Deciding when patients can't

End-of-life considerations for incapacitated patients without surrogates.

It can be a physician's worst-case scenario: When a dying patient with no family becomes incapacitated without designating a surrogate, who decides about care, and how?

Not infrequently, physicians are faced with being the chief decision maker, and the available guidelines on how they should proceed can be contradictory, overly stringent or nonexistent.

Physicians and institutions must balance the need for rapid decision making with fairness multidisciplinary input and due process says Dr White Photo by Comstock Complete
Physicians and institutions must balance the need for rapid decision making with fairness, multidisciplinary input and due process, says Dr. White. Photo by Comstock Complete.

Douglas B. White, MD, MAS, assistant professor of pulmonary and critical care medicine and bioethics at the University of California, San Francisco, recently led a study to determine how often this situation arises and how it's handled. In the multicenter, prospective trial, which was published in the July 3 Annals of Internal Medicine, Dr. White and his colleagues used questionnaires to gather information about the care of 3,011 critically ill adults in the ICUs of seven U.S. medical centers. According to their survey results, one in 20 deaths in these ICUs occurred in incapacitated patients without surrogates, and physicians made most of the decisions about life support in these cases without formal review.

Dr. White recently spoke to ACP Hospitalist about his study and offered suggestions on improving the decision-making process affecting this vulnerable population.

Q: Were your results what you expected, or was it a surprise that decisions about life support in patients without surrogates happen fairly often in the ICU?

A: It was surprising how often these situations are confronting physicians in ICUs, but it was also somewhat surprising that there's so little oversight of these decisions.

Q: Your study was done in six states on the East and West Coasts. Do you think your findings are typical throughout the U.S.?

A: We think the findings are generalizable because the laws that govern the states were quite heterogeneous. Some states allowed physicians to make the decision, and others said that physicians couldn't withdraw life support unless there was a very high level of evidence that the patient would want that. The heterogeneity of the laws combined with the relative homogeneity of the approaches to decision making suggests that the findings are generalizable, i.e., that the laws don't really seem to affect how these decisions are made.

Q: For end-of-life decisions in incapacitated patients without surrogates, the American Medical Association recommends ethics committee or judicial review, ACP recommends judicial review and the American Geriatrics Society recommends peer review. What level of review do you think is most appropriate?

A: I don't think there's one right answer, because each has its pros and cons. I do think this is something that can be guided by evidence. Pilot studies of each of these approaches could be used to flesh out the relative merits and weaknesses of each approach in practice. Until that happens, it would be a reasonable compromise to have the hospital ethics committee involved in the decision making. That will give an element of due process. The downside is that the training of hospital ethics committee members varies, and if the membership is not well versed in these issues, they may not be able to substantively improve the process. There's also a potential concern about conflict of interest, because hospital ethics committees are committees for the hospital, so that relationship should be known from the outset. It's important to note that conflicts of interest can exist for other methods too. As we think through the best approach, these are the issues that have to be balanced.

Q: Can stringent review help maintain patients' rights more than other methods?

A: Imagine that you have a patient who never wanted to be on life support, but you don't know this because the patient's unable to speak for himself and doesn't have a surrogate. Then the judicial review process takes three months and the patient is on life support, which is continued because in the absence of clear evidence, the courts generally err on the side of preserving life. In that setting, the argument that due process is protecting the patient's rights is probably not true. That's the hypothetical concern. It's been well documented in the last 20 years that most patients do not want to die receiving a lot of highly technological, sometimes burdensome care. On balance, when physicians are thinking that it may be reasonable to withdraw life support and focus on comfort care, most of the time their attitudes are probably in accordance with the patient's. This is tricky, though, because several studies suggest that physicians in general desire less life support than their patients do.

Q: Your study mentions that critical care societies don't offer guidelines on this topic. Do you think that having such guidelines would help?

A: If the guidelines were able to respond to the difficulties that physicians face and develop an approach that's easily implemented, then I do think a guideline from a critical care society would be valuable. But the existing guidelines from the prominent medical societies are generally not followed, so it may be a matter of physicians' attitudes that will not be affected by guidelines. Physicians may simply believe that these mechanisms are not necessary or helpful. If that's the case, then broader input into the development of other mechanisms that do meet physicians' concerns while still achieving the goal of some due process for this decision would be really valuable.

Q: How much do living wills or advance directives help in this situation?

A: For our study, patients who had advance directives were not eligible. But I think they would be valuable in the same way that they're valuable for any patient, which is to provide general information about whether the patient would accept some limitation of life support in the face of severe illness. My experience in intensive care units, though, is that there's always some process of deliberation with the family to figure out whether the current situation matches what the advance directive covers, whether the values expressed in the advance directive are contemporary and when the patient's perceived desire should be implemented if there's some uncertainty about the outcome. The boilerplate language of “If I'm in a coma or some other irreversible state then let me die” generally doesn't apply because things aren't black and white in ICUs. Advance directives are simply a starting point, and when there's no family, you're left with this somewhat vague document that may or may not be very helpful.

Q: Besides conducting studies of each review method, is there any other research that you'd recommend?

A: Interviews with physicians would be valuable, really talking to physicians in intensive care units about their attitudes and what barriers they perceive to involving other people in the decision-making process. It would also be helpful to conduct semi-structured interviews with the population that makes up these patients without surrogates in ICUs. If broad consensus emerged that this population doesn't want physicians to make the decisions or conversely that they really don't want the courts involved, that could guide policy making. This is a very socially marginalized population, and some of them have come in contact with the legal system in their past, through homelessness or crime. They may view judicial review or guardianship as an intrusion on their liberty. That would certainly shape policy if the mechanism that we're putting in place to protect them is one that they fear.

Q: What advice would you give to physicians in dealing with this situation?

A: First, we as physicians have to be careful not to put our own beliefs about what is appropriate life support onto our patients. We need to know that patients generally desire somewhat more life support than physicians do when faced with similar circumstances. Second, in practical and procedural terms, physicians need to know the professional society recommendations and their state laws. They also need to understand that in general, most people advocate for some sort of due process mechanism that takes some of the burden off the physician and makes sure decisions are made in a fair manner. Physicians and institutions should cooperate to develop some institutional wisdom about what procedures work best. A reasonable place to start is to recognize the limitations of the environment and to start to design procedures that balance the need for relatively rapid decision making with fairness, with multidisciplinary input and with due process.