Urban75 Home About Offline BrixtonBuzz Contact

Coronavirus in the UK - news, lockdown and discussion

maomao said:
The app is Bluetooth based isn't it? So it works by recording who you've been physically close to. Relies on everyone having Bluetooth turned on all the time.
Click to expand...

Yes, but it will use Bluetooth LE which uses lots less power than standard bluetooth. It's not very secure though.
 
littlebabyjesus said:
It's detailed information about who you associate with, when and for how long. You don't see a problem with the government gathering that information?
Click to expand...

I don’t like the idea of it at all, but when I try to examine that reaction for logical reasons the only one I can come up with is they’d lose the data and it would become publically available, which could be bad if I’d been up to something clandestine I didn’t want somebody to know about.

If think about whetherI’m worried that the govt will get all stasi-like, then I rapidly decide that if that’s the route they’re going down the existence or otherwise of this data/app isn’t going to make it more likely (and I don’t think they are going to go stasi-like).

But maybe I’m suffering from a failure of imagination?
 
Neither of my (both retired) parents have smart phones. My g/f's dad (early 60's) doesn't have a smart phone.
 
maomao said:
The app is Bluetooth based isn't it? So it works by recording who you've been physically close to. Relies on everyone having Bluetooth turned on all the time.
Click to expand...
As far as I've read, if your phone spends 15 minutes within 2 meters of the phone of someone who chooses to tell the app they have symptoms, you will get a message advising you to self-isolate. If that person gets the all clear, you get another message saying you're good.

It doesn't seem completely watertight at tracking transmissions.
 
prunus said:
I don’t like the idea of it at all, but when I try to examine that reaction for logical reasons the only one I can come up with is they’d lose the data and it would become publically available, which could be bad if I’d been up to something clandestine I didn’t want somebody to know about.

If think about whetherI’m worried that the govt will get all stasi-like, then I rapidly decide that if that’s the route they’re going down the existence or otherwise of this data/app isn’t going to make it more likely (and I don’t think they are going to go stasi-like).

But maybe I’m suffering from a failure of imagination?
Click to expand...

Just to reiterate, this doesn't record location data or personal details of who you have been in contact with. It also, obviously, doesn't know about what interaction you have had with the positive person. You could have been stood next to them on the tube or you could have been scoring off them. It won't know.

Obviously, GCHQ and the NSA already have the location data and who you score off so you are already compromised. :thumbs:
 
Don't forget that people with smartphones are humans and also talk to each other!

So if someone gets a notification through the app to self isolate then they are highly likely to tell other people they've been in contact with, whether those other people are using the app or not, so hopefully they'll be people self isolating through being told by secondary normal conversations. In 'meat space' as the geeks say I think.
 
Anyone make any sense of this? I can't.

www.independent.co.uk

Antibody tests would put up to quarter of those told they were immune at risk of infection, government advisers warn

Scientists have warned the government that even highly accurate antibody tests for coronavirus could leave more than a quarter of people who were told they were immune at risk of infection.
www.independent.co.uk www.independent.co.uk
Scientists have warned the government that even highly accurate antibody tests for coronavirus could leave more than a quarter of people who were told they were immune at risk of infection.

New documents published today reveal the problem exists even for antibody tests that are 98 per cent accurate. Even if the tests are 99 per cent accurate almost 10 per cent of people who were told they were immune would be put at risk.

The conclusions, considered by the government’s Scientific Advisory Group for Emergencies, or SAGE, suggest the much anticipated development of home antibody kits may not be the route to exiting the coronavirus lockdown many had hoped.

The problem comes from the number of patients who may receive a false result even though the tests are so accurate.

The SAGE paper says the performance of the tests depend on how many people who actually have antibodies test positive, this is called the sensitivity, and how many people without antibodies test negative, which is described as the specificity.

Alongside this is the spread of the disease within society and the number of people who actually have antibodies – the prevalence.

The SAGE document said for a test which is more than 98 per cent accurate and where 5 per cent of the population are thought to have antibodies, a test would lead to 68 people being told they had had the virus and were immune.

Of these, 19 people, or 28 per cent would not have antibodies and would be put at risk of infection if allowed to go back to work or mix in public with those carrying the virus. At least one person out of 1,000 would be wrongly told they had not had the virus and would be forced to stay in lockdown.
Click to expand...

I particularly don't understand the text from "68 people being told they had the virus ..." :confused:
 
LynnDoyleCooper said:
No. AFAIK it only requires a certain % of the population to use it for it to be effective anyway.
Click to expand...
Yeah but once you take out.
people who don't have smart phones, children, elderly, technophobes, other reasons;
Morons who seem not to have seen the news for the past 8 weeks and won't even be aware of this;
People with overriding suspicion of it's potential usage down the line;
People who CBA;

That could be quite a gap. I thought I heard a similar app in Singapore had a take up of less than 20%.
 
Amongst other things, absolute transparency regarding the data processing, and storage is necessary to sell this thing to asm many as possible.
 
two sheds said:
Anyone make any sense of this? I can't.

www.independent.co.uk

Antibody tests would put up to quarter of those told they were immune at risk of infection, government advisers warn

Scientists have warned the government that even highly accurate antibody tests for coronavirus could leave more than a quarter of people who were told they were immune at risk of infection.
www.independent.co.uk www.independent.co.uk


I particularly don't understand the text from "68 people being told they had the virus ..." :confused:
Click to expand...

I think it's missing that that is out of 1000 people. If 5% have had the virus, if the test was 100% accurate then you'd have 50 people testing positive and 950 testing negative. IN fact the test will give 68 people a positive result. So 18 of those people will have been given the wrong information.
 
Monkeygrinder's Organ said:
I think it's missing that that is out of 1000 people. If 5% have had the virus, if the test was 100% accurate then you'd have 50 people testing positive and 950 testing negative. IN fact the test will give 68 people a positive result. So 18 of those people will have been given the wrong information.
Click to expand...

Yes that's good, ta. That gives 1.8% of people the wrong information. I still don't understand the bit earlier though:

"Even if the tests are 99 per cent accurate almost 10 per cent of people who were told they were immune would be put at risk."
 
two sheds said:
Yes that's good, ta. That gives 1.8% of people the wrong information. I still don't understand the bit earlier though:

"Even if the tests are 99 per cent accurate almost 10 per cent of people who were told they were immune would be put at risk."
Click to expand...

It comes about because one section of the population (people who haven't been infected) is, in the assumed scenario, so much higher than the other (people who have been infected). If it's only returning 2% inaccurate returns wrt positive tests that doesn't seem too bad, right? But as it's 2% of a the much larger group the number of innaccurate returns is a much more significant number of the total number of positive returns than 2%.
 
two sheds said:
Anyone make any sense of this? I can't.

www.independent.co.uk

Antibody tests would put up to quarter of those told they were immune at risk of infection, government advisers warn

Scientists have warned the government that even highly accurate antibody tests for coronavirus could leave more than a quarter of people who were told they were immune at risk of infection.
www.independent.co.uk www.independent.co.uk


I particularly don't understand the text from "68 people being told they had the virus ..." :confused:
Click to expand...

It's a really badly written article. Looks like it was published quickly and I expect it'll get edited soon to make more sense!
 
two sheds said:
Anyone make any sense of this? I can't.

www.independent.co.uk

Antibody tests would put up to quarter of those told they were immune at risk of infection, government advisers warn

Scientists have warned the government that even highly accurate antibody tests for coronavirus could leave more than a quarter of people who were told they were immune at risk of infection.
www.independent.co.uk www.independent.co.uk


I particularly don't understand the text from "68 people being told they had the virus ..." :confused:
Click to expand...
No, it is confusing.

The figures at the end only make sense if both specificity and sensitivity are under 100%.

However, the claim from the Swiss company that's made the test is that it has 'at least' 99.8 % specificity and 100% sensitivity. If they're right, then everyone with the antibodies will be correctly identified, while 2 out of 1000 without them will be incorrectly told that they have them.

So addressing that claim instead, if 5 % of the population has the antibodies, then you'll expect 95 negative tests per 100. Out of those 95, 2/1000 will get the wrong result and be told they have the antibodies when they don't. So on average you'll get roughly 94.8 negative tests per 100 instead of 95. Everybody told they're immune will be immune, and only a relative handful will be told they are immune when they are not. Put another way, 50/52 of everyone testing positive would actually be positive. Seems pretty good to me, if the claim is correct.

Roche’s COVID-19 antibody test receives FDA Emergency Use Authorization and is available in markets accepting the CE mark

You might want to check my workings, but I think that's right. :D It's actually a lot easier to work out if one of sensitivity or specificity is 100%.
 
Last edited:
xenon said:
That could be quite a gap. I thought I heard a similar app in Singapore had a take up of less than 20%.
Click to expand...

Yes, that was the case but it was a few weeks ago now and before (or at least around the same time) as Singapore was going into a more widespread lockdown. The locals may feel a bit different about it now if it is seen as tied into a release from lockdown strategy.

xenon said:
Amongst other things, absolute transparency regarding the data processing, and storage is necessary to sell this thing to asm many as possible.
Click to expand...

Absolutely. I have some minor concerns about the practicality of it but by far and away my biggest concern is data security both from government misuse but more from hackers. I just don't have much faith in government IT projects and certainly one that has been rushed like this. Hopefully it will be a decent and effective tool but I fear the take up might be so low as to render it worthless.
 
LynnDoyleCooper said:
www.wired.co.uk

Everything you need to know about the new NHS contact tracing app

The app behind the NHS Test and Trace project failed. A second NHS Covid-19 app was released on September 24. Here's how it works
www.wired.co.uk www.wired.co.uk

Yeah, my mistake. Here's a good article about it. Apparently stores no personal data.
Click to expand...

Here's another, apparently it wont work: UK finds itself almost alone with centralized virus contact-tracing app that probably won't work well, asks for your location, may be illegal

But there is a problem with the NHS's approach: it probably won't that well work on your phone, and probably won't be terribly accurate at measuring the spread of the virus.

That's because the proposed system will only work in the way the UK government claims it will if everyone does what it says: a classic failing of the Whitehall mindset that stretches back to the World War One trenches and further back still to the days of Great Houses and Men Who Knew Better.

Despite what the NCSC has continued to imply, the app will not, as it stands, work all the time on iOS nor Android since version 8. The operating systems won't allow the tracing application to broadcast its ID via Bluetooth to surrounding devices when it's running in the background and not in active use. Apple's iOS forbids it, and newer Google Android versions limit it to a few minutes after the app falls into the background.

That means that unless people have the NHS app running in the foreground and their phones awake most of the time, the fundamental principle underpinning the entire system – that phones detect each other – won’t work.

It will work if people open the app and leave it open and the phone unlocked. But if you close it and forget to reopen it, or the phone falls asleep, the app will not broadcast its ID and no other phones around you will register that you've been close by. There is even a handy video of someone in Australia showing this (Australia has gone for a similar system with its COVIDSafe app.)

We cannot state it plainer: on iPhones, apps cannot send out their IDs via Bluetooth when the software is in the background, and on newer Android builds, IDs cannot be transmitted after a few minutes in the background. And Apple and Google have refused to allow the tracing app to send out IDs in the background.


The NHS has insisted its engineers have worked around this problem "sufficiently well" by waking the app after it detects itself running on a nearby phone emitting an ID: the software is blocked from sending out its ID when in the background but it can passively listen for IDs of apps still allowed to broadcast. However, this assumes there are a sufficient number of phones running the tracing app nearby still broadcasting to keep enough people's apps awake: there needs to be a critical mass of users while we're all supposed to be socially distancing. If two or more people pass each other and their apps have stopped broadcasting, the software will never know they came in contact.


And it could be a battery hog, which may make people leave the app off, preventing the app on other phones from waking up.

Little choice

What Levy doesn’t say is that he – and NCSC and the UK government – are assuming that when people are moving around, and so are close to one another, they are likely to be on their phones or have recently opened the app. It’s an assumption they have no choice to make because otherwise they don’t get the data. By contrast, the Apple-Google solution that Germany, Austria, Switzerland and Ireland, among others, are following will allow the IDs of phones to be recorded in the background all the time, due to being built into the operating system, so it will be more accurate and kinder to battery life.
Click to expand...
 
two sheds said:
Ta for that. Even that figure of 28% (19/68) error seems artificial though, we're still talking about 1000 people being tested so I don't see why the important figure isn't 19/1000.
Click to expand...
It means that nearly a third of those told they're immune are not immune. So, knowing that, if you're told you're immune, you have to assume a 2:1 chance that you're actually not. It's told you a lot in terms of changing your knowledge cos it's changed your chances of being immune from 1/20 to 7/10, but you'd need another test to improve that knowledge more.

At the very least, with a test of that level of accuracy, all positive tests would need to be followed up by a second test, which would bring that figure right down and give something more like 99% confidence in the result.
 
Last edited:
Teaboy said:
I just don't have much faith in government IT projects and certainly one that has been rushed like this. Hopefully it will be a decent and effective tool but I fear the take up might be so low as to render it worthless.
Click to expand...

Yes, as much as I dislike Apple and Google, presumably they have some fairly good developers working for them. Almost certainly a cut above our government's. In a situation like this it seems bonkers to turn down the offer. I think Germany were going down our route then reversed and going with the actual market leaders in their field.
 
littlebabyjesus said:
It means that nearly a third of those told they're immune are not immune. So, knowing that, if you're told you're immune, you have to assume a 2:1 chance that you're actually not. It's told you a lot in terms of changing your knowledge cos it's changed your chances of being immune from 1/20 to 7/10, but you'd need another test to improve that knowledge more.

At the very least, with a test of that level of accuracy, all positive tests would need to be followed up by a second test, which would bring that figure right down and give something more like 99% confidence in the result.
Click to expand...

Ah ok ta (I think, although it's still a relatively small number given 1000 tested).

And yes 99% confidence unless the same people give false positives/negatives for some reason.
 
two sheds said:
Ah ok ta (I think, although it's still a relatively small number given 1000 tested).

And yes 99% confidence unless the same people give false positives/negatives for some reason.
Click to expand...
Point is to do with the confidence you yourself can have in the result. There are tests for very rare diseases that, when they come back positive, still actually mean you probably still don't have it. In this instance, a 7/10 chance that you're immune isn't a great number if you're basing life choices around it.

Anyhow, if Roche's claims are correct, it's all rather moot. The test is a lot better than that.
 
Last edited:
Looby said:
From what I’ve heard it’s the opposite so even if your employer requests the test for you, the results will only be given to you.
Sounds like either she or the person she spoke to have got the wrong end of the stick.

My friend tried to book for our local centre and they had no available appointments. We’re assuming they’ve run out rather than being at capacity as it hasn’t been that busy.
Click to expand...


need to book tests for carer, no car, how do I go about doing it?

I am also employing a new carer, however I didn't know she wasn't here in the city, but at her folks in Norwich, she seems to think she has to wait for restrictions to ease, but she is an essential worker, also her main home as a post grad is here as well, can't she just come back when she wants to
 
treelover said:
need to book tests for carer, no car, how do I go about doing it?

I am also employing a new carer, however I didn't know she wasn't here in the city, but at her folks in Norwich, she seems to think she has to wait for restrictions to ease, but she is an essential worker, also her main home as a post grad is here as well, can't she just come back when she wants to
Click to expand...

They have to book their test online and they can post it out afaik. Look at the gov.uk website, it explains it all.

www.gov.uk

Testing for COVID-19

Find out who can get free NHS tests, how to get tested, and what your test result means.
www.gov.uk www.gov.uk
 
You must log in or register to reply here.

Similar threads

elbows
UK Inquiry Module 3: Impact on UK Healthcare Systems
2
Replies
37
Views
618
InArduisFouette
I
elbows
UK Inquiry Module 1: Resilience and preparedness
2 3 4 5 6
Replies
165
Views
5K
elbows
elbows
elbows
UK Inquiry Module 2: Decision-making & political governance
2 3 4 5 6
Replies
173
Views
4K
elbows
elbows
frogwoman
Shanghai lockdown
2
Replies
55
Views
2K
skyscraper101
skyscraper101
editor
Professional football and coronavirus
Replies
6
Views
527
elbows
elbows
Back
Top Bottom