Technically, a “case” is confirmed, symptomatic infection.
A positive PCR test includes viral loads too low to cause infection, harmless fragments of DNA from previous but now recovered infections, asymptomatic and mild infections that will result in immunity, and symptomatic and life threatening cases.
The large and growing split between PCR positives (“cases” as incorrectly reported in the press), and death rates is a clue here that we’re getting this wrong.
The failure of many models to account for this is either corruption or incompetence.
We now know that, so far, the mortality rate in the USA – the probability that you will die (looking backward over the year) is not 1%, or 3%, as was assumed in Mar, but 0.06%.
If you get virus in the USA (i.e. become infected, including asymptomatic and mild scenarios), the probability that you will die in is 0.2% (CDC, June, with a wide margin of error). Over time, the CDC has been revising this number down.
If you become significantly symptomatic in the USA (the correct definition of a “case”), the probability that you will die is 3%.
The vast majority of people who become infected do not develop significant symptoms.
All of these numbers are averages across the entire population. If you break them down by age, the vast majority of cases and deaths are in the 60+ age demographic. The numbers for folks in their 30s are roughly an order of magnitude more favorable.