One of the most important factors influencing public policy for handling the COVID-19 virus is the mortality rate, or case fatality rate.
The fatality rate is the percentage of people who will die if they are infected. Let’s say that for every thousand people infected with a virus, ten die. That would mean the fatality rate is 1%.
Early estimates of the fatality rate of the COVID-19 virus were in the range of 2-3%. For comparison, the annual flu has a fatality rate of under 0.1%. That means that for every 1000 people who get the flu, one will die.
If the COVID-19 virus has a fatality rate that’s 20 or 30 times as high as the flu, that could make it a truly terrifying disease.
The problem is, how do you compute the fatality rate? In most parts of the world, we have a pretty good idea of how many people have died. But we don’t know how many people were infected, so it’s hard to compute the fatality rate.
Two doctors in Southern California looked at the results of widespread testing, and concluded that the number of people infected with COVID-19 is far higher than previous reported. If they are right, that means the death rate is far lower than experts believe… much lower than the flu. Based on their reasoning, it’s safe to re-open businesses.
The problem is, their analysis is wrong.
To understand where they went astray – and why their claims are completely wrong – we need to understand how we estimate the number of people infected.
If we could test every single person for COVID-19 antibodies (evidence that a person is infected or has been infected), we’d know exactly how many people were infected.
But we can’t do that. So let’s look at the next best approach. Suppose we randomly chose 1,000 people and tested them, and found that 200 people had antibodies. That would mean that 20% of people were infected. Since California (for example) has 39.5 million people, that would mean 7.9 million people were infected. And we could then use the number of deaths reported, and compute the fatality rate.
Sounds simple, right?
That’s what Drs. Dan Erickson and Artin Massihi claim to have done. Using the results of testing in California, they claim that 12% of Californians – 4.8 million people – have been infected. Since the number of deaths in California at the time of their report was roughly 1,200, they computed a fatality rate of 0.03% – much lower than the flu.
But here’s the fundamental flaw in their reasoning: the people who have been tested were not a random sample of Californians!
So far, tests have been given to people who had a high likelihood of infection. They were people who had symptoms, or people who were exposed to people known to be infected.
Given the people who have been tested, we would expect a very high infection rate. But that rate has nothing to do with the general population. Statisticians would say it is not a representative sample. Inferring anything based on a non-representative sample is just fake science.
Dr. Carl Bergstrom, a University of Washington biologist who specializes in infectious disease modeling, likened their extrapolations to “estimating the average height of Americans from the players on an NBA court.” And most credible studies of COVID-19 death rates are far higher than the ones the doctors presented.
“They’ve used methods that are ludicrous to get results that are completely implausible,” Bergstrom said.
The deceptive claims of the Bakerfield doctors have been condemned by experts, including the American College of Emergency Physicians, and the American Academy of Emergency Medicine .
YouTube has removed videos posted by the doctors for violating its community guidelines, which have been strengthened to prevent coronavirus misinformation from spreading rampantly on the platform. A YouTube spokesman said “We quickly remove flagged content that violate our Community Guidelines, including content that explicitly disputes the efficacy of local healthy authority recommended guidance on social distancing that may lead others to act against that guidance. However, content that provides sufficient educational, documentary, scientific or artistic (EDSA) context is allowed — for example, news coverage of this interview with additional context. From the very beginning of the pandemic, we’ve had clear policies against COVID-19 misinformation and are committed to continue providing timely and helpful information at this critical time.”
Sadly, the doctors seem to have accomplished one thing: gaining enormous publicity for their privately owned clinic, at the expense of scientific integrity and public health.
So What’s the Real Fatality Rate?
As explained above, we don’t have good data on how many people have been infected, so it’s very hard to determine the fatality rate. Data collected from around the world shows a fatality rate of anywhere from 0.5% to 6%. It will be months before we have a final answer, but scientists agree that the COVID-19 virus is far more lethal than the flu .
- Cue the debunking: Two Bakersfield doctors go viral with dubious COVID test conclusions, CalMatters.
- Joint Statement on Physician Misinformation, ACEP-AAEM.
- Antibody tests support what’s been obvious: Covid-19 is much more lethal than the flu, Washington Post
When to Share
Share whenever you see people using spreading the fake science claims of Drs. Dan Erickson and Artin Massihi.
When sharing, please remember our recommendations for talking about science.