Dr. Cillian De Gascun:

Of the three main approaches to testing at this point in time, as Deputy O'Rourke said, the gold standard is PCR. It is important to highlight that at this point we have the best test available. It is the most sensitive and the most specific. There is a lot of interest in antigen testing on the basis of the facts. PCR, for clarity, identifies the genetic material of the virus, or the viral RNA. What antigen testing looks for are viral antigens, typically virus proteins. These proteins tend to be present in people in whom the virus is actively replicating. They tend to be less sensitive purely because there is not necessarily as much protein there as there might be viral RNA. The difficulty with the WHO criteria at this point is that they refer to a sensitivity of 80%. The WHO is speaking to the whole world, not just to developed countries that can afford to have a PCR testing system in place. The problem with a sensitivity of 80% is that, in essence, one in five cases will be missed, and that is not good enough. Therefore, in order to compensate for that lack of sensitivity, there are a couple of options. These tests can potentially be used in an outbreak setting, whereby one tests a large number of people who have symptoms and then the risk of getting lots of false negatives is significantly reduced. If, however, one does identify somebody who is part of that outbreak or cluster in whom the antigen test does not detect virus antigen, one needs to go on and do a PCR test to facilitate the contact tracing process after that. The other alternative approach to compensating for the lack of sensitivity is repeated testing, which is another area in which people have expressed an interest.

The problem is we have no evidence at this point in time that antigens work in that setting. For example, if I take an antigen test today and it is not detected-----