A covid update?

[NoelC]

As the government appears once again to be well behind the curve as the second wave gets well and truly started, I fear we will be put into an effective age related lock down.  Que another quick check on the net to catchup with what's happening...

I'm signed up with the Covid 19 symptom study run by UCL/Zoe, 4M people login and their figures on infection rates are generally several times higher than the governments figures (because you don't have to queue for a test and their app is somewhat simpler than the NHS one).  At present they are holding the daily infection rate at 12,698/day verses government's figures of 4,926.  See their website for updates:-
https://covid.joinzoe.com/data
Make sure you know the symptoms (they keep changing) - as it's important if you suspect you have it or know someone who isn't well. Signing up for the app will allow you to go through their symptom checker (and may offer you a test).

What's the risk from Covid-19?  The ALAMA website is the detailed approach to estimating individual risk of mortality from Covid for the UK medical community:-
https://alama.org.uk/covid-19-medical-risk-assessment/
For the average white male at 60 the risk is 0.7%, at 70 that rises to 2%, women are nearly half that risk, all the other factors are listed on the ALAMA page to allow you to arrive at a 'Covid age' which relates your risk to an average white male.

However there is good news too.  In a recent small trial in the University Hospital Cordoba (Spain) they ran a clinical trial on the effects of large doses of Vitamin D on the outcomes for hospital admissions for Covid-19 (i.e. all diagnosed with the disease).  The outcomes were improved by 90% (2% required ICU and none died as opposed the control group in which 50% required ICU and 16% died).  This was a very small trial, but as in other areas where Vitamin D has been shown to be benficial, this should lead to further investigation into the benefits of Vitamin D supplementation. According to the BMJ, PHE has been looking at this since June (with no outcomes yet).
https://www.sciencedirect.com/science/article/pii/S0960076020302764#

The following article discusses the benefits of Vitamin D in preventing accute respiratory illness:-
https://www.thelancet.com/journals/landia/article/PIIS2213-8587(20)30268-0/fulltext

The Journal of the American Medical Association reportedly carried an article looking at Covid-19 susceptability in patients who had been diagnosed deficient in Vitamin D  (based on 486 cases); their findings were that on average deficient patients were 1.77 times more likely to die from the disease.  In their article they noted that 80% of Black people were deficient, 70% of the hispanic comunity were deficient verses 42% in the general population. Given the greater sucseptability to Covid among the black community these findings were considered significant.

The appropriate dose is around 1000iu (25mcg/day) - as it is incredibly cheap in comparison to pharmaceuticals and does not damage at these levels, it would seem sensible to issue everyone in the country with it!  Get your orders in now, because if the media talk this up much more; it will go out of stock.

Vaccine trials look set to be mired in the US election fall out:-
https://www.cbsnews.com/news/covid-19-vaccine-when-will-be-available-ready/
And the UK vaccine has hit some snags:-
https://medicalxpress.com/news/2020-09-astrazeneca-covid-vaccine-trial.html
Leading to doubts about whether it will make it.
Lets hope we don't have to fall back on the Russian alternative:-
https://www.rt.com/russia/501417-putin-free-sputnik-vaccine-un/
Not that I'm criticising Vladimir of course! However I suspect the free supply of vaccine might reduce the number of critics the Kremlin's foreign policies attract (permanently).

Any other interesting articles?

[Roberto]

I'm signed up with the Covid 19 symptom study run by UCL/Zoe, 4M people login and their figures on infection rates are generally several times higher than the governments figures (because you don't have to queue for a test and their app is somewhat simpler than the NHS one).  At present they are holding the daily infection rate at 12,698/day verses government's figures of 4,926.  See their website for updates:-
https://covid.joinzoe.com/data

Noel

I could not see from the link above but are these just reported cases, as if I feel I have Covid and report and it increases the number by 1 or confirmed cases?   If confirmed, are there any statistics on the % of false positives?

Roberto

[Carole]

I'm signed up with the Covid 19 symptom study run by UCL/Zoe, 4M people login and their figures on infection rates are generally several times higher than the governments figures

I have been participating in this since it first came out (which reminds me I don't think I have reported in today yet).  On one occasional I wasn't feeling well with a tummy upset and reported this on the daily return.  the following day I got an E mail asking me to take a Covid test which I and Adrian duly did(we did a postal one).  This was back in July, and we got the results by text 2 days later (negative).  I think they were looking for other symptoms that might be associated with Covid.

In the last 2 days I have walked down to our local shops a couple of times for essential reasons, and i could not believe the lack of social distancing, the number of people who overtook me too close (and I didn't see them coming), or people walking towards me and only I took evasive action.  I passed 2 workmen drilling close to the kerb earlier today so I walked as far to the left as I could.  On the way back, there were 2 other men (not workmen) standing chatting to them on the pavement far too close.  Then as if that wasn't bad enough 2 Asian people walked past the 4 of them even closer to the ones on the pavement (I had crossed the road to avoid this congestion) and yet Asian people are supposed to be more vulnerable.  Could not believe what I saw.

Has the general public just given up on social distancing?  I shall avoid walking anywhere now (apart from the park) unless absolutely essential, yesterday's trip was essential as I had to visit the dentist with a gum abscess and it was very painful.

Carole

[Carole]

Hmmm,

I reported that I had a painful gum abscess a couple of days ago on the Covid19 app and today I have been asked to take another Covid test.  This will be my 3rd test including the one I had in hospital a few weeks ago.  Still need to support the research, so Adrian and I will be trotting off to Cotmandene testing centre this afternoon.

Carole

[NoelC]

Roberto RE Covid-19 App (https://covid.joinzoe.com/)
Not sure how it works, can't see an exact description of methodology on Tim Spectre's site.  The figures were produced by a team of epidemiologists at Kings College London. 
I assume its algorithmic; i.e. if you fill out the symptoms, they will assign your response a weighting based on the number of positives they have had historically with the same symptoms.  But as you can see from Carole's experience (which is the same as mine - I was offered a test when I had a cold in May), they do cross check for correlation with PCR testing and I assume highlight new symptoms if they occur (such as the rash).  I assume they are getting tier 2 allocations for this.   
But, my understanding is these numbers are essentially symptomatic Covid verses tested Covid (which the government supply) and they do only offer this as an 'estimate' in the same way as they offer their prevalence data.  The reports go to SAGE and are important because they have sufficient numbers on the app to break down infection hotspots by area.  In my area (Gravesham) there are 1,500 contributors with an estimate of 67 cases (0.1%). In Bromley there are 9669 contributors with an estimate of 734 cases (0.3%).

Noel

[Roberto]

Thank you Noel.  My issue with the "sudden" jump in cases - after PCR testing - is the margin of error which arises from false positives.  As I understand it, the current method has a false positive incidence which is well above (multiples) the sampling rate.  So for example, your 0.1% in Gravesham could be swamped by a false positive rate of 0.8% from the PCR test.  In a 10,000 sample this is a difference between 10 - real - cases and 80 false positives!

Roberto

[NoelC]

Roberto
The false positives argument would not apply to these numbers (unless they correct the numbers as a result of testing) since they are only estimates based upon symptomatic infection rates - although false positives could affect the algorithm.  There is in any case little correlation between these estimates (currently at 16,130) and the governments figures (6,178).  The difference being: not everyone who thinks they've got it get tested, even if they do it will be up to a week before their figures register in the .gov.uk figures, by that time - at the current rate, the symptomatic figures will be 30,000+

Noel

[MarkS]

The two numbers are measuring different things.

The Gov't figure is the number of people testing positive.  The covid.joinzoe figure is an estimate of the actual number of new cases - probably most of them have not had tests.

The Gov't figure will always be constrained by the number of tests available.  Demand for tests will be greater than supply right through the Winter months.  They need to be targeted where their effect will be most beneficial e.g. keeping the NHS, schools and essential services running as near normal as possible and this time around to properly "ring-fence" care homes.

Mark

[Roberto]

I know the numbers are different. What I'm arguing is that both are inflated (witness people reporting symptoms every time they feel discomfort; are there even cohorts in the summary numbers?). And then the government's numbers have the risk of being full of false positives.
I completely agree tests should be targeted and the vulnerable should be protected but all measures have so far been applied without any sense of statistical significance.

[Carole]

Thanks for explaining that in plain English Mark, my brain was getting fried with the the statistical jargon.

Carole

[Rick]

Has the general public just given up on social distancing?

The number of folk wearing masks in shops (and wherever) round here seems pretty low. I gather (from a friend who drives busses) that "almost nobody" is wearing a mask while on public transport.

And then the government's numbers have the risk of being full of false positives.

Given the level of fines they've introduced, as I see it, a far higher risk (people being people) is that many fewer people with possible symptoms will come forward to be tested. The folk who will come forward to be tested are the ones who're pretty sure they're not infected. Same goes for the contact-tracing. Far as I can tell, it's all stick and no carrot, and they're starting from a position where four out of five people who should be self-isolating (or in quarantine, or getting tested) aren't in fact doing so.

[Mac]

Has the general public just given up on social distancing?

[RANT]

The thing I noticed in spain was EVERYONE was wearing a mask, at all times, the exceptions were if you were in a restaurant or bar, doing exercise or on a beach.
Even the local toe rags on the corners were all wearing one, Spanish law states €1000 if you are not wearing one and its enforced, no exceptions.
Social distancing was pretty much common place. Even at the school pick up, which they staggered over a period of 30 mins.

The problem over here is no one gives a $hit, people know they wont get a fine, they just dont care, probably because of all the media and government junk thats thrown
down their necks, with constant changing rules, you cant meet with more than 6 people, unless your in a pub, or onpublic trasport, families cant meet with more than 6, but school children can mix with 30,
oh and dont worry we'll close the pubs early. Oh any you only need to wear your masks in shops and public trasport, but you dont if you work  there, oh and somepeople dont have to wear masks.

Hello people, a lot of people sleep under sheets at night, guess what, it doesnt kill you or stop you breathing, same as a mask......

What needs to be done is for someone with a large pair of B0110cks to stand up and just say.

Masks are mandatory everywhere (execpt home), £1000 for not wearing one, and bloody enforce it. Doesnt matter if you are working or not.
If little Johnny on the corner with his 7 gang bangers dont want to wear one, thats ok, they will just have to pay £8000 for the privilege, they will all be wearing one tomorrow.
If our numbers start to spike, start an immediate lock down for four weeks, if no one moves for four weeks, neither does the virus.
Four weeks should we more than enough for anyone to catch the virus, have symptons and get over it. (thats what the government tell us  )

None of this, "well, if, maybe, it might, well wait and see, its looking like." Just lock it down, 4 weeks is not going to kill the economy, but 6-18 months of fuddy duddying around like a wet fart will.
Whats the point if in 24 months the government can stand up high and say we beat the virus, sorry 12 million of you lost their jobs, and the fact that we lost most of our airlines, i know 30% lost your houses because we didnt pay you because you were self employed or on a furlo scheme, oh by the way, we have put the tax up to 75% to cover the costs, so there is going to be a lot more hardship.

If people have to use the excuse Critical workers (and i enclude myself in that bracket) then they should have to under take covid tests every couple of days at work,
If your company is classed as Critical, then the government should implement covid testing so that critical workers dont become the carriers. (Asymptomatic carriers)
Its not hard to take someones temperature, and if its above a threshhold instant test. we've been told they can do it in 15 mins now.

The other way is just to sat Fk it, and let evreyone get back to normal, yes covid will spread like wild fire and there will be a lot of people dying with underlying health issues, but once herd immunity kicks in, and everyone has had it, game over.

Dont forget the AVERAGE number of uk deaths per week is 10000 (1973 to 2019), pre covid.
So those extra 50 deaths this week or 400 next week, are they part of the 10000 or on top of it?

Im not a betting man, but i predict at the end of this year the total number of deaths for this year will not be the highest ever which was 1993.

1993 576K deaths highest for one year
2011 483K deaths lowest for one year

Jan 2001 20500 deaths in a week
Dec 2013 6500 deaths in a week

[/RANT]

Mac.

and breath..............................................

[NoelC]

Roberto
W/R to False Positives; Spiegelhalter was clattering on about this on Radio 4's More or Less this week, could I take up the argument about false positives-

Using the test sensitivity (ability to identify positives) of approximately 99.2% to estimate the false positive rate (which has been suggested to be 0.8%) is incorrect.  The test specificity (ability of the test to identify negatives) would give the false positive rate and is much more accurate for the current PCR test (but is not published). When observed empirically it was <0.08%.  See ONS report on Covid infection survey:
https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/bulletins/coronaviruscovid19infectionsurveypilot/englandwalesandnorthernireland25september2020
Para 5. Test sensitivity and specificity

Doesn't this study demonstrate a strategy for testing for statistical significance? 
Granted the government's figures do not, and the ONS figures tend to be out of date, but from the above article it would appear the ONS figures on infection rate are inflated by less than 0.1% by false positives.  Please jump in here if I'm wrong.

Interestingly in the ONS results for their survey London was returning 2.5% positive on the 19th of September, which is in the same ballpark as the App.
ONS Survey results:
https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/bulletins/coronaviruscovid19infectionsurveypilot/englandwalesandnorthernireland25september2020#test-sensitivity-and-specificity

W/R to the App, here's an article giving some insight to their approach:-
https://www.nature.com/articles/s41591-020-0916-2
As you say, it's not quite the same level of confidence you get from a PCR test, but as a crowd source data project it's pretty good.  I think ZOE work a lot in countries with far less testing, so the approach could help save lives.

I think I disagree that there has been no sense of statistical significance in the measures applied; I suspect SAGE has been entirely driven by the stats, but between them and the government (who are driven by economic considerations) they have been unable to convey their strategy clearly or consistently or in a timely manner. 
Have the government ignored the current stats in their recent minor amendments to the regulations, or are they fully aware of the cost of the measures they must take and are delaying implementation until the public recognise, and support the need for more extreme measures?
A second lock down could put us in a position of hyper inflation as borrowing spirals, failure to address the pandemic will lead to 400,000 dead people - not a choice to be taken lightly.

Noel

[NoelC]

Mac
Are you back in the country (I'm surprised they let you back in)?
Magnificent Rant!
A lot of good points there - be careful what you say though, in the US stuff like that could get you elected. 

Noel

[Carole]

Totally agree Mac, needed saying.

Carole