Why the numbers we’re given don’t tell us anything meaningful
In many countries around the world the mass media is having a great time sensationalizing the latest in a long line of new viruses. Forget MERS and SARS and H5N1; today’s you-must-be-terrified-about-this virus is COVID-19 and it’s enabling the media to boost ad revenues nearly as much as a Presidential assassination.
Speaking of which, why hasn’t anyone yet bothered to….? But I digress.
Politicians, who always feel obliged to respond to whatever the mass media is pushing lest anxious citizens accuse them of insufficient activity, are exacerbating the problem as they so often do. And so the fear factor ramps ever-higher.
Let’s pause and ask: what do we really truly know about the heath risks associated with COVID-19?
In fact, not very much.
It appears that in many people the virus is asymptomatic, which means they don’t even get a sniffle. In many other people it manifests as a cold or a mild flu. These aren’t very sensational things so they rarely get a mention from earnest reporters talking straight to camera with a gauze mask fatuously obscuring their nose and mouth.
Some people do indeed die from infection: the very old and frail, and those with compromised immune systems.
Guess what? These are the very same people who die every year from regular flu. Up to 650,000 people a year worldwide die of flu and more than three billion people contract it. But we’re not told to panic and run around like headless chickens by the mass media because they know they can’t create sensation from regular flu. We’re too familiar with it. We know most people will, at worst, feel rotten for a few days and then recover and get on with life as normal. But COVID-19 is new and unknown, so it can be sensationalized.
After a lull in which it seemed there was insufficient fear to be peddled, the mass media has now returned to full-on mode, gleefully publishing statistics about infection rates and deaths. It appears infections are spreading and deaths are increasing at a terrifying rate.
But are they really?
Here’s the thing: the more testing that is carried out, the more cases will be found. And the more cases that are found, the more deaths can be attributed to the disease. It seems as if COVID-19 has been around for many months now, as belated testing in the USA appears to confirm.
Let’s take an example of what happens when we test more people. Country X hears about COVID-19 in January and tests 1,000 people coming back from China. 250 test positive. After a while, five very old and frail people die. Sad, but everyone dies eventually. All in all, there’s a bit of media sensation to be had here but it is slim pickings.
Now let’s move forward in time. Country X tests 50,000 people in March (a fifty-fold increase in testing) and now 15,000 people test positive, of whom eighty subsequently die.
We’d expect a strong correlation between number of people tested and number of people in whom the virus is detected. It’s just the same as if I look at road deaths only in Oregon (maybe 1,800 per year) and then later look at road deaths in California and Washington also (maybe 15,000 per year). Because my sample size is larger, my positive results will be larger too. But it doesn’t mean that road deaths are increasing exponentially and that we’re all going to be killed in auto crashes.
Unfortunately the mass media doesn’t bother to explain this important point (maybe the journalists are just too simple-minded to understand anything at all, least of all the harm they are doing with their sensationalism) so all we get is the headline:
CORONAVIRUS DEATHS INCREASE FROM 5 TO 80 IN JUST ONE MONTH!!!!
Put like that, it seems at first glance like an epidemic is truly on its way. Imagine: maybe next month there will be 100,000 deaths and if we keep on extrapolating then by the end of the year there will be more deaths than there are people alive on Earth today! Our tiny ape-brains always draw straight lines into the future because we’re not evolved to deal with complex systems in which there are many interdependent variables all acting on each other in complex ways. So we’re easy to scare because we don’t really understand data. We only understand anecdote, and that’s a very unreliable basis for anything at all.
As more countries test more people it’s obvious we will be regaled with ever-greater numbers of people who’ve contracted the virus and ever-greater numbers of people whose deaths will be attributed to it.
But is it really a cataclysmic event if Auntie Eve, who is 87 and extremely frail, dies tomorrow from COVID-19 or dies next month from heart failure? Our modern obsession with hiding death away in hospitals and pretending there’s an (expensive) cure for everything means we’re uniquely incapable of accepting the inevitability of death.
That in turn means we’re easily tricked into panicking over what are statistically meaningless risks. When up to 650,000 people die of flu each year and more than 1,000,000 people die each year in auto accidents and several million die of obesity-related diseases, the ultimate death toll wrought by COVID-19 won’t even be a rounding error in the daily mortality statistics.
At some point the true morbidity value of the virus will be known. As we have so much data on regular flu we know that it kills between 0.1% to 0.2% of people infected. At present, data from Western countries seems to indicate that the coronavirus kills around 0.4% of people infected, but there are two caveats: the first is that many infected people don’t even know they have the virus because they remain symptom-free, so the death rate estimate is too high; and the second caveat is that some people who are still alive may ultimately die because of the virus, which would lead the death rate estimate to be a little too low. But, best guess, is that COVID-19 is about four times as deadly as regular flu if you’re very old or have a compromised immune system.
So yes, many of us likely will contract COVID-19 in the weeks and months ahead. But nearly all of us will either experience no symptoms, mild symptoms, or a few rotten days just like with regular flu. Some very old and compromised people will die, just as happens every year with regular flu.
The world is not, in fact, coming to an end.
We really ought to know better than to allow ourselves to be stampeded by a mass media obsessed with generating ratings, but it seems that no matter how often we’re sensationalized into blind panic, we never learn our lesson.