Coronavirus Quickies

This post is by Phil Price, not Andrew.

There a couple of things that some people who comment here already know, but some do not, leading to lots of discussion in the comments that keeps rehashing these issues. I’m hoping that by just putting these here I can save some effort.

1. The ‘infection fatality rate’ (IFR) is not an endogenous number that describes the virus: it depends on the people and their circumstances.

By ‘it depends on the people’ I mean it is a much higher number for a typical group of old people than for a typical group of young people. It is a much higher number for a typical group of diabetics than typical group of non-diabetics.

By ‘it depends on their circumstances’ I mean it is a higher number for people who get no medical help than for people who do.

2. The IFR does not characterize how dangerous the disease is at a societal level, even if we knew the accurate number for a given population and their circumstances. That’s because the IFR quantifies the probability of dying once a person is infected; it says nothing about how likely it is that they will get infected.

Even if COVID-19 is like a slightly-worse-than-average seasonal flu in terms of IFR, it would be much much worse from a societal standpoint: it seems that nearly nobody is immune (except perhaps those who have already had it), whereas in any given year many people are immune to the flu, either because they were vaccinated and the vaccine was effective for them or because they had a similar strain of the flu in the past and are still immune. (I’m aware that the distinction between ‘immune’ and ‘not immune’ is not so clear-cut, but that doesn’t invalidate the point).

A virus with an IFR of 40% in a given population, but that only infects 0.1% of the people exposed to it, would not become an epidemic because it would not infect enough people. But a virus with an IFR of 0.1% that infects 40% of the people exposed to it would be a public health disaster and would kill millions of people. If someone says ‘coronavirus is like the seasonal flu, just look at the IFR’, they do not understand. To be like the seasonal flu in the way they mean, it would need to be like the seasonal flu both in terms of IFR and in terms of the number of people it will infect.

Those are my main points. But since I’m here I’ll go on with one more thing…hey, the rule of three, gotta do it:

3. According to the “Worldometers” data aggregation site, coronavirus deaths per million in the Republic of San Marino is over 1200. Even under the assumption that everybody there has been infected, that implies that in that population, with whatever medical care they received, the IFR was 0.12%. It’s one of the wealthiest countries in the world, and does not have a population that is highly skewed towards old people, which suggests to me that for the U.S. population as a whole — if everyone were infected, or if a simple random sample of people were infected — the IFR would be over 0.1%, even if every infected person got good medical care. This is also suggested by data from Spain and Belgium, where deaths per million are above 500 even though (I think most people agree) fewer than half the people in those countries have been infected.

This post is by Phil.