Hacker News Reader

Comment by dehrmann | Hacker News Reader

Only related because NY, but NY state covid 19 new cases have been incredibly flat since June. I would think they'd trend closer to zero, up some, or just bounce around more. Normally, new cases are driven up by complacency and down by fear or policy, but there doesn't seem to be either happening quickly and subtly enough to keep new cases between 500 and 1000 for so long, and r0 should be metastable, so keeping it in that range for so long is unexpected.

View on HN

Yes there's something going on we don't understand, because some places that were hit very hard first time round are not seeing a second wave.

Perhaps the "something" is simply antibodies and herd immunity?

1. In April (!!), 1 in 5 NYCers tested positive for antibodies: https://www.nytimes.com/2020/04/23/nyregion/coronavirus-anti...

2. Given asymmetric social contact patterns, experts estimate that ~20% may be the threshold for herd immunity of COVID: https://theconversation.com/coronavirus-could-it-be-burning-...

I support looking at the science, and this means looking at it both ways; not just cherry-picking doom and worst case scenarios while ignoring best case scenarios.

We can't afford to let an "overabundance of caution" kill businesses and our economy, for good.

That seems plausible. Both New York and New Jersey have lots of high density areas, yet their numbers are very flat now. There's also some speculation that some people might have a natural immunity.

It likely happened in Florida and Texas too. Their profile basically matches New York.

That might explain it--this is just a baseline level of spread and herd immunity knocks it down. It also begs the question of how to loosen restrictions because if NYC actually hit herd immunity for the current level of social interaction, not relaxing restrictions is unethical. You can't keep everything closed forever, and if there's a sign you've hit herd immunity, 3+ months of this is it. It might also be what spread looks like post-vaccine. How to know you're at that point where you can reopen is tricky.

To add to this, Sweden isn't seeing the big surge that France and Spain are seeing (among others, including Belgium and the Netherlands, both at or near record highs). Sweden's cases have dropped lower and have stayed there. Given their comparatively open approach to the virus, they should be seeing a surge. While it's possible it'll show up in the near future as it again spreads across parts of Europe, I think it's also plausible they've reached a partial herd immunity. The commonly repeated premise that you need ~60% to reach herd immunity is plainly false, that's a calculation meant for a vacuum. In reality people aren't numbers on a sheet or living in a vacuum, all people don't behave the same way, don't spread a virus the same way, aren't all similarly at risk, some large percentage of the population may already have immunity thanks to prior exposure to other coronaviruses, and so on.

The 60 percent figure is people living their lives normally. Currently most people are social distancing, wearing masks and large gatherings have been cancelled. Under these circumstances 20 percent may well be herd immunity. But if Madison square garden was open and the subway was packed with people not wearing masks, its not clear that 20 percent would be enough.

I'm one of those who tested positive for IgG antibodies in April. Same in June, August. These were Mount Sinai's quantitative research tests, with a high titer. I donated plasma twice. The only symptoms occurred in mid March. Then I got an Abbott qualitative IgG test earlier this month, because I was about to fly to Italy. Still positive. Once there, to end quarantine before the customary two weeks, I got two negative swabs and another Abbott test, but this time they also tested IgM (short term response), which quite surprisingly still came back positive. Since this was the first time it ever got checked, it's not clear if I was IgM positive for the entire six months or I got some exposure in August or early September that didn't result in symptoms or infection.

I don't know what this means or if I'm just a rare case, but my inkling is still to explain NYC's current rates with mask wearing and reduced indoor interactions. I guess we'll see in the next four to six weeks.

> my inkling is still to explain NYC's current rates with mask wearing and reduced indoor interactions

Stockholm, Sweden has better numbers than New York yet people don't wear masks here nor are stuff closed. So you can't really say for sure that New York's numbers are thanks to those.

I don't know how to compare and control for different population mobility between the two or any other pair of cities in the world, which is why I mentioned that we'll see in a few weeks, i.e. comparing NYC to itself.

It's not antibody immunity. That's why I said we don't understand. 20% is nowhere near enough to explain the patterns we're seeing.

The good news is that even though we don't understand it, the NY experience does point towards a potentially more muted second wave in areas which were hit hard the first time round (as long as they take precautions at least), but we just don't know enough to be sure - anyone who is sure about the outcome at this point is probably overconfident and wrong.

20% of the total population is probably 60% of choiceless and careless. Percentage of the population that are exposed to a potential Covid infection, either due necessity (essential workers and so on) or stupidity is much less than 100%.

Someone who has been barricaded inside of their home since February most likely doesn't have antibodies, but at the same time is not a potential Covid spreader.

20% is nowhere near enough to explain the patterns we're seeing.

... according to epidemiologists who have been utterly discredited, no. According to other epidemiologists (and just lay amateurs) who have built better models, 20% appears to be about the right threshold when taking into account non-homogenous social mixing and t-cell immunity.

But you don't really need anything fancy to figure this out. Just look at the observed data, messy and contradictory though it is. When antibodies reach 20% the epidemic stops, in many places not just NY.

cross-immunity

Getting hit hard is probably why they don't see a second wave. They got the whole thing over with in one wave.

I think spring was "the wave".

In Nashville they closed bars and restaurants for 6 months because .004% of the cases came from bars and restaurants.

We need to give people choice.

If we wait until we get a vaccine it will be Fall 2022 and many Americans won't get it. What then? No bar and gym re-openenings until X percent of the population is verified to be vaccinated? Mandatory vaccination checks on entrance? Punt again and find reasons not to re-open until 2022?

Let people and businesses make their own choices.

Allow all re-openings and let people choose.

> In Nashville they closed bars and restaurants for 6 months because .004% of the cases came from bars and restaurants.

You're taking the number of cases that they've managed to contact trace to bars (difficult, especially if lots of cases all at once). You're then dividing that by the total number of cases, including the ones after they closed bars and concluding that it was the incorrect choice.

There’s email chains that came out showing that they knew the bar numbers were low and buried them. Then lied about it when asked.

Whether you think it was the correct choice or not is politics. It was not data driven nor supported by any actual exposure data that they collected, they knew that was the case, and tried to cover it up.

No am I am not but I understand your assumptive nature. I did speak incorrectly. Here is a repost of my previous comment:

https://www.dailymail.co.uk/news/article-8743605/Nashville-m...

In Nashville, TN (USA) bars and restaurants are closed to stop the spread of the virus. In e-mails obtained by a local television station, and later confirmed authentic by legal council, the e-mails discussed how only 80 of the 20,000 positive cases could be traced to bars and restaurants.

> Tennessee Lookout reporter Nate Rau asks, “The figure you gave of 'more than 80' does lead to a natural question: If there have been over 20,000 positive cases of COVID-19 in Davidson and only 80 or so are traced to restaurants and bars, doesn’t that mean restaurants and bars aren’t a very big problem?"

A day after the story broke they allowed bars and restaurants to re-open:

https://www.tennessean.com/story/money/2020/09/17/nashville-...

Indoor recreation was shut down in this city for 80 cases in a city of 2 million aka .004% of the population.

Point zero zero four percent.

https://news.ycombinator.com/item?id=24535010

> a city of 2 million aka .004% of the population.

Oh so it's the population now? I forgot that the whole city of Nashville was infected.

Not going to respond further.

You're done because we both know the math is correct, my argument is sound, and you don't like that. I see it often in these types of debates. It's easier to walk away from a losing position than to defend it.

I'm not sure what you think I changed but the math is fairly straight forward.

Nashville metro is 2 million people. There were 80 cases confirmed traced to bars and restaurants. They were keeping bars and restaurants closed because they were "hotspots".

When you take 80 and divide by 2,000,000 you get .00004 or .004%. Therefore, the government was keeping bars and restaurants shut down because .004% of the infections came from bars and restaurants.

They restricted indoor recreation for 2 million because of 80 cases.

Actually your argument is not sound as the person you are replying to already pointed out in a previous response to you that you seem to have missed. You actually contradicted yourself. Note:

"There were 80 cases confirmed traced to bars and restaurants."

"[80] of the infections came from bars and restaurants."

Those two statements are not the same as each other - the first, from the authorities, is true, the second, from you, is almost certainly not true unless every single infection from a bar or restaurant was actually traced.

Please, please don't use "you stopped responding, so I'm right". It is one of the most infuriating (and also pretty insulting) things you could have said. People stop responding because they have better things to do with their time, and/or because they know an interaction is bad for them or for the other person (e.g. is enraging them), and/or because they took a while to work out that they were arguing with someone who cannot change their mind, and/or even because they can't face being wrong. There are dozens of reasons for refusing to engage in a pointless and enraging discussion.

And provide a safety net for those people who choose not to risk their health to work in bars and gyms, right?

I wonder how much people changed their behavior in sunbelt states in July when cases were on the rise. I don't remember lots of new restrictions or lockdowns. New cases just kinda peaked.

I'm cautiously optimistic that the past week in the US is a blip from Labor day. Both peaks and troughs are improved from two weeks ago, but up from the week before.

There's definitely a dynamic with x% of the population distancing always, x% of the population not social distancing always, and then a certain percent that are oscillating between getting afraid and then feeling safe again - and that's what is keeping the replication rate around 1.

You can get insights into behavior with rt.live

There were supposed to be spikes after spring break, Memorial Day, 4th of July, Labor Day, back to school, the Sturgis bike rally. There were a blip...if that.

Rather than biting our nails at daily numbers it's time to allow for individual choice and remove the government restrictions. It's clear that the virus is not nearly as potent as people think and we're prolonging this due to fear and politics - not science or numbers.

I am pro-choice.

Let us choose.

Individual choice is not really applicable in this situation. Everyone that gets sick becomes very likely to get someone else sick. You're not just choosing for yourself, you're choosing for them, which isn't good.

And there have been lots of spikes among the groups that actually changed behavior drastically, especially in places like colleges.

For some definition of "hard time:" Sweden. On the other hand Spain and France are pulling an America right now. They're currently worse than the US for new cases per capita.

The positivity rate of testing is 1% and below. Guess what the false positive rate of the actual test is?

What’s more likely, the virus actually has a perfect 1.00 Rt factor, or there’s a systemic error introduced by humans?

Reply on news.ycombinator.com