Yes, it's a single test, so you take the results along with the type of patient you are testing. If the probability of that person's having COVID is low—the person is asymptomatic and is feeling fine—and the test is negative, yes, the sensitivity says that we might miss a few of them and that that negative might actually be a positive, but the probability of that person going in being positive isn't that high to begin with.
You have to combine those things. There is also the use of serial testing, similar to what they're doing in Calgary—day two and day seven. That also helps with increasing that sensitivity. Even if you got caught too early on that first test, you still have time to get a viral load that's higher on the second test and get picked up.
Again, these aren't perfect tests. You'd treat them as positives, but you'd still do a PCR on them. We have other models of infectious diseases where we do a screening test and a gold standard test for positives. There certainly are ways to make it work such that you don't necessarily overcall positives.
In terms of the negatives, yes, it is the right context. Serial testing helps, but you could certainly miss a couple of positives here and there. It's much less likely in people whom you don't suspect as being positive to begin with, though. If I'm in an emergency room testing people and I get a negative for someone who has a fever and a cough, I'm not going to rely on that result. If I'm walking out on the street and I swab someone and it is a negative, and their probability of having COVID is zero and they're feeling fine, then, yes, I'm going to rely on that negative as a real result.