Hard Choices: Who really has the last word on a DNR?

Published on: 

A brother and sister clash over their mother's final wish. Our panel of physicians and ethicists parses the conflict.

A brother and sister clash over their mother's final wish.Our panel of physicians and ethicists parses the conflict.

This is the second of three articles* exploring ethical dilemmas fortoday's physicians. Some are issues that have been faced for years; othersarise from the era and aura of managed care. And still others--like theone presented here--are spawned by medical advances that prolong life. Thearticles are based on roundtable discussions organized by Senior EditorAnne L. Finger.

Willard Gaylin, the panel moderator, is a clinical professor ofpsychiatry at Columbia University College of Physicians and Surgeons, NewYork City and co-founder of The Hastings Center, a bioethics institution.

The other participants:

Jordan S. Busch, MD, an internist, medical director of Beth IsraelHealth Care, in Brookline, MA, and an instructor in medicine at HarvardMedical School.

Daniel Callahan, PhD, co-founder of The Hastings Center, and currentlydirector of its international programs and senior associate for health policy.He is also a visiting scholar at Harvard Medical School.

Harvey R. Gross, MD, chief, Department of Family Practice, Englewood(NJ) Hospital and Medical Center, and assistant clinical professor of medicine,Mount Sinai School of Medicine, New York City.

BonnieSteinbock, PhD, professor and chair, Department of Philosophy,State University of New York at Albany, with joint appointments in the departmentof Public Administration and Policy and Health Policy, Management, and Behavior.

*The first, "Would a cost-conscious physician order this MRI?"appeared in the Aug. 9, 1999 issue.

The case: Among internist Philip Weber's nursing home patientsis 87-year-old Catherine Mallon, who has Alzheimer's disease, diabetes,and chronic renal failure. Mallon has an advance directive stating thatshe doesn't want resuscitation. DNR orders have been completed by her daughter,Patricia, who serves as Mallon's health care proxy. The DNR appears on thepatient's chart.

When Mallon is found unresponsive in her bed one morning, the nursingsupervisor calls Weber and 911. EMS technicians begin CPR, and a paramedicintubates her. Weber arrives and shows the advance directive to the EMSstaff, but they refuse to stop the resuscitation, maintaining that theydon't know whether the document is valid. They tell Weber that if he insiststhey stop, they'll call the police to escort the patient to the nearesthospital. He does insist, and Mallon is transferred to a hospital and placedon a ventilator.

Family members are notified, and they converge on the hospital. Patriciais horrified that her mother has been intubated. "Everyone knew whatmy mother wanted, and they knew how to reach me!" she cries. "Whywasn't I called when they found Mom?" She insists that Mallon be removedfrom the ventilator.

But Paul, Patricia's younger brother, who has flown in from his Chicagohome, disagrees. "Mom never said anything to me about not wanting tobe resuscitated," he tells his sister. "You're just tired of takingcare of her."

"Mom never told you about anything important," his sister responds."She knew I was the one she could depend on."

"Well, she sure can't depend on you now," Paul snaps back,"or on this doctor who's supposed to help her." Paul threatensto sue the nursing home and Weber if his mother is allowed to die.

Willard Gaylin: Let's talk first about the issues raised in thebeginning of this scenario. Dr. Gross and Dr. Busch, would an EMS groupoverride a physician in a resuscitation process?

Harvey R. Gross: Yes, this happens frequently. The job of theEMS people is to resuscitate and transport the patient to the hospital.They really can't make a determination of whether or not the patient isdead. They also can't always make a determination of whether the DNR orderis valid. In my state, New Jersey, we're working on a uniform DNR documentthat everybody will recognize as valid.

Gaylin: Dr. Busch, do you concur with this observation?

Jordan S. Busch: Where I practice, in Massachusetts, we have auniversal form that would be accepted. Could this situation happen if youhad some other documentation instead? I'm not sure. If there's a physicianon site, I believe he can take over and make such decisions.

More typically, by the time somebody says, "I think the patienthas a DNR order," the EMS people have already started to intubate her.The patient ends up in the emergency room, and somebody says, "I havedocumentation that this patient was supposed to be DNR." Now what dowe do?

It's a lot easier to make a decision when the patient is first encountered.To intubate and then extubate her is an active process; that's more difficultthan saying the decision was made to let nature take its course.

Gross: Here's another complicating factor. You have a valid DNRin the nursing home, and a valid one in the hospital. But what happens inthe ambulance? If the patient starts to code, do you resuscitate her there?

This goes back to the physician's relationship with the nursing homeand the family. Many nursing homes have a "do not transfer" or"do not hospitalize" order. Nurses know that the patient is tobe allowed to die in dignity.

Busch: Doctors, nursing homes, patients, families--all are increasinglyaware that end-of-life decisions need to be made proactively. They're tryingto make and document the right decisions. But I don't think we'll ever havea system that can eliminate all mistakes. Sometimes, in the heat of themoment, there's a degree of uncertainty, and heroic measures work only ifyou do them then. Sometimes people don't feel they have adequate information,and they have to revert to thinking, "My job is to perform CPR on someonewho needs it."

Gaylin: We have to assume that Philip Weber, our prototypic doctorin the Mallon case, is as knowledgeable as you are about what goes on ina nursing home, and that he's totally dependent on the person who's in charge,whatever that individual's biases are.

Is there anything the doctor might have done? Would ordering a slow code--goingthrough the motions of resuscitation while letting the patient die--be justifiable?

Busch: This is an issue that ethicists have helped me understand.In my mind, we have DNR orders because we want to respect the wishes ofa patient who feels that the risk*or the pain and suffering*would outweighany benefit. With the slow code, you prolong the pain and suffering fromthe procedure, and if you're successful, you increase the risk of long-termdamage to the patient. I can find no ethical basis for ever performing aslow code.

Gross: DNRs and advance directives are very clear: Either we'regoing to resuscitate somebody or we are not. If we communicate with thefamily about what we're going to do in this situation, and we know how we'regoing to respond, and the nursing home knows how it's going to respond,then there really isn't any reason for a slow code.

Busch: It comes up consistently in two situations. In one, nodiscussion of the patient's wishes has occurred. If the doctor believesthere's no reasonable hope, or the quality of life would be so bad thatresuscitation is futile, he may be thinking along the lines of a slow code.

In the other situation, the doctor and the family have had the discussion,and the doctor firmly believes that CPR and heroic measures are not indicated*forexample, the patient has end-stage cancer and is gasping for her last breath.The family says, "Don't let Mom die," but the doctor feels keepingthe patient alive is inflicting pain and suffering. So he thinks, "MaybeI'll just do this slowly."

Gaylin: Let me ask our two ethicists: Do you also agree with thephysicians at this table that a slow code is never justified? And pleaseexplain the moral differences between the slow code and the "DNR notindicated."

Daniel Callahan: The idea of slow codes developed some time inthe 1960s or '70s. The bias then was always "When in doubt, treat*veryaggressively." A lot of physicians had a dilemma with that, and theslow code was a way to squirm out of it.

Now the laws and regulations make the slow code unnecessary.

There's another interesting debate among jurists: The person at the primeof life decides, "If I ever become demented, let me go." Thenyou see the person in this state, and he is obviously happy living. Is thereal self the one that thought about everything? And is that a more importantself than the demented self?

Bonnie Steinbock: It's been referred to as the "then self"and the "now self." Why does the past, intellectual self get tomake decisions for the contented, demented self? That's not an easy question.

Gaylin: The ethicists are cautioning you physicians not to takeDNRs or advance directives as a way off the hook. When physicians gentlyand benignly did not expose all the truth to patients, they were calledliars. Physicians responded by what I call truth dumping: They told patientseverything, which in my mind was treating the patients with just as littleconsideration.

One nice thing about these theoreticals is that we can change the case.We can now make Mrs. Mallon's condition not quite so severe. Let's say she's82 and has mild senile dementia*not Alzheimer's*diabetes, and chronic renalproblems. She's in discomfort, and she has ulcers on her legs. Before thenursing home, she had lived with her daughter, but now there's an emergency,and we can't reach the daughter.

Mrs. Mallon signed the DNR when she was 52*and hasn't looked at it since.Does a directive signed 30 years earlier have the same weight as a morerecent one? Are all directives directive--or are they simply excuses torelieve doctors from having to make tough moral choices? What if the doctorreally disagrees with the decisions of the directive and of the family?

Busch: I think advance directives need to be an ongoing process.The mere fact that somebody at one time specified a DNR*yes, that gets doctorsoff the hook. They now don't have to do things. And they really don't haveto think about things, either.


But how did the directive come about? Suppose a physician said: "Whenyou're having the 'big one,' and you can't breathe, do you want a bunchof doctors to run in, pump on your chest, break all your ribs, stick a tubedown your throat, and have a machine breathe for you, which undoubtedlywill hurt with all those broken ribs? Is that what you want your last hoursto be like?"

The patient may say, "No, I don't want those things done to me."But the doctor hasn't asked the right question, which is: "If we thoughtthat you could recover from some event, but you might not be able to takecare of your own activities of daily living, how would you feel about that?"To this, a patient may say, "Well, if there was a reasonable chancethat I could recover, I'd like you to try. But if it looks like I'll havepermanent brain damage, never mind."

In most cases, it's not a question that can be answered Yes or No. Itdepends on the circumstances and prognoses for the acute problem and anychronic processes. Yet the DNR is black or white; there's an order or thereis not. Legally, doctors should document the conversations they've had andnote what that DNR really means.

Gaylin: Dr. Busch describes how the procedures might bepresented to the patient years before the fact. But we don't know then whatthe situation will be. It might simply require an intubation; it might needan invasive procedure. You are allowing the doctor enormous entry for hisbias into the treatment of the patient.

A concept called "authenticity" applies here. "Autonomy"just means that Mr. Jones signed the advance directive and it was witnessed.But "authenticity" means: "Does this sound like Mr. Jones?Is it consonant with the rest of his life?" Can't we ask those questions?

I don't think it would be terribly difficult to institute proceduresso that when someone is admitted to a nursing home, you go over the advancedirective with him: "Here is what you wrote in 1952, and here is whatyou wrote in 1972."

Callahan: I agree, and probably you'd want to do it again aftersomebody has been in a nursing home for three or five years. But none ofthis works very well if there hasn't been some ongoing discussion betweenthe family and the physician.

Busch: I might be talking with a patient who has a malignancyabout the possibility that this disease will overwhelm him at some point.I might ask, "Would you want us to do heroic measures?" The patientmight say, "Well, Doc, by that time I suspect you'll have done everythingto try to combat this disease, and if I'm losing the battle, just let mego naturally. Don't put me through this type of end-of-life torture."

Then the patient has a heart attack, unrelated to what he was thinkingabout in that discussion. So we need to talk again. Every time I admit apatient to the hospital, or I think the health status has changed in somesignificant way, I remind the patient about the conversation. It needs tobe an ongoing discussion.

Gaylin: Why is such a distinction made between intubation andreversing the intubation? Life is dependent on the tube.

Gross: It's easier to think about intubating than extubating.We're trained to act to save people's lives, and intubation is one meansto do that.

Gaylin: Is there a difference between not intubating and extubating?

Busch: Clinically, I don't draw the distinction. But I believethat my patients' families do, so if they're encouraging me to intubate,I sometimes need to help them understand that we are about to do thingsthat the person they feel responsible for wouldn't want. Families sometimesfind it easier not to intubate, and let nature take its course, than toextubate, which feels like a more active process.

Personally, I think about the goal of therapy: what we are trying todo, and whether the best treatment may be to do nothing--or to stop somethingthat we are doing. We don't distinguish between intubation and extubation.

Gaylin: So you see it as an unsophisticated and morally irrelevantdifference? Then you would try and educate. Dan, do you agree?

Callahan: There's an important, pragmatic reason for saying thatit doesn't make an ethical difference. There are times when you ought toput someone on a respirator, and if you're worrying that you might not beable to take him off somewhere down the line, you may not do the right thingto begin with.

Gaylin: Do I understand from this group of four that an old argument--whichused to have people at each other's throats--has disappeared? Once you determinethat the quality of a person's life isn't worth continuing, there's no realmoral distinction between withholding dialysis and stopping it.

Busch: The only problem with the way you've described it is: Whois the "you"? Is it the doctor, the patient, or the health careproxy? I think you have to rely on the patient and the proxy. Your job asthe physician is to try to work with them to understand the benefits oftreatment and the patient's goals and objectives for the therapy.

Gaylin: Let's turn now to the second part of this dilemma. Whatdo we do when we have conflicting reports from the patient's children?

Gross: Patricia, the daughter in the case we've been discussing,has the health care proxy. There seems to be a legal DNR. Everything seemsto be in order to make that decision at this point.

Steinbock: Mrs. Mallon lived with the daughter. Paul, theson, just came in from Chicago.

Gross: It's more upsetting if he shows up with the healthcare proxy, and the daughter has been living with the mother and doesn'thave the proxy. That happens all the time.

Busch: Another problem occurs when the family member or friendwho has the health care proxy says, "You know, now that I see Dad--demented--watching'Jeopardy,' he looks pretty happy. I'd bet if he could have seen himselflike this. . . ." And they want to change the directive. In my mind,the proxy gets to make the decision; that's the responsibility the patienthas given to that person.

The role of the physician is to try to work with the health care proxyand their knowledge of the patient's preferences to say, "This is notreally what was supposed to happen, so let's think this through."

Gaylin: Let's change the case. No one has a proxy, and there'sno advance directive. The mother has been living with the daughter. Thedaughter has gotten progressively irritable; it's difficult to take careof a dying person. But she loves her mother. Paul only hears the cheeryvoice on the phone, but he is a good son. He comes in twice a year to visithis mother.

Gross: The question becomes "What would Mother have wantedwhen she was well?" I was presented with that situation when I wentinto practice: One daughter threatened to sue me if I removed her motherfrom the respirator; another threatened to sue if I didn't. At that point,I acted to save life--not remove life.

Steinbock: The importance in the case as you presented it, Will,is not so much a matter of who takes care of Mom and changes her diapersand cleans up her vomit, but which of her children you believe knows herbetter.

If the daughter says, "In the 10 years I've been living with Momand taking care of her, we've discussed this and she has told me she doesn'twant to live the way Mrs. O'Grady lived in her last years"--that'simportant.

Gaylin: Suppose there haven't been such clear-cut discussions*asthere usually haven't. Should the weight of 10 or 20 years of giving carecount in these decisions?

Steinbock: Absolutely. What we know about a person--whether it comesfrom doctor-patient communication or from the family--often doesn't comefrom talking about these kinds of things explicitly. It comes from whatyou know about that person's values and what kind of life she wants to lead.

Busch: Sometimes, when you talk about what the quality of lifehas been, and what it's likely to be after all this illness*typically, itwon't be as good*you realize that you're dealing with the old conflictsbetween siblings. At our medical center, that's often when the ethics committeecomes in. They'll meet with the entire team and the family and try to talkthrough the issues and how we can proceed in a way that feels comfortablefor everybody.

Gaylin: Would Mrs. Mallon's advance directive or the daughter'shealth care proxy tip the scale?

Gross: It really might, and if I disagreed with brother Paul,I would go to the ethics committee.

This did happen: I was working with three wonderful, supportive childrenwhose mother had terminal cancer. All of a sudden, a son showed up. He didnot have a good relationship with any of the family, but he had the proxy.It took a week or two--working with him and the ethics committee--to gethim to accept that we were going to have a DNR on his mother.

Busch: The difficulty occurs when you sense that everybody seemsto know the dying patient's values except the person who has the healthcare proxy, but the law favors that person.

Gaylin: Let me sum up today's consensus: In a delicately balancedcase, the proxy becomes a convenient way to solve the problem. On the otherhand, if Mrs. Mallon's son had held the proxy, but the daughter had spent20 years tending their mother and sharing thoughts with her, the moral choicemight outweigh the legal one.

That leaves usas most moral discussions dowith a group of entirely newquestions.

What do you think?

How would you have responded to Weber's dilemma? Have you developed anyguidelines for end-of-life care that help prevent such problems? What didyou think was the most valuable insight expressed in this article? Do youhave any suggestions that weren't mentioned here? Please fax us at 201-722-2688,ATT: ALF, or e-mail to,and let us know.

Who really has the last word on a DNR?. Medical Economics 1999;19:116.