Interesting paradox. You suggest that a false "positive" is totally unacceptable, for some very good reasons. However, what if there was no such thing as a "positive" result reported? If the result was instead reported as "suspect" or "possible", with the instruction to get to a proper testing centre for a definitive result? Do you think that would be just as undesirable, in that people would jump to the conclusion that "suspect" meant positive anyway?
The reason I ask is that I tend to view sensitivity as more important in screening tests in general. Because, if the negatives are entirely reliable, then you have very significantly reduced the number of patients that need to be put through the professional testing system. And the patients who are presented include the ones who need attention.
Conversely, if it's the negatives that aren't wholly reliable, have you gained much? If all the negatives have to proceed to further testing, then you probably haven't cut down the professional lab workload by much. Plus, the people you don't see (the positives) are exactly the people you really want to get into the system.
Having said that, though, the sensitivity and specificity figures quoted about for this test were very impressive. There were indeed no false positives being reported, and the percentage of false negatives was so small that it's likely very very few infected patients would be missed.
So in this case it really does seem to be a sociological and psychological argument.
Rolfe.