Something to Consider for Engineers Among Us

[Handy Artie]

https://www.aier.org/article/lockdowns-of-young-people-lead-to-more-deaths-from-covid-19/

 

[Crissa]

Further, the results in Sweden were not a catastrophe. As of October 10, the total mortality from Covid-19 was 583 deaths per million population.

I don't believe this is 'data' but 'opinion'. And not a good one, either.

'Not a catastrophe' says the guy lauding the death rate in the country with the 15th highest death rate. https://www.statista.com/statistics/1104709/coronavirus-deaths-worldwide-per-million-inhabitants/

This is why you don't get epidemiology from economics guys who make repeated assertions. And don't support their assertions with anything.

-Crissa

 

[ajdelange]

I believe that if you and I run into one another the probability that one of us will be infected by the other is 2*pi*pc where pi is the probability that an individual is infected and pc is the probability that our contact is close enough to transfer infection. 2*pi(1-pi) ~ 2pi is the probability that at least one of us is infected but not the other. I believe that 1 - 2*pi*pc is the probability that I will not be infected in an encounter with you and that (1 - 2*pi*pc)^M is the probability that I will not be infected in M such encounters and that 1 - (1 - 2*pi*pc)^M ~ 2*M*pi*pc if pi*pc is small is the probability that I will be infected. If that's the probability of an individual being infected then the expected number of new infections in a population of N persons is approximately 2*N*M*pi*pc. So that says that a place with few people is the place to be and that wherever you are you should minimize your contacts and limit them to persons unlikely to be infectious and that you should minimize pc by keeping distance, wearing a mask, keeping contacts short, washing hands, meeting outdoors etc. So there's the math that supports what common sense should already tell you. The problem is that common sense isn't very common any more. If you allow people to come together in populous areas and they become casual about pc then rates are going to go up. Duh!

Now, of course, the nuance comes in when we consider that pi isn't a simple constant but that it changes as people contract the disease and become immune or die from it or as a large infected sub population (migrant fruit harvest workers in Vermont, for example) moves into the area and so on.

Anyway, yes, I believe the data but I have done enough modeling in my lifetime to know what the limitations of models are and understand that statistics apply to ensembles - not individuals. The simple math makes it clear what I should do and the supplied data, interpreted properly, is sufficient to support the simple model.