https://inews.co.uk/news/politics/pillar-2-testing-data-local-government-covid-19-test-figures-local-lockdowns-what-explained-460435. If that is indeed so, it doesn’t make any sense (except to a politician) to count Pillar 2 tests in the “tests done.”

And further, one of Sir David Norgrove’s criticisms in his letter I linked to is: “the top summary presents the number of positive results from diagnostic tests (pillars 1 and 2) alongside the total number of tests across all pillars. This presentation gives an artificially low impression of the proportion of tests returning a positive diagnosis.” In other words, to assess fairly the ratio of positive tests to all tests, only the diagnostic tests should be included in the tests counted. Pillars 3 and 4 should be excluded.

Reply to  Neil Lock
July 2, 2020 8:04 am

“If that is indeed so, it doesn’t make any sense (except to a politician) to count Pillar 2 tests in the “tests done.””

Well, the tests were done. The fact that results are not well communicated to the public doesn’t change that. They were done for the benefit of the patients.

I think both the 4 pillars total and the 3 pillars total are meaningful – the total of tests with results available to the public is also useful. The fact that the govt lists both is fair, and is clearly set out in the page I linked.

Reply to  Nick Stokes
July 2, 2020 9:18 am

Indeed so, Nick; but the title of my article was “Is the UK government misleading the public on COVID tests?”

Separately, the totals are meaningful. They might even be close to the reality! But to aggregate them into a headline number, without being clear about what it is that you are aggregating into that headline, is misleading in my opinion. And that was the essence of Sir David Norgrove’s criticism.

Yooper
July 2, 2020 5:06 am
Jerry Mead
July 2, 2020 5:06 am

The UK’s Royal Mail sent out an expensive courier specifically to collect my and my wife’s carefully taken swabs.

That was four weeks ago. I’m still waiting to discover whether or not we were positive. Then.

Vuk
Reply to  Jerry Mead
July 2, 2020 6:22 am

Hi Jerry
Perhaps you need to check with the register of diseased at your local town hall, just in case you have died since from the dreaded CV infection. / sarc
http://www.vukcevic.co.uk/EuropeCV.htm

Rod Evans
Reply to  Jerry Mead
July 2, 2020 9:54 am

Jerry,
I was part of the random statistical group tested by Imperial. sent my swab back by courier on Friday and received a result withing the week on Wednesday. A pointless exercise as it was an antigen test. Why won’t they do antibody testing and learn something useful?

JohnM
Reply to  Rod Evans
July 3, 2020 1:08 am

Too expensive

July 2, 2020 5:16 am

We have a clear indication of the intention to mislead. I regularly produce maps based on the official dashboard data on positive test results by local authority across England, and across the smallest geographical reported areas for GB as a whole.

The current kerfuffle over extending the lockdown in Leicester seemed to me to be much ado about nothing. Leicester simply doesn’t figure as having an alarming level of infections or in the rate of change of infections based on the publicly available data. See this map of recent reported new cases

https://datawrapper.dwcdn.net/B7qlJ/1/

And this one showing rates of change

https://datawrapper.dwcdn.net/zYt9u/1/

You need to be very familiar with the geography to pick out Leicester.

Here is a chart of the daily cases by date of sampling from the same PHE data pushed to the public

https://datawrapper.dwcdn.net/K3Wry/1/

Not declining towards zero, but certainly no upward spike, and a fifth of peak levels.

Then yesterday the cat was let out of the bag.

https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/897128/COVID-19_activity_Leicester_Final-report_010720_v3.pdf

The published data only refer to Pillar 1 testing, and there is a huge level of positive pillar 2 tests that are nowhere else published by locality. Not that, if you read the report, there is much clue on how to interpret the results.

I still have all the downloads I’ve made, so it would be possible but laborious to examine how the data have changed.

Reply to  It doesn't add up...
July 2, 2020 7:51 am

Thanks for the link to that report. And you’re right, they don’t seem to have a clue as to how to interpret the results. Nor, quite honestly, do I, from what I read. Somebody is going to have to do some detailed digging to find out what’s going on in Leicester.

Steve Borodin
July 2, 2020 5:31 am

One problem is that we have the latter-day Lord Haw-Haw, alias the BBC, critically criticizing everything the hated UK government does. They are milking the pandemic avidly. Had the Government fallen behind in testing they would have had a field day.

Martin Howard Keith Brumby
July 2, 2020 5:56 am

The most worrying Pandemic isn’t the ChiCom-19, bad though that obviously is.

It is the rampant Pandemic of Crass Stupidity, sweeping remorselessly across (particularly) the Western World.

My hunch is that this Stupidity will lead to way more deaths than the ChiCom virus.

Many people, thankfully, survive the latter.

There is no cure for Stupid.

Roger welsh
Reply to  Martin Howard Keith Brumby
July 2, 2020 10:51 am

Well put

fretslider
July 2, 2020 6:55 am

Is the UK government misleading the public on COVID tests?

One may just as well ask, Is the Pope Catholic?

Kevin kilty
July 2, 2020 7:13 am

With the help of a Swedish commenter at WattsUpWithThat, I found that the issue seems to arise because the Swedes allocate each positive test to the date the test was done, whereas Our World in Data (whose data, if I understand right, comes via the World Health Organization) allocates each positive test to the date the test was reported, which is often days or even weeks later.

Colorado allows a person to choose whether to view date of result or date of onset of symptoms. A comparison is instructive to say the least. Reporting by date of onset gets ride completely of the fantastic one-week period swings so prominent in most jurisdiction’s reporting.

decnine
July 2, 2020 7:41 am

The UK Civil Service is having a terrible epidemic. Everything they touch turns to cr*p. Public Health England, the agency supposed to be prepared for epidemics, wasn’t. They have spent most of the last decade appeasing ‘activists’ who want to control people’s intakes of salt, sugar, fat, meat, you name it, they want to restrict it; except insects. The approach to testing was concocted on the backs of several advisor’s envelopes. Whose envelope got chosen to be believed on any given day seems to have been chosen with the aid of a Ouija board. The politicians are getting fed up with having to face the press to take the flak that ought to be hitting the bureaucrats. Prominent Civil Service heads are beginning to roll. The Civil Service has every incentive to make the numbers look as good as possible. But most of them read Classics at Oxford. Numeracy is not their strong suit.

n.n
July 2, 2020 10:51 am

Counting twice, thrice, dates of conception and confirmation, and a conflation of causes.

Keith Peregrine
July 2, 2020 3:17 pm

If the UK graph number catches one’s eye, so too does Costa Rica. Wonder what’s happening there.

DP
July 2, 2020 8:05 pm

I’ve been using https://covid19info.live since Day 6 – Spreadsheet available to anyone interested

newt2u
July 3, 2020 1:29 am

An interesting analysis of the accuracy and specificity of the PCR test… or how the numbers can mislead even if they are accurate:

https://wwwkevboyle.blogspot.com/2020/07/covid-19-testing-is-pure-insanity.html

William Astley
July 3, 2020 1:03 pm

Let the Bells Ring and the Banners Fly!!!! GREAT NEWS!!!!

A game changer. Cut the covid death rate by 50%

It as unequivocally been shown, that HCQ Reduces the Covid Death rate by more than 50% when it is given to the patient, early, …

… well before the patient is near death.

The studies that alleged HCQ increased the risk of heart damage and death, gave HCQ to patients who were on the verge of death, which is close to criminal….

As, it appears, those gave HCQ to patients who were near death, …..

….their objective, was to hide the effectiveness of HCQ by given it purposely to patients who were near death.…. Is everything about politics or is it money?

A study of 1300 patients showed, unequivocally, that HCQ does not cause heart damage.

The same study shows HCQ when it is given to the patient early, reduces the patient’s chance of death by more than 50%.

A 50% reduction in death rate, with no serious side effects, is a stunning improvement in covid disease outcome.

https://www.foxnews.com/politics/hydroxychloroquine-helped-save-coronavirus-study

Hydroxychloroquine helped save coronavirus patients, study shows; Trump campaign hails ‘fantastic news’

Hydroxychloroquine lowers COVID-19 death rate, Henry Ford Health study finds
Researchers at the Henry Ford Health System in Southeast Michigan have found that early administration of the drug hydroxychloroquine makes hospitalized patients substantially less likely to die.

Reply to  William Astley
July 4, 2020 5:06 am

Richard Horton has a lot of deaths on his hands having published the now retracted paper in the Lancet that caused research on HCQ to be halted in several projects. Publication would appear to have been politically motivated to judge from Lancet editorials. He has no place running a supposedly scientific journal.

William Astley
July 3, 2020 1:45 pm

This is link to one of the new HCQ studies.

This is a big deal as it looks as if the Covid death rate can be reduced by 70% and…

It appears, negative HCQ, results were generated by, in appropriately, giving HCQ to patients, who had covid lung damage and were struggling do to lack of oxygen. HCQ can prevent covid damage, it cannot reverse covid damage. There is no medical reason, therefore to give HCQ to patients who are near death and have serious covid lung damage. It is too late, and the weaken, oxygen starved heart is adversely affected by HCQ.

HCQ & Azithromycin given to patients early in the covid disease progression, showed a 71% ‘hazard reduction ratio’.

The patients treated with HCQ & Azithromycin early, had a 71% less chance of dying and those that did not die, recovered earlier with less serious damage.

The study proved that HCQ does not damage the heart, …. no patient had documented ‘torsades de pointes’. The reason why that is true, is the HCQ has not given to patients who were near death.

If the patient is near death, HCQ and Azithromycin is not effective and should not be given to the patient, as the covid damage was been done, (HCQ is effective in preventing damage, it cannot reverse damage), ….

… after covid has damaged the lungs, the heart is struggling due to a lack of oxygen. For patients whose hearts are struggling for oxygen, HCQ adversely affects the oxygen starved heart.

Hydroxychloroquine provided a 66% hazard ratio reduction, and hydroxychloroquine + azithromycin 71% compared to neither treatment (p < 0.001).

https://www.ijidonline.com/article/S1201-9712(20)30534-8/fulltext

Treatment with Hydroxychloroquine, Azithromycin, and Combination in Patients Hospitalized with COVID-19