t-threads
private beta
search about

Assistant Professor of Biostatistics at @UF specializing in emerging infectious diseases and vaccine study design. @HarvardBiostats PhD. Tweets my own. She/her.

Jun 17, 2020 2:40 PM
copied to clipboard

In our continuous search for metrics to track, I’d like to see more discussion of median age of new diagnoses. Lower median age suggests: (1) we are better protecting the elderly, (2) we are testing enough to capture infections in the age range most likely to be missed.


This is not a perfect example, but we can compare South Korea and Canada, and look at the difference in cases in the elderly. My thought is to use more recently reported data (not the entire epidemic). https://www.statista.com/statistics/1107149/covid19-cases-age-distribution-canada/ https://www.statista.com/statistics/1102730/south-korea-coronavirus-cases-by-age/

tweet media tweet media


We saw this earlier in the epidemic, that countries that are doing a lot of testing have an age distribution that is shifted towards the younger ages. Where we are only testing the most severely ill, the distribution is shifted towards the older ages.


This is just one dimension to the numbers, on top of everything else (Rt, test positivity, numbers hospitalized). Of course an explosive outbreak in young people is bad. But it gives some accountability to people saying they can selectively protect the elderly.