Waisel and Truog's article [1] misrepresents the Seattle Veterans Affairs (VA) Medical Center policy on do-not-resuscitate orders in two important respects. First, the VA policy acknowledges that values enter into determinations of quantitative and qualitative futility. Rather than allowing the physician's values to determine judgments of medical futility, the policy promotes shared decision making. It requires that physicians discuss their judgments of medical futility with the patient or surrogate and obtain agreement before writing a do-not-resuscitate order. In the absence of agreement, the policy outlines a conflict resolution process. As an academic teaching facility, the VA Medical Center uses this policy to teach physician trainees about the role of values in medical decision making and about the importance of open communication with patients and families.

Second, the Seattle VA policy recognizes the uncertainty inherent in judging medical futility. In fact, uncertainty is inherent in most medical decisions because physicians inductively use clinical evidence, aggregated data, and probabilities to make medical decisions. Again, the VA policy teaches physician trainees to acknowledge uncertainty and to draw on the best available evidence to make medical futility and other ethical judgments in medicine.

Finally, the VA policy attempts to fairly and consistently address cases in which patients or families ask for treatment that probably will not succeed or will result in a life that seems devoid of human value. These are the "futility cases" that the Seattle VA Medical Center most commonly encounters. Although Waisel and Truog advocate the idea of "physiologic futility," such an approach focuses attention on the least contentious element of medical futility. In our experience, most futility cases do not center on determining the effects of treatments but rather on determining whether treatment effects represent patient benefits.