Carl Hennigan

So the problem is some of these models come down with too much certainty and don't reflect the sort of wide spectrum of uncertainties right now.

And if you thought about that, we should have never got to the point where we said it's 80% infected,

Half a million deaths.

It should have been actually somewhere between 10 and 80 percent and 20 and half a million.

The problem is right now, some of this this fight over data is not helping us clinically on the ground when we've got to deal with the situations.

Look, I do want to caveat this is this is a difficult issue to deal with right now.

We are seriously going to have to get our heads together because I think what the problem is, is that the viral epidemiologists are completely different from what I consider the clinical epidemiologists in understanding and interpreting the data.

And I think what needs to happen here is the two groups need to get together.

The people who think about what happens clinically look at a much wider range of uncertainty and would be trying to answer this in a slightly different way than the people who are the viral epidemiologists who seem to say it's all about the deaths and all about the numerators.

And what John's saying is we need to create some denominators so we can understand what's going on.

And I'm totally with him on that.

So it's quite interesting.

What they have an advantage over is they've got a defined population.

That's not too large, aren't they?

They can set a denominator of about 350,000 people in Iceland, and it's much easier to understand what's going on at that level.

Multiply that by a factor of 10 and 100, so you're at 50, 60 million new people.

It becomes much harder to do what they're doing because they can know exactly how many people came into the country from which areas around the world.

You couldn't answer that in London Heathrow.

It'd be impossible.

But I think there are a couple of points that are really interesting.