Imagine we develop a vaccine against the coronavirus (COVID-19). Suppose the vaccine has some very small chance of some serious side effects, for instance seizures. However, this vaccine can save millions of lives globally, in the same way as other vaccines do. You are the prime minister and you have to decide whether to make the vaccine compulsory. You know that many people are opposed to vaccines and especially compulsory vaccination. They would not vaccinate even if failure to vaccinate enough people would result in a pandemic that would kill millions of people worldwide and devastate the world’s economy. They think that the state should have no business in telling us what should go into our bodies, that vaccines are harmful, that infectious diseases are no riskier than vaccines, and that the claims about the COVID-19 outbreak have been exaggerated or even made up to promote compulsory vaccination campaigns by Big Pharma.

The example at this point is no longer purely hypothetical, unfortunately. Even now, as the coronavirus pandemic is getting worse and worse and killing an increasing number of people every day, there are those who would not want to vaccinate against COVID-19 if this vaccine was available, for all these reasons. One of us wrote a short article three years ago defending mandatory vaccination policies. As the latest contagion started spreading in the US, people started commenting on that article again, suggesting that this virus is a scam and that various people have engineered it to boost pro-vaccine campaigns. For instance, someone wrote: “What a crock (…) this is from 2017 and three years later Italy is right up there with S Korea and China for Coronavirus cases. Yeah that crap doesn’t work”. Other comments are more explicit.

Now let’s jump back in time. Suppose you are the prime minister in the early 1990s. You are dealing with a different public health problem: the very high number of victims of car accidents. Cars are not as safe as they would be in 20 or 30 years (no airbags, no anti-lock brake systems, and so on). However, some other countries have started introducing seat belt requirements. You look at the figures and you see that seat belt requirements significantly reduce the number of injuries and deaths from car accidents. For example, it is estimated that risk of death is reduced by 45% and risk of serious injury by 50% in a country like the United States. However, you know that many people are opposed to seat belt mandates, because they think that the state has no business in telling us what risks we can take for ourselves and for our children, that it is not true that seat belts improve safety, and that seat belt requirements are the result of lobbying from the seat belt industry. You might be a bit more surprised than in the case of vaccination to learn that also this example is not purely hypothetical; this is precisely what happened when seat belt requirements were introduced between the ’80s and the ‘90s.

Vaccines are like a seat belt against infectious diseases, and that vaccination mandates are justified for the same reasons seat belt mandates are. Actually, the justification is stronger when it comes to vaccine mandates. If you fail to buckle up yourself or your child, you are putting yourself and your child at unnecessary risk, and perhaps other people at some small risk too (for instance, those who are not buckled up in the back seat are more likely to kill or injure those in the front seat in case of accident). But if you fail to vaccinate yourself or your children there is a much larger risk of harming other people who are not vaccinated (for example those who are immunosuppressed, young children or those whose immunity has waned over time), besides imposing an easily preventable risk on you and your children.

Now, here is the elephant in the room. Vaccines might entail some small risks. For instance, it seems that in very rare circumstances (one to two cases per million doses) the flu vaccine could cause Guillain-Barré Syndrome, although the evidence is not clear and the cases are so few that it is difficult to provide a statistical estimate.

And here is where the analogy with seat belt requirements becomes very effective. Seat belts, in rare circumstances, can cause injuries and even deaths that would not have occurred if people had not been buckled up. Sometimes the dynamics of car accidents are such that people would be better off without seat belts, e.g. when seat belts prevent them from getting out of the car fast enough to avoid drowning or fire. And there are injuries that form a specific injury profile labelled “seat belt syndrome,” which can be serious and require surgery.

And yet, the vast majority of people support seat belt mandates and eagerly buckle up these days. Wearing seat belts has become not only a legal requirement, but also a social norm in the vast majority of countries. And this is how it should be, because seat belts do save many lives. As vaccines do, in spite of very small risks, in both cases. Injuries from car accidents are way more common and way more likely to be lethal than injuries from seat belt use. And in the same way, many infectious diseases, such as measles, are way more likely to harm and to kill than vaccines.

If the very small risks of seat belts are not a good enough reason against seat belt requirements, then the very small risks of vaccines are not a good enough reason against mandatory vaccination.

This opinion piece reflects the views of the author, and does not necessarily reflect the position of the Oxford Martin School or the University of Oxford. Any errors or omissions are those of the author.