To control for country/region-specific observation biases, we included the length of time (measured in days) since the first confirmed COVID-19 case in each country/region and the number of COVID-19 tests conducted (as a measure of sampling effort) as covariates.