Preliminary test results suggest 21% of NYC residents have Covid antibodies • Hacker News Board

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

#481

Post by nodamage » Thu, Apr 23, 2020, 11:30 PM UTC

__blockcipher__ wrote: post #387

slg wrote: post #360

Earlier quoted context omitted.

>There are many more infections than we know about and the CFR is not anywhere close to the actual IFR. That is the optimistic takeaway. The pessimistic takeaway is that even the hardest hit areas are nowhere near herd immunity levels and that we are either going to be isolating until a vaccine is created or we can expect to see a lot more death once nonessential people are forced back to work.

At this point people advocating the position you’re advocating for are in a state of denial (this is my opinion, not a matter of fact, obviously). Your assumption is that we can effectively prevent the majority of the nation from exposure via lockdown. Not only does evidence seem to point against that, but when you do the math on mortality due to suicide and overdose it’s not clear that containment would even save mo…

> As others have said, for those around age 45 or less, Covid is equally or less dangerous than Influenza. And particular for those under 30 the flu is an order of magnitude more deadly at least.

Can you share the specific numbers you used to reach this conclusions? The reason I ask is because I'm seeing a lot of people in this thread comparing:

1) Overall influenza fatality rates to age stratified Covid fatality rates, and

2) Symptomatic influenza fatality rates to symptomatic + asymptomatic Covid fatality rates.

Both of which are highly flawed comparisons for reasons that should be obvious.

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

#482

Post by arcticbull » Thu, Apr 23, 2020, 11:38 PM UTC

camgunz wrote: post #480

__blockcipher__ wrote: post #448

Earlier quoted context omitted.

Well even with your qualifier of "no comorbidities", the evidence is pretty unanimous that Covid and Influenza are not even the same ballpark. Worrying about a healthy young person dying of Covid is more akin to worrying about a healthy young person dying of cancer than it is to bacterial meningitis or something. As I've said elsewhere, Influenza is defined by being deadly to the very young and very old, and so-so to…

> Unfortunately, and I know people read this stuff and their quack heuristics start firing, the reason people are so afraid about Covid's impact on young people is because the mainstream media has intentionally promulgated a narrative that "young people are at risk too" because they fear that otherwise young people wouldn't submit to glorified house arrest for months straight. They've reported that because young peop…

Young people are also drastically less likely to get severe disease, and drastically less likely to be hospitalized, see FIGURE 1. [1] I find it very suspicious that young people who don't go to the hospital with severe disease would require oxygen or end up with life-long lung injuries.

> Everyone needs to stay practice social distancing and shelter, because everyone can carry and spread.

Sweden has demonstrated at national scale that's not the case. [2,3]

The reality is about 60% of us are going to get it one way or the other, so let's control which 60%, and in what order, and on what timeline before everyone stops listening and just walks out.

[1] https://www.cdc.gov/mmwr/volumes/69/wr/mm6915e3.htm

[2] https://www.bloomberg.com/news/articles/2020-04-19/sweden-sa...

[3] https://theconversation.com/coronavirus-are-we-underestimati...

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

#483

Post by fspeech » Thu, Apr 23, 2020, 11:40 PM UTC

__blockcipher__ wrote: post #464

fspeech wrote: post #461

Earlier quoted context omitted.

In San Diego people under 30 are close to 20% of total confirmed and over 5% of hospitalized cases. About 7% of confirmed are hospitalized for this age group. Our positive test ratio is under 7%. Tests are constrained by policy but not dramatically so. https://www.sandiegocounty.gov/content/sdc/hhsa/programs/phs... Another data point is the US military. I believe they test much more widely due to their living conditi…

Young people interact with more individuals, and engage in more behavior that exchanges microbes such as unprotected sex, protected sex, sharing pipes/vapes, sharing utensils, etc. We should expect that the proportion of people with Covid should be heavily biased towards the young for that reason, as it is for any disease that does not have a biased transmission. Most indications I've seen is that the young are not h…

For any test that is not given randomly or universally, there is a selection bias. A test given based on severity of symptoms bias heavily against patients with light symptoms, presumably younger people. But to get from 7% (hospitalized to confirmed) to 1 in 500 is a factor of 35. BTW the city of Canton (Guangzhou) just tested all their high school seniors in preparation for school reopenning and found zero positive case so far. And they did find cases among other groups.

I don't think anyone enjoys lockdown. At a minimum we need sufficient testing, sufficient ppes, sufficient surge capacity and training for standard supportive care. We are still quite a ways from that. Even if death is unavoidable people deserve to have reasonably good care.

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

#484

Post by DenisM » Thu, Apr 23, 2020, 11:41 PM UTC

This might be a bit blunt but I am not jesting - an article about disease testing that does not contain the words "false positive" and "false negative" in some form is literally not worth reading.

Even decent antibodies tests can have 10% false positive rates (bodysphere), and lots of those tests have dodgy provenance and 20% false positive is not out of the question.

Here's a better article, thou still not great (paywall): https://www.nytimes.com/2020/04/23/nyregion/coronavirus-anti...

The closest they come to discussing the errors is to acknowledge there might be some: State officials said the test had been calibrated to err on the side of producing false negatives — to miss some who may have antibodies — rather than false positives, which would suggest a person had coronavirus antibodies when they did not. Not very informative, I know.

A word of caution from the NYC health officials: https://www1.nyc.gov/assets/doh/downloads/pdf/han/alert/2020...

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

#485

Post by swat535 » Thu, Apr 23, 2020, 11:42 PM UTC

taeric wrote: post #303

bryansum wrote: post #259

Earlier quoted context omitted.

Absence of evidence is not evidence of absence.

But there is evidence in all of the asymptomatic people getting tested and not having any affects worth making it into a report. Could they have hidden damage that we can't detect? Plausible. But that is true of many activities. And illnesses.

> Could they have hidden damage that we can't detect? Plausible. But that is true of many activities. And illnesses

The difference is that we know plenty about those illnesses and activities. This is a novel virus and we are just speculating.

We simply don't know the long term effects of this virus, even in recovered cases.

Until we know this virus well enough to make such statements, we should play it safe.

I for one am not willing to risk a lung because you think it's fear mongering

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

#486

Post by GhostVII » Thu, Apr 23, 2020, 11:43 PM UTC

Uhhrrr wrote: post #468

nateberkopec wrote: post #406

Earlier quoted context omitted.

It's more or less epidemiologically impossible that you will see an attack rate of 100% (as you just assumed). Once attack rate reaches >40~60%, the reproduction number drops below 1 because there just aren't enough uninfected hosts.

It will probably be significantly higher than 60%, since COVID-19's R0 without lockdown seems to be pretty high. The CDC has estimated it at 5.7 [0]. Herd immunity threshold = 1 - 1/R0, so the threshold would be 82.4%. [0] https://wwwnc.cdc.gov/eid/article/26/7/20-0282_article?deliv...

That R0 assumes we are taking essentially no measures to stop the virus though, right? I don't no much about it but it seems like the R0 would drop significantly if we keep doing social distancing, expand testing, make people wear masks, etc.

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

#487

Post by hitchhiker999 » Thu, Apr 23, 2020, 11:44 PM UTC

sjg007 wrote: post #52

Is this antibody test specific for covid-19 or would it be positive on other coronaviruses? That's a real concern since the other 4 coronaviruses are endemic.

What's terrifying is with all these professional analysts in here, yours is the first comment I've seen that asks this obvious question.

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

#488

Post by qqqwerty » Thu, Apr 23, 2020, 11:44 PM UTC

mlyle wrote: post #473

qqqwerty wrote: post #470

Earlier quoted context omitted.

> If you miss a constant percentage of cases We are talking about 1 serological study at a single point in time. We have no idea what percentage of cases we were missing before or after that point.

There's no reason to interpret the serological study to think a higher R0 is implied, which is what you said. The only way that missing cases affects estimates of R0 from time series is if we're missing a much bigger proportion of cases now than we were in the past. All the evidence I've seen leans the other way-- more testing, better testing, more cases diagnosed by "presumed" criteria. So, it actually implies the o…

I am not interpreting it that way, I believe in the consensus estimates. The parent commenter (articbull) thinks these serological studies (SC, LA, and to a lesser degree NY) prove that the IFR is much lower than we thought because the actual infection numbers are much higher. The NY study, using reasonable assumptions, looks like the IFR might be between 0.5% - 1.0%, but the SC study was claiming 0.1-0.2%. And I am arguing, in the hypothetical case that the SC study is correct (I think it is flawed), then we have under estimated R0.

Considering we have been under lockdown for over a month (which greatly impacts effective transmission), and considering our estimates of latency hasn't changed dramatically, I think my above assumption is fair.

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

#489

Post by slg » Thu, Apr 23, 2020, 11:47 PM UTC

mlyle wrote: post #479

slg wrote: post #465

Earlier quoted context omitted.

First off, all those numbers are once again absolute deaths and not related to case numbers so my prior complaint still stands. If you look at the first chart, you will see that COVID-19 deaths didn't start becoming a problem until March. Distancing practices were not widespread until mid-March. Meanwhile flu deaths weren't dramatically cut until the most recent timeframe. So taking into consideration both the first…

We've said approximately as dangerous. It clearly is in a very similar ballpark. We don't have enough data right know to know if it's twice as dangerous for people under age 30, or half as dangerous. (We do already know it's -way less dangerous- for people under age 18).

Ok, maybe this dispute is based on different definitions. I don't think 2X is "approximately", but I do agree that we don't have enough data right now to know the exact difference so I can't say for sure that COVID-19 is more than 2X as dangerous.

I also agree that COVID-19 does appear to be way less dangerous for the under 18 group. However practically none of the under 18 group is able to isolate themselves away from the 18 and over group so there is no real societal benefit to this distinction.

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

#490

Post by greedo » Thu, Apr 23, 2020, 11:55 PM UTC

__blockcipher__ wrote: post #439

slg wrote: post #410

Earlier quoted context omitted.

I agree with some of what you said here, but there is one specific point I want to disagree with. All the projections for unemployment related deaths are based on society functioning as it previously did. I think many of the people who are suggesting we are in for long term isolation are also suggesting a much bigger increase in the social safety net to help people through these difficult times.

Right, but I think some of that suicide rate comes from the lack of "purpose" (it's silly that we rely on our jobs for purpose but we really do), or more broadly the desynchronization of one's internal schedule that many of us have experienced (which leads to worse sleep and therefore higher mortality). I also think that given we know about the level of competence of our government, it will be very difficult for "rea…

You are missing the point in that if nothing was done in terms of trying to (at least temporarily) contain COVID, then the economy would be just as bad off.

We're currently at 50K deaths in the US, largely in a four week period. I think it's safe to say that the shutdown has cut the death rate in half, if not more. If 100K people die in 4 weeks in the US, the people will shut the economy down whether the government initiates it or not.

Either way we're screwed. And I think your data is far off in terms of IFR and CFR. Imagine the IFR being a magnitude higher for COVID versus influenza. I personally think this is a safe guess based on what we've seen in a wide variety of environments, complicated by different testing methodologies etc etc. So we'll use 1% for the IFR.

Now current thought is the R0 is between 2 and 5. Hopefully closer to 2 since R0 dramatically affects the percentage of infections required to reach herd immunity. But to be simple, we'll assume we need 50% to reach herd immunity. So we'll use 164M as the number of Americans needing to be infected before we can "return to normal."

With 164M cases, and an IFR of 1%, 1.6M Americans will die. An additional 9% will need hospitalization, but survive. That's 14.8M.

Say we're off by a lot and "it's just a flu" with an IFR of 0.1%. You'll have 164K fatalities, and we'll still have the 14.8m hospitalizations. Say the hospitalization rate is off by a magnitude, it's still 1.48M.

There's no way our country can operate "normally" with these types of numbers.