Lecture 7 — End-of-Life Care

Part A: Death, Dying, and Advance Planning

Brendan Shea, PhD

An Introduction to Bioethics.

A Woman Who Is Neither Awake Nor Dead

A Case to Sit With

In January 1983, 25-year-old Nancy Cruzan lost control of her car on an icy Missouri road and was thrown into a ditch. Paramedics restarted her heart, but her brain had gone too long without oxygen. She never woke up.

Years later she still breathes on her own. Her eyes open. She has cycles of sleep and waking. But she never tracks a face, never responds, never shows a sign that anyone is home. Her parents say she would have hated living this way, and they ask the hospital to stop the feeding tube that keeps her alive.

Write or discuss:

1. Is Nancy alive? Is she still there?
2. Should her parents’ word be enough to stop the feeding tube?
3. Who, exactly, should get to decide — and on what basis?

After You’ve Discussed

Note

Nancy’s case hides two different questions that people constantly run together.

• Is the person gone? This is a clinical question — about brains, breathing, and consciousness.
• Who decides what happens next, and on what basis? This is a legal and ethical question — about consent, evidence, and family.

A patient can be clearly gone with no decision left to make (she is simply dead), or fully present but unable to speak for herself, or — like Nancy — somewhere in between, where both questions are live at once. Today we take them one at a time.

Today

By the end of Part A, you will be able to:

• Distinguish coma, persistent vegetative state, minimally conscious state, and locked-in syndrome by their clinical signs.
• Explain the whole-brain standard of death, where it came from (Harvard 1968), and how the law codified it.
• Apply the two surrogate standards — substituted judgment and best interest — to a patient who cannot speak.
• Tell a living will, a durable power of attorney, and a POLST apart: what each is, who fills it out, and when it takes effect.
• Explain why the Cruzan ruling drove the spread of advance directives.
• Say what made the Quinlan, Cruzan, and Schiavo cases different from one another.

A Note on Language

The everyday words for this topic are loaded. “Vegetable,” “life support,” and “pulling the plug” all smuggle in judgments before the argument starts.

• We will use the clinical terms — vegetative state, mechanical ventilation, withdrawing treatment — not because they are colder, but because they are more precise.
• Precision matters here because the words decide real outcomes. Whether a body counts as a living patient or a dead body changes who may touch it, who may stop, and who may grieve.

The vocabulary is often where the ethics is hiding.

Question One: Is the Person Gone?

Two Things We Mix Up

Arousal vs. awareness

Consciousness has two parts, and they can come apart.

• Arousal is wakefulness — eyes open, sleep-wake cycles. It is run by the brainstem.
• Awareness is content — actually experiencing something, recognizing a face, feeling pain as pain. It depends on the cortex.

A patient can be aroused but not aware (eyes open, no one home). That combination is what makes end-of-life cases so hard, and why “unconscious” is too blunt a word for these states.

The Spectrum of Consciousness

These are not stops on a single dial. Locked-in patients are fully aware, off on their own axis — proof of how wrong a bedside guess can be.

Coma

A coma is a state of deep unresponsiveness: no arousal and no awareness. The eyes stay closed, and there are no sleep-wake cycles.

• It is caused by damage to the brainstem or to large areas of both hemispheres.
• Coma is usually temporary. Within two to four weeks, most patients either recover some function, progress to a vegetative state, or die.
• A coma is not the same as brain death, and it is not the same as deep anesthesia or sleep, though it can look similar from the bedside.

Persistent Vegetative State

In a vegetative state, the brainstem recovers enough for arousal but the cortex does not recover awareness. The patient’s eyes open, they have sleep-wake cycles, they may grimace or moan — but there is no sign of experience.

The word persistent means it has lasted a while; permanent is a prediction that it will not reverse — generally after 3 months (non-traumatic) or 12 months (traumatic injury). (The Multi-Society Task Force on PVS 1994)

Important

This is where Nancy Cruzan was.

Her eyes opened. She had sleep-wake cycles. Reflexes worked. To a visitor she could look almost awake — which is exactly why her family’s request to stop her feeding tube struck many people as killing, not letting die.

Minimally Conscious State

A minimally conscious state is a step up from vegetative: the patient shows inconsistent but real signs of awareness — following a simple command once, tracking a person with their eyes, reaching for a cup.

• The signs come and go, which makes the diagnosis genuinely hard.
• The stakes of getting it wrong are enormous. Studies have found that a large share of patients labeled “vegetative” are, on careful testing, actually minimally conscious.
• “Minimally conscious” still means conscious. Someone is in there, at least intermittently — and that changes every decision that follows.

Locked-In Syndrome

Locked-in syndrome is the mirror image of the vegetative state. Damage to the ventral pons in the brainstem paralyzes nearly every voluntary muscle, while the cortex is completely intact.

The patient is fully awake and aware — thinking, feeling, hearing every word said over the bed — but can often move only their eyes. They are not gone at all. They are trapped.

Why it belongs here

Locked-in patients have been mistaken for vegetative for months. From the outside, a fully conscious person and an unaware one can look the same — so any confidence about “no one is home” should carry a margin of doubt.

From a Stopped Heart to a Stopped Brain

For most of history, death was obvious: the heart stopped, breathing stopped, the body went cold. You did not need a definition.

Two mid-century technologies broke that simplicity.

• The mechanical ventilator could keep a body breathing and a heart beating after the brain had been destroyed. Now there were warm, pulsing bodies with no brain function at all.
• Organ transplantation raised the stakes: organs had to be taken from a body that was legally dead, but still perfused. When someone counted as dead suddenly had life-or-death consequences for others.

The Harvard Criteria (1968)

“A Definition of Irreversible Coma”

In 1968 an ad hoc committee at Harvard Medical School proposed a new basis for declaring death: not a stopped heart, but a permanently nonfunctioning brain. (Ad Hoc Committee of the Harvard Medical School to Examine the Definition of Brain Death 1968)

Their tests were practical and strict — no response to stimuli, no movement or breathing, no reflexes, a flat EEG — repeated after 24 hours to confirm irreversibility.

This was a genuine shift. For the first time, a breathing, beating body could be declared dead on the basis of its brain.

Whole-Brain Death Becomes Law

By 1981 the idea had matured into a legal standard. The President’s Commission endorsed it (President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research 1981), and the Uniform Determination of Death Act put it into statute across the states (National Conference of Commissioners on Uniform State Laws 1981).

A person is dead who has sustained either:

1. irreversible cessation of circulatory and respiratory functions, or
2. irreversible cessation of all functions of the entire brain, including the brainstem.

The second clause is whole-brain death. Both routes are legally death — equally final, equally complete.

How Brain Death Is Tested

Declaring whole-brain death is not a guess; it follows a fixed clinical protocol.

• No responsiveness and a known, irreversible cause.
• No brainstem reflexes — pupils fixed, no gag, no cough, no response to cold water in the ear.
• The apnea test — the ventilator is briefly removed; if carbon dioxide rises and the patient still makes no effort to breathe, the brainstem is not driving respiration.

Only when every part fails, with no reversible cause, is the patient declared dead — even while the ventilator keeps the chest rising.

Is Death a Fact or a Line We Draw?

The whole-brain standard (the law)

Death is the irreversible loss of the brain’s role as the body’s integrator. It is a biological fact we discover.

The challengers

• Higher-brain view: what matters is the person — consciousness. On this view a permanently unaware patient is, in the sense that counts, already gone.
• Circulatory view: bodies with brain death can be maintained for years, so “brain = death” may be a useful convention, not a discovery. (DeGrazia 2021)

The legal line and the philosophical question do not perfectly match — the law fixes a usable boundary where the metaphysics stays open.

Make It Contemporary

In 2025, Adriana Smith, a pregnant nurse in Georgia, was declared brain-dead after a medical emergency. By the whole-brain standard, she was legally dead.

But she was about nine weeks pregnant, and hospital staff read Georgia’s strict abortion law as requiring them to keep her body on support to let the fetus develop. Her body was maintained for months over her family’s objections.

Discuss:

1. If Adriana was legally dead, whose decision was it to continue support?
2. Does a brain-death declaration end a person’s interests, or can the law assign her body new obligations?

Make It Contemporary (cont.)

The Adriana Smith case is a collision of three bodies of law that rarely meet.

• Brain-death law said she was dead and her treatment could stop.
• Abortion law was read to say the pregnancy must be protected.
• Advance-directive law normally lets a person or family refuse continued intervention — but the law was unclear whether a dead woman could still “refuse.” (Watson 2025)

Discuss: When these three frameworks conflict, which should win — and who should decide that they conflict at all?

After You’ve Discussed

Note

The Smith case turns on a distinction worth stating outright.

• Being dead is a fact about the body — Adriana met the legal definition of death.
• Being allowed to die is a decision about a still-living patient — the question Nancy Cruzan’s family faced.

These are not the same problem. Confusing them is the single most common mistake in end-of-life debates: arguing about what someone is when the real disagreement is about what we are permitted to do.

Question Two: Who Decides?

When the Patient Can’t Speak

A competent adult may accept or refuse any treatment, including life-sustaining treatment — that is settled ground. The hard cases begin only when the patient loses the capacity to decide.

• Capacity is decision-specific: it is the ability to understand the choice, weigh it, and communicate it.
• When capacity is gone and cannot be regained in time, the decision must pass to someone else.
• The question then is who that someone is, and how they are supposed to decide.

The Surrogate Hierarchy

When a patient cannot decide, most states follow an ordered chain to find a surrogate — a stand-in decision-maker.

• First, anyone the patient named in advance (a healthcare proxy).
• If no one was named, a default hierarchy kicks in: usually spouse, then adult children, then parents, then siblings.
• The exact order is set by state law and varies, which is why two families in two states can face the same crisis and get different answers about who is in charge.

Who Decides? — A Flowchart

Standard One: Substituted Judgment

Substituted judgment

The surrogate asks: “What would this patient have wanted?” — not what the surrogate wants, and not what is best in the abstract.

It is an attempt to honor the patient’s own values when she can no longer voice them. Past statements, religious commitments, and how she lived all become evidence.

Its weakness: we are often guessing. Studies show surrogates predict patients’ wishes only a little better than chance. The harder the case, the thinner the evidence tends to be.

Standard Two: Best Interest

Best interest

When we have no idea what the patient would have wanted — a young child, a lifelong intellectual disability, a stranger with no record — substituted judgment has nothing to work with.

Then the surrogate falls back to asking: “What is best for someone in this situation?” — weighing benefits, burdens, pain, and prospects for recovery.

This standard is more objective, but it is also more paternalistic: it can override what an unusual patient might actually have chosen for herself.

The Two Standards Side by Side

	Substituted judgment	Best interest
Question asked	What would this patient have wanted?	What is best for someone here?
Needs	Evidence of the patient’s values	A judgment about benefits/burdens
Used when	We know something about the patient	We know little or nothing
Risk	Guessing, projection	Paternalism

Courts generally prefer substituted judgment when there is real evidence, and treat best interest as the fallback — which is exactly why written evidence became so valuable.

Cruzan Reaches the Supreme Court (1990)

Nancy’s parents asked Missouri to let them stop her feeding tube. The state refused, and in 1990 the case became the first “right to die” dispute the U.S. Supreme Court ever heard. (Supreme Court of the United States 1990)

The Court’s holding was narrow but pivotal:

• A competent person does have a constitutional right to refuse treatment, including artificial nutrition.
• But a state may require “clear and convincing evidence” of an incompetent patient’s own wishes before a surrogate stops treatment.
• Missouri’s standard was constitutional — and Nancy’s family had not yet met it.

Cruzan’s Ending

The ruling sent the family back to find better evidence — and they did.

• After the decision, former coworkers came forward with specific memories of Nancy saying she would never want to live “like a vegetable.”
• A Missouri court found this met the clear-and-convincing standard.
• Her feeding tube was removed in December 1990, and Nancy Cruzan died on December 26, 1990, nearly eight years after the crash.

The case that demanded evidence of her wishes is the reason millions of Americans now write those wishes down.

The Paperwork Cruzan Created

Why Directives Exploded After Cruzan

If the law would honor a patient’s wishes only with clear evidence, the obvious response was to create the evidence in advance.

• Within months of the ruling, Congress passed the Patient Self-Determination Act of 1990, requiring hospitals to ask every admitted patient whether they have an advance directive — and to tell them they may make one.
• An advance directive is simply a set of instructions, made while you have capacity, for a time when you do not.
• There are two main kinds, plus a third document that works differently. (Davis 2024)

The Living Will

A living will states, in advance, what treatments you would or would not want if you become unable to decide and are terminally ill or permanently unconscious.

• It speaks for you when you cannot — “no ventilator, no feeding tube, no CPR,” or the opposite.
• Its blind spot: it can only address situations you foresaw. Medicine is full of cases no one wrote down, and a living will goes silent on those.
• It also requires someone to find it and read it at the right moment — which does not always happen.

The Durable Power of Attorney

A durable power of attorney for health care does something different: instead of guessing every scenario, you name a person to decide for you.

“Durable” means it stays in effect after you lose capacity — which is the whole point.

The named agent can respond to the actual situation, ask questions, and reason the way you would, rather than being limited to a checklist written years earlier.

Note

Why many ethicists prefer it

A scenario you never imagined cannot be covered by a living will. But a trusted person who knows your values can adapt. Naming an agent transfers judgment, not just instructions.

POLST: A Different Kind of Document

A POLST (Physician Orders for Life-Sustaining Treatment; MOLST in some states) is not an advance directive. The difference matters.

• A living will is a patient’s wish. A POLST is a medical order, signed by a clinician, that other clinicians must follow immediately.
• It is meant for the seriously ill — people for whom these decisions are not hypothetical but likely within the year.
• It travels with the patient across settings (home, nursing facility, ambulance) and tells a paramedic at 3 a.m. exactly what to do, with no time to hunt for a lawyer’s document.

Three Documents at a Glance

	Living will	Durable POA	POLST / MOLST
What it is	Your stated wishes	A named decision-maker	A doctor’s standing order
Who fills it out	Any adult	Any adult	Clinician + patient
For whom	Anyone planning ahead	Anyone planning ahead	The seriously/terminally ill
When it kicks in	You lose capacity	You lose capacity	Immediately, as an order
Binds EMS?	Not directly	Through the agent	Yes, directly

What a Directive Looks Like

A simplified living-will excerpt (illustrative only)

If I am terminally ill or permanently unconscious, and two physicians agree my condition will not improve, I direct that:

• I do not want cardiopulmonary resuscitation (CPR).
• I do not want a breathing machine to be used or continued.
• I do want medication to keep me comfortable, even if it shortens my life.

I appoint my sister, Maria, as my health care agent to make any decision not covered above.

Real forms are longer and state-specific — but this is the shape of the thing.

Where Directives Fail

Advance directives are powerful, but they are not a solution to everything.

• Most people never fill one out — estimates put completion well under half of adults.
• A directive that exists but cannot be found in the emergency does no good.
• Many are too vague (“no heroic measures”) to guide a concrete decision.
• And a family that disagrees can stall or override even a clear document at the bedside.

These gaps are where the worst end-of-life conflicts begin.

Thought Question

Your father has had a massive stroke and cannot communicate or swallow. The team asks whether to place a feeding tube. He left no living will, but years ago he named you as his health care agent.

You remember him saying, after visiting his own mother in a nursing home, “Don’t ever let me end up like that.” But you are not certain he meant this.

1. Are you reasoning by substituted judgment or best interest right now?
2. How strong is your evidence — would it meet “clear and convincing”?
3. Does being the named agent change what you are allowed to decide, versus a default surrogate?

Make It Contemporary

States have tried to make the 3 a.m. version of your father’s case easier.

• POLST programs (now in most states) would let his physician convert “no feeding tube” into a portable medical order that EMS must honor.
• Default-surrogate statutes name, in law, who decides when no agent was appointed — but the order of relatives differs by state.

Discuss: Would a POLST have made your decision easier, or just moved the hard judgment earlier — onto the doctor who signed it?

After You’ve Discussed

Note

A distinction clinicians feel even though the law treats the two alike:

• Withholding treatment is never starting it (not placing the feeding tube).
• Withdrawing treatment is stopping something already running (removing the tube).

Ethically and legally, these are considered equivalent — if it is permissible not to start, it is permissible to stop. But many caregivers feel withdrawing is far harder, as if it causes the death. That gap between law and feeling is where Part B begins.

When the Tools Aren’t Enough: Schiavo

Theresa Schiavo

In February 1990, 26-year-old Terri Schiavo collapsed at home in Florida from cardiac arrest. Like Nancy Cruzan, her brain was starved of oxygen, and she entered a persistent vegetative state.

• Her husband, Michael, became her legal guardian.
• She left no living will and had named no health care agent.
• For several years the family was united in hoping she might recover.

The tools we just studied — the directive, the named agent — were exactly what she did not have.

The Conflict

By the late 1990s the family had split, and the split was total.

• Michael Schiavo said Terri would never have wanted to be kept alive this way, and asked to remove her feeding tube — a substituted-judgment claim based on past remarks.
• Her parents, the Schindlers, disputed that she was truly vegetative, believed she responded to them, and wanted treatment continued.

With no written directive, the case became a contest of credibility: whose account of Terri’s wishes should the court believe?

The Escalation

What made Schiavo unique was not the medicine — it was how far the fight traveled. (Quill 2005)

• Florida courts repeatedly found, on the evidence, that Terri was in a PVS and would have refused the tube.
• The Florida legislature passed “Terri’s Law” (2003) to let the governor order the tube reinserted; the state supreme court struck it down.
• In March 2005, Congress passed and the President signed a law moving the case to federal court.
• The federal courts declined to overturn the findings.

The Autopsy

Terri Schiavo’s feeding tube was removed, and she died on March 31, 2005, fifteen years after her collapse.

• The autopsy found her brain had atrophied to about half its normal weight — damage that was massive and irreversible.
• It confirmed she had been blind, making claims that she “tracked” visitors impossible.

The autopsy settled the medical facts. It did not settle the ethical and political dispute — by then the case had become a national argument about life, disability, and who should decide, and the autopsy changed almost no one’s mind.

Three Cases, Three Lessons

Case	Year	State	The new question it raised
Quinlan	1976	NJ	May a surrogate withdraw a ventilator? (Yes.)
Cruzan	1990	MO	May a state demand clear evidence of wishes? (Yes.)
Schiavo	2005	FL	What happens when the family itself splits?

Quinlan opened the door to refusing treatment. Cruzan set the evidentiary price. Schiavo showed that even with both settled, a divided family and a missing directive can still tear everything apart.

(Karen Ann Quinlan, notably, kept breathing for nearly a decade after her ventilator was removed — a reminder that “removing life support” does not always mean immediate death.)

Why These Cases Keep Happening

The Schiavo circus was extreme, but its ingredients are ordinary.

• Most people leave no directive — so there is nothing clear to follow.
• Families disagree, often out of love pulling in opposite directions.
• The diagnosis itself can be contested, especially the PVS/minimally-conscious line.

Every tool in this lecture exists to keep an ordinary tragedy from becoming a Schiavo. None of them works if the documents are never filled out.

Mini-Recap

We took two questions in turn. Is the person gone? led us through coma, vegetative and minimally conscious states, and the whole-brain definition of death. Who decides? led us through the surrogate hierarchy, substituted judgment versus best interest, and the documents — living wills, durable powers of attorney, and POLST — that Cruzan made essential.

Every case so far shared one feature: the question was whether to let the dying happen — to stop or withhold treatment. Part B crosses a sharper line: not letting die, but actively ending a life on request. The same patient, the same suffering — but a question that divides medicine far more deeply.

Review Questions

Recall

Name the two clauses of the Uniform Determination of Death Act, and explain how a vegetative state differs from whole-brain death.

Apply

A patient with advanced dementia and no advance directive develops pneumonia. Her three children disagree about antibiotics. Which surrogate standard applies, and what would you need to know to use it?

Debate

Should states be allowed to require “clear and convincing evidence” of a patient’s wishes before a family may stop treatment, as Cruzan permits — or does that set the bar too high?

Key Terms

• Coma — deep unresponsiveness with no arousal and no awareness; usually temporary.
• Vegetative state — arousal (eyes open, sleep-wake cycles) without awareness.
• Minimally conscious state — inconsistent but genuine signs of awareness.
• Locked-in syndrome — full awareness with near-total paralysis.
• Whole-brain death — irreversible loss of all brain function, including the brainstem; legally death.
• Surrogate — a stand-in decision-maker for a patient who lacks capacity.
• Substituted judgment — deciding as this patient would have wanted.
• Best interest — deciding what is best for someone in the patient’s situation.
• Advance directive — instructions made while competent for a time of incapacity.
• Living will / Durable power of attorney / POLST — wishes / a named agent / a portable medical order.

Further Reading

• President’s Commission. Defining Death. U.S. Government Printing Office, 1981.
• Cowart, Dax, and Robert Burt. “Confronting Death: Who Chooses, Who Controls?” (Cowart and Burt 1998)
• DeGrazia, David. “The Definition of Death.” Stanford Encyclopedia of Philosophy. (DeGrazia 2021)
• Davis, John K. “Advance Directives and Substitute Decision-Making.” Stanford Encyclopedia of Philosophy. (Davis 2024)

References

Ad Hoc Committee of the Harvard Medical School to Examine the Definition of Brain Death. 1968. “A Definition of Irreversible Coma.” JAMA 205 (6): 337–40. https://doi.org/10.1001/jama.1968.03140320031009.

Cowart, Dax, and Robert Burt. 1998. “Confronting Death: Who Chooses, Who Controls?” Hastings Center Report 28 (1): 14–24. https://doi.org/10.2307/3528337.

Davis, John K. 2024. Advance Directives and Substitute Decision-Making. The Stanford Encyclopedia of Philosophy (Fall 2024 Edition), Edward N. Zalta and Uri Nodelman (eds.). https://plato.stanford.edu/archives/fall2024/entries/advance-directives/.

DeGrazia, David. 2021. The Definition of Death. The Stanford Encyclopedia of Philosophy (Summer 2021 Edition), Edward N. Zalta (ed.). https://plato.stanford.edu/archives/sum2021/entries/death-definition/.

National Conference of Commissioners on Uniform State Laws. 1981. Uniform Determination of Death Act. Uniform Law Commission. https://www.uniformlaws.org/committees/community-home?CommunityKey=155faf5d-03c2-4027-99ba-ee4c99019d6c.

President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research. 1981. Defining Death: Medical, Legal and Ethical Issues in the Determination of Death. U.S. Government Printing Office.

Quill, Timothy E. 2005. “Terri Schiavo—a Tragedy Compounded.” New England Journal of Medicine 352 (16): 1630–33. https://doi.org/10.1056/NEJMp058062.

Supreme Court of the United States. 1990. Cruzan v. Director, Missouri Department of Health, 497 u.s. 261. Supreme Court of the United States.

The Multi-Society Task Force on PVS. 1994. “Medical Aspects of the Persistent Vegetative State (First of Two Parts).” New England Journal of Medicine 330 (21): 1499–508. https://doi.org/10.1056/NEJM199405263302107.

Watson, Katie. 2025. “Brain Death in Pregnancy—Abortion, Advance-Directive, or End-of-Life Law?” In New England Journal of Medicine, No. 4, vol. 393. https://doi.org/10.1056/NEJMP2507544.