The journalists here haven't mentioned what kinds of tests each the 4 unnamed sources – 2 doctors, 2 citizens – had at each stage of their 'positive then negati

The journalists here haven't mentioned what kinds of tests each the 4 unnamed sources – 2 doctors, 2 citizens – had at each stage of their 'positive then negative then positive again' journey. They haven't mentioned if any of the individual tests were followed-up with multiple confirmatory tests of alternate methods. To the extent they mention test types at all, it is quotes from "February" or "a professor... by email" – nothing about these persons' tests. Yet, the tests in use have some level of false-positive and false-negative. Some variants have had quite high error rates – which may still be acceptable for mass-screening, but not for understanding the course of a single person's infection status. So, among many millions of people tested in China, a handful have had alternating results not correlated with their symptoms? Not a surprise, even with tiny test error rates, unless each contributing result to the "+/-/+" pattern was reconfirmed by multiple-tests/multiple-methods. (Note this also suggests many accounts of 'asymptomatic cases' may just be false-positives who never had Covid-19.)