Santa Maria Sun / Letters to the Editor

Know the unknowns

Jan Lipski, Vandenberg Village -

Let’s say you have a koi pond with 100 fish, and one day you go out and you find one of the fish belly up. You test it and find out it died of a viral infection. At that point you have a 1 percent fatality rate. If there were 1,000 fish and that was the only dead fish, move the decimal point to the left for a .1 percent. Out of 10,000 fish that would be .0001 percent. But for simplicity let’s just use our 100 fish base.

So as you proceed, you notice a couple of fish looking like they’re pretty close to going belly up so you test them and both have the virus. So here’s what we know. Three percent of your pond has been infected, that we know of, and you have had one fatality. You might venture that there’s a 33 percent fatality rate of infected fish. But that would be false because you haven’t tested every fish. 

If you managed to test 50 percent of your fish and found all 50 were infected, the stats change. It would be wrong to project the infection rate suddenly increased from 3 percent to 50 percent of the population because you had no idea when they became infected. Only if you can test all the fish will you know how many are not infected. Only from that base can you continue testing on an ongoing basis. Only by comparing a changing infection rate can you make a valid calculation of the increase. Finding a new case when you don’t know when that fish did not have an infection is meaningless to determine a rate of contagion. You can only say that the infection is widespread, but we have no idea at what rate.

Getting back to our base 100 and a second fish dies. Some would say the fatality rate has increased 100 percent, doubled if you like. Having tested only 50 percent, you can say erroneously there’s 4 percent mortality rate. But that’s only of the known infected fish. You also have other factors as to how lethal the virus is. How many of the deaths were fish that were 150 years old and had other underlying diseases. 

In summary, be wary of facts expressed as percentages. Be very wary of numbers or percentages expressed with an unknown base. There are known unknowns.

Coronavirus is dangerous, but are things a little out of whack?

Jan Lipski
Vandenberg Village