4 ms·

As we do serological testing, it is being revealed that the death rate was even lower than previously imagined -- potentially by a factor of 50x [1]. [1] https

As we do serological testing, it is being revealed that the death rate was even lower than previously imagined -- potentially by a factor of 50x [1]. [1] https://www.nature.com/articles/d41586-020-01095-0 https://www.nature.com/articles/d41586-020-01095-0 Most places (outside of super dense metros) actually don't need quarantine at all. Maybe for people working in eldercare. With these new statistics the risk of death for those <65 is about the same as the normal risk of random death through other circumstances over the course of 2 months.

elgenie 7y ago
That preprint [1] is a great example of how far ahead of the skis you can get with motivated reasoning, bad math, and no peer review. They recruited people over Facebook at a time that California's COVID tests were backlogged (with ads not presented in the paper, with links to signup that could, anecdotally, be shared out); in that situation, avoiding selection bias towards people that were sick with something previously and want to know about potential immunity would be rather tough. (Per WTO protocol, they asked about symptoms in their sample but poof for some reason that data isn't in the paper). They got a 1.5% positive rate in their sample with 50 individuals testing positive. They then immediately applied demographic weighing to move that raw number to around 3%. Only then did they contemplate the false positive rate, taking the data of the manufacturer and 30 samples as a point estimate to figure out a 95% CI of positives around that 3%. That's a no-no for correctly propagating uncertainty. The statistically legitimate way to do that would've been to generate the 95% confidence interval for belief about the test's false positive rate and carry that forward in the demographic reweighing. That basic math would've avoided the frankly laughable result that a randomly selected person in Santa Clara county could have a 5% chance of having had COVID-19 while in the same county people seeing a doctor and getting tests through their doctor are only around 10% positive for COVID-19 [2]. Another critique: [3] [1] https://www.medrxiv.org/content/10.1101/2020.04.14.20062463v1 https://www.medrxiv.org/content/10.1101/2020.04.14.20062463v... [2] https://www.sccgov.org/sites/covid19/Pages/dashboard.aspx#testing https://www.sccgov.org/sites/covid19/Pages/dashboard.aspx#te... [3] https://medium.com/@balajis/peer-review-of-covid-19-antibody-seroprevalence-in-santa-clara-county-california-1f6382258c25 https://medium.com/@balajis/peer-review-of-covid-19-antibody...
- stanfordkid 7y ago
  Wow -- really enjoyed the write up by Balaji S. -- I had heard about these flaws earlier, but didn't realize how great they really were. I still maintain the gist of my point though -- we don't have sufficient evidence for widespread quarantine. New York and Italy are outliers, and should be understood as such. Even if the results are 5x or 10x instead of 50x -- it still means decisions were made on grossly inflated worst case data.
  - 52-6F-62 7y ago
    There is some data. https://toronto.citynews.ca/2020/04/20/new-modelling-shows-peak-in-community-spread-of-coronavirus-growth-in-long-term-care-homes/ https://toronto.citynews.ca/2020/04/20/new-modelling-shows-p... Ontario is largely a rural province. > New modelling predictions show that the wave of new, community spread cases of coronavirus in Ontario appears to have peaked, but the spread in long-term care homes and other group settings is still growing. > Earlier models predicted the peak would be hit in May, however according to data released by the province’s Ministry of Health, public health measures and social distancing have brought it forward. In spite of this we are not yet moving to reopen, rather deciding not to cut the parachute just because it started to slow our fall.
    - stanfordkid 7y ago
      Totally... no one is saying that we get off scot-free if we stop quarantine. The quarantine can't last forever, and the virus at some point will spread... so even if you are at-risk if you don't get it now you might get it 6 months from now -- unless of course, younger folks get out there and build herd immunity.
- stanfordkid 7y ago
  I'll add that this high rate is mirrored in several countries doing anti-body tests on blood donors (though one could argue blood donors are a self-selecting group): >In this sample 11% has antibodies and the test detects 70-80% of people with antibodies. So, that leaves us with 14.6% immune. With a population of 2,377,081, that means 347053 people would be immune in Stockholm now, for an IFR of 0.27%. [1] https://www.reddit.com/r/COVID19/comments/g4znbg/at_least_11_of_tested_blood_donors_in_stockholm/ https://www.reddit.com/r/COVID19/comments/g4znbg/at_least_11...

elgenie 7y ago

That preprint [1] is a great example of how far ahead of the skis you can get with motivated reasoning, bad math, and no peer review. They recruited people over Facebook at a time that California's COVID tests were backlogged (with ads not presented in the paper, with links to signup that could, anecdotally, be shared out); in that situation, avoiding selection bias towards people that were sick with something previously and want to know about potential immunity would be rather tough. (Per WTO protocol, they asked about symptoms in their sample but poof for some reason that data isn't in the paper). They got a 1.5% positive rate in their sample with 50 individuals testing positive. They then immediately applied demographic weighing to move that raw number to around 3%. Only then did they contemplate the false positive rate, taking the data of the manufacturer and 30 samples as a point estimate to figure out a 95% CI of positives around that 3%. That's a no-no for correctly propagating uncertainty. The statistically legitimate way to do that would've been to generate the 95% confidence interval for belief about the test's false positive rate and carry that forward in the demographic reweighing. That basic math would've avoided the frankly laughable result that a randomly selected person in Santa Clara county could have a 5% chance of having had COVID-19 while in the same county people seeing a doctor and getting tests through their doctor are only around 10% positive for COVID-19 [2]. Another critique: [3] [1] https://www.medrxiv.org/content/10.1101/2020.04.14.20062463v1 https://www.medrxiv.org/content/10.1101/2020.04.14.20062463v... [2] https://www.sccgov.org/sites/covid19/Pages/dashboard.aspx#testing https://www.sccgov.org/sites/covid19/Pages/dashboard.aspx#te... [3] https://medium.com/@balajis/peer-review-of-covid-19-antibody-seroprevalence-in-santa-clara-county-california-1f6382258c25 https://medium.com/@balajis/peer-review-of-covid-19-antibody...

stanfordkid 7y ago
Wow -- really enjoyed the write up by Balaji S. -- I had heard about these flaws earlier, but didn't realize how great they really were. I still maintain the gist of my point though -- we don't have sufficient evidence for widespread quarantine. New York and Italy are outliers, and should be understood as such. Even if the results are 5x or 10x instead of 50x -- it still means decisions were made on grossly inflated worst case data.
- 52-6F-62 7y ago
  There is some data. https://toronto.citynews.ca/2020/04/20/new-modelling-shows-peak-in-community-spread-of-coronavirus-growth-in-long-term-care-homes/ https://toronto.citynews.ca/2020/04/20/new-modelling-shows-p... Ontario is largely a rural province. > New modelling predictions show that the wave of new, community spread cases of coronavirus in Ontario appears to have peaked, but the spread in long-term care homes and other group settings is still growing. > Earlier models predicted the peak would be hit in May, however according to data released by the province’s Ministry of Health, public health measures and social distancing have brought it forward. In spite of this we are not yet moving to reopen, rather deciding not to cut the parachute just because it started to slow our fall.
  - stanfordkid 7y ago
    Totally... no one is saying that we get off scot-free if we stop quarantine. The quarantine can't last forever, and the virus at some point will spread... so even if you are at-risk if you don't get it now you might get it 6 months from now -- unless of course, younger folks get out there and build herd immunity.
stanfordkid 7y ago
I'll add that this high rate is mirrored in several countries doing anti-body tests on blood donors (though one could argue blood donors are a self-selecting group): >In this sample 11% has antibodies and the test detects 70-80% of people with antibodies. So, that leaves us with 14.6% immune. With a population of 2,377,081, that means 347053 people would be immune in Stockholm now, for an IFR of 0.27%. [1] https://www.reddit.com/r/COVID19/comments/g4znbg/at_least_11_of_tested_blood_donors_in_stockholm/ https://www.reddit.com/r/COVID19/comments/g4znbg/at_least_11...