Thread by @hhesterm

A small thread on comparing covid numbers across borders - and why it has become more difficult rather than easier (thread)

First up: how many cases are there. The number of choice is the incidence rate. Cases per 100,000 on a 7-day average (or something along those lines). /2

Asymptomatic cases have meant that these numbers can differ with the number of tests (and this fact in and of itself can also be used as an excuse). However, you can compensate to some extent with the positivity rate of tests. But can you still? /3

Just as an example: country A hands out loads of free tests for school kids (in some German states 3 LFT per kid per week). Country B does official tests. Result? /4

In country B the positivity rate will be low, in country A asymptomatic kids will only ever get an official test entering statistics if the at-home LFT was a positive, resulting in a far higher positivity rate. /5

Apparently there are also official tests where samples of a whole class are taken and mixed and individual tests are done if that class sample was positive. Good luck trying to see how this messes with statistics. /6

But measuring successful vaccination campaigns has also become more difficult. In the past you could look at the vaccination rate and say: hooray: 70% vaccinated and compare to see whether that‘s good/bad/fast/slow /7

With boosters that‘s more problematic. Boosters are given some months after the original vaccines. A perfect booster campaign following a slow original campaign might be slower than a mediocre booster campaign after a fast original campaign /8

And then of course there‘s the question which vaccine was given in the original campaign, something that might have an impact on the period that has to elapse between original vaccination and booster.