Herd immunity and risk characteristics of 'healthy' people

[JonathanH]

Congratulations on your ability to see into the future!

Clearly I can't see into the future any more than you can. I am taking a view, just like you are. Our views are different and based on what we both are hearing from experts.

For all your negativity, we simply don't know what's going to happen next. Most experts agree that we're nearing the peak, indeed we may already have reached it given that the reported daily figures often lag behind the actual stats. From the government's daily briefings, new cases of the virus seem to be stabilising at the 4,000 a day mark.

Once that peak is reached, it's perfectly rational to assume that the number of new cases will fall, as the less people catch the virus, the less people there are to spread it. There is also the theory that the warmer weather will reduce the number of new cases, just as most flu viruses tend to die out in the summer.

It is not a flu virus. There is a hope that it diminishes in the warmer months and clearly people should be less susceptible to other illnesses but it is in no way certain that transmission of Covid-19 will drop. Indeed, there is probably no chance that it would be gone by the autumn either.

Yes, we're by no means out of the woods yet, but I remain confident that the worst of the pandemic will soon be behind us.

As 'Yorkie' has just posted, if the aim is to get the R0 value to 1, ie peak and then plateau, then maybe we are away from the worst of the initial shock of a pandemic starting but it isn't going away.

 

[Mogster]

It’s an interesting piece of pure mathematical modelling but it discounts the effects of climate and immune pressure on the virus.

Is the healthcare system close to collapse? I don’t see it, or the exponential increase in hospitalizations. Are the extra beds in the Nightingale hospitals going to be needed? The Guardian ran an article a couple of days ago saying even in London the hospitals haven’t been pushed past 75%. It is good to have the capacity though, just in case.

As JH has said this probably isn’t going away completely although I still don’t think you can’t discount immune pressure and climate leading to a burn out as the virologists call it. SARS never became endemic, although it was much more virulent. We’ll know by May if summer weather is going to affect this markedly.

 

[Mainline421]

Not so.

https://www.telegraph.co.uk/global-...number-one-important-covid-19-sweeps-britain/


Once it's under control I believe the aim will be to get to just under (but very close to) 1.

Even if the RO did fall below 1 (which doesn't seem very likely considering that only 28% of people live alone in the UK) it would only take 1 case to cause an entire 2nd wave which could just cause a continuous cycle for maybe 5 years or so until herd immunity is achieved. (Also either way it's still exponential so I assume you meant to highlight the accelerating comment).

 

[hwl]

It’s an interesting piece of pure mathematical modelling but it discounts the effects of climate and immune pressure on the virus.

Is the healthcare system close to collapse? I don’t see it, or the exponential increase in hospitalizations. Are the extra beds in the Nightingale hospitals going to be needed? The Guardian ran an article a couple of days ago saying even in London the hospitals haven’t been pushed past 75%

First admissions to Excel this afternoon/ evening. Part of the plan at this stage seem to be to give some breathing space to other hospitals.

 

[Mogster]

First admissions to Excel this afternoon/ evening. Part of the plan at this stage seem to be to give some breathing space to other hospitals.

Yes, that’s the plan I’d imagine. They have the space so why not use it. It’ll give the staff at the Excel something to do and a relatively calm period to make sure everything is in place and working OK.

Manchester Central should be open by the weekend.

 

[hwl]

Even if the RO did fall below 1 (which doesn't seem very likely considering that only 28% of people live alone in the UK) it would only take 1 case to cause an entire 2nd wave which could just cause a continuous cycle for maybe 5 years or so until herd immunity is achieved. (Also either way it's still exponential so I assume you meant to highlight the accelerating comment).

R0 is an average...

The maths doesn't work in the way you think it does.

 

[181]

I'm not an expert (so I'm open to correction provided the people correcting me know what they're talking about), but my understanding is that:

a) If you can keep R0 below 1 for long enough you can eliminate a disease -- isn't that what happened with the original SARS? (although that was easier because it was a less infectious virus). The problems are that this might require a long lockdown with, towards the end, not many cases to motivate people to stay locked down; and that until every country in the world achieves elimination there's always a risk of new infections being imported. (I can envisage international travel restrictions going on much longer than local ones).

b) Immunity, whether from having had the disease or from vaccination, doesn't have to be complete to make a difference. Even if it makes you only 50% less likely to catch the disease, or means you catch it but less badly, that's still very useful; but on the downside it might increase the number of asymptomatic carriers,

 

[Chester1]

I'm not an expert (so I'm open to correction provided the people correcting me know what they're talking about), but my understanding is that:

a) If you can keep R0 below 1 for long enough you can eliminate a disease -- isn't that what happened with the original SARS? (although that was easier because it was a less infectious virus). The problems are that this might require a long lockdown with, towards the end, not many cases to motivate people to stay locked down; and that until every country in the world achieves elimination there's always a risk of new infections being imported. (I can envisage international travel restrictions going on much longer than local ones).

b) Immunity, whether from having had the disease or from vaccination, doesn't have to be complete to make a difference. Even if it makes you only 50% less likely to catch the disease, or means you catch it but less badly, that's still very useful; but on the downside it might increase the number of asymptomatic carriers,

My (non expert) two cents are

A) An R of below zero eventually would mean zero cases but only with no reinfection from other countries. However, not possible to completely seal off the UK for practical reasons and because of the Irish border. Besides which, this virus spreads much easier than SARS and short of literally blocking off peoples front doors as in Wuhan it would take many months if not years to get to zero purely through isolating people. What worked for one province of a country of 1.4 billion people under a dictatorship cannot be implemented for the UK (although our response has similarities). Testing and contact tracing are great tools but look at Taiwan and Singapore, they are starting to fail.

B) The more people have immunity the slower the virus can spread. Herd immunity refers to this reaching a tipping point where a virus (or bacteria) has too few carriers to spread. The Chief Medical Officer said he wasn't concerned about isolated examples of people getting it twice because that always happens to some extent. Any proportion of people getting temporary (or long term but not necessarily life long immunity) slows future spread. I don't think herd immunity was ever the governments plan for the immediate crisis but the most likely of the three possible ways in which the crisis ends (alongside vaccination and eradication). When we moved from contain to delay stage it was an admission that eradication was no longer possible. Its going to become an endemic disease in the developing world and spreads too fast to permanently eradicate it in the developed world.

While the government has failed over testing and that is damaging in other ways, we don't need huge levels of UK testing to have the general data. Germany will probably soon be in a position to provide accurate data to the world, telling us about immunity, asymptomatic case proportions etc.

 

[Mainline421]

R0 is an average...

The maths doesn't work in the way you think it does.

I'm aware R0 is an average as I actually mentioned in my first post. I wouldn't swear by everything I posted earlier and I'm aware this is a massive over-simplification of reality, but if you think it works differently please elaborate.

Chester1 has provided a very good explanation of my points above. It would be extremely naive to think it possible to "starve" out the virus as it were on a global scale and the only other way to keep R0 <1 is herd immunity as the government actually seem to understand for once.

 

[Chester1]

I'm aware R0 is an average as I actually mentioned in my first post. I wouldn't swear by everything I posted earlier and I'm aware this is a massive over-simplification of reality, but if you think it works differently please elaborate.

Chester1 has provided a very good explanation of my points above. It would be extremely naive to think it possible to "starve" out the virus as it were on a global scale and the only other way to keep R0 <1 is herd immunity as the government actually seem to understand for once.

There are two massive potential problems with herd immunity. Firstly we don't know the extent or longevity of immunity. Second, we don't know the proportion of asymptomatic cases or non diagnosed minor cases. The UCL prediction of 400,000 deaths was based on 1% fatality rate and herd immunity being achieved at about 70% of population coming into contact with it. I think that was based on about 1 in 5 cases resulting in some medical intervention. If that proportion is different (which results from countries more reliable than China should tell us soon), then the maximum number of deaths could be lower or higher. The Chief Medical Officer and Chief Scientific Advisor both seem to think that asymptomatic or minor cases are in excess of 9 out of 10, potentially as high as 19 out of 20. For herd immunity to end this crisis without collosal societal and economic damage they need to be approximately right and for infection to provide immunity for as long as required to develop a vaccine or effective treatment. If herd immunity isn't provided by infection we are basically at the stage of praying that scientists invent a vaccine (and without the option of infecting us with a weakened virus).

 

[plugwash]

a) If you can keep R0 below 1 for long enough you can eliminate a disease

But this is extremely difficult to actually do in practice because R0 is not uniform. So as you eliminate the disease from places with low R0 the overall R0 increases.

 

[edwin_m]

No, but doing it right now is a little worrying. The growth is exponential so assuming the infection rate is still (averages) each infected person infects 2 others and that everyone they infect is within the first 2 weeks before they self-isolate themselves it would take 50 weeks from the first case to when we start achieving her immunity. (1x2^25=33,555,432). Obviously that's still very flawed in reality for many reasons, but it means it likely won't be until the autumn the lockdown will end. If the infection rate reduced it will take longer but it will still be exponential so will continue "accelerating" every day as there's still gonna be more people spreading the disease.

My calculation was based on keeping the number of hospital cases below the likely number of beds available, and I came up with about 5 years to get to herd immunity on that basis. Obviously a lot of scope to change this figure by altering the assumptions, but I think it's in the right ball park. Probably we'll have a vaccine before then.

Yes very good piece of analysis.
The reckoning is that we are seeing R down to around 0.6-0.65 in UK.
Keeping R just below 1 will be very hard in the early rounds of easings and will need lots of thought and preparation (e.g. The INEOS hand sanitiser factory needs to have flooded the distribution channels for a long time to improve hygiene in the community.) + all the antibody testing.

Keeping R just below 1 is essentially the same thing, managing the rate of spread to stay within the capacity of the NHS.

So I think the way forward is to keep some level of restriction until we get a vaccine, probably sometime next year.

 

[Mogster]

Germany will probably soon be in a position to provide accurate data to the world, telling us about immunity, asymptomatic case proportions etc.

Not just Germany, I understand PHE have started limited antibody screening for data also.

 

[edwin_m]

Is the healthcare system close to collapse? I don’t see it, or the exponential increase in hospitalizations. Are the extra beds in the Nightingale hospitals going to be needed? The Guardian ran an article a couple of days ago saying even in London the hospitals haven’t been pushed past 75%. It is good to have the capacity though, just in case.

We have to bear in mind that nearly everyone admitted to hospital is going to be there for some time, so when the admission rate levels and falls the load on the hospitals will continue to increase for maybe a couple of weeks afterwards.

a) If you can keep R0 below 1 for long enough you can eliminate a disease -- isn't that what happened with the original SARS? (although that was easier because it was a less infectious virus). The problems are that this might require a long lockdown with, towards the end, not many cases to motivate people to stay locked down; and that until every country in the world achieves elimination there's always a risk of new infections being imported. (I can envisage international travel restrictions going on much longer than local ones).

b) Immunity, whether from having had the disease or from vaccination, doesn't have to be complete to make a difference. Even if it makes you only 50% less likely to catch the disease, or means you catch it but less badly, that's still very useful; but on the downside it might increase the number of asymptomatic carriers,

You don't have to knock the cases right down to zero, although if you get it low enough it would be possible to revert to a strategy of contact testing and tracing.

However once the capacity has been set up to treat large numbers, it can remain in place or be re-activated if numbers pick up again. There is then the choice between keeping the restrictions tight and casualties low, or relaxing them and accepting a higher but manageable rate of casualties which will also progress faster towards herd immunity. I think testing becomes less important if we take the second path, other than as a check on how many people have been infected possibly without symptoms and therefore what proportion are likely to be immune.

 

[hwl]

We have to bear in mind that nearly everyone admitted to hospital is going to be there for some time, so when the admission rate levels and falls the load on the hospitals will continue to increase for maybe a couple of weeks afterwards.


You don't have to knock the cases right down to zero, although if you get it low enough it would be possible to revert to a strategy of contact testing and tracing.

However once the capacity has been set up to treat large numbers, it can remain in place or be re-activated if numbers pick up again. There is then the choice between keeping the restrictions tight and casualties low, or relaxing them and accepting a higher but manageable rate of casualties which will also progress faster towards herd immunity. I think testing becomes less important if we take the second path, other than as a check on how many people have been infected possibly without symptoms and therefore what proportion are likely to be immune.

Average ICU stay for CV-19 is ~13days
Oxygen system pipework and supply systems are starting to become an issue in many hospitals. The Nightingales will be useful relief in this regard.

 

[Bletchleyite]

However once the capacity has been set up to treat large numbers, it can remain in place or be re-activated if numbers pick up again. There is then the choice between keeping the restrictions tight and casualties low, or relaxing them and accepting a higher but manageable rate of casualties which will also progress faster towards herd immunity. I think testing becomes less important if we take the second path, other than as a check on how many people have been infected possibly without symptoms and therefore what proportion are likely to be immune.

The second path can also offer an element of choice, particularly if the furlough scheme was still in place for those who felt they would rather stay isolated, though careful consideration would have to be given to avoiding too much abuse of this.

 

[edwin_m]

The second path can also offer an element of choice, particularly if the furlough scheme was still in place for those who felt they would rather stay isolated, though careful consideration would have to be given to avoiding too much abuse of this.

In principle encouraging the vulnerable to remain isolated helps with this approach, because the probability and duration of hospitalization will fall so the manageable rate of infection will increase. This means less restriction and quicker herd immunity. But it's probably very unclear how individuals will react to this suggestion. I'm beginning to think the real moral judgements on this are still in the future.

 

[Peter Sarf]

In principle encouraging the vulnerable to remain isolated helps with this approach, because the probability and duration of hospitalization will fall so the manageable rate of infection will increase. This means less restriction and quicker herd immunity. But it's probably very unclear how individuals will react to this suggestion. I'm beginning to think the real moral judgments on this are still in the future.

Talking of vulnerable. Sadly we have just seen 15 deaths at a care home in Luton with five of them testing positive for Covid-19. This indicates to me not just how vulnerable they are but more importantly that many people really are already infected.

I would be interested to see how Korea got on top of it as I assume they are still in lockdown. If we can believe China's figures then, even if they are playing things down, it will still be obvious if the end of lockdown in Wuhan results in a lot more people getting seriously ill. Give it two weeks to see if there are signs of herd immunity. I hope it turns out that relaxing lockdown does not result in a significant rise in cases. That will then show her immunity as in most of the rest of the population, that were not ill, had in fact caught it and shrugged it off.

Wishful thinking on my part I fear. Otherwise social distancing is really going to drag on for a long time. This then begs the question - how likely is it that Covid-19 is going to mutate but still stay dangerous. It has got more time and hosts to do it with if tings drag on. Perhaps it is here to stay. I mean will it be like the common cold - there will be another one or more along next winter. If that is so then maybe the human race will adapt (as in those who get it bad don't re-produce because they are dead). Maybe if we could look back far enough we would see the advent of the first common cold and see that it too was a new shock to the gene pool. But nowadays it is just a nuisance to most of us - well ignoring the vulnerable.

 

[yorkie]

I would be interested to see how Korea got on top of it...

By doing things that cannot be done here, such as 'contact tracing'; there are various articles on this subject e.g.
https://www.nature.com/articles/d41586-020-00740-y

When a person tests positive, their city or district might send out an alert to people living nearby about their movements before being diagnosed. A typical alert can contain the infected person’s age and gender, and a detailed log of their movements down to the minute — in some cases traced using closed-circuit television and credit-card transactions, with the time and names of businesses they visited.

If we can believe China's figures then, even if they are playing things down, it will still be obvious if the end of lockdown in Wuhan results in a lot more people getting seriously ill. Give it two weeks to see if there are signs of herd immunity. I hope it turns out that relaxing lockdown does not result in a significant rise in cases. That will then show her immunity as in most of the rest of the population, that were not ill, had in fact caught it and shrugged it off.

Wuhan will not have herd immunity as they also used contact tracing methods.

 

[Peter Sarf]

By doing things that cannot be done here, such as 'contact tracing'; there are various articles on this subject e.g.
https://www.nature.com/articles/d41586-020-00740-y


Wuhan will not have herd immunity as they also used contact tracing methods.

I suppose that means that Wuhan is bound to be heading for another (mini) peak then now that the restrictions are being lowered. I think it is unlikely that the virus has been completely eradicated in/around Wuhan. But in any case all it takes is someone from outside bringing it back in and off they go again. Of course there maybe some more herd immunity happening at the same time which will show up soon.

But it does look like contact tracing is the way to beat it efficiently. Herd immunity is not looking very likely/useful ?.

 

[Tom B]

Maybe if we could look back far enough we would see the advent of the first common cold and see that it too was a new shock to the gene pool. But nowadays it is just a nuisance to most of us - well ignoring the vulnerable.

I'm sure I remember reading that one of the reasons that contact with isolated tribes in the south americas was discouraged was that things like the common cold - a mild inconvenience in the western world - could prove fatal to people who had never encountered it.

 

[Chester1]

I suppose that means that Wuhan is bound to be heading for another (mini) peak then now that the restrictions are being lowered. I think it is unlikely that the virus has been completely eradicated in/around Wuhan. But in any case all it takes is someone from outside bringing it back in and off they go again. Of course there maybe some more herd immunity happening at the same time which will show up soon.

But it does look like contact tracing is the way to beat it efficiently. Herd immunity is not looking very likely/useful ?.

Herd immunity through infection was never a viable short term option but because global eradication seems impossible then the only long term ways it can be beaten are either herd immunity through vaccination or if this is too late, through infection. Everything else helps reduce deaths and delay spread but won't stop it. Contact tracing and mass testing are not a pancea, they are starting to fail in Taiwan and Singapore because the virus spreads too easily, eventually they will end up in our situation.

 

[ashworth]

In principle encouraging the vulnerable to remain isolated helps with this approach, because the probability and duration of hospitalization will fall so the manageable rate of infection will increase. This means less restriction and quicker herd immunity. But it's probably very unclear how individuals will react to this suggestion. I'm beginning to think the real moral judgements on this are still in the future.

Encouraging the vulnerable to remain isolated helps but for how long? This is especially a problem when they are living with others who are perhaps not quite so vulnerable. My 85 year old mother is living with me. She hasn’t been out beyond the garden for the last 3 weeks and because she is not so mobile these days that hasn’t been too much of a problem for her, although she does miss a ride out in the car and visit to a garden centre etc.

The big question is, as time goes on, what should I do? I have been going out most days for a walk but as we live in a rural area I can walk over fields where I see no one. I have limited my food shopping to just 2 visits a week to small local shops, mainly shopping mid afternoon when they are almost empty and I’m only out of the house for a short time. With this there is still a risk of me catching the virus and passing it on to her but I’m being careful and washing my hands etc thoroughly when I return home. How long will this have to go on?

My mum could quite easily continue socially distancing herself and could probably be able to go for short walks out of the garden and around the village without coming into close contact with anyone.
What about me in future months. Will I feel that I could go on holiday in the autumn or even this time next year. After having travelled by train to another part of the country, stayed in a hotel, visited restaurants etc will I be able to come back safely into the house with my mother or will I have to self isolate for 14 days on my return.

 

[cactustwirly]

There are two massive potential problems with herd immunity. Firstly we don't know the extent or longevity of immunity. Second, we don't know the proportion of asymptomatic cases or non diagnosed minor cases. The UCL prediction of 400,000 deaths was based on 1% fatality rate and herd immunity being achieved at about 70% of population coming into contact with it. I think that was based on about 1 in 5 cases resulting in some medical intervention. If that proportion is different (which results from countries more reliable than China should tell us soon), then the maximum number of deaths could be lower or higher. The Chief Medical Officer and Chief Scientific Advisor both seem to think that asymptomatic or minor cases are in excess of 9 out of 10, potentially as high as 19 out of 20. For herd immunity to end this crisis without collosal societal and economic damage they need to be approximately right and for infection to provide immunity for as long as required to develop a vaccine or effective treatment. If herd immunity isn't provided by infection we are basically at the stage of praying that scientists invent a
vaccine (and without the option of infecting us with a weakened virus).

Infection should provide immunity for a long time, as it effectively operates in the same way as a vaccine.
A vaccine is just a weakened form if the virus, which stimulates the immune system to produce an immune response.
Actually infection will technically give a better immunity, as the initial immune response will be much bigger than a vaccine.

 

[Chester1]

Infection should provide immunity for a long time, as it effectively operates in the same way as a vaccine.
A vaccine is just a weakened form if the virus, which stimulates the immune system to produce an immune response.
Actually infection will technically give a better immunity, as the initial immune response will be much bigger than a vaccine.

Hopefully. The Chief Medical and Scientific Officers have both made the point that if infection doesn't provide at least medium term immunity then the options for a successful vaccine become drastically reduced. There are some other options for a permenant vaccine but the biggest route would be gone. It was a polite way of them saying that if infection doesn't eventually provide herd immunity to protect those at high risk then don't bet on it happening through vaccination either.

If infection only provides immunity for a year or so then we could develop a production line and give it alongside annual flu vaccines. Repeated vaccination might be required anyway because of mutations.

 

[edwin_m]

I suppose that means that Wuhan is bound to be heading for another (mini) peak then now that the restrictions are being lowered. I think it is unlikely that the virus has been completely eradicated in/around Wuhan. But in any case all it takes is someone from outside bringing it back in and off they go again. Of course there maybe some more herd immunity happening at the same time which will show up soon.

But it does look like contact tracing is the way to beat it efficiently. Herd immunity is not looking very likely/useful ?.

Because of the effort needed to do it, contact tracing is only practicable if the number of infections is very low. I think the long and severe lockdown in Wuhan may bring them back into that range, but they may still need some travel restrictions in case the number of people moving into and out of the area is too large to manage by contact tracing. A test that gave an instant result might help with this, but one that has to be sent away to a lab doesn't seem to me very helpful as people can't immediately get the OK to travel and who's to know if they haven't caught it between doing the test and getting the result?

To get back to contact tracing in the UK I think we'd need a much longer and more severe lockdown. You need to stop virtually all transmission for at least the incubation period.

 

[Peter Sarf]

Because of the effort needed to do it, contact tracing is only practicable if the number of infections is very low. I think the long and severe lockdown in Wuhan may bring them back into that range, but they may still need some travel restrictions in case the number of people moving into and out of the area is too large to manage by contact tracing. A test that gave an instant result might help with this, but one that has to be sent away to a lab doesn't seem to me very helpful as people can't immediately get the OK to travel and who's to know if they haven't caught it between doing the test and getting the result?

To get back to contact tracing in the UK I think we'd need a much longer and more severe lockdown. You need to stop virtually all transmission for at least the incubation period.

Oh well it all brings it back round to the fact that all we can do is not overload the NHS.

 

[hwl]

Some "opinion poll" type work from Austria which suggest herd immunity is potentially very low:

Less than 1% of Austrians infected with coronavirus, study shows

Research appears to scotch hopes of countries being close to relying on ‘herd immunity’

www.theguardian.com

Less than 1% of the Austrian population is “acutely infected” with coronavirus, new research based on testing a representative sample of more than 1,500 people suggests.
The government-commissioned study, reportedly the first of its kind in continental Europe, was led by the polling company Sora, which is known for projecting election results, in cooperation with the Red Cross, the Medical University of Vienna, and other institutions.
The study made it possible to estimate the prevalence of acute coronavirus infections in Austria among those not in hospital at the beginning of April, and was designed to provide a clearer picture of the total number of infections, given gaps in testing.
The research, if replicated and confirmed elsewhere, would appear to scotch hopes of countries being remotely close to relying on “herd immunity” – where enough of the population is exposed to the virus to build up a combined immunity – as a viable policy option.

But test results from heavily impacted countries (UK US France Spain Italy etc) have +ve test results for 0.1-0.4% of their total population (mainly those hospitalised). Which suggests the Austrian estimate is under recording given that the majority with it don't seek hospital treatment and hence testing.

 

[edwin_m]

That article also says:

However, researchers, speaking at a press conference to release the results, said the study provided only a “snapshot” and did not account for asymptomatic infections, or people who were immune.


“We did not find out how many people are immune, but only how many people in Austria are currently acutely infected,” said Günther Ogris, from Sora.

So it appears they used an antigen test which would only confirm infection at the time not past infection or immunity. For that reason I don't think it tells us anything very useful about herd immunity. Also it's probably not too surprising that the number of "acute infections" they found outside hospitals was small, since most people acutely infected would want to be in hospital.

 

[Mogster]

It sounds like PCR, although the articles don’t really say. Maybe they sent out home swab kits then tested them on return.

Whatever, PCR will only detect people who are infectious in the short window of a few days (mostly, there will always be people who don’t conform to “normal”) when they have circulating virus. I also understand that the way the swabs are taken is quite critical and has an important effect on the sensitivity of the test. If they’ve asked people to swab themselves then that could influence the results, at the very least people need to be swabbed in a systematic way.

Anyway the Guardian says this wasn’t an antibody test so really the results don’t tell us that much more imo.