Some bad arguments about COVID-19

People are understandably getting antsy, and would like the COVID-19 restrictions lightened. I would too.

I keep seeing a few claims, however, that are simply not true, or irrelevant to the discussion.

#1 The death rate is low (or lower than we thought it would be)

Repeat after me: this doesn’t matter. There’s only three numbers that matter in the short term:

1) how many people (not the RATE: i.e., deaths per thousand, but how many in absolute numbers) die

These are important because they are unlikely to be missed, and so help us get a true sense of the disease’s prevalence.

2) how many people are hospitalized;

3) how many people are in ICU;

Why? Because the whole reason for the lockdown is to spread out the cases that will get sick over time. So, it doesn’t matter if the rate of those cases is 0.01% or 0.1% or 1% or 10%. If too many people get sick, and they swamp the health system, more people will die.

People will die for at least four reasons:

  • there will be insufficient resources to treat everyone in the best way–so some will die because they couldn’t get the benefit of full modern care;
  • people will die if they have other conditions (e.g., trauma) if resources are overwhelmed. If you are in a car accident, and need an ICU, but the ICU is full of COVID-19, you will die from the trauma, but because of COVID-19.
  • If all the deaths occur suddenly, at the beginning, the system will not be prepared, and there will be deaths that could have been prevented
  • Assuming that treatments or strategies are found that help in at least some cases, if everyone gets sick all at once in the beginning before those treatments are found, then more people will die that could have been saved.

So, the rate doesn’t matter in the short term, only the absolute numbers. If there’s a 10% death rate, 1000 people sick at a time, and 1500 ICU spots, we’re fine from a systems point of view.

If there’s a 0.1% death rate, 1000 people sick at a time, and 500 ICU spots, we have a huge problem from a systems perspective.

We’ve always known the death rate is going to be lower than what has been measured–because mild cases are less likely to be measured. The fact that this is happening is not evidence of a conspiracy or incompetence–it’s a known thing that everyone has expected.

We’ll figure those numbers out eventually, but for now they don’t matter much.

#2 More people than we thought have antibodies

This likewise doesn’t really matter in the short term. This is for several reasons:

  • we don’t yet know if antibodies are protective, or to what degree they are protective. There have been reports of repeated infections, though its not clear what’s behind them;
  • the kind of spread and death rate that is being seen in places like New York is notwithstanding whatever the background rate of antibodies is. It doesn’t really matter–the problem is what it is;
  • as I have discussed here, if the background rate of antibody presence is low (say, 15%), then 1 out of every 4 “Positive” antibody tests will in fact be a false positive. Below 15%, those numbers get even worse. Antibody tests just aren’t accurate enough at this point. There’s no way around that–it’s the basic statistics of this sort of thing, and the technical limits of antibody testing.

#3 Opening things up will at least do better for the economy

The first two are slam dunk, no question. This last one has more uncertainty, but it bears thinking about.

There’s no question that shutting down the economy is costly. The question is, would just letting the pandemic run be more costly? It is generally accepted in epidemiology that an unchecked pandemic causes more financial dislocation and harm to the economy than a controlled one with a lockdown.

So, what we are experiencing may be the least of a couple of really terrible evils. I wrote about this factor and the uncertainties around it here. You can also read here to see a published, peer-reviewed analysis that suggests that the lockdown has saved the US economy 5 trillion dollars. If you assume Canada is about 10% the size, that would be about 500 billion dollars (US).

Conclusion

People are suffering enormously over the last 4 weeks, in many ways. It is natural to want to reduce that suffering, in ourselves and others.

Whether we should open the economy further is a legitimate question and point for debate, about which no one has perfect data.

It is unwise, however, to appeal to clearly false claims as we discuss it. We don’t have perfect data and we won’t. But, we should use the data we do have wisely.