Using random testing to manage a safe exit from the COVID-19 lockdown

Markus Mueller; Peter Derlet; Christopher Mudry; Gabriel Aeppli

The idea of statistical testing is to randomly select people 5 and test them for infectiousness. 6 5 It is important that the set of randomly selected people must change constantly, so that it should happen extremely rarely that a given person is tested twice. 6 Here, we solely focus on a person being infectious, but not on whether the person has developed antibodies. The latter test indicates that the person has been infected any time in the past. Serological tests for antibodies and (potential) immunity have their own virtue, but aim at different goals from the random testing for infections that we advocate here. By following the fraction of infections as a function of time, we can determine nearly instantaneously the growth rate of infections, k(t), and thus assess and quantify the effectiveness of socio-economic restrictions through the observed changes in k following a change . CC-BY-NC-ND 4.0 International license It is made available under a author/funder, who has granted medRxiv a license to display the preprint in perpetuity.