Epidemiology

Today, dear reader I am going to play an epidemiologist. Given how poorly they have performed to date, I don’t see how I could do any worse. I also realize that many of my assumptions here will be wrong ... but, hopefully, they will be offsetting errors.

Way back on February 5th, I published a blog that, using data from a Taiwan newspaper, I estimated that China’s coronavirus numbers were off by a factor of 10 for reported cases ... and by a factor of 100 for deaths ... see: Straight Skinny. (I am now backing off this 100x mortality factor and now estimate it was more like 50x.) Current “official statistics” have China cases at 81,907 and deaths at 3,336. We know that these are fudged numbers and that the real statistics should be more like 820,000 cases and 115,000 deaths. One more adjustment: a German investigation has shown that there are about three asymptomatic cases for every symptomatic one ... see: Reason.com Article. Ergo, China’s actual cases were likely more like 2,460,000. This would place China’s mortality rate at 4.5% ... close to what it allows in its “official” numbers.

Now let is look at things from a different angle:

Wuhan, China (an urban area of some 21 million people) now seems to achieved herd immunity since things have been reopened there completely.

Coronavirus might produce herd immunity at about a 33% penetration rate (a number developed during previous pandemics)

Meaning 7 million people in Wuhan might have been infected ...including asymptomatics

Assuming 66% were asymptomatic (German study) ... or 2.3 million exhibited symptoms

Assume  mortality rates for symptomatic coronavirus patients approximates 1% ... 10x seasonal flu rates

Meaning that there had been about 230,000 deaths in Wuhan

So one model says 115,000 actual deaths ... the other one comes up with 230,000 deaths ... exactly twice that number. Let’s split the difference ... guessing the actual number of Chinese deaths was 172,000! This is a little over 0.01% of the total Chinese population.

This would suggest that we might have that same mortality proportion here in the US ... with similar mitigation rules as China ... producing around 52,000 (+ or - 17,000) COVID-19 deaths ... or even more on the downside due to our better testing and therapeutics. We’ll see if I should take up this new profession ... or slink away in shame!

Afterward: Dt. Oz said on TV last night that herd immunity took 65-70% being infected. My numbers might still work out if one assumes that  9 or 10 million fled Wuhan before it was shut down.