Clinical Decisions

Care of an Unresponsive Patient with a Poor Prognosis — Polling Results

Patricia A. Kritek, M.D., Ed.M., Arthur S. Slutsky, M.D., and Leonard D. Hudson, M.D.

N Engl J Med 2009; 360:e15March 5, 2009

Article

Interactive Map

Clinical Decisions Polling Results by Country.

In late January, we presented the case of an unresponsive, 56-year-old homeless patient who had a ruptured aneurysm, a high probability of cancer, and a best prognosis of severe disability that would leave him dependent on care by others in Clinical Decisions,1 an interactive feature designed to assess how readers would manage a clinical problem for which there may be more than one appropriate solution. There was evidence that the patient would not want aggressive medical care. The patient's mother and brother agreed with the care team to shift to comfort care, but the patient's son, the legal next of kin, wanted full aggressive-care measures to be taken, even though he had initially been in agreement with the mother and brother.

A total of 6332 votes and 436 comments were received; 6120 votes could be attributed to a continent or region (Figure 1Figure 1Percentage of Participants Choosing Each Option for the Care of an Unresponsive Patient with a Poor Prognosis.). The response was notably international, with readers from more than 120 countries casting votes. About half the respondents were from the United States, with the next largest contributions coming from readers in the United Kingdom and Canada.

The votes were not evenly divided among the three care options. The first option, to continue aggressive care and pursue an ethics consultation with the patient's surrogate, and the second option, to write a do-not-resuscitate order and transfer the patient to a skilled-nursing facility, each received about 25% of the votes, whereas the third option, to withdraw life support on the basis of substituted judgment, received the most votes — 49%. Although this was certainly not a scientific survey, we found marked geographic variation among these responses, as can be seen on the interactive map, available at NEJM.org. The option to withdraw life support was the least commonly chosen by respondents from Asia (24%) and South America (34%) and the most commonly chosen in Australia and Oceania (68%) as well as North America, Europe, and Africa (53% of the respondents on those continents combined).

Many of the comments associated with the second option, a do-not-resuscitate order and transfer to a nursing facility, reflected what the respondents considered to be local legal limitations regarding the withdrawal of aggressive support. Similarly, a majority of the comments by those who opted for aggressive care and an ethics consultation expressed a sense of commitment to the legal health care proxy, the son. There was unwillingness to go against the son's wishes without trying harder to achieve consensus. The comments associated with choosing to withdraw life support were more varied. They included a sense that the patient's wishes (not to receive life support) should be honored, that other options would prolong dying and suffering, that continuing supportive care was a poor use of resources, and that the physician ultimately had the responsibility for making the choice.

Comments related to all three choices reflected a desire to better understand the son's perspective, particularly in light of his changed opinion. Of note, several comments reflected dissatisfaction with any one choice. Many of these readers advocated for the involvement of palliative or hospice care. Others felt that an ethics consultation would be appropriate as part of either a do-not-resuscitate order and transfer or a withdrawal of life support, with an emphasis on gaining consensus before moving forward.

What practitioners say they do in surveys can differ dramatically from what they are observed to do in practice. We are left to wonder whether caregivers favoring the withdrawal of life support (especially in North America) would indeed carry out this option in practice. Many respondents who selected one of the other two options expressed the desire to withdraw life support but believed they were limited, by legal concerns, in making this choice.

References

References

1. 1

   Slutsky AS, Hudson LD, Dubler NN, Weijer C, Tonelli MR. Care of an unresponsive patient with a poor prognosis. N Engl J Med 2009;360:527-531
   Full Text | Web of Science | Medline