Technological and biomedical advancements are undeniable blessings in today’s health sector. These advancements have improved life-prolonging technologies. So, clinicians can often sustain biological life long after meaningful recovery is unlikely. Consequently, there is a subtle shift happening in modern medicine, where we are no longer asking whether we can keep people alive but who gets to decide when not to.

This shift becomes clearer when viewed through Ronald Dworkin’s distinction between “critical interests” and “experiential interests.” In this framework, decisions about the end of life are not only about preserving biological existence or relieving suffering in the present moment. They also involve honoring the deeper narrative structure of a person’s life: the values, commitments, and identity-shaping choices that give that life coherence. From this perspective, autonomy is not merely about contemporaneous consent, but about whether the trajectory of dying aligns with the kind of life the individual has reasoned themselves to have lived.

Following this, end-of-life decision-making is no longer understood as the prerogative of a single actor. Contemporary bioethics, law, and clinical practice now frame it as a structured, shared process that distributes authority among patients, surrogates, clinicians, and legal frameworks.

Notably, across intensive care, pediatrics, dementia care, and emergency settings, shared decision-making is widely advocated as the ideal model, even if actual practice varies significantly across institutions and cultures. Yet “shared decision-making” often covers underlying issues about authority, responsibility, and moral burden.

A helpful way to approach the question is to distinguish the layers of decision-makers:

• The competent patient
• The surrogate or family decision-maker
• The clinician and care team

The competent patient: primary moral authority

Legal and ethical frameworks in many jurisdictions recognize that competent adults have the right to refuse or discontinue medical treatment, even if this refusal leads to death. This applies not only to terminal illness but also to decisions involving ventilation, dialysis, and artificial nutrition and hydration.

Ethically, this rests on:

• Respect for autonomy: Individuals are entitled to shape the course of their own dying in line with their values.
• Bodily integrity: Treatment without consent may amount to a violation.

Fortunately, patient autonomy is not unlimited. It does not compel clinicians to provide treatments that are non-beneficial or medically inappropriate. Consequently, patients can choose among reasonable options, but cannot demand every possible intervention.

Moreover, the ethical grounding of patient autonomy is best understood within a broader principlist framework, where autonomy, beneficence, non-maleficence, and justice operate not as hierarchical rules but as competing moral claims that must be balanced in context. In end-of-life care, autonomy often carries rhetorical dominance, but it is always ethically constrained by the clinician’s duties of beneficence and non-maleficence. This tension explains why autonomy in clinical practice is never absolute. It is always negotiated within the limits of medical reasonableness and professional responsibility.

Surrogates and families: speaking for those who cannot speak

When patients lose decision-making capacity, surrogate decision-makers, usually family members or formally appointed proxies, step in. They are expected to rely on two main standards:

• Substituted judgment: What the patient would have chosen.
• Best interests: What would most reasonably benefit the patient now.

But then, in reality, this role is not as neutral as it seems. Surrogates often face:

• Emotional distress and grief
• Conflicts between prior patient wishes and present realities
• Personal, cultural, or financial pressures

In some cases, surrogates may reinterpret, or even resist, advance directives when they conflict with their own preferences or fears. Notably, legal frameworks also shape how much authority surrogates truly have. In some jurisdictions, different types of surrogates (for example, default family members vs. formally appointed proxies) have different powers, particularly when it comes to withdrawing life-sustaining treatment. It is very important to point out that these differences are highly jurisdiction-specific and not universally consistent. For example, in the United Kingdom, the guiding legal principle is the “best interests” standard under the Mental Capacity Act, which places strong emphasis on clinical judgment combined with consideration of the patient’s values and previously expressed wishes. In contrast, many United States jurisdictions place greater procedural weight on surrogate authority and advance directives, with variation across states in how conflicts between clinicians and families are resolved.

Furthermore, although substituted judgment is often presented as a mechanism for preserving patient autonomy through surrogates, it rests on a contested assumption: that a person’s preferences are stable, knowable, and translatable across contexts. However, relational autonomy theory challenges this assumption by arguing that decision-making capacity is socially embedded rather than purely individual.

From this perspective, surrogate decisions are not simply “representations” of the patient’s will but interpretations shaped by relationships, memory, emotional bonds, and moral identity. This helps explain why surrogates frequently struggle not because they misunderstand the patient, but because the patient’s values themselves were never static or fully articulated.

Clinicians: gatekeepers of medical indications and limits

The interesting part about clinicians is that they do not simply carry out decisions. They shape what decisions are even available.

Their responsibilities primarily include:

• Clarifying prognosis and clinical realities
• Defining treatment goals
• Assessing whether interventions are beneficial, burdensome, or inappropriate

The concept often used here is “futility,” but this term is itself contested. There is no universally accepted definition, and it may refer to physiological ineffectiveness, low probability of success, or poor quality outcomes. Some clinicians interpret futility in physiological terms, where an intervention cannot achieve its intended biological effect. Others adopt probabilistic definitions, focusing on extremely low likelihoods of success. A third approach introduces value-based judgments, considering whether a treatment meaningfully contributes to quality of life. This last interpretation is particularly controversial because it risks shifting medical judgment into subjective assessments of worth or desirability of life, which raises ethical concerns about bias and value imposition.

Despite this and as a matter of discipline, clinicians are typically the primary assessors of medical indication and potential futility. However, their authority to act on these judgments is legally and institutionally constrained. In some systems, they may limit treatment after following formal processes. In others, consent or transfer is required. In pediatric care, clinicians often follow a process involving early planning, information exchange, and eventual shared decisions with parents, and sometimes with the child, depending on age and capacity.

Additionally, the role of the clinician can be more deeply understood through Edmund Pellegrino’s conception of medicine as a “moral enterprise.” In this view, clinical judgment is not merely technical decision-making about biological outcomes, but an ethical act grounded in the fiduciary relationship between physician and patient. The physician is therefore not a neutral technician but a moral agent responsible for determining what constitutes appropriate medical good in a given context. This explains why judgments of futility are not purely empirical, but are embedded in professional interpretations of benefit, harm, and the limits of medicine itself.

It is equally important to observe how this subject matter reflects an unresolved tension between liberal legal theory and paternalistic state interests. Liberal approaches prioritize individual autonomy and bodily integrity, emphasizing the right to refuse treatment even when death is foreseeable. In contrast, paternalistic or communitarian approaches justify certain constraints on autonomy in order to preserve life, protect vulnerable persons, or uphold perceived social and moral goods. End-of-life law therefore does not simply codify ethical consensus. It institutionalizes ongoing philosophical disagreement about the limits of individual self-determination.

Interestingly, in practice, the biggest challenges are not even theoretical. They usually manifest in more human and less predictable ways, such as uncertainty about prognosis, limited understanding of treatment options, emotional distress often mixed with mistrust, and a lingering fear of legal consequences. These realities shape decision-making in ways guidelines cannot fully anticipate, often leading to delayed choices, demands for aggressive treatment, or tension between clinicians and families.

Moreover, no discussion of end-of-life decision-making is complete without addressing distributive justice. In real-world clinical settings, decisions about life-sustaining treatment occur against a backdrop of finite intensive care unit (ICU) beds, ventilators, nursing staff, and hospital budgets. A family’s demand for aggressive, non-beneficial treatment for a patient with no meaningful recovery prospect does not occur in a vacuum. It consumes resources that could benefit other patients with reversible conditions. While clinicians rarely invoke resource scarcity explicitly at the bedside, systemic constraints shape what is offered, for how long, and under what conditions. Justice requires that end-of-life frameworks acknowledge this reality, even if the primary ethical obligation remains to the individual patient. When such resource limits are ignored, there is the risk of presenting an idealized picture of decision-making that does not exist in under-resourced health systems.

But then, across different settings, one finding remains consistent. Effective communication is at the heart of good end-of-life care. This means starting conversations early before crises force rushed decisions, clearly defining the goals of care whether that is cure, prolongation of life, or comfort, and ensuring that treatments are aligned with those goals. Yet communication is not a panacea. When communication fails, due to mistrust, cultural divergence, emotional breakdown, or conflicting values, the structured frameworks of law and clinical ethics must provide guidance. The article’s emphasis on communication is vital, but it should not obscure that some disputes are irresolvable through conversation alone.

Finally, end-of-life decision-making exposes a deeper philosophical tension in modern medicine, which is the transition from medicine as purely life-preserving to medicine as value-sensitive practice. The question is no longer only what can be done biologically, but what should be done morally, socially, and legally. This shift places clinicians not only as scientific decision-makers but as mediators of competing values under conditions of uncertainty. Ultimately, the moral challenge of end-of-life care lies not in choosing between life and death, but in cultivating the clinical and ethical discernment required to act faithfully within the meaning of a life that is approaching its natural or medically recognized end.

Chinmeri Nwuba is a health policy writer.