This article mentions, off hand, that the study was run with an antibody test where the false positive rate is unknown.

This seems to me to be a PRETTY BIG problem. https://twitter.com/latimes/status/1252331574215000067

(Or more specifically: they say there are “concerns” about the false positive rate.)

If you’re trying to measure a small number in a population with no inherent pre-test probability (that is, no reason to think any one of them were infected), a small number of false positives can absolutely dwarf the result.

I just really want to wait for the information about false positives before relying on a study like that, because even 1-2% false positives means that the result is completely off.