Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Covid-19 Antibody Seroprevalence in Santa Clara County, California

medrxiv.org

Post reply on HN
Print view View original

131–140 of 209 posts

Re: Covid-19 Antibody Seroprevalence in Santa Clara County, California

#131

Post by jboy55 »

tom-thistime wrote: post #124
crdrost wrote: post #111

Earlier quoted context omitted.

That can't be true for the official confirmed numbers. Like, it was for a while impossible to get a test if you are not experiencing multiple symptoms AND can make a case that you might have been exposed via interacting with someone who has traveled. Even now I don't think they're giving the test if you don't have a fever. It's implausible that 50% of the people tested by the official pathway had no symptoms because…

Ok, we can't conclude anything about the rate of symptoms from testing that was only conducted on people showing severe symptoms. I don't think we can conclude much of anything from the OP either, because its rate of positives seems compatible with the rate of false positives from the test. They could have 0% infected, they could have 5% infected. An example where testing was conducted without requiring symptoms firs…

Also, they weren't able to factor out a bias towards people seeking confirmation tests due to perceived COVID symptoms.

This is in the bay area, where people were very quick to self-quarantine. Most people have been working from home since early March. They were asking people to break their self-quarantine to get tested without having any symptoms?

Re: Covid-19 Antibody Seroprevalence in Santa Clara County, California

#132

Post by jboy55 »

ajross wrote: post #108
robotcookies wrote: post #104

"Participants were recruited using Facebook ads targeting a representative sample of the county by demographic and geographic characteristics." How is it representative when it's only Facebook users in this sample?

In the south bay? Pretty good, unless you want to argue for a demographic skew because of Facebook's poor uptake among iOS users. I mean, no. It's not perfect. It will miss demographics like the elderly with lower social media use, but that group tends to be well-sampled already due to their risk profile. It will probably miss some immigrants too, which seems like a bigger problem. But really, it's a pandemic. It doe…

They also were unable to remove any bias towards people self-selecting to be part of the test because they felt they had symptoms of COVID. The bay area has been in lock down since early March, its really hard to imagine someone participating if they didn't feel they had some symptoms and wanted to 'know for sure'.

Re: Covid-19 Antibody Seroprevalence in Santa Clara County, California

#133

Post by contravariant »

MiguelVieira wrote: post #109
erentz wrote: post #79

It’s hacky and I’m happy to be told how wrong it is. But based on these recent antibody studies in Germany, Finland, and now here in CA, I’ve been assuming an actual fatality rate of about 0.4, and thus an actual infected rate of 250x our known deaths, which is a much firmer number. This is only a small comfort since it means we may have had about 8.75 million infected and presumably now immune in the USA. Or about 2…

I've been using 0.5% for mental math, rounding down from the lower bound from this study in The Lancet: https://www.thelancet.com/journals/laninf/article/PIIS1473-3... I would be surprised if, after this current wave of infections, the percentage of people with antibodies in the US is higher than the low single digits. Herd immunity without a vaccine is a pipe dream. Our best bet is to massively ramp up testing and c…

Herd immunity isn't a pipe dream it is just one hell of a sacrifice. And we'd better check the immunity actually lasts before we try that. Frankly I've never been a great fan of the 'flatten the curve' explanation, sure the graphs seem nice but I reckon the peak in those graphs is drawn too low by several orders of magnitude.

In the Netherlands a fatality rate of 0.5% would mean it'd take about a year for herd immunity to kick in assuming we managed sustain the peak that occurred about 2 weeks ago (and it's somewhat dubious whether the healthcare system actually can sustain that peak).

Re: Covid-19 Antibody Seroprevalence in Santa Clara County, California

#134

Post by hcknwscommenter »

MiguelVieira wrote: post #109
erentz wrote: post #79

It’s hacky and I’m happy to be told how wrong it is. But based on these recent antibody studies in Germany, Finland, and now here in CA, I’ve been assuming an actual fatality rate of about 0.4, and thus an actual infected rate of 250x our known deaths, which is a much firmer number. This is only a small comfort since it means we may have had about 8.75 million infected and presumably now immune in the USA. Or about 2…

I've been using 0.5% for mental math, rounding down from the lower bound from this study in The Lancet: https://www.thelancet.com/journals/laninf/article/PIIS1473-3... I would be surprised if, after this current wave of infections, the percentage of people with antibodies in the US is higher than the low single digits. Herd immunity without a vaccine is a pipe dream. Our best bet is to massively ramp up testing and c…

It's not a pipe dream. It's just that it needs to happen in a different way. We need to massively increase number of isolation rooms, icu beds etc. (just like China did building that hospital in 10 days), massively increase number of clinicians (take every one in their last year of nursing school and med school and get them in the hospital, not making decisions but carrying out basic tasks; use a draft if you have to, just like Italy did), and massively increase testing capacity (~4x from today's capacity in US), and massively increase ventilators, add ubiquitous temperature checking (to enter any building other than your house, you need to confirm temp) and then just open up the economy and let it rip. With enough testing and enough surge healthcare capacity, we can do this.

Re: Covid-19 Antibody Seroprevalence in Santa Clara County, California

#135

Post by ignoramous »

fiftyfifty wrote: post #115
ignoramous wrote: post #106

Earlier quoted context omitted.

> So we are still a long way off herd immunity. But at the same time closer than we thought we were. Yes and no. A disease is endemic when R0 x S = 1 [0]. R0 for SARS-CoV-2 is estimated to be 2.5 to 3.5 and some models predicting a much higher 5.7 [1]. The herd immunity threshold is given by (1 - S)% , which implies when R0 = 2.5, 60% of the population would need to be immune; similarly 71.42% and 82.45% for R0 s, 3.…

What percentage of immunity can we expect to see a slowdown in the spread of the disease? While we may need 60% or more to achieve full herd immunity surely somewhere lower than that we can expect to see a noticeable reduction in the spread?

> ...noticeable reduction in the spread?

I believe the SIR model [0] answers those questions. 3blue1brown's video on the topic is super digestible and informative [1].

In short, the discovery of a prophylactic if not a therapeutic treatment are our best immediate bets, absent which, quarantines and lockdowns will have to persist to curb the spread of a disease as infectious as covid-19 whilst we patiently wait for an effective vaccine against it [2].

[0] https://en.wikipedia.org/wiki/Compartmental_models_in_epidem...

[1] https://www.youtube.com/watch?v=gxAaO2rsdIs

[2] https://en.wikipedia.org/wiki/Vaccine_efficacy

Re: Covid-19 Antibody Seroprevalence in Santa Clara County, California

#136

Post by guenthert »

cameldrv wrote: post #103
cuchoi wrote: post #8

Caveat from the paper: "This study had several limitations. First, our sampling strategy selected for members of Santa Clara County with access to Facebook and a car to attend drive-through testing sites. This resulted in an over-representation of white women between the ages of 19 and 64, and an under-representation of Hispanic and Asian populations, relative to our community. Those imbalances were partly addressed…

Yes, and look at the way they corrected for this. They rebalanced for demographics, and that doubled their estimate. This is exactly the opposite of what they should have done. For example, zip codes farther from the testing sites were less likely to get tested, and more likely to be positive. What you see is that certain groups were less likely to go get tested, UNLESS they had a good reason to think that they were…

With the average adult getting two bouts of common cold each year and mild COVID symptoms resembling a common cold, I'd think a large fraction of the population might think they had it. I wouldn't dear to guess in which way this shifts the estimate.

I think it's understood, that with such a small sample size, results are to be taken cautiously. You have to start somewhere though.

Re: Covid-19 Antibody Seroprevalence in Santa Clara County, California

#137

Post by fspeech »

guscost wrote: post #125
fspeech wrote: post #34

Earlier quoted context omitted.

How do you square that with the fact that there is no outbreak in any overseas Chinese community? Not Taiwan or Hongkong or Macau, nor in the epicenter of Italy ( https://www.reuters.com/article/us-health-coronavirus-italy-... ), nor San Francisco Chinatown ( https://www.nytimes.com/2020/04/17/us/san-francisco-coronavi... )? All of these communities have far more ties to mainland China.

It could have already passed through all of those communities basically unnoticed, and/or confused with other ILIs. There's no way to rule out that possibility, and when you consider how rare cardiovascular diseases (the most common comorbidity) are in those communities, you have to conclude that it's not vanishingly unlikely: https://minorityhealth.hhs.gov/omh/browse.aspx?lvl=4&lvlid=4... Obviously mitigation effort…

Maybe Chinese emigrants in Italy are young so they didn't notice the wave. But Taiwan/Hongkong/Macau? They are testing and tracing. When their recent travelers to Europe and America came back, a lot of them were infected and caused a spike in cases. https://www.nytimes.com/interactive/2020/04/09/world/asia/co...

Re: Covid-19 Antibody Seroprevalence in Santa Clara County, California

#138

Post by Uhhrrr »

erentz wrote: post #79

It’s hacky and I’m happy to be told how wrong it is. But based on these recent antibody studies in Germany, Finland, and now here in CA, I’ve been assuming an actual fatality rate of about 0.4, and thus an actual infected rate of 250x our known deaths, which is a much firmer number. This is only a small comfort since it means we may have had about 8.75 million infected and presumably now immune in the USA. Or about 2…

This is what I've been assuming - 200 to 250x. But 250xdeaths gives you the infected number from two weeks ago.

To get current infected you then have to multiply it by communication^(days to death), something like 1.1^14, with the first term varying depending on population density and lockdown measures, and the second on how on the ball that locality is for detecting cases and keeping people alive.

Re: Covid-19 Antibody Seroprevalence in Santa Clara County, California

#139

Post by timr »

pkaye wrote: post #105
timr wrote: post #48

Earlier quoted context omitted.

All of the flu surveillance numbers suggested a historically bad season this year, and there was a big peak in flu-like mortality in February. It's entirely possible that we were mis-characterizing the data: https://gis.cdc.gov/grasp/fluview/mortality.html

Stanford did a study looking for Coronavirus cases in Jan and Feb in the Bay area. http://med.stanford.edu/news/all-news/2020/04/testing-pooled... > The researchers found that the burden of COVID-19 in the Bay Area prior to mid-February was low. Only two of nearly 3,000 people with respiratory-disease symptoms who were tested in early 2020 at Stanford Health Care or affiliated clinics for common respiratory viruses w…

Sure, but the bay area has very few detected infections, in general.

It seems pretty obvious that we need to do the same experiment on samples from New York, Washington, Michigan, etc.

Post reply on HN