A test of 200 people just outside Boston found that 32% had been exposed to the coronavirus

[Yo_Mama_Been_Loggin]

compared to an official rate of 2%

A snapshot test of people in a town just outside of Boston found that around one-third of them had been exposed to the coronavirus, even though none showed symptoms.

The results from Chelsea, Massachusetts, have led some experts to conclude that the coronavirus outbreak there, and potentially elsewhere, had touched many more people than first thought.

The study was carried out by medics from Massachusetts General Hospital, and was reported by the Boston Globe newspaper.

Chelsea is just outside Boston, and has one of the most severe outbreaks in the state.

The study asked 200 people to provide a drop of blood to be examined by a rapid testing kit, which delivered a result in 10 minutes. The Globe said that the tests are not FDA-approved, but are considered reliable by Massachusetts General.

https://news.yahoo.com/test-200-people-just-outside-144732703.html

Meaning we need more testing but we won't get it from President "Shit for Brains".

[LisaL]

1. How much do we need for herd immunity?

[JenniferJuniper]

2. It can go into effect at 40%, but usually it's at least 80%

[DrToast]

5. It depends on the R0

If R0 is 2.2, herd immunity is estimated to require 55% of the population.

If R0 is 5.7, herd immunity is estimated to require 82% of the population.

https://www.forbes.com/sites/tarahaelle/2020/04/07/the-covid19-coronavirus-disease-may-be-twice-as-contagious-as-we-thought/#1ed439c29a6a

[Celerity]

7. +1

[Celerity]

6. It depends on the R- naught value (how contagious it is)

R-naught (R0) of 2.0 needs 50% viral penetration
R0 of 2.5 needs 60%
R0 of 3.0 needs 67%

And so on. Lots of arguing over the R0 atm, but its probaly around 2.7/2.8, so 63/64% viral penetration needed for herd immunity to occur.

The highest R0 I have seen pushed is 5.7, which would require 82% viral penetration. I take that with a grain of salt as some of the sources pushing were of the same archetype of those who were wildly and irresponsibly flinging around 7, 8, even 10, 12 million deaths in the US alone just a few weeks ago. Not saying an R-naught of 5.7 is impossible, not at all, but I need to see a LOT more evidence.

[Shermann]

3. Amazing if accurate

Really good news and bad news at the same time.

[FreeState]

8. Problem is it's a biased study - volunteers

They didn’t survey a non-biased group - they asked for volunteers. It’s wasn’t a study of the general public.

“The project, carried out by Massachusetts General Hospital, took blood from 200 volunteers and tested it for antibodies produced after exposure. 32% had the antibodies.“

[Igel]

9. The question is how the sample would have been biased.

The ruled out those who'd been tested (by self-report, I assume).

That left those who hadn't been tested. Half of those who were found to have antibodies reported having had some symptoms--so that's a clear source of bias. 64 had antibodies, so that's 32. Out of 200 that leaves 168 who didn't fess up to having symptoms.


That leaves 168, and it's unclear to me what bias would be in there. Assuming they're accurate in reporting no symptoms. Of those, 32 had antibodies. That's 19%. That would be a good start.

Do need a random survey, but I'm not sure how to get that. Even picking random blocks and going door to door would include some bias. A large enough sample would be good but impractical. I guess take what you can get, record demographics, and weight things accordingly?

[FreeState]

10. Yes and no

The problem is when people can’t be tested and you ask for volunteers you will get a high percentage of people who had reason to believe they had it and wanted to be tested.

One non-biased way would be to study groups of random(ish) people - like everyone who worked at a factory that shut down before any cases were reported or similar. The Chinese have done this and found only about 3% have antibodies.

Edit to add I can’t find the article about the factory but here is a more random study currently underway:

https://www.bloomberg.com/news/articles/2020-04-18/china-tests-thousands-to-calculate-true-spread-of-coronavirus

[Baclava]

4. Exposed meaning antibodies found, but not infectious? Are they sure?