Page 1 of 12
Perspective Roundtable: Physicians and Execution
DR. ATUL GAWANDE: Welcome to a Perspective Roundtable from the New England
Journal of Medicine. I am Atul Gawande, a staff surgeon at the Brigham and Women’s
Hospital here in Boston and an associate professor at the Harvard Medical School and the
Harvard School of Public Health.
In 1977, Oklahoma became the first state to adopt lethal injection as a method of carrying
out the death penalty, in the belief that it would be more humane than electrocution. The
design of the original protocol for the procedure was written quickly, without any prior
study, by A. Jay Chapman, Oklahoma’s chief medical examiner at the time. And
Chapman’s approach became the de facto standard, as other states followed Oklahoma in
switching to the new method.
That method and the attempt it represents to medicalize execution, in order to make it
more morally acceptable, have led to ongoing tensions between the legal community and
the medical community over whether physicians and other health care professionals
should be involved in the process of putting convicted criminals to death.
On January 7, 2008, the U.S. Supreme Court heard oral arguments in Baze v. Rees, the
case of two Kentucky death-row inmates who argued that the current three-drug lethalinjection
protocol violates the Constitution’s Eighth Amendment guarantees against cruel
and unusual punishment.
Page 2 of 12
With me today to discuss this important case and its implications for the health care
community, we have Professor Deborah Denno, Arthur A. McGivney Professor of Law
at Fordham University School of Law and an expert on death-penalty law, Dr. Bob
Truog, professor of medical ethics, anesthesiology, and pediatrics at Harvard Medical
School and also a coauthor of a brief to the U.S. Supreme Court arguing that the lethalinjection
protocol, as currently constituted, does violate the ban on cruel and unusual
punishment. And we also have Dr. David Waisel, associate professor of anesthesiology at
Harvard Medical School and also the author of a much-discussed article arguing that
physicians should be involved in executions, in order to relieve suffering. Thank you, all
three of you, for joining today.
2. The Lethal-Injection Protocol
DR. GAWANDE: I want to start with you, Professor Denno. Now, as I understand it, the
case of Baze v. Rees doesn’t ask if the death penalty is unconstitutional; it doesn’t ask
whether lethal injection is unconstitutional in general; it asks only whether the specific
formula used in the state of Kentucky is unconstitutional. So the puzzle of it, to me as a
doctor, and to plenty of people who are not lawyers, is what is the big deal about this
PROFESSOR DEBORAH DENNO: Well, the big deal about this case is immediately
when the method was adopted, A. Jay Chapman conceded that there were problems with
this formula. There certainly have been problems within the 30 years that the formula has
been examined and used. And now the Supreme Court is confronting that. The biggest
problem with the formula is the second chemical, pancuronium bromide, that is a
paralytic agent. There seems to be no real reason for its use. There’s evidence that it
paralyzes an inmate, so that inmate is not able to — to express himself if in fact he’s
suffering from the effects of the other chemicals.
DR. GAWANDE: So then, let me ask a little bit about the formula, Dr. Waisel. Can you
describe that three-drug protocol and how it’s at least supposed to work?
Page 3 of 12
DR. DAVID WAISEL: Yes. The three-drug protocol is based, at least a great deal, on
what was considered a normal induction of anesthesia when it was developed. First
comes thiopental, also known as sodium thiopental or pentothal, which is a barbiturate,
which is designed to put you to sleep, create amnesia and anesthesia. Second comes
pancuronium bromide, which is designed to paralyze the muscles. And the third drug,
which is not a drug used in anesthesia, is potassium chloride, which is designed to rapidly
stop the heart. The doses used are massive compared to the doses that would be used in a
normal anesthetic induction.
DR. GAWANDE: You raised, Dr. Truog, whether these are the right drugs in the first
place, in your brief.
DR. ROBERT TRUOG: We’ve taken a pretty strong stand that paralytic agents have no
role in end-of-life care. The concern, at the end of life, is that they can mask the
behavioral signs that we look to, as to whether or not a patient is comfortable. And we are
deeply committed to making sure that patients are comfortable and as free of pain and
suffering as possible during the dying process. And since we have medications that do
relieve pain, that do sedate perfectly adequately, there’s no need to be introducing
paralytic agents into end-of-life care.
DR. GAWANDE: One thing that came up in the course of the oral arguments before the
Supreme Court was: Why do they administer pancuronium as part of the protocol? Can
you explain a little bit about the thought that emerged from their discussion about this,
PROFESSOR DENNO: Well, according to the state, pancuronium bromide is used in
order to enhance the dignity of the inmate who’s dying, because evidence came out that
without pancuronium there might be some jerking or involuntary movements that would
disturb some of the witnesses. So this would enhance the dignity. That I find problematic,
and Justice Stevens certainly did.
Page 4 of 12
DR. GAWANDE: Dr. Truog, is there a medical reason behind it? And has it lasted only
just because of this until-now-unstated role of making things look dignified?
DR. TRUOG: Well, from the point of view of the inmate, the argument seems bizarre, at
best. You know, imagine saying to the inmate, “You have a choice. You can either be
assured of a pain-free death and you may have some twitching and grimacing, or we can
expose you to the risk of an excruciating death, but we’ll make sure that you don’t twitch
or grimace.” I can’t imagine that an inmate would actually consider that to be a real
So if we’re talking about the dignity of the inmate, it’s only in the eyes of those who are
watching that — and, in fact, you know, if that’s all you cared about, don’t even bother
with any of the other drugs, just paralyze the inmate. They will look just as peaceful. So I
think it’s completely specious and has no weight at all.
The number one alternative that’s been proposed has been a very large dose of a
barbiturate. And, you know, I know that there’s a number of experts who have said that 2
or 3 or 5g of pentothal is absolutely going to be lethal. The fact is that, at least in this
country, if there’s anyone who’s ever had experience with giving a huge dose of
pentothal and watching an otherwise healthy person die — now, I’d be very interested in
knowing the circumstances — I mean the fact is none of us have any experience with
this. And my concern is that if you look at a country where they really do have some
experience with it, their findings are pretty concerning.
So, if we go to Holland, where euthanasia is legal, and we look at a study from 2000 of
535 cases of euthanasia, I was stunned to see that in 69% of those cases, they used a
paralytic agent. Now, what do they know that we haven’t figured out yet? I think what
they know is that it’s actually very difficult to kill someone with just a big dose of a
barbiturate. And in fact they report that in 6% of those cases, there was problems with
completion — you know, getting the person dead. And in — what was it? — I think five
Page 5 of 12
of those, the person actually woke up, came back out of coma, you know, despite an
intention to give a lethal dose.
3. Cruel and Unusual Punishment
DR. GAWANDE: Professor Denno, in turning to this three-drug protocol back in 1977,
one of the things you’ve written is you said, “The law turned to medicine to rescue the
death penalty.” What did you mean by that?
PROFESSOR DENNO: By virtue of coming up with a method of execution that makes
an inmate look serene, comfortable, and sleeping during the death process, by virtue of
using a paralytic agent, the death penalty in this country was rescued. The humane
application of a method of execution was a key goal. And the presence of doctors, their
involvement, and the association with medicalizing the procedure enhanced its
DR. GAWANDE: What does it mean to be not cruel and not unusual punishment?
PROFESSOR DENNO: This is what the petitioner was asking the court to do: “Please
provide us some Eighth Amendment guidance, so states can know how to judge whether
lethal injection is cruel and unusual punishment.” So the court is in the position, or at
least we hope, of answering this question.
The Eighth Amendment has never said, nor have the petitioners ever argued, that
executions are to be pain-free. The question is whether or not that pain is unnecessary,
whether there are alternatives.
DR. GAWANDE: Chief Justice Roberts asked, “Do you agree that if the protocol is
properly followed, that there is no risk of pain?” And the attorney for the prisoners
waffled. And I think he waffled because it’s a medical question. And so I want to turn to
Page 6 of 12
both of you, Dr. Waisel and Dr. Truog. What’s your take on the answer? Is there no risk
of pain with the three-drug protocol, if properly followed?
DR. WAISEL: Define “properly followed.” In other words, the protocols list that this
should happen and that should happen. But does that mean if everything happens
correctly, there are no problems with insertion of intravenous catheters, if there’s no
problems with mixing up the medications, there’s no problem with delivery of the
medications, then yes, it would be pain-free.
DR. TRUOG: I think I agree with Dave. You know, thousands of times, every day,
people are getting anesthetics around this country. Many of them involve the use of
paralytic agents. The risk of awareness under anesthesia is very, very small. It does occur,
and that’s, you know, that’s the risk of the anesthesia being too light under the cover of a
paralytic agent. And that’s what, you know, anesthesiologists live in fear of that
happening. And fortunately, it occurs very, very rarely.
4. Risk of Errors
DR. TRUOG: I think the issue, here, is that people go to school for a long time and do
years of training in order to be able to do this well. And certainly, everything that I’ve
read is that the adequacy of the training for the people that are doing it in lethal injection
is nowhere near adequate.
DR. GAWANDE: So there is a practical empirical question, which is: How likely is it
that errors will occur with the current process of lethal injection? By one measure, there
have been 40 clear botched executions out of a little over 900, which suggests a 4 to 5%
rate of failure. But then, Dr. Waisel, you started to talk about other problems that occur
along the way, starting with drug preparation, where we’ve seen problems appear,
perhaps, at an even higher rate. What’s your sense of how often problems occur, and
what the problems are?
Page 7 of 12
DR. WAISEL: We have no idea what the error rate is, because there is no oversight,
there is no public reporting. So there’s no way of knowing what’s happening where. And
the information you hear worries me, that the process is of less concern. So, for example,
I believe the case was from Missouri, in which they pushed the three drugs, and the
inmate didn’t go to sleep. He was sitting there looking at them.
And he realized the strap holding the arm was functioning as a tourniquet, not permitting
the drugs through. So they loosened it up, all the drugs came in at once. Now in that case,
I’m highly confident that the inmate experienced a great deal of pain from the potassium
chloride. And yet I believe it was the sheriff who was quoted as saying, “Eh, no big deal.
It’s not like we hurt the guy,” or something like that. And so I think that your 4 to 5%
number is dramatically underestimated.
Second, I would argue that, you know, when we talk about medical error, we also talk
about the consequences of if that medical error occurs. We should approach this with the
gravest manner possible. And this should be done perfectly.
DR. GAWANDE: Dr. Truog, we’re talking about you put an IV in, you give some
medications, that’s a routine kind of procedure that occurs thousands of times a day in
any typical hospital across the country.
DR. TRUOG: First of all, putting an IV in is not as easy as it may sound. And being
certain that it continues to remain in the same place also requires, actually, quite a bit of
experience, because these catheters can become dislodged, they can go into the tissue,
and then they won’t work anymore. Furthermore, we know that many of these inmates,
by virtue of their past history of drug abuse or obesity or being muscular, can be very
difficult to start IVs in.
Now, in a hospital setting, we have a lot of different ways of approaching the situation
when we can’t get an IV in or it’s going to be difficult. Most commonly, we’ll just put in
a central venous line. But there again, that requires a great deal of training, far beyond
Page 8 of 12
anything that would be readily available outside of the medical profession. The mixing,
the administration of the medications, routine in any operating room in this country, but
far from routine if you haven’t done it before.
And indeed one of the mistakes that I know has occurred happened to me, as an
anesthesiologist early in my training, when I injected the paralytic agent too quickly after
the pentothal, and they precipitated in the tubing. The tubing turned into a piece of
concrete. Suddenly, I had no IV. And, you know, thank goodness I was surrounded by
very experienced anesthesiologists who stepped in, within moments had another IV, fixed
the problem, and it’s never happened to me since. I learned my lesson. But I know that
that has happened in executions, and it could be a disaster.
5. Physician Involvement
DR. GAWANDE: When Justice Ginsberg asked of the petitioners — she asked, “Is there
a way to ensure proper use of the three-drug regimen?” the petitioners said, “Yes, with
the direct involvement and control by medical professionals.” And so now, then, of
course we come to this fundamental question that all three of you have weighed in on in
various ways. And that’s whether physicians should, therefore, take charge to make death
less painful in these instances.
DR. GAWANDE: Dr. Truog, what’s your take on that, I think, fundamental question: If
you were to have to be executed, wouldn’t you rather have a capable, specialized
physician doing this job?
DR. TRUOG: Sort of as a philosopher, if I think of the kind of a hypothetical where you
have an inmate who is about to be executed and knows that this execution may involve
excruciating suffering, that inmate requests the involvement of a physician because he
knows that the physician can prevent that suffering from occurring, and if there is a
physician who is willing to do that, and we know from surveys that many are, I honestly
Page 9 of 12
can’t think of any principle of medical ethics that would say that that is an unethical thing
for the physician to do.
DR. GAWANDE: Let me ask a follow-up on that, because I’d be curious to hear from a
nonmedical person’s perspective, but also an expert in the law on this, how you take the
role that physicians should play in this. You’ve written a recent law review where you
would like physicians to actually play more of this role, at least in constituting the
protocol to minimize suffering. Is that still your take on what our role should be?
PROFESSOR DENNO: If we’re going to be executing people, I would prefer to have a
method of execution where a doctor did not have to be involved, where medical expertise
would not have to be necessary. If in fact we’re going to, however, have a method that
would be cruel and constitute suffering if we didn’t have doctor involvement, then it
suggests to me that if there are physicians in the country who are willing to be involved,
or medical personnel, then I would like to think that they would not be chastised or lose
their license or punished by the medical profession for volunteering to take part in an
execution, to relieve suffering.
DR. GAWANDE: Well, there is an argument that I think we have to grapple with.
Steven Miles has made it, a physician and medical ethicist. When he was looking into the
records of what happened at Abu Ghraib, the very exact same question happened. It was:
if you’re a prisoner who’s about to be tortured, wouldn’t you rather have a doctor
available, to help you survive the torture, so that it could be titrated in ways that avoid
killing you inadvertently and also provide some guidance on how it might be made more
effective in various ways?
DR. TRUOG: I’ve thought a lot about Steve Miles’ work on torture, and actually I
welcome the analogy to torture, because there’s been a lot written, there’ve been
symposia about whether physicians should participate in torture. And I think it all sort of
misses the point. Of course, physicians shouldn’t participate in torture. But,
fundamentally, it’s because torture is wrong. And this is sort of returning, now, to kind of
Page 10 of 12
my views about physician involvement in capital punishment. While I think at one level,
we can justify it, as Dave does very, very well, but I think it’s to miss the bigger picture. I
really believe that capital punishment is ethically wrong.
And, you know, I think that living in the bubble of the United States, as we do, it’s easy
to lose sight of just how much of an outlier our country is. You know, the United Nations
has recently voted to ban capital punishment worldwide. Over 100 countries have. We
stand among a small group of countries that still do capital punishment that I really don’t
think we want to be, you know, in their company.
DR. GAWANDE: So then, when we come to this question of where can the remedy be
found, the directions that seem to be posed are we involve physicians more and let them
treat the prisoner as a patient, or we come up with alternative protocols that don’t involve
physicians at all. The judges in the oral arguments seemed very uncomfortable with
trying to reinvent the protocol, for the reasons Dr. Truog has just mentioned. That is, it’s
not clear that any alternative protocol has enough experience to show that it works 100%
of the time and it’s pain-free. And so the natural place the discussion tends to go is
towards trying to make sure there’s enough professional involvement. Is that right,
Professor Denno? Was there a disinclination among the justices to be reinventing the
protocol on the spot?
PROFESSOR DENNO: I think there was a disinclination. I think what became clear
during some of the arguments is there’s probably not enough information for the justices
to determine what the next direction should be. You know, my recommendation has been
that there be a panel of experts who would propose what would be a viable method of
execution and offer information that seemed to be sorely needed during the oral
Page 11 of 12
DR. GAWANDE: I have to say, it makes me deeply concerned, though, imagining us
sitting around a table at a conference, trying to figure out various ways of executing
people, and then the prospect of what that becomes, that we either figure out that
physicians have to be continually actively involved, and we create a specialty of the
It may not be possible for the court to say that doctors would be allowed to really treat
inmates as patients. And so then my question to you, Professor Denno is: Is it a realistic
thing that a physician could treat an inmate as a patient and that the court would let them
control the protocol, make judgments about how to make the suffering less or more, and
leave them free to have that professional role?
PROFESSOR DENNO: I guess my best answer to you is that they’ve been doing that
for 30 years. There have been physicians, as you know, involved in lethal injection since
the very first execution in 1982 in this country, in the involvement of Dr. Ralph Gray. We
don’t, because of secrecy and the lack of information, we’ll never know, at least up to
this point, the full involvement of doctors. But we have many examples of doctors having
been involved — the doctor in Missouri, Dr. Carlo Musso in Georgia, etc., who have
made these kinds of discretionary judgments about drugs or chemicals and what should
DR. GAWANDE: So Dr. Truog, if the court says, “We need this to go to an expert
panel, with physicians, lawyers, public citizens, to determine a new protocol for
execution,” would you participate on that panel? And should other physicians participate
on that panel?
DR. TRUOG: I would not participate on that panel, because I don’t think that capital
punishment is ethical. I think other physicians should be free to participate on that panel.
And while I wouldn’t want to prejudge how they might come out, certainly, from
everything I’ve read, I can’t imagine that they are going to be able to develop an evidence
Page 12 of 12
base for any other approach that is likely to be successful without the immediate presence
of a physician. And then, I think we have to grapple with the ethics of that.
DR. GAWANDE: Dr. Waisel, can I ask you the same question? If they say there has to
be some expert panel weighing this question, would you participate on the panel to come
up with a better execution method?
DR. WAISEL: I agree that it should be wholly permissible for other physicians to
participate, if they wish. I would have to think about it very carefully. A large part would
be depending on the intellectual freedom involved in the panel, the ability to write a
dissenting opinion from what the panel comes up with, and moving away from certain
constraints that are put around this that seem not to permit what I would consider to be
successful ways of nonphysician involvement.
DR. GAWANDE: Well, a decision from the Supreme Court in Baze v. Rees is expected
this spring. And whatever the decision is, it is bound to have important implications for
physicians and the entire health care community about our role in punishment. I want to
thank all three of you for taking the time to sort through these issues, their complexities
— Professor Deborah Denno, from Fordham University School of Law, Drs. Robert
Truog and David Waisel from Harvard Medical School. For the New England Journal of
Medicine, I am Atul Gawande.