Rapid tests are already being used to screen people without symptoms for the coronavirus — even though they're not cleared for this purpose, and the data in asymptomatics is sparse.
Some of that data is emerging. It might not be what some hoped. 1/
nytimes.com/2020/11/02/hea…
Some of that data is emerging. It might not be what some hoped. 1/
nytimes.com/2020/11/02/hea…
Is there still a role for rapid tests? Absolutely. More data will be needed to figure out where they fit in best. But for now, it's crucial for people to understand that not all coronavirus test negatives are created equal. 2/
Testing negative on less sensitive tests, for example, might not mean you're virus-free.
Could it mean you're not infectious? Maybe. That's really, really hard to test.
It's very likely to be the case that people will less virus in their bodies are less infectious. 3/
Could it mean you're not infectious? Maybe. That's really, really hard to test.
It's very likely to be the case that people will less virus in their bodies are less infectious. 3/
But infectiousness isn't a binary; there's no hard cutoff. And there will probably be exceptions to that trend. Transmission is also not easy to measure. 4/
When PCRs and rapid tests disagree, it opens up a lot of questions. Which test was "wrong"? It sort of depends what question you were asking. Was the person infected? Was the person posing a huge risk to others? Test results are a snapshot; it's hard to extrapolate from them. 5/
Say you're PCR positive but negative on a rapid antigen test. Are you at the very start of an infection, or at the very end? Those could mean very different things for your own health, and the risk you pose to those around you. 6/
Here's where data from repeat testing could be really useful. There isn't much of that yet.
This all gets a lot harder if rapid tests aren't as useful in people without symptoms. 7/
This all gets a lot harder if rapid tests aren't as useful in people without symptoms. 7/
(It's not that the test changes or that the virus changes when there aren't symptoms; it's that the population you're looking at has so much more variation.) 8/
Interested to see what happens next, but as many wonderful experts told me, this is why it's so crucial to collect data and think hard about it; to walk, not run, when it comes to rolling out testing, as @DienBard said. 9/
Thank you also to @omaigarner @SBtotheDub @Linoj_S @BenMazer and others who chatted with me about this data. 10/10
• • •
Missing some Tweet in this thread? You can try to
force a refresh
