Chat about Novel Coronavirus - 2019-nCoV - COVID-19

There seems to be a concentration of hospitalised C-19 cases around the Cheltenham races as of April 3rd.

https://www.gloucestershirelive.co.uk/news/cheltenham-news/leaked-map-shows-postcode-next-4071259

The Cotswolds (where most visitors stay) is badly effected relative to the average for Gloucestershire.

I think all the antibody tests looked at so far are only 60% ish accurate. Risk of false positive and false negatives. So I wouldn't bother using these at the moment, especially at the price private clinics are probably charging.

Yes, just checked, PHE say that use of these is not allowed though didn't specify reason here-
https://www.gov.uk/government/publications/covid-19-rapid-tests-for-use-in-community-pharmacies-or-at-home

Ah, there's more about this in an FT article from yesterday:

"The National Covid Testing Scientific Advisory Panel in Oxford this week released the results of its evaluation of nine commercial antibody tests being considered by the UK government. None came close to meeting the required specifications for testing individuals for their immune response to Covid-19."

https://www.ft.com/content/a93e6b28-3778-4089-8d80-4e6775996aea

https://www.manchestereveningnews.co.uk/news/uk-news/tired-mum-named-shamed-facebook-18143150

Bit of a rag link to a thread on mumsnet originally. Mum 'shamed' on facebook for not clapping the nhs and 'letting the street down'. WTF is wrong with people?

It's worth noting that due to how Bayesian probability works, having a test that is 60% accurate does not mean that, should you test positive, you have a 60% chance of having had Covid19.

Unless you know the true prevalence of infection in the population you cannot state with any accuracy what your chance of having had it are.

The maths is counter intuitive and I certainly cannot write out an explanation here. Happy to Google links for people...

It's worth noting that due to how Bayesian probability works, having a test that is 60% accurate does not mean that, should you test positive, you have a 60% chance of having had Covid19.

In essence, its not that you have 60% chance of having it, its 60% its right?

Not so simple:

An 18th century concept in statistics, known as Bayes’s theorem, can help us. This theorem tells us how to calculate the probability of an event given that another event has happened. For example, say people in a particular colony are being tested, and 20% of them actually have the disease. Next, say the sensitivity (probability of a positive result given the disease is present) of the test being used is 80% and its specificity (probability of a negative test result given the disease is not present) is 90%. A little bit of math yields the following probabilities:

True positive = 0.16
False positive = 0.08
False negative = 0.04
True negative = 0.72

According to Bayes’s theorem, the probability that the disease is present given a negative test result can be obtained by multiplying the probability of disease in the locality (0.20) and the probability of a negative result given the disease is present (0.20), then dividing this by the probability of a negative test result (0.76). This value comes out to be 5.26%. That is, a little more than 1 in 20 people who test negative may actually have the disease. Similarly, the probability of ‘no disease’ given a positive test result is 33.3%.

If the disease’s prevalence in the colony rises to 50%, these two figures become 18.2% and 11.1%, respectively. If the prevalence increases to 80%, these figures become 47.1% and 3%, respectively.

From here: https://science.thewire.in/the-sciences/covid-19-rt-pcr-serological-tests-false-positive-false-negative-rate-bayes-theorem/

Please do.
I read something earlier in all of this. It was great. But if someone else could explain it simply, to try and help out. That would be grand.

Pretty sure Ben Goldacre talks about this in one of his books; I'd imagine he's discussed it on his blog somewhere. I think he used the example of a 99% specific/sensitive HIV test. Because the underlying rate of HIV in the population is so low, even if you get a positive result you're still more likely to have had a false positive than to actually have the disease. Statistics are counterintuitive and weird.

(Edit: I find visualisations help with this more than reading the explanations, i.e. 1000 squares coloured according to how many in the population actually have the disease, and how many the various tests will show up as positive/negative)

Letting Cheltenham go ahead seemed like a bad idea at the time let alone now; 250,000 pissed up people cramming into the paddock for each race was never going to be helpful. Distinctly remember the ITV rhetoric being along the lines of the meet being a final hurrah before those bloody politicians end our fun - strange

How simply?

On the most basic level, it's like this:

• Say you get a positive result. This could be due to you being positive and the test working correctly (true positive), or you being negative and the test working incorrectly (false positive).

• Say you get a negative result. This could be due to you being negative and the test working correctly (true negative), or you being positive and the test working incorrectly (false negative).

In terms of being one sample out of an entire population, you have a certain probability of being positive or negative (a priori probability). The test then also has a certain probability each of producing the four different scenarios outlined above (true / false positive, true / false negative).

So when you test people and get positive / negative numbers, what you get is the combination of these different probabilities, and you don't just know which bit is due to what.

Cheltenham was a nightmare, I had to travel from home in Teignmouth to our Bristol office on Wednesday and Friday that week , normally a pretty empty train but that week it was packed from Taunton to Bristol, Bristol train station was a nightmare with everyone trying to get on the train. The train's home especially the Friday were disgusting , tables full of empty beer cans and wine bottles, loads more all over the floor. Pretty sure it was doing that journey on the Friday I picked up the stupid cough I had for a few days afterwards.

Reminds me of a great line at the start of this video: https://www.youtube.com/watch?v=7GgLSnQ48os

"What are the chances of finding me here?"
"It's the first time I've found you here, so I don't know"

It’s more like 70k but yeah, agreed. 250k is unique visitors to the course that week.

There was probably more transmission in the rammed pubs and betting shops. But that’s every pub within a 20 mile radius for four days of solid drugs & alcoholic before dispersing across the UK and Ireland. Madness 📈

.
Wrong thing edited

On the sky lanterns chat, anybody using them should pay high fines.
People selling them should go to prison.
The fire that killed 30 animals on new years in the Krefeld Zoo was caused by those.

I live near Cheltenham and looking at that map the two areas with the highest number of cases also appear to be those with the highest population density. The racing shouldn't have gone ahead though and one of my work colleagues and his wife who live around there have both had Covid-19 (although that might be coincindental).

And because of the NYTimes article warning that covid pneumonia can go undetected for a while this landed here today.
Probably insane, but I did not quit cigarettes 3 months ago after 20 years of smoking so some virus fucks my lungs even more.

1 Attachment

Anyone worked out how we get the light in the lungs yet?

I presumed it was some sort of garbled message to his Evangelical followers,
implying 'Belief in The Lord' would be enough.

Cheers for the offer but I think she's gone to Mellis' having tracked down some obscenely expensive artisanal flour and going down that route. "Essential travel" I'm not so sure but they're open so...

I'm glad she's finding entertainment in such simple pleasures in these trying times so I feel it would be cruel of me to take that eventual satisfaction of finally hitting the right recipe away from her too...

There is experimental treatment for dementia involving light therapy that someone might be receiving. Intracranial and intranasal infrared lights.

And, I guess, we are all familiar with the benefits of daylight correcting light sources during northern winters to ward off SAD.
I thought Trumps claims were based upon a dismally basic paper from the DHS, (Dept. of Homeland Security)?