Now that Pfizer and Moderna have begun distributing their COVID-19 vaccines to hospitals across the U.S., a new question has gained national attention—how will the vaccine be allocated?
Together, Pfizer and Moderna expect to produce a total of 40 million doses for the American market by the end of 2020. This means about 20 million Americans will be vaccinated by year end, since each person requires two doses of vaccine, separated by a few weeks. The federal government has begun allocating initial doses to states roughly in proportion to the size of their adult population.
The C.D.C. plans to offer a broad framework for which groups to prioritize in each phase of vaccine distribution, but states have broad leeway to implement their own allocation plans. On December 1st, advisers at the CDC released their recommendation that the first wave of vaccines (phase 1a) should be allocated to (1) healthcare personnel and (2) residents of long-term care facilities. State officials have already begun implementing their distribution plans for the Pfizer vaccine.
I had a conversation with Professor Glenn Cohen, one of the world’s leading experts on the intersection of bioethics (sometimes also called “medical ethics”) and the law, to discuss some of the ethical implications and considerations for vaccine distribution.
Authors note: this interview took place on December 1, 2020 (prior to the CDC releasing their distribution plan and the FDA approving the Pfizer and Moderna vaccines). Responses have been edited for clarity.
There has been a lot of talk about who to prioritize for vaccine distribution, with consensus from the CDC that medical workers and the elderly should be first in line. There are other priority mechanisms though, for instance targeting vaccination to curb the spread or to reduce the racial/socioeconomic impact that COVID has exacerbated. Are we prioritizing the right groups and what factors would you considering when thinking about vaccine distribution ?
First, the problem of allocation is one that we face time and time again in the medical world. We have seen similar issues with organ allocation, ICU beds, and pandemic flu vaccine allocation problems.
Second, it’s important to distinguish what are called ‘temporary’ versus ‘persistent’ rationing systems. In a temporary rationing system, there will eventually be enough of the resource for everyone, but we need to prioritize who gets the initial doses. In a persistent rationing system, there will always be a shortage of the critical resource, and some individuals will never be helped. The COVID vaccination plan falls somewhere in between these two. People are actually at risk, and the vaccine would reduce risk, but they’re not in a so-called “rule of rescue situation,” where you have somebody who will perish unless you act. The hybrid nature of this challenge makes things a little more complicated.
There are a number of traditional principles that the medical community relies on when tackling questions of resource allocation. You can find a good summary of these principles in a 2009 paper by Govind Persad, Zeke Emmanuel, and the late Alan Wertheimer, entitled Principles For Allocation of Scarce Medical Interventions. The principles are:
- First Come First Served
- Sickest First
- Youngest First
- Number of lives saved
- Number of life-years saved
- Instrumental Value
Usually, when people design ideal systems, they argue for some mix of the latter six (though you will find some defenders of lotteries, and they are useful once you have exhausted all other principles and still face too many claimants). For COVID, in the major policy documents that have been released thus far, there has been consensus in giving the vaccine to frontline health care workers who face elevated COVID exposure first, which is typically justified on some combination of instrumental value and reciprocity. Doing this is not only good for these health care workers, it’s good for everybody if it means they can continue to do their jobs in the face of the pandemic and help the rest of us. Most everybody agrees on this.
Things get more complicated past that point. In particular there are trade-offs between focusing on people who are most at risk, and people who are likely to be the greatest spreaders. If you’re really trying to prevent those who are likely to die because there are at higher risk, you’re going to focus on individuals who are much older or face demonstrated risk-enhancing comorbidities. On the other hand, if what you care about is saving the most lives or the most life-years, you might design a strategy to focus on vaccinating people who are the most likely to spread COVID.
The second scenario that you highlighted is the question about whether to prioritize communities of color or poor communities. Here it’s important to understand why people think that might be worthwhile to do, because there are multiple overlapping justifications. One possibility: this is another version of sickest first. This is the group of people who, if they get COVID, are most likely to die or suffer serious adverse health consequences based on current epidemiological data. There is also evidence that community spread may be highest for these populations, which provides a separate reason for prioritization.
A different reason to favor these communities is unrelated to COVID, it is about past injustice more generally. That view would be that these communities are entitled to a certain amount of extra protection for vaccination for COVID as a way of recompensing them for other kinds of mistreatments historically. There are some people who think that’s the reason to focus on these populations. Others disagree.
Another complicating factor, is that communities of color are more skeptical about the medical system in general, so the rate of vaccine refusal there is predicted to be higher. In particular, African American communities have a long history of being subject to research experimentation in terrible unethical ways—like the Tuskegee syphilis experiment. One problem is that if you offer more vaccinations to these communities, and try to prioritize them ahead of others, it might be interpreted in a negative way by the community. You can see why this gets complicated.
These are just some of the considerations that are floating around that make this issue of vaccine distribution so complicated.
Expanding on that, given there is a lot that we still don’t know about these vaccines, and that they’re held only to the FDA’s emergency standard, is there an ethical problem with prioritizing the more vulnerable populations?
Yes, these are emergency use authorizations. Yes, the data has a certain number of months of follow-up and not the optimal non-emergency use amount of data. However, we actually have a lot more data than many in the scientific community thought would be available. Even so, it is generally accepted that most vaccines in the past have caused some injury to some subset of the people who are vaccinated. That will likely hold true with the COVID vaccine.
The groups I’m most worried about are pregnant women and children. Clinical trial recruitment companies tend to avoid those two groups because there’s a higher risk of an adverse event. We also know that those groups’ metabolization and the way their bodies work is quite different. So it would not shock me to learn that a vaccine that’s terrific for the general population has a slightly different profile as applied to these two groups.
This is a real problem. If we approve this vaccine, but not for pregnant women and not for children, will there be enough of an incentive to follow up studies and look specifically at these populations or not?
What do you think will happen if people refuse vaccination? Will states turn to vaccine mandates?
It will be a problem if people don’t get vaccinated at all. The clinical trials are not a great indicator for this, because the clinical trial populations are small and well-managed. We are talking about vaccinating basically the entire United States population, so it’s quite hard to know at this point how many people will refuse vaccination.
There is, however, a political reality here. People are already balking at the idea of mask mandates, so it’s likely that state governments would have a much harder time, politically speaking, with actual vaccination mandates. A vaccination mandate would invade bodily integrity in a much more serious way than a mask mandate, and this is something people are more concerned about.
Rather than government mandates, you might see schools or employers requiring vaccination records—Massachusetts and other states already do this. There is an ongoing debate about the constitutional challenges that may arise if we begin to see those mandates. For regular vaccines, some states permit exemptions for moral and religious reasons, but they do it as a matter of legislative grace, not because they think religious exemptions are constitutionally required. So this will be an interesting issue to watch.
Specifically on incentivizing individuals to get the second shot after they have received the first—could the government have people sign a contract promising to get the second shot?
I’m skeptical. How are you going to enforce that? If they did this, would it be constitutional? Complex questions arise. In reality, the government would never try to do this. Especially in phase 1a, where we want as many people vaccinated as possible, we don’t want any barriers or any hesitation.
Last question: as we begin ramping up vaccination efforts, will we run into issues of immunity passports? Given some people will safely be able to participate in workforce, and return to social life, how do we manage them when we are still at the height of the pandemic?
To the extent existing public health measures do not make exceptions for people who can prove vaccination, those will likely be challenged. That may raise an interesting set of questions around the standard of review and how those decisions will be considered. On the one hand, it would be burdensome for a police officer to break up a gathering and analyze everyone’s vaccination cards, separating those who are and those who are not vaccinated. On the other hand, think about a church which wants to host a religious gathering with only vaccinated individuals, and they are willing to prove ahead of time that everyone is vaccinated. If this church asks for an exemption, and the state refuses, that state is going to be in a tough position legally.
You may also have interesting questions arising in the disability discrimination context if, for instance, you have a private employer who tries to hire only vaccinated individuals. The system of vaccination will likely favor the elderly and poorer communities, so the people who would be challenging lack of access will be young, wealthy individuals. One thing to watch is whether this effects the doctrinal analysis. These issues will, of course, be most pressing in the period where not everyone is vaccinated.
 Deputy Dean and James A. Attwood and Leslie Williams Professor of Law, Harvard Law School.
Faculty Director, Petrie-Flom Center for Health Law Policy, Biotechnology & Bioethics.
 Professor Cohen sits on the UNOS/OPTN ethics committee that helps advise on, among other things, how organs are allocated.