Forum menu
I’m not particularly interested in defending Prof Gupta. In fact I’m trying to avoid taking sides at all and listening to as many credible sources as I can. But we know there is a big variability in how people react to infection. Ranging from barely any symptoms through to death. As far as I know, we don’t know what causes that variability (genetic?, previous exposure to something else?). Given that, is it not possible that there is a fraction of the population that are effectively immune despite not showing the specific antibodies that are looked for in these tests? I don’t know, it seems unlikely, but not impossible.
If she is right and these models are just simple compartment models (with the population divided into well mixed compartments, with rate constants defining movement between the compartments and the whole thing described by a load of first order differential equations) then I can we’ll believe that very different solutions give equally good fits to the data. I have some experience of these models in a different context and once you get beyond a couple of compartments you do need very good data or things can get pretty unstable. But she could be wrong on that too, for all I know.
By the way, I’ve not seen anything from the Swedes saying they were wrong, so it would be good to see some links to that. Last I heard they still thought they were on the right track but admitted they hadn’t done enough to project people in care homes.
Given that, is it not possible that there is a fraction of the population that are effectively immune despite not showing the specific antibodies that are looked for in these tests? I don’t know, it seems unlikely, but not impossible.
The seroprevalence antibody tests are consistent with ONS swabbing PCR of 0.1-0.3% having active infection in a two-week sampling period.
Sweden - Tom Britten is doing the modelling
“It means either the calculations made by the agency and myself are quite wrong, which is possible, but if that’s the case it’s surprising they are so wrong,” he told the newspaper Dagens Nyheter. “Or more people have been infected than developed antibodies.”
https://www.dn.se/nyheter/sverige/tester-73-procent-bar-pa-antikroppar-i-stockholm/
Whilst there is some time delay to develop IgG antibodies, Robust models are based on orders of magnitude, not absolute values, so 0.1, 0.3, 1, 3, 10, 30, 100% prevalence. ALL studies suggest the needle is at the 3-10% not the 30-100% that Oxford and Sweden would wish. That's a BIG difference that is not a result of "test inaccuracy".
In exponential land, it's easy to be "quite wrong" 😉
Thanks. Doesn't sound as though he is accepting that they are wrong, just accepting that it is a possibility. But that's a start 🙂
So, am I right in thinking that nobody is seriously suggesting that there could be a fraction of the population who are immune to this disease despite not having contracted it? I thought one of the Swedes was hinting at this (or at least that there may be other forms of immunity that don't show up on these tests) but I can't remember where I saw that.
It is not impossible that past coronavirus infection MAY confer some protection (indeed past SARS-COV-1 infection probably does). Equally, having low levels of antibodies that bind to virus may instead make infections worse (antibodies help shuttle virus into cells).
There are four other circulating human coronaviruses. I doubt this fifth one is particularly different with regards to immunity. One needs some perspective, not just data acquisition bias (looking too hard at THIS virus at the expense of human experience with all other respiratory viruses).
Yes, that’s a fair point. There may be something particularly unusual about this virus, but I wouldn’t bet on it.
There may be something particularly unusual about this virus
There is with regards to innate immune response as I linked yesterday. Really nice paper. There is a lot of "firing from the hip", I've probably been guilty of it at some points, but I always try and ask "How wrong would I have to be to make a wrong decision?". @thecaptain has done the same. And the answer for me is information could not be THAT wrong to interpret the Oxford position.
roverpig, I'm not at all saying that London has reached herd immunity (for normal behaviour), far from it, but if the behavioural changes including lockdown are sufficient to reduce R to about 0.8ish in most of the country, then an immune fraction of 17% of the most active people would potentially cause a significantly more rapid drop in cases in London compared to outside London.
(There is actually some ambiguity in what people mean when they talk about R. From the modelling perspective, it makes most sense IMO to think of it as meaning the number of new cases that would be created per current case in a fully naive population. But some of the estimates are based on the number of new cases that are actually created per current case. I think when SAGE is talking about R being different in London, they mean the latter, not the former.)
Some evidence on masks:
https://medicalxpress.com/news/2020-05-hamster-masks-coronavirus-scientists.html
It's convinced me, I'm gonna start wearing one away from home when other people are about.
Well I guess that's one way to meet a target.
I think when SAGE is talking about R being different in London, they mean the latter, not the former.)
One of the reasons that politicians blathering on about "the R" has annoyed me so much. R₀ and Rₑ are two different things.
R₀ refers to the reproduction in a susceptible population with no immunity - this is a property of the virus and can't be changed. Rₑ is the actual reproduction seen in a particular context and can obviously be changed by the prevalence of immunity and the mitigations taken.
I’m gonna start wearing one away from home when other people are about.
We all will be soon. On my ride yesterday I saw more riders wearing masks than helmets.
The science on their use by us untrained idiots is still wobbly, but as a tool for reassuring people when the time to start mingling again is here, they’ll prove invaluable.
Mrs OTS is making them for us. I'm on version 2 using neck tube material to give some stretch. I do look like Hannibal Lecter in it though.
The science on their use by us untrained idiots is still wobbly,
Some of that seems to be based on the medical profession asserting their right alone to 'do medicine'. I saw one doctor telling people on Facebook not to wear gloves unless they are 'medically trained' for example.
@thecaptain I did carefully avoid using the phrase herd immunity, but thanks for the clarification. As you say adding a bit of immunity to the other measures could make the difference, but I don’t know enough about the models to know if a 17% immunity is enough to explain the difference in the curves between London and elsewhere.
Fully agree about R. I assume most of the numbers bandied about at the moment are Re rather than R0
I saw one doctor telling people on Facebook not to wear gloves unless they are ‘medically trained’ for example.
That’s because, if not trained, the public using gloves is more likely to increase rather than increase spread.
All that really matters is whether Re > 1, = 1 or < 1, and the confidence of that statement. The exact value is immaterial. Modelling has over-sold the utility in trying to explain the assumptions of the model.
Obviously as R comes down, the herd immunity threshold comes down too. But since that is based on behaviors, we know how well reliance on behaviors lasts (third week in January, normally) 🙂 .
Rather than say "don't use gloves", a little instruction is better placed. I learned something.
Indeed. Training. They become an essential tool then. Or two essential tools, if you count like the government.
That’s because, if not trained, the public using gloves is more likely to increase rather than increase spread.
How does that work then?
Multi-centre longditudinal study of the use of hydroxycloroquine etc in Covid. Not exactly a home run for Trump's favourite medicine...to put it mildly.
https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)31180-6/fulltext
How does that work then?
The link that Tired provided is a very good starting point if you really want to understand…
How does that work then?
Because member of general public takes off gloves incorrectly, assumes that their hands are 'clean' and then promplty goes around touching stuff with infected hands. simples
That link does not claim what you said about gloves 'increasing the risk of spread' in (or should that be on?) untrained hands.
The argument is not about whether medically trained people can use gloves 'better' but whether Joe average is putting himself and/or others more at risk by wearing gloves.
I can’t understand it for you.
Is there some way of preventing a Second Wave of questions, roughly 6 weeks after the First Wave? 🙂
Because member of general public takes off gloves incorrectly, assumes that their hands are ‘clean’ and then promplty goes around touching stuff with infected hands. simples
You are missing another premise there, perhaps something like non-trained people will wear gloves in lieu of appropriate hand washing? If that is true you may be correct but otoh it could be a prejudiced and unfounded assertion.
Is there some way of preventing a Second Wave of questions, roughly 6 weeks after the First Wave?
I keep resisting posting “read this thread, it’s been covered already”, because that would become tiresome very fast.
It isn't a question of understanding some inferred meaning but statements of facts. If you can link the quote regarding the increased risk of transmission it would be welcome. I can see how you have jumped to that conclusion but it's not actually what it says.
Read this thread, it’s been covered already.
The link first and foremost cautions against public use because the gloves are a scare resource which should be reserved for the health services. In passing it is noted that glove use by the public may increase the risk of transmission based on an assumption about missed opportunities for hygiene. That latter hypothesis is total speculation and pulled out of nowhere apparently. In academia, the use of 'may' indicates the least secure kind of knowledge.
Going back to Sweden:
Perhaps another explanation is that Sweden has a very different way of perceiving the current crisis. Instead of seeing it as a national emergency or a fight against an “invisible enemy”, there seems to be a tendency to regard coronavirus just as a serious public health problem. It is viewed as something that requires the careful observance of rules set out by health experts, rather than an existential problem that calls for the state to suspend civil liberties for the sake of national security.
From: https://www.theguardian.com/commentisfree/2020/may/22/sweden-u-turn-controversial-covid-19-strategy
The evidence for Vitamin D:
I suspect most STWers are outside all the time and don't need to worry too much about Vitamin D. Certainly I can't imagine I'm lacking Vit D in summer. It's still interesting, there's certainly correlation, maybe there's causality as well.
It is viewed as something that requires the careful observance of rules set out by health experts
I'm starting to think that's where the rest of Europe is going to end up. We're just going to be really careful and follow some rules. (My predictions are always wrong.)
Obviously as R comes down, the herd immunity threshold comes down too. But since that is based on behaviors, we know how well reliance on behaviors lasts (third week in January, normally) 🙂 .
You wouldn't have to do much to make a big difference though.
IR cameras on public transport and airports.
Make sure people wear a mask/stay at home if they're ill.
Perhaps some other easy, cheap measures.
But my prediction is that our government is a bit too dim to do any of these.
antiviral wipe (sorry if it's already done)
On the masks and gloves issue. I agree that there should be some comprehensive advice on how they are best used. Certainly not to do what I saw a guy do in morrisons on Tuesday. He came out of the shop with gloves and mask on. Opened his car and put all of his shopping in the back, pulled down the mask to his throat, got on the front and drove off. All with the gloves still on.
In Italy, where masks are worn as a matter of course in public places, they have tried to introduce a scheme to ensure that the can be bought at no more than 50c each (roughly 45p). Not surprisingly this hasn’t gone well, manufacturers and retailers are not complying and people are reusing disposable masks
I can’t understand it for you.
Well that’s quite defeatist. And therefore not British old chap.
On the masks and gloves issue. I agree that there should be some comprehensive advice on how they are best used. Certainly not to do what I saw a guy do in morrisons on Tuesday. He came out of the shop with gloves and mask on. Opened his car and put all of his shopping in the back, pulled down the mask to his throat, got on the front and drove off. All with the gloves still on.
But is that more likely to lead to transmission than if he didn't wear the gloves.
There is no question that certain practices optimise the benefits of gloves but are gloves by default riskier?
Nice to see Dominic Cummings has been abiding by lockdown- guardian reports the police spoke to him as he left London to visit his folks in Durham while having coved symptoms
I think I'll enjoy seeing how Cummings wheedles out of this one 🙂
Clearly rules is for proles not incredibly important people like Dom.
How did cummings and his wife get from London to Durham? If by car, were they in a fit state to drive? Did they stop en route?
If not by car, how?
Being denied that police talked to him; did they talk to his wife?
Reported by Guardian, Mirror and Laura Kuenssberg that trip was for child care; LK saying that downing st intend to 'tough this out'.
Yet again, do as I (we) say not as I (we) do.
Tomorrow's headlines now being hastily re-written.
There is no question that certain practices optimise the benefits of gloves but are gloves by default riskier?
No, but it's similar to the mask question. False sense of security, incorrect use etc. etc.
YMMV, basically.
I really hope that **** ends up Mussolini'd.
I think I’ll enjoy seeing how Cummings wheedles out of this one 🙂
Could this be their chance to drop the evil overlord?
onewheelgood, I'm going to pick a nit about your assertion that R0 is a property of the virus that can't be changed. It most definitely depends on the behaviour of the population and can vary significantly around the world. As SAGE say "Both the reproduction number and doubling time are dependent on the characteristics of the population so may be different in the UK, and may be different in different groupings within the UK." (the point of comparison here being Wuhan, before it reached us).
Though I suspect the low numbers people calculated for China early in the outbreak were probably just wrong. The doubling time in the deaths didn't seem very different to what we got here. People tried all sorts of fancy stuff with detailed models rather than plotting the numbers and putting a line through them.
Reported by Guardian, Mirror and Laura Kuenssberg that trip was for child care; LK saying that downing st intend to ‘tough this out’.
'A source close to Mr Cummings defended his actions, saying he made the trip because his parents could help with child care while he and his wife were both ill with symptoms of coronavirus.'
So not essential travel, they didn't isolate as a family, and put parents (in their 70s at a guess) at risk?