Almost simultaneously with the news that Biden had been elevated to president elect, the news reached me that my grandson Marius was infected with the virus. He lives in Vienna and is serious about football, in a football school where they are regularly tested. He and two others were found to be positive. No serious symptoms. Mother also infected, father not and brother maybe. Austria was locked down on the same day. A critical transition for the family. From being in the middle of a pandemic that is becoming more and more serious, but of which you do not notice anything you can see into being at home in quarantine and being obliged to keep your distance from your parents and your brother.
For some time now, I have been working on whether I can independently form an opinion about COVID-19 as a pandemic and about what I can do to stay healthy and cheerful under it. In any case, making computer models keeps me awake. The past few days I have been working on a mutation of the program that I already had, but now in such a way that I could also easily process data from other countries. From Austria, for example.
One of the ways to compare how COVID operates in different countries is to look at the total number of contaminations per 100,000 inhabitants. In Fig. 1, those values are shown for nine countries and in the first 44 weeks of the pandemic. In alphabetical order of country codes, these are Austria, Brazil, China, Germany, India, South Korea, the Netherlands, New Zealand and the USA. It is clear that Brazil, the US and India are showing linear growth, that China, South Korea and New Zealand have the virus under control and that Austria, the Netherlands and Germany have had the virus under control between weeks 16 and 35 but that it then started to show all the signs of exponential growth again. A lock-down was therefore announced last week in those three countries.
In Fig. 1 is the black line for Austria. It ends in week 44 in the middle ground, at 1269 infections per 100,000 Austrians. That number is between the Netherlands (2146) and Germany (669). The number (the infection number) varies widely between the different countries, from 6 in China to 2,807 in the US. And that number is very uncertain, because COVID infections can really only be determined by testing, and not everyone is tested. Fig. 2 provides a picture of how COVID has been faced in and by Austria.
When setting up Fig. 2 I have assumed that numbers of COVID deaths are tracked much better than numbers of infections. That idea is supported by showing those numbers together in a graph. That is what is in Fig. 2. The red line here is for the registered infections (divided by a hundred or in percentages). The black is for the recorded deaths. If those lines coincide, it is feels reasonable to think that 1 percent of infections will result in 1 death.
It is not surprising to suspect that there are regularities that link the degree of lethality of the virus with the number of infections. This is reflected in similar shapes in the two lines. And such similarities can actually be seen, both at the beginning and at the end of the period. But there is a remarkable difference: in the beginning, the lethality of COVID appears to be something in the order of 3%, and at the end it is something like 0.5%. That is a factor of six lower.
In the overview in Fig. 2 a green slider is visible that is set to 1%. I can use the slider to see how the two lines fit together. In fig. 3 I show what that is like for 3% and also for 0.5%.
Indeed, Fig. 3 suggests that the risk of death was much higher at the start of the pandemic than at the end. But that difference can of course also be due to the fact that the scope of registering who is and who is not infected has improved and expanded. It is tempting to draw conclusions from the patterns that become so visible. But a warning applies. As far as I can see, the detection and registrations of infections is still very incomplete, so that the reality behind the alleged similarities between the two lines may turn out to be very different.