Abstract:  The debate over futility is not new nor is it settled.  There is a growing consensus that stakeholders should focus less on defining the concept and more on establishing a process-based approach to clinical decisions that involve futility determinations.  Notwithstanding this consensus, two questions remain.  One question pertains to whether unilateral decisions should still be made despite a due process; the second question relates to the ethical aspects of a process-based approach.  The former question has been the subject of much of the debate to date; the latter question is the subject of this essay.  More specifically, what are the substantive ethical aspects of a process-based approach to clinical decisions involving a futility determination?  Ethical aspects of a process-based approach to decision-making based on futility judgments involve two major substantive issues: futility and authority.  By clarifying futility correlative to specific goals as well as authority and its corresponding powers, an ethical process-based approach will anchor decision-making around specific goals and will scope each stakeholder’s responsibility in the process.

Keywords:      futility—medical futility, authority, power, decision-making

Introduction

The debate over futility is not new nor is it settled.  There is no consensus on what the term should mean.  However, there is a growing consensus that stakeholders should focus less on defining the concept and more on establishing a process-based approach to clinical decisions that involve futility determinations.  For example, the American Medical Association (AMA) guidelines on medical futility and the Texas Advance Directives Act reflect this emerging consensus toward a process-based mechanism.[1]  Notwithstanding this consensus, two questions remain.  One question pertains to whether unilateral decisions should still be made despite a due process; the second question relates to the ethical aspects of a process-based approach.  The former question has been the subject of much of the debate to date;[2] the latter question is the subject of this essay.  More specifically, what are the ethical aspects of a process-based approach to clinical decisions involving a futility determination?[3]

Ethical aspects of a process-based approach to decision-making based on futility judgments involve two major substantive issues: futility and authority.  By clarifying futility correlative to specific goals as well as authority and its corresponding powers, an ethical process-based approach will anchor decision-making around specific goals and will scope each stakeholder’s responsibility in the process.  In exploring these substantive issues, I will pursue the following objectives: to analyze four descriptions of a process-based approach, to explore the meaning and function of various concepts of futility, to identify the authorities and powers operative in clinical decision-making (regarding futile treatment decisions); and to conclude with a description of a process-based approach to clinical decision-making involving futility determinations.

Process-based Approaches

As mentioned, the consensus has shifted from trying to define futility to elaborating a process-based approach to decisions involving futility judgments.  The purpose of this section is to describe six features of such processes and to suggest opportunities where ethical reflection could benefit such processes.[4]  To begin, each process is inclusive in its communication, which aims at appropriate participation in the decision-making process.  Two, each process acknowledges a lack of a universal definition of futility, yet a state or an institution can implement a socially conditioned, operational definition of futility (e.g., for the purposes of the policy).  States and institutions recognize social consensus and community standards as ways to legitimately address the ambiguity in defining futility.  Three, each process suggests that the role of ethics committees is to provide an opportunity for added reflection and robust participation beyond the bedside.[5]  Four, each process invokes time parameters that catalyze decision-making and implementation so the issues do not linger indefinitely.  Five, each institutional process appropriates legal parameters in state law governing end-of-life decision-making.  The Texas law gives physicians legal immunity for both civil and criminal charges.[6]    Finally, commentators note that health care decisions based on futility judgments are compatible, in principle, with the Catholic tradition.[7]  Here, commentators connect the ordinary-extraordinary distinction (with correlative obligations of patients and physicians) to end-of-life decisions involving futility judgments.

Reflection on these features reveals several opportunities for ethical reflection.  First, though some policies may define or relativize futility in specific circumstances, the policies seem to link futility inadequately to the specific goals of each case.  According to the descriptions, providers are to give reasons justifying the judgment of futility (often such reasons are required to be in writing).  Are these reasons meaningfully connected with the goals of the medical treatment(s) or the overall care plan?  Have those goals been sufficiently discussed throughout the course of the patient’s care?  In addition, the discussions on futility may conflate the fact-value distinction,[8] are potentially vague regarding goals (and might fail to distinguish between effects and benefits), and are (therefore) likely ambiguous regarding futility.  Thus, these discussions need anchoring points in order to be practical and meaningful.    Second, arbitration and negotiation[9] models are not likely to be fruitful when they confuse authority and power.  On the one hand, the processes reflect attempts to negotiate the meaning of futility (and therefore the decision to forego or withdraw treatment).  On the other hand, they also express a last resort to arbitration (institutional or judicial) to settle the dispute.  Here, the authority that physicians and patients purport to negotiate is authority as control (i.e., force of will or expertise).  This obscures two sets of ideas: (a) the different kinds of authority that might be operative in the physician-patient relationship and (b) that authority is power.[10]  In this way, the bargaining zones in process-based futility decisions are likely to be negative bargaining zones: neither party will find sufficient or acceptable terms.  One needs to distinguish the different domains of power between physicians (the power to heal) and patients (the power to self-actualize); this difference is significant for how one understands the power to heal with biomedical technology (instruments of that power).  Thus, futility cases are not disputes over authority per se, but they are disputes regarding the power to control biomedical technology without appropriate authority over it.  Likewise, the ascription of healing power to patient’s self-actualization (i.e., through absolutizing patient autonomy) medicalizes healing beyond the physician’s authority.  This reflects a failure to heal the person integrally and adequately considered.[11]  Processes that fail to reach mutual decisions based on futility reflect bilateral failures to engage effectively in healing the whole person in his or her totality.  This means that while physicians’ authority seeks the medical good and the ultimate telos of medicine (i.e., the good of the patient), the pursuit of the medical good has to be conditioned by the other goods authoritatively known and pursued by the patient (or his or her surrogate decision-maker).[12]

Futility

The first substantive element to a process-based approach is the concept of futility itself.  The intent of this section is to describe alternative conceptions in a way that shapes the conditions of a due process to futility-based decisions.  That is, specifying different kinds of futility will provide anchors to the discussion in the decision-making process.

To begin, any approach to futility-based decisions needs to recognize the formal conditions of futility.  While this is the most general and often the most easily agreed upon, current practice often does not integrate these formal conditions.[13]  Slosar writes:

…the concept of futility is relevant when three formal conditions are met: 1) there is a goal; 2) there is an action aimed at achieving the goal; and, 3) there is virtual certainty that the action will fail in attaining the goal in question.[14]

In this sense, futility is fundamentally a teleological concept wherein one is never able to reach the telos.  One can understand the teleology of futility in two different ways: empirically and morally.[15]  One concept of futility recognizes an agent’s intent in the goal whereas the other concept recognizes that an end may be independent of an intended goal.  Commentators [sometimes implicitly] acknowledge this difference by distinguishing effects and benefits.[16]

Nevertheless, in order for futility to be meaningful and practical, one needs to specify the material conditions of futility.[17]  J. Walter states:

It is important to note at the outset that nothing is said to be “futile” unless it is judged to be so in relation to some type of goal.  Thus, nothing is futile in the abstract, but a proposed medical intervention can become futile when it is determined that its goal cannot be achieved, no matter how many times the intervention is repeated.[18]

As one moves from the formal conditions to alternative material definitions of futility in the clinical setting, agreement becomes increasingly difficult.  Slosar writes:

A review of the literature reveals that there have been and still are at least four main definitions of medical futility competing for general acceptance: 1) physiological futility, in which the goal of the intervention is the somatic effect of the intervention (e.g., the return of spontaneous respiration and circulation is the physiologic goal of CPR); 2) imminent demise futility, in which the physiologic effect is achieved, but does not result in a significant duration of survival; 3) lethal condition futility, in which the patient survives for some period of time as a result of the treatment, but will die in the not too distant future; and, 4) qualitative futility, in which the patient survives, but with an “unacceptably” poor quality of life.[19]

Nevertheless, two problems persist in determining the material conditions of futility: different goals and different epistemic criteria for knowing whether they will be or should be achieved.  Generally, the different intended goals operative in a dispute over futile treatment are both moral in the sense that medicine is a moral practice grounded in the medical good of the patient (e.g., informing a physician’s goal) and that a patient’s personal life is lived in the pursuit of goods or values, including but not limited to the medical good (e.g., informing a patient’s goal).[20]  J. Walter states:

Two types of goals are almost always involved in medical decision making.  First, the caregiver will have in mind a clinical goal that he or she will seek to achieve on behalf of the patient.  But some clarity is needed here.  Physicians may seek to produce a “medical effect” rather than a “medical benefit” for the patient, and these are not always the same thing.[21]

Later, he continues:

The patient’s personal goal, the second type of goal in most clinical encounters, may or may not be the same as the physician’s clinical goal.  The patient’s goal may be a quality-of-life judgment about the life he or she will have if the caregiver intervenes with medical technology.[22]

While the goal of the physician may be to benefit the patient (not merely producing medical effects), that goal may not support the patient’s personal goals.  Following Pellegrino’s account of the telos of medicine, Robert Walter identifies the ultimate end of medicine to be “serving the best interest of the patient through some action that promotes his or her right and good healing.  All other ends or goods of medicine (intermediary ends or goods) are aimed toward this telos.”[23]  R. Walter cites four goods internal to the practice of medicine: the medical good, the patient-preference good, the autonomy good, and the ‘last resort’ good.  R. Walter writes:

This is the use of medicine’s scientific and technological knowledge to cure, contain, ameliorate, or prevent illness.  It is the judicious application of this knowledge determined in the clinical situation by “medical indications” (which Pellegrino defines as “a statement of those clinical characteristics that make the application of given therapeutic modality worthwhile in certain classes of patients”).[24]

In addition, there are different epistemic criteria involved, which influence the moral quality of the choice.  Here, an evidence-based conception of empirical realities refers to physiological futility (for example); and an evidence-based conception of moral realities refers to patient preferences, quality of life, etc.[25]  The moral relevance of this second difference centers on the degree of certitude one has for each goal: medical (per the physician) and personal (per the patient).

Based on the different teleological understandings mentioned above, one can distinguish between empirical futility and moral futility.[26]  Empirical futility is a descriptive account of a phenomenon (or series of phenomena) that does not achieve a particular end.  Moral futility is a prescriptive account of a phenomenon (or series of phenomena) that does not achieve an intended end.  Based on the different goods mentioned above, one can distinguish two kinds of moral futility: medical futility and integrative futility. On the one hand, medical futility obtains when a treatment (or series of treatments) do not achieve the intended medical good for the patient.  On the other hand, integrative futility obtains when a treatment (or series of treatments) do not serve the good of the patient.  In other words, integrative futility occurs when the patient does not benefit in his or her totality.[27]

There are interesting epistemological questions in these descriptions that have moral relevance.  In short, a question is the degree of certitude one must have in order to ‘know’ that a treatment is futile.  One can frame this epistemological question in an account clinical judgment.[28]  Furthermore, one recognizes that within a physician’s clinical judgment the distinction between moral and absolute certitude is important relative to the corresponding normative questions of culpability and responsibility (e.g., praiseworthiness and blameworthiness).  In other words, with all its problems, what remains in effect is an evidence-based approach to empirical futility, which informs (though does not determine) medical and integrative futility judgments.

Arguably, the primary purpose of futility determinations is to justify authority for unilateral decisions to forego or withdraw treatment.  As a consequence, futility-based decisions may mask and distort discourse over the goals of treatment, which may be a more fruitful purpose than legitimizing unilateral decisions.  Nevertheless, futility remains an a posteriori judgment relative to mutually recognized goals, which may implicitly embrace the notions of empirical and moral futility.  Moreover, the struggle over authority is arguably misplaced: it equivocates the concept of authority as control and the distinction between authority and power.  Physicians and other providers (including health care institutions) invoke futility determinations in an effort to limit (forego or withdraw) medical treatment (typically without adequate communication).  In this sense, this is a power struggle over the use of biomedical technology that is obscured by the struggle over appropriate authority to make such decisions.

Authority

The second substantive element of a process-based approach to decisions involving futile treatment is authority.  Throughout the debate over futility, there has been resistance to physician authority in light of the history of paternalism.  Robert Veatch critiqued paternalism as exercising the fallacious ‘generalization of expertise’ wherein physicians presume to know what is best for their patients by virtue of their medical expertise.[29]  This ‘generalization of expertise’ confuses the medical good described by R. Walter, Pellegrino and others, with the personal goods of individual patients.[30]  Slosar also sees the legitimate authority of physicians, but recognizes the difficulty of describing such authority in the post-modern context.[31]  For Slosar, qualitative futility judgments fail for the same reasons modern moral theory fails: there has been a fragmentation of value, which also gives rise to “various conceptions of moral rationality.”[32]  Nevertheless, does the fragmentation of value and the “various conceptions of moral rationality” require one to abandon the concept of futility altogether?  Does this mean that there is no such thing as physician authority (or patient authority)?  For Slosar, futility retains conceptual significance, but it has little practical import.[33]  While there may be a fragmentation of value, this does not undermine a process-based approach to futility determinations that acknowledge physician and patient authority. 

Daniel Sulmasy identifies three conceptions of authority: authority as control, as expertise, and as warrant.[34]  First, he describes authority as control: “The most typical meaning of authority refers to force of will or the ability of one person to control another’s thoughts, words, or deeds…  To be in authority is to have control.”  Second, Sulmasy explains authority as expertise; here, the term is “frequently used when referring to knowledge, skills, precedents, and conclusive statements…  To be an authority is to possess knowledge and skills superior to others, often rendering the others dependent upon the authority for access to some good or service.”  Lastly, Sulmasy expresses authority as warrant: “A less common but by no means archaic use of the word refers either to the freedom granted by one who is in control, or to actions carried out with conviction…  To act with authority is to act in the freedom granted one by someone else or to act with an apparent sense of legitimacy or conviction.”[35]

For Sulmasy, a physician’s authority is an authority as warrant based in the physician-patient relationship.  The ‘authority as control’ concept dominates contemporary philosophical discussions, yet for Sulmasy it is ultimately an inadequate conception of authority in the physician-patient relationship.  When one conceives of authority as control, conflicts with autonomy naturally arise.[36]  Ultimately, Sulmasy argues that the ‘authority as control’ model is inadequate for the physician-patient relationship for anthropological reasons (the primary human drive is not control over others, but to be in loving relationships with others), metaethical reasons (the control model “assumes that all moral truth in medicine resides in the subjectivity of the autonomous individual,”[37] there is no objective moral truth), and normative reasons relative to the practice of medicine (medicine is a practice, which is not a form of political or organizational power, but oriented toward particular goods—it is “inherently prescriptive”[38]).  Sulmasy concludes:

…I am now convinced that the model so constrains the rich reality of the doctor-patient relationship that it is inadequate.  A solution is not to be sought by accepting the basic correctness of the model and merely suggesting a shift from “unquestioning acceptance of physician authority, as embodied in the Parsonian model” to a “more egalitarian bargaining” state.  The problem lies with the Hobbesian assumptions of the sociological model itself.[39]

Instead of seeing physician authority as control, Sulmasy argues that a physician’s authority is an authority as warrant.

In demonstrating a physician’s authority as warrant, Sulmasy draws upon the New Testament’s terms exousia and dynamis (authority and power, respectively).  In particular, Sulmasy notes the use of these terms,

When Jesus sent his disciples out into the world, he gave them “power and authority to overcome all demons and to cure diseases.  He sent them forth to proclaim the reign of God and heal the afflicted” [Lk. 9:1-2; cf. Mt. 10:1 and Mk. 3:15]. In this passage it is important to note that Luke attributes to Jesus a distinction between the power that heals (dynamis in the Greek) from the authority (exousia) to heal.  This is a distinction which is made with remarkable consistency throughout the writings of the New Testament.  In making this distinction, it would seem that Scripture is suggesting that neither force of will nor the power of expertise is at issue in a discussion of authority to heal.[40]

Sulmasy mentions a few important elements to the New Testament uses of exousia and dynamis.  First, there is a distinction between power and authority to heal.  Through this distinction, one can see that to heal with authority is not the same as an ability to heal—or the ability to exercise power in healing.  Second, dynamis is the power to heal.  “Dynamis is power for, not power over, others.”[41]  This is important for two reasons: the power is other-oriented and the power is bound up with the warrant to heal.  That is, when one heals with authority, one exercises the power to heal for others: he or she heals them.  Third, exousia is an ‘authority as warrant’ that (a) presupposes dynamis and (b) while not a virtue exousia is oriented toward a telos.  Sulmasy writes:

Exousia presumes dynamis, but not vice-versa.  In Greek usage exousia was an illusion if not backed by real dynamis.  Exousia meant “the warrant or the right to do something”.  Thus, exousia is really closest in meaning to the third definition of authority set forth at the beginning of this essay.

Sulmasy notes that while exousia is not a virtue itself, it is an orientation to virtue and a telos.[42]

If this telos is the good of the patient, and the good of the patient that physicians have power to pursue is an intermediary good (i.e., the medical good); then, when this telos is no longer in reach by the power to heal, medicine has reached its end based on its own reason for being.[43]  Does this also mean that when physicians lose the power to heal, patients can no longer demand of them an authoritative practice of medicine?

Sulmasy links exousia with the practice of medicine.  He writes:

To practice medicine with exousia is to ordain one’s practice to the good of the patient and to ordain one’s practice for the good of the patient to the glory of God.  In this way, the dynamis to heal, which is already given in nature and in human reason, not only becomes actual but has a context and an ultimate orientation, emerging from God and leading back to God.  Exousia therefore demands the virtues of practice: wisdom, equanimity, selflessness, trustworthiness, concern, and fidelity.[44]

A physician practices medicine with authority when he or she organizes his or her practice toward the good of the patient.  Furthermore, when the physician exercises a power to heal, the virtuous physician directs such power toward the total good of the patient.  What kind of power to heal does the physician have?

Physicians and patients have qualitatively different authorities.  On the one hand, physicians have the authority to heal as a warrant to heal. This authority presumes the power to heal via biomedical technology.  This authority and power are other-oriented; one finds their true meaning in promoting the good of the patient, especially seeking the medical good.  On the other hand, patients have the authority to self-actualize as control to pursue life’s values necessary for human fulfillment.  This is one understanding of personal autonomy and freedom that providers seek to respect by in informed consent procedures.  This ‘control’ is neither control over others by force of will nor—like the physician’s authority—other-oriented; this authority is over self-control.[45]  In illness, authority is threatened indirectly, but it is not destroyed.  However, the patient’s personal power to seek the goods necessary for fulfillment is radically threatened; the power to pursue the goods that fulfill human needs requires external assistance.  Since, medical goods fulfill vital needs, patients enter into healing relationships with physicians in order to seek those goods.  In this sense, futility cannot be about disputes over authority: (1) physicians and patients have qualitatively different authorities and (2) their powers intersect when a patient seeks the healing power of the physician.  In these ways, futility disputes are never truly about authority, but they are about the quality of the power to heal a physician has at his or her disposal.

As mentioned above, the medical good is dependent upon medical knowledge and skill, but the medical good was only one of four intermediary goods subsumed under the ultimate telos of medicine: the good of the patient.  How then, ought one to reconcile the authority of a physician with the freedom of the patient?  Sulmasy argues:

Exousia implies that authority is an assertion of the other in freedom.  It is not mere dynamis, which is really indifferent to the will of the other, nor is it a coercive use of power, which is the assertion of personal will against the will of the other.  Exousia is authority which addresses human freedom and human reason.  Exousia is authority which assumes a mutual orientation toward a tertium quid.  Exousia is always at the service of others and their freedom.[46]

For Sulmasy, the relationship between exousia and the patient’s freedom is one that recognizes a mutual orientation to the ultimate telos of medicine and its intermediary goods.  Therefore, the overall goal of medicine—the good of the patient in his or her totality—conditions the pursuit of the medical good.[47]  This mutual orientation is a mutual recognition of shared goals.  If decision-making is to be successful, then this mutual recognition should be explicit and not presumed.

This understanding futility disputes still leaves open questions about situations when no mutually recognized goals are present.  Here, one could appeal to the physician’s authority as warrant to heal coupled with power to heal.  What grounds the physician’s power to heal?  How does a physician concretely exercise this power to heal?  If the physician’s authority presumes the power to heal, then a physician must exercise that power in some way as an embodiment of the practice of medicine.  The power to heal, in essence, is the artful application of biomedical technology[48] through sound clinical judgment.  Here, biomedical technology is an “instrument of power” in the healing arts.  Ian Barbour explains:

…technology is neither inherently good nor inherently evil but is an ambiguous instrument of power whose consequences depend on its social context.  Some technologies seem to be neutral if they can be used for good or evil according to the goals of the users.  A knife can be used for surgery or for murder.  An isotope separator can enrich uranium for peaceful nuclear reactors or for aggression with nuclear weapons.  But historical analysis suggests that most technologies are already molded by particular interests and institutional goals.  Technologies are social constructions, and they are seldom neutral because particular purposes are already built into their design.  Alternative purposes would lead to alternative designs.  Yet most designs still allow some choice as to how they are deployed.[49]

Biomedical technology is not value-neutral because the purposes of the healing arts are built-in to the very structure of the technology (or techniques).  Thus, when patients vie for control over the biomedical technologies, demanding—as it were—their use, they are going beyond the boundary of their freedom and usurping the authority of physicians: they are taking away from physicians the warrant to heal for themselves.

Medically futile treatments exist when the authority as warrant to heal remains without the power to heal.  While physicians retain their authority, their power is transformed from a positive action (to heal) to an inability, a negative action (not to heal)—they have lost the power to heal.[50]  Sulmasy writes:

The freedom of exousia is the freedom which comes with liberation from self-preoccupation.  It is the freedom which only loving service can bring.  It is also the free acceptance of human nature with all its inherent limits, including death.  It is liberation from the punishment of Sisyphus, condemned to the eternal trial of attempting to make those limits disappear.  It is therefore liberation from both the entrepreneurial approach to medicine often assumed by physicians and the consumerist approach to medicine often assumed by patients.  Because the doctor does not own the authority to heal, the doctor cannot put healing up for sale on the market.  Because the patient cannot purchase immortality, the patient need not expend all his or her human resources on a grandiose death-denying delusion.[51]

Likewise, patients and their decision-makers do not have authority (as control) to demand empirically futile treatment because it will not serve their interests (i.e., their pursuit of values).  Hence, physicians do have authority to say, “No,” because there is no warrant to use empirically futile treatment because it will not heal. 

Conclusion

There is no consensus on the definition of the term “futility” or even “medical futility.”  However, many commentators propose shifting the burden of the decision from the concept of futility itself to a process that mediates disputes over so-called futile treatment.  However, this process often requires some concept of futility and presupposes a conflict over authority in decision-making.  I argue that such process-based approaches benefit from ethical reflection regarding concepts of futility as well as authority and power operative in clinical decision-making.[52]  What is the practical import of these reflections?[53]  First, stakeholders[54] should anchor processes in values-based discussions about goals.[55]  The hope here is that stakeholders can come to a mutual understanding of the goals before a judgment about futility.  Likewise, this might help caregivers and patients shift from a curative paradigm in choosing treatments to a more palliative one should it be medically and morally appropriate.  Another aspect of anchoring the discussion around goals in an open, values-based discussion is trust wherein the relationship between providers and patients is strengthened, not threatened.[56]  Setting up a process around futility suggests that the stakeholders may already be in an antagonistic position relative to the other.  In addition, it might reinforce defensive behavior and combative dispositions. 

Second, stakeholders should scope out each one’s responsibility in the decision-making process.  This includes the mutual recognition of the domains of decision-making based on the correlative authorities and power.  Hopefully this will result in mutual respect between the stakeholders.   Each stakeholder can then recognize the other’s obligations (cf. the ordinary-extraordinary means of sustaining life), which includes obligations to provide and to accept treatment as well as obligations to share trustworthy evidence of the patient’s values (or preferences) and evidence of what the medical good is and how best to achieve that good.[57]  Both sets of participants share the evidence not in a purely descriptive way; rather they do so in the context of a values-based discussion.

Finally, physicians and institutions should integrate the following characteristics into any process-based approach to mediating futility disputes.  First, the best way to handle futility cases is to prevent them; in this way, process-based approaches should be proactive.[58]  Second, process-based approaches should be goal-oriented; this will support futility judgments insofar as both the means and ends enter into the conversation explicitly.  Third, process-based approaches should be inclusive in order to maximize the viewpoints of the goods considered as well as expanding access to the evidence-base.[59]  Fourth, process-based approaches should be communicative both substantively as well as structurally.  That is, stakeholders should frequently revisit and discuss the goals and the on-going treatment plan.  This characteristic should also reflect cultural and linguistic competency for patients and providers.  Fifth, process-based approaches should be time-sensitive so that stakeholders make and implement decisions in a timely manner.  Finally, process-based approaches should be evidence-based in two ways: empirically (e.g., relating the probabilities of causing a particular clinical effect) and morally (e.g., evaluating the risk, the quality of life, and the preferences of patients in the pursuit of the good of the patient).  Here, it will be important to recognize that an agent needs moral certitude, which is not always equivalent to absolute certitude.[60]

There are several questions that remain unanswered and many issues will militate against this view of futility, authority, and power in clinical decisions involving a futility claim (especially unilateral decisions).  Perhaps the greatest impediment to resolving futility cases is the legal, social, and cultural context in health care that is hostile to values-based discussions at the end of life.  Governments, institutions, and providers may be limited in their ability to effectively address futility until there is significant transformation throughout society at large.  Patients and the medical profession will benefit from such transformation.

Acknowledgements

I would like to acknowledge two colleagues, Robert Van der Waag and Jennifer Shaw, for their constructive feedback on earlier drafts of this essay.  I would also like to thank Providence Health & Services in Southern California and St. Vincent Medical Center in Los Angeles, CA.  The reflections contained in this essay would not have been possible without the collaboration of the health care providers and ethics committees at those hospitals.