Covid19 Reflections #11

Today’s Insight

Today’s insight is just a brief one, it is about exponential growth. We spend a lot of time talking about how the growth of Covid19 is exponential. But exponential growth is not intuitive. To illustrate this to myself, I built a very simple model.

I looked at the difference in growth rates in deaths in Australia. Today, the 31st March 2020, 19 people have died from Covid19 in Australia. If that number grows 10% a day, then by mid May, we have around 1,200 deaths from Covid19 in Australia. If that number grows by 20% a day, then by mid May, there would be around 25,000 deaths from Covid19 in Australia. There is a lot at stake in reducing the growth rate.

Cases of Covid19 in Australia have grown by an average of 12% a day over the last week. The previous week the average daily growth rate was 23%. It’s an improvement over a few days, but I really hope that this reduction in cases is real, and not just an artefact of who is being tested, and that it will continue.

Today’s link

This Wired article is a really good survey of the mathematics being used by modellers around the US – and how shortage of good data makes it harder. It also makes the point that if your model predicts a terrible outcome, and then behaviour changes (eg by locking things down) and the outcome doesn’t happen, that means the model has done its job.

Those problems can tend to catastrophize, to present a worst-case scenario. Now, that’s actually good, because apocalyptic prophecies can galvanize people into action. Unfortunately, if that action works, it makes the model look as if it was wrong from the start. The only way these mathematical oracles can be truly valuable is to goose people into doing the work to ensure the predictions don’t come true—at which point it’s awfully difficult to take any credit.

Do read the whole thing – it is a great investigation into the intersection of modelling and policy, both when it has worked, and when it hasn’t. And I can’t help reading that kind of article with climate change in the back of my mind. I’m hoping that models will become a whole lot more respected after all of this.

Life Glimpses

My mother, who is in her 70s, has given blood all her life. A while ago, she switched to being one of those super donors who gives platelets every 2-4 weeks So we had quite a debate with her over the last week about whether she should do that yesterday. She promised she would drive there, and not talk to anyone until she got in the door, but the collective medical professionals in the family were all horrified. So she stayed home, despite the Red Cross now being very short of blood due to a combination of excess donations during the bushfires, and all those people staying at home.

Unfortunately I can’t take her place.  Because I lived in the UK for more than six months between 1980 and 1996, the Australian Red Cross will not accept my blood. This is due to the almost entirely (but not quite) theoretical risk of what is colloquially called mad Cow disease (vCJD) According to this Lancet article, a total of four people have ever caught vCJD from a blood donation. It seems to me that now would be a good time to relax that restriction. When they put it in place, back in the 90s, they reduced their donor pool by 5% overnight. They probably wouldn’t get all of those people back again, but I suspect the risk of not having enough blood would be greater than the risk of transmission of vCJD.

But rant over, the blood bank is short of blood, so if you can give blood, try and make the time!

Bit of Beauty

Today’s picture comes from my morning run, it was a glorious day in Sydney this morning. Even compared with two days ago people were being more scrupulous about social distancing; stepping off the path when coming the other way so that we could stay 1.5m away from each other.