Prioritarianism pertains to the generic idea that it matters more to benefit people, the worse off they are, and while prioritarianism is not uncontroversial, it is considered a generally plausible and widely shared distributive principle often applied to healthcare prioritisation. In this paper, I identify social justice prioritarianism, severity prioritarianism and age-weighted prioritarianism as three different interpretations of the general prioritarian idea and discuss them in light of the effect of pandemic consequences on healthcare priority setting. On this analysis, the paper arrives at the following three conclusions: (1) that we have strong prioritarian reasons for special concern about the vulnerable and socially disadvantaged in reference to pandemic effects, (2) that severity of illness is an important factor in identifying the worse off in priority setting but that this must not over-ride the special priority to the socially disadvantaged and (3) that the maximisation rationale of the age-weighted view runs against the core prioritarian idea, and the age-weighted prioritarianism is thus unfitting as a prioritarian response to the COVID-19 case.
- allocation of health care resources
- distributive justice
- political philosophy
This article is made freely available for use in accordance with BMJ’s website terms and conditions for the duration of the covid-19 pandemic or until otherwise determined by BMJ. You may use, download and print the article for any lawful, non-commercial purpose (including text and data mining) provided that all copyright notices and trade marks are retained.https://bmj.com/coronavirus/usage
Statistics from Altmetric.com
The application of prioritarianism has become common in the healthcare priority setting literature.1–5 Generically, prioritarianism is coined around the following moral principle:
It matters more to benefit people, the worse off they are.6
For prioritarianism to be distinct from egalitarianism and utilitarianism, it is important to emphasise two qualifications. First, benefits given to worse-off people have more intrinsic moral weight than benefits to better-off people. This makes prioritarian weighting distinct from utilitarianism and egalitarianism, which gives merely indirect priority to the worst off for reasons related to decreasing marginal utility7 or the reduction of inequality. Thus, even when benefiting the worst off is not utility efficient or inequality reducing, we have prioritarian reasons to favour it. Second, prioritarians’ concern for the worse off refers to their absolute, not comparative, level of well-being. That is, prioritarians could prefer benefiting the worse off even if the comparative difference between them and the better off increases as a result, which makes prioritarianism distinct from egalitarianism.
One important distinction within prioritarianism is between lifetime and time-slice views. Where time-slice views are concerned with the worst off at some specific time, lifetime prioritarianism is concerned with distributions over complete lives. In contrast to time-slice views, lifetime prioritarianism could accept priority to some person A instead of B, despite B being worse off than A at this point in time, with reference to the fact that A has been worse off than B in previous parts of life.8–10 While lifetime prioritarianism faces its own set of problems,5 I take time-slice prioritarianism to be the weaker view, since it allows giving priority to people with only temporary suffering, even if they are extremely well off in other parts of life, which most prioritarians find implausible. For this reason, my arguments shall mainly concern lifetime prioritarianism, although much of what I will argue apply for time-slice views as well.
The prioritarian ideal is thought to fit well the context of healthcare priority setting, but it has taken a number of more specific applications. First, what I shall call social justice prioritarianism identifies the worst off on a wide social dimension, which implies a special moral duty to protect the well-being of the socially disadvantaged.11 Second, severity prioritarianism focuses on giving priority according to severity of illness and implies that we have special moral duties to help the most severely ill.12 13 Third and finally, age-weighted prioritarianism holds young people to be relevantly worse off than older people, because they have had less life years, and that it matters more, for this reason, to help people the younger they are.14–17
All three applications of prioritarianism have relevant implications in the case of pandemic healthcare priority setting. First, in light of the COVID-19 outbreak, it became indubitably apparent that health risks concentrate unevenly on the socially vulnerably and disadvantaged. Second, the pandemic also caused a state of emergency resulting in almost triage-like rationing decision making that invoked the concern of priority for the most severely ill. Finally, the COVID-19 outbreak raised questions about the role of age in the rationing of intensive care resources, questions to which age-weighted prioritarianism provides answers.
This paper asks what we learn from the pandemic experience about the interpretation of prioritarian healthcare priority setting. The paper concludes: (1) that the social effects of the pandemic raise special prioritarian concerns for the socially disadvantaged; (2) that severity of illness is an important factor in deciding who is worse off but that this should not be taken to over-ride the importance of social disadvantage; and finally (3) that age-weighted prioritarianism, when applied to the COVID-19 case, runs counter to the strong fundamental social commitment of prioritarianism in a way that has so far been neglected in the way the view has been put forward, and that, for this reason, we should not consider the age-weighted view a proper prioritarian response to the pandemic.
Healthcare prioritarianism and the COVID-19 pandemic
Social justice prioritarianism
Social justice prioritarianism implies that goods and opportunities in society should be distributed according to the priority principle, with intrinsic moral weight put on benefiting the worst off. While the term ‘prioritarianism’ is today often referred to as a form of consequentialism,8 10 social justice prioritarianism can also be defended from a contractualist justification.18 19 We should prefer a principle of justice that allows special concern for the worse off, where being worse off refers to absolute, arbitrary disadvantages, because rational and reasonable persons would find those disadvantages unfair. Had we chosen the principles for priority setting unaware of our own position, we ought to prefer, contractualist prioritarians argue, a principle that gave special priority to the worse off. The principle could thus be based on consequentialist or contractualist reasoning, but for my purpose, it suffices to acknowledge that it can walk on two legs.
How does this respond to the pandemic experience? One side of the prioritarian response to the COVID-19 outbreak is to identify how socially unbalanced the consequences of the outbreak are. This involves several dynamic effects, some of which unknown, but there are some tendencies we can elaborate,20–22 and although many of them are similarly problematic on egalitarian and utilitarian grounds, they raise special commitments for prioritarians due to the intrinsic and absolute characteristic of our moral duties towards the worst off.
One relevant phenomenon is the dependency of health deficiency and vulnerability on social determinants of health. The socially disadvantaged are more vulnerable to health threats, they have significantly shorter life expectancy and lead less healthy lives than the socially privileged, and, as is well-known today, this phenomenon not merely depicts a difference between the rich and the poor but a social gradient of health running continuously through society.23–25 Moreover, while there are certainly important local differences, the social gradient of health appears in all countries.
Through social determinants of health, we can expect social standing to significantly influence vulnerability to the pandemic threat.21 22 Aside from age, suffering from underlying medical conditions is the most important explanatory factor for COVID-19 caused severe illness and hospitalisation. Conditions such as heart diseases, obesity and type 2 diabetes are reported to significantly increase risk of hospitalisation in case of infection. These conditions are much more prevalent among the socially disadvantaged. Additionally, comorbidity further increases the risk of severe illness and is much more frequent among the socially disadvantaged. Thus, vulnerability is extensive, and the health-related consequences of pandemics are much worse for the people at the bottom of the social hierarchy.
Another evident result of the outbreak is the socially unequal distribution of health risks. Not only are the socially disadvantaged more likely to become severely ill, they are moreover more exposed to virus and less able to protect against it than the socially privileged. There are several explanations accounting for this. First, low-income families generally have more persons per household than middle-income and high-income families, which makes infection entry into the family more likely, and low-income households are also often placed in areas with higher density, which may increase the risk of contagion. Second, many low-income jobs involve work in the frontline (eg, bus drivers and cleaning personnel) and therefore with a relatively high risk of contagion. Health professionals are an exception here. Nurses and medical doctors have been much exposed to the virus without opportunity to stay home to avoid viral exposure. However, unlike many socially disadvantaged, health professionals have been carefully instructed to take all necessary precautions, and they are well equipped to comply to health authorities’ directives.
Finally, many countries have experienced an economic down-flow as a result of the COVID-19 outbreak. When countries are required to fully or partially ‘lock down’ to break or slow virus spread, it has significant economic consequences. It negatively affects the export, it limits the domestic production and consumption, and the heavy restrictions on travel and traffic draws tourism to a minimum. All this affects the economy negatively, and unemployment immediately rises. This falls especially on the worse off. Low-income staff generally hold less specialised jobs, which are much easier to cut off than more specialised positions. Thus, unemployment costs also fall primarily on the worse off. In addition, the hardship of unemployment also tends to follow a social gradient. Socially disadvantaged people will generally face graver economic hardship by the loss of income than better-off people, and their opportunities for getting other jobs are more limited.
Together, all these points paint a pretty one-sided picture. We have strong prioritarian reasons for concern about pandemic distributional effects because they tend to lay additional burdens primarily on the socially disadvantaged. As mentioned, these effects would also be problematic on egalitarian and utilitarian grounds, but prioritarianism must respond to them with special intensity. It is evident, therefore, that prioritarians of the sort responding to concerns of social justice have strong reasons to insist that the healthcare system should rectify these unfair harms and that we have special moral duties to protect the worst off in society against pandemic effects.
Another common interpretation of prioritarianism to healthcare priority setting implies special weight to treating the most severely ill. While this view has not stayed clear of criticism,26–28 most find that it should play some role in prioritarian deliberation. One interpretation of severity prioritarianism is to define ‘worse off’ narrowly in terms of health deficiency so that the relevantly worst off are simply those that are most severely ill. This is easy to interpret as a form of time-slice prioritarianism—what matters most is to treat the most severely ill at this point. In principle, of course, we could also imagine a lifetime prioritarianism narrowly focused on severity of illness, in which case the worst off are those who experience most severe illness over the course of a complete life. However, this is an odd view that few prioritarians would accept. The moral importance of treating one person’s severe illness now seems untouched by the fact that others have suffered severe illness in the past (in so far as they do not now).
There are other plausible ways, however, to embrace severity of illness from the perspective of lifetime prioritarianism. First, the fact that a person is worse off than others now certainly has some impact on this person’s standing compared with others over a complete lifespan. This could of course be over-ridden by other factors on a lifetime view, but severe illness is not a trivial moral factor; it makes life significantly worse. Second, severe illness in many cases—and specifically in the COVID-19 case—involves a significant risk of death. Thus, on a lifetime prioritarian view, being severely ill with COVID-19 or other life-threatening diseases is likely to make people worse off than most other people even on a lifetime scale. This last point inevitably raises considerations regarding the importance of age, because from this perspective severity of illness will be worse for people, the younger they are (I discuss this in the next section).
It seems clear, then, that severity of illness is a central component in accounting for disadvantage on any prioritarian view, but this raises the question of how it should be weighed against our prioritarian commitment to benefiting the socially disadvantaged. What would be the right prioritarian scheme for setting priorities between severely ill patients of different social standing? There seems to be two relevant arguments for giving priority to the socially disadvantaged available. First, consequentialist prioritarianism would find it of more moral weight to offer treatment to the disadvantaged than to the privileged because of the higher intrinsic moral worth of benefiting the worse off. Hence, when setting priorities between people of different social standing with equally severe conditions—such as patients with COVID-19 equally in need of intensive care—we should give priority to helping the underprivileged. Moreover, while severity of illness is a very weighty factor, and thus should be central to our priority setting, lifetime prioritarianism could find reason to give priority to socially worse-off people, even when these are somewhat less severely ill than better-off people, because whereas pandemic illness is immediate, their social disadvantage has been ongoing.
Second, a contractualist account would justify higher priority to the socially disadvantaged over the privileged in allocating healthcare resources between patients with equally severe illness when it would be prudent for reasonable people to prefer access to treatment for pandemic infection as a member of the socially disadvantaged group over access to the same treatment as a member of the privileged group (when in making this choice, they do not know their social standing), and, as my analysis of the pandemic effects illustrated, this would indeed be prudent. Because the health risks of the pandemic are much more prevalent and severe among the socially worse off, prudence would incline reasonable and risk-averse people to demand stronger insurance against need of pandemic-related treatment for people, the more socially disadvantaged they are.
One could object here that contractualist prioritarianism does not commit to this implication, since giving special priority to the socially disadvantaged would be unfair to the better off, because it includes information in the moral equation that is arbitrary and irrelevant for the decision in question, that is, which should ideally be filtered out by a veil of ignorance. It is a way, critics could say, of giving more weight to some people due to the arbitrary fact that they happen to be members of one particular social group (ie, the socially disadvantaged), which was exactly the kind of arbitrariness that contractualism was meant to avoid. However, it seems evident that the distribution of pandemic effects is indeed not arbitrary in the relevant sense but in fact follows a very systematic social pattern and thus would be something that we should allow people to insure against on a prudential account. If, for example, the risk of needing intensive care as a result of COVID-19 is twice as high for socially disadvantaged people as it is for the socially privileged, it would indeed be prudent to prefer insurance against this higher risk. Special priority to the socially disadvantaged in distributing healthcare resources for pandemic treatment could be justified on this contractualist reasoning. Thus, even allowing prioritarianism to stand on two legs—consequentialism and contractualism—the reaction to pandemic-related severity is one sided. Severity of illness is indeed an important component in determining absolute disadvantage, which should be taken seriously, but it should be weighed against, and not over-ride, our prioritarian commitment to the worst off on a wider social dimension.
The third and final application of prioritarianism is the age-weighted view according to which it matters more to treat people the younger they are. Importantly, some sort of prioritarian age weighting is already implied by lifetime prioritarianism because, other things equal, untreated conditions early in life will make a life worse over a lifespan than the same condition later in life. However, age-weighted prioritarianism also gives intrinsic moral weight to treating the younger, because younger people suffer the unfairness of not yet having had the opportunities in life that older people have already enjoyed.14–16 Thus, even if total lifespan outcome will be the same, we should still favour treating the younger over the older on this account. This is very intuitive, and at this point, I think it fits prioritarianism nicely. However, as it has been laid out in the literature, the age-weighted view comes to focus so narrowly on the maximisation of age-weighted life years that it neglects the fundamental moral commitment of prioritarianism towards the socially disadvantaged. For this reason, I argue, it is an unfitting application of prioritarianism to COVID-19 healthcare priority setting.
To see this, suppose we must prioritise intensive care resources between the following three patients with COVID-19:
A 20-year-old individual with life expectancy to 80 (60 quality-adjusted life years (QALYs)).
A 20-year-old individual with life expectancy to 60 (40 QALYs).
A 40-year-old individual with life expectancy to 80 (40 QALYs).
According to age-weighted prioritarianism, a justified scheme for healthcare priority setting must aim to maximise age-weighted life years, where early life years count for more than later life years.15 16 Thus, the age-weighted view implies the following combination of fairness and maximisation reasons. It gives priority to A over B because of the extra gain of 20 additional QALYs. It prefers B over C despite the fact that QALY benefits are equal, because QALYs have more moral worth for people in earlier stages of life, since young people have had less life opportunity than older people. Finally, it would treat A rather than C both because of the larger number of QALYs and their extra age-weighted moral worth.
This is in many ways a plausible account of how age should matter for priority setting. As an application of prioritarianism to the COVID-19 case, however, it faces the problem of running against the strong social commitment of social justice prioritarianism.29–31 To see this, imagine that the difference in life expectancy between patients A and B is explained by social determinants of health. This is not farfetched. As Marmot25 depicts, in London Borough of Westminster, UK, we find a health gap of about 18 years based on social standing, and in Baltimore, USA, the gap is 20 years. This fits perfectly the gap introduced between patients A and B, and it is thus not unreasonable to expect socioeconomic factors to give rise to such a difference. However, here, as mentioned, age-weighted prioritarianism implies priority to the person with higher social status (A) over the socially disadvantaged person (B).
This brings forward a problem for the application of consequentialist prioritarianism.32 Both age and social status on the one hand invoke fairness reasons and thus influence our determination of who is the relevant worse off, and on the other hand, significantly affects the cost-effectiveness ratio, and while this is unproblematic for age since the priorities point in the same direction—priority to the younger is both fair and cost-effective—it presents a problem of conflict for social status. Here, our prioritarian reasons of fairness to give priority to the worse off conflicts our reasons to maximise (age-weighted) health outcome, which underlies the age-weighted view. What should a prioritarian believe?
The general answer to this must be disappointingly undecisive. Certainly, efficiency will be one significant factor on all prioritarian accounts. Hence, it is unreasonable to insist on priority for the worst off if the benefits we can offer them are incomparable to potential benefits given to better-off people. Thus, for example, if our choice is between prolonging the life of a better-off patient with 20 additional life years and prolonging a worse-off person’s life with merely 1 month, any prioritarian could reasonable be willing to favour the more health-efficient outcome. However, this does not alter the fundamental moral commitment of prioritarians to assist the worst off. It is still the case that it is more intrinsically moral important to benefit people the worse off they are, and this fundamental requirement has been neglected in the way age-weighted prioritarianism has been unfolded.15 16
Based on the analysis of the previous sections, it seems clear that the prioritarian reasons for special concern for the vulnerable and socially disadvantaged in the case of a pandemic carries significant weight and that the probability of developing severe illness with need for hospitalisation and intensive care is higher among the worse off. When all this is taken into account, and we assume that the differences in life expectancy between A and B is a result of social determinants, it seems evident that simply giving priority to treating A (the privileged) over B (the disadvantaged) merely because it is more (age-weighted) QALY efficient is problematic from a lifetime prioritarian perspective. The prioritarian duty to benefit the socially disadvantaged must be taken into account given the outbreak circumstances that put them at risk in the first place. This, certainly, does not imply that we should abandon age rationing32; however, it seems to imply that age-weighted prioritarianism, as it has been put forward, is unfitting as a prioritarian response to the COVID-19 case.
Ethicists can learn much from studying the effects of pandemics. In this paper, I have shown that the COVID-19 experience projects three different interpretations of prioritarianism, adhering to social justice, severity of illness and age, respectively. I have argued that prioritarians have strong reasons to be concerned about the pandemic effects on the vulnerable and socially disadvantaged in our society, that severity of illness is an important moral factor but that this should not override the prioritarian duty towards the socially disadvantaged and that age-weighted prioritarianism, as laid out in the literature, is an unfitting prioritarian response to the COVID-19 case. If sound, this argument has both theoretical and practical implications. Theoretically, it implies that social disadvantage should be a primary concern for prioritarianism even when narrowly applied to healthcare priority setting. As a practical implication, the argument suggests that welfare state health systems take special interest in the protection of the socially disadvantaged and that they initiate health policies to mitigate the unfair social harms caused by the pandemic.
I am grateful to Andreas Albertsen, Anna Christine Dorf, Anne-Marie S Christensen, Didde Boisen Andersen, Kasper Lippert-Rasmussen, Sigurd Lauridsen, Thomas Søbirk Petersen and two anonymous reviewers for very useful comments.
Contributors LN is the sole contributor.
Funding The work was conducted as part of a research project funded by the Independent Danish Research Fund, Grant Number 9017-00007B.
Competing interests None declared.
Patient consent for publication Not required.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement There are no data in this work.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.