Musée d’Orsay/Musée du Louvre, Paris/RMN–Grand Palais/Art Resource

Georges Seurat: Anaïs Faivre Haumonté on Her Deathbed, 1887

In his newest and best book, the surgeon Atul Gawande lets us have it right between the eyes: no matter how careful we are or healthful our habits, like everyone else, we will die, and probably after a long period of decline and debility. The average American, he tells us, spends a year or more disabled and living in a nursing home. Furthermore, the medical system will be of very little help at the end. In Gawande’s words:

The waning days of our lives are given over to treatments that addle our brains and sap our bodies for a sliver’s chance of benefit. They are spent in institutions—nursing homes and intensive care units—where regimented, anonymous routines cut us off from all the things that matter to us in life. Our reluctance to honestly examine the experience of aging and dying has increased the harm we inflict on people and denied them the basic comforts they most need.

Gawande wants us to know that the tragedy of old age and death cannot be fixed by modern medicine, so we better find some other way to deal with it. He divides his book into eight beautifully written chapters that follow the trajectory from independence to death. Being Mortal, the most personal book he has written, ends with the long dying of his own father.

In the first chapter, Gawande introduces us to his wife’s grandmother, Alice Hobson, age seventy-seven when he first met her in 1985. At that time, she was completely independent—living alone, healthy, and active. Gawande’s parents were immigrants from India, and he contrasts Hobson’s kind of independence with his own grandparents’ lives in India, where

it was understood that parents would just keep living in their home, assisted by one or more of the children they’d raised. In contemporary societies, by contrast, old age and infirmity have gone from being a shared, multigenerational responsibility to a more or less private state—something experienced largely alone or with the aid of doctors and institutions.

This change from the traditional way of dealing with old age has its reasons: there are now more old people and fewer young ones; families require incomes from both the husband and wife, so there are few leftover daughters and daughters-in-law to care for parents and grandparents at home; and family members might not live even in the same state, much less the same town. Still, as Gawande points out, if you are healthy and can afford it, “there is arguably no better time in history to be old.”

He is right about that, I think. In fact, as shown in the Harvard Grant Study,1 old people who are healthy and have the means are generally more content than they were when young. They have either achieved their ambitions or are resigned to not achieving them, so they live in the present rather than the future. Furthermore, they no longer have responsibility for children, and all the worries that go with that. “Whenever the elderly have had the financial means,” writes Gawande, “they have chosen what social scientists have called ‘intimacy at a distance.’”

But that string runs out sooner or later. And so it did for Alice Hobson. When we next meet her, it is 1992 and she is eighty-four. She has become unsteady and her memory is failing, but evidently everyone in the family, including Hobson herself, is utterly unprepared for her decline. They, like most Americans, have somehow imagined that it would not happen, that she would live in good health until someday in the distant future she would suddenly and neatly die. According to Gawande, “the prevailing fantasy is that we can be ageless,” and “we regard living in the downhill stretches with a kind of embarrassment.”

Gawande describes three patterns of decline. The first results from a fatal disease, such as incurable cancer; treatment may stave off the terminal phase for months or even years, but eventually the body is overwhelmed and death comes fairly quickly. In the second pattern, there is a chronic disease, such as emphysema, that is treatable, but characterized by repeated relapses until the body can no longer withstand even minor stresses. And finally, there is the pattern of old age called “frailty.” Here there is no life-threatening disease, but instead what Gawande calls the “accumulated crumbling of one’s bodily systems.” “Eventually,” he writes, “one too many joints are damaged, one too many arteries calcify. There are no more backups.”

The major threat to older people is that they will fall down; most falls are associated with muscle weakness and poor balance, and often with taking multiple medications. The consequences can be disastrous. About 20 percent of elderly people who fall and fracture a hip die within the following year from complications, and many of the rest never regain full function.


Despite her efforts, Alice Hobson eventually becomes unable to continue living on her own, and is moved from her home to a small apartment in a “senior-living complex,” which, says Gawande, “involved the imposition of more structure and supervision than she’d ever had to deal with before.” The next year, she fell and broke her hip, and was moved to the facility’s skilled nursing floor, where “all privacy and control were gone.”

She woke when they told her, bathed and dressed when they told her, ate when they told her. She lived with whomever they said she had to…. She felt incarcerated, like she was in prison for being old.

Nursing homes arose through federal legislation in 1954, not as a place where the elderly would be content, necessarily, but as a solution for hospitals that needed to empty their beds of patients who, with the closing of poorhouses, had nowhere else to go. “This is the consequence of a society that faces the final phase of the human life cycle by trying not to think about it,” Gawande writes.

We end up with institutions that address any number of societal goals—from freeing up hospital beds to taking burdens off families’ hands to coping with poverty among the elderly—but never the goal that matters to the people who reside in them: how to make life worth living when we’re weak and frail and can’t fend for ourselves anymore.

The family is the usual alternative to institutions for the elderly who can no longer live independently. “Your chances of avoiding the nursing home are directly related to the number of children you have,” says Gawande, “and, according to what little research has been done, having at least one daughter seems to be crucial to the amount of help you will receive.”

But since daughters now have to work to help support their own families, they have little time for elderly parents. At about the time most women became employed, beginning in the 1980s, “assisted living” facilities sprang up to help with parents and grandparents. “The concept of assisted living became so popular,” writes Gawande, “that developers began slapping the name on just about anything. The idea mutated from a radical alternative to nursing homes into a menagerie of watered-down versions with fewer services…a mere layover on the way from independent living to a nursing home.”

The quality of institutions for the elderly varies widely. In the worst of them, residents sleep two to a room, and during the day are often medicated to near oblivion (nursing homes are a big market for drug companies), strapped to Geri-chairs, with television the constant and only backdrop. One feature they nearly all have in common is that safety takes precedence over autonomy, since they are designed to appeal to the residents’ children, not the residents themselves. As one of the originators of the assisted living concept, quoting a colleague, told Gawande, “We want autonomy for ourselves and safety for those we love.” In addition, permitting residents to take risks exposes the institutions to the omnipresent threat of lawsuits.

Gawande finds a few facilities that try to introduce more autonomy and stimulation, even at the price of less safety. In one, for example, a man who prefers not to use a wheelchair is allowed to use a walker instead, even though that increases his risk of falling. Another sets out to bring life to its eighty residents—literally—by providing them with one hundred parakeets, four dogs, two cats, a colony of rabbits, and a flock of laying hens, plus a garden (all of which they have to care for with the help of the staff). But the innovative programs Gawande describes were all partly financed by state grants or philanthropy, and he doesn’t give that the emphasis it deserves. No matter how successful, they could not be widely adopted without some change in the system for financing long-term care. A single-payer health system that includes long-term care, delivered as much as possible in the home, would, I believe, be an improvement over parakeets and dogs, but unfortunately Gawande does not pay much attention to the economic system in which his account of aging and death unfolds.

Rembrandt: Bust of an Old Bearded Man Looking Down, Three-Quarters Right, 1631

In his most powerful chapter, called “Letting Go,” a version of which was published in The New Yorker in 2010, Gawande deals with the awful dilemma of deciding when to stop trying to prolong life, or in his words, “When should we try to fix and when should we not?” He begins with the story of Sara Monopoli, a thirty-four-year-old woman with inoperable lung cancer. Her story seems a little out of place in a book on aging and death, since I would expect her course to be different from that of an elderly person with the same diagnosis, and in fact it was—longer and worse. It took her eight months to die, during which she underwent four rounds of chemotherapy, whole-brain radiation, and at least two hospitalizations.


Maybe that’s why Gawande chose to feature her in this chapter—because it shows how bad a modern death can be. “These days, swift catastrophic illness is the exception,” he writes.

For most people, death comes only after long medical struggle with an ultimately unstoppable condition—advanced cancer, dementia, Parkinson’s disease, progressive organ failure (most commonly the heart, followed in frequency by lungs, kidneys, liver), or else just the accumulating debilities of very old age. In all such cases, death is certain, but the timing isn’t. So everyone struggles with this uncertainty—with how, and when, to accept that the battle is lost.

In Sara Monopoli’s case, the battle was lost with the diagnosis. The question was when continued life came at too high a price in suffering.

There are many reasons why burdensome treatment continues even when it’s futile and creates extra layers of suffering. Partly it’s because patients and their families simply don’t want to give up. We all have a strong biologic imperative to survive, whatever it takes. But in addition, patients and families are often unrealistic about what is possible, and one reason for that, says Gawande, is that physicians themselves are unrealistic. He cites a study of five hundred terminally ill patients whose physicians on average predicted their survival time would be more than five times longer than it turned out to be.

But even when physicians do know how bleak the outlook is, they often keep it from their patients. As Gawande says, “Most are reluctant to give a specific prognosis, even when pressed. More than 40 percent of oncologists admit to offering treatments that they believe are unlikely to work.” Gawande does not let himself off the hook. When Sara Monopoli was found to have thyroid cancer in addition to her lung cancer, “my solution was to avoid the subject altogether,” rather than tell her she would die from the lung cancer before the thyroid cancer could cause any problems. “I even raised with her the possibility that an experimental therapy could work against both her cancers, which was sheer fantasy.” It is much easier and faster to offer one more technological fix than to talk about death with a dying patient.

Gawande does not spare us the full horrors of Sara Monopoli’s death, and I believe he is right not to sidestep it:

Clinic notes from December describe shortness of breath, dry heaves, coughing up blood, severe fatigue. In addition to the drainage tubes in her chest, she required needle-drainage procedures in her abdomen every week or two to relieve the severe pressure from the liters of fluid that the cancer was producing there.

“We imagine that we can wait until the doctors tell us that there is nothing more they can do,” he says, “But rarely is there nothing more that doctors can do. They can give toxic drugs of unknown efficacy, operate to try to remove part of the tumor, put in a feeding tube if a person can’t eat: there’s always something.” Even at the end, when Sara Monopoli developed pneumonia, she was hospitalized and treated with antibiotics. Pneumonia in those circumstances was once called “the old man’s friend,” because it is a relatively quick and peaceful way to die. But it is now regarded as just one more treatable complication, never mind the situation in which it occurs.

Survival statistics form a bell-shaped curve, in which there are a small number of people who survive much longer than expected—the tail of the curve. Gawande believes that “we have failed to prepare for the outcome that’s vastly more probable…. We’ve built our medical system and culture around the long tail. We’ve created a multitrillion-dollar edifice for dispensing the medical equivalent of lottery tickets.” At some point, when suffering outweighs benefits, he clearly favors palliative care, and he cites a study from the Massachusetts General Hospital of patients with advanced cancer that showed that “those who saw a palliative care specialist stopped chemotherapy sooner, entered hospice far earlier, experienced less suffering at the end of their lives—and they lived 25 percent longer” (Gawande’s italics).

In Chapter 7, Gawande begins the touching story of his father’s final illness and death. Atmaram Gawande was a urologist in Athens, Ohio, who died of cancer on August 10, 2011, at the age of seventy-six, after a long and difficult course. Clearly, he was a strong and loving presence in his son’s life, and like Alice Hobson at the beginning, he was independent and active (in his case, professionally as well as physically). But in his early seventies, he developed neck pain and tingling in his fingers, which was initially chalked up to arthritis. Eventually, an MRI of his neck revealed a tumor in his spinal cord, which turned out to be a rare, slow-growing cancer. As the tumor compressed the nerves in his neck, he was gradually becoming quadriplegic, and he faced the difficult decision of whether and when to undergo the highly risky surgery that would be necessary to remove or decompress the tumor. He put off surgery for four years, during which he heroically managed to live with his increasing disabilities.

Atmaram Gawande survived the surgery, but the cancer could not be completely removed, and after six weeks of radiation, he was left with persistent pain, as well as nausea, fatigue, and loss of appetite. A follow-up MRI showed that the cancer was still growing, now upward into his brain. This was the beginning of the inexorable and rapid downhill course Gawande earlier described as the first of the three patterns of deterioration in old age. There was a series of falls, and once he was too weak to get off the floor. He told his son then, “I’m so scared. I’m becoming paralyzed. I can’t do this. I don’t want this. I don’t want to go through this. I want to die rather than go through this.” Gawande describes his father’s condition unflinchingly: “He was so weak his speech sometimes slurred. He had trouble getting food into his mouth, and his shirt was smeared with his dinner. He needed help standing from sitting.” He had reached a fork in the road, writes Gawande.

Atmaram Gawande enrolled in home hospice, which better regulated his medications and instructed him not to try to get up without someone with him. Gawande writes that it “astonished me to see the difference the hospice’s two simple instructions made.” But that respite did not last long. Within a few months, his father’s pain medications were making him “fuzzy, foggy, heavy,” he was sometimes confused, and he was having increasing difficulties balancing pain control with oversedation.

One morning, his wife, also a physician, couldn’t wake him, and so she called an ambulance to take him to the hospital—something hospices strongly discourage. In the hospital, he was given naloxone to counter the sedating effects of the narcotics, and also, like Sara Monopoli, he was given antibiotics for pneumonia, even though the only conceivable effect of this treatment would have been to deny him the fastest, most peaceful death. There was even talk of moving him to the ICU and putting him on a ventilator, something his wife wisely refused. When Gawande arrived at the hospital, he found his father alert, but very unhappy about waking up there, “Why are you letting me suffer?” At his insistence, he went home, where he died a few days later.

“Endings matter,” says Gawande. They certainly do, and what he does not say is what he thinks of his father’s ending. Would he have done anything differently? If so, what? I know that is a lot to ask from a son talking about the death of his beloved father, but given the subject of the book, it seems an important omission. Although the facts of the elder Gawande’s death are painful even to read, Gawande does make it clear that there were times of pleasure for his father, even in his last weeks; he had dinner parties, he watched movies at home with his wife, he kept up e-mail correspondence, and he talked with his grandchildren by phone. Was that enough?

Gawande only briefly discusses physician-assisted dying (sometimes called “assisted suicide”)—the practice, now legal in five states, of allowing physicians to give dying patients a prescription for a lethal dose of barbiturates to take if and when they choose to bring about a faster, more peaceful death. This is the weakest section of the book. He writes, “For the terminally ill who face suffering that we know will increase, only the stonehearted can be unsympathetic,” but then goes on to say, “I fear what happens when we expand the terrain of medical practice to include actively assisting people with speeding their death. I am less worried about abuse of these powers than I am about dependence on them.”

The implication is that we might begin to substitute assisted dying for palliative care and hospice. He points to the experience in the Netherlands, where he says the fact that “one in thirty-five Dutch people sought assisted suicide at their death is not a measure of success. It is a measure of failure.” (The Dutch usually use euthanasia, which involves the physician administering the lethal medication, instead of assisted suicide, in which the patient must ingest it orally, but the intent is the same.) Later, however, he says:

Given the opportunity, I would support laws to provide these kinds of prescriptions to people. About half don’t even use their prescription. They are reassured just to know they have this control if they need it. But we damage entire societies if we let providing this capability divert us from improving the lives of the ill.

This discussion is questionable on several counts in addition to his confusing vacillations. There is absolutely no evidence from any of the five states where assisted dying is legal (Oregon, Washington, Vermont, Montana, and New Mexico) that it has displaced good palliative care and hospice. On the contrary. The Center to Advance Palliative Care in 2011 issued a state-by-state report card evaluating the quality of palliative care. Oregon, where assisted dying had then been legal for thirteen years, rated an A, whereas Massachusetts (Gawande’s state, and mine) rated a B. And nearly all Oregonians who ended their lives were receiving hospice care at the time of their death.

Gawande says he would support a law like Oregon’s “given the opportunity,” but he did in fact have the opportunity in 2012 when such a bill appeared on the Massachusetts ballot (and was narrowly defeated).2 It is odd that he doesn’t mention this or say what he did then.

Why, moreover, does Gawande simply assert that the one in thirty-five assisted deaths in the Netherlands are too many? Given the prevalence of terrible deaths from cancer, as Gawande describes so well in his book, why is it not the right number? In Oregon, the number is one in five hundred deaths. Is that the right number? The problem, it seems to me, is the dreadful disease, not the number of patients who choose to bring about an inevitable death a little sooner. Moreover, giving patients that choice honors the autonomy that Gawande so powerfully champions throughout his book.

On November 1, twenty-nine-year-old Brittany Maynard, with the strong support of her husband and mother, died peacefully after taking a lethal dose of barbiturates prescribed according to Oregon’s Death with Dignity law. As widely reported in the media, she was suffering from incurable brain cancer, and, along with her family, relocated from California to Oregon so she would be able to obtain the medication. Afterward, she went through what she called a “bucket list” of things she wanted to do, including a trip to the Grand Canyon with her husband, mother, and stepfather, only ten days before she died. As she made clear in her online postings, she loved life, but was unwilling to wait to die until after the cancer had destroyed the person she was. I can’t help comparing Brittany Maynard with Sara Monopoli. Which death was preferable? For me, there is no question, but dying patients should be able to decide for themselves.

My husband, Arnold Relman, who died of metastatic melanoma on June 17, 2014, wanted the choice Brittany Maynard had, but because he lived in Massachusetts, not Oregon, he did not have it. He was assured he would not suffer, but that promise cannot always be kept, and in any event, what he feared most was not suffering, but the effect on his mind of the drugs he would need to prevent suffering—what Atmaram Gawande struggled with at the end. A month before he died, my husband enrolled in home hospice. His major problem then was debilitating pain from spinal stenosis (of uncertain relation to the cancer), but he was also becoming weak and short of breath, as the tumor filled his lungs. The hospice nurse started him on gradually increasing doses of methadone, and over the course of three weeks, the pain almost disappeared, but as was the case for Atmaram Gawande, his thinking became clouded and he was sometimes confused.

The last four days of his life were, rather abruptly, unspeakably difficult. He stopped eating and drinking altogether, he was in and out of consciousness, and his breathing was increasingly labored. Since there was no IV and he could no longer swallow, small amounts of liquid narcotics—to reduce his air hunger and, let’s face it, to hasten his death—had to be squirted under his tongue, as was done for Gawande’s father. I doubt that helped much.

I see no reason why my husband, and the people who loved him, had to endure those last days. Gawande speaks of the “fork in the road” that led his father to start home hospice. I think there is sometimes another fork in the road, the one my husband reached in his last days. There is no doubt in my mind that hospice is the best option for helping dying patients during their last months, but I am not at all sure it is adequate for the death itself. I have long supported physician-assisted dying (and was a lead petitioner in getting a Death with Dignity Act put on the Massachusetts ballot). More than ever, I believe dying patients should have that choice.

But after my husband’s death, I have come to favor euthanasia as well, for home hospice patients in the final, agonal stage of dying, who can no longer ingest medication orally. These patients are usually no longer mentally clear enough to give contemporaneous consent, but if they have earlier made it known that this is what they would wish, I believe that a duly appointed proxy should be able to have that wish carried out.

Whatever his view of assisted dying, Gawande has provided us with a moving and clear-eyed look at aging and death in our society, and at the harms we do in turning it into a medical problem, rather than a human one. I wish he had spent more time on what could be done to make things better (parakeets and dogs aren’t enough), but he has certainly shown us what is wrong. Narrowing the lives of the aged down to mere existence in institutions and submitting the dying to the full panoply of procedures that modern medicine has to offer cause enormous, unnecessary suffering. There is no way, of course, to make old age and all its infirmities disappear; it is what life deals us. But what Gawande shows us in this admirable book is that we could handle it a lot better.