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

Preliminary test results suggest 21% of NYC residents have Covid antibodies

6sqft.com

Post reply on HN
Print view View original

561–570 of 679 posts

Re: Preliminary test results suggest 21% of NYC residents have Covid antibodies

#561

Post by mantap »

capkutay wrote: post #556
aazaa wrote: post #317

> In his press conference today, Governor Cuomo revealed the preliminary results of a first-phase antibody test that surveyed 3,000 New Yorkers over two days in 19 counties at 40 locations that included grocery and big-box stores. The sample suggests that 13.9% of New York State residents have the antibodies, meaning they had the virus at one point and recovered. Of the regions tested–Long Island, NYC, Westchester/Ro…

> This still is not a random sampling. It only samples from shoppers at grocery stores and big-box retailers. Imagine doing the same study, but of people who ordered groceries online only. Would you expect to see big differences in exposure rate? I think so. I'd expect people who still do their own grocery shopping and go to big box retailers to be a better sample of the population than hacker news' audience definiti…

It's not black and white. A person who goes to the store more often has a greater chance of being tested, than someone who goes infrequently.

There's also the issue that people are trying to avoid other human beings right now, including the human beings administering tests. People who are social distancing more effectively are less likely to agree to a test.

Re: Preliminary test results suggest 21% of NYC residents have Covid antibodies

#562

Post by ineedasername »

greedo wrote: post #518
Symmetry wrote: post #311

Earlier quoted context omitted.

I hope that the numbers stay at .6% and that the difference from previous estimates might even be improvements we've made in figuring out how to care for people suffering from Covid-19. But I remember being happy at how low South Korea's numbers were when their epidemic peaked and receded then watching the case fatality rate slowly climb up to the 2.2% it's at today. So I'm not counting any chickens just yet.

Be careful not to conflate CFR with IFR. CFR is the number of deaths divided by the number of confirmed cases. IFR is deaths divided by total infections. It's hard to determine IFR, but CFR is almost always an exaggeration of the risk.

Yes. Hiwever, it is valid to compare the CFR of one pandemic to the CFR of another pandemic if you find the CFR from the same relative point in time of that other pandemic.

Re: Preliminary test results suggest 21% of NYC residents have Covid antibodies

#563

Post by mirimir »

aazaa wrote: post #317

> In his press conference today, Governor Cuomo revealed the preliminary results of a first-phase antibody test that surveyed 3,000 New Yorkers over two days in 19 counties at 40 locations that included grocery and big-box stores. The sample suggests that 13.9% of New York State residents have the antibodies, meaning they had the virus at one point and recovered. Of the regions tested–Long Island, NYC, Westchester/Ro…

Sure, but it's still data for the groups that they sampled. It would be better to have data for truly random samples.

But even then, you'd want demographic data for each subject. And then you could compare appropriate subsets of the results with this study.

With such data, you could explore risk factors, effectiveness of quarantines, and so on.

Re: Preliminary test results suggest 21% of NYC residents have Covid antibodies

#564

Post by mantap »

zyxzevn wrote: post #512

Good news. Herd-immunity is growing. Also, this means that the death-rate is far lower than reported. And this is nice too. Maybe we can save people by starting the society again. In other news, we see that hydroxychloroquine is actually working according to doctors. But there was a very bad study that claimed that it caused 2x as many deaths. It should have said instead that it was mainly used on patients that were…

The presence of antibodies is not sufficient for herd immunity. For herd immunity you need those antibodies to stick around.

Re: Preliminary test results suggest 21% of NYC residents have Covid antibodies

#565

Post by wtvanhest »

mannykannot wrote: post #550
koheripbal wrote: post #546

Earlier quoted context omitted.

As a (40-something) NYC resident, I don't believe that is accurate. Half of the people I know are working from home and stocked up on 1-2 months of food at the outset of this. The other half left the city and are now staying with friends/family in other states or have rented homes elsewhere. Those that did stay and still need food (ie produce) are having it delivered. People who are shopping in stores are absolutely…

Thanks for your observations. If, however, behavior in NYC is more or less the same as elsewhere, then what might account for the difference in the results between store-customer sampling in the city and elsewhere? The results may well be biased from sampling shoppers, but if the bias is assumed to be from shopping behavior, and that behavior is essentially the same everywhere, why does it not wash out? One explanati…

Just out of curiosity, what studies are you asking the OP to compare?

Re: Preliminary test results suggest 21% of NYC residents have Covid antibodies

#566

Post by np_tedious »

Karrot_Kream wrote: post #401
TheBlight wrote: post #348

Earlier quoted context omitted.

Sure all of these studies are potentially flawed but they're all generally pointing in the same direction. There are many more infections than we know about and the CFR is not anywhere close to the actual IFR. NYC is an outlier with a 21% infection estimate but for the rest of NY (outside NYC metro/Westchester/LI) the estimate is 3.6%. Santa Clara estimate was 3%. LA County estimate was 4%. Seems like a trend is deve…

> Sure all of these studies are potentially flawed but they're all generally pointing in the same direction That's unfortunately not the way statistics works. Combining multiple bad tests just makes the results incorrect or highly uncertain; the devil often is in the details. The Stanford/Santa Clara County study is a good example of how the details can really undermine a study. Things do indeed seem to be converging…

> Combining multiple bad tests just makes the results incorrect or highly uncertain

But it's not just, say, taking a bunch of faulty studies and average them and saying voila. They are in agreement. They'd have to be bad in similar ways in order to converge like that.

Re: Preliminary test results suggest 21% of NYC residents have Covid antibodies

#567

Post by umeshunni »

mrb wrote: post #279
izend wrote: post #198

Earlier quoted context omitted.

Making it the worse pandemic in decades but no where near the Spanish Flu.

I disagree. It's simply too early to make that determination. The Spanish Flu had a estimated infection fatality ratio of >2.5% ( https://wwwnc.cdc.gov/eid/article/12/1/05-0979_article ). As another commenter pointed out, the COVID-19 IFR is 0.9% in NYC. The difference between 2.5% and 0.9% is only 2.8x. However the IFR is not only underestimated due to the lag between case detection and death, but it's also probably…

Another factor to consider is that healthcare today is much better than when the Spanish flu occurred and that could explain the 2x difference.

Re: Preliminary test results suggest 21% of NYC residents have Covid antibodies

#568

Post by wtvanhest »

koheripbal wrote: post #546
mannykannot wrote: post #499

Earlier quoted context omitted.

NYC is not suburbia. I believe a higher percentage are still working outside of their homes (and, in many cases, that also means travelling on public transport), and a smaller percentage have the financial and transport resources to significantly reduce the frequency of their shopping trips. Apartment living means a greater exposure to your neighbors whenever you do enter or exit. The sampling in NYC, while far from…

As a (40-something) NYC resident, I don't believe that is accurate. Half of the people I know are working from home and stocked up on 1-2 months of food at the outset of this. The other half left the city and are now staying with friends/family in other states or have rented homes elsewhere. Those that did stay and still need food (ie produce) are having it delivered. People who are shopping in stores are absolutely…

I’m a resident of SF downtown and previously DC and Boston. Everyone I know has stopped leaving their apartments/condos except for fresh air and are trying not to go to any stores. When possible are having food delivered. My friends in NYC have all left except for one who has a large apartment and no family without taking a flight.

I think people see these studies of places they visited once and determine that those areas are way more dense and that they have seen the news and that all those deaths are due to tons of people having it.

That is just not true. The studies are biased and likely this will go on for a long-time. We know this is way more deadly than the flu. That is obvious when you look at simple examples like the princess cruise in Japan. How often do 7+ people die on a cruise ship on a single voyage?

Re: Preliminary test results suggest 21% of NYC residents have Covid antibodies

#569

Post by misun78 »

newacct583 wrote: post #529
CubsFan1060 wrote: post #373

Earlier quoted context omitted.

Which is about 1/2 of what die in a normal year. The real question is how much do those circles overlap. Are we saving lives, or are people now dying in June that would have died in November anyway?

This is like the third or fourth spot I've seen on this very site where people are postulating that covid deaths would have really died in "months" anyway. Why is this so weirdly consistent? To be clear: there is absolutely no evidence for a number like that, none at all. Typical at-risk patients with comorbidity like advanced age or obesity are routinely expected to live for decades. Actual conditions with a death-i…

If you look at the median age of covid fatalities in Italy at age 81.5, this question is entirely valid. Some of the numbers I read pointed to a 16-20% all cause mortality rate for that group irrespective of covid. So by definition the circles overlap to a meaningful extent.

Re: Preliminary test results suggest 21% of NYC residents have Covid antibodies

#570

Post by zamfi »

timr wrote: post #495
zamfi wrote: post #427

Earlier quoted context omitted.

> Are all three methods flawed in some way? Sure. They're not just all flawed in some way -- along at least one dimension they are flawed in the same way , that is, they all have a biased sample of the population that seems more likely to be infected than a true random sample. So, sure, they're "created differently", i.e., they're not all facebook ads, or grocery stores, but all those methods bias towards higher infe…

These studies have completely different methodologies, and the only way you can plausibly call them similar is by arguing that anyone willing to get serological testing is substantially more likely to be infected than those who are not. That's little more than a fancy way of disregarding any testing result you don't happen to like. We've been systematically biasing toward testing the old and infected for months now.…

> arguing that anyone willing to get serological testing is substantially more likely to be infected than those who are not.

Yes, which is clearly true, no?

> That's little more than a fancy way of disregarding any testing result you don't happen to like.

We all have biases. I'm not sure how pointing out that adverse selection is a thing, though, is a bias.

Any stats can be quibbled with, and I have no personal skin in this game -- if anything, I would love (like almost everyone) for the IFR to be > and yet all are pointing in the same direction: a systematic under-count of cases

We already know we're undercounting, because we're only explicitly testing people with symptoms, and not even all people with symptoms! We didn't need these studies for that.

These studies would ideally be true random samples so that we can know what the true infection rate is. Some locales are doing that kind of testing, and I'll be much more interested in those reports than studies in which respondents can opt into participation.

Post reply on HN