Some Useful New Thoughts on Death Rates and Other Matters

This piece is written by a retired pathologist who does a far more cogent job than I could at explaining issues in calculating death rates from coronavirus.  (Dr. Lee article)  He more artfully explains some of the same issues I have tried to warn about.  If you are looking at death rates as a percent of infections, you must be aware that the number of reported infections is far, far lower than actually exist in the population.  If you are looking at any death rate statistic, you must understand the difficulties in attributing death to coronavirus.  As the doctor says, there is a difference between dying because of coronavirus and dying with coronavirus.  A variety of techniques can be used to ferret out and estimate the true attribution of death to coronavirus and when that is done, the number will be significantly lower than is currently reported, especially because this virus is very good at making older and/or people with underlying illnesses very sick.

Next, I have also attempted to raise awareness about the difficulty of ascertaining true rates of a number of aspects of the epidemic, not just death rates.  An article in the New England Journal of Medicine describes the kinds of studies we need to get accurate data on these various rates.  Please note that, as I very poorly attempted to describe in an earlier post on making sense of coronavirus numbers, there doesn’t appear to be any way to really know how many people have been exposed to the virus.  Not everyone exposed gets infected.  We can get a better estimate of the percent of the population infected with the studies described in the article.  But it would be useful to know how many people are exposed but don’t get infected.  Right now it appears reasonable to infer that a very large proportion of the people who are exposed don’t become infected.  Why?  (NEJM Article)

And finally, here is one group’s take on a plan for reopening society and the economy, led by Scott Gottlieb, the former FDA commissioner.  (AEI Report)