You Need to Check the Experts’ Math

This offers a perspective on covid survival rates, but screws up some simple statistics:

0-19	20-49	50-69	70+
100.000%	100.000%	100.000%	100.000%
-99.997%	-99.98%	-99.5%	-94.6%
0.003%	0.02%	0.5%	5.4%

It’s official data.  It purports to be from CDC.  The author implies possession of a MD.

The math is screwed up.  By a factor of 100.  I learned the difference between decimals and percentages in the fifth or sixth grade – this isn’t a mistake at a graduate stats level, or even freshman stats. It appears someone releasing official data screwed up.  We need to check the math even on official data.

This site https://lincolnmtcovid.com/ has local numbers – and you can contrast them against the CDC statistics:

The local numbers show some anomalies when we compare and contrast them with CDC statistics.  The Libby area shows a cumulative 1,190 cases (in a population of 9,772  that’s 12.2%).  North County shows 467 cases (in a population of 6,470 that’s 7.2%) and Troy shows 258 cases (in a population of 3,435 that’s 7.5%). 

Lincoln County death rates can’t be contrasted with the CDC percentages – the tyranny of small numbers makes it impossible.  That said, in the 70+ age range that the CDC figures identify as a (corrected) 5.4% infection fatality rate, Lincoln County’s charts show 24 deaths in 311 cases – 7.8% – 44% more fatalities than national statistics.  The 3 deaths in the 50-69 age range, with 557 total cases work out amazingly close to the national 0.5% infection fatality rate.

There’s not enough data for me to infer causality.  It is good to have local data available – and I do wonder why the infection rate is higher in Libby.  Checking the math when you can is a good idea.