2020 has been a truly strange year. Of all the articles that I never expected to write, one of the weirdest ones has to be a myth-busting piece about how herd immunity probably isn’t going to save us. I’m famously a fan of herd immunity — my first ‘viral’ article was about how herd immunity is pretty great — and when it comes to vaccines, I’m all for the idea.
Unfortunately, when it comes to COVID-19, herd immunity has never been the solution some people make it out to be.
The basic idea of herd immunity is simple — if enough people are immune to a disease, then even if a person comes into the community carrying the illness, they won’t spread it. The few people who aren’t immune are protected, because most of the people they interact with are immune and so they can’t get the disease.
This works really well if you have a vaccination for infections. Measles, for example, used to cause massive outbreaks and kill hundreds of thousands a year, but in most countries cannot spread further than a few kids these days. That’s herd immunity at work, and it’s a wonderful thing to see.
The problem with COVID-19 is that we don’t have a vaccine. To be immune to the disease, you have to get it and then recover, and we know that coronavirus kills a pretty scary proportion of people who get it.
We’ve known this for months. So why are people suddenly saying that herd immunity is just around the corner?
There are actually two arguments that have been going around implying that herd immunity is almost here. The first is pretty simple, and very obviously wrong: that 50% of people are already immune to COVID-19 due to pre-existing t-cell reactions that were probably caused by other coronaviruses (the ones that cause the common cold). This, the idea goes, means that only a few percent need to catch the disease to reach the herd immunity threshold, and so we’re probably already there and there’s nothing more to worry about. While it’s a popular line, it makes very little sense for a number of reasons. If nothing else, we already know, from superspreading events, that most people can catch COVID-19 if they are exposed and that pre-existing t-cells might make COVID-19 less deadly but it certainly doesn’t make you immune.
So t-cells, despite being very cool, aren’t our ticket out. But there’s another idea about herd immunity that’s very popular, which is to do with population heterogeneity. While it sounds complicated, heterogeneity basically just means “differences”, so population heterogeneity means “differences within the population.
See, traditionally the herd immunity threshold — the proportion of people who have to be immune to a disease for it to stop spreading — is thought of as a relatively fixed number. It varies, but it’s mostly decided by how infectious a disease is — more infectious diseases like measles require nearly everyone to be immune, while less infectious ones such as rubella infect fewer people.
But the problem with this idea is that it assumes that every person has the same chance to pass on the disease as every other person. So a recluse living in the wilderness is just as likely to spread disease as a server at a restaurant, which obviously isn’t true. This is the population heterogeneity that people are talking about — differences in how different people behave that may impact the herd immunity threshold and mean that fewer people need to get COVID-19 to stop the disease from spreading.
So, the theory goes, what if we took this heterogeneity into account? Well, some people have done that using mathematical models, and it turns out that they seem to show a lower number of people needing to get infected before the disease peters out.
The problem is that these models are, like all models, very susceptible to assumptions. One key assumption is that while this heterogeneity exists, it is pretty fixed — i.e. some people are more prone to spreading coronavirus, and that this is fixed through time. So, for example, people working at a meat-packing plant might be more at risk of spreading COVID-19, but once all meat-packing workers are infected the disease will slow down considerably.
Unfortunately, this assumption may not be true at all. As the authors of the paper linked above note, if people move around inside society a lot — if they change jobs, join choirs, get over their personal space issues and start hugging more — suddenly the herd immunity threshold jumps right back up. Given that many of the places where COVID-19 spreads best are typically hotbeds of casual work, with people moving in and out all the time, it’s not unreasonable to say that heterogeneity might not impact the spread of coronavirus all that much.
The answer to this seems to be that some places are already experiencing a form of herd immunity. Sweden, for example, has been painted as a notoriously lax country where coronavirus was allowed to run free, and they are seeing relatively few new cases of COVID-19 now. This is evidence, the line goes, for herd immunity even at low levels of infection (Sweden has an estimated ~15% of the population with antibodies to the coronavirus).
Except, this explanation doesn’t actually make much sense. Firstly, as laid out in the original Swedish plan, they still have a number of COVID-19 restrictions, making Sweden actually more restricted right now than many other places in the world. For example, it is illegal to have more than 50 people at a public gathering, and Sweden has strict rules that all restaurants must follow.
And despite this now relatively high level of coronavirus restrictions — more intense than many states in the U.S. — Sweden is still experiencing hundreds of new cases a day. Yes, the rate has subsided a bit, but the outbreak is still ongoing. If herd immunity had kicked in, as so many have argued, the rate of infections would mostly die down — instead, it’s still increasing, albeit a bit more slowly.
This actually gets to the heart of the issue with herd immunity for COVID-19. It doesn’t really make sense to call a situation where a disease might slow down its spread somewhat ‘herd immunity’ in the first place. The whole idea of herd immunity is that it is a long-lasting phenomenon where the disease won’t spring up again — hence the word ‘immunity’. That’s why we vaccinate, because we want the disease to die down or potentially die out entirely.
The situations that most people are describing as herd immunity are actually more like endemicity, where a disease stays prevalent in a population, and sometimes dies down, but never completely goes away. Even in places where 20–30% of people have been infected, and 10,000s have died — like New York City — we would expect there to be some transmission of COVID-19 and regular outbreaks if all restrictions are relaxed.
So no, we haven’t reached herd immunity yet. It is, in fact, quite unlikely that we will somehow see the end of this disease with only a small proportion of people infected, until we have a vaccine. It’s possible, based on some mathematical models, but it’s definitely not certain and I would say not even that likely.
The reality is that herd immunity remains a dangerous path to tread without a vaccine. We know that a large proportion of people who catch COVID-19 die, and that the disease is deadly even for people in their 40s and 50s. The fact is that large numbers of infections inevitably means large numbers of deaths, and even death is just one outcome for coronavirus. We still don’t know exactly what long-term impacts the disease will have for people who catch it and recover, but we know that there probably will be some.
Shockingly, a strategy that requires most people to get infected with a dangerous disease might not be the best idea. Go figure.
Ultimately, the best solution is probably to keep at it despite our exhaustion — keep social distancing, wear a mask, and try to stay away from house parties. Herd immunity may sound like a silver bullet to our current plight, but as I wrote in April it requires us to sacrifice the vulnerable on the altar of the economy. That hasn’t changed, and probably never will.