A test has to be reliable against both kinds of error, and, under the usual standards of p-hacking, these are. "The Abbott BinaxNOW test gives an answer in fifteen minutes. It is reported to be 97.1% accurate for COVID-positive persons and 98.5% for negatives. This sounds really good. And, as medical tests go, it is." Stronger power against false negatives is particularly desirable, but even at any one test being 98.5% accurate, as soon as fifty people have been screened and allowed inside, it's better than a coin toss that a Wuhan coronavirus carrier got in. Develop your intuition.
This doesn’t mean the test is bad. It’s actually a very good test. But testing can’t be effectively used for screening. You need 100% reliability and specificity for screening, and no test is that good. When you have to deal with a lot of people, you will get exposed. You just can’t avoid it. It’s in the wild.Yes, and with the testing still focusing on people who have symptoms, it's unlikely anyone will ever get a handle on how many people might have already been exposed, their bodies shut it down (for a variety of reasons) and they're immune for the time being.
The only real question remaining is what happens after the virus bites you, because it will. And that answer is, at best, very uncertain. We know that between 50% and 60% of the population has “T-Cell Immunity.” This appears to be acquired by historical exposure to related viruses. If you are under age 60, you really don’t have much to worry about. But if you’re high risk, get to the doctor soonest. Let him worry about the latest literature on treatments.I suspect that the people who make policy will figure out, I hope sooner rather than later, that younger people ought not be confined to quarters or deemed non-essential workers.
No comments:
Post a Comment