What's interesting is that the rate of negative results seems to be roughly constant. I would expect that with an expansion of testing we would see many more ne

What's interesting is that the rate of negative results seems to be roughly constant. I would expect that with an expansion of testing we would see many more negative results as a proportion (if we were allocating tests early only to the most likely cases and are now expanding the reach of testing)