Hacker News Reader

Comment by jupp0r | Hacker News Reader

Keep in mind that "herd immunity" isn't really immunity, it's the point at which Rt (the average number of people each infected person passes the infection on to) drops below 1.0 and the spread shrinks instead of growing. Rt is dependent on how people behave. When behavior changes, Rt can change as well. Each herd immunity level is thus dependent on health measures, which is why "reaching" herd immunity and then loosening up health measures won't work.

Edit: removed wrong information on R0 that's not really essential to my point

View on HN

When non experts talk about herd immunity they mean herd immunity given no to minimal precautions. This is a well defined concept.

If that's an informal lay interpretation I don't think it can be well defined, by definition.

Yeah. The term "endemic steady state" would be more accurate, but "herd immunity" is the term everyone ended up using.

> R0, which is a constant inherent to the virus strain

I think you just devalued your argument.

https://wwwnc.cdc.gov/eid/article/25/1/17-1901_article

Thanks for pointing that out, I corrected the post. It wasn't really important to my point though, which is that we might have enough immunity to let the infections die without football games, while we still have growth with football games.

It would have been more effective if you left it at that. I'm trying very hard to bow out of HN discussions on COVID, you were making a good point in a bad way and it would be a waste to see it lost. Thank you for the correction.

I think a big part of the confusion here is how do you clarify the meaning of R0 versus Rt in a context where people don't know the difference, and are using R0 in a context where there is either a time component or obvious public health interventions.

R0 may be a weak theoretical construct, but saying that it's also contextual and not inherent to the virus doesn't do much to clear things up.

http://web.stanford.edu/~jhj1/teachingdocs/Jones-on-R0.pdf

R0 is not a constant inherent to a virus strain. It's a contextual number, determined by population and behavior, population immunity and other factors.

Rt is simply notation of an estimate of R0 at a particular time.

Either way, you're correct that "herd immunity", as used here, means the point at which time the infection rate begins to decline, and this is conditional on population behaviors. If people mix more freely, the estimate changes.

However, the observation that people don't mix uniformly still applies, even if they mix a bit more than they do now. To put it in a CS context, it's like debating the magnitude of the constant, when the algorithm has a fundamentally different asymptotic behavior.

> R0 is not a constant inherent to a virus strain. It's a contextual number, determined by population and behavior, population immunity and other factors.

From Wikipedia: In epidemiology, the basic reproduction number, or basic reproductive number (sometimes called basic reproduction ratio or basic reproductive rate), denoted {\displaystyle R_{0}}R_{0} (pronounced R nought or R zero),[20] of an infection can be thought of as the expected number of cases directly generated by one case in a population where all individuals are susceptible to infection.

https://en.wikipedia.org/wiki/Basic_reproduction_number

You can think of it that way, but we never know, in practice, what level of actual immunity exists in a population.

In any case, it's not relevant to my point: R0 is a contextual number, always defined by empirical data. It's not a fixed feature of the virus.

Exactly. OP is simply wrong about this.

> can be thought of as the expected number of cases directly generated by one case in a population where all individuals are susceptible to infection

Yeah, and that is contextual number, determined by population behavior and other factors.

Yes, population immunity should not be on that list, but the population behavior, weatcher and what not are still influencing it a lot.

Population immunity has to be on the list: if the population has some level of immunity, it affects the observed R0. And we can't really measure immunity, other than in very crude ways (i.e. antibody tests for specific epitopes), so we can't control for it.

This isn't a political statement of any sort.

The next paragraph calls out that R0 is not a biological constant specific to a pathogen as it is affected by environment and behavior.

> Each herd immunity level is thus dependent on health measures, which is why "reaching" herd immunity and then loosening up health measures won't work.

Yep. Here's a very approachable and well written paper on the topic: https://academic.oup.com/cid/article/52/7/911/299077

And my comment on it from 6 months ago: https://news.ycombinator.com/item?id=22818413

> Each herd immunity level is thus dependent on health measures

This seems like something you made up. Can you cite your source?

How exactly does your rule apply to non-humans? What "health measures" do packs of wild horses take when disease comes to the herd?

When cows get mad cow, you quarantine and then kill potentially infected cows. Wild horses have behaviors. They may not change their behaviors when they find a virus, but that doesn't mean that the reproductive rate is independent of the horses behaviors. It absolutely still is.

It doesn't apply to non-humans because herd immunity is not something that is at all related to non-human populations. The term was coined to describe a trait of human populations and was found to be difficult if not impossible to achieve in humans until we started to vaccinate.

https://en.wikipedia.org/wiki/Herd_immunity#History

Reply on news.ycombinator.com