Possible reasons why the R value is lower than expected in some places

[yorkie]

I'm curious to learn why the R value in places like London and Stockholm is reportedly now very low, when these are the sort of places there the R value was previously very high and I found some research that indicates it could be that they are approaching herd immunity with a much lower proportion of the population being infected than simplistic models had predicted.


https://arxiv.org/abs/2005.03085

The disease-induced herd immunity level for Covid-19 is substantially lower than the classical herd immunity level

https://www.medrxiv.org/content/10.1101/2020.04.27.20081893v2

...simple calculations suggest that herd immunity to SARS-CoV-2 requires 60-70% of the population to be immune. By fitting epidemiological models that allow for heterogeneity to SARS-CoV-2 outbreaks across the globe, we show that variation in susceptibility or exposure to infection reduces these estimates. Accurate measurements of heterogeneity are therefore of paramount importance in controlling the COVID-19 pandemic.

https://judithcurry.com/2020/05/10/...covid-19-is-reached-much-earlier-than-thought

In my view, the true herd immunity threshold probably lies somewhere between the 7% and 24% implied by the cases illustrated in Figures 4 and 5. If it were around 17%, which evidence from Stockholm County suggests the resulting fatalities from infections prior to the HIT being reached should be a very low proportion of the population. The Stockholm infection fatality rate appears to be approximately 0.4%,[20] considerably lower than per the Verity et al.[21] estimates used in Ferguson20, with a fatality rate of under 0.1% from infections until the HIT was reached. The fatality rate to reach the HIT in less densely populated areas should be lower, because R0 is positively related to population density.[22] Accordingly, total fatalities should be well under 0.1% of the population by the time herd immunity is achieved. Although there would be subsequent further fatalities, as the epidemic shrinks it should be increasingly practicable to hasten its end by using testing and contact tracing to prevent infections spreading, and thus substantially reduce the number of further fatalities below those projected by the SEIR model in a totally unmitigated scenario.

https://www.independent.co.uk/news/...london-lowest-rate-infection-uk-a9515761.html

London’s rate of coronavirus infection has fallen to less than 24 cases a day, the lowest in the UK, nearly two months after the region hit a high of 200,000 new cases in a day at the start of the nationwide lockdown.

Research by Public Health England and the University of Cambridge's MRC Biostatic Unit showed the number of daily infections in London now halving every 3.5 days, which means coronavirus could be wiped out there within just two weeks.

 

[6862]

I think while there is some evidence to suggest that R is not quite so simple as we perhaps once first thought (i.e. different people contribute differently to the overall R, and hence may lower the herd immunity threshold), we probably need to observe the effects of releasing lockdown for a while longer before being confident that the lower R is due to variable individual 'spreading ability' as opposed to the fact that about 80 % of people are staying at home more than they used to (and a significant minority are so terrified that they won't go out at all).

 

[yorkie]

I think while there is some evidence to suggest that R is not quite so simple as we perhaps once first thought (i.e. different people contribute differently to the overall R, and hence may lower the herd immunity threshold), we probably need to observe the effects of releasing lockdown for a while longer before being confident that the lower R is due to variable individual 'spreading ability' as opposed to the fact that about 80 % of people are staying at home more than they used to (and a significant minority are so terrified that they won't go out at all).

I'm not suggesting the current situation is changed before 1st June, though clearly some easing needs to take place after that date.

But if it was true that the R value is so much lower when more people stay at home, then does that mean more Londoners are staying home and there is more social distancing in London than elsewhere? I find that hard to believe; anecdotal evidence appears to suggest the opposite of that. And how could you explain that the R value was so much higher in London previously?

 

[Belperpete]

Agreed. It is noticeable that in the UK, those areas that had it worst (London and the Midlands) seem to be seeing the numbers falling fastest, whereas in other areas (such as the NE and SW) the numbers hardly seem to be falling at all. However, in France, Paris still seems to be the worst affected part of the country. I too find it hard to believe that social distancing can be better in London, considering the number of people using the tube and buses.

It may well be that the herd immunity for this virus is much lower than 60-70%, but intuitively I find it hard to believe that a level of 20% could be effective - let alone 7%! My experience of lockdown is that I am interacting with a small pool of the same people (generally the same few checkout operators when I visit the supermarket after 9pm, and such-like). During lockdown, when people are interacting with a small pool of people, it may well be that you only need one two in that pool to be immune to significantly reduce the infection rate.

 

[Simon11]

I wonder if it is because there is more risk of one care worker spreading the virus to a care home with say 50 residents with close contact, than there is one person spreading the virus on a tube train where most people are being careful?

With London having a younger generation, there are likely to be far more older people and care homes outside of London?

Just my thought?

 

[bramling]

I'm not suggesting the current situation is changed before 1st June, though clearly some easing needs to take place after that date.

But if it was true that the R value is so much lower when more people stay at home, then does that mean more Londoners are staying home and there is more social distancing in London than elsewhere? I find that hard to believe; anecdotal evidence appears to suggest the opposite of that. And how could you explain that the R value was so much higher in London previously?

Could it be that it was, hitherto, spreading heavily via public transport - which despite some media reports concentrating on a few exceptions - has tended to remain very empty in London since the lockdown came in?

 

[yorksrob]

All very interesting.

I can't think of any reason why the NE appears to be so much higher than London in terms of transmission. I guess community antibody testing on a wide scale might help to explain things.

Could it be that it was, hitherto, spreading heavily via public transport - which despite some media reports concentrating on a few exceptions - has tended to remain very empty in London since the lockdown came in?

Public transport's been emptier everywhere else, so I don't see what that would have to do with it.

 

[bramling]

All very interesting.

I can't think of any reason why the NE appears to be so much higher than London in terms of transmission. I guess community antibody testing on a wide scale might help to explain things.



Public transport's been emptier everywhere else, so I don't see what that would have to do with it.

But could this have essentially forced London into staying at home as many more people in London don’t have cars? This combined with these other areas being behind London in terms of time.

 

[yorksrob]

But could this have essentially forced London into staying at home as many more people in London don’t have cars? This combined with these other areas being behind London in terms of time.

But for the first month and a bit of lockdown, no one's been driving anywhere around here either.

Transport just doesn't seem to be a driver of this.

 

[Mogster]

Maybe London with a younger demographic has had far more asymptomatic/mild cases than other parts of the country. S Korea seems to indicate that the “club scene” may be a major source of cases, there’s plenty of that in London but it’s mainly a young persons bag.

 

[Mag_seven]

Is the "R" value estimated to be in force at any one time at any one place actually published anywhere? If not it should be.

 

[Belperpete]

... these other areas being behind London in terms of time.

The lockdown was imposed at the same time everywhere in the UK.

There are just so many reasons why London should in theory have higher transmission (public transport usage, higher population density, people travelling to and from Heathrow to return to their home country, etc), it just seems so counter-intuitive that it has the lowest transmission rate.

 

[PTR 444]

Could it be that London has already achieved herd immunity?

I think the virus has been circulating in London for much longer than first thought, and that has allowed it to spread to almost all of its population via the tube and mass gatherings throughout February and March. With a high proportion of the London population being under 40, I wouldn’t be surprised if most have already had the virus without realising because they showed no symptoms.

 

[Belperpete]

Could it be that London has already achieved herd immunity?
I think the virus has been circulating in London for much longer than first thought, and that has allowed it to spread to almost all of its population via the tube and mass gatherings throughout February and March. With a high proportion of the London population being under 40, I wouldn’t be surprised if most have already had the virus without realising because they showed no symptoms.

I very much doubt that it has spread to almost all of London's population. But it may have reached sufficient of those who use public transport to achieve herd immunity in those people.

 

[yorksrob]

Again, it all comes down to that antibody testing. Just having an idea of who has had it, where, will help us begin to put the jigsaw puzzle together.

 

[ChrisC]

Could it be that London has already achieved herd immunity?

I think the virus has been circulating in London for much longer than first thought, and that has allowed it to spread to almost all of its population via the tube and mass gatherings throughout February and March. With a high proportion of the London population being under 40, I wouldn’t be surprised if most have already had the virus without realising because they showed no symptoms.

According to an article in today’s Daily Telegraph figures show that by the time the country went into lockdown the reproduction ‘R’ rate was already falling in London but was still high in the regions. I find that very interesting and I do wonder at what rate they were falling and how quickly they would have continued to fall without the lockdown. The article also suggests that infection rates are now so low in London that the virus could almost disappear from London within 2 weeks.

I’m Sorry that I’m not any good at posting actual quotes but perhaps someone else will be able to find the article and post the relevant parts here.

 

[AM9]

Don't forget that the cases counted for London are those where the patient lives in the city. A very large proportion of London's workforce lives outside the GLA boundaries, hence the volume of long-distance commuting. Additionally, if commuting is considered to be a major vector of infection, then a large portion of those long-distance commuters use LU and or buses for the last leg into work. Once mass homeworking started and commuting largely stopped, there were far less opportunities for transmission so the R figure for residents fell rapidly.
It may be that the long-distance commuters were significant factors in carrying the infection to provincial towns, whilst the London residential population was isolated soon after.

The problem with opening London back up, even if the R figure is the lowest in England, is that commuters from the provinces would bring it back up because the reinfection rates are still higher outside the capital.

 

[yorkie]

Could it be that London has already achieved herd immunity?

Not yet, but the research does appear to suggest that it's heading that way.

I very much doubt that it has spread to almost all of London's population....

But have a look at the research linked to above; there is a possibility that a smaller percentage of the population needs immunity for there to be a significant effect.

I don't think there can be any doubt that the original models were far too simplistic; the more data that can be gathered the more accurate the modelling will be. Clearly a lot more research is needed, but it does sound promising to me.

 

[Yew]

If the findings in the Judith Curry article, and the Independent article are both true, It could simply be that London has reached a point where (the early stages of) herd immunity has an effect.

Edit:

In more detail, Lewis suggest that instead of ~60% infection rate before herd immunityis reached, their model suggests that due to the actual coefficient of variance (basically, how many different people you see) being lower than assumed in earlier models, it is more likely to be in the range of 6.9-23.6%. Given that research from the University of Manchester (https://www.manchester.ac.uk/discover/news/over-25-of-the-uk-likely-to-have-had-covid-19-already/) suggests that there is scope for up to 25%.

There are still some flaws in drawing this information together, especially as the I'm not entirely certain under which conditions their coefficient of variance assumptions are actually valid. But prima face it stands to reason that modelling the UK as 70 million people all randomly bumping into each other, is overly simplistic, as in reality we are geographically constrained, and mainly staying around the same colleagues/friends/family members. Understandably though, there is a lot of variation in this, someone who drives to work in a small office, and goes to the same quiet pub on Fridays, has a significantly different risk profile to someone who commutes on the tube, to a public facing role and goes out to packed night clubs. It'd be interesting to run their models with varying coefficients of variance, to try and represent this difference.

 

[DynamicSpirit]

According to an article in today’s Daily Telegraph figures show that by the time the country went into lockdown the reproduction ‘R’ rate was already falling in London but was still high in the regions. I find that very interesting and I do wonder at what rate they were falling and how quickly they would have continued to fall without the lockdown.

My experience in London was that a fair few people were starting to deliberately limit their social contact because of the virus a couple of weeks before the country went into lockdown. I would suspect that was not happening to such a high degree in the rest of the country, and that could well account for the falling R rate for London at the time you mention.

 

[MarkyT]

Maybe London with a younger demographic has had far more asymptomatic/mild cases than other parts of the country. S Korea seems to indicate that the “club scene” may be a major source of cases, there’s plenty of that in London but it’s mainly a young persons bag.

A significant new 'superspreader' outbreak occurred in Seoul's clubland recently but has hopefully been shut down fairly quickly and efficiently. An asymptomatic individual visited a series of clubs one night in early May but a week later was tested positive. A small number of rare new infections were traced to the establishments he reported visiting. A cascade of thousands of his first order and subsequent contacts has since been identified and traced by various means revealing a string of new infections. While mask-wearing is still very common in South Korea and is often necessary to enter nightclubs, people inevitably remove them once inside. Energetic and intimate socialising is clearly very risky for transmission of this virus, so clubs and bars have been temporarily closed down again in the capital until further notice.

 

[Bald Rick]

A thought, not based on much evidence.

I seem to remember that those with chronic diseases are much more likely to suffer from the infection; ie if you catch it, you’re more likely to get the symptoms, and the symptoms to be bad, if you have a chronic disease. Something like 80% of chronic diseases are linked to lifestyle, with obesity and smoking being particular triggers.

Now it might be just my experience, but people in London tend to be, dare I say, slimmer and fitter than in many northern cities; and their seem to be fewer smokers. I certainly feel like I stand out whenever I’m in Blackpool and the West Mids; and let’s just say the salad section in the Tescos I go to in up there (and in Scotland) is about a third of the size of those for similar sized stores in London. My parents (W Mids) routinely comment on ‘how thin everyone looks’ when they come down south.

If you follow this logic, then for the population as a whole, those in London may be less likely to present with symptoms than those further north, and thus be less likely to be tested.

Notably S Korea has some of the lowest rates of obesity, smoking, etc, and has done remarkably well with the virus - helped by technology and testing of course, but notably didn’t do much in way of a lockdown.

Doesn’t explain the Paris thing though, where most people are similarly svelte.

 

[yorkie]

Doesn’t explain the Paris thing though, where most people are similarly svelte.

If I lived in Paris my mental and physical health would not be in a good state, and if I'm feeling happy and not stressed then I tend not to get viruses. I wonder if this might have something to do with it; a harsh lockdown is hardly going to get anyone's immune system into good shape!

 

[Greybeard33]

I'm curious to learn why the R value in places like London and Stockholm is reportedly now very low, when these are the sort of places there the R value was previously very high and I found some research that indicates it could be that they are approaching herd immunity with a much lower proportion of the population being infected than simplistic models had predicted.

https://arxiv.org/abs/2005.03085

https://www.medrxiv.org/content/10.1101/2020.04.27.20081893v2

https://judithcurry.com/2020/05/10/...covid-19-is-reached-much-earlier-than-thought

https://www.independent.co.uk/news/...london-lowest-rate-infection-uk-a9515761.html

Of the four links in the OP, only the first two are to peer reviewed scientific papers. The third is to a blog by a climate researcher (Nicholas Lewis), while the Independent story does not mention the herd immunity theory.

The first paper estimates a reduction in the herd immunity threshold from 60% to 43%. However, it lists a number of caveats and concludes:

The present study highlights that the disease-induced herd immunity level is substantially smaller than the classical herd immunity level. To try to quantify more precisely the size of this effect remains to be done, and we encourage more work in this area.

43% is well above most estimates of the percentage who have been infected in London.

The second paper estimates that the herd immunity threshold could be as low as 10-20% in a heterogenous population. However, one of the co-authors of the paper, Caetano Souto Maior, has cautioned, in his Twitter feed, against misinterpreting this result:

As a coauthor in this study I'm going to start by saying what its conclusions are NOT: (i) it does NOT conclude that lockdowns/social distancing are unnecessary; (ii) it does NOT say "herd immunity" is ~20% regardless. In fact that number is conditional on social distancing 1/6

The study does say that given observed trajectory, a SECOND wave may be considerably smaller IF there is a large variance in the distribution of both biological susceptibility AND exposure/social contact. There's a lot to unpack there 2/6

As @joel_c_miller put it recently, heterogeneity/variation is complicated, and often doesn't change significantly the conclusions from "homogeneous" models. The point is that here it just might, but we are mostly ignorant of the actual parameter values 3/6

Two corollaries to the main results information on the distribution of biological susceptibility to #coronavirus could be very useful (but it is also very hard to infer), and that social distancing has a larger effect than its average may suggest -- and that's good news. 4/6

The flip side is that relaxing lockdowns/social distancing measures will have a more detrimental effect than what is predicted by the average (bad news). That's because of the dependence of reproductive numbers on degree of exposure 5/6

So like everything in science, this is not the ultimate truth. This is not saying everyone else was wrong and we are right, and especially is not saying public health policy must conform to this one study, but that this is likely an important factor that deserves attention. 6/6

In other words the lockdown itself reduces the herd immunity threshold. As I understand it, the lockdown confines the virus to hotspots such as hospitals and care homes, within which herd immunity can develop, with few opportunities to escape to the wider community. Conversely, if the lockdown is lifted prematurely, infections can be seeded from those hotspots, then grow exponentially again in the wider community, where the proportion who have been infected is still much lower. So the herd immunity threshold will rebound back up towards 60%.

 

[Belperpete]

In other words the lockdown itself reduces the herd immunity threshold. As I understand it, the lockdown confines the virus to hotspots such as hospitals and care homes, within which herd immunity can develop, with few opportunities to escape to the wider community.

That does seem intuitively correct. Likewise, if public transport is a hotspot, but on a larger scale, then the same should apply.

 

[philosopher]

There is an article in the Daily Mail today saying the sunny April may have helped reduce the number of cases in the UK as it would have increased levels in Vitamin D and so strengthened people’s immune systems. Furthermore this could partly explain why the R value is lower in southern parts of the UK such as London as sunlight will be stronger there.

 

[yorksrob]

In other words the lockdown itself reduces the herd immunity threshold. As I understand it, the lockdown confines the virus to hotspots such as hospitals and care homes, within which herd immunity can develop, with few opportunities to escape to the wider community. Conversely, if the lockdown is lifted prematurely, infections can be seeded from those hotspots, then grow exponentially again in the wider community, where the proportion who have been infected is still much lower. So the herd immunity threshold will rebound back up towards 60%.

So in that scenario, presumably London itself is one of the former "hotspots" with herd immunity.

Presumably them, if you know where those hotspots are likely to be (e.g. care homes and hospitals) you can throw a lot of testing at them.

 

[Bald Rick]

There is an article in the Daily Mail today saying the sunny April may have helped reduce the number of cases in the UK as it would have increased levels in Vitamin D and so strengthened people’s immune systems. Furthermore this could partly explain why the R value is lower in southern parts of the UK such as London as sunlight will be stronger there.

Whilst there may be something in that, similarly there maybe something in it being warmer down south (the virus apparently doesn’t survive so long in the warm), there’s a huge caveat on this being a report in the Daily Mail. Was it quoting any reliable medical experts?

 

[Bletchleyite]

Whilst there may be something in that, similarly there maybe something in it being warmer down south (the virus apparently doesn’t survive so long in the warm), there’s a huge caveat on this being a report in the Daily Mail. Was it quoting any reliable medical experts?

It also doesn't tally with Italy and Spain, Southern European countries that are typically considerably warmer than any part of the UK.

 

[takno]

It also doesn't tally with Italy and Spain, Southern European countries that are typically considerably warmer than any part of the UK.

Most of the deaths in Italy were in Lombardy and in Spain were in Madrid, neither of which are very warm or sunny in February or early March. Numbers in both are down very significantly since then.