Once concern I've had since the beginning of COVID19 showing up in the US is that some number of our tests should be dedicated to randomized testing of the population. Without any idea of what the level of "background radiation" exists  any data from testing of symptomatic or "exposed" individuals is crippled by our complete ignorance of the "underground" prevalence of the disease in the general population. Our understanding and conclusions drawn from our limited data set are bound to lead to faulty conclusions of how best to deal with the disease. I'm still surprised that I have not heard or read of a single epidemiologist proposing or referring to randomized survey of COVID 19 infection.