Huh. That seems... bad. I would think a test threshold is a tradeoff between false positives and false negatives (alpha and beta error) and that you would want to really concentrate on reducing false negatives, which result in sick people spreading contagion, even at the cost of increasing false positives, which are inconvenient and maybe generate more costs for unnecessary follow-on treatment.
A test with no false negatives sounds to me like it is calibrated wrong.
we've had a fair number of people that had to be re-tested, possibly because quality control metrics were not met for a particular batch of samples/tests -- I'm not sure of the specifics because i'm not involved in the testing.
False negatives is a concern to me because someone can be infected but not shedding virus in sufficient quantities yet.
One interesting thing is there are distinct regional differences in the virus genome in the US. It looks like like the virus made its way into the US from multiple sources and you can trace that lineage.