Policy Forum
Sep 2006

Why Physicians Avoid Straight Talk about CPR

Leslie J. Blackhall, MD, MTS
Virtual Mentor. 2006;8(9):602-608. doi: 10.1001/virtualmentor.2006.8.9.pfor2-0609.


A resident observing the misuse of cardiopulmonary resuscitation (CPR) as practiced in the 1980s, Leslie J. Blackhall, MD, MTS, spoke out, writing “Must We Always Use CPR?” which was published in the New England Journal of Medicine in 1987. Still concerned about medical ethics, Dr. Blackhall teaches communication skills to a new generation of physicians and helps them to improve the quality of care at the end of life for patients and family alike. She sees less misuse of CPR today, but problems persist. “There has been little improvement in dealing with the idea of medical futility,” she says. “Doctors are more likely to tell patients that they would probably not survive CPR than that CPR would be medically futile.” She also points out troublesome inequities in care options for the rich and poor. Recently Virtual Mentor interviewed Dr. Blackhall and asked her to update readers on her concerns about the ethical consequences of how end-of-life care is delivered.

Q. How have things changed since you wrote your article?

A. At the time I wrote the article, use of CPR on people for whom it had very little hope of success was common practice. Today, doctors and other health professionals are far more willing, I think, to admit to themselves when this is the case. In my experience, very few oncologists, for example, want to do CPR on someone with advanced metastatic cancer, even if they still believe in continuing chemotherapy. And for the most part they’re willing to convey that decision to their patients, so their patients are not ending up in the intensive care unit at the end of their lives or receiving CPR when they’re dying of aggressive metastatic cancer. By contrast, when I wrote the article it was extremely common to have people with end-stage cancer or people with advanced Alzheimer’s disease receiving CPR, so I think things have improved.

On the other hand, there has been little improvement in dealing with the idea of medical futility. Doctors are more likely to tell patients that they are probably not going to survive CPR than that CPR would be medically futile. And there is still a subgroup of people who, even though told by doctors that they would be unlikely to survive CPR, continue to insist on it. We’re sort of at a standstill on what to do in those cases.

Q. What are those cases?

A. It’s not any particular type of case. But take, for example, someone whose diagnosis is widely metastatic cancer and whose need for resuscitation probably arises from that diagnosis rather than from any other cause. The issue here is not disagreement about the case itself—everyone agrees that the likelihood of this person surviving CPR and being discharged from the hospital is extremely low. But disagreement arises over whether we can ever say that there is zero chance someone will survive CPR. We can’t say that. Recently there was that patient who woke up after having been in a coma in a minimally conscious state for 19 years. I remind people that this is only one case. But because of one case here and there—and everyone wants to be that one guy—we can’t say there is zero chance. So that makes this a difficult issue.

And there’s a second reason that many of us are reluctant to “ration” CPR, if you will. And that reason has to do with the fact that people with personal financial resources can afford and demand treatment that is clinically futile, which makes physicians uncomfortable in denying that care to poorer patients—even when we believe it is futile. I heard a presentation at a bioethics conference by some people who are dealing with a medical futility law in Texas. Let’s say you have a patient in an intensive care unit with multiple organ systems failure who, all the physicians agree, is dying. The family refuses to allow withdrawal of therapy even though all the physicians believe that it is of no benefit. In such cases we usually call the ethics committee. Most ethics committees see it as their job to clarify the ethical issues, establish who the appropriate decision maker is and facilitate communication among all participants. If, despite this intervention, the patient and family continue to insist on care that the physicians feel is futile (however they define that term) there has been no recourse.

So in Texas they developed a legal procedure for such cases. After calling an ethics committee in clinical circumstances like I just described, you may call a prognosis committee that comprises physicians who are not currently involved with the case but have specializations in the area or areas of the patient’s disease. This committee decides whether or not this is a case of futile treatment, and, if they agree it is, the family has so many days to accept or reject the decision. If they accept the prognosis committee’s decision, care is withdrawn. If they don’t accept the decision, they have the right to transfer the patient to another institution. But here’s the problem: this situation often occurs among patients who don’t have good health insurance; they have Medicaid or they have very little money. So other hospitals don’t want to accept that patient in transfer.

We had a case at our institution very recently concerning a patient with metastatic lung cancer and a tracheoesophageal fistula. Nobody at our institution thought the patient was a candidate for surgery, but this patient was not poor; he had lots of money, and his daughter managed to get him airlifted to another institution. If that patient had had no insurance or minimal insurance, there’s no way the airlift would have happened or that another hospital would have accepted him. So the person who gets care withdrawn tends to be poorer. And this brings up uncomfortable questions of power relationships, rationing of health care and health care inequity.

Even though the care might have been futile, it makes a lot of people uneasy to let the doctor have the final say about that term “futility.” First, because we can never say any intervention has an absolute zero chance of working, and, second, because if the patient has resources, he or she may have the option to go to another hospital and get services that a less wealthy person cannot get.

Q. So there is a double standard?

A. When patients with resources can get services—even futile ones—that poor patients cannot, it raises that question. Everybody seems to agree about the right to refuse therapy. That’s the American way. But, we’re having a lot harder time dealing with the other side: the fact that one group of people who tend to have power dictate what futility means (in effect, they refuse to provide treatment) to the people who have less power.

When physicians use the argument with patients or the families that CPR is unlikely to be of help to them, most people can accept that. That’s why we do less futile CPR now. There will be a group that for one reason or another is willing to hold out for that .001 percent chance that they might have some meaningful life. So the question of how much money is too much to spend on the care, or how much time is too much or how many medical resources are too many—those questions remain unaddressed. It’s not primarily a question of when to use CPR.

Q. What would you propose if you were writing about this topic today?

A. As a palliative care physician, I think the problem is mainly one of communication. Often when I talk to people, I can help them make a reasonable choice about their resuscitation status because I’m straightforward about the likelihood that they will survive—not just survive CPR, but survive their illness. I think that part of this is about physicians’ reluctance to be forthright. I once had a resident tell me that I was the first physician he had ever heard use the word “dying.” The truth is that most patients and families have thought about death. But a lot of younger physicians and those in residency programs are hesitant to be straightforward about it.

A great study was done at Brigham and Women’s Hospital in Boston by CM Lilly asking whether physicians could decrease the amount of futile care that goes on in the ICUs [1,2]. There’s a high rate of mortality in the ICUs; people there are gravely ill. There comes a point when the physicians realize that we probably need to pull back, but the patient and the family generally take a little longer to come to that point. So these researchers did an intervention in which every patient had to have a family meeting within 72 hours of being admitted to the ICU to clarify goals of care. The meeting included the patient, if possible, the family, physicians, nurses and other members of the multi-disciplinary ICU team to make sure that everyone was on the same page. For example, it might have been agreed that the goal of care was to help this person with severe chronic obstructive pulmonary disease (COPD) recover from pneumonia and return to his or her prior health status. Participants from the initial meeting then had follow-up meetings every few days to ask, “Are we meeting these goals of care?” It’s far easier to have that discussion at first rather than just plowing ahead and then having that discussion a month down the line when everyone has just been going along without asking about the goals.

Similarly when I meet with a new patient in the Palliative Care Clinic of the cancer center at the University of Virginia, I ask them what they understand about the chemotherapy and their cancer. For patients who have metastatic cancer, I try to explain that there is almost no chance that the chemotherapy is going to cure them. The purpose of their therapy is either to prolong their life or shrink their tumor and improve their symptoms. Thinking about it this way allows us to ask, “How much is enough?” If the goal of chemotherapy is to improve your symptoms, and yet the chemotherapy itself is diminishing your quality of life, then maybe we don’t want to continue it. At the very least they can switch from thinking “I have cancer, I must get chemotherapy,” to “this is the purpose of the chemotherapy; what’s the likelihood it will provide that benefit?” Talking about goals of care and whether what we’re doing meets those goals is the type of communication people need to make decisions. If you have those conversations, most people make reasonable decisions. There will always be the family in which three of the five daughters say let mom go, but two say keep going because mom has always been a fighter. There will be a certain percentage of the time when family members will simply disagree, but studies show we can minimize that if we improve communication about the goals of care.

Q. Should physicians refrain from mentioning CPR or mention it and explain why it is not an option?

A. I come right out and say that we usually don’t do CPR on people with advanced cancer because it doesn’t work. This is how I put it. I say, look, if your cancer were to grow so much that it caused your heart and lungs to stop, doing CPR and putting you on a ventilator tends not to be beneficial because it doesn’t take the cancer away. If people are still getting chemotherapy, we can tell them we’ll do everything possible to keep them from getting to the point where the cancer causes their heart or lungs to stop; we can treat every infection. But if, despite all that, they get to that point, then in my experience CPR doesn’t help; the literature says it doesn’t work and we don’t recommend it. I always bring the topic up and talk with them about it. Everybody watches TV, and everybody knows what CPR is. And, as we all know, TV CPR is much more effective than real CPR, which gives people the wrong idea of what CPR is [3].

Q. Just to sum up, what would clinically appropriate CPR treatment look like?

A. For people who are experiencing chronic illnesses like cancer and Alzheimer’s for whom we can look ahead and see the CPR question coming, we must simply begin to discuss goals of care. Here is what you have. We don’t have a cure for this, but here’s what we can do to help it. I actually believe that many doctors think the outcomes are obvious, but for most patients they’re not obvious. Most doctors would know that widely metastatic pancreatic cancer is a terminal diagnosis, but not all patients do. That doesn’t mean that such patients shouldn’t be receiving Gemzar or other chemotherapy; it just means they have a right to know what they’re buying and what their doctor hopes it will do for them. So, frequent and early discussions of the goals of care and progress toward those goals is helpful. Then CPR becomes just another one of those discussions.

For people who’ve undergone acute devastating injury—for example, one moment they were sitting up talking and the next moment they’ve had a massive intracranial hemorrhage—those people need to have a decision-making process also. Two days ago “dad” was talking and planning to go on vacation and today he’s completely unresponsive in the neurosurgical intensive care unit. Family members need to have their questions answered. Once again, start from the beginning, discuss goals of care and the likelihood of reaching them. Update the family when there are changes. This way, you’re less likely to get into the situation where you or the nurses feel as though you are torturing the patient with futile care and the family feels angry on the other side.

In the small number of cases when I’m called to the intensive care unit because the family has talked with every specialist and still insists on doing everything, I ask them to tell me what they understand about what the doctors have said. If it seems they haven’t gotten the doctor’s message quite right, I try to restate what the doctors think, and I admit we are at an impasse. The doctors don’t want to provide a certain treatment anymore. You—family members—disagree and are mad about it. Let’s be mad at each other, but let’s be open about it. We don’t want to do it because we feel like we’re hurting your dad, and you want to do it because you feel like he has a chance for life. We’re having a disagreement about our values here. One of our values is we don’t like to hurt people by continuing medical care that doesn’t help them. Tell me why you want us to keep doing these things that we believe are doing more harm than good. That’s the best we can do in these difficult circumstances; be straightforward with each other and continue the conversation.

Q. How can medical students learn these communication skills?

A. At University of Southern California and now at University of Virginia we have used standardized patients. In its most extensive form at University of Southern California, we had a little grant money to try this, so we had about five students at a time with five standardized patients (actors and actresses) all playing the same part and videotaped in different rooms with a very realistic-looking set-up. After the students interviewed the standardized patients, they watched their videotapes. And then we talked about what went wrong and what went right. This is time-consuming, but I think it works.

Q. How did the CPR scenarios play out?

A. One of the medical students was talking about intubation with a “patient” who had ALS. We looked at the tape and the medical student was sitting there holding his head in his hands. He was just talking around the subject and could not be straightforward about it, and therefore nothing got decided. When it was over the medical student said that a friend of his had had cancer, and the doctors told him he was dying, which was a terrible thing for his friend to hear. The medical student promised himself he would never tell a patient that he was dying. But he realized as a result of watching the videotape that he couldn’t keep that promise; that there were times he would have to tell people that they were dying. And in the next patient encounter you could see that, although it was difficult for him, he was able to get straight to the point and talk to the patient in a way that the patient could understand. That was an instance where you could see somebody acquiring an incredibly difficult skill set. The training has improved since I was a medical student, but we could do more. I think it’s almost abusive to put young doctors into positions where they have to talk with the family of a dying patient when they are unprepared to do so.

Q. Dr. Weismann, who wrote for Virtual Mentor, came to the conclusion that nobody really wants to talk about death.

A. Nobody wants to face it. I give a “breaking bad news” presentation. When I give that talk, the first slide says “People want to know the truth.” The second slide says “They want the truth to be good news.” They want both of those things, which they can’t always have. The same is true for doctors—you want to tell the truth—that’s part of your ethos—but you want to tell good news. There tends to be collusion—there’s a great paper about this actually [4]—between doctors and patients with a critical illness to just sort of “not deal with this now” because any encounter in which you do deal with it is going to be anxiety-provoking on both sides, and human beings are animals that try to avoid unpleasant emotion. I feel like it’s my job to be the one who says, “I’m not avoiding unpleasant emotions.”


  1. Lilly CM, De Meo DL, Sonna LA, et al. An intensive communication intervention for the critically ill. Am J Med. 2000;109(6):469-475.
  2. Lilly CM, Sonna LA, Haley KJ, Massaro AF. Intensive communication: four-year follow up from a clinical practice study. Crit Care Med. 2003;31(5 Suppl):S394-S399.
  3. Diem SJ, Lantos JD, Tulsky JA. Cardiopulmonary resuscitation on television. Miracles and misinformation. N Engl J Med. 1996;334(24):1578-1582.
  4. The AM, Hak T, Koeter G, van Der Wal G. Collusion in doctor-patient communication about imminent death: an ethnographic study. BMJ. 2000;321(7273):1376-1381.

Editor's Note

Dr. Blackhall interviewed by Philip A. Perry, MSJ.


Virtual Mentor. 2006;8(9):602-608.



The viewpoints expressed in this article are those of the author(s) and do not necessarily reflect the views and policies of the AMA.