3 ms·

> It's probably better to have a 50% hospitalization rate and an R value of 0.8 than a 5% hospitalization rate and an R value of 2. After 3 'generations' of spr

> It's probably better to have a 50% hospitalization rate and an R value of 0.8 than a 5% hospitalization rate and an R value of 2. After 3 'generations' of spreading the second situation already has more hospitalizations. Sure, but those aren’t the only choices. Over the last 30 days the daily case average has increased by 500%. During that same time the daily average hospitalized has risen 100%. And the test positivity rate has increased to almost 30%, which implies an almost comical undercount in actual cases. The contagiousness and severity are diverging in an extraordinary way. We’re likely undercounting cases by 1/5 or 1/10th.

ianhawes 5y ago
The testing positivity rate is relatively useless at this point. For example, in my State, there is no way to report a positive test from a rapid result. Only PCR tests administered by a medical professional are considered valid for statistical purposes, even though the government is giving out free rapid tests.
- parkingrift 5y ago
  I don't think it's useless. I think it's a pretty good indication that there are many multiples of actual cases for every confirmed case. If covid cases have risen 5,000% in 30 days, but hospitalizations have only risen 100%, that's an extraordinarily positive outcome.
- thomaslord 5y ago
  This is for good reason. Rapid tests are substantially less sensitive and often give false negatives, especially for asymptomatic cases. There are also a lot of rapid "at home" tests out there that have high false positive rates, so even if non-PCR tests were added to the data the only reasonable approach would be to still require them to be administered by a medical professional to make sure the test is correctly administered and that the test itself is relatively high quality.
  - ianhawes 5y ago
    The idea that rapid tests are less sensitive and therefore not valuable is summer-2020 thinking and outdated. Rapid tests are perfectly fine and should be counted by health departments. Case in point. My family of 4 all tested positive on rapid tests. 3/4 of us had symptoms. None of us received PCR tests because the next available appointments (other than a 6 hour wait at a state run testing center) were a week away. Thats 4 people not counted. And we were part of an outbreak that included 3 other families that all tested positive via rapid tests.
TobTobXX 5y ago
Unknown cases don't use up ICU and hospital beds and don't result in deaths, so who cares?
- parkingrift 5y ago
  It's pretty important to determine actual severity rates.

ianhawes 5y ago

The testing positivity rate is relatively useless at this point. For example, in my State, there is no way to report a positive test from a rapid result. Only PCR tests administered by a medical professional are considered valid for statistical purposes, even though the government is giving out free rapid tests.

parkingrift 5y ago
I don't think it's useless. I think it's a pretty good indication that there are many multiples of actual cases for every confirmed case. If covid cases have risen 5,000% in 30 days, but hospitalizations have only risen 100%, that's an extraordinarily positive outcome.
thomaslord 5y ago
This is for good reason. Rapid tests are substantially less sensitive and often give false negatives, especially for asymptomatic cases. There are also a lot of rapid "at home" tests out there that have high false positive rates, so even if non-PCR tests were added to the data the only reasonable approach would be to still require them to be administered by a medical professional to make sure the test is correctly administered and that the test itself is relatively high quality.
- ianhawes 5y ago
  The idea that rapid tests are less sensitive and therefore not valuable is summer-2020 thinking and outdated. Rapid tests are perfectly fine and should be counted by health departments. Case in point. My family of 4 all tested positive on rapid tests. 3/4 of us had symptoms. None of us received PCR tests because the next available appointments (other than a 6 hour wait at a state run testing center) were a week away. Thats 4 people not counted. And we were part of an outbreak that included 3 other families that all tested positive via rapid tests.

TobTobXX 5y ago

Unknown cases don't use up ICU and hospital beds and don't result in deaths, so who cares?

parkingrift 5y ago
It's pretty important to determine actual severity rates.