3 ms·

It's important to remember that the positivity rate is a measure of how much testing you're doing, not a measure of how much virus is around. Unless you're read

It's important to remember that the positivity rate is a measure of how much testing you're doing, not a measure of how much virus is around. Unless you're reading articles about the politics of rapid tests or the like, you should mostly just ignore the pos rate, it doesn't tell you anything useful otherwise.

theli0nheart 6y ago
If one was coming to this conclusion from a single data point, you might be right, but the OP is referring to the period-over-period change. This is a significant trend.
- onychomys 6y ago
  But only if the testing rate (which is the denominator) stays the same. And it never does, because as fewer people get sick, fewer people get tested.
40four 6y ago
I'm reading numbers from my states department of health site, and they only report molecular tests. You are right percent positive, and even daily new cases, it's not a measure of how much virus is around, and should taken with a grain of salt. But they are some of the only metrics we have, and if you keep the caveats in mind, I don't agree that they aren't useful. What else would you suggest looking at? Deaths? In my state, the daily new deaths numbers are following the same trend. So if you keep in mind percent positive is heavily influenced by total number of tests, and the daily new deaths chart is following the same trend, then that seems like very positive new to me.
- onychomys 6y ago
  Oh, don't get me wrong, I'm very happy with the way all our numbers are heading. I just don't quite understand how it is that in the US we've come to latch onto the positivity rate as a meaningful measure of stuff. In the UK and Europe, they routinely report the R0 (...or other related measures), which is a zillion times more useful when thinking about how the pandemic is going.
jkubicek 6y ago
I think you may have this backwards? As testing rates increase, we would expect the positivity rate to go down, as you are now testing more people who have never contracted COVID. The more testing we're doing, the more we expect the positivity rate to approach the actual percentage of infected folks. If you're only testing people in hospitals and reporting systems, you're going to have a much higher positivity rate.

theli0nheart 6y ago

If one was coming to this conclusion from a single data point, you might be right, but the OP is referring to the period-over-period change. This is a significant trend.

onychomys 6y ago
But only if the testing rate (which is the denominator) stays the same. And it never does, because as fewer people get sick, fewer people get tested.

40four 6y ago

I'm reading numbers from my states department of health site, and they only report molecular tests. You are right percent positive, and even daily new cases, it's not a measure of how much virus is around, and should taken with a grain of salt. But they are some of the only metrics we have, and if you keep the caveats in mind, I don't agree that they aren't useful. What else would you suggest looking at? Deaths? In my state, the daily new deaths numbers are following the same trend. So if you keep in mind percent positive is heavily influenced by total number of tests, and the daily new deaths chart is following the same trend, then that seems like very positive new to me.

onychomys 6y ago
Oh, don't get me wrong, I'm very happy with the way all our numbers are heading. I just don't quite understand how it is that in the US we've come to latch onto the positivity rate as a meaningful measure of stuff. In the UK and Europe, they routinely report the R0 (...or other related measures), which is a zillion times more useful when thinking about how the pandemic is going.

jkubicek 6y ago

I think you may have this backwards? As testing rates increase, we would expect the positivity rate to go down, as you are now testing more people who have never contracted COVID. The more testing we're doing, the more we expect the positivity rate to approach the actual percentage of infected folks. If you're only testing people in hospitals and reporting systems, you're going to have a much higher positivity rate.