International Mortality Rates Updated; What Do They Tell Us?

Our friend Brian Sullivan has updated the international mortality table that I posted a few days ago, through yesterday. It shows deaths per million of population in 12 Western European countries, South Korea and the U.S. The blue bar shows the per capita death rate (per million) as of March 16, the orange bar as of March 20. Click to enlarge (slightly):

The U.S. has the lowest per capita death rate of this group, tied with Austria, at 0.7 per million, up from 0.3 four days earlier.

Italy has the highest mortality rate by far, at 67 per million. It is too early to gauge the ultimate course of the epidemic, but in crude terms it looks as though Italy is heading for a death rate of something like 100 per million. It could easily go higher than that.

I don’t think the U.S. will experience a mortality rate anywhere near as high as Italy’s, for a number of reasons including our younger population, far fewer smokers, lower population density, a better health care system and early deployment of anti-viral drugs, some of which are likely to prove helpful. But let’s assume the U.S. ultimately sees a mortality rate of 100 per million. That would be 143 times the current U.S. rate, not outside the realm of possibility. Do the math: if we have around 330 million people, and 100 die per million, that equals 33,000, which would be equivalent to the deaths from an average seasonal flu season. Maybe it’s worse than that; maybe by the time it runs its course, the death toll from COVID-19 rises to 200 per million, 286 times the current rate. That would still be less than the death toll from flu in the U.S. just two years ago.

Maybe the Wuhan virus will prove much worse than any of those crude assumptions suggest. It is too early to rule out that possibility. But policymakers need to consider the possibility that the damage done by the extreme measures being taken to slow the spread of the virus will ultimately prove to be greater than the harm done by the virus itself.

Responses