Rt vs percent infected, and Cases per Fatality vs percent infected, for all 50 states. Faint lines on the Cases per Fatality plot are isocountours of equal disease burden, and are spaced apart at 2x changes in burden. A methods thread 1/
For these plots, percent currently infected is 3.8 times the number of new cases in the past 10 days. 3.8x is the difference between national CFR and CDC claimed IFR. It is an estimate scaled from cases per capita. 2/
Rt is a 10 day windowed sum of new cases, ratioed to a similar sum 14 days earlier, and then pro-rated to a 5.2 day period. A 6.5-7.5 day period would be better, but I am staying consistent and using an April number. 3/
Cases per Fatality is a 14 day sum of new deaths, ratiod to a 14 day sum of new cases time shifted 20 days earlier. It is a laggy measure!!! You can remove the lag by relating it to the fraction of new cases over age 65 (an exercise for the reader). 4/
Cases per capita, and Cases per Fatality, together, capture ALL the cross-region variance in disease burden measured by death (by definition). On the Cases per Fatality plot, I draw faint lines at each 2X change in disease burden. They are curved. 5/
The implication is that states with high Cases per Fatality will be over-represented when people try to find states in trouble using cases per capita. This prediction is born out particularly well. What is Cases per Fatality? 6/
Because SARS-COV-2 is so age sensitive, Cases per Fatality is sensitive to the ages of people getting infected. If you have nursing home problems, or an older age demographic (larger % over age 65), you will have a lower Cases per Fatality. 7/
Cases per Fatality also goes up when you test the less vulnerable (younger) people better, and find infections better. Good testing finds infections. Low test positivity is demonstrably good at finding people who are NOT infected. Which one do you want in your state? 8/
Any reasonable comparison of disease burden across regions REQUIRES considering both cases per capita and cases per fatality. 9/9
Share this Scrolly Tale with your friends.
A Scrolly Tale is a new way to read Twitter threads with a more visually immersive experience.
Discover more beautiful Scrolly Tales like this.