Cross Post: Pandemic Ethics: Vaccine Distribution Ethics: Monotheism or Polytheism?

Written by Alberto Giubilini, Julian Savulescu, Dominic Wilkinson

(Oxford Uehiro Centre for Practical Ethics)

(Cross-posted with the Journal of Medical Ethics blog)

Pfizer has reported preliminary results that their mRNA COVID vaccine is 90% effective during phase III trials. The hope is to have the first doses available for distribution by the end of the year. Discussion has quickly moved to how the vaccine should be distributed in the first months, given very limited initial availability. This is, in large part, an ethical question and one in which ethical issues and values are either hidden or presented as medical decisions. The language adopted in this discussion often assumes and takes for granted ethical values that would need to be made explicit and interrogated. For example, the UK Government’s JCVI report for priority groups for COVID-19 vaccination reads: “Mathematical modelling indicates that as long as an available vaccine is both safe and effective in older adults, they should be a high priority for vaccination”. This is ethical language disguised as scientific. Whether older adults ‘should’ be high priority depends on what we want to achieve through a vaccination policy. And that involves value choices. Distribution of COVID-19 vaccines will need to maximize the public health benefits of the limited availability, or reduce the burden on the NHS, or save as many lives as possible from COVID-19. These are not necessarily the same thing and a choice among them is an ethical choice.

There may be two phases in distribution.

In the first phase, there is likely to be more people wanting the vaccine than there is vaccine. The question is: how should we prioritise access to the vaccine? For example, should we prioritize the elderly, or other vulnerable groups, or the younger, or essential workers, or some other group?

In the second phase, as production increases, there may be more vaccine than people willing to take it. The question then is: how should vaccine uptake be ethically be increased? For example, should vaccines be made mandatory, or nudged, or incentivised? We won’t consider this phase here but one of us has recently written about it.

The Current Approach 

The current UK approach is to distribute to those most likely to become ill and health care workers. The policy essentially prioritises those at greatest risk of dying. Thus, because we know that the group at higher risk are the elderly in care homes, the JCVI recommends that the first group to receive the vaccine should be aged 80 or over resident in care homes and care home workers. Health Secretary Matt Hancock said the Government will send NHS staff into care homes to carry out the vaccinations.

The German Federal Ministry of Health commissioned report sets the following goals: “1. Prevention of severe courses of COVID-19 (hospitalisation) and deaths. 2. Protection of persons with an especially high work-related risk of exposure to SARS-CoV-2 (occupational indication) 3. Prevention of transmission and protection in environments with a high proportion of vulnerable individuals and in those with a high outbreak potential. 4. Maintenance of essential state functions and public life”. From this, it concludes that “[a]s a result, priority should be given to those individuals at the highest risk of death and serious illness from a disease such as COVID-19”

Thus, the two approaches are very similar. According to both, the most vulnerable should have priority access.

However, this approach may be questioned. Vulnerability is only one of the many factors to consider, and potentially not even the main one, in deciding whom should be prioritized. We need to maximize the public health benefit of the vaccine, but what counts as public health benefit is not only measured in terms of numbers of lives saved from COVID-19.

One Value? Rejecting monotheism of value

The JCVI approach considers one value: probability of dying. Another word for it is ‘vulnerability’. It is a kind of monotheism of value. Many people would simply assume that this is the right approach because the more one reduces the probability of dying, the more lives that will be saved. It is a basic ethical principle that limited resources should be used to save the greatest number, as two of us have recently argued However, even assuming this principle, number of deaths is determined by many factors that require to look beyond mere vulnerability. For example, how likely are different groups to spread the virus? How effective is a vaccine on different groups? How effective is a vaccine at preventing contagion vs preventing individuals from getting sick?

Numbers

Essentially, the older you are, the more likely you are to die from COVID-19. The risk especially increases over 85, which accounts for about 45% of the mortality. The risk is higher among the elderly who reside in care homes, where the risk of infection is higher than in the case of elderly living in isolation or with few contacts.

Other risk groups include being the recipient of an organ transplant, haematological cancers, certain neurological conditions, chronic kidney disease, immunosuppression, dementia, stroke, diabetes, chronic pulmonary disease, obesity, malignancy, liver disease.

Implicit in the idea of prioritising care home residents and workers, as well as health care workers is that this will significantly reduce infections among those at the highest risk. This principle could be extended to anyone who has contact with vulnerable people, such as carers.

If we prioritize this criterion, then we are likely to maximize the number of lives saved

However, there are plausibly other values, apart from saving lives, which should be included, or at least considered.

Years of Life Saved

It is plausible that what matters is not only whether a person’s life is saved, but how long it is saved for. It is morally different to save a person for 50 years than it is to save them for one year.

In a related but different way, those who have most life years to gain are usually those who have so far lived the least. Thus, age itself may be an independent value on the grounds of desert – the young deserve to have their life saved because they have had less life.

These values (probability of survival and length of survival) can be combined into the notion of “expected life years gain”. Expected life years gain is calculated by multiplying probability of survival x duration of survival.

Here is an example of a quick calculation of how this might work. The mortality rate from COVID-19 for 60-64 year olds has been estimated (in absence of other risk factors) as somewhere between 1% and 3.6%. Assuming 100% vaccine efficacy, and given an average life expectancy for a 62 year old male of 85, a vaccine would give this person an extra 23 years. The expected utility, where utility is defined in terms of life years gain, is 0.69 per case of COVID-19 avoided.

On the basis of the same calculation, the expected utility for people aged 80-84 years old (age specific mortality rate around 15% according to the most pessimist estimates), is 1.2. For 40-49 years old, it would be 0.156.

This simple calculation suggests that the greatest benefit might be achieved by targeting the vaccine to the elderly as recommended by the JCVI. Formal cost effectiveness modelling has been attempted. This is essential in a situation of limited availability of resources. Current estimates suggest that vaccination is most cost effective in terms of QALYs gained if targeted at groups with higher risk of hospitalization and death. Indeed, vaccinating patients at highest risk from COVID-19 would potentially save more money that it cost. The model suggests that the vaccine would still be cost effective if targeted at adults aged 50-64, although each QALY would in that case be more costly – which in conditions of limited resources means that less QALYs can be gained, other things being equal.

However, other factors need to be considered.

Vaccine Effectiveness

A vaccine may have different levels of effectiveness in different groups. Flu vaccination, for instance, is much less effective in the elderly as their immune systems fail because of immunosenescence. Elsewhere we have argued that this factor should be taken into account in flu vaccination policies in order to maximize the public health benefits of the flu vaccine: targeting the young might better protect the vulnerable via indirect protection. We have also argued more recently that the same approach should at least be considered for the future COVID-19 vaccine if such vaccine will turn out to be similar to the flu vaccine in this respect (which might or might not be the case for any of the future COVID-19 vaccines). For example, if the vaccine were only 50% effective in over 80 year olds, it could make sense to prioritise the 60 year olds if the vaccine is fully effective in them. In this case, vulnerability is not the only or even the primary criterion for prioritization in access to the vaccine.

Recent UK vaccine prioritisation modelling suggests that as long as a vaccine were more than 20% effective (i.e. only working 1 in 5) it would still save most lives and life years if older age groups were targeted.

But are there other values that we also need to consider?

Quality of Life

It is standard in health economics and the allocation of limited health resources to take into account not only probability of beneficial effect and length of effect, but also the expected quality of the expected life.

This is combined in the concept of a Quality Adjusted Life Year (QALY). A QALY is given by multiplying a year of life x its quality (on a scale of 0-1). In the UK, treatments will only be funded (with a few exceptions) if they cost less than £30 000 per QALY.

Thus, for instance, if we assumed QALYs as the only criterion for value, unconsciousness is plausibly the same value as death: by definition, there is no quality of life in being unconscious. Thus a person who is unconscious (for example, from end stage dementia) would have a quality of life of 0. If we consider QALYs, these people would not be a priority candidate for vaccination because even giving them additional years of live would be of very little comparative value. The extent to which they are vulnerable to COVID-19, the effectiveness of the vaccine on them, and the years gained would not affect prioritization.

Other forms of severe cognitive disability would affect the number of QALYs a person could gain by surviving COVID-19, and therefore the level of priority such a person would be given in accessing the vaccine. This would clearly be relevant to a policy of prioritising patients in nursing homes. Approximately 70% of residents of nursing homes in the UK have dementia. It is unclear whether this was taken into account in modelling or decision-making by the JCVI.

Discriminating on the basis of disability would be controversial, but it is important to point out that the existing strategy for vaccine distribution already discriminates. It chooses to give some people priority for vaccination (those who are older, or who have certain health conditions, or are health workers). It is not egalitarian – putting everyone in a lottery for access to the vaccine. The question is not whether we should discriminate – but how, and whether the basis for choosing is ethical.

The criterion for deprioritization in this case would not be disability per se, but the way some disabilities would affect quality of future life. If we defined quality of life in such a way that certain disabilities would not significantly impaired quality of life, considering QALYs as a relevant prioritization criterion would not affect the priority level of people with disabilities.

Dependents

We need to consider not only how the vaccine would protect the most vulnerable and would slow down the spread of the virus, but also broader societal and individual benefits that it could bring. Those who have dependents might be given extra priority, and those who have dependents who are in COVID-19 vulnerable groups might be given even higher priority. The health of dependents should be factored in into the health costs of COVID-19, as a sick carer might (at least temporarily) undermine the health of their dependents, too.

Societal Value

We might want to take into account the contribution to society that different people can make. Those whose activities are more important for society might permissibly be given higher priority if their getting sick would translate into significant societal harm.

Although the idea might appear controversial, the public does give moral weight to “social worth”

Adopting this criterion, we might for instance give priority to “key workers” (e.g. those working in supermarkets, garbage collectors, those in delivery services, etc) who must go to work to keep the economy functioning and expose themselves to greater risks.

Conclusion

Our point is not to settle which values should influence prioritisation. It is merely to show that we need not be monotheists about prioritization. There are many ethical values that could be included. Here we have mentioned some, but there might be others. Ultimately, all these values might determine what counts as ‘public health’ and therefore what it means to maximise the public health benefit of the vaccine. Questions such as whose health, what degree of health (or quality of life), what length of healthy life, as well as others are all relevant to a definition of public health and presuppose value judgments about what, ethically, matters the most.

How to prioritize access to the COVID-19 vaccine is not an exclusively political or medical decision, but an ethical decision. It must be informed by good science, but science cannot tell us who should get the vaccine first. We have to decide.

  • Facebook
  • Twitter
  • Reddit

One Response to Cross Post: Pandemic Ethics: Vaccine Distribution Ethics: Monotheism or Polytheism?

  • Ian says:

    Reflecting upon the plausibility of the statement ‘It is morally different to save a person for 50 years than it is to save them for one year.’ revealed other arguments considerably different, which then pointed to many similar ethical tensions within the article.
    To explain: There is no moral difference between saving the life of ‘a other’ or ‘b other’ be it for one hour or fifty years, (the ordinary emotional content of the end of a life event for all those involved may be perceived to validate that statement – although it is recognized those emotions may be ameliorated by other generated emotions should that route be chosen. i.e. it was a relief for them, they could not have survived for long, it would be cruel for them to have continued, and so forth. It is acknowledged that the existence of those thoughts before or after the event may mask other emotions in differing ways)
    Continuing with that explanation across many of the points presented: Arguments do exist (which hold validity in certain worldviews) that ethical criteria constructed with a focus upon measurable material objects, outweigh other moral considerations. And that may be perceived to be the major perspective the blog article itself appears to be predicated upon. When that point is carefully considered it more clearly illustrates frequent differences between a produced science of ethics and human morality, which the quoted statement deliberately conflates, thereby facilitating an ignorance of what is actually occurring.
    From those worldviews preferring measured ethical paradigms this particular response may be perceived as perversely disabling all measured ethical inquiry, rather than differentiating between outlooks, one looking to justify both any life and any death, and a different positively motivating morality for life which accepts humanities diversity (in this sense in the potential of lifespan and approach to death) and deaths eventual finality. (Many faiths take similar stances, but this one is not faith based.)

    To start the critique: Partially because of those issues the presented ethics discussed in the blog article become self limiting in application within national boundaries and one immediately available vaccine (Which in fairness is pointed out as the focus at the beginning of the blog article, along with other rationales for those reading more deeply). Although understandable in the context of nationalistic perspectives and the constraints and abilities for immediately realistic action the global ethical contexts of this pandemic become lost. And even within the context of the nation state there is a latent limitation on discussion due to a largely material focus which constrains justifications. This may be illustrated by considering issues arising from similar sentences:
    Following the science but managing the numbers;
    Managing the science but following the numbers;
    Reading both statements in various ways reveals the existence of circumstances and worldviews in which each could be ethically justified, but when applied to life and death situations and morally (or scientifically) considered within a living diverse society all interpretations feel wrong.
    Looking to the considerations presented at the conclusion of the blog article and removing any national aspect then applying a focus on the vaccination for a global pandemic and including a perspective where densely populated areas could be shown to have greater risk and vulnerability, then deploying the vaccine at relevant times in those areas of highest population density in the world would clearly maximize all the areas mentioned in the blog, and yet that argument fails to gain any traction because of the articles approach, and itself would be ethically/morally wrong. Whilst allowing individuals to make their own vaccination choices, relying upon their inherent values would create a certain level of chaos because of the diversity of values, and not provide the wider common protections being sought within the article.

    In conclusion: The ‘chosen’* focus limits the presentation of options, even in the brief blog article (where attempts are ostensibly made to promote the inclusion of other views) by limiting considerations within structures of thought which support particularly desired outcomes inside the societal constructs in which they reside. Motivationally this could be expected in the interests of social stability and a drive to achieve a specified objective, but should morality (or ethics) discussions be wholly driven and defined by quite rigid structures? Or does morality contain a softer aspirational aspect arising out of a common humanity which cannot so easily be reflected by photographing a located society to illustrate immediately espoused situationally inspired remarks which ignore more diverse cultural implications whilst attempting to attain a favoured resolution which inherently excludes other groups?

    Moving into the future, comprehending the answer to that last question does seem more important in the world of closely connected humanity that exists today….

    *The word ‘chosen’ is intended to be read in a loose way as including unrecognized cultural bias as well as conscious choice.

    N.B. This response was being delayed further because of its length and the intrusions that poses within the blog, and for those my apologies are offered. But because of feedback beginning to be privately proffered in other media and in different ways from other perspectives it was considered of material relevance to post it now.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use the <em>, <strong> and <blockquote> tags. Links have been disabled to combat spam.

Notify me of followup comments via e-mail. You can also subscribe without commenting.

Authors

Affiliations