Mark Twain once said, “There are three kinds of lies: lies, damned lies, and statistics.” He was right. It’s all about how you look at data. During the virus induced panic, public policy is being set by elected officials who want to save lives. I believe they are well intended. However, I do not believe they are well informed. For this reason, the policies appear to be rooted in only fear, not facts. This is a basic lesson in estimated numbers vs confirmed numbers.
Though the term mortality rate is being used regularly, most people have no idea what it means. I know that sounds crazy, but let’s look at these numbers to determine a real mortality rate.
Looking at confirmed cases only (not estimates) here is a look at flu vs coronavirus in the US for 2019-2020:
Mortality rate of flu: 10%
Mortality rate of the coronavirus: 1%
Crazy? Not at all. That’s because these percentages only use confirmed numbers, or what clinicians and researchers would call the “case fatality rate.” That is the number of confirmed cases that resulted in a death. Here is where those percentages come from, using CDC data.
Confirmed cases of flu in US this season: 222,552
Confirmed cases of coronavirus in US: 22,132
Confirmed deaths from flu in the US this season: 23,000
Confirmed deaths from coronavirus in the US: 282
So, when policy makers compare the .1% flu mortality rate to the 1% mortality rate of the coronavirus, they are comparing the estimated flu mortality rate to the confirmed coronavirus mortality rate. In clinical terms, it means they are comparing the estimated infection fatality rate to the confirmed case fatality rate. This is a horribly misleading and inaccurate comparison. And, without a true infection mortality rate with coronavirus, we are all flying blind.
To use the estimated .1% flu mortality rate as a baseline for policy making, we need to be comparing infection fatality rates of both viruses. But we can’t. Why? With the flu, we have years of testing, spread factors, demographically applied models and all the other data that goes into providing a solid estimated mortality rate for a long standing virus in the US population. With the coronavirus, we don’t have any of this. No expert has this information yet, so no model to build proper mortality rate estimates can be created.
For the flu, estimates are most often cited because a ton of people get the virus and are never formally tested or diagnosed. This is why the CDC estimates 38 million people had the flu this season, even though less than a quarter million of those people were actually tested. We simply don’t have this ability with the coronavirus yet.
Because the current coronavirus testing has almost exclusively focused on the sickest people, we have absolutely no clue how many people actually have the coronavirus. To do this properly, we need to be testing much larger groups of people–not just the sick. And, we need to following positive tests with detective work into the positive individual’s social circles with both virus tests and antibody tests. Only with this kind of detective work can we begin to have any sense of the spread factor and then properly estimate a mortality rate for this novel virus.
BTW, don’t hold your breath waiting for the CDC to do this, as they have been the primary hurdle to proper containment strategies and policy making from day one. For this to be done right, local medical professionals and labs will have to partner up to get the work done in a short period of time. They can report their information to the CDC, but shouldn’t wait for them.
Bottom line, let’s stop comparing the apples of estimated rates to oranges of confirmed rates. And let’s definitely stop making economically crushing decisions with no data or justification to do so.
