This is a clear, accessible explanation--in the Economist, paywalled--of the paradox of Covid-19, and the reason no one should be reassured by the overall low case fatality rate: to the contrary, in fact. https://twitter.com/NAChristakis/status/1292831244636639232

I'm shortening the argument perhaps to the point of rendering it incomprehensible, but the low CFR means that vastly more people will get it--versus, say, Ebola, which is so deadly that it tends to kills off all the hosts before the disease can spread around the world.

Covid-19 is in that bitter-spot such that it's deadly, but not so deadly as to quickly take its victims out of the population. The worst kind of virus to confront. For the human population at large, it's far *worse* than a pathogen like Ebola.

Covid-19 seems to people much milder, he writes. But in reality it's far worse: "No one would prefer to be in a group facing the second pathogen [with a CFR of 1%] rather than the first [with a CFR of 10%]—with ten times as many people infected."

Covid-19 affects people so differently, with so many having mild (or no) symptoms, that it spreads like fire. And "because, for many, the symptoms resemble a mild cold, there is a tendency for the public and politicians to take it less seriously [making] controlling it harder."

If the incubation period of a virus is longer than the latent period (as in HIV), you get many asymptomatic carriers. The incubation period of SARS-CoV-2 tends to be longer than the latent period, he notes, which is why it's been far more devastating than SARS-CoV-1,

even though SARS-CoV-1 has a much higher CFR. SARS-CoV-1 kills about 11% of people who catch it, v. an estimated 0.5-1.2% for SARS-CoV-2.

Finally, the greater the variation in R-nought (the dispersion factor: kappa or k) matters. If k = 1, R-nought is the same for everyone: the number of people infected by any infected person is about the same for every infected person. (Like the 1918 flu.)

As k approaches 0, the variability of the contagiousness of any given patient grows higher. A low k means most infect few, but few infect many; this kind of epidemic progresses in a stochastic, discontinuous way--by clusters, in super-spreader events.

But it also means that most infection chains gutter out.

SARS-CoV-1 & SARS-CoV-2 seem to have a similar R-nought (estimated between 2.2 and 3.6), but SARS-CoV-2 seems to have a higher k.

SARS-CoV-1 was chiefly spread by super-spreading events.

SARS-CoV-2 spreads in super-spreading events *and* everyday transmission. This makes it harder to control, because the everyday transmission chains are less likely to gutter out.

So what does this mean? It's bad news.

Since detection is difficult, he notes, self-isolation and mandatory quarantine make sense.

And low k means it won't work just to focus on super-spreading events. You need *much* more testing and contact tracing.

Above all, it means you can't just test people when they go to the doctor because they feel sick. You can't rely on symptoms to identify infectious people.

We need massively widespread testing with instant results.

(Fortunately--this is me speaking now, not @NAChristakis---this technology is coming online fast--it exists; it's being used already to save lives in India; and the FDA seems to be on the verge of understanding this:

It's beginning to approve rapid testing devices that don't rely on difficult-to-acquire reagents and other supplies. Why it's taken the FDA so long to understand this (this tech has been available for four months) is something I'd like to learn more about.

****NB: If anyone out there has a story of trying to get FDA approval for rapid paper (or other) testing and running into a brick wall, please get in touch with me. I'd very much like to talk to you.****

@NAChristakis concludes:

But (this is me speaking, not him), this is not necessarily the only outcome. If people just hold on and "behave prudently," that is, social distance, wear masks, remain vigilant--we're going to have a lot of much better options soon. Like rapid, cheap, on-the-spot testing.

That will change everything.

And it's very likely we'll have a safe and effective vaccine in the coming year: https://www.nytimes.com/interactive/2020/science/coronavirus-vaccine-tracker.html It may not provide perfect immunity. It may not provide permanent immunity,

but it will probably provide enough immunity that if we can produce and distribute it quickly

(vote for Biden--seriously: Trump has demonstrated that he will not be capable of overseeing this);

and if we can get everyone to take it--

(Vote for Biden, again: no one will trust a vaccine with Trump's imprimatur);

and if we can manufacture and distribute it globally (again, Biden: Trump literally won't understand why the whole world needs it, not just Americans);

we can reduce Covid-19 to a nuisance.

And we will do this soon enough that *yes,* it makes sense to live this way--that is, socially distanced, masked, and even, if necessary, with our kids uneducated--until we do. It's not forever, it's just for now. So hang on and wait, because hope is at hand, for sure.

Don't hold out hope for "herd immunity," though, because that's *not* coming: https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)31483-5/fulltext

Just keep trying to stay alive and keep your family alive. For a little while longer. Don't give up. It's worth it.

(I should be more clear--don't hold out for herd immunity that arises because people have been exposed to the virus naturally. Obviously, hold out for herd immunity by means of vaccination.)