Myth vs. Reality of the Coronavirus: A Statistical Thinking Perspective

Editor's Note

This article lays little claim to originality or profundity. But it provides a much-needed synthesis of what the medical and scientific communities have been attempting to tell us.

It should be mentioned–indeed, emphasized–many conservative & liberal publications including The Daily Beast, Buzzfeed, The Wall Street Journal, Axios and others have written about the need for responsible comparisons of the coronavirus to the current flu epidemic. As written in Wired:

“For example, The Daily Beast advised: ‘Don’t worry about the new coronavirus, worry about the flu.’ Self-magazine had a calming message for Americans: ‘For perspective,’ the publication recently tweeted, 'the flu is a bigger threat in the US [than the coronavirus].’

Even the U.S. Surgeon General has gotten in on this idea: 'There are as many as five million severe cases of the flu worldwide each year, and 650,000 deaths'; 'In other words,' says Axios, ‘if you’re freaking out about coronavirus but you didn’t get a flu shot, you’ve got it backwards.’”

Properly viewed, the flu is a worldwide pandemic with serious risks. Yet we still fly, take public transportation, go to the movies, attend concerts and all the rest. In short, we maintain our normal lifestyles.

At this point in time, we have a relatively effective flu vaccine. We do not have a coronavirus vaccine. Yet more than 80% of reported coronavirus cases are asymptomatic, relatively mild and don't require major medical interventions.

According to Dr. Anthony Fauci, Director of National Institute of Allergy & Infectious Diseases at the National Institute of Health (NIH), “There are several vaccines that are very promising and one that will be tested quickly to see if it's effective, that is, protective." “But,” says Dr. Fauci, “it will probably take, at least, one year before it becomes available.”

But–and this is a big but–Dr. Fauci reports several antiviral medications–one in particular (i.e., remdesivir)–are being successfully used to treat people already afflicted with the coronavirus.

On March 12, 2020 the Associated Press reported "amid all the fears, quarantines and stockpiling the food, it has been easy to ignore the fact that more than 60,000 people have recovered from the coronavirus spreading around the globe."

Other very recent reports including one from John Hopkins ups the number of recoveries reported by the AP to be rapidly approaching 70,000 people.

The point? The antiviral medications described by Dr. Fauci are proving to be miracle drugs for those afflicted with the coronavirus.

Introduction

More than 125 years ago, H.G. Wells wrote: "Statistical thinking will one day be as necessary for efficient citizenship as the ability to read and write."

The dangers of statistical ignorance have never been greater. With respect to the coronavirus, many people are confused because they lack basic understanding of elementary statistical concepts.

Let's begin with the basic ratio error taught to millions of college/university students: Absence of a normal or failure to use a control group when needed can lead people to jump to the wrong conclusions.

To be told: 9% of the physicians who died in 2019 failed to reach the age of 50. The inference people would draw from a statement like this is that physicians die prematurely.

Appropriate control/comparison groups, however, might very well show that a higher proportion of deaths in other professions (i.e., history professors, lawyers, accountants) occurred before age 50.

Indeed, it's even possible that for the general population, about 20% of the deaths in, say, 2019 occurred before the age of 50. (This example is for teaching purposes only, and we have not researched the actual death rates used in the above examples).

Take-home message so far: A meaningful comparison is always needed to provide statistical information with meaning.

No number in itself has meaning. Numbers can only be meaningful when it is compared with another number. Statements of numbers without appropriate comparisons can be very misleading.

Dispelling Coronavirus Myths & Misconceptions

Hysteria has, in many instances, replaced useful statistical information that can be used to help Americans make rational decisions about the coronavirus with respect to their health and the health of their families.

At the very beginning of the coronavirus coverage, not much information existed about what was happening and what was apt to happen.

Now, more information is becoming available that should ease many concerns and eliminate/reduce the cause for panic.

It seems politicians and many in the media don't express the same kind of concern over the annual flu season, because there is truly no reason to make Americans panic.

Simply put, many now believe the coronavirus fears have been overhyped. This article attempts to put the coronavirus outbreak in perspective.

Reviewing the Meaning of the Word “Rate”

The word “rate" always refers to a ratio. A ratio is a statement of comparison between two numbers (6/12) or symbolically (A/B).

The birth rate, for example, is the ratio of the total number of births to the total number of women of childbearing age; the rate of speed is the ratio of distance over time; the net defect rate is the ratio of the total number of defective parts to the total number of parts produced in a given time period.

The Exaggerated Coronavirus Death Rate

The death rate related to the coronavirus has been greatly exaggerated because of the absence of a reliable denominator (the bottom part of the ratio).

Let's explain. The coronavirus death rate can be defined as: The total number of reported deaths due to the coronavirus to the total number of confirmed cases of people afflicted with the virus.

This bears repeating, no one really knows what the denominator (the bottom part of the ratio) is right now. We don't know in other words how many people have been infected.

We won't know the true death rate until epidemiologists can determine a reliable estimate of the denominator, meaning how many people actually have been infected with the coronavirus.

The number will include people who never had symptoms, or had a flu-like illness but were never tested for the coronavirus.

Most infectious disease experts predict the overall mortality rate in the U.S. is very likely to be far lower than what's being claimed now because the number of asymptomatic or mildly ill cases will be several times greater than the number of reported cases. It's that tricky denominator again!

Simply put, at this point in time, the true death rate is an unknowable statistic. Indeed, statisticians/data scientists would say what's being touted as the coronavirus death rate is a Meaningless Statistic.

Do You Remember The Swine Flu Epidemic?

Little reference is made to the 2009/2010 "swine flu" pandemic in the United States.

In mid-March 2010, the U.S. Centers for Disease Control & Prevention (CDC) estimated that about 59 million Americans contracted the H1N1 virus, 265,000 people were hospitalized as a result and 12,000 died.

Factually, the swine flu didn't dominate the news headlines–certainly not in the way the coronavirus has. More importantly, people continued to live their normal lives, that is, attended concerts, took public transportation, traveled by plane and the like.

We are not underestimating the dangers of the coronavirus. But we do believe the current hysteria which characterizes reactions to the coronavirus needs to be put into perspective.
 

What We Now Know to Be True & Not True

Many uninformed news reports still refer to the coronavirus death rate announced by the World Health Organization (WHO) which was guesstimated at 3.4% .

This is considered untrue for all the reasons discussed above.

Of late, many experts have recently claimed a much better estimate of the coronavirus death rate is probably less than 1%.

Even better news. In a recent article (March 6, 2020) in The Wall Street Journal (by Holman W. Jenkins, Jr. it was stated:

"In the New England Journal of Medicine last week, U.S. experts predicted that the coronavirus fatality rate may prove closer to the flu’s 0.1%."

To repeat: The mortality rate for the seasonal influenza (a reliable number) is 0.1%. (That's 1/10 of 1%). That's a lot lower than 1%.

The trickle of information on the coronavirus has now become a raging torrent. With each passing day, it seems the threat is lessening.

Things to Ponder About the Coronavirus

1. The following table tells a story about the 2018-2019 flu epidemic in the United States.

Source: The Next Revolution with Steve Hilton, March 8, 2020.

This was quite serious. Yet it didn't dominate the news headlines. And everybody went about their business with relative calm.

2. Flu season is hitting its stride right now in the U.S. So far, the CDC has estimated (based on weekly influenza surveillance data) that at least 12,000 people have died from influenza between October 1, 2019 through February 1, 2020, and the number of deaths may be as high as 30,000.

The CDC also estimates that up to 31 million Americans have caught the flu this season, with 210,000 to 370,000 flu sufferers hospitalized because of the flu virus.

So far (this year) 105 children have died from the flu. This does not fault Americans for anxiety over the coronavirus outbreak.

It must be mentioned–indeed emphasized "many Americans are being driven exclusively by what some term misleading headlines… by reporting that creates hysteria."

3. The coronavirus, of course, has now become center stage. As of May 15, 2020, the number of coronavirus fatalities climbed to 83,947, with 1,384,930 confirmed cases in the U.S. This still pales in comparison to the flu virus.

To repeat what we said above: medical experts recently wrote in The New England Journal of Medicine predicted the coronavirus fatality rate may prove closer to the flu's 0.1% death rate given the most up-to-date data available.

4. A very large percentage of the newly estimated coronavirus death rates occurs in the elderly and those in high risk categories due to weakened immune systems, cardiac conditions and a host of other afflictions.

To be blunt: Many in the media do not express the same type of concern over the flu season. People only know and react to what's reported to them. What is not reported to them is ignored or not factored into their decision-making process.

The official toll of the 2019-2020 flu season won't be known for months. But as always, the flu season will end in the near-future. And as predicted by many well-informed experts, so will the coronavirus season.

Using the Decision Sciences to Aid in Determining the Best Course of Action for America

Professional statisticians have a language unto themselves. Several different criteria (e.g., minimax and maximin) for selecting the best course of action have been put into very mathematical form but are quite useful when explained in plain language.

Abraham Wald (1902-1950), a mathematical statistician, suggested usage of what he called the maximin criterion. Under this method, selecting "the best of the worst" is sometimes the best solution to a complex problem.

In a mostly non-fiction movie (2014), Imitation Game, English mathematical genius Alan Turing cracks (during World War II) the German Enigma code with the help of his fellow mathematicians.

Turing, now known as the father of computer science, because of his top-secret work as a code breaker of unparalleled genius, shortened World War II by years, saved millions of lives... and was so central to the Allied victory authoritative historians claim the war could not have been won without the work of Turing and his team. 

Once the code was broken, both England, America and its allies were made aware of future German offensive and defensive plans which, among a host of other actions, included attacks on American and Allied vessels.

Next came a truly tough set of decisions. If the code breakers used their information to prevent future German attacks, the Germans would quickly know the code had been cracked and they would have changed it immediately.

So what did they do? They did allow a certain number of selected ships to be torpedoed and other targets annihilated which led to major casualties. But the losses would have been far, far bigger if the Germans changed the code.

In essence, this was a maximin solution. They selected the best of what they considered the worst possibilities, including losing World War II.

In the Holman Jenkins Jr. Wall Street Journal article mentioned above, Jenkins brilliantly describes what we would term a practical "maximin" viewpoint:

"Containment means quickly tracing the contacts of sick people and encouraging them to quarantine themselves, as well as closing schools, workplaces and public events...

… In the U.S. and other countries, it probably will not involve the forcible imprisoning of healthy people in their homes as adopted in parts of China...

…In contrast, mitigation means accepting that the virus is running flu-like through society and focusing on the severely ill. As with the flu, the elderly and those in bad health are most in jeopardy…

… Hundreds of such people die a week from “acute respiratory distress” in the best of times ... Now this question becomes societal...

… At what point should we stop working so hard to prevent transmission to people who most likely will have a mild flu- or cold-like experience in hopes of preventing a small percentage of severe cases that require costly medical intervention?…

…Containment, after all, has costs for people’s well-being too: It deprives them of jobs and income as travel is curtailed, events are called off and restaurants and other businesses empty out in ways that don’t happen with the flu...nothing similar has been tried with respect to flu outbreaks that kill thousands a year because, until now, nobody thought it worth doing."

In Conclusion

Not one reader of this article, we are reasonably sure, will agree with every point just made. But, as Peter F. Drucker repeatedly reminded us, the most serious mistakes are never made as the result of wrong answers. The truly dangerous thing is asking the wrong questions.