3 ms·

There's been a lot of speculation and conflicting information, but I recently read that a supposedly random sample of people in NY (I assume state, not city) su

There's been a lot of speculation and conflicting information, but I recently read that a supposedly random sample of people in NY (I assume state, not city) supermarkets found about 16% had antibodies for covid. So it's my belief currently that the reported case numbers are on the order of 1/10th the total and one in six people probably has it. Edit: is this controversial? Do people not believe NY state did such a study? WTF??

mapgrep 7y ago
Yes, this was a state test: https://www.nytimes.com/2020/04/23/nyregion/coronavirus-antibodies-test-ny.html https://www.nytimes.com/2020/04/23/nyregion/coronavirus-anti... “ In New York City, about 21 percent tested positive for coronavirus antibodies during the state survey. The rate was about 17 percent on Long Island, nearly 12 percent in Westchester and Rockland Counties and less than 4 percent in the rest of the state. “State researchers sampled blood from the approximately 3,000 people they had tested over two days, including about 1,300 in New York City, at grocery and big-box stores. The results were sent to the state’s Wadsworth facility in Albany, a respected public health lab. This was characterized as Covid spreading “far more widely” than previously believed in NY. So I agree national rate probably higher than people think. Of course the 1/5 figure would not be expected outside the hotspot of NYC. Probably a national high in fact. But 1/10 as a long term national number seems plausible.
eyegor 7y ago
Guessing the downs are because you acknowledged that speculation and conflicting information is problematic, then tossed out specific numbers and attributed "they say". Should probably cite if you want to malign a lack of citations.
- perl4ever 7y ago
  I overlooked the ambiguity in saying "I recently read", where I meant in an article in the media, not according to some random gossip.
michaelmrose 7y ago
There are numbers coming in from NY and CA that are very interesting but not dispositive. We don't know what the rate of false positives is, we don't know if the tests were sufficiently representative, we don't know what degree of immunity is actually conferred. An intelligent critique of the CA analysis 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... At least the new york research didn't rely on a facebook ad which offered free covid testing but it was conducted by surveying people shopping in big box stores. It would be unsurprising if idiots out shopping had a higher prevalence than the general population. Covid deniers have latched firmly onto the "aspirational" idea that many many more people have had it thus it really and truly isn't any more deadly than the flu even though the CA numbers are so badly analyzed that the submitter probably ought to be fired and the New York numbers only show that the infection fatality rate may be as low as 1/2 of 1% instead of just shy of 1% as previously theorized. Covid deniers usually move right from fudging the denominator in the deaths/cases calculation to proving that we should just all go back to work with an optional stop about it being a democratic plot to destroy our economy to make everyone dependent on the gubmint. I recognize that this isn't your shtick but its pretty common among your slightly more educated ignoramus recently.
- Gibbon1 7y ago
  UCSF has preliminary results from it's mass testing effort in the mission. They used both RNA and Serological tests to test about ~4200 people. This is a dense neighborhood that's over represented in detected cases. https://missionlocal.org/2020/05/preliminary-results-of-mission-covid-tests-show-95-percent-of-positive-cases-were-latinx/ https://missionlocal.org/2020/05/preliminary-results-of-miss... Short: 2% infection rate. Bad news for team nutter.
stevage 7y ago
>There's been a lot of speculation and conflicting information, [here, have some more confilcting information]
Zanni 7y ago
Here's an article on why that figure might be overblown (tl;dr - sampling bias): https://towardsdatascience.com/were-21-of-new-york-city-residents-really-infected-with-covid-19-aab6ebefda0 https://towardsdatascience.com/were-21-of-new-york-city-resi...

mapgrep 7y ago

Yes, this was a state test: https://www.nytimes.com/2020/04/23/nyregion/coronavirus-antibodies-test-ny.html https://www.nytimes.com/2020/04/23/nyregion/coronavirus-anti... “ In New York City, about 21 percent tested positive for coronavirus antibodies during the state survey. The rate was about 17 percent on Long Island, nearly 12 percent in Westchester and Rockland Counties and less than 4 percent in the rest of the state. “State researchers sampled blood from the approximately 3,000 people they had tested over two days, including about 1,300 in New York City, at grocery and big-box stores. The results were sent to the state’s Wadsworth facility in Albany, a respected public health lab. This was characterized as Covid spreading “far more widely” than previously believed in NY. So I agree national rate probably higher than people think. Of course the 1/5 figure would not be expected outside the hotspot of NYC. Probably a national high in fact. But 1/10 as a long term national number seems plausible.

eyegor 7y ago

Guessing the downs are because you acknowledged that speculation and conflicting information is problematic, then tossed out specific numbers and attributed "they say". Should probably cite if you want to malign a lack of citations.

perl4ever 7y ago
I overlooked the ambiguity in saying "I recently read", where I meant in an article in the media, not according to some random gossip.

michaelmrose 7y ago

There are numbers coming in from NY and CA that are very interesting but not dispositive. We don't know what the rate of false positives is, we don't know if the tests were sufficiently representative, we don't know what degree of immunity is actually conferred. An intelligent critique of the CA analysis 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... At least the new york research didn't rely on a facebook ad which offered free covid testing but it was conducted by surveying people shopping in big box stores. It would be unsurprising if idiots out shopping had a higher prevalence than the general population. Covid deniers have latched firmly onto the "aspirational" idea that many many more people have had it thus it really and truly isn't any more deadly than the flu even though the CA numbers are so badly analyzed that the submitter probably ought to be fired and the New York numbers only show that the infection fatality rate may be as low as 1/2 of 1% instead of just shy of 1% as previously theorized. Covid deniers usually move right from fudging the denominator in the deaths/cases calculation to proving that we should just all go back to work with an optional stop about it being a democratic plot to destroy our economy to make everyone dependent on the gubmint. I recognize that this isn't your shtick but its pretty common among your slightly more educated ignoramus recently.

Gibbon1 7y ago
UCSF has preliminary results from it's mass testing effort in the mission. They used both RNA and Serological tests to test about ~4200 people. This is a dense neighborhood that's over represented in detected cases. https://missionlocal.org/2020/05/preliminary-results-of-mission-covid-tests-show-95-percent-of-positive-cases-were-latinx/ https://missionlocal.org/2020/05/preliminary-results-of-miss... Short: 2% infection rate. Bad news for team nutter.

stevage 7y ago

>There's been a lot of speculation and conflicting information, [here, have some more confilcting information]

Zanni 7y ago

Here's an article on why that figure might be overblown (tl;dr - sampling bias): https://towardsdatascience.com/were-21-of-new-york-city-residents-really-infected-with-covid-19-aab6ebefda0 https://towardsdatascience.com/were-21-of-new-york-city-resi...