Law Professor Seth Chandler is truly among the most brilliant individuals that I have ever met. I consider my B+ in his Civil Procedure class back in 1994 to be one of my all-time greatest academic accomplishments.
Chandler has several areas of expertise. Two of them are a) health law; UHLC has one of the nation’s top programs in that field, and b) the use of mathematical modeling in the law; he’s a bit of a pioneer in that field; he started doing that back in the mid 90s at a time when hardly anyone was doing it.
He has used his mathematical modeling skills to devise a new model to estimate the number of COVID 19 infections in the USA.
Based on his model, the rate of COVID 19 infection in the USA is about FOUR TIMES what is being reported. He regards the currently quoted number of 650,000 active infections to come from more than 2.5 million+ ACTUAL infections.
If there is anyone here who is good at this sort of thing (I’m certainly not), maybe you can look at his mathematical model, a link to which is provided in the article, and analyze/critique it. Thanks!
Here’s the link to Chandler’s mathematical model. That stuff is all completely over my head, so someone with a little more expertise in that area will have to look at it and sort it out.
I can believe that since testing percentage is so low and some infected persons show no symptoms.
What doesn’t get much press is the positive rate for people being tested. Its 20%. This thing has already spread.
I’m interested what the antibody studies show. The early test are hard to believe, but if true then you would be right, the disease has spread through the population. Not sure how long it will take to prove out but this where i see the most hope.
I appreciate the academics especially from UH. It validates what we already knew. Plenty of people are home with COVID19 and recovered without a test and without a doctor / hospital visit.
the 20% positive rate makes sense because testing is currently limited (in most places, like Minnesota) to testing people who need to be hospitalized for their symptoms or symptomatic healthcare workers with exposure. This significantly increases the pretest probability of a positive.
there is a pre-peer-review, prepublication Stanford antibody study that has the infection rate at 2.5-4.2% of the population, but there area few problems with the design of the study that could make it artificially high. Will be interesting to see where it lands after peer review. Im guessing somewhere near the lower boundary of their range.
Also, there have been some anti-body tests approved that would give positives based on other sars viruses so would be wrong. It is a side effect of the FDA lessening their restrictions for approval to try and hurry development.