Introduction
The case fatality rate of any disease is easy to calculate: just divide the number of those who died of the disease by the total number of confirmed cases. The infection fatality rate is a very different figure. It is the number those who died of a disease divided by the number of who were infected (most of whom might never know about it).
I was reading an online paper the other day when I came across an article that discussed the findings of a new study yet to be peer-reviewed that estimated the infection mortality of Covid-19. Here's a link to the study titled Statistical techniques to estimate the SARS-CoV-2 infection fatality rate:
https://www.medrxiv.org/content/10.1101/2020.11.19.20235036v1.full
Here's a link to the article, which I had translated by Google Translate:
The study was based on tests of Covid-19 specific antibodies in the blood
It said that the group used a serological antibody test on a large sample of the population on various countries to find out how many have already have Covid-19. The group developed a mathematical model to correct errors caused by various biological and administrative delays between infection and reporting.
The results and conclusions
The result was that in Western societies the infection mortality of Covid-19 is 0.4%. That is, one out of 250 who get Covid-19 die of it. The 90% confidence interval was 0.2% to 0.8%, which means that the likelihood of the true figure being within that interval is 95%.
The infection mortality of seasonal flu is around 0.1%. Based on this study, Covid-19 is a few times deadlier than the seasonal flu.
I've read somewhere that the antibodies against Covid-19 may not stay in the blood for longer than a few months but that the If that is the case, then any study based on antibodies in the blood as evidence of someone having had the disease will underestimate the number of people in the population who has had the disease and thus overestimate the infection mortality of it. Cell-mediated immunity means the type of immunity that is based on memory cells of the immune system retaining memory of a once-encountered pathogen, which enables the rapid production of antibodies against it if it is encountered again.
It will be very interesting to find out what the actual infection mortality rate(s) is/are when medicine finally gets to the bottom of the disease.