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

Covid herd immunity is probably impossible

nature.com

Post reply on HN
Print view View original

161–165 of 165 posts

Re: Covid herd immunity is probably impossible

#161

Post by xyzzyz »

maxerickson wrote: post #158
rhino369 wrote: post #157

Earlier quoted context omitted.

A lot of Europe tried. It's really hard unless you can control borders. Even if the US locked down for 3 months and got to zero, it could be undone by one group of illegal immigrants crossing the border with COVID. The US and EU were already pretty well infected before they realized what was happening. Only China, S. Korea, and Australia had major outbreaks that were contained, and even those, the outbreaks were fair…

There is some threshold below which contact tracing and isolation work. A few dozen infected people crossing the border could be dealt with. We've not bothered to do much support for isolation, have high case counts and a year of encouraging people to respond poorly to contact tracing, as if consenting to disease control measures is some unforgivable affront to autonomy.

> A few dozen infected people crossing the border could be dealt with.

There are something between 50 000 and 100 000 people each year who are caught trying to illegally cross southern US border. The number of people who are not caught is obviously a multiple of that. This probably could be dealt with, if the government was actually trying to do that.

Re: Covid herd immunity is probably impossible

#162

Post by Out_of_Characte »

cogman10 wrote: post #100
Out_of_Characte wrote: post #78

Earlier quoted context omitted.

In the Netherlands, shopping carts and masks became mandatory in shops. The shopping cart is/was used as a way to keep track of the amount of people allowed in the store. This had the counter-productive effect of having to have someone disinfect shopping cart handles with alcohol or in worst case a do-it-yourself system where disinfectant is provided but only AFTER you've already had to touch the cart of physically p…

Copper helps, but it isn't magic. It takes about 1->2 hours to self disinfect. You'd still need a step of disinfecting the handles. That said, I'm not sure how often flu goes from cart handle to person, I've my doubts that it's a major disease vector. I'd think just being in a store with a bunch of idiots coughing on each other probably spreads the flu more readily.

I think its a better option than do-it-yourself disinfection.

Really, coronavirus presents itself as a virus that is not stopped by any means we currently have employed. on the micro level all these techniques are tried and tested (by the likes of almost perfectly sterile enviroments as surgical rooms) but at the macro level coronavirus spreads trough humans in a way that cannot be seen or prevented.

Its not like we ever stopped flu season.

Re: Covid herd immunity is probably impossible

#163

Post by lamontcg »

TMWNN wrote: post #151
corty wrote: post #146

Earlier quoted context omitted.

Maybe not even that, it could be a typical childhood ailment that you get once in your life and never again.

I suspect that, in retrospect, the first inclination of many parents when the pandemic began to hold "COVID parties" for children like the familiar "chicken pox parties"—inclination that wasn't followed, because of dire warnings about long-term health effects—was the right one.

Fairly definitively that was always the wrong answer.

In the future:

1. Most people will get SARS-CoV-2 antibodies from their mothers

2. Early life exposure will happen

3. The SARS-CoV-2 virus will be force to make expensive mutations to avoid immune escape will come at a cost to fitness.

None of the conditions can be created by what you suggest. You have the problem of having the relatively most-lethal / most-transmissible version of the virus spreading unhindered in a population of >40 year olds who have no natural immunity. The "COVID party" solution maximizes the death and cost of the virus.

Re: Covid herd immunity is probably impossible

#164

Post by a-dub »

varajelle wrote: post #136
a-dub wrote: post #103

Earlier quoted context omitted.

yes but we are unable to vaccinate fast enough. virus is n^2, vaccine is n you cannot beat exponential growth with linear growth

Where did you get that virus is n^2 ? If I'd have to guess, I'd say it is O(R^n)

it's tempting to think that, but no, in the simple non-mutating case, the exponent is fixed. the growth rate is exponential but the exponent does not grow with spread. they say the exponent ranges from 2-5 for covid, and drops with mitigation methods. that seems to clearly put it in n^2 to me.

that said, the mitigations do cause the exponent to fluctuate... and in countries where people are willing to cooperate it has dropped below 1 and the epidemic has been mitigated.

it is possible that this is actually a case calling where asymptotic analysis is inappropriate and in fact you have to resort to the raw data and measurement.

now that i think of it though, if you allow for mutations that are more infectious, then the exponent is not fixed and increases with spread... yikes.

Re: Covid herd immunity is probably impossible

#165

Post by varajelle »

a-dub wrote: post #164
varajelle wrote: post #136

Earlier quoted context omitted.

Where did you get that virus is n^2 ? If I'd have to guess, I'd say it is O(R^n)

it's tempting to think that, but no, in the simple non-mutating case, the exponent is fixed. the growth rate is exponential but the exponent does not grow with spread. they say the exponent ranges from 2-5 for covid, and drops with mitigation methods. that seems to clearly put it in n^2 to me. that said, the mitigations do cause the exponent to fluctuate... and in countries where people are willing to cooperate it ha…

I'm somehow still not convinced by your explanation. What exactly is N and what exactly is this measuring. I thought that N was a time period, and that the equation is something like

    Infected count = O(R^t)
But apparently you mean that N is the number of infection? And what are you measuring?

    ??? = O(n^R)
Where the ??? is also linear with the amount of people vaccinated?
Post reply on HN