Virtual Roundtable: Who Gets the Vaccine First?

Thursday, September 17, 2020
Tatyana Makeyeva/Reuters
Ezekiel Emanuel

Vice Provost for Global Initiatives, University of Pennsylvania


Senior Fellow for Global Health, Economics, and Development, Council on Foreign Relations

Global Health, Economics, and Development Roundtable Series

As vaccine nationalism rises, the question looms: who gets the vaccine first? Dr. Ezekiel Emanuel discusses future vaccine manufacturing, distribution, and roadblocks. 

Thomas Bollyky: Great. Welcome. My name is Tom Bollyky. I’m the director of the Global Health program at the Council on Foreign Relations, and it's my pleasure to welcome you to this roundtable on vaccine allocation entitled, “Who Gets the Vaccine First?”

The one part of the global response to this pandemic that has worked well has been the science. Within just four days of the publication of the genetic sequence by China, potential vaccine candidates had already been identified and development and had begun. Sixty-six days later, the first clinical trial started. Today, we find that there are 188—180 rather—vaccine candidates overall in development, thirty-five of those certain clinical development, eight of those are in the latest stage of clinical development. Russia and China have already granted their leading candidates authorization for emergency use. Just this week, the United Arab Emirates granted emergency use authorization of China's vaccine. So already, we're starting to see the first vaccines deployed.

As you know, in this country, in the United States, there has been some question about whether or not a vaccine will be granted or authorized for emergency use by the election. Who knows whether that will occur, but the fact that it's even possible is astonishing. The previous record for a novel vaccine for development was four years. Of course, an emergency use authorization or an approval would just be the beginning, leaving the more complicated and even more difficult questions of how it will be allocated and distributed, both internationally and within countries. And today we're going to focus on the first of those questions—the allocation part—and we have a wonderful speaker, Ezekiel Emanuel, to help us guide through this question.

All of you have Zeke’s very impressive bio, so I will just grossly truncate it in the interest of time. He is the vice provost of global—for global initiatives at the University of Pennsylvania. He's also the Diane v.S. Levy and Robert M. Levy university professor and the codirector of their Institute on Healthcare Transformation. As many of you know, Zeke served in the Obama administration as a special advisor on health policy to the director of the Office of Management and Budget, and most importantly, he was a member of the CFR-sponsored Independent Task Force on The Emerging Global Health Crisis: Noncommunicable Diseases in Low- and Middle-Income Countries that I had the honor of chairing, and he was a terrific contributor, so it's a great pleasure to welcome him back to the Council.

We will start with ten minutes of opening remarks from Zeke, both outlining the current situation on allocation, as well as a terrific new publication that he has authored in Science on putting forward a fair priority allocation model.

After he gives us opening remarks, I will ask a few questions to get us started, and then turn it on over to all of you. Please keep in mind that this meeting will be for attribution. Anything you say can and will be used against you, and possibly read back to you at your confirmation hearing. And with that, let me turn it over to Zeke to get us started with some initial remarks.


Ezekiel Emanuel: Tom thank you very much, and thank you for making this possible, and you didn't mention my membership in CFR, which I hold very dearly.

So, let's begin with the problem. We have, as Tom mentioned, a rush to get a vaccine, and we are literally doing it in record time. When a vaccine will prove itself in phase three trials to have sufficient effectiveness and sufficient safety is still unknown. In both the Russian case and the Chinese case, they haven't actually done a phase three trial. They have granted these emergency use authorizations after phase two trials. I know that the Russians are in the midst of trying to conduct a phase three trial to world standards, and we'll see how that goes. But we shouldn't—they're not proof that we have a safe and effective vaccine.

Once we're through the first phase of these trials, we need to be very, very clear what's going to happen. Right now, the trials are 30—40,000 people. 20,000, roughly 15,000 in the U.S. getting the COVID vaccine; 15,000 or 20,000 getting a “placebo.” And we're looking at something like 150 events—people converting from COVID negative to COVID positive to be the assessment, to lead to an assessment of whether the vaccine is safe and effective.

There are many things that need to be looked at for this. First of all, we need to understand whether it actually is effective and at what percentage is it effective. Is it 60 percent effective in preventing infection? 75 percent? 90 percent? Very, very different. Most flu vaccines 50 to 60 percent—annual flu vaccines 50 to 60 percent effective.

We won't have sufficient, long term safety data. We will have immediate safety data, but very comparatively short-term safety data. Normally on vaccines, you need two, three years. The great vaccine developer Maurice Hillman, who worked at Merck and developed many of the vaccines we use and probably has saved more lives in human history than anyone else, used to say that he didn't sleep comfortably until three million people got his vaccine. So, it gives you a sense of what, you know, a vaccine developer who isn't being politically pressured thinks you need to do to establish real safety. We won't have that. We're going to have literally a few months on a few thousand people of safety data.

In addition, one thing we are—or, two other things we won't know when we look at these data, even if things go well and this is proven effective, is we will not know whether it's effective at preventing transmission. We’ll mostly know is it effective at preventing infection, but not transmission. And there are plenty of vaccines that prevent infection but aren't necessarily good at preventing transmission. There are also vaccines that may not be great at preventing infection but reduce the severity of the illness, so that people who might have gotten a severe illness now just get a mild illness. That is even a lesser standard than not preventing infection. That will shape how you actually use the vaccine and who you immunize because you won't know if you can actually use it to prevent transmission.

The third—or the second thing that we also won’t know as a result of these early trials is the durability of these vaccines. How long will this immunity, if it’s there, last? I was having a conversation with some experts earlier this morning, and they were talking about, you know, got to be a minimum of twelve months. The FDA standard, the World Health Organization standard is six months, but I agree that I think twelve months is much more realistic. That's more like a flu vaccine cycle. And if you think of six months, that would mean that every American, every year would need, at least with the first set of vaccines, four shots. That just seems to me very unlikely from a logistical and also just a personality, a willingness, behavioral economic standpoint that Americans are going to do that—four shots of this vaccine. It’s going to be very hard to get up to 250 or 300 million people vaccinated if we need four shots in a year.

The last point I would make a just about the vaccine is we should not think that the first vaccine approved is going to be the best vaccine. There are many reasons to think it won't be, one of which is they need two shots—you need two shots twenty-one or twenty days apart. The tortoise in this race may win and may be a much better vaccine. I would just know that, for example, Merck is putting its chips on developing an oral vaccine, so no shots needed, and the vaccine that reproduces, so it's durable for much, much longer than what we've been seeing here. So, you know, this may compound problems of actually rolling out a vaccine, making the logistics even more complex.

I assembled a group of public health people, epidemiologists, bioethics experts, not-at-bioethics ethics and political theory people because there is a very unsolved problem about how to distribute a vaccine among countries. We have 7.8 billion people in the world, even under optimal, rosy scenarios we might have a billion, two billion doses by the end of 2021—at most, you know, either 15 or 20 percent of the world's population. How do you distribute a vaccine among countries?

This has been a question that has been around, not just about vaccines but about other treatments, and yet no, I could find no scholarly paper that definitively answered it and mostly pronouncements. Now, the good news is a lot of these pronouncements have coalesced around we need a fair and equitable distribution system internationally among countries, but the problem is, what is fair and equitable mean? And so that we thought we were expert in fairness and equity, and we could answer this better than other people. So, we went about asking that question and trying to answer it. It’s a very complicated question. It involves, you know, attention to economics, international relations, and ethics itself.

We came up first with three fundamental principles, which are accepted very widely and not by any particular theory like utilitarianism or other theories. These three principles are benefiting people in limiting harm. This seems widely viewed. It's not just a utilitarian principle; every defensible ethical theory has to have this. Second, mitigating and certainly not compounding disadvantage. People who are disadvantage should get some priority, and other words that has been widely used in global health. It's a pretty good standard principle in global health. And the third principle is equal concern and respect, sometimes known as equity, sometimes known as fairness. But fundamentally, in this context, it's a nondiscrimination principle. We should not use irrelevant factors like sex, race, religion to distribute. On the other hand, it does say what relevant factors, and there may be relevant factors, that are important, like who's more vulnerable, that can be used to distinguish who gets a vaccine and who doesn't.

If you take those three principles, your next question is what are the biggest harms or goals of using a vaccine? And our group said, you know, death is most important. And why is death—premature death—most important? Well, we identified three reasons. First, it's irreversible. Lots of goals could be long term and even irreversible, but death certainly is. Second, it's devastating. And lots of other problems can also be devastating—being thrown into poverty we know can be devastating for a long time, having a severe comorbidity. But the last thing differentiated mortality, and that is that it's non-compensatable. You cannot compensate a person once they're dead, they're gone. Whereas, you can compensate for unemployment, you can compensate for poverty. I'm not saying it's easy, but I am saying it’s compensatable. So, we said we should focus in on premature deaths.

We also wanted to focus in on premature death because there is a moral difference. All of us live one life, if we live a shorter amount of time, we don't live the full amount of time, and that's a deprivation. That's a disadvantage, and all of us personally and collectively recognize that. You know, if we had to invest our money for our health, we would all invest it as a young age, to ensure that we live to an old age, rather than to invest it at eighty live to eighty-five. And so, premature death is a very important consideration.

The second goal, we argue, was economic—relieving economic deprivations and social deprivations. Things like unemployment, poverty, a loss of education.

And the last one was to achieve herd immunity.

And then what we did is to select principles or metrics that should be used to assess these. And again, part of the assessment is if I give a vaccine to a country, how much am I going to decrease premature mortality? How much am I going to decrease the new cases of poverty or increase the income of people in the country? And that is a critical question.

This leads to some interesting approaches that are very different—and with this I’ll conclude—from the way the, for example, WHO has initially determined to allocate vaccine. In their initial suggestion, the WHO said they we should distribute evenly across all populations, up to 20 percent of the population in every country. So that their notion, and this appears to be, well this is equal concern, everybody, every country is the same, and they all get up to initially 3 percent of the population, and then up to 20 percent of the population.

I say it appears to render equal moral concern, but remember when we were laying out that principle, one of the things we said is being at a hotspot, being at higher risk is a relevant moral consideration, and this principle of every country, getting up to 20 percent—enough vaccine for up to 20 percent of the population really doesn't reflect equal moral concern. If you're in a country like Brazil that is having a lot more COVID, it seems a vaccine is more relevant than if you're in Taiwan, where it had less than 500 cases and seven deaths altogether.

So, I'd like to make the analogy to show that this principle or this way of allocating is really ethically flawed. Think about the following scenario. You're an emergency room doctor. The emergency room is packed with people. You don't go out to all the people and say, “Alright, everyone's getting five minutes.  It's equal. I'm going to give everyone the same.” No, you go out there and you say, “Alright, who's got severe illness? The person with a heart attack, they get time. The person who has a sore throat or recently broke their arm, they're going to have to wait until I can attend to the people who are sicker.”

Distributing your time or a vaccine equally is clearly not ethical in that context. We would seriously criticize that emergency room doctor at 5 percent, no matter how sick you are. And you can see that if you focus on that, New Zealand gets just as much vaccine as Panama does, even though the outbreaks are very, very different between those two countries.

Our approach is: where are the hotspots? Where can the vaccine do the most good in terms of reducing premature deaths? And then, after that's, you know, largely taking care of reducing poverty, increasing income, and getting kids back to school. And those are very big, different moral approaches to distributing among countries. Now, once you're within a country. The country distributes to individuals, but the distribution across countries is really focusing on where countries go even though the ultimate recipients are individuals—on, which country gets a vaccine, even though the ultimate recipients are individuals themselves.

So, with that I'll stop. Probably, I’ve gone on too long, Tom. Anyway, over to you.


Thomas Bollyky: No, that was perfect, and I think you laid out the ideas very well. So, thank you for that. Let me ask first, a foundational question, and then I'll ask more about the proposal.

So, you laid out a number of scientific questions we don't know the answer to, yet. We don't know how effective the vaccine will be and at what. We don't know how durable the protection will be and in whom. Many of these clinical trials, of course, like many clinical trials, involve healthier subjects, so they're not broadly representative of the population.

Connect how the uncertainty around the science affects allocation. How will the answers to those questions—how should they affect allocation and on what assumption, are we making those allocation decisions now, with the state of knowledge we're likely to have in a couple of months?


Ezekiel Emanuel: That's a fantastic question. And so, I like to say that the ethics has to be married to the appearance, and you cannot simply layout ethical principles and suggest you don't attend to the evolving epidemiology and, as you point out, science around the vaccines.

So, to give a concrete case, as the epidemiology changes, you may change your allocation. What we do is we define the goals of the allocation, based upon ethical principles, and we define a metric that should be used. So, if the science change, if we learn about transmission, and we can decrease the premature deaths by targeting transmission rather than targeting those who are at high risk, that's what we should do. So, we don't try to be overly prescriptive.

Let me contrast that with the view that has often been out there, and I didn't elaborate this, but that we should give vaccine first to cover health-care workers and those people over sixty-five. Well, that reflects a judgment on two levels. One that health-care workers are at high risk, and two, that people over sixty-five are at high risk.

Health-care workers may, in fact, not be at high risk. You know, today with good PPE and knowing how to put it on and take it off, the health-care infection rate in American hospitals plummeted to the health-care workers. It's not at all clear to me that they’re at high risk, and they should be at the top of the priority list. And yet, commonly, the National Academies report had healthcare workers at the top. The WHO talks about that the way they get to their 3 percent of the population initially is by figuring health-care workers and people over sixty-five.

A second contrast is if you look at distributing on the basis of health-care workers and people over sixty-five, there are groups that are discriminated against. The disadvantage are—the disadvantages compound, not relieve. So, you know, who’s in the healthcare worker category? Well, it tends to be white upper-middle-class people. People over sixty-five—minorities, you know that the mortality rate in the United States, among minorities—30 percent of the mortality among blacks is under sixty-five. 13 percent of mortality in whites is under sixty-five. So, if you say we're going to focus on people over sixty-five, you’re compounding disadvantage. You're not relieving disadvantage, and those seem the wrong goals, in our opinion.

That is why the metric we use is actually not age related; it's years of life lost. It’s how much before your life expectancy—not your life expectancy, worldwide life expectancy—might you die. That's a much more important number, and it doesn't compound disadvantage. It treats every person, regardless of country, the same because we use a worldwide standard of mortality. We think that's a much better way to go. And as the epidemiology changes, as I mentioned, the characteristics of actually becoming clearer, that will get factored in.

You'll ask the question. Alright, let's project if we use this vaccine in a country, how many deaths can we prevent? How much poverty can we alleviate? We think that's a much better approach than classifying, once and for all, health-care workers at the top, people over sixty-five at the top, or whatever category you want.


Thomas Bollyky: Great. On the health-care worker side, let me ask a follow up. So, the rationale that's often put forward for prioritizing health workers is the indirect health toll that would come from if many health workers succumb. So, there's of course the direct toll of the virus. But if, ultimately, you have health systems collapse, you would have a knock-on effect. How does your model take into account that?


Ezekiel Emanuel: Well, first of all, if, if—you know we saw this in HIV. And I think this is where we're seeing this, that a lot of the people who got infected in sub-Saharan Africa were health-care workers, and it did lead to undermining of health-care systems that already had too few health-care personnel. But in the United States, as we mentioned and in many countries, health care, the risks to health-care workers aren’t large, and so you're not going to undermine the health-care system.

Early on when we didn't understand much about this illness, PPE wasn't available, people didn't know how to don it, then they were at high risk. But we are, you know, six months in, we've learned a lot, and we've learned a lot about how to care for people, and I don't think they're at high risk. You might say, well, we want to recognize their sacrifice. But, you know, there are lots of people who've sacrifice in this situation, you know, kids who are learning online, people who are working in grocery stores or in meatpacking plants. You know, it's not clear to me health-care workers are pivotal in the situation and at high risk that would justify them being at the top of the priority list.


Thomas Bollyky: Great, I'll have two more questions, and I'm going to turn it over to the audience. And I already see questions, hands raising. So well, I'll try to keep them relatively brief, so we can have as many people ask questions as possible.

So, WHO’s proposed allocation approach of 3 percent to each country seems to me like classic WHO, or in two respects. One, it’s a member state-driven organization. So, they would like to put forward a solution that potentially offers something to all member states, which is often what you see in WHO policies. The second is, of course, they are trying to get governments to sign up to COVAX, so having something where the benefits for signing up are quite immediate makes that task easier.

How, from an ethical perspective, how does one weigh those two sets of concerns? This policy enables WHO to be at the lead of this because they're addressing the needs of the broader World Health Assembly, and second, there’s the practicalities of getting people to actually commit to a multilateral approach.


Ezekiel Emanuel: Tom, you've answered your question. Its political expediency dressed up as ethics invoking equity. I have no problem saying, look, we've got this problem of political—this political problem. We've got to get all these countries signed up. Although remember, of the 170 plus that they're going after or think they're going to get, almost all of those are going to be countries that are going to need the vaccine, not that they're going to finance the vaccine or pay. And so that, it's not clear, you know those countries have to be on anyway. It's the U.S., the EU, Australia, Japan, those are the countries that could contribute here—either vaccine or, more likely, money to buy vaccine—that really need to be brought on. And saying, you get 3 percent or 20 percent not going to be persuasive to them.

But I, what I do want, and I think it's really important to be clear, is that it is a political consideration. I'm not saying it's irrelevant. I'm a realist just like other people. That's a political consideration. Don't dress it up as an ethical—we've made an ethical judgment here, this is the ethical approach. It isn't the ethical approach, as I pointed out, and I really, as an ethicist and a person with a PhD in political science, I know how to differentiate political expediency from ethical principles. This is political expediency. Let's just call it what it is and say, yes, we'll get to ethics once we have people signed up and we have a pool of money.

You know, my hope is that, you know, the United States, the European Union recognize the importance of taking world leadership, and part of taking world leadership is, you know, we have to contribute to the COVAX facility. I've called for this too. What does that mean?

Well, they're not asking for a huge amount of money. I mean, I don't know how many trillion dollars we're already into this thing. They're asking $16 billion for vaccines. You know, that should be relatively easy for the United States. Part of our vaccine thing is we're going to give you, you know, if we don't do the whole amount, we're going to give you $10 billion because it's important for us. This is a worldwide pandemic. We want to get back to normalcy. We need the world to be vaccinated. Otherwise, you're not going to have travel, you're not going to have tourism, you're not going to have lots of other things that are essential, both to our economy, but also to the way the world works.

So, it makes perfect sense for us to say $10 billion. What is $10 billion in this moment? It's nothing. And if we contribute it, we show up, demonstrate our leadership in the world. We put pressure on the EU, Japan, and others to contribute. I think writing a $10 billion check would be, you know, the easiest thing we could do, and you wouldn't have to say to the American public, “Guess what? We're taking vaccine away from you.”


Thomas Bollyky: Great. Well, those of you following along, who aren't following the debate around vaccines as closely as Zeke and I, unfortunately the president has announced the U.S. will not, in fact, participate in COVAX, given that it is, and this is a quote, being organized by the corrupt World Health Organization and heavily influenced by China. So the US will sit this one out. Obviously, they are providing some support to GAVI as they regularly do, but in terms of COVAX, we will sit that one out. Just to make sure everyone knows that.

One last question, which you actually address in your paper, but I want to draw you out because I think it'll save you some time as we get through the Q&A. I'm sure it has not escaped people's notice that the countries that have the greatest coronavirus burden are countries like the United States, currently France, Spain, the UK, Brazil, India—all countries with some resources. Is this an issue for your model, where the premature deaths that are likely most likely to occur are occurring in countries that may have other avenues for addressing this pandemic other than taking up vaccines that could be devoted to countries that may not have those resources to purchase them?


Ezekiel Emanuel: So, you know, you have to send the vaccines where they're going to do the most good, and if they're going to do the most good in Brazil or Peru, that's where you have to send them. And it seems to me that's fair.

You know, is it make any sense to send it to, you know, Zambia, if Zambia’s not got a problem? If for whatever reason, and we still don't know, and it does appear that, you know, Africa is not having a big problem. Contra, and it's not just not testing, contra to everything people have expected, it just doesn't seem to be. Maybe it's they have cross immunity because of high levels of other infections, who knows. Well, what sense would it make to send a vaccine to Zambia under those conditions, just because it's poor?

Look, I want to, as I mentioned, we want to mitigate disadvantage. But if the disadvantaged countries aren't suffering here, it doesn't make sense to prioritize them. You have to prioritize the hotspots where the vaccine is going to do the most good in terms of premature death, alleviating poverty, etc. Now, by the way, again, by the way we measure morality—premature mortality—and poverty and income, those disadvantaged countries, if they become hotspots like in Latin America, they do rise up, and they will get priority on the vaccine, and that's super important.

Secondly, if places like Zambia or Ethiopia are having economic difficulties, not necessarily directly from COVID, but indirectly because of trade and stuff, in the second stage where we're trying to alleviate economic disadvantage, they might be at the top. Again, it depends upon the empirics. What we try to do, again, is set out the principles, set out the goals, set out the metrics, but the data has to help determine how those are going to be implemented at the moment when we have vaccine.


Thomas Bollyky: Great. You've been patient long enough, gentle audience. I will call on you now in the order that I see your hand. If you'll just say your name and affiliation when you start, and also please make your questions sound like a question, so we can get to as many people as possible. And I will start first with Peter Katona.


Ezekiel Emanuel: Unmute yourself.


Peter Katona: Got it. Hi, nice to see you again, Tom. This is Peter Katona. I chair the infection control working group for UCLA to bring the campus back on track. My question is—


Ezekiel Emanuel: So, you know more than I do. Alright, now I'm going to be embarrassed.


Peter Katona: I know much less than you do. But my question has to do with the allocation issues you've mentioned. You know, we have 30—40,000 people being tested in phase three studies seems like a large number. But only about 150 of those are going to move in one direction or another direction. So you're looking at a very small number that that is it basically encompassing a large population. My question is the subpopulations that you've mentioned, you know, whether they’re years of life lost or whether they’re health-care workers or whether they're over sixty-five. When are we going to have the power from the numbers to be able to make a comment about those subpopulations, as opposed to, yes the vaccine works great, let's go with it?

So that's my question. I might also add that my daughter's a hospitalist who takes care of COVID patients, and she feels much safer in the hospital than she does in the grocery store.


Ezekiel Emanuel: There you go, because she has full PPE, right? So, I think your point is exactly on target, which is, you know, we're going to make an initial judgment on the basis of 150 events. That is nowhere near enough to know, does it work equally well in minority populations or not? How does it affect, work in elderly populations? People with comorbidities? People who have obesity? That, frankly, even in the 30,000 we're not going to have enough. We're going to have to follow those 30,000 out for twelve to, I think most of the trials are, twenty-four months.

But you're right, we're only going to begin knowing that if we have good data collection from the actual real-world immunization. And that gets back to lots of other, you know—I'm an oncologist. I don't know if Tom mentioned that, but I'm an oncologist and we do a lot of trials, you know, cancer patients, and then you let out your chemotherapies that have been proven into the real world, and inevitably, the results are a little different. Because you're doing it in a wider population. There are administration issues that creep in, because you don't, can’t spend this much to make sure everyone's getting admitted.

And here, I've mentioned one that I'm particularly worried about, which is how many people are going to get one dose and the second dose? They're going to be lost to the follow up. How much are we going to spend in terms of effort to make sure everyone gets the two doses? We have some vaccines that required two doses, you know, some of the infant vaccines. The shingles vaccine required to doses. HPV.

But, you know, that is going to be a major problem, especially, you know, if you look at the preliminary Pfizer data, a large number of people get muscle aches and chills. And it may be that after one round of muscle ache and chills, people say, I'm not sure I want to rush in and get that second round of muscle ache and chills. So, it is going to depend upon real world data collection, and that's going to take a whole year, at least.


Thomas Bollyky: Let me ask a science question, then I'll turn over to the next in line. As you mentioned, there's going to be multiple vaccines probably that emerge from this process, and the later ones will be better. How will that be addressed in terms of the allocation and distribution? Will people be mixing and matching different vaccines? Would you let the audience know whether that's advisable, and then also, you know, what the current capacity is for tracking vaccine administration?


Ezekiel Emanuel: Well, Tom. Currently, it won't be advisable, because we'll have zero data if you start out with Pfizer vaccine and you get the Moderna is that okay? Or you start with Pfizer and J&J gets approved, we have no idea. Well, and it may well be that some combination turns out to be better than, you know, two doses of a Pfizer vaccine for a whole series of reasons. But that's not a good thing to try on your own. Your own self experiment is a very bad idea. And it'll take time before we know what the optimal combination is, or which vaccine is really truly much more effective.

So, you know, and part of it is we may be willing to trade off some effectiveness for ease of administration, you know. One dose, if you had one dose, you might say, well, a one-dose vaccine that's, you know, 65 percent effective may be worth a two-dose vaccine that 75 percent effective just because it's so much easier from a logistical distribution administration standpoint. And I think those are going to be judgments that we'll have to see later on as things abound.

And also, as I mentioned, another key element is going to be does, which one prevents transmission.


Thomas Bollyky: Makes sense, thank you. Thank you for being patient. I have Yanzhong next and then we'll turn to Bob after that. If anybody else would like to get in the queue, please press the raise hand function at the bottom. Please go ahead, Yanzhong.


Yanzhong Huang: Thank you, Tom. Thank you, Ezekiel. This is a very informative, insightful presentation. Really appreciated that. I just, I have a question about the clinical trial in terms of like who are under the trial, because we know that typically, right, a vaccine will be tested first in healthy adults. Then, potentially, what other segment of the population like pregnant women, children, elderly people, and that is probably why the U.S. the pharmaceutical companies they have delayed the testing on these groups. But my question is that, is it ethical or considered ethical of having or administering the vaccine on this group of the population—of prioritizing this group of the population—without even conducting clinical trial on this so-called high-risk population?


Ezekiel Emanuel: Well, first of all, you are right in traditional ethics of research. You put children after you get a signal that it's safe in adults. We don't want to expose them before we have a deeper understanding of how safe’s what we're doing. In this case, we are enrolling older people, but I think as the previous question, or Peter made clear, you know, we don't have a full complement, and we don't have a comprehensive group, and it, it is one of the issues that we're going to have.

You know, we're going to, given the exigencies of the situation, we are going to use a vaccine early on that mainly will have been given to healthy, relatively healthy people between nineteen and sixty-five. Some of the people over sixty-five will be, will have gotten it. Some people with some comorbidities will have gotten it. But it won't be a large number. And so, we will be making some leaps, as it were, especially given the priority groups.

Now the first priority groups worth about 5, 6 percent of the population are health-care workers and first responders, who are relatively—not all of them, some of them will have, you know, diabetes, or might have some other health problem, might have been cancer survivors. So we will have some, most of the people in that first tranche, defined by the National Academy, will fit this category. And we'll have more experience, and then get more experienced with the clinical trials.

The one I'm most concerned about is, you know, moving to kids and children, because I think eighty million kids, they're an important group, especially if you want to reopen schools or if they may be critical to transmission. So starting trials on them is very important, and we shouldn't give them a vaccine that we haven't tested in kids. But then again, you're going to have a situation of parents saying well, I'm happy if it's proven safe and effective in kids to give it to my kids, but you know enrolling them, maybe, maybe not. And that's a serious problem. Given, you know, the worries and doubts, we're experiencing about the process of approving a vaccine.


Thomas Bollyky: Zeke, I have Bob and Gary next, but I wanted to just make sure you had a chance to clarify something for the audience because you referenced the National Academy and their tranches.


Ezekiel Emanuel: Oh, yeah.


Thomas Bollyky: You've talked about what should happen. What is going to happen domestically in terms of allocation? What are we saying? How does that process fit into those?


Ezekiel Emanuel: Good question. We first, first of all, the CDC and the NIH asked the National Academy of Science, Engineering, and Medicine to look at the allocation of vaccine, and the ethics of it, but also some of the practicalities of it. They have, in rapid time—and I really commend them, it's just been remarkable how fast they've worked—in rapid time, they have produced a draft report, which was out for public comment last week, I think. And rapidly refined it in light of that public comment. And they divided into four tiers, and the first tier is divided into an “A” part and a “B” part, and the “A” part are health-care workers and first responders. The “B” part are mainly people of any age who are at high risk for complications and death from COVID, so people who have comorbidities—severe, known severe comorbidities like cancer, like heart disease, like sickle cell anemia, and those kind of diabetes, obesity.

So, that that first group of health-care work, frontline health-care workers and first responders are about seventeen million, and the next group is, I think, it's another thirty plus million. So, the first two groups are round about fifty million people, which ironically is pretty close to what we're expecting to have available either before the end of the year or early next year. Again, lots of the productions of how much we’ll have, they’re projections, and we haven't really tested the system, yet.

Now, the National Academy doesn't determine the priority groups. The priority groups, the ACIP, this advisory group to the CDC on immunization policy, they have a final determination, but then the CDC has got to put a blessing on it and tie a bow. There's also NVAC, which is the National Vaccine Advisory Committee, I think, and they are also looking at this question.


Thomas Bollyky: Right. I lied. I have one quick follow up, just so people know. How will we know if you fall in one of those categories? What U.S. database will you have to prison proof? How will they actually identify who has comorbidities, who's a health worker?


Ezekiel Emanuel: Tom, that is a really good question, and I have to say I'm scratching my head, especially if you think the first problem is that, say, the Pfizer vaccine is the first out of the box, and, you know, it's the one that we start with.

You know, the Pfizer vaccine can't be done in physician offices or your corner CVS, it has to be done in larger facility, just because of, it has to be held it you know negative seventy degrees centigrade, round about ninety-five, a hundred degrees, negative hundred degrees Fahrenheit. Physician offices can’t handle it. Your corner CVS can’t handle. It's not even clear how many of your hospitals can handle it. And also, therefore, when it's being distributed to be, as I mentioned, large quantities, a minimum of a thousand doses. So, physician offices are definitely not taking it, and those doses are only good for ten days as far as we can tell, in terms of stability.

Yeah, it's unclear to me what a person—how are you going to turn determine a person with two comorbidities? They're going to come in with here's my insulin, and here's my, you know, cholesterol medication? I just don't fully understand. Maybe their doctor will certify. Maybe that's what we're going to do—we're going to create a forum for doctor certification, but I do think this implementation thing is non-trivial. You know worker status, age, those are things we can we can pretty much do. You know, health status is something we pretty much can't do, easily.


Thomas Bollyky: Good news. We have five weeks, apparently, to sort that out. So that'll—


Ezekiel Emanuel: Trivial problem.


Thomas Bollyky: I have tested Bob’s patience for too long. Please go ahead. You’re next.


Bob Sorenson: That's where the unmute is. Yeah, I have just one point of clarification before my question, if I may. As I read the National Academies study the health-care workers category is cast very broadly to include a lot of people like, say, home health-care workers and nursing home health-care workers who would be, you know, a much less protected and much less privileged class than just doctors and nurses, as I think you had put it earlier. So that's I think that's less problematic than a very narrowly cast health-care worker category.

But the question I had was, was quite different. I know you've been at those policymaking tables in Washington, and I have some concerns both about actually calculating the metrics that you're proposing in the paper or the latest paper. I just—do we actually have any idea how years of life lost will, per dose, will vary in various countries or, you know, what the actual magnitude of income either, you know, increased income in? As we go through this allocation, I assume that has to go through some kind of modeling process. So, I'd like to hear a little more about how that's going to be calculated and, you know, the secondary, the follow-on problem, of course, is the policymakers who are going to say, “Ha. What's this?” You know, you know how these meetings are, you've got people from all these different agencies, and I think that's true in every country. “What do you mean? What do you mean the amount of income we saved per dose of vaccine?” And you say, how you got that. So, I think that's good. The question should be clear enough.


Ezekiel Emanuel: Yeah, no, I look, I think you're right in expressing some skepticism. On the other hand, I will tell you as part of this process we did contact some people who are doing modeling. Some of them are doing modeling on the basis of comorbidities, and I asked, you know, could you adopt the approach we're suggesting into your models. And part of this is a chicken and egg.

If you don't say, you know, we want to use SEYLL, standard expected years of life lost, people won't model it. On the other hand, if you say this is going to be a pivotal effort, the modelers out there will do it, and they did not think it was a problem. One of the reasons we picked SEYLL, standard expected years of life lost, is because it's already used in global health. It's used in the global burden of disease. We didn't invent it, it's not, you know, complicated, you know, new idea out of Zeke Emanuel’s brain. No, you know, we took what was available and what seemed to be the best in terms of able to use and fulfilling the ethical principles we identified. And so that was the goal. And I think, again, again, it's partially an incentive structure. If you say this is going to be a really important criteria, you get a lot of modelers working on it.

The second thing is you already have the World Bank looking at poverty and drop in GNI from COVID across countries. You know, you have the world, and now they're disaggregating it across country. So, I think there, again, I think it's a valid point—do we have the modeling capacity? Now once you get in a room, you know, again, part of part of what those committees do is, what is the model that's available to them and that gives them data? And I think that if we had modelers doing this, it would be good.

You also have to contrast it with what's the alternative? Is the alternative population based? Is the alternative over sixty-five based? All of the—I won't say off, that's a mistake—many of the alternatives that have been batted around, I think, are you know, full stop, not fulfilling the ethical criteria that we've put out. If you allocated by strictly by population, I've explained why I think that’s unethical. If you allocated by health number of health-care workers or number of people over sixty-five, big developed countries win that race. That doesn't seem to correspond with mitigating disadvantage. So, I would doubt.

I will tell you one, which I think is a little more complicated, Bob, that I've just begun to think about. If you look at the data, and it's, I would say, good but not great, it appears that men are much worse off than women. We don't know why. Okay. So, does that mean that, you know, under equal moral concern, maybe in these groups, men ought to get priority over women? I don't know that we have, you know, I said at the start, you know, equal moral concern is, in essence, a nondiscrimination principle. Well, and then I said, you know, things like sex, geography, or race and religion are not. Well, maybe sex is a relevant principle. I just, you know, so I'm raising this. I haven't thought it through. But I think it's non-trivial, especially if the delta between men and women is big.


Thomas Bollyky: Provocative. That’s really interesting. Let me turn it over to Gary to ask the next question. Just as a reminder to the audience. If you'd like to get in the queue. We have a little time. So, there's more time for other questions. Otherwise, you'll be forced to listen to mine. Please put the raise hand function. With that, I'll turn it over to Gary.


Gary Cohen: Well, thank you, Thomas, and thank you, Ezekiel. And Thomas, I was so happy to see that mute unmute appear, just as you asked the question. So, anyone who wants to ask the question, it comes up right when Thomas calls on you.

So, I'm going to share a basic principle, and I'm not a hockey fan. I don't follow professional sports, but it's going to come from sports. That you go where the puck goes, not where the puck is. And I'm wondering whether the methodology that you shared, more or less goes where the puck is rather than anticipating where the outbreaks might occur, which could be particularly relevant for something like vaccination because it's a prevention measure? And the second part of that question is by going to the hotspots, are we not in some respect, if you will, rewarding those who've exhibited the worst behaviors? If you imagine, you know, countries like our own, which have mismanaged this would be prioritized because we've created these hotspots. So, are either of those factors relevant considerations?


Ezekiel Emanuel: So, I think both of them are worthy of consideration. There's a limit to how well we can predict where the next hotspot is going to be. You know, almost all of us involved in this situation pretty confident we're going to have an uptick in the Northern Hemisphere come November, December and that has to factor into how you think about it. How bad that uptick is going to be, no one can precisely tell you. We have projections, but those projections are projections.

So, absolutely agree with you. Part of what you want to know is not, you know, you use the past to predict the future, but what you want to know is if I give a million doses to Greece, how many premature deaths, how much poverty am I going to alleviate? And given the situation when we can do it, which is, you know, in a few weeks that's exactly right. Will there be uncertainty? Yes. Are there some factors we can bring into account in modeling out the future and whether Greece is going to be on the increased number of cases or on the downside? Absolutely. So, I think there is uncertainty, as you point out, in knowing where the hotspots are going to be. But there it but we get, we're going to have to take into account that as we try to calculate things out. So, I think that's right.

Big question, and we've been asked this over and over again, and we have a small section in the paper about, you know, the incentive structure. Look, you know, you and I, and Tom, we have all, you know, we're as powerful as you get in America, and we have no influence on how this administration's responded to this pandemic. Right. Americans shouldn't suffer if their government is doing a bad job, and the ultimate recipients of a vaccine that you want to benefit are individuals. So, I don't think it giving it to countries that have done a bad job is unethical. In fact, I think it's ethical even if those governments could have done a better job.

But I also think there's not really an incentive, as you might say, well, you know, maybe Taiwan now will do a bad job. First of all, it would be bad from a health standpoint. It would be bad economically. And there has been huge soft power advantages to countries that have responded well. The stock in New Zealand, you know, the people looking at New Zealand, looking at the Prime Minister, they're like awestruck by how well she's done.

Take another country, Greece. Right, Greece, it's done way better than we have, way better than many of its northern European friends, who were just holding it on the financial short hairs, not too long ago. Government actually executed well. Turned out really valuable internally, domestically from a political standpoint, economically internally, but also in the European community turned out to be incredibly valuable from a soft power standpoint of the Greek people looking at Greece and realizing, yeah, they can actually manage many of these crises.

So, I think it's unlikely that the incentive structure—now, that's different than the incentive structure of a November 3, election, and but, you know, what can I do about that?


Thomas Bollyky: Great. We have two questions. I'm going to take them both. Please make them short, and then we’ll have Zeke get the last word. We have Shannon, and then we'll have Chris afterwards.


Shannon Kellman: Thanks so much. Hi, I'm Shannon Kalman. I'm the policy director at Friends of the Global Fight Against AIDS, Tuberculosis, and Malaria. We are advocates on behalf of the Global Fund, and so a slight pivot away from vaccines to think a little bit more about diagnostics and therapeutics. How do we make a determination around where to send diagnostics and therapeutics and PPE, and given the—let's call it an uphill—challenge to convince policymakers of the importance of that part of the response, how would you pitch that to a U.S. policymaker about the importance of that?


Thomas Bollyky: Great. Chris Elias?


Chris Elias: Yeah. Thanks, Tom. Thanks, Zeke. A great discussion. Just Zeke, in terms of operationalizing your principal around disadvantaged areas. Here in the U.S., have you looked at the area deprivation index or the CDC social vulnerability index (SVI)? Do you think one's better than the other? Do you think either is implementable in that, in the kind of case that we're talking about a roll out? Over.


Ezekiel Emanuel: Really good question. So, let me start with the latter. I'm not an expert in those indices. There are disagreements, and I think the CDC SVI, there's been some criticism of it. They have a new COVID, I think, disadvantage metric, which may turn out to be better. But I actually think they are useful in this context of recognizing that minority communities have been hard hit, probably deserve as a community. But I also think you can get at the minority issue by, you know, one of the reasons that they're so hard hit is a lot of their, a lot of minorities are in the essential workforce that is exposed a lot—grocery store workers, meatpacking workers, etc.—and targeting high risk jobs, congregate housing situations, etc. will get at a lot of the disadvantage. And I think that's very, very important. And you can target and put extra emphasis on getting it into disadvantaged communities.

Shannon, you know, I thought this was hard enough to think about vaccines in the international setting. And I think you are challenging the right question, which is, is the framework, the right one for looking at diagnostics? And by the way, you know, the moment we get point of care tests that are reliable like pregnancy tests or like malaria tests, you know, we're going to have a shortage, because there's no way of going from zero to billions produced monthly, that would be needed for them. I'm not, I'm not 100 percent sure of the whole mechanics.

I am 100 percent sure of the three ethical principles we identify as critical. The benefit people and limiting harm, mitigating disadvantage, and equal moral concern as being the right principles. Then, we have to ask, what's the goal here, are the goals here, and what are the right metrics for using for distributing across countries? I’ll just confess I haven't worked all that out, and you know, it's a really important project that needs to be addressed because we are going to have shortages, and we have them, and we're going to continue to have them in the diagnostic and PPE category.


Thomas Bollyky: Great, thank you, Zeke. Not only have you delivered provocative, insightful comments, you have delivered them exactly on time. And that's what we aim for at the Council, so thank you. I hope all of you will join me and miming applause for Zeke for these great remarks, and thank you all for joining today. If you haven't had a chance to read Zeke's paper in Science, I really do recommend it and look forward to connecting with you all at the next meeting.

Top Stories on CFR



As armed conflict rages on, Sudan’s devastating humanitarian crisis continues to worsen. More than one million people could flee to neighboring countries, threatening regional upheaval.

Diamonstein-Spielvogel Project on the Future of Democracy