Hacker News new | past | comments | ask | show | jobs | submit login
[flagged] Herd Immunity (wikipedia.org)
46 points by EndXA on March 14, 2020 | hide | past | favorite | 72 comments

Reading the simple formulas in this article, it looks like the present strategy of "social distancing" is intended to reduce R(0), the number of persons infected by each infected person. Pretty obvious, but still it's good to have a clear understanding.

It would be good if the same strategy could somehow be applied to the person-to-person spread of misinformation. But that's harder, witness some of the comments here.

> person-to-person spread of misinformation

You know, for as much as I've heard about the supposed danger of "misinformation", so many factual observations have been labeled "misinformation" early on, only to be later confirmed by WHO, a major national authority, or a medical journal.

The type of misinformation matters a lot, because if people take the concerns about permanent lung damage, reproductive damage, and the extreme contagiousness of the virus seriously, it doesn't matter if they were "misinformed" to some extent. If the misinformation encourages them to not take things seriously, or to breach social distancing (see Rudy Gobert, who should hope he hasn't committed homicide for a laugh), that's when it is the most dangerous.

There's been talk that previous coronaviruses don't cause immunity and there already seems to have been cases with this one where a person got it twice. Does anyone have insight?

I have yet to hear this from a credible source. It appears to be a self perpetuating rumor.

All it would take us a few people who catch Coronavirus in addition to one of the many things floating around with similar symptoms to think they caught it twice. It’s just too easy.

There have only been a handful of such reports with COVID-19, and they’re probably false test results.

> "there already seems to have been cases with this one where a person got it twice"

Just so people know: this has not been confirmed and has been an unsubstantiated rumor for a while. It is pretty unlikely this will happen considering what we know about how things like this function on a physiological level. It is much more likely due to the changes is testing (mostly where on the body samples were taken), confusion about symptoms or false negatives.

I've read that coronavirus is a near relation to the common cold virus. How significant is it that catching a cold doesn't seem to give you immunity from catching another?

That's because the common cold mutates very quickly, so by the time you catch another cold it's effectively a new virus.

The "common cold" is a catch-all phrase for many different viruses, which are not all closely related to one another. The coronaviruses that circulate in humans (other than SARS-CoV-2) are common colds. My understanding is that they do not mutate quickly, but that for whatever reason, one infection does not confer lasting, complete immunity to later reinfection (though later infections may be less severe).

Is coronavirus different in that respect?

If we can use data related to Human Coronavirus strains (HCov), which are responsible for many colds, then not that much different.

Of course, SARS-CoV-2 could be different. For SARS-CoV and MERS-CoV, there was no data on this.

Over 200 different viruses can cause a cold.

Rhinoviruses are a totally different "animal" compared to COVID19.

You can get colds from different viruses.

The second infection is likely about the virus reaching the lungs while it did not make it there previously. Meanwhile, asymptomatic carriers etc. are still being researched.

We don’t know but it is expected that immunity lasts for at least 6 months. Time will tell, this thing hasn’t been out long enough.

citation please

No reported reinfections and we’re 4-5 months in.

Have people actually been infected twice? I know people have been hospitalized for COVID-19, assessed as recovered, and released only to test positive later. But have they proven that's from a second infection, rather than a premature release or residual virus bits?

there may be people who continue to test positive, but there have been no confirmed instances of reinfection.


Not an expert, but this sounds like social media hysteria rather than science.

This is the best solution we can hope for now with Coronavirus IMO.

Slow the infection rate (but not too much) till we hit enough cases to have herd immunity, which will then slow the rate significantly till it mostly dies out (I suspect this has happened in China and South Korea, if not they will suffer the consequences later as the virus is reintroduced unnoticed from another country).

Since it won't be stamped out worldwide this year, there's no point in attempting complete eradication and tracing all contacts. That means the aim of reducing the spread should not be to avoid infections, but to to reach herd immunity in an ordered way without taxing healthcare systems too much on the way.

> I suspect this has happened in China and South Korea

How could it have if only a negligible fraction of the population caught it?

No an epidemiologist, but I think the theory is you only need herd immunity in the local area, if travel is also restricted, for the infection to burn itself out.

We really have no idea what fraction of the population caught it, since so many show minimal symptoms.

If it were not a significant fraction in infected areas I’d expect it to flare up again quickly. Quarantine simply cannot be maintained indefinitely, nor is it going to stop all spread.

Do we know how many people are already susceptible to it? It’s plausible that some percentage of people are immune to it already — maybe they recently caught a similar coronavirus or some other natural resistance to it.

Something seems a little off. If this thing truly incubates for up to 14 days, during which time it can be transmitted, it seems odd that so few people are ill.

Is it possible that many people, in fact, are infected but do not get sick?

It is likely that the early symptoms get ignored, as they are similar to a cold - except the nasal discharge and phlegm, which SARS-CoV-2 does not seem to cause. (It gives dry cough instead of wet cough.)

14 days is not typical. It's more a conservative upper bound used for quarantine.

Most people do not get sick.

That's a very good question. I have not seen anyone address it yet. I guess there should be probably similar research on other viruses.

This recipe is a fantasy.

And it is not what has happened in any population except maybe a few groups of medical workers who were treating patients in outbreak zones.

But it's a fantasy because of how slow the controlled spread would have to be.

Some of the best medical systems in the world have been pushed into chaos with less than 1/1000th of the population infected. This level has increased the mortality rate by two to ten fold, depending on age group (with younger groups seeing a much greater increase).

If each 1/1000th of the population takes weeks to be infected and recover, then this plan will take at least a decade to reach the levels of exposure required for herd immunity. And by that point, the virus will likely have mutated into something that those previously infected can catch, destroying the benefit of the plan.

The only solution right now is total quarantine. Even maintaining infection rates of 1/1000th of the population is going to require that.

Some of the best medical systems in the world have been pushed into chaos with less than 1/1000th of the population infected

It's really impossible to give figures like that with any certainty, given so many people seem to be asymptomatic, and how concentrated some of the outbreaks have been. Which area are you considering where you think infections are 1/1000?

Whether it's a fantasy or not, our best hope at this stage is controlling spread. Total quarantine is simply not possible, since people need to go out for food, and will also increase infections within family groups stuck together in close quarters. There is no simple solution.

the best option is antivirals AFAIK, which will probably become available before the vaccine

The best solution will be mass vaccination. What you’re proposing amounts to millions people dead worldwide and a total collapse of healthcare.

How do you imagine "total collapse"? Hospitals falling down, no doctors left? It sure can and does overwhelm the healthcare system. Maybe you can even call it short-term collapse, but if it keeps functioning it is not total collapse.

Mass-produced vaccines are probably around a year away, in the meantime we need a plan.

yes, sit on your ass and wait until everyone (quite literally) can be tested and immediately quarantine any and all suspected infections and their contacts, all the while planning the logistics of vaccinating hundreds of millions if not billions of people before the next flu season.

To be clear, a strategy of allowing this to spread unchecked will result in 10s of millions of people dying around the world.

Herd immunity is the end-game scenario. All other measures should be done to decrease the infection speed to prevent the health system from melting down.

If you’re referring to the UK strategy, it is not to be “unchecked”.

I'm not sure how sound is UK's strategy.

Current advice is to self-isolate for 7 days if persistent cough suddenly appears and/OR a temperature of 37.5°C (that is not high temperature).

That sound like good advice, but they are not going to test those cases unless it ends in hospital. And they don't want you to call to the NHS help line or go to the doctor (again, not bad advice but I see some people getting anxious about this; IMHO at least the help line should be available because most people aren't doctors).

Considering that it could be a mild case of the disease or any other common seasonal disease (I have two toddlers at home: they get ill), how can we tell we had passed the disease and we have some immunity?

Besides, if testing and tracing stops, it means there won't be information about the progression of the epidemic in the UK; which sounds to me like a bad idea (but I'm not an expert).

Considering how the Conservative government has been underfunding the public health system for 10 yeas, I understand how their approach can be seen by some as untrustworthy.

EDIT: I see this may be off-topic re: herd immunity, sorry about that.

> Besides, if testing and tracing stops, it means there won't be information about the progression of the epidemic in the UK;

Yes, there will. It will be inferred from hospital stats.

Testing and tracing (apparently; I'm no expert, parroting what I've read from experts) is only useful in the contain phase. After that, you're better focusing on slowing the spread in the delay phase.

That said, Singapore and China have shown success with testing, tracing, and extreme measures. The assumption of the UK Government, however, is that in the long-term these measures will fail and kill more people than following a delay phase strategy.

Early on, doctors were quick to note that there is no herd immunity with this particular coronavirus.

No _existing_ herd immunity. That doesn't mean "herd immunity doesn't work for this virus" which is I suspect what you're thinking.

> Early on, doctors were quick to note that there is no herd immunity with this particular coronavirus.

Of course there was no herd immunity at the start - nobody had had it yet!

How can we be immune to a brand new disease?

It's currently unknown if there is an immunity after getting Corona.

That is just not true. There were a few number of cases, where a second test was positive after initial curing. This is being investigated and can be attributed to issues with the test, false positive diagnosis in the first place or various individual reasons. There is no evidence for a fundamental issue with adapted immunity to Sars-CoV-2.

You forget something. Maybe after first curing people get immunity and can get the virus again but it does not get active and test gets positive. People can have viruses and antivirus or a non active virus without consequences and then the test will be positive.

All signs point to the coronavirus infection inducing normal immunity. The number of apparent reinfection is extraordinarily low, consistent with being due to testing error, and in any case not relevant even if real if it only applies to a small fraction of the population. Furthermore, the virus does not have a segmented genome like influenza (thought to be responsible for its rapid evolution and ability to evade vaccine) and is a single "species" (unlike the huge family of viruses, mostly rhinoviruses, that cause the unvaccinable common cold).

That's great news. Got any links to the data?

I just found some strong counter evidence in tension with everything I'd been hearing:

Francois Balloux (Computational biologist, director of UGI at UCL): "How long immunity lasts for following covid-19 infection is the biggest unknown. Comparison with other Coronaviridae suggests it may be relatively short-lived (i.e. months). If this were to be confirmed, it would add to the challenge of managing the pandemic.


It's not inconsistent per say, since it still seems that the recent reinfections were likely due to test error or rare poor immune response, but my initial comment is misleading enough that I retract it (and would delete it if I could).

Corona viruses also have an error correction enzyme in their rna transcription so their rate of mutation is much lower than most viruses, it’s likely that both herd immunity and vaccination would be effective as a consequence.

An interesting side discussion to this regarding vaccines: Does the presence of vaccine antibodies exhibit selection pressure on viruses? That is to say, do vaccines inadvertently and indirectly create an 'arms race' scenario?

NOTE: I am not anti-vaccine. This is intended as a nuanced point

In some cases it does, in some it does not. And yes, the western world taking their flu shot does mean that those who can't afford it might have a problem or a worse problem compared to what would otherwise be. ( Also why many advocate you should only take it if you need it. )

This virus mutates slowly due to proof reader RNA. It most likely mutates too slowly to evade eradication, just like polio or smallpox.

Herd immunity until now referred to immunisation (ie vaccination). It is not at all clear that allowing people to get infected and recover will achieve the same result.

Herd immunity is a term describing a steady-state distribution and one avenue (the typical one) of getting there is vaccines. Typically when 80% of the population\* is immune for one reason or the other a population is said to have herd immunity. The idea is statistical safety for susceptible individuals (yes, typically those that cannot be vaccinated for $reasons) thanks to being significantly outnumbered by those immune/resistant to infection - or a buffer population acting as a firewall, if you will.

I haven’t seen the term used for small pox in the Modern Era, but I believe it would be fair to say even though the small pox vaccine hadn’t been invented (and wouldn’t be until 1796), there were definitely periods of herd immunity in major European cities/ports of the time as it absolutely destroyed the previously unexposed Native American population.

But when dealing with “learned immunity” due to previous exposure as compared to natural/genetic immunity to the same diseases and the antibodies are not passed down from mother to child, what you’re going to see is a resurgence in cases as each new generation loses that immunity. In the present case, the unvaccinated herd immunity would be something to keep spread in check until a vaccinated herd immunity can take its place (assuming the vaccine isn’t developed before 80% of the remaining population has been exposed to and has recovered from this particular variant of the coronavirus).

\* at least that’s the number we were taught in my livestock/farm animals veterinary medicine class a decade ago.

The exact percentage required depends on the R0 of the disease, the number of people each contagious person can be expected to spread the disease to assuming no prior immunity. I think you can think of it this way: if each person would otherwise spread the disease to two other people, but half of the population is immune, they're only going to be able to pass it on to one other person. In reality it's probably more complicated than that though.

Thanks! That makes a lot of sense. Certainly more than the pundits currently lambasting the UK health services for their strategy of flattening the curve. The current argument against being (I kid you not) that the health authorities are somehow stupid enough to think that immunity can be passed between individuals like the virus itself.

I appreciate you taking the time to correct me.

No problem. I wasn’t yet aware of the UK drama when I wrote that.

If this is going to happen and we don't get a vaccine in time one interesting possibility is to look at stuff like hydroxychloroquine as a prophylactic that seems pretty good in vitro and could maybe reduce the infection from very nasty to not too bad in the appropriate dose. I'm hoping for some trial and error research here.

paper showing it good in vitro https://www.ncbi.nlm.nih.gov/m/pubmed/32150618/

medcram talking about that https://youtu.be/vE4_LsftNKM?t=200

some semi evidence of it working in real life - the lupus patients didn't get it in Wuhan here https://www.jqknews.com/news/388543-The_novel_coronavirus_pn...

I was wondering how you could encourage people to try it and see if it works. My idea - if you got a thousand health frontline workers to try and compared with a similar number who didn't you may get decent data in a week or two if lots of one group got sick and the other didn't. Or it might not work. Don't know till you try though and could save many lives.

There was a lot of talk about this early on in China’s epidemic. Is there any newer evidence anyone’s seen? Seems to have gone quiet.

I wonder if this is would be more effective in the second stage of getting COVID-19, which is the immune system going haywire rather than the virus directly causing problems.

There's this trial where chloroquine worked well for treatment. (4.4 days to negative tests) http://news.southcn.com/nfplus/gdjktt/content/2020-03/09/con...

Tocilizumab seems good for the later stages http://www.ansa.it/english/news/science_tecnology/2020/03/11...

Prophylaxis, taking the drugs before catching the virus, remains a bit untested.

Thanks. The first one seems positive. I wonder why no other countries are using it then?

(Here’s a translate link for anyone who can’t read mandarin. https://translate.googleusercontent.com/translate_c?depth=1&...)

The second one is a different drug isn’t it? An arthritis drug rather than an anti-malarial.

It's getting very hard to believe the UK story. I mean, i get exceptionalism, but this is going too far.

The UK has some of the best epidemiology schools and experience, thanks to a tradition harking back to the British Empire expanding in very dangerous lands.


Imho, precisely this wealth of experience is driving a lot of people to overthinking it, looking too much at the long-term big picture and not weighting the immediate costs correctly. Some people have ended up in Strangelove territory and we risk paying a big price for it. Saving more people right now is worth, say, a longer impact on the economy or longer quarantine measures, if anything because those are social constructs that can adapt - whereas our bodies work only one way.

> The UK has some of the best epidemiology schools and experience, thanks to a tradition harking back to the British Empire expanding in very dangerous lands.

This is not an argument in 2020. This is the kind of exceptionalism i m talking about

I’m not the one saying that, Italian experts are saying it, because it increases the level of bafflement they experience when looking at the (non-)approach the UK is taking at the moment. I am extremely critical of this (non-)approach, I’m just trying to figure out where it comes from and this might well be a factor, sadly. After 2016, I’ve learnt never to underestimate the often-misplaced sense of superiority that the English upper classes have drilled into the wider population for the last 200 years, and the degree of detachment of their elites.

If they really know something the rest of Europe doesn't, don't they have a moral duty to release their data and models? This might save millions of lives when all the other countries realize they're doing it wrong.

Without such transparency, it's easy to assume they've calculated they'll lose more votes to a recession that they would to predominantly Tory-voting elderly people dying. It's not saying this is their actual motivation, but secrecy breeds suspicion.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact