INTRODUCTION — The concept of medical futility is an ancient one. Hippocrates counseled clinicians not to treat patients who were “overmastered by their disease.” For centuries, most codes of medical ethics have stipulated that doctors should not recommend or provide treatments that, by their clinical judgment, would not help a patient.
Modern controversies about medical futility arise when doctors and patients (or family members) disagree about whether a particular treatment is futile. In some analyses of such controversies, the word “futility” is avoided, and treatments are instead described as inappropriate, inadvisable, nonindicated, potentially inappropriate, or nonbeneficial. Regardless of which word is used, controversy arises because of different perspectives about what, exactly, constitutes a benefit to a particular patient. Physicians, patients, and family members may have different views about, say, the value of a brief prolongation of life when death is inevitable. At times, the quality of a patient’s life (eg, lack of awareness of surroundings and/or inability to interact with surroundings without anticipated recovery) would lead many to determine there is little value in continuing life-sustaining interventions. In short, medical technological capacities may provide the opportunity to challenge moral and ethical beliefs and frameworks.
This topic review covers the concept of medical futility as it applies to palliative care patients. Ethical issues related to futile and potentially inappropriate treatments that are specific to issues that arise in the intensive care unit from a critical care perspective are presented separately, as is a more extensive discussion of ethical issues that arise in palliative care. (See "Ethics in the intensive care unit: Responding to requests for potentially inappropriate therapies in adults" and "Ethical issues in palliative care".)
AUTONOMY AND MEDICAL FUTILITY — Controversies about medical futility are entwined with the emergence, in the last half of the 20th century, of patient autonomy as the dominant principle in medical ethics. Prior to that, doctors routinely made unilateral decisions about what treatments to offer and what to withhold [1,2]. There was no legal or (perceived) ethical need to discuss such decisions with patients or families. That began to change in the 1960s, when the concept of “informed consent” became embedded in the law governing doctor-patient communication. To ensure informed consent, the doctor was required to explain the risks and benefits of each available treatment to patients and family members and then defer to patient choices. (See "Ethical issues in palliative care", section on 'Framework for ethical reasoning' and "Informed procedural consent".)
The late 20th century also saw increasing use of life-sustaining medical technologies that enabled the prolongation of life . For dying patients, there are now almost always choices to be made about which treatments to provide or withhold, and patients, families, and physicians must engage in a process of shared decision-making .
Shared decision-making and informed consent or refusal has advantages and disadvantages over the older paternalistic approach. Patients and families are now allowed to discontinue treatments they deem to be at odds with their goals. They are also informed about and allowed to request treatments that clinicians believe are inappropriate or futile. In the 1980s and 1990s, published reports emerged of patients and families insisting on medical interventions that would not be able to achieve the ends of reversing the dying process and sustaining meaningful life. Many physicians and bioethicists argued that clinicians should be allowed to unilaterally withhold treatments that they considered futile or potentially inappropriate . Some intractable disagreements went to court and led to some landmark decisions defining the issues in this contentious domain.
Representative cases — The following cases highlight the ways in which controversies about medical futility arose and played out in United States courts:
●Helga Wanglie – Helga Wanglie was 86 years old when, in December 1989, she fell at home and broke her hip. After receiving care at the county hospital, she was discharged to a nursing home for rehabilitative care. There, she developed respiratory distress and was transferred back to the hospital where she was subsequently intubated and ventilated. She was unable to breathe on her own and remained on the ventilator for many months. She was transferred to another long-term care facility for chronic respiratory failure. During this time, she was conscious and aware of her environment. Ultimately, during attempts at weaning and separating from ventilation, she suffered a cardiopulmonary arrest with severe anoxic brain injury.
Mrs. Wanglie was diagnosed as being in a persistent vegetative state (PVS). She underwent months of ongoing support with mechanical ventilation, gastrostomy feeds, and antibiotics for multiple recurrent infections. Her husband and children acknowledged the severity of her brain injury and recognized the medical view that this was irreversible.
The medical staff believed further care to not be in the best interest of Mrs. Wanglie. Her family believed that the medical staff should not “play God” and that stopping medical care would be against their religious views. Multiple family conferences were convened, and an ethics consult was obtained. The family eventually agreed to placement of a do-not-resuscitate (DNR) order but did not want her removed from the ventilator or other treatments withheld. Medical staff and administration agreed with the ethics committee recommendation to stop interventions. The hospital and family sought another hospital to take over Mrs. Wanglie’s care, which could not be achieved [6,7].
The hospital requested that the family obtain a court order to continue care. When the family did not, the hospital filed a petition with the court. The court appointed Mr. Wanglie as his wife’s guardian and supported his decision to continue care. The decision came three days before Mrs. Wanglie's death on July 4, 1991, at the age of 88 from sepsis .
●Baby K – Stephanie Keene, referred to as Baby K during the ethical debates that surrounded her life, was an infant with the prenatal diagnosis of anencephaly, with a brainstem to maintain heart rate, blood pressure, and some respiratory effort but no higher cognitive abilities. Her mother refused the recommendation to terminate the pregnancy on religious grounds, and Baby K was born on October 13, 1992.
Consistent with her underlying pathology, Baby K had significant respiratory insufficiency at birth and throughout her life. Her mother refused to place a DNR order as advised by medical staff. Baby K was on mechanical ventilator support for the first six weeks of life, during which time attempts were made to transfer her to another hospital, but no other facility would accept her. She was eventually weaned from the ventilator and transferred to a long-term care facility for other medical needs. However, she had continual episodes of respiratory distress for which she was frequently transferred back to the hospital for ventilator support. Baby K’s mother insisted upon continued resuscitation and intubation. Baby K’s father, who was not married to the mother, supported the hospital in its objections to ongoing resuscitation.
The hospital filed a petition with the courts to appoint a guardian and to allow the hospital to provide only palliative care. The American Academy of Pediatrics wrote a brief in support of the hospital. The court gave a sweeping decision that in providing only palliative care against the wishes of the mother, the hospital would be in violation of multiple federal laws, including the Emergency Medical Treatment and Labor Act (EMTALA), the Americans with Disabilities Act, and the Rehabilitation Act, as well as the mother’s constitutional rights as protected by the 14th Amendment . The decision was affirmed by the federal court of appeals on only the basis of EMTALA .
There was clear disagreement within the medical ethics community, with Dr. Michael Grodin, a pediatrician and head of the Program on Medical Ethics at the Boston University School of Medicine and Public Health, stating, “For a baby that has no brain and is born dying, life for just another minute or another day is not an appropriate goal. It’s saying that it’s the technology that has life, not the patient.” On the other hand, Dr. Robert M. Veatch, director of the Kennedy Institute of Ethics at Georgetown University, stated that the mother’s and court’s decision “is well within the range of parental discretion.” The discrepant opinions that surrounded this case reveal the heated debate over the concept of medical futility and how it is defined and applied .
Notably, in both cases, the courts ruled in favor of family members advocating for continued medical intervention and against the medical providers who were arguing that such intervention was “futile” or not warranted due to the compromised state of the lives of the individual patients in question. The unique nuances of each case make it hard to generalize to broadly applicable policies. The decision in Baby K relied on EMTALA, but that law applies only to emergency department patients. Since most treatment conflicts concern inpatients, EMTALA is often not relevant. In Wanglie, the hospital asked the court to rule on the propriety of the surrogate instead of the propriety of the treatment itself.
While courts have often ruled in favor of families and against unilateral physician determinations of futility in many of the early cases, that has begun to change. In more recent cases, they sometimes allow unilateral determinations of futility . One notable high-profile case in which courts struggled with this issue is the case of Tinslee Lewis.
●Tinslee Lewis – Tinslee Lewis was born prematurely on Feb 1st, 2019, in Texas. She was diagnosed with Ebstein’s anomaly with pulmonary atresia. She underwent a number of cardiac surgeries to palliate her heart disease but developed worsening chronic lung failure with severe pulmonary hypertension. She remained in the intensive care unit, ventilator-dependent and heavily sedated. When she reached six months of age, her doctors’ opinions were that further interventions could not save her. They believed even routine care caused pain and suffering. They recommended withdrawing life support, but the mother disagreed. The doctors followed the processes outlined by Texas law for such cases (see 'Texas Advance Directives Act' below). They sought and received the concurrence of the hospital ethics committee. They notified the mother of this. They reportedly contacted 20 other hospitals to see if they would accept Tinslee in transfer. None would .
A lower court ruled in December 2019 that it would be legally permissible to remove Tinslee from the ventilator. But in July 2020, the Texas Court of Appeals reversed, sending the case back for trial in the Tarrant County District Court . The appellate court held that since the physicians and hospitals would be “state actors” when withdrawing treatment over parental objections pursuant to the Texas Advance Directives Act, they must comply with requirements of constitutional due process. The hospital had failed to do that because they did not provide the mother with sufficient notice or opportunity to participate in the ethics committee deliberations. Moreover, the Court of Appeals averred, the committee lacked sufficiently objective standards by which to reach its decision.
The mother appealed. But both the Texas Supreme Court and the United States Supreme Court declined to review the case. A trial on the constitutionality of the Texas Advance Directives Act was scheduled in Tarrant County District Court for early 2022. But the conflict was mooted by April 2022 when Tinslee was discharged from the hospital and transferred home. Thus, in this case, the court leaned towards empowering doctors to remove life support over the objections of the parents, but never issued a final, determinative opinion.
The case raises complex issues about the authority of hospital ethics committees and the constitutionality of the Texas law that permits unilateral decisions by doctors to remove life-support over the objections of patients and family members. Tinslee’s survival to discharge also raises questions about the accuracy or meaning of a clinical determination that continued treatment would be futile. (See 'Definitions of futility' below.)
Representative United Kingdom cases — Because intractable British medical futility conflicts are typically resolved publicly in court, there is more transparency and uniformity. The following cases highlight the ways in which controversies about medical futility arose and were played out in United Kingdom courts:
●Charlie Gard – Charlie Gard was born on August 6, 2016. At two months of age, he became lethargic and developed respiratory failure. He was admitted to the intensive care unit at Great Ormond Street Hospital (GOSH) in London and started on mechanical ventilation. He was eventually diagnosed with a mitochondrial deoxyribonucleic acid (DNA) depletion syndrome (MDDS), a rare clinically and genetically heterogeneous condition .
Through their own research, Charlie’s parents discovered an experimental therapy called “nucleoside therapy.” They asked the doctors at GOSH to consider a trial of nucleoside therapy, and the physicians agreed. But Charlie developed status epilepticus, suffering further brain damage. Based on that new status, Charlie’s doctors concluded that nucleoside therapy would thus be futile, and they recommended withdrawal of life support. The parents requested instead that Charlie be transferred to the United States and to the care of a doctor who was willing to try nucleoside therapy. The doctors at GOSH thought that the treatment would be futile and thus transfer would not be in Charlie’s best interest. So, GOSH petitioned the United Kingdom High Court of Justice to allow Charlie’s mechanical ventilation to be discontinued over the parents’ objections.
The High Court concluded that Charlie was suffering and therefore it was in his best interests to have life support withdrawn. The High Court judge also stressed that the decision was not based in any way on the cost of treatment .
The parents appealed. Charlie’s lawyers argued that even a small chance of benefit from a trial of therapy outweighed the potential harms. More importantly, they argued that the well-informed and caring parents, in collaboration with a doctor of their choice, should make the judgment about the balance of harms and benefits. The Court of Appeals upheld the High Court’s opinion . The United Kingdom Supreme Court decided not to hear an appeal . As a last resort, the parents appealed to the European Union Court of Human Rights. That court concluded that the decision to withdraw life support did not violate Charlie’s human rights . Eventually, the outside consultant came to agree that Charlie’s disease had progressed to the point where therapy would no longer be beneficial. The parents gave up their fight, and Charlie was transferred to hospice, where he died.
This case has implications for the process of shared decision-making in such circumstances. Generally, when physicians disagree about the care of a child, we defer to parents and allow them to choose which physician’s recommendations to follow . In the United States, this is true even in states such as Texas that permit unilateral physician decisions to withdraw life support . Under Texas law, when doctors determine that further life support is inappropriate, parents are given 10 days to find another physician. If they do, their child will be transferred to the care of that physician . In California, this was even true for a child who was declared brain dead . The precedent in the Gard case suggests that, instead, courts may make an independent evaluation of which medically endorsed treatment is best.
●Tafida Raqeeb – In dozens of cases, many with extremely public profiles (like Alfie Evans and Archie Battersbee), British courts sided with clinicians in medical futility disputes as they did in the Charlie Gard case . But in 2019, the court did not agree with clinicians at Barts Health NHS Trust that stopping life-support treatment was in the best interest of five-year-old Tafida Raqeeb . Tafida had suffered from a brain arteriovenous malformation, which resulted in a burst blood vessel in her brain. Key to the court’s decision was the family’s Muslim beliefs and their plan to transfer Tafida to a hospital in Italy. The court found Tafida had a minimal awareness of her surroundings, was possibly able to interact with them, and was not in any pain. The court further found that while Tafida was unlikely to recover significantly, she might be able to breathe without ventilation. Finally, the court noted that children in the United Kingdom in similar conditions to Tafida are already receiving treatment of a similar nature. The United Kingdom cases demonstrate that neither physician nor parental authority is absolute, and that patient nuances must be considered in respect to other cultural values.
DEFINITIONS OF FUTILITY — Medical futility has proven difficult to define. In the Oxford English Dictionary, the word futility is defined as “leaky, vain, falling out of the desired end through intrinsic defect.” The word is derived from the Latin “futtilis,” meaning a vessel wide at the top and narrow at the bottom that tipped over and poured out its contents whenever it was filled. These definitions and word origins invoke the images of the Danaïdes from Greek mythology, eternally filling an unfillable vessel to atone for their sins . Doctors and nurses caring for extremely ill patients could resonate with the feeling of carrying water in a leaky bucket, continually working towards a goal that could not be achieved. Attempts to provide clarity to the definition of futility have led to multiple qualifiers of the term. The six most significant terms and concepts are described below.
Physiological futility — Physiological futility describes medical interventions that could not possibly achieve their intended physiological goal [27,28]. Using a specific antimicrobial, for instance, for a bacterial infection that is resistant to that antimicrobial is physiologically futile. Another example is performing cardiopulmonary resuscitation (CPR) on a patient with a ruptured myocardium or a patient with multisystem organ failure on maximal ventilatory and vasopressor support.
This definition carries two assumptions: that each medical intervention has a specific physiological goal and that the clinician can determine if achieving that goal is possible. An action or treatment can only be considered futile in relation to a specific goal.
Critics of the use of the term “physiological futility” argue two points:
●Goals may exist other than the perceived physiological goal. For example, in performing CPR on a terminally ill patient, perhaps the true goal is not to restore a spontaneous heartbeat but rather to give the family a measure of certainty that everything possible has been attempted. In this instance, CPR could be considered “physiologically futile” for achieving return of circulation but not necessarily futile in helping the family cope, psychologically, with the death of a loved one.
●The term “physiological futility” implies that a clinician knows with certainty the likelihood that an intervention cannot reach a physiological goal. However, the clinician’s predictive abilities may not be that absolute. Even when multiple physicians agree that a treatment is futile and death inevitable, they can be proven wrong if treatment is continued . However, physicians often disagree with one another about whether a particular treatment for a particular patient is futile .
Quantitative futility — The term quantitative futility focuses on the numeric probability of achieving the intended goal of therapy, which is generally presumed to be cure and survival . It is used interchangeably with physiological futility by some, although others argue that quantitative futility, the likelihood or chance that a treatment will benefit a patient, differs from physiologic futility, which measures the outcome achieved, whether likely or not, by the effect on the patient’s physiology . Properly understood, physiological futility is quantitative futility with a probability of zero.
Those who advocate the use of this term recognize that a possibility of death or survival is not absolute and that some probability threshold should be established. This approach utilizes published data on medical outcomes incorporated with personal clinician experience to determine a probability of survival following a given treatment.
While this mechanism of applying data attempts to make decisions more straightforward, it requires a judgment call on what level of likelihood should be the threshold for making a decision. Some recommend that clinicians use empiric evidence, including personal experience or published data, to assess the likelihood of a treatment being successful. One proposal is that if the last 100 cases experienced or reported indicate that the treatment did not achieve the desired end, then it could be regarded as futile . However, is ≤1 percent possibility of survival an appropriate threshold ? What if another clinician or family member believes that a 0.5 percent chance of survival warrants aggressive therapeutic interventions? Whose threshold should be used for making the decision? Investigators have shown wide variations in the threshold of judging quantitative futility when physicians and laypeople are compared .
Another problem is that most published studies that purport to guide determinations of futility are based on insufficient data to provide statistical confidence for clinical decision-making . In addition, available published data on survivability may already be skewed by health care perceptions on futility. Published mortality rates for a specific patient demographic may be 100 percent or close to it because attempts at improving this outcome are not made because of the poor prognosis, thus creating a self-fulfilling prophecy.
Qualitative futility — Viewing futility from a qualitative viewpoint shifts the focus from achieving a level of certainty to a focus on the quality of the potential benefits . In other words, there is a distinction made between an effect of treatment that is limited to some part of the patient’s body and a benefit that improves the patient as a whole that can be perceived and appreciated by that patient .
If a therapeutic intervention might have a good chance to prolong survival even a small amount, with minimal suffering and an improvement in quality of life, it would not meet criteria for physiologic or quantitative futility. On the other hand, if an intervention would cause significant suffering while offering only a brief extension to the length of life, one could argue that it should be considered qualitatively futile. This means of determining futility requires individuals to make judgment calls on what constitutes a sufficiently low quality of life. Such judgments inevitably raise fears that such qualitative futility judgments may mask discriminatory attitudes about people with disabilities. This became especially pertinent during the coronavirus disease 2019 (COVID-19) pandemic when critics highlighted the ways that some frameworks for rationing, that were based on objective measures of illness severity, incorporated judgments about quality of life for people with disabilities . Even seemingly objective measures as age or body mass index can be used in ways that exacerbate disparities and thus are unjust .
Invoking qualitative futility sometimes leads to disagreements between clinicians, patients, and family members. Different people may view the balance of benefit and harm from certain therapeutic options quite differently. In most cases, doctors, patients, and family members agree. For example, there is a broad consensus that withdrawal of life-support is appropriate when a patient is permanently unconsciousness . But even a broad consensus doesn’t eliminate the possibility of different views.
Imminent demise futility — Other ways to define futility include “lethal condition futility” and “imminent demise futility” . In both these situations, the underlying condition is terminal in the short term. In “lethal conditions,” that is usually weeks or months. In “imminent demise,” it is more likely to be hours or days. Implicit in the determination of futility is the fact that the dying process cannot be reversed or significantly altered by treatment. While an intervention may have a physiological impact upon the patient, such as CPR restoring a spontaneous heartbeat, they argue that the intervention will not alter the trajectory of the underlying illness and is therefore futile.
Two states have implemented this concept of futility in their Provider Orders for Life-sustaining Treatment (POLST) laws. Vermont permits a physician to write a do-not-resuscitate (DNR) order without consent when CPR would not prevent “imminent death” . Similarly, Maryland permits a physician to write a DNR order without consent when CPR would be “medically ineffective,” which means that it would not prevent “impending death” .
Potentially inappropriate treatment — In 2015, five critical care societies, the American Thoracic Society (ATS), the American Association of Critical-Care Nurses (AACN), the American College of Chest Physicians (ACCP), the European Society of Intensive Care Medicine (ESICM), and the Society of Critical Care Medicine (SCCM), officially endorsed new and specific terminology that should be used when discussing and categorizing treatments of questionable benefit  (see 'ATS/AACN/ACCP/ESICM/SCCM (2015)' below):
●The five-society statement maintains that the term “futile” should be used only in the rare circumstance that an intervention definitely cannot accomplish the intended physiological goal. The statement provides the example of performing CPR on a patient with a ruptured myocardium. (See 'Physiological futility' above.)
●The term “potentially inappropriate” should be used, rather than “futile,” to describe the more common situation in which treatments have at least some chance of accomplishing the effect sought by the patient, but clinicians believe that competing ethical considerations justify not providing them. In contrast to “futile” treatment, disagreements over a “potentially inappropriate” treatment are value based.
We agree with these definitions.
Other treatments for which benefit is considered unlikely, uncertain, inadvisable, or controversial should be discussed when applying the terminology of potentially inappropriate, rather than futile, treatments. These issues, as well as suggestions as to appropriate responses to requests for inappropriate/inadvisable treatments in the intensive care unit, are discussed in more detail separately. (See "Ethics in the intensive care unit: Responding to requests for potentially inappropriate therapies in adults", section on 'Definitions' and "Ethics in the intensive care unit: Responding to requests for potentially inappropriate therapies in adults", section on 'Responding to requests for potentially inappropriate interventions'.)
Proscribed and discretionary treatment — In addition to distinguishing futile and potentially inappropriate treatment, the five-society statement (ATS/AACN/ACCP/ESICM/SCCM) also carves out two other categories :
●“Legally proscribed” treatments are those that are prohibited by applicable laws, judicial precedent, or widely accepted public policies. The statement offers the example of circumventing organ allocation policies.
●“Legally discretionary” treatments are those for which there are specific laws, judicial precedent, or policies that give physicians permission to refuse to administer them. The statement offers the example of POLST laws in Maryland and Vermont that permit clinicians to write DNR orders without patient or surrogate consent when CPR would not prevent the patient’s “imminent death.”
FUTILITY VERSUS COST-EFFECTIVENESS AND RATIONING — Some believe that the issue of futility will inevitably be submerged in the debate about health care reform and utilization of limited resources, and the need for society to determine where resources are appropriately spent . Societies with nationalized health care have certainly made decisions about how money and resources should be allocated based upon cost-effectiveness. Providers working in specific locations or epochs with limited resources may be more reflective on what constitutes futile treatment.
However, there is a clear distinction between cost-effectiveness and futility, and between futile and rationed treatments . Clinicians should not label interventions as futile simply because they believe the treatment is not cost-effective, or because the treatment has limited availability and must be rationed. Care that may be expensive or resource intensive may not necessarily be futile, as it may provide benefit to the patient and achieve specific goals. Focusing on the impact to society minimizes the values of specific benefits or harms to the patient, which should be the primary factors considered.
Nonfutile treatments, however, may be denied during the allocation of scarce resources based on other ethical principles. During “crisis standards of care,” allocation paradigms may be altered to exclude those who are less likely to benefit and/or have a protracted course. This was seen with the allocation of advanced medical technology, as well as resources such as intensive care unit beds, during the coronavirus pandemic. Patients who may have shown some benefit were likely denied medical modalities due to resource scarcity. Confusing rationing decisions with decisions based on futility could result in misallocation of lifesaving treatments once resource limitations have resolved. This type of rationing must have strong oversight to minimize inequities .
RISK OF BIAS — A key justification for the renewed emphasis on vocabulary (futile versus potentially inappropriate) is to mitigate the risk of bias. Permitting clinicians to too-readily apply the label “futility” to clinical situations leaves value-laden judgments unrecognized and unexamined, and some of those judgments may be colored by implicit bias. For example, in 2020, the case of Michael Hickson was widely reported . Litigation and regulatory agency investigations continue to examine the extent to which clinicians stopped Hickson’s life-sustaining treatment because of stereotypes about what life is like with a disability. One clinician told the patient’s wife: “Because he's paralyzed with a brain injury, he doesn't have quality of life." The National Council on Disability rightly calls for checks on a physician’s subjective judgments about whether a patient’s life is “worth living” .
Importantly, even a fair dispute resolution process that comports with procedural due process might be unable to completely eliminate the risk of bias. Ethics committees or other institutional reviews typically examine the appropriateness of treatment for one specific patient in one specific dispute. Left unexamined are the broader ways in which clinicians invoke a conflict resolution policy . For example, a hospital may have two clinically identical patients each with families requesting life-sustaining treatment. But the clinician resists only one family and claims treatment is inappropriate for only one patient.
COMPLEXITY OF PROGNOSTICATION — One of the major barriers to the definition and acceptance of medical futility and potentially inappropriate treatments is the medical community’s struggle with prognostication. Some degree of uncertainty almost always exists in any prognostic estimate that is delivered to patients and their families. Clinicians often struggle with this uncertainty and with how to honestly communicate this appropriately to families [49,50]. As an example, in Kalanithi’s book, When Breath Becomes Air, Dr. Kalanithi, a neurosurgeon, is diagnosed with metastatic cancer. He asks his oncologist how long he has to live. She refuses to give him a number and says, instead, that he needs to think about what he wants to do while he is alive rather than how long he will live . The tension is clear. In order to deem a treatment futile, doctors must be precise in their prognostication. But prognostication is inevitably a matter of probabilities and therefore always somewhat imprecise. (See "Communication of prognosis in palliative care", section on 'The science of estimating prognosis' and "Survival estimates in advanced terminal cancer", section on 'Accuracy of survival estimates'.)
Attempts have been made to use individual patients’ data to predict the likelihood of survival. Many of these have been used to predict the likelihood of survival among critically ill patients in the intensive care unit. Predictive scoring systems such as these are designed to estimate survival or mortality for populations of patients and are highly inaccurate when used to predict outcomes in a specific individual patient . They should not be used for clinical decisions about futility. (See "Predictive scoring systems in the intensive care unit".)
MAJOR POLICY STATEMENTS — Due to the need for consensus on how to address concerns regarding futile or potentially inappropriate treatments, multiple groups, including professional societies, have established policies or guidelines.
American Medical Association (1997) — In 1997, based on the report from its Council on Ethical and Judicial Affairs, “Medical Futility in End-of-Life Care,” the American Medical Association (AMA) released its policy in Opinion 2.037 . It states: “When further intervention to prolong the life of a patient becomes futile, physicians have an obligation to shift the intent of care toward comfort and closure.” To help with this process, the AMA recommends that all hospitals develop medical futility policies and suggests an algorithmic approach to ensure due process. This guidance is included in the AMA Code of Medical Ethics.
American Academy of Hospice and Palliative Medicine (2011) — The American Academy of Hospice and Palliative Medicine (AAHPM) has an official position on withholding and withdrawing nonbeneficial medical treatments, which are considered ethically indistinguishable and both ethically and legally permissible . The policy, however, does not discuss utilizing the concept of futility, and the term is not addressed. It does state that an ethics consultation should be considered if disagreements exist between the health care team and patients or their surrogates.
American Thoracic Society (2015) — In addition to being part of the five-society statement on potentially inappropriate treatment  the American Thoracic Society (ATS) also has a 2015 policy statement on managing “conscientious objections” (ie, situations where the clinician has a moral objection to providing or disclosing information about a legal, professionally accepted, and otherwise available medical service) in intensive care medicine . The recommendations are based on the dual goals of protecting patients’ access to medical services and protecting the moral integrity of clinicians. Their approach is outlined in an algorithm (algorithm 1). This statement emphasizes that an individual clinician should not use their conscientious objection to justify unilaterally deciding to forego what they perceive to be futile treatment.
ATS/AACN/ACCP/ESICM/SCCM (2015) — In addition to distinguishing futile, proscribed, discretionary, and potentially inappropriate treatment vocabulary, as described above, the five-society statement (ATS, American Association of Critical-Care Nurses [AACN], American College of Chest Physicians [ACCP], European Society of Intensive Care Medicine [ESICM], Society of Critical Care Medicine [SCCM]), offers recommendations for how clinicians should respond to each type of request :
●Clinicians should not provide futile or proscribed interventions. In responding to requests for futile, proscribed, or discretionary treatment, clinicians should carefully explain the rationale for treatment refusal.
●Responding to requests for potentially inappropriate treatment is more complicated. These value-based disagreements require a fair-process approach for conflict resolution. The recommended approach, as reflected in an algorithm to address disputed treatments in the intensive care unit (algorithm 2), includes the following seven-step process:
•Enlist expert consultation to continue negotiation during the dispute-resolution process.
•Give notice of the process to surrogates.
•Obtain a second medical opinion.
•Obtain review by an interdisciplinary hospital committee.
•Offer surrogates the opportunity to transfer the patient to an alternative institution.
•Inform surrogates of the opportunity to pursue extramural appeal.
•Implement the decision of the resolution process.
The five-society statement recognizes that time pressures may sometimes make it infeasible to complete all steps of the process. In these cases, clinicians may refuse to provide the disputed treatment if they have a high degree of certainty that the requested treatment is outside accepted practice and if they complete as much of the seven-step process as possible.
Society of Critical Care Medicine (2016) — Building on the five-society statement, SCCM recommended that intensive care unit interventions should generally be considered inappropriate when there is no reasonable expectation that the patient will improve sufficiently to survive outside the acute care setting, or when there is no reasonable expectation that the patient’s neurologic function will improve sufficiently to allow the patient to perceive the benefits of treatment  (see "Ethics in the intensive care unit: Responding to requests for potentially inappropriate therapies in adults"). The same year, SCCM also provided new intensive care unit admission and discharge guidelines .
American College of Physicians (2019) — The 2019 statement from the American College of Physicians (ACP) states the following :
“In the circumstance that a specific treatment desired by the patient is medically ineffective and potentially harmful, the physician is not ethically obliged to provide such treatment (although the physician should be aware of any relevant state law). The physician need not provide an effort at resuscitation that cannot conceivably restore circulation and breathing, but he or she should help the family to understand and accept this reality. The more common and much more difficult circumstance occurs when treatment offers some small prospect of benefit at a great burden of suffering… but the patient or family nevertheless desires it. If the physician and patient (or appropriate surrogate) cannot agree on how to proceed, there is no easy, automatic solution. Consultation with learned colleagues or an ethics consultation may be helpful in ascertaining what interventions have a reasonable balance of burden and benefit. Timely transfer of care to another clinician who is willing to pursue the patient’s preference may resolve the problem. Infrequently, resort to the courts may be necessary. Some jurisdictions have specific processes and standards for allowing these unilateral decisions… A physician who writes a unilateral DNR order must inform the patient or surrogate and give a detailed explanation when doing so.”
OTHER POLICY STATEMENTS — While the policy statements described above are the most recent and relevant, other groups and professional societies have established policies or guidelines. We describe some of the more significant statements below in chronological order.
●President’s Commission on Biomedical Ethics (1983) – The Presidential Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research was convened in 1978 and continued its work through 1983. Its report, “Deciding to Forego Life-Sustaining Treatment,” addresses many of the evolving ethical concerns around advancing medicine technology . The report stressed the importance of patient and surrogate autonomy in making decisions on life-sustaining treatment. It is the clinician’s responsibility to ensure that ethically appropriate decisions are made, and the clinician is not required to provide therapy that they find ethically objectionable; however, the clinician must not abandon their patient and instead should transfer care to another provider. While somewhat dated, this is still an important document for setting out the framework of futility controversies and potential resolutions.
●San Diego County Medical Society (2009) – The San Diego County Medical Society created its own policy addressing the use of nonbeneficial treatments . This policy is unique in that it recognizes that not all disagreements about medical futility are between doctors and patients or family members. They categorize three types of disagreements: intraprofessional (between members of the treating team), intrafamilial, and between the treating team and the patient/surrogate. The policy outlines means of mediating the different types of conflicts.
●American Academy of Pediatrics (2007, 2017) – The Committee on Fetus and Newborn of the American Academy of Pediatrics (AAP) addressed nonbeneficial treatment. In their opinion, “The physician is not obligated to provide inappropriate treatment or to withhold beneficial treatment at the request of the parents. Treatment that is harmful, of no benefit, or futile and merely prolonging dying should be considered inappropriate. The physician is not obligated to provide inappropriate treatment or to withhold beneficial treatment at the request of the parents. Treatment that is harmful, of no benefit, or futile and merely prolonging dying should be considered inappropriate. The physician must ensure that the chosen treatment, in his or her best medical judgment, is consistent with the best interest of the infant” . This view is reiterated in a statement from the AAP Committees on Hospice, Palliative Care, Bioethics, and Child Abuse. They write that unilateral withdrawal of life-sustaining medical treatment “…may be ethically supportable in rare circumstances of extreme burden of treatment with no benefit to the patient beyond postponement of death” .
●Renal Physicians Association (2010) – The Renal Physicians Association recognizes that conflicts may occur between the patient/legal agent and the renal care team about whether dialysis will benefit the patient. In addition, conflicts also may occur within the renal care team or between the renal care team and other health care providers. A 2010 clinical practice guideline advises physicians to “try to understand [the patient/agent] views, provide data to support his/her recommendation, and correct misunderstandings. If dialysis is indicated emergently, it should be provided while pursuing conflict resolution, provided the patient or legal agent requests it” . (See "Kidney palliative care: Withdrawal of dialysis" and "Kidney palliative care: Principles, benefits, and core components".)
●National Council on Disability (2019) – The National Council on Disability (NCD) recognizes that when a physician decides that providing or continuing health care treatment would be “medically futile,” there are a number of objective, evidence-based factors that can impact this decision. But the NCD is concerned that a physician’s subjective judgments about whether a patient’s life would be “worth living” may go unrecognized and can also play a role in decision-making . Some health care providers, most unknowingly, harbor biases and misperceptions about the quality of life and capacities of people with disabilities. These assumptions can and have impacted physicians’ willingness to provide or continue life-sustaining care.
●Swiss Academy of Medical Sciences (2021) – The Swiss Academy of Medical Sciences (SAMS) released an English language version of its report, “Ineffectiveness and Unlikelihood of Benefit – Dealing with the Concept of Futility in Medicine.” Based on discussions that intensified during the pandemic, the SAMS developed clinician-focused recommendations. These recommendations offer orientation for practice and aim to encourage a discussion on how to deal with ineffectiveness and unlikelihood of benefit .
NOTABLE US STATE LAWS
Texas Advance Directives Act — Recognizing the struggle to define futility, the state of Texas took a unique approach. The Texas Advance Directives Act (TADA) was passed in 1999 and amended in 2003 with the addition of pediatric patients. It provides definitions of “irreversible conditions,” “terminal conditions,” and “life-sustaining treatment.” It delineates steps for approaching disagreements, including first having the case reviewed by an ethics or medical committee [64,65]. The results from this meeting are then presented to the patient or patient surrogate, and they can be asked for in written form. If the clinician and the patient or surrogate continue to disagree, the clinician will attempt to transfer the patient to another clinician or facility. Life-sustaining support will be maintained during any pending transfer and up to 10 days following the written decision. If an accepting clinician or facility cannot be found within those 10 days, the clinician and facility are not required to continue support other than artificial nutrition and hydration .
TADA also provides legal support to the health care team and guarantees that a patient will be supported for a period of time while alternative treatment facilities are considered [67,68].
Arguments against TADA claim that the hospital ethics committee is given too much power and is representative of the health care team, not society as a whole. Opponents argue that these cases would best be handled by judicial courts .
Virginia Health Care Decisions Act — In 2018, Virginia enacted a statute that authorizes a dispute resolution mechanism like TADA. First, all Virginia hospitals must develop a policy regarding the medical and ethical appropriateness of proposed medical care. The policy must provide for a second opinion, a review by an interdisciplinary review committee, and a written explanation of the committee’s decision. If a clinician determines that treatment requested by a patient or surrogate is medically or ethically inappropriate, the clinician must inform the patient or surrogate and provide a copy of the hospital’s policy.
If the conflict persists, then the hospital must give the patient or surrogate 14 days to attempt transfer to a facility that is willing to provide the disputed treatment. During this period, the current treating facility must both continue the disputed treatment and allow access to the patient’s medical records. If the patient is not transferred by the end of the 14 days, the hospital may stop treatment. Like TADA, artificial nutrition and hydration can be unilaterally withdrawn only when it is physiologically futile. But unlike TADA, Virginia allows court review of the hospital’s decision .
California Health Care Decisions Act — The 1999 California Health Care Decisions Act provides that “a health care provider or health care institution may decline to comply with an individual health care instruction or health care decision that requires medically ineffective health care or health care contrary to generally accepted health care standards applicable to the health care provider or institution” . Furthermore, when doing so, “the health care provider or institution is not subject to civil or criminal liability or to discipline for unprofessional conduct” .
Treatment consent laws — While Texas and Virginia empower clinicians, other states have taken an opposite approach and have empowered patients and families. As an example, in 2012 Idaho enacted the Discrimination in Denial of Life-Preserving Treatment Act . This statute mandates that health care “may not be withdrawn or denied if its provision is directed” by the patient’s surrogate. Oklahoma enacted similar legislation, the Nondiscrimination in Treatment Act, in 2013 .
Simon’s laws — More recently, states have enacted laws prohibiting clinicians from writing do-not-resuscitate (DNR) orders without consent. For example, between 2017 and 2022, several states enacted versions of a bill called “Simon’s Law.” The law is named after Simon Crosier, a baby with trisomy 18 for whom physicians instituted a DNR order without his parents’ consent or knowledge. Alabama, Arkansas, Georgia, Kansas, Missouri, and Iowa enacted stronger versions of the law, directed at prohibiting unilateral DNR orders for minors . Other states like Arizona, Idaho, and South Dakota enacted weaker versions directed at informing parents but not requiring their consent.
Simon’s Law typically imposes three mandates on providers. First, it requires hospitals and medical facilities with policies about withholding life-sustaining treatment to disclose those policies upon request. Second, the law prohibits physicians from instituting a DNR order on a minor patient unless at least one parent has first been informed both orally and in writing. Third, in its stronger version, Simon’s Law prohibits physicians from instituting a DNR order if either parent refuses consent.
Other DNR consent laws — Similarly, but going beyond the pediatric context, in a special 2017 legislative session, Texas enacted a statute that requires clinicians to obtain patient or surrogate consent for DNR orders, except in one situation . That exception is when in the reasonable medical judgment of the patient’s attending physician, both 1) the patient’s death is imminent, regardless of the provision of cardiopulmonary resuscitation (CPR); and 2) the DNR order is medically appropriate. Still, even in that situation, clinicians may not write a DNR order if that would be “contrary to the directions of a patient who was competent at the time the patient conveyed the directions.”
CLINICAL SCENARIOS IN PALLIATIVE CARE SETTINGS
Prematurity — The limit of viability is defined as the stage of fetal maturity that ensures a reasonable chance of survival. Determining the limit of viability is desirable so that futile, costly, and painful interventions can be avoided if an infant cannot survive. However, deciding upon a threshold of viability is challenging because it remains uncertain which extremely preterm infant has a reasonable chance of survival.
Factors that affect survival rates for extremely premature infants (gestational age below 24 weeks) include gestational age, birth weight, sex, multiple gestations, and the use of antenatal corticosteroid therapy to hasten lung maturation in the fetus.
Gestational age is the major factor in determining viability (table 1). It is generally accepted in 2022 that babies born at <22 weeks of gestational age cannot survive and thus that life-supporting treatment for such babies is futile. Survival rates are improving for babies born at 22 weeks . Furthermore, the margin of error in prenatal estimates of gestational age is one to two weeks.
Every facility that cares for high-risk pregnancies should have a consensus approach to the management of extremely premature infants that is consistently used by all providers. As of 2022, in the United States, many such facilities are revisiting their policies about the borderline of viability as evidence accumulates that babies born at 22 weeks can survive and thrive . Futility issues surrounding critically ill newborns were specifically addressed by the Presidential Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research . Several professional organizations have issued recommendations in this area, including the American Academy of Pediatrics. These issues are discussed in greater detail separately. (See "Periviable birth (limit of viability)" and "Preterm birth: Definitions of prematurity, epidemiology, and risk factors for infant mortality".)
Chromosomal abnormalities — In the past, certain chromosomal syndromes, such as trisomy 13 and 18, were thought to be incompatible with life and treatment was thought be futile. However, some parents successfully insisted on treatment, and it often led to better survival rates. This illustrates the worrisome phenomenon by which determinations of futility can become self-fulfilling prophecies. When patients are not treated because treatment is thought to be futile, their deaths can be interpreted as confirmation of the correctness of the futility assessment. Published data on low survival rates can then be used to “prove” the lethality of the disease.
We now know that, for babies with these trisomy syndromes, survival rates beyond infancy are approximately 20 percent . Life-supporting treatment, including surgery, can no longer be considered futile for such babies. Furthermore, many families report that their infants’ quality of life is better than that reported in much of the medical literature . Clinical decisions for such babies must be made on a case-by-case basis using the model of shared decision-making. Doctors and parents should consider not just the chromosomal diagnosis but should base decisions on the baby’s complete clinical situation to decide whether treatment is appropriate . This controversial area is discussed in detail separately. (See "Congenital cytogenetic abnormalities", section on 'Trisomy 18 syndrome' and "Congenital cytogenetic abnormalities", section on 'Trisomy 13 syndrome'.)
Persistent vegetative state — Resuscitation from cardiac arrest or other cardiopulmonary catastrophe may be complicated by hypoxic ischemic brain injury. Obtundation or coma is frequent early on, followed by recovery or evolution to brain death or persistent vegetative state (PVS). Based on available clinical data, PVS is judged to be permanent after three months if induced non-traumatically. (See "Hypoxic-ischemic brain injury in adults: Evaluation and prognosis", section on 'Prognosis assessed on clinical features'.)
PVS, increasingly referred to as unresponsive wakefulness syndrome , results in complete unawareness of oneself and one’s environment. Patients are incapable of interacting with others; they have no sustained, reproducible, purposeful, or voluntary behavioral responses to stimuli; and they have no evidence of language comprehension or expression. They have the presence of sleep-wake cycles with the preservation of hypothalamic and brainstem function. Because the patient is unaware of themself or their environment, they are not able to suffer or experience pain, despite what may appear as grimaces, crying, or other expressions of discomfort. The perception of pain and suffering is a conscious experience governed by the cerebral cortex, while the expression of pain may be elicited at any level of the nervous system, including the motor/behavioral, endocrinologic, and autonomic responses that may occur as reflexes in the absence of consciousness. (See "Hypoxic-ischemic brain injury in adults: Evaluation and prognosis", section on 'Family counseling'.)
While aggressive treatments for individuals with PVS would not qualify as physiologically futile, as treatment might be able to maintain cardiopulmonary function, they may be considered qualitatively futile because of the lack of benefit in terms of quality of life. Common ethical challenges in these patients revolve around the pursuit of surgically placed feeding tubes and tracheotomies with chronic ventilator support. While the patient may not be considered to either suffer or benefit from any of these aggressive treatments because they are in a PVS, there is clearly an impact on the family, health care team, and society. (See "Ethics in the intensive care unit: Responding to requests for potentially inappropriate therapies in adults" and "Withholding and withdrawing ventilatory support in adults in the intensive care unit" and "Gastrostomy tubes: Uses, patient selection, and efficacy in adults".)
Decision-making regarding withdrawal of various levels of treatment is dependent on accurate prognostication (see "Hypoxic-ischemic brain injury in adults: Evaluation and prognosis", section on 'Prognosis based on clinical findings'). Once this is established, discussions with surrogate decision-makers should focus on the patient’s known preferences or advance directives, ie, whether ongoing life-supporting treatments are consistent with their prior stated wishes. However, these decisions can be very difficult. Some families, given time and information, can agree that pursuit of these treatments is not beneficial. However, some families and patients possess religious or cultural beliefs that support continuation of life even if the patient is unaware . Disagreements between providers and families are common in these emotionally and spiritually charged circumstances. Most disagreements can be resolved, however, when clinicians and families meet and talk and consider the patient’s clinical situation and previously expressed wishes . Some doctors, patients, and families find decision aids helpful.
Brain death (death determined by neurologic criteria) — Brain death (death by neurologic criteria) is defined as the irreversible cessation of cerebral and brainstem function. There is no respiratory drive, and thus there are no spontaneous breaths regardless of hypercarbia or hypoxemia. There are no responses arising from the brain (including cranial nerve reflexes and motor responses) to stimuli, although spinal reflexes may persist. One is legally dead in the United States when neurologic criteria for death have been demonstrated. However, vital organ function may be maintained in these patients, sometimes for long periods, through artificial ventilatory support. Requirements for the declaration of brain death are discussed in detail elsewhere. (See "Diagnosis of brain death".)
The diagnosis of death using neurologic criteria is an area where, until recently, the meaning of futility was very clear. The concept first came into the literature in the 1950s, and in 1968 the Harvard criteria for brain death were published, putting forth an agreed-upon definition of the term and recommendations for its use . The committee stated: “Death is to be declared and then the respirator turned off. The decision to do this and the responsibility for it are to be taken by the physician-in-charge, in consultation with one or more physicians who have been directly involved in the case. It is unsound and undesirable to force the family to make the decision.” Therefore, since the conception of this definition of death, it was deemed appropriate for clinicians to stop medical intervention following the declaration of death by neurologic criteria. (See "Diagnosis of brain death".)
Since then, declaring death by neurologic criteria has legally been accepted as a means of declaring death (the other being cardiopulmonary death) in all United States jurisdictions. While cases have been brought to court challenging a clinician’s ability to declare a patient dead by these criteria and to stop futile medical intervention, courts have generally honored the legal acceptance of brain death. One narrow exception is New Jersey which allows for religious exemptions among adherents to religions that do not accept neurologic criteria for defining death. Three other states (New York, California, Illinois) require brief reasonable accommodation .
Controversies about legal definitions of death led to a widely publicized controversy in the case of Jahi McMath . Jahi was a 13-year-old girl who suffered a cardiopulmonary arrest in the wake of surgery for sleep apnea with an anoxic brain injury that progressed to death (as defined by neurologic criteria). She remained on mechanical ventilation with stable heart rate and blood pressure. Following the declaration of death, her family requested and was granted a restraining order preventing the hospital from disconnecting Jahi from the ventilator. They argued that their religious beliefs did not accept brain death. Although the diagnosis of death by neurologic criteria was confirmed by consulting clinicians, the court order was extended while the McMath family sought and found an accepting health care facility in New Jersey. She subsequently received a surgically placed feeding tube and was maintained through a tracheostomy on a ventilator for five years. Those who supported the family stated that only patients and their surrogates should have the right to decide when medical treatment stops, not medical personnel .
Since the Jahi McMath case, other cases (mostly pediatric) have also reached United States, Canadian, and British courts. These cases raise three questions [84,87]. First, must clinicians obtain family consent before administering brain death testing ? Second, must clinicians accommodate religious objections to determination of death on neurologic criteria ? Third, do accepted medical criteria for determining brain death comply with the legal standards in the 1981 Uniform Determination of Death Act (UDDA)? The Uniform Law Commission is drafting a revised UDDA to address these problems.
Extracorporeal membrane oxygenation — The use of extracorporeal membrane oxygen (ECMO) raises unique issues because many centers have limited supplies of ECMO equipment and therefore must sometimes make decisions based on concerns about the allocation of scarce resources. This was especially true during the COVID-19 pandemic.
EMCO can be utilized for patients with severe cardiac and/or pulmonary failure when other standard treatments, such as vasoactive support and mechanical ventilation, have not reversed the patient’s decline. In the 1980s and 1990s, it was predominantly used for neonates. In the 2000s, studies showed its benefits for adults [90,91]. It is now more frequently used for pediatric and adult patients as a bridge to recovery from a life-threatening disease or as a bridge to transplantation.
Although ECMO may be viewed as “life-saving” for some, it is a very complex and resource-intensive intervention that introduces new significant risks to the patient. Patients being considered for ECMO are frequently in a rapid state of decline with developing multiorgan dysfunction. It is challenging to prognosticate who will or will not benefit from its use and thus when ECMO will be beneficial or futile . ECMO can only be delivered in an intensive care unit, so it is different from other life-support technologies such as dialysis or mechanical ventilation that can be provided to patients at home.
The ethical dilemmas surrounding ECMO became more complex during the COVID-19 pandemic. With initial absence of knowledge on the progression of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), it was unclear whether the use of ECMO was futile. The additional burden of significant resource limitations and high demand added to decision-making complexity. While one study showed improved survival with ECMO for patients with COVID-19 , many potentially eligible patients were not offered ECMO because they were deemed too sick and ECMO was thought to be futile. Thus, we still do not have robust data showing which COVID-19 patients are unlikely to benefit from ECMO .
Slow, show, and short codes — In the 1970s and 1980s, the practice of unilateral, undisclosed do-not-resuscitate (DNR) orders and “slow codes” was discovered. Clinicians were independently deciding either not to perform cardiopulmonary resuscitation (CPR) or to deliberately provide inadequate or half-hearted CPR (a “slow code”), knowing that it would not be effective [1,95]. This practice has been deemed unethical , and most bioethicists and professional medical societies condemn the practice. The American College of Physicians Ethics Manual states, “Because it is deceptive, physicians or nurses should not perform half-hearted resuscitation efforts” .
Some distinguish three related concepts: “slow codes,” which refer to how timely CPR is administered; “show codes” or “Hollywood codes,” which refer to inadequate or half-hearted CPR; and “short codes,” which refer to the duration of CPR. While slow codes and show codes are condemned, the potential role of a “short code” continues to be debated .
When a family member refuses to consent to a DNR order in the instance of a patient with certain imminent mortality, the clinician is faced with either placing a unilateral DNR order or performing futile CPR . If it is known that the CPR will be futile, it may be argued that adjusting the length or extent of the CPR might meet the family’s need (seeing that everything has been done) while limiting the suffering of or battery to the patient . Short codes should not necessarily be considered morally equivalent to slow codes. It is the responsibility of the physician in charge of any resuscitation to determine when efforts to revive a patient should be discontinued. This could legitimately vary depending upon the circumstances that led to the need for resuscitation.
Geriatrics — Patients cannot be discriminated against because of their age. Therefore, issues surrounding futility determinations should be no different for older adult patients. However, one important issue that arises in older patients is the need to accurately assess decision-making capacity (see "Assessment of decision-making capacity in adults"). Increasingly, patients thought to be incapacitated can still participate in their own medical decision-making with supported decision-making .
But many older patients will lose decision making capacity. Therefore, clinicians should encourage them to make known their long-term wishes and appoint a surrogate, should they become incapacitated. End-of-life planning should be pursued by clinicians, giving time for patients to consider their wishes and values and to be as informed as possible about different medical interventions that could be pursued or refused. The fact that survival following out-of-home cardiac arrest decreases significantly with age should be disclosed when discussing DNR orders [100,101]. (See "Geriatric health maintenance", section on 'Advance directives and health care proxy' and "Advance care planning and advance directives" and "Medical care in skilled nursing facilities (SNFs) in the United States", section on 'Advance care planning' and "Hospital management of older adults", section on 'Advance directives and goals of care'.)
PROSPECTIVE DECISION MAKING FOR INCAPACITATED PATIENTS
Advance directives — Deciding upon issues of medical futility can be challenging enough when a patient is able to communicate their own goals and values. It is even more challenging when there is speculation upon what the patient would wish. Advance directives (ADs) are the documents a person completes while still in possession of decisional capacity about how treatment decisions should be made on their behalf in the event they lose the capacity to make such decisions. They are legal tools meant to direct treatment decision-making and/or appoint surrogate decision-makers. They have been associated with decreased utilization life-sustaining treatment and greater use of hospice and palliative care . (See "Advance care planning and advance directives", section on 'Advance directives'.)
However, it is not clear how much impact ADs would have on futility issues should they be more broadly adopted. One aspect is that physicians may not be aware of the presence of an AD . Another problem is that disease progression can be unanticipated and prognosis may change over time. The AD, however, may remain stagnant and not accurately reflect the wishes of the patient at the moment when it is needed . Concerns from clinicians that an AD may not always reflect the patient’s wishes in the exact clinical scenario often limit their utility. Still, it is generally better to have some understanding of the patient’s values and wishes than to have none, and if an AD is available, the physician should consider its contents, along with other factors, in making clinical decisions about life-sustaining treatment. (See "Legal aspects in palliative and end-of-life care in the United States" and "Advance care planning and advance directives".)
Provider Orders for Life-Sustaining Treatment (POLST) — While all adults should have an advance directive, those who are seriously ill should supplement their advance directive with a POLST. POLST forms are completed by health care professionals after discussion with the patient or their surrogate and are immediately actionable and transportable. As medical orders, POLST forms are followed by emergency medical service personnel in times of crisis and by treating health care professionals once the patient has been transported to a health care facility. In short, seriously ill patients can reduce the chance of unwanted treatments by completing a POLST with their health care professional . (See "Advance care planning and advance directives", section on 'Physician Orders for Life-Sustaining Treatment'.)
Do-not-escalate-treatment orders — Some surrogate decision-makers remain uncomfortable with being asked to withdraw or withhold treatments but may understand that certain limits of therapeutic intervention have been reached. As an example, if a patient is already on high doses of continuous inotropic support, the care provider can inform the family that increasing the medications beyond the current doses is unlikely to provide physiologic benefit and should not be pursued. Reframing the conversation to identifying where limits should be set, rather than asking for agreement to stop treatments altogether, may help to resolve conflicts.
For families, complying with a physician’s recommendation on these limits may feel less burdensome than making the decision to withdraw life-sustaining support. Some families may verbalize that withdrawing support feels like they are actively killing their family member. But placing limits to escalation can diffuse some of that burden. Do-not-escalate-treatment (DNET) orders may prolong the dying process as compared with immediate withdrawal of support, providing the family more time with their loved one. However, it may also increase the possibility of pain or suffering in the dying patient, and therefore the order should also include plans on what will be offered, including what medications will be provided to alleviate any suffering .
COMMUNICATION AND RESOLVING CONFLICT — Issues that arise surrounding futility may be indicative of a breakdown in communication and trust. Maintaining excellent communication between health care providers and patients/family members seems paramount in resolving conflict and reaching joint decisions. (See "Communication in the ICU: Holding a meeting with families and caregivers", section on 'Importance of communication'.)
However, good communication alone is not sufficient to avoid conflicts that arise about medically futile or inappropriate/inadvisable therapies, as was shown in the SUPPORT study . The study had a prospective observational (phase I) component that followed outcomes in 4301 adults hospitalized with one of nine life-threatening diagnoses, then a randomized controlled clinical trial (phase II) component in which another 4804 patients hospitalized with one of nine life-threatening illnesses were randomly assigned to either an intervention or a control group. The intervention included providing the patients’ clinicians with frequent prognostic data, obtaining and updating the chart with patient and family care preferences, and providing skilled nursing to organize and help conduct family meetings. The phase I observational study revealed significant shortcomings in communication, as well as knowledge of patients’ wishes regarding cardiopulmonary resuscitation (CPR), and an overall high frequency of aggressive treatments. The hope was that the improved availability of prognostic information and communication with patients and families would improve these shortcomings. However, the investigators found that the intervention failed to improve care or patient outcomes, and it did not reduce use of hospital resources.
Fins and Solomon, in their studies of communication to resolve futility disputes, found that, in addition to good communication, other needed elements were clinicians’ self-awareness, early recognition of impending disputes, psychological insight into the cause of the dispute, and an institutional culture that promotes good communication . Some physicians may not possess these qualities and may benefit from referral to or assistance from colleagues who better exemplify these characteristics; others may benefit from assistance from trained mediators. The New England Journal of Medicine has created podcasts to help physicians learn better communication skills.
Strategies to avoid futility disputes — Prendergast outlined seven strategies to avoid futility disputes:
●Keep patients and families informed.
●Identify other staff members, such as other physicians or nurses with whom the patient or family has a better rapport who can assist in facilitating good patient relations.
●Promote realistic expectations.
●Strive for accuracy in prognosis.
●Maintain continuity of care.
●Be compassionate and flexible.
●Show firmness about limits and beware of making decisions based on economic market forces .
Goal formulation is also pivotal. As discussed in the definitions of futility, a treatment may be deemed futile if it does not achieve a specific goal. While we may assume that the patient’s or family’s goal is recovery or survival, sometimes it may be to have a few more days or weeks with their loved one, or to die knowing they “fought” until the end, or to ensure that there is minimal pain or suffering. We must inquire about what a patient would like to achieve, provide the realistic framework to shape their expectations, and hopefully arrive at a mutually shared goal of therapy . If an agreed-upon goal can be achieved, futility issues can be minimized.
Unfortunately, there are times when escalation of ethical concerns is unavoidable. Pragmatic steps can be taken to prospectively be prepared for these instances. Most hospitals have policies that specifically deal with these situations.
Hospital futility policies — As listed above, many major professional associations have advised that all hospitals develop formal policies to address futile and potentially inappropriate/inadvisable therapies [53,110]. (See 'Major policy statements' above.)
The Ethics Committee of the Society of Critical Care Medicine (SCCM) listed five criteria for these polices : “a) be disclosed in the public record; b) reflect moral values acceptable to the community; c) not be based exclusively on prognostic scoring systems; d) articulate appellate mechanisms; and e) be recognized by the courts.” It has also been suggested that hospitals, particularly those in close geographical proximity, work together to ensure uniformity and standardization, which will provide clarity for the general public . Some communities, such as San Diego and Houston, have taken it upon themselves to adopt futility policies, as discussed above. These types of community-driven policies can ensure that hospitals and clinicians are acting in a manner that is consistent with the community’s values and that members of the community may be educated on medical futility issues.
In some cases, it may also be appropriate for clinicians to make a recommendation about limiting treatment without specifically asking surrogates to explicitly verbalize agreement. This is referred to as the “nudge/assent model.” The patient may veto but is not asked for consent ex ante, only assent. The patient or surrogates’ “informed non-dissent” may be a way of accepting limits of therapeutic intervention without necessarily explicitly endorsing those limits [111,112]. This method may be particularly useful when the surrogate’s guilt keeps them from being able to make a decision but they are aware and able to acknowledge the patient is dying (as long as others make the decision).
Framework for clinicians to use in resolving conflict — Written policies should outline a clear, stepwise approach to negotiating disputes about futile and potentially inappropriate/inadvisable treatments. An algorithmic approach to managing disputed treatment requests in the intensive care unit, which is also applicable to other health care settings, is outlined in the algorithm (algorithm 2) . Recommended steps for clinicians to use to resolve conflict when a requested treatment is deemed potentially inappropriate are outlined in the table (table 2) .
Developing a clear process for deliberation that is transparent to health care professionals, patients, and families may alleviate some of the distress that accompanies these difficult situations and help obtain a viable solution.
Approach to patients who “want everything done” — Communication is paramount in situations where, when asked about setting limits on medical treatment in the face of a severe, life-threatening illness, patients and/or their families respond that they want to continue providing all available life-prolonging treatments until death has occurred. Although physicians might reasonably object to providing every imaginable treatment no matter how harsh, invasive, or unlikely to be of benefit and be tempted to tell this to patients and families, a more appropriate response after acknowledging the request would be to discuss the patient’s underlying treatment values and nonmedical concerns (table 3).
The idea of “doing everything” or “providing all available life-sustaining treatments” means different things to different people. The clinician must figure out what it means to the particular individuals involved in a particular case. Often, it can mean “do everything that has a reasonable chance of helping my loved one.” Once the meaning is clarified (table 4), the clinician must provide accurate information about the patient’s illness and prognosis and the treatment options that seem appropriate or inappropriate . With that information, the physician can make more informed recommendations, and patients can make better decisions about what should (and should not) be done to help them achieve their goals. An algorithmic approach to such patients is outlined in the algorithm (algorithm 3). A separate topic review on discussing goals of care is available. (See "Discussing goals of care".)
Time-limited trials — Given the difficulties with prognostication and a patient’s or family’s need for time and experience to feel truly informed and to come to terms with medical realities and grief, some have advocated for the use of time-limited therapeutic trials . Such trails may be beneficial in scenarios where treatment may be considered futile or inappropriate/inadvisable therapies and disagreement exists between parties. The goal of the trial should be determined and agreed upon between the health care team and the patient or surrogate. A specific timeframe should be determined that reflects the clinical situation, the treatments being proposed, and the nature of the underlying illness . Sometimes, a written document specifying the nature of the time-limited trial, and the endpoints to be used in determining whether the trial has or has not been successful, can be useful. During the trial, all parties will have the opportunity to get a better sense of prognosis and efficacy of treatment and to adjust to illness realities. If the goal of therapy is not achieved, all parties may be more confident in not pursuing further interventions [116,117].
Palliative care consultations — For non-palliative-care-trained clinicians struggling with communication and consensus with patients and families around end-of-life issues, obtaining a palliative care consult can benefit communication. Individuals trained specifically in death and dying can provide a different perspective to families and help families and patients understand that transition from pursuing aggressive medical interventions to pursuing control of pain and suffering is not “giving up” on a patient but is, instead, providing the best treatment when cure is not possible . (See "Benefits, services, and models of subspecialty palliative care" and "Palliative care: Issues in the intensive care unit in adults".)
Ethics consultations — Ethics committees and consultations have been discussed above, both as components of the Texas Advance Directives Act (TADA) and as part of the American Medical Association’s (AMA) recommendations for futility negotiations. Many articles support the involvement of ethics consultations prior to escalation of issues to that level . Asking ethics consultants to get involved earlier, particularly ones trained in conflict mediation, may help gain resolution earlier while maintaining positive relationships between all parties . (See "Ethics in the intensive care unit: Responding to requests for potentially inappropriate therapies in adults", section on 'Six-step process to conflict resolution'.)
SUMMARY AND RECOMMENDATIONS
●Conflict regarding futility – Controversies about medical futility arise when doctors and patients (or family members) disagree about whether a particular treatment is futile or inappropriate/inadvisable in a particular circumstance. (See 'Introduction' above.)
●Definitions – In 2015, five critical care societies, the American Thoracic Society (ATS), the Association of Critical-Care Nurses (AACN), the American College of Chest Physicians (ACCP), the European Society of Intensive Care Medicine (ESICM), and the Society of Critical Care Medicine (SCCM), officially endorsed new and specific terminology that should be used when discussing and categorizing treatments of questionable benefit  (see 'Definitions of futility' above):
•The five-society statement maintains that the term “futile” should be used only in the rare circumstance that an intervention definitely cannot accomplish the intended physiological goal. (See 'Physiological futility' above.)
•The term “potentially inappropriate” should be used, rather than “futile,” to describe the more common situation in which treatments have at least some chance of accomplishing the effect sought by the patient, but clinicians believe that competing ethical considerations justify not providing them. In contrast to “futile” treatment, disagreements over a “potentially inappropriate” treatment are value based. (See 'Potentially inappropriate treatment' above.)
•“Legally proscribed” treatments are those that are prohibited by applicable laws, judicial precedent, or widely accepted public policies. “Legally discretionary” treatments are those for which there are specific laws, judicial precedent, or policies that give physicians permission to refuse to administer them. (See 'Proscribed and discretionary treatment' above.)
●Prospective decision making
•Having an advance directive (AD) clearly decreases the use of life-sustaining treatment and increases the use of hospice or palliative care. However, ADs can be vague and may not always reflect the patient’s wishes in the exact clinical scenario that they face. This limits their utility. (See 'Advance directives' above.)
•Seriously ill patients can materially reduce the chance of unwanted treatments by completing a Provider Orders for Life-Sustaining Treatments (POLST) form. POLST forms are immediately actionable and transportable. (See 'Provider Orders for Life-Sustaining Treatment (POLST)' above.)
•Most disagreements about futile or potentially inappropriate treatments can be resolved with time and continued efforts to come to consensus. Some surrogate decision-makers remain uncomfortable with being asked to withdraw or withhold treatments but may understand that certain limits of therapeutic intervention have been reached. Reframing the conversation to identifying where limits should be set (ie, do-not-escalate-treatment [DNET] orders), rather than asking for agreement to stop treatments altogether, may help to resolve conflicts. (See 'Do-not-escalate-treatment orders' above.)
●Policy statements and state laws – Policy statements on dealing with futile and potentially inappropriate therapies are available from several groups. In addition, several states have passed laws mandating consent for withdrawal of treatment and DNR orders. (See 'Major policy statements' above and 'Notable US state laws' above.)
●Approaches to resolving conflict
•An algorithmic approach to managing disputed treatment requests in the intensive care unit, as recommended by five joint critical care societies, is provided in the algorithm (algorithm 2); this approach may also apply to other health care settings. (See 'ATS/AACN/ACCP/ESICM/SCCM (2015)' above.)
•A framework approach for clinicians to use in resolving conflict regarding potentially inappropriate treatments, also as suggested by the five joint critical care societies, is provided in the table (table 2). (See 'Framework for clinicians to use in resolving conflict' above.)
•A time-limited therapeutic trial may be beneficial in scenarios where treatment may be considered futile or inappropriate/inadvisable and disagreement exists between parties. The goal of the trial should be determined and agreed upon between the health care team and the patient or surrogate prior to the trial. (See 'Time-limited trials' above.)
•If conflicts over the aggressiveness of care cannot be resolved, then the courts can be a last resort. Ethics committees and consultations should be sought prior to escalation of issues to that level. Asking ethics consultants to get involved earlier, particularly ones trained in conflict mediation, may help gain resolution earlier while maintaining positive relationships between all parties. (See 'Ethics consultations' above.)
7 : Bioethics and the law: the case of Helga Wanglie: a clash at the bedside -- medically futile treatment v. patient autonomy.
8 : Bioethics and the law: the case of Helga Wanglie: a clash at the bedside -- medically futile treatment v. patient autonomy.
15 : A novel mutation of the RRM2B gene in an infant with early fatal encephalomyopathy, central hypomyelination, and tubulopathy.
16 : A novel mutation of the RRM2B gene in an infant with early fatal encephalomyopathy, central hypomyelination, and tubulopathy.
17 : A novel mutation of the RRM2B gene in an infant with early fatal encephalomyopathy, central hypomyelination, and tubulopathy.
18 : A novel mutation of the RRM2B gene in an infant with early fatal encephalomyopathy, central hypomyelination, and tubulopathy.
19 : A novel mutation of the RRM2B gene in an infant with early fatal encephalomyopathy, central hypomyelination, and tubulopathy.
21 : Institutional policies on determination of medically inappropriate interventions: use in five pediatric patients.
22 : Institutional policies on determination of medically inappropriate interventions: use in five pediatric patients.
23 : Institutional policies on determination of medically inappropriate interventions: use in five pediatric patients.
24 : Institutional policies on determination of medically inappropriate interventions: use in five pediatric patients.
25 : Institutional policies on determination of medically inappropriate interventions: use in five pediatric patients.
26 : Institutional policies on determination of medically inappropriate interventions: use in five pediatric patients.
30 : Differences between attendings' and fellows' perceptions of futile treatment in the intensive care unit at one academic health center: implications for training.
34 : Can physicians' judgments of futility be accepted by patients? A comparative survey of Japanese physicians and laypeople.
39 : Defining Futile and Potentially Inappropriate Interventions: A Policy Statement From the Society of Critical Care Medicine Ethics Committee.
40 : Defining Futile and Potentially Inappropriate Interventions: A Policy Statement From the Society of Critical Care Medicine Ethics Committee.
41 : Defining Futile and Potentially Inappropriate Interventions: A Policy Statement From the Society of Critical Care Medicine Ethics Committee.
42 : An Official ATS/AACN/ACCP/ESICM/SCCM Policy Statement: Responding to Requests for Potentially Inappropriate Treatments in Intensive Care Units.
55 : An official American Thoracic Society policy statement: managing conscientious objections in intensive care medicine.
56 : ICU Admission, Discharge, and Triage Guidelines: A Framework to Enhance Clinical Operations, Development of Institutional Policies, and Further Research.
58 : Protection of human subjects; reports of the President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research--Office of the Assistant Secretary for Health, HHS. Notice of availability of reports
59 : Protection of human subjects; reports of the President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research--Office of the Assistant Secretary for Health, HHS. Notice of availability of reports
66 : Procedural due process and intramural hospital dispute resolution mechanisms: the Texas Advance Directives Act.
69 : Counterpoint: The Texas advance directives act is ethically flawed: medical futility disputes must be resolved by a fair process.
70 : Counterpoint: The Texas advance directives act is ethically flawed: medical futility disputes must be resolved by a fair process.
71 : Counterpoint: The Texas advance directives act is ethically flawed: medical futility disputes must be resolved by a fair process.
72 : Counterpoint: The Texas advance directives act is ethically flawed: medical futility disputes must be resolved by a fair process.
73 : Counterpoint: The Texas advance directives act is ethically flawed: medical futility disputes must be resolved by a fair process.
74 : Counterpoint: The Texas advance directives act is ethically flawed: medical futility disputes must be resolved by a fair process.
75 : Counterpoint: The Texas advance directives act is ethically flawed: medical futility disputes must be resolved by a fair process.
77 : Assessment of an Updated Neonatal Research Network Extremely Preterm Birth Outcome Model in the Vermont Oxford Network.
80 : Attitudes of neonatologists toward delivery room management of confirmed trisomy 18: potential factors influencing a changing dynamic.
82 : Toward shared decision making at the end of life in intensive care units: opportunities for improvement.
83 : A definition of irreversible coma. Report of the Ad Hoc Committee of the Harvard Medical School to Examine the Definition of Brain Death.
87 : Determination of Death by Neurologic Criteria in the United States: The Case for Revising the Uniform Determination of Death Act.
88 : Legal and Ethical Considerations for Requiring Consent for Apnea Testing in Brain Death Determination.
89 : Frequency of use of the religious exemption in New Jersey cases of determination of brain death.
90 : Efficacy and economic assessment of conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multicentre randomised controlled trial.
93 : Survival benefit of extracorporeal membrane oxygenation in severe COVID-19: a multi-centre-matched cohort study.
103 : Advance directives for seriously ill hospitalized patients: effectiveness with the patient self-determination act and the SUPPORT intervention. SUPPORT Investigators. Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatment.
104 : Association between Physician Orders for Life-Sustaining Treatment for Scope of Treatment and in-hospital death in Oregon.
105 : Easing the burden of surrogate decision making: the role of a do-not-escalate-treatment order.
106 : A controlled trial to improve care for seriously ill hospitalized patients. The study to understand prognoses and preferences for outcomes and risks of treatments (SUPPORT). The SUPPORT Principal Investigators.
109 : The Goals of Patient Care project: implementing a proactive approach to patient-centred decision-making.
110 : Consensus statement of the Society of Critical Care Medicine's Ethics Committee regarding futile and other possibly inadvisable treatments.
112 : Informed non-dissent: a better option than slow codes when families cannot bear to say "let her die".
114 : Time-Limited Trials Among Critically Ill Patients With Advanced Medical Illnesses to Reduce Nonbeneficial Intensive Care Unit Treatments: Protocol for a Multicenter Quality Improvement Study.
115 : Time-Limited Trials of Intensive Care for Critically Ill Patients With Cancer: How Long Is Long Enough?
116 : Beyond the futility argument: the fair process approach and time-limited trials for managing dialysis conflict.