Interesting article, indeed. And it hints at a plausible mechanism.
I dont see any science there, just hunches and straw grabbing.
The effect is well known, though it involves some fairly heavy statistics – apologies in advance!
I googled for papers on the subject, and came up with this sample:
Specifically with Covid19 there is eg this article
Estimating the overdispersion in COVID-19 transmission using outbreak sizes outside China
where they talk about the “k” number as a measure of “over-dispersion” in the
Negative Binomial distribution
which is the statistical distribution they use to model random transmission of infections.
It is more properly called the “Gamma-Poisson mixture” distribution and it is used to make the Poisson distribution “more random”. The basic Poisson process models random events (eg infections) occurring over a time period at an average rate. The downside, from a modeller’s point of view, is that its variance equals that rate, ie the rate determines the standard deviation.
The Poisson process itself is realistic, and the “Poisson rate” can be linked to the “R number”, but transmission is more random in the real world. The Gamma-Poisson mixture allows you to increase that variance, without changing the rate, and so that is used instead.
Finally we get to the “k number” – the Poisson variance is increased by by the factor (k+1)/k.
So when the paper above says they found k=0.1 fitted the data, they mean the variance was 11 times higher than Poisson, and the standard deviation about 3.3 times higher.
[edited to correct: that factor should be 1 + rate/k, so the variance is increased, but it’s not as simple as 11 times]
The consequence, in their model, was “80% of secondary transmissions may have been caused by a small fraction of infectious individuals (~10%)”
If that 10% is a distinct group, then overall herd immunity can be achieved once that 10% group is largely immune, rather than the whole population.
School children and flu is well-known example:
super spreader children should get flu vaccination to protect grandparents at christmas
This paper discusses super-spreading in other diseases and finds it is a normal feature of disease transmission.
Superspreading and the effect of individual variation on disease emergence
The effect has been known for a long time – here is the 1997 origin of the 80/20 rule
Heterogeneities in the transmission of infectious agents: Implications for the design of control programs
Finally, here is a typically breathless acount of the “k number” from the
BBC
(equally breathless stuff from other news outlets is available).
Alternatively, I posted something similar here (without the heavy statistics) on the 15th August :-)
Yep, I follow the logic. Simplistically, the children become herd immune, so they arent infected, and dont infect their parents, and grand parents, who continue to isolate to some degree. That make sense, except I dont think there is much evidence that immunity rates are all that high in any group of the population yet? Even with a group, take for example the 15 to 25 year olds, there would surely need to be an immunity level of at least 50% before the virus is suppressed in this group. I dont believe we are close to this level in the UK yet.
I think the fear is once the infection rate starts rising in this group, and they bringing the virus into close quarters as people move in doors in the Autumn, children wish to see parents and grandparents etc., the typical rise in Influenze we see in the winter may also be mirrored by COVID.
Never the less, while it is far from perfect, it seems to me Europe is doing a better than expected job of suppressing the virus, and I sense better that the experts expected. If nothing else I feel it demonstrates that the measures taken by large numbers of the population are effective and that when it comes to population the spread of the virus can be controlled so long as complacency does set in too much.
Fuji_Abound wrote:
I dont see any science there, just hunches and straw grabbing.
I didn’t claim there was any science there. Just the fact that the epidemic ended by itself with no more than 0,165% excess deaths. You may draw your own conclusions.
Another interesting scientific insight about Covid 19, pertaining to the role of bradykinin:
https://elemental.medium.com/a-supercomputer-analyzed-covid-19-and-an-interesting-new-theory-has-emerged-31cb8eba9d63
I wonder if that even explains the possible effectiveness (or ineffectiveness) of chloroquine or hydroxychloroquine, cf. https://www.jbc.org/content/267/1/303.full.pdf
Effect of Chloroquine on Metabolic Processing of Intracellular pH]BK – Chloroquine penetrates cells and equilibrates the pH of cellular compartments with that of the cytosol. Treatment of DDTl MF-2 cells with 100 p~ chloroquine for 30 min at 37 “C resulted in a 30-40% decrease in the amount of [3H]BK specifically bound to the cell surface at 4 “C (data not shown).
[…]
Addition of chloroquine resulted in a decrease in the degradative activity in both compartments. These results suggest that the plasma membrane and microsomal compartments that BK is routed through are acidic and that the enzymes responsible for degradation of BK in these compartments are partially dependent on these acidic conditions. The cellular routing of BK through these compartments does not appear to depend on the acidic nature of the compartments because chloroquine was without effect on the amount of BK-derived activity present intracellularly and released into the media and on the distribution of activity between the plasma membrane and microsomal compartments.
Back in the early days of all this, I got frustrated by the standard epidemiological models which essentially assume that every infected person is just as likely to infect every susceptible person. This is clearly nonsense – if I’m sick it’s likely I’ll infect my wife, less likely I’ll infect Peter in whatever rural backwater he’s in (and who I’ve never met). So I came up with the idea of “fractal herd immunity”. I spent quite a lot of time writing simulation model (individually simulating up to 10 million people, not some kind of mathematical model). It proved what I thought. Glad to see I’m not the only one who has thought of this.
I think we should stop over thinking this and lock down. Then throw the key away. Anything else and you are a NAZI bent on killing civilization.
172driver wrote:
I am well over 50 and I have to – the best of my recollection – never had the Flu. I might have had it as a small child, but never within my living memory.
If you have never had the following symptoms then you are truly lucky and should give your plasma to science.
Sudden onset of the following symptoms:
Fever above 100F
Severe pains in muscles and joints ( I’veNever had it severe so its patient dependent)
Fatigue
Flushed skin and red watery eyes
Chills
Headache
Dry cough initially
Sore throat and runny nose
red watery eyes
Only when I got the divorce bill
