Do not resuscitate orders sit at the intersection of nearly every major tension in medical ethics: a patient’s right to choose versus a physician’s duty to preserve life, the hope of families versus the realities of what CPR can accomplish, and the ideal of individualized care versus the pressures of overburdened health systems. What makes DNR orders so ethically fraught is not any single issue but the way these tensions compound one another, often in high-stakes moments when time, information, and emotional clarity are all in short supply.
What CPR Actually Does and Why That Matters
Most ethical dilemmas around DNR orders begin with a misunderstanding about what resuscitation involves and how often it works. CPR on television succeeds far more often than CPR in a hospital, and the gap is enormous. A survey of public expectations found that about 96 percent of respondents held unrealistically optimistic views of CPR’s effectiveness, regardless of where they got their information.1PubMed. Public expectations of survival following cardiopulmonary resuscitation That mismatch shapes everything downstream. If you believe CPR is likely to bring someone back to a meaningful quality of life, declining it feels like choosing death. If you understand that survival rates after in-hospital cardiac arrest are considerably lower than popular culture suggests, and that neurological outcomes can be poor even when a heartbeat is restored, the decision looks very different.
This perception gap creates real ethical problems. When families or patients are asked whether they want “everything done,” many say yes based on an image of resuscitation that does not match clinical reality. The consent they give is technically informed in a legal sense but practically uninformed in a medical one. Researchers have pointed out that this dynamic can lead people to elect CPR for themselves or family members in situations where recovery, not just survival, is unlikely.1PubMed. Public expectations of survival following cardiopulmonary resuscitation
The Consent Problem
For a DNR order to be ethically sound, the person making the decision needs to understand what they are agreeing to. In practice, that understanding is often missing. A study examining resuscitation preferences found that almost half of the people surveyed remained undecided about their own resuscitation wishes, and many lacked accurate knowledge about what DNR status means, including what resuscitation physically involves and what outcomes to expect.2Patient Education and Counseling. Associations with resuscitation choice: Do not resuscitate, full code or undecided That same research found that people who had spoken with a physician about resuscitation and who felt confident in understanding their own health were more likely to choose DNR status, while those whose impressions came primarily from media were more likely to remain undecided or choose full code.
The implication is uncomfortable but clear: a significant portion of both DNR decisions and full-code decisions are being made on shaky informational ground. The ethical ideal is that a DNR conversation happens early, calmly, and with honest information. The clinical reality is that these conversations often happen in crisis moments, in unfamiliar hospital rooms, with family members who are frightened and overwhelmed. When the conversation does not happen well, or does not happen at all, the resulting order may not reflect what the patient would actually want if they understood the situation.
When Someone Else Decides
Many DNR decisions are not made by the patient at all. People who are unconscious, cognitively impaired, or too ill to communicate rely on surrogates, usually family members, to decide on their behalf. This introduces a distinct ethical layer. Research involving terminally ill cancer patients and their families found that a common assumption underlying DNR discussions was that the decision amounted to choosing between life and death. In making that choice, patients and families were implicitly required to render moral judgments about the value of the patient’s remaining life, including the significance of their relationships with others.3PubMed. Choosing between life and death: patient and family perceptions of the decision not to resuscitate the terminally ill cancer patient
That is an extraordinary burden to place on a grieving spouse or adult child. Surrogates often do not know with certainty what the patient would want, and they carry guilt regardless of which direction they choose. Some feel they are “giving up” by agreeing to DNR; others fear they are prolonging suffering by refusing one. The ethical framework assumes surrogates act as faithful agents of the patient’s values, but the psychological reality is that their own values, fears, and relationships inevitably shape the decision.
Unilateral DNR Orders and Medical Futility
Perhaps the most contentious ethical question in this space is whether a physician can write a DNR order without the agreement of the patient or family. These unilateral orders are rare but real, and they tend to arise when the medical team believes CPR would be futile. In oncology and hematology settings, clinicians have described their guiding ethical values as avoiding harm and ensuring a peaceful, dignified death for dying patients.4PubMed Central. The ethics of DNR-decisions in oncology and hematology care: a qualitative study When a patient is actively dying and CPR would cause rib fractures and organ damage without any realistic chance of restoring heart function, clinicians can feel that performing it violates their own oath to do no harm.
The trouble is that “futility” is far easier to define in theory than in practice. There is broad agreement that CPR is futile when it has no physiological chance of working, but once you move beyond that narrow definition, the concept gets contested quickly. During the COVID-19 pandemic, some institutional guidelines suggested that CPR could be considered futile in a “significant portion” of elderly, critically ill patients with underlying conditions. Critics of those guidelines argued that this language went well beyond the small number of truly uncontroversial cases and risked applying futility judgments too broadly.5Journal of Law, Medicine & Ethics. The Ethics of Unilateral Do-Not-Resuscitate Orders for COVID-19 Patients The worry is that once futility becomes a basis for overriding patient or family wishes, the line between genuine medical judgment and resource-driven rationing blurs.
The Ripple Effect on Overall Care
One of the less obvious ethical concerns about DNR orders is their tendency to affect more than just the resuscitation decision. A DNR order technically applies only to cardiac arrest: it says that if your heart stops, the team will not attempt to restart it. It does not say anything about antibiotics, blood transfusions, dialysis, or any other treatment. But in practice, research has documented what has been called a “ripple effect,” where the presence of a DNR order can reduce the overall aggressiveness of care a patient receives. As illustrated in the case of intracerebral hemorrhage, the culture surrounding DNR orders could influence treatment intensity even for patients who do not have one.6PubMed Central. Do-not-resuscitate orders, unintended consequences, and the ripple effect
This is a serious ethical problem because it means a patient who agrees to forgo resuscitation may inadvertently receive less aggressive treatment for treatable conditions. A DNR order is supposed to be one specific instruction, not a signal to “wind down” all care. When clinicians, consciously or unconsciously, interpret it as the latter, patients can be harmed in ways they never consented to.
Racial and Ethnic Disparities
DNR decisions do not occur in a vacuum. They take place within health systems that have long-standing disparities in how different communities are treated, and those disparities show up clearly in end-of-life care. A large study of terminal hospitalizations in New York State found that Black patients were less likely than non-Hispanic White patients to have DNR status, with adjusted odds roughly 9 percent lower, while Hispanic patients were slightly more likely to have DNR status.7PubMed Central. Disparities in End-of-Life Care for Minoritized Racial and Ethnic Patients During Terminal Hospitalizations in New York State Research in surgical intensive care settings found an even starker gap: after adjusting for clinical factors, Black patients were roughly half as likely as White patients to be DNR at the time of death.8Trauma Surgery & Acute Care Open. Racial disparities in end-of-life suffering within surgical intensive care units
The reasons behind these patterns are complex and layered. Historical mistreatment by medical institutions has bred justified mistrust in some communities, making families less willing to agree to the withdrawal of any intervention. Cultural and religious values play a role. And systemic differences in access to palliative care consultations mean that some patients simply never have the conversation in the first place. The ethical concern is that DNR patterns should reflect individual patient values and medical circumstances, not structural inequities in who gets offered what kind of care and how those conversations are conducted.
Religious and Cultural Dimensions
For many people, the question of whether to accept a DNR order is inseparable from their spiritual beliefs about life, death, and the body. Research has shown that certain religious practices strongly predict the belief that DNR decisions are morally wrong, including frequent meditation, frequent contemplation of God, and endorsement of the idea that one’s faith restricts personal action.9PubMed. Effects of religiosity on patients’ perceptions of do-not-resuscitate status In traditions where life is understood as a gift from God that humans have no right to relinquish, agreeing to DNR status can feel like a violation of sacred duty.
A study of intensive care physicians in Palestine found that roughly 73 percent said religion affected their views on DNR orders, and about 70 percent said culture influenced their decision-making.10Indian Journal of Palliative Care. Intensive Care Unit Physician’s Attitudes on Do Not Resuscitate Order in Palestine Research from Taiwan found that patients with Buddhist or Daoist religious backgrounds were significantly less likely to have a DNR order during their surgical ICU stay.11PubMed Central. The Associations Between the Religious Background, Social Supports, and Do-Not-Resuscitate Orders in Taiwan: An Observational Study These findings underscore that what looks like a straightforward medical decision from the outside may be a deeply loaded moral and spiritual question for the patient and family involved. Ethical practice requires acknowledging that reality, not dismissing it.
Dementia and the Problem of Changing Selves
Advance directives are supposed to solve the problem of what happens when a patient can no longer speak for themselves. But dementia introduces a philosophical puzzle that advance directives were never designed to handle. A person who writes a DNR directive while cognitively healthy may, years later, be living with advanced dementia and showing no apparent desire to die. The “now” patient can have very different sensibilities than the “then” patient who wrote the directive, and clinicians lack clear means of following a directive to die rather than to live with advanced dementia.12The American Journal of Medicine. The Limits of Advance Directives in Maintaining Autonomy in Patients with Advanced Dementia
The ethical tension deepens when a person with dementia actively expresses wishes that contradict their earlier directive. If someone signed a DNR years ago but now, in their diminished state, seems to want treatment, whose wishes count? Ethicists have struggled with this question, and healthcare workers and families face it in real time. The debate centers on whether “precedent autonomy,” the idea that your earlier, competent self can bind your later, impaired self, holds moral weight when the later self appears to be a meaningfully different person.13Frontiers in Psychiatry. That was just your life: rethinking dementia for advance directives There is no consensus, and the practical result is that clinicians often find themselves navigating case by case, with little clear guidance.
Pediatric DNR Decisions
When the patient is a child, the ethical landscape shifts in fundamental ways. Children generally cannot make their own medical decisions, so parents serve as surrogates. But unlike with adult surrogates, who are asked to represent the patient’s known wishes, parents are making decisions for someone who has never had the chance to form or express preferences about death. The ethical considerations in pediatric DNR decisions stem from the special three-way relationship between physician, child, and parent, and from the legal and moral standing of parents as decision-makers.14Annals of Emergency Medicine. Pediatric DNR considerations
When parents and the medical team disagree, the situation becomes especially painful. A family that insists on full resuscitation for a child with a terminal illness may be making a decision that clinicians view as prolonging suffering. Yet overriding parental wishes through a unilateral DNR order is an extreme step. Ethicists have argued that such unilateral orders in pediatric care should be reserved for very limited circumstances, and that the emphasis should instead be on open discussion, shared decision-making, and preserving the clinician-parent relationship, while also supporting clinicians who feel moral distress about parental decisions.15PubMed Central. Is There Ever a Role for the Unilateral Do Not Attempt Resuscitation Order in Pediatric Care?
Suicide Attempts and the Autonomy Paradox
Few scenarios expose the limits of DNR ethics more starkly than a suicide attempt by someone with a valid DNR order. The standard reasoning behind DNR orders rests on autonomy: a competent person has the right to refuse medical intervention. But suicide is widely regarded in medicine as something that calls autonomy into question, since it often occurs in the context of treatable psychiatric illness. When a person who has a DNR for legitimate medical reasons then attempts suicide, emergency physicians face a direct collision between two ethical imperatives.16PubMed. Suicidal Patients with a Do-Not-Resuscitate Order
Some clinicians believe that a suicide attempt invalidates the assumption of rational decision-making that underlies any DNR order, and therefore all suicide attempts should be treated regardless of pre-existing directives. Others argue that overriding a DNR in this context violates the patient’s autonomy, especially if the patient had a terminal illness and the suicide was a foreseeable extension of their desire not to prolong suffering. The ethical literature frames this as a genuine dilemma between upholding the DNR and risking a preventable death, or suspending the DNR and risking a violation of the patient’s self-determination.17PubMed. Do-Not-Resuscitate Orders After Suicide Attempts: To Uphold or Suspend? Most emergency departments default to resuscitation in these cases, but the ethical tension remains unresolved.
Moral Distress Among Clinicians
The ethical weight of DNR orders does not fall only on patients and families. Nurses and physicians who carry out these orders, or who are asked to perform CPR they believe is futile, experience real psychological consequences. Qualitative research with critical care nurses found that participating in DNR execution was described as depressing and morally distressing. Nurses reported feeling as though they could do something but were not allowed to, or felt guilty because “no one wants to die,” or experienced conflict between DNR orders and their personal beliefs and values.18PubMed Central. Moral distress among critical care nurses when executing do-not-resuscitate (DNR) orders in a public critical care unit in Gauteng
The distress runs in both directions. Clinicians also suffer when they are required to perform CPR on patients for whom resuscitation has no realistic chance of benefit. Scholars have argued that moral harms arise from performing CPR on a person who is effectively already dead, and that the legal fears driving such “defensive” resuscitations are largely unfounded.19PubMed. Raising the Dead? Limits of CPR and Harms of Defensive Practices In many hospitals across multiple countries, CPR remains the default for all patients in cardiac arrest unless a DNR order exists, regardless of whether CPR is medically appropriate.20PubMed Central. Slow Codes are symptomatic of ethically and legally inappropriate CPR policies This default-to-resuscitate framework puts clinicians in the position of performing interventions they know are harmful, simply because the paperwork to decline them was never completed.
Legal Fears and How They Distort Care
Fear of legal liability shapes physician behavior around DNR orders in ways that do not always serve patients. A majority of physicians surveyed, about 52 percent, believed that the legal risk of keeping someone alive against their wishes was lower than the risk of failing to resuscitate.21PubMed Central. Physician perspectives and compliance with patient advance directives: the role external factors play on physician decision making In other words, when in doubt, doctors tend to err on the side of doing more rather than less, not because more is better for the patient, but because it feels legally safer. This creates a systematic bias toward overtreatment at the end of life, one that runs counter to the wishes of many patients who have explicitly asked not to be resuscitated.
The legal landscape itself adds to the confusion. In the prehospital setting, EMS providers face significant variation in what counts as acceptable DNR documentation from state to state, leading to situations where a patient’s wishes may not be honored simply because the paperwork does not match local protocol requirements.22PubMed. Do not resuscitate (DNR) emergency medical services (EMS) protocol variation in the United States Documents like living wills, which were not specifically designed for emergency use, can create interpretation challenges for paramedics who need to make split-second decisions.23ResearchGate. Variation of Do Not Resuscitate (DNR) Documentation in Statewide EMS Protocols Inside hospitals, confusion around DNR orders is also widespread. A survey found that more than 70 percent of respondents recalled situations in which confusion about a DNR order led to problems in patient care.24PubMed. Identification of inpatient DNR status: a safety hazard begging for standardization
POLST and the Attempt to Bridge the Gap
The Physician Orders for Life-Sustaining Treatment program, commonly known as POLST, was developed partly to address these communication failures. Unlike a standard advance directive, which is a legal document expressing general wishes, POLST is a set of actionable medical orders that travels with the patient across care settings. It was designed to ensure that treatment preferences are clearly communicated and honored whether the patient is at home, in an ambulance, or in a hospital.25PubMed. Use of the physician orders for life-sustaining treatment program in the clinical setting: a systematic review of the literature
Systematic reviews of POLST have shown that treatment limitations on the form are generally associated with less in-hospital death and less high-intensity treatment, particularly in prehospital settings. However, in acute care hospitals, a meaningful number of patients received care that did not match their POLST orders.26PubMed Central. The Influence of POLST on Treatment Intensity at the End of Life: A Systematic Review And there is a deeper concern: while POLST does alter treatment in ways consistent with what the form says, evidence that the form accurately reflects what the patient or surrogate actually wanted has been harder to establish.25PubMed. Use of the physician orders for life-sustaining treatment program in the clinical setting: a systematic review of the literature In other words, POLST may be very good at enforcing a documented order, but the question of whether that order truly captures the patient’s informed wishes loops back to the same consent problems that plague DNR discussions more broadly.
Crisis Standards and Group-Based Decisions
The COVID-19 pandemic forced many of these ethical tensions into the open, particularly around the question of whether DNR orders could be applied categorically rather than individually. Some hospitals and health systems considered or adopted policies that would have applied blanket DNR status to certain patient groups, such as the very elderly or those with specific comorbidities, as a way of managing scarce resources like ventilators and ICU beds. Ethicists pushed back hard, arguing that while prioritizing resources during a catastrophe is necessary, fairness requires basing decisions on clinical criteria applied to the individual rather than assumptions that favor or disfavor entire groups.27PubMed Central. Universal Do-Not-Resuscitate Orders, Social Worth, and Life-Years: Opposing Discriminatory Approaches to the Allocation of Resources During the COVID-19 Pandemic and Other Health System Catastrophes
The danger of group-based DNR policies is that they cross the line from medicine into social valuation. Once you start assigning resuscitation status based on age, disability, or projected “life-years,” you are no longer making a medical judgment about whether CPR will work for a specific patient. You are making a societal judgment about whose life is worth saving, a calculation that has historically fallen hardest on the people already most marginalized by health systems. The pandemic made this tension visible in a way that peacetime medicine usually keeps hidden, but the underlying ethical fault line was always there.