by Haron » Sat Nov 14, 2020 4:43 pm
The virus discriminates. I would think that was the case for most virus, but I could be wrong. Anyway, you are right that this is a factor. If, for some reason, the increase in infected only comes in certain resilient age groups, the death rates would be lower than if infections were spread equally across all age groups. And to some extent this may actually be the case now, as young adults generally are more careless than elder people. And to the extent that this is the case, I would agree that it is good news; not so many people in risk groups get infected. There is a substantial risk that even if the increase in infection starts in this age group, though, it will spread from those to others. And it anyway means that the risk of being infected increases for everyone.
As for my graph, that was not based on any specific model I've read about. It was simply to illustrate why death rates and numbers of infections are not correlated. The deaths today depend on how many was infected every day for the last month or so. So if the numbers of infected is increasing now, but was decreasing for around a month ago, then the death rates could be relatively constant. However, if the number of infected continues to increase, at one point the death rates will increase as well.
The virus may indeed mutate into a less lethal variant with time. That may be good news, however, that new mutation will likely also be more contagious. So hard to tell.
A lot of the increase in the number of infections during the last period, is because more are being tested. During the summer, that was probably an important factor. However, lately the number of infections are increasing faster. What looks like exponential growth. And the number of tested is more constant these days. So this last surge of infected is probably a real, and relatively recent, increase. And the death rate has not yet increased much, since this is recent. Few of the newly infected have died yet, and there was an underlying decrease due to the decrease in infections one or two months ago.
You have to look at the death rates and infection rate and see how one follows the other. The exact way to model this is hard. You'd need to know not only the mean or median time it takes from infection to death, but the actual probability distribution. Also, the number of reported new infections depends on several factors, like how many tests are conducted. So this number is uncertain, unfortunately.
As for the 20%, I believe that was in the period after the first surge in USA until early October. The part of the year before this was not included. Until that first surge, approximately as many died this year as an average year. But after the first Corona surge, about 20% more than usual have died. It is natural to attribute this to direct and indirect effect of Corona.
The T'zak Ryn offers Naval Combat Solutions for the Quality Conscious Customer