We should not vaccinate the young to protect the old: a response to Giubilini, Savulescu, and Wilkinson.

A recent article by Giubilini, Savulescu, and Wilkinson posed a timely question: should we vaccinate children to save the old? According to the authors, this option could be acceptable if a number of conditions applied. In this piece, I hold that its main claim—that using children to fight epidemics is morally acceptable if this only causes them little harm—is fair. However, this is only true if the idea of ‘small risk’ does not equate to high risk to a small number of children. On the other hand, my reply shows that the acceptability of vaccination of children should not be based on making exceptions to the Kantian categorical imperative, but to the simple fact that we can rely on a presumption of consent to altruistic behavior by children. Nevertheless, I also state that this option would be the best strategy in the COVID-19 scenario. Isolating children through the performance of tests in schools and institutes could be a better alternative, since it would hinder their spreading of the disease without depriving the elderly of a scarce resource, namely the vaccine.

I. INTRODUCTION

A recent article by Giubilini, Savulescu, and Wilkinson exposes an ethical issue that deserves careful analysis: should we vaccinate children to save the old? The answer provided by the authors is ‘yes’ under the condition that some circumstances apply, namely: (i) the vaccines provide better results in children than in old people; (ii) children are a vector for the contagion of the elderly, who are not adequately protected by the vaccine; (iii) there is only a small harm to the children; and (iv) this demonstrates to be the most effective strategy to save the greatest number of people. If all of these conditions apply, then we should target children for vaccination instead of the elderly, even though this might be considered as using children as mere means. In this comment, I would like to address some of the statements made by the authors, clarify some misconceptions and provide reasons to support that vaccinating children instead of the elderly is not a good idea in the case of COVID-19.

II. CLARIFYING A MISCONCEPTION. VACCINATING CHILDREN WOULD NOT BE ‘TREATING THEM AS MERE MEANS’

First, the article makes quite a strong claim by stating that sometimes ‘treating people as means in a way that is harmful or disrespectful to them and without their consent may be considered permissible’. I cannot agree with the authors on this. I certainly hold that Kant’s categorical imperative (‘[s]o act that you treat humanity, whether in your own person or in the person of any other, always at the same time as an end, never merely as a means’) is a boundary that we should not cross. However, this does not imply that we should not vaccinate children. Contrary to Giubilini, Savulescu, and Wilkinson, I do not consider that we will defy this rule if we proceed to do so.

I concede that this issue is open to discussion since the concrete implications of the categorical imperative are hard to draw. Indeed, philosophic discussion about the meaning of the term ‘merely’ in this mandate continues and will probably last long. Nevertheless, I think that the examples provided by the authors to show that sometimes breaking the Kantian mandate might be permissible are not good examples. The case of the folic acid intake in pregnant women only shows that mothers are willing to sacrifice themselves and take some risks to benefit their children. The second case—the pediatric bone marrow donation—seems closer to the ‘using somebody as a mere means’ scenario. However, I do not think that the analogy works either. Lots of people donate bone narrow, blood, or even organs, and we usually do not consider those actions as a kind of exploitation but an expression of altruism.

Of course, in the case of children there is a major difference compared with adults: they cannot give consent. However, this does not imply that they are used as mere means. We can apply the presumed consent concept perfectly in this case. We can assume that if children were able to consent to altruistic initiatives that only create a small risk or harm to them, they would do so. We proceed this way, for example, when we decide to isolate them even though the disease probably causes them less damage than the process of isolation. Thus, we can agree on the idea that children would accept a small sacrifice that could serve well to alleviate other people’s suffering. Said in other words, we could consider that children would not be reluctant to behave as altruistic people. Therefore, unless we consider that any form of altruism implies treating ourselves or our descendants as mere means, vaccinating children instead of the elderly does not imply breaking the categorical imperative.

III. THE SMALL RISK ISSUE

Giubilini, Savulescu, and Wilkinson wrote in their paper: ‘If the COVID-19 vaccine is similar to flu vaccine, then by “using” children as a means to protect the elderly and the vulnerable, we are using children as means in a way that also actually benefits the children, as they acquire protection against COVID-19. Even if COVID-19 seems to be not life-threatening for young children, it still is a disease that can cause some complications and some level of pain or discomfort. And however small, there is some risk of severe complications and death even in children’.

The authors’ argument, therefore, is that subjecting children to a small risk, even if it is not in their best interests, would not be unethical if it benefits society significantly. I totally agree with them on this point. Indeed, I find similar initiatives, such as the idea of changing the strategy to fight seasonal flu by involving children in vaccination programs, certainly attractive. However, there is an essential issue that must be kept in mind. When we speak of small risk, we must distinguish between the risk for a group and for specific individuals. In the case of some vaccines, such as influenza, the situation is ideal, because the risk for the children is extremely low. Even better, vaccination brings a certain benefit for them. In the case of COVID-19, the situation remains unclear. It may be that the vaccine has a high risk of producing relatively mild side effects (fever, diarrhea, vomiting). This means that complications will often happen, causing some suffering to the subject, but only mildly. In my opinion, vaccination might be ethically acceptable under such conditions: we can expect children to be altruistic enough to endure a little inconvenience to help the elderly. However, it may also be the case that the vaccine has a low risk (say 1/50,000) of causing profoundly serious side effects, including death. This is much more worrying. If we know that this probability exists, we can assume that in a country where there are 10 million inhabitants, of which 2 million are children, and we vaccinate all of them, 40 children will die. Since COVID does not usually cause such consequences in healthy children, this means that we would be assuming the death of those children in exchange for saving the lives of thousands of people, certainly.

In my opinion, this would be ethically unacceptable. Indeed, this issue reminds me a lot to John Harris’ survival lottery. For those who are not familiar with his provocative thought experiment, this mental experiment suggests that since organ donation saves lives, and donors are always scarce, we should draw a lottery whenever donations are needed. Those selected would give up their lives to allow several people to live. Both in Harris’ proposal and the fictional case of coercive vaccination of children against COVID described above, some people are sacrificed for the sake of society. In both, the ex-ante cost is also small (given the very low probability of being selected to donate organs or to die because of the vaccine), but we have to face a large post facto cost and this seems to be unacceptable. Therefore, I think that we should clearly keep in mind that when we talk about ‘small risk’ we are talking about mild consequences to children, even though this includes a large percent of them, but not in any way severe side effects, even though in a minimal number of them.

IV. THE REASON OF MY REFUTATION: THE EXISTENCE OF A BETTER ALTERNATIVE

There is, finally, a third consideration that I would like to introduce. The proposal by the authors is based on a postulate: that causing children a small harm could be justified if this were the best alternative to protect society. But is this the case? I do not think so. Even in the best-case scenario (the vaccine is effective in 95% of children and prevents them from spreading the disease), some children (in this case 5%) could yet transmit COVID. This means that they could cause serious harm to elderly people who had been deprived of the scarce resource—namely, the vaccine—in their place. Therefore, harming children in this case would not serve to protect the old at all, especially if we consider that such an allocation would continue to make the elderly vulnerable against the greater threats posed by other people, such as young adults, for instance.

To this must be added two other factors that play against vaccinating children to protect the elderly. First, according to current data, children probably play only a minimal role in the spreading of the infection to others. Therefore, they are not in reality a great threat to the elderly. Second, to my knowledge, the clinical trials related to the current group of COVID vaccines do not involve children. Consequently, we will not have any evidence of the risks involved in the vaccination of children before implementing our policies. This is an extraordinarily strong reason to oppose the use of vaccines in children at this moment: ‘it is imperative that any policy mandating the use of a COVID-19 vaccine follow the fullest assessment of safety’.

Keeping this in mind, I think that the main claim in the article that I am commenting on is not tenable. Furthermore, I consider that there are better strategies to protect both children and the elderly. I will propose one: let us vaccinate the old as commonly recommended. Then, let us carry out proper tests to ensure that children are not infected by the virus. How can we do this? In practice, this is quite simple. Most children go to school. If we undertake testing at least twice every week, they could hardly escape surveillance and monitoring. Now that pool PCR with saliva becomes a real possibility, this scenario seems particularly reachable. If we can implement it, it will be easy to detect and trace positive cases. Therefore, we could separate infectious children from the elderly and thus avoid transmission between them, without posing an unnecessary risk or harm to the children and without depriving the elderly of a useful resource, such as the vaccine.

V. CONCLUSION

Keeping in mind all the arguments showed, my conclusion is that Giubilini, Savulescu, and Wilkinson’s general claim—that using children to fight epidemics if this only causes them little harm—is fair. However, this is not due to the acceptability of making exceptions to the Kantian categorical imperative, but to the simple fact that we can rely on a presumption of consent to altruistic behavior by children. Nevertheless, I do not think that this would be the best strategy in the COVID-19 scenario. Isolating children through the performance of testing in schools and institutes could be a better alternative, since it would prevent them from spreading the disease without depriving the elderly of a scarce resource, namely the vaccine.

FUNDING

European Commission (H2020 SWAFS Programme, PANELFIT Project, research grant number 788039), Eusko Jaurlaritza (Ayudas a Grupos de Investigación IT-1066-16), and La Caixa Foundation (LCF/BQ/ DR20/11790005).