7 ms·

Makes it a bit harder to deny the validity of the Stanford/Santa Clara results. They're virtually the same.

Barrin92 7y ago
also in line with the dutch results a week ago (https://www.nytimes.com/reuters/2020/04/16/world/europe/16reuters-health-coronavirus-netherlands-study.html https://www.nytimes.com/reuters/2020/04/16/world/europe/16re...)
lalaland1125 7y ago
One issue is that the study design and probably analysis is very similar to the Santa Clara study (they are using the same test kits and have shared coauthors on both studies). It's not really independent replication when both are potentially using flawed kits and analysis.
tunesmith 7y ago
Looking forward to what the science heads on my twitter feed say about it. They were positively withering in their criticism regarding the Santa Clara study.
tempsy 7y ago
How do you explain the low number of hospitalizations and deaths in LA vs NYC? If these numbers hold then you’d expect NYC to be close to herd immunity given the difference in fatalities.
- TheBlight 7y ago
  See: https://news.ycombinator.com/item?id=22928754 https://news.ycombinator.com/item?id=22928754
  - tempsy 7y ago
    More questions than answers. It hasn’t been hot in LA so suggesting it has to do with heat doesn’t make sense imo.
scott_s 7y ago
I'm not sure if we can consider this independent support: both studies use the same antibody test. Any problems with the test itself (or our understanding of problems with the tests) will be present in both studies.
strangeloops85 7y ago
Well, a lot of the same people are involved in both..
dgacmu 7y ago
That's because the methods and authors have major overlap. The USC study reported here is a collaboration with the same group at Stanford: https://news.usc.edu/168810/usc-covid-19-antibody-researcher-answers-questions-about-testing-in-l-a-county/ https://news.usc.edu/168810/usc-covid-19-antibody-researcher... The participant selection here was done differently, but the same problems arise with the false positive rate of the tests.
microdrum 7y ago
No. They both have to be rejected in order to affirm our religiously held priors. The virus has IFR of 2% or more, millions of Americans will due, and we must stay in our homes.* * Not FedEx employees though! /s

also in line with the dutch results a week ago (https://www.nytimes.com/reuters/2020/04/16/world/europe/16reuters-health-coronavirus-netherlands-study.html https://www.nytimes.com/reuters/2020/04/16/world/europe/16re...)

lalaland1125 7y ago

One issue is that the study design and probably analysis is very similar to the Santa Clara study (they are using the same test kits and have shared coauthors on both studies). It's not really independent replication when both are potentially using flawed kits and analysis.

tunesmith 7y ago

Looking forward to what the science heads on my twitter feed say about it. They were positively withering in their criticism regarding the Santa Clara study.

tempsy 7y ago

How do you explain the low number of hospitalizations and deaths in LA vs NYC? If these numbers hold then you’d expect NYC to be close to herd immunity given the difference in fatalities.

TheBlight 7y ago
See: https://news.ycombinator.com/item?id=22928754 https://news.ycombinator.com/item?id=22928754
- tempsy 7y ago
  More questions than answers. It hasn’t been hot in LA so suggesting it has to do with heat doesn’t make sense imo.

scott_s 7y ago

I'm not sure if we can consider this independent support: both studies use the same antibody test. Any problems with the test itself (or our understanding of problems with the tests) will be present in both studies.

strangeloops85 7y ago

Well, a lot of the same people are involved in both..

dgacmu 7y ago

That's because the methods and authors have major overlap. The USC study reported here is a collaboration with the same group at Stanford: https://news.usc.edu/168810/usc-covid-19-antibody-researcher-answers-questions-about-testing-in-l-a-county/ https://news.usc.edu/168810/usc-covid-19-antibody-researcher... The participant selection here was done differently, but the same problems arise with the false positive rate of the tests.

microdrum 7y ago

No. They both have to be rejected in order to affirm our religiously held priors. The virus has IFR of 2% or more, millions of Americans will due, and we must stay in our homes.* * Not FedEx employees though! /s