Atul Gawande: “Letting Go: What Should Medicine Do When It Can’t Save Your Life?”

Summary: In the August 2 issue of the New Yorker, Boston surgeon Atul Gawande writes about the ambiguities that plague end-of-life care.  With the candor that makes his writing both so credible and so moving, he acknowledges that, as a physician, he too has great difficulty “letting go” of a dying patient. 

Ultimately, Gawande suggests, the problem with the way we deal with death today is that we have forgotten the art of dying.  Palliative and hospice care can help us recover a lost art, he adds, though not quite in the way most of us expect.  The goal is not to cut costs and curtail suffering by shortening the process of dying. In fact research shows that those who receive hospice or palliative care often live longer than other patients.

The aim of hospice and palliative care is to help patients “have the fullest possible lives,” during their final weeks or months, Gawande explains. “That means focusing on objectives like freedom from pain and discomfort, or maintaining mental awareness for as long as possible, or getting out with family once in a while. Hospice and palliative-care specialists aren’t much concerned about whether that makes people’s lives longer or shorter.” 

As it turns out, improving the quality of life  for patients suffering from a fatal disease does, in fact, reduce the cost of dying. Patients are less like to return to an ER in the middle of the night, or wind up in an ICU.  But saving money is not the goal of palliative care, it is simply a welcome by-product that, by now, HealthBeat readers should expect. Once again, higher quality care translates into more affordable care.  First, aim for better care. The savings will come.


Gawande begins “Letting Go” with the story of Sara Thomas Monopoli, 39 weeks pregnant with her first child “when her doctors learned that she was going to die. It started with a cough and a pain in her back. Then a chest X-ray showed that her left lung had collapsed, and her chest was filled with fluid. A sample of the fluid was drawn off with a long needle and sent for testing. Instead of an infection, as everyone had expected, it was lung cancer, and it had already spread to the lining of her chest.” She and her husband Rich were bewildered by the diagnosis. “She had never smoked, or lived with anyone who had. She exercised. She ate well.”

The doctors induced labor, and Sara and Rich decided that, for one day, they would enjoy the birth of their first child. “’Sara and I looked at each other,’ Rich recalled, ‘and we said, “We don’t have cancer on Tuesday.  It’s a cancer-free day. We’re having a baby. It’s exciting. And we’re going to enjoy our baby.’” On Tuesday, at 8:55 P.M., Vivian Monopoli, seven pounds nine ounces, was born. She had wavy brown hair, like her mom, and she was perfectly healthy.”

On Wednesday, they would confront Sara’s cancer—or not.

       What Our Culture Teaches Us

Sara’s oncologist, Dr. Paul Marcoux, told her that the cancer was inoperable. But there were chemotherapy options, “notably a relatively new drug called Tarceva, which targets a gene mutation commonly found in lung cancers of female non-smokers. Eighty-five per cent respond to this drug,” and, Gawande reports, “her oncologist couldn’t help but add, ‘some of these responses can be long-term.’” How could he not tell her that? She had a right to know.

Still, “Words like ‘respond’ and ‘long-term’ provide a reassuring gloss on a dire reality,” Gawande writes.  “There is no cure for lung cancer at this stage. Even with chemotherapy, the median survival is about a year. But it seemed harsh and pointless to confront Sara and Rich with this now. Vivian was in a bassinet by the bed. They were working hard to be optimistic. As Sara and Rich later told the social worker who was sent to see them, they did not want to focus on survival statistics. They wanted to focus on ‘aggressively managing’ this diagnosis.”  Unfortunately, as Gawande later notes, most patients and families think that “long-term” means ten years—not one year.

Sara understood that her disease was incurable,” Gawande adds. “The week after she was given the diagnosis and delivered her baby, she spelled out her wishes for Vivian’s upbringing after she was gone. She had told her family on several occasions that she did not want to die in the hospital. She wanted to spend her final moments peacefully at home. But the prospect that those moments might be coming soon, that there might be no way to slow the disease, ‘was not something she or I wanted to discuss,’ her mother said.” It is one thing to know that you are battling a disease that, ultimately, will win. It is something else to recognize that death is imminent.

Sara’s father, Gary, and her twin sister, Emily, “still held out hope for a cure,” Gawande notes. “The doctors simply weren’t looking hard enough, they felt. ‘I just couldn’t believe there wasn’t something,’ Gary said.” 

This is an American way of denying death that  we have been taught, by our culture, our media, and the medical profession itself:  If you don’t give up, if you fight, if you search hard enough, if the doctor is good enough, if the hospital is the best, if you are willing to spend enough —there is always something to be done.  An insurance company who says that  the treatment won’t do any good is simply trying to save money.

        Why Physicians Deny Death

Gawande acknowledges that physicians also have difficulty navigating those dark “shoals,” where we all founder, that shallow water where even big ships like our health care system can’t anchor, and the dying can’t catch their breath: “At least two things get in the way. First, our own views may be unrealistic. A study led by the Harvard researcher Nicholas Christakis asked the doctors of almost five hundred terminally ill patients to estimate how long they thought their patient would survive, and then followed the patients. Sixty-three per cent of doctors overestimated survival time. Just seventeen per cent underestimated it. The average estimate was five hundred and thirty per cent too high. And, the better the doctors knew their patients, the more likely they were to err.

“Second, we often avoid voicing even these sentiments. Studies find that although doctors usually tell patients when a cancer is not curable, most are reluctant to give a specific prognosis, even when pressed,” he reports.  More than forty per cent of oncologists report offering treatments that they believe are unlikely to work. In an era in which the relationship between patient and doctor is increasingly miscast in retail terms—‘the customer is always right’—doctors are especially hesitant to trample on a patient’s expectations. You worry far more about being overly pessimistic than you do about being overly optimistic. And talking about dying is enormously fraught. When you have a patient like Sara Monopoli, the last thing you want to do is grapple with the truth. I know, because Marcoux [her oncologist] wasn’t the only one avoiding that conversation with her. I was, too.”

Gawande explains: “Earlier that summer a PET scan had revealed that, in addition to her lung cancer, she also had thyroid cancer, which had spread to the lymph nodes of her neck, and I was called in to decide whether to operate. This second, unrelated cancer was in fact operable. But thyroid cancers take years to become lethal. Her lung cancer would almost certainly end her life long before her thyroid cancer caused any trouble. Given the extent of the surgery that would have been required, and the potential complications, the best course was to do nothing. But explaining my reasoning to Sara meant confronting the mortality of her lung cancer, something that I felt ill prepared to do. . . .”

“Sitting in my clinic, Sara did not seem discouraged by the discovery of this second cancer,” Gawande recalls. “She seemed determined. She’d read about the good outcomes from thyroid-cancer treatment. So she was geared up, eager to discuss when to operate. And I found myself swept along by her optimism. Suppose I was wrong, I wondered, and she proved to be that miracle patient who survived metastatic lung cancer?

“My solution was to avoid the subject altogether. I told Sara that the thyroid cancer was slow-growing and treatable. The priority was her lung cancer, I said. Let’s not hold up the treatment for that. We could monitor the thyroid cancer and plan surgery in a few months.”

Meanwhile, Sara had undergone three rounds of chemotherapy with limited, if any, effect. At that point “Perhaps her oncologist could have discussed what she most wanted as death neared and how best to achieve those wishes,” Gawande observes. “But the signal he got from Sara and her family was that they wished to talk only about the next treatment options. They did not want to talk about dying.”

A doctor cannot force people to talk about death. If a physician tries, the family is likely to label him “not up to snuff,” fire him, and find a new physician. This does the patient no good. So when Marcoux met with Sara and her family, “he took the measure of the room,” Gawande says. “With almost two decades of experience treating lung cancer, he had been through many of these conversations.”

“’I know that the vast majority of my patients are going to die of their disease,’” Marcoux later told Gawande. “The data show that, after failure of second-line chemotherapy, lung-cancer patients rarely get any added survival time from further treatments and often suffer significant side effects.’ But,” Gawande notes, “he, too, has his hopes.”  He could not be sure that a new, experimental therapy would not help Sara.

Marcoux told the family that, “at some point, ‘supportive care’ was an option for them to think about. But, he went on, there were also experimental therapies. He told them about several that were under trial. The most promising was a Pfizer drug that targeted one of the mutations found in her cancer’s cells. Sara and her family instantly pinned their hopes on it. The drug was so new that it didn’t even have a name, just a number—PF0231006—and this made it all the more enticing.

“There were a few hovering issues . . .” Gawande writes. “The critical problem was that the rules of the trial excluded Sara because of the pulmonary embolism she had developed that summer. To enroll, she would need to wait two months, in order to get far enough past the episode. In the meantime, [her oncologist] suggested trying another conventional chemotherapy . 

“It’s worth pausing to consider what had just happened,” Gawande adds. “Step by step, Sara ended up on a fourth round of chemotherapy, one with a minuscule likelihood of altering the course of her disease and a great likelihood of causing debilitating side effects. An opportunity to prepare for the inevitable was forgone. And it all happened because of an assuredly normal circumstance: a patient and family unready to confront the reality of her disease.”

Gawande asked Marcoux “what he hope[d]to accomplish for terminal lung-cancer patients when they first come to see him.” Marcoux’ reply:  “I’m thinking, ‘Can I get them a pretty good year or two out of this?’ Those are my expectations. For me, the long tail for a patient like her is three to four years.’”  For a young mother like Sara, the difference between two years and four years would be enormous. Four years would give her the opportunity to get to know her daughter as a person, and to leave her with the firm memory of a mother who loved her. Little wonder that Sara was eager to try anything.

Moreover, when a doctor talks to a patient and her family about “more time,”  they realize that he or she doesn’t really know what the outer limit of  “more “ might mean in an individual case. “’They’re thinking ten to twenty years,’ Marcoux confided.  "You hear that time and time again. And I’d be the same way if I were in their shoes.”

Meanwhile, Sara grew sicker. “A CT scan in December showed that the lung cancer was spreading through her spine, liver, and lungs. When we met in January, she could move only slowly and uncomfortably,” Gawande remembers. “Her lower body had become swollen.  . . .

“And still,” he reports, “Sara, her family, and her medical team remained in battle mode. On February 12th, she completed five days of radiation treatment, which left her immeasurably fatigued, barely able get out of bed. She ate almost nothing. She weighed twenty-five pounds less than she had in the fall. She confessed to Rich that, for the past two months, she had experienced double vision and was unable to feel her hands.

“‘Why didn’t you tell anyone?’ he asked her.

“I just didn’t want to stop treatment,’ she said. ‘They would make me stop’”

Her chances were rapidly dwindling,” Gawande admits. “But who was to say they were zero?

        “The Median Is Not the Message” But “Hope Is Not a Plan”

When he thinks about his terminally ill patients, Gawande tell us that he cannot help but recall an extraordinary essay entitled “The Median is Not the Message,” written by the paleontologist and writer Stephen Jay Gould in 1985 “after he had been given a diagnosis . . . of abdominal mesothelioma, a rare and lethal cancer usually associated with asbestos exposure.  He went to a medical library when he got the diagnosis and pulled out the latest scientific articles on the disease. ‘The literature couldn’t have been more brutally clear: mesothelioma is incurable, with a median survival of only eight months after discovery,’ Gould wrote. The news was devastating. But then he began looking at the graphs of the patient-survival curves.

“Gould was a naturalist, and more inclined to notice the variation around the curve’s middle point than the middle point itself,” Gawande points out.  “What the naturalist saw was remarkable variation. The patients were not clustered around the median survival but, instead, fanned out in both directions. Moreover, the curve was skewed to the right, with a long tail, however slender, of patients who lived many years longer than the eight-month median. This is where he found solace.

“He could imagine himself surviving far out in that long tail. And he did. Following surgery and experimental chemotherapy, he lived twenty more years before dying, in 2002, at the age of sixty, from a lung cancer that was unrelated to his original disease.”

In his 1985 essay, Gould wrote: “‘It has become, in my view, a bit too trendy to regard the acceptance of death as something tantamount to intrinsic dignity. Of course I agree with the preacher of Ecclesiastes that there is a time to love and a time to die—and when my skein runs out I hope to face the end calmly and in my own way. For most situations, however, I prefer the more martial view that death is the ultimate enemy—and I find nothing reproachable in those who rage mightily against the dying of the light.”

Gawande confesses: “I think of Gould and his essay every time I have a patient with a terminal illness. There is almost always a long tail of possibility, however thin. What’s wrong with looking for it? Nothing, it seems to me, unless it means we have failed to prepare for the outcome that’s vastly more probable. The trouble is that 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—and have only the rudiments of a system to prepare patients for the near-certainty that those tickets will not win. Hope is not a plan, but hope is our plan.”

For Sara, Gawande tells us, “there would be no miraculous recovery, and, when the end approached, neither she nor her family was prepared.”

She had wanted to die, peacefully, at home.  Instead, one morning “three days before she was to start her new round of chemo, Rich awoke to find his wife sitting upright beside him, pitched forward on her arms, eyes wide, struggling for air. She was gray, breathing fast, her body heaving with each open-mouthed gasp. She looked as if she were drowning. He tried turning up the oxygen in her nasal tubing, but she got no better.

“’I can’t do this,’ she said, pausing between each word. ‘I’m scared’”

“He had no emergency kit in the refrigerator. No hospice nurse to call,” Gawande writes. “He called 911, . . .A few minutes later, firemen swarmed up the stairs to her bedroom, sirens wailing outside.”

In the hospital, Gawande explains,  her primary care physician intervened, and finally Sara’s family “let him call the palliative-care team to visit. The team prescribed a small dose of morphine, which immediately eased her breathing. Her family saw how much her suffering diminished, and suddenly they didn’t want any more suffering. The next morning, they were the ones to hold back the medical team.

“‘They wanted to put a catheter in her, do this other stuff to her,’ her mother, Dawn, told  Gawande . ‘I said, “No. You aren’t going to do anything to her. . . ”’  

“In the previous three months, almost nothing we’d done to Sara—none of our chemotherapy and scans and tests and radiation—had likely achieved anything except to make her worse,” Gawande admits. “She may well have lived longer without any of it. At least she was spared at the very end.

“That day, Sara fell into unconsciousness as her body continued to fail. Through the next night, Rich recalled, ‘there was this awful groaning. “Whether it was with inhaling or exhaling, I don’t remember, but it was horrible, horrible, horrible to listen to. . .’. At the end, Rich knelt by her bed, weeping,  and whispered in her ear “It’s okay to let go.”  Eventually she did.  But this was not how she had wanted to die.

Who Should ‘Win’ the Patient or the Payor?  “We’re “Debating the Wrong Question”

What could Sara’s doctors have done differently “Or, to put it another way,” Gawande asks, “if you were the one who had metastatic cancer—or, for that matter, a similarly advanced case of emphysema or congestive heart failure—what would you want your doctors to do?

“The issue has become pressing, in recent years,” he continues, “for reasons of expense.  . . .  Twenty-five per cent of all Medicare spending is for the five per cent of patients who are in their final year of life . . .  Spending on a disease like cancer tends to follow a particular pattern. There are high initial costs as the cancer is treated, and then, if all goes well, these costs taper off. Medical spending for a breast-cancer survivor, for instance, averaged an estimated fifty-four thousand dollars in 2003, the vast majority of it for the initial diagnostic testing, surgery, and, where necessary, radiation and chemotherapy.”  Many of these patients live; those health care dollars are not wasted.  But “for a patient with a fatal version of the disease,” Gawande explains, “the cost curve is U-shaped, rising again toward the end—to an average of sixty-three thousand dollars during the last six months of life with an incurable breast cancer. . . .  ultimately, death comes, and no one is good at knowing when to stop.” Arguably those dollars are squandered.  Yet when the spending begins, no one knows which patients will survive. And even when the outlays  begins to rise on the other side of the “U,” it’s not clear when to stop. Maybe this is just a set-back; maybe the situation will turn around . . .

Gawande notes that when we debate end-of-life care in our society, we wind up focusing on the money, and ignore the issues that cut much further into our national culture and psyche: “The subject seems to reach national awareness mainly as a question of who should ‘win’ when the expensive decisions are made: the insurers and the taxpayers footing the bill or the patient battling for his or her life.  Budget hawks urge us to face the fact that we can’t afford everything. Demagogues shout about rationing and death panels. Market purists blame the existence of insurance: if patients and families paid the bills themselves, those expensive therapies would all come down in price. But,” Gawande argues, “they’re debating the wrong question. The failure of our system of medical care for people facing the end of their life runs much deeper. To see this, you have to get close enough to grapple with the way decisions about care are actually made.

“These days, swift catastrophic illness is the exception; for most people, death comes only after long medical struggle with an incurable condition—advanced cancer, progressive organ failure (usually the heart, kidney, or liver), or the multiple debilities of very old age. In all such cases, death is certain, but the timing isn’t,” he notes. “So everyone struggles with this uncertainty—with how, and when, to accept that the battle is lost. As for last words, they hardly seem to exist anymore. Technology sustains our organs until we are well past the point of awareness and coherence. Besides, how do you attend to the thoughts and concerns of the dying when medicine has made it almost impossible to be sure who the dying even are? Is someone with terminal cancer, dementia, incurable congestive heart failure dying, exactly?

“This is a modern tragedy,” he continues, “ replayed millions of times over. When there is no way of knowing exactly how long our skeins will run—and when we imagine ourselves to have much more time than we do—our every impulse is to fight, to die with chemo in our veins or a tube in our throats or fresh sutures in our flesh. The fact that we may be shortening or worsening the time we have left hardly seems to register. We imagine that we can wait until the doctors tell us that there is nothing more they can do. 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. We want these choices. We don’t want anyone—certainly not bureaucrats or the marketplace—to limit them. But that doesn’t mean we are eager to make the choices ourselves. Instead, most often, we make no choice at all. We fall back on the default, and the default is: Do Something. Is there any way out of this?”

   Hospice and Palliative Care – Finding A Way Out

Some of us wonder if for-profit insurance companies ever “add value” to our health care system. Gawande reminds us that there are exceptions to all of our generalizations when he offers an example of an insurer who has, in one instance, expanded our understanding of how to create a more compassionate, more patient-centered, more affordable system.

“In late 2004, executives at Aetna . . .  started an experiment.” Normally, a patient is not eligible for hospice until he is willing to forego all treatment. This is why most patients don’t enroll in hospice until the very end.  But “Aetna decided to let a group of policyholders with a life expectancy of less than a year receive hospice services without forgoing other treatments. A patient like Sara Monopoli could continue to try chemotherapy and radiation, and go to the hospital when she wished—but also have a hospice team at home focusing on what she needed for the best possible life now."

A two-year study of this “concurrent care” program found that enrolled patients were much more likely to use hospice," Gawande reports. "The figure leaped from twenty-six per cent to seventy per cent. That was no surprise, since they weren’t forced to give up anything. The surprising result was that they did give up things. They visited the emergency room almost half as often as the control patients did. Their use of hospitals and I.C.U.s dropped by more than two-thirds. Over-all costs fell by almost a quarter.

(It's worth noting, as Gawande did in an online interview responding to questions about this article, that while Aetna has not expanded this experiment to its customers, the new healthcare legislation includes a Medicare pilot project that would let 12 communities across the country test this approach of not requiring terminal patients to sign away their ability to recive curative treatment (however unlikely to be succesful) in order to get hospice services.  If the pilot is succesful, Medicare could roll it out nationwide, without needing approval from Congress.)

The results of Aetna's experiment were "stunning, and puzzling," says Gawande, "it wasn’t obvious what made the approach work." Aetna then decided to see if palliative care would have a similar effect, running a “more modest concurrent-care program for a broader group of terminally ill patients. For these patients,” Gawande points out, “the traditional hospice rules applied—in order to qualify for home hospice, they had to give up attempts at curative treatment. But, either way, they received phone calls from palliative-care nurses who offered to check in regularly and help them find services for anything from pain control to making out a living will. For these patients, too, hospice enrollment jumped to seventy per cent, and their use of hospital services dropped sharply. Among elderly patients, use of intensive-care units fell by more than eighty-five per cent. Satisfaction scores went way up. What was going on here? The program’s leaders had the impression that they had simply given patients someone experienced and knowledgeable to talk to about their daily needs. And somehow that was enough—just talking.

“The explanation strains credibility,” Gawande acknowledges, “but evidence for it has grown in recent years. Two-thirds of the terminal-cancer patients in the Coping with Cancer study reported having had no discussion with their doctors about their goals for end-of-life care, despite being, on average, just four months from death. But the third who did were far less likely to undergo cardiopulmonary resuscitation or be put on a ventilator or end up in an intensive-care unit. Two-thirds enrolled in hospice. These patients suffered less, were physically more capable, and were better able, for a longer period, to interact with others. Moreover, six months after the patients died their family members were much less likely to experience persistent major depression." In other words, people who had substantive discussions with their doctor about their end-of-life preferences were far more likely to die at peace and in control of their situation.

Here, it is crucial to understand that the goal of Hospice is not to speed  the process of dying—or to save money.  The difference between hospice and other care,” Gawande explains is a matter of “priorities. In ordinary medicine, the goal is to extend life. We’ll sacrifice the quality of your existence now—by performing surgery, providing chemotherapy, putting you in intensive care—for the chance of gaining time later. Hospice deploys nurses, doctors, and social workers to help people with a fatal illness have the fullest possible lives right now. That means focusing on objectives like freedom from pain and discomfort, or maintaining mental awareness for as long as possible, or getting out with family once in a while. Hospice and palliative-care specialists aren’t much concerned about whether that makes people’s lives longer or shorter.

“Like many people,” Gawande writes, “I had believed that hospice care hastens death, because patients forgo hospital treatments and are allowed high-dose narcotics to combat pain. But studies suggest otherwise. In one, researchers followed 4,493 Medicare patients with either terminal cancer or congestive heart failure. They found no difference in survival time between hospice and non-hospice patients with breast cancer, prostate cancer, and colon cancer. Curiously, hospice care seemed to extend survival for some patients; those with pancreatic cancer gained an average of three weeks, those with lung cancer gained six weeks, and those with congestive heart failure gained three months. The lesson seems almost Zen: you live longer only when you stop trying to live longer.”  

My guess is that with supportive hospice therapy, patient are no longer flailing about, they are no longer wrestling with the Angel of Death. They save their strength for a good death.

     When Atul Gawande Cannot Find the Right Words

Gawande ends his story with great humility, describing the scene, last winter, when he tried to tell a patient suffering from colon cancer that she was dying.  After surgery, he met with her and explained that her cancer “had spread not only to her ovaries but also to her lymph nodes. I said that it had not been possible to remove all the disease. But I found myself almost immediately minimizing what I’d said. ‘We’ll bring in an oncologist,’ I hastened to add. ‘Chemotherapy can be very effective in these situations.’

“She absorbed the news in silence, looking down at the blankets drawn over her mutinous body. Then she looked up at me. ‘Am I going to die?’

“I flinched. ‘No, no,’ I said. ‘Of course not.’”

“A few days later, I tried again. ‘We don’t have a cure,’ I explained. ‘But treatment can hold the disease down for a long time.’ The goal, I said, was to ‘prolong your life’ as much as possible.

“I’ve seen her regularly in the months since, as she embarked on chemotherapy. She has done well. So far, the cancer is in check. Once, I asked her and her husband about our initial conversations. They don’t remember them very fondly. “That one phrase that you used—‘prolong your life’—it just . . .” She didn’t want to sound critical.

“‘It was kind of blunt,’ her husband said.

“‘It sounded harsh,’ she echoed.  She felt as if I’d dropped her off a cliff.”

Once you realize that Atul Gawande, couldn’t “find the right words,” you begin to understand: talking to a patient about dying is an art, like dying itself. It is not a matter of simply asking for an end-of-life directive. And it is probably not something that every physician can or should try to do. But every terminally ill patient needs a caregiver who is trained in helping a patient deciding how he wants to die, while helping the family accept the fact of death. As Dr. Susan Block, a palliative-care specialist at Gawande’s hospital later explained to him: “‘A family meeting is a procedure, and it requires no less skill than performing an operation.’

“One basic mistake is conceptual,” Gawande writes. “For doctors, the primary purpose of a discussion about terminal illness is to determine what people want—whether they want chemo or not, whether they want to be resuscitated or not, whether they want hospice or not. They focus on laying out the facts and the options. But that’s a mistake, Block said.

“‘A large part of the task is helping people negotiate the overwhelming anxiety—anxiety about death, anxiety about suffering, anxiety about loved ones, anxiety about finances,’” she explained. “There are many worries and real terrors.” No one conversation can address them all. Arriving at an acceptance of one’s mortality and a clear understanding of the limits and the possibilities of medicine is a process, not an epiphany.

“There is no single way to take people with terminal illness through the process, but, according to Block, there are some rules. You sit down. You make time. You’re not determining whether they want treatment X versus Y. You’re trying to learn what’s most important to them under the circumstances—so that you can provide information and advice on the approach that gives them the best chance of achieving it. This requires as much listening as talking. If you are talking more than half of the time, Block says, you’re talking too much.  . . .

“Given how prolonged some of these conversations have to be,” Gawande adds, many people argue that the key problem has been the financial incentives: we pay doctors to give chemotherapy and to do surgery, but not to take the time required to sort out when doing so is unwise. This certainly is a factor. (The new health-reform act was to have added Medicare coverage for these conversations, until it was deemed funding for ‘death panels’ and stripped out of the legislation.) But the issue isn’t merely a matter of financing. It arises from a still unresolved argument about what the function of medicine really is—what, in other words, we should and should not be paying for doctors to do.”

I agree. We need to give up the notion that the problem at the center of our health care system is all about money—having enough, not having enough, whether reform will boost the deficit.  The truth is that we have enough money.  What we spend now, per capita, far exceeds spending in any other developed country. The question is not funding—though we must think clearly about how we distribute our health care dollars, and try to be sure that we are spending them in ways that do patients more good than harm.

 But as Gawande suggests, the toughest issues we face turn on questions about the purpose of medicine, and, I would add, the relationship between doctor and patient. Ultimately, that connection must be based on trust, founded on honesty, and tempered by humility. This means acknowledging that often, medicine is not about “cure;” it is about talking and listening, comfort and care. When we let technology take over medicine, we turn a physician into a technician, and a patient into an object—a body that succeeds or disappoints, depending on whether it responds to treatment.  

Gawande sums up his argument about the function of medicine: “The simple view is that medicine exists to fight death and disease, and that is of course, its most basic task. Death is the enemy. But the enemy has superior forces. Eventually, it wins. And, in a war that you cannot win, you don’t want a general who fights to the point of total annihilation. You don’t want Custer. You want Robert E. Lee, someone who knew how to fight for territory when he could and how to surrender when he couldn’t, someone who understood that the damage is greatest if all you do is fight to the bitter end.

“More often, these days, medicine seems to supply neither Custers nor Lees.We are increasingly the generals who march the soldiers onward, saying all the while, ‘You let me know when you want to stop.’  All-out treatment, we tell the terminally ill, is a train you can get off at any time—just say when. But for most patients and their families this is asking too much. They remain riven by doubt and fear and desperation; some are deluded by a fantasy of what medical science can achieve. But our responsibility, in medicine, is to deal with human beings as they are. People die only once. They have no experience to draw upon. They need doctors and nurses who are willing to have the hard discussions and say what they have seen, who will help people prepare for what is to come—and to escape a warehoused oblivion that few really want.”

Sara’s story is only one of many stories that Gawande tells in “Letting Go.” To read the full article, click here.

26 thoughts on “Atul Gawande: “Letting Go: What Should Medicine Do When It Can’t Save Your Life?”

  1. The approach to end of life care in LaCrosse, WI makes a lot of sense, in my opinion. It should be quite easy to replicate elsewhere. I just wonder why it hasn’t been. Are local practice patterns and medical cultures so rigid that they can’t learn anything from outside their own area?

  2. Chris, Barry —
    The link is to the New Yorker article Maggie is talking about, which discusses a program at Gunderson Clinic to improve management of terminal illnesses. Maggie has the link at the end of her discussion, or you can just go to New Yorker on line.
    Gunderson Clinic in La Crosse has implemented a program whereby patients are asked and helped by their doctors to create a “living will” using a series of questions. As a result, 85% of Gunderson patients have living wills, and this results in much more rational management of terminal patients, better for patients and their families and also saving lots of money.
    As to why this is not applied more widely: Gunderson is another of the large multispecialty clinics in the upper midwest where values of cooperation, consultation, teamwork, and being part of a medical community are stressed over ego, independence, and mini-kingdom creation. As a result, Gunderson is in position to adopt and implement institution wide practice standards and to have them adopted by providers. In most more independent practice structures, providers are free to decide they don’t want standards dictated to them and to reject implementation. Since, as the Gawande article stresses, discussions involving planning for the death of patients are well outside the comfort zone for most doctors as well as outside their training experience, many simply choose to ignore the notion, or as Gawande admits about himself, botch and chicken out of discussions.
    These type of discussions are routine in many other counties, and contribute to better, less expensive care.
    As to what we can do in the US, we can get rid of the death panal stupidity, use payer systems to encourage and pay providers to offer these services, and train providers including doctors, nurses, and others to be prepared to offer the services.
    All this is just another way in which our love affair with high tech medicine has prevented us from offering better and more appropriate care and harmed patients as well as the economy.

  3. Barry, Pat S., Chris
    Barry– If you read Gawande’s article (and I urge you to use the link and read the whole piece in the New Yorker), you will see that just giving a patient a living will does not begin to solve the problem.
    As Gawande and the palliative care specialist (Susan Block) who he quotes make clear, having a discussion about end-of-life issues with the patient and family is a process that involves many dicussions–and the doctor talking to the patient has to use “the right words” and know how to broach the topic.
    As Gawande stresses it’s all about “just talking” but it’s not one short conversation. It’s a process, not unlike pshychotehrapy, which allows the patient to set the agenda, and begin to open up.
    At Gundersen, Linda Briggs, assoc. director of their end-of-life program explains:
    “Good advance-care planning involves much more than filling out a form. It should include healthcare providers and the patient’s family. Patients need to understand their illness and their options — that they can have comfort care, or that certain treatments may have side effects, or may prolong life without improving it. These discussions are especially important at certain critical times, such as when people are diagnosed with a life-threatening illness, a chronic condition worsens, or elderly people become more frail, Briggs says.
    And, at Gundersen, they do not steer patients away from further treatment. For example: ” Recently, she worked with a patient whose kidneys were failing, and was considering dialysis. He had to make his decision within two or three months. When she met with him to talk about his future care, he told her, ‘All I know is, I don’t want to go on dialysis.’ She explained he would die if he didn’t go on dialysis, and talked about what dialysis would be like. They talked about hospice and about palliative care. They also talked about starting dialysis, with the option to go off. Then they talked about at what stage he might want to go off. His wife took part in the discussions. He hasn’t made a firm decision yet, Briggs says, but is leaning toward trying dialysis.”
    Secondly as Pat S. points out, the medical culture has to support acceptance of death.
    This is not what we teach our medical students in med school. We teach them that their job is to “save” the patient. Death equals defeat.
    In places like New YOrk City oncologists, surgeons and others regularly refuse to let a palliative care specialist come near their patients.
    In other words, yes, local practice patterns and medical cultures are extremely rigid. It will take years to change the way many doctors (not to mention patients) think about death.
    As Pat S. says the medical culture in much of the Northwest is unusual. It’s collaborative. Many fewer small private practicxes. Big egos are not encouraged. Competition is actively discouraged. At Gundersen, if a doc says “I could make more money in Chicago,” he is told: “You should go there.” The medical culture is far more patient-centered.
    This is why, as I’ve tried to explain in the past, Kaiser Permanente works so
    well in Northern California and failed in Texas.
    The medical culture in Texas, Florida and most of the corridor from Boston to D.C. will resist most of the health reform ideas that you, I, Chris Johnson, Pat S. or Don Berwick would endorse.
    (It’s worth noting that both Chris and Pat S. have spent their careers practicing medicine in the Northwest. Don Berwick has spent much of his career fighting the arrogance and lack of patient-centeredness at our marquee academic medical centers. (He and Jack Wennberg have been disappointed that academic medical centers didn’t embrace the values of reform long ago. Unfortunately, most AMCs are too competitive and too focused on collecting revenues, amassing endowments, etc. There are, of course, exceptions: UCSF, Dartmouth . . . Note that they are not located in the highest-spending regions.
    The medical culture in the South is dominated, to a fair degree, by for-profit-hospitals which, by and large, are not leaders in palliative care. In the South religious beliefs and racial gulfs also complicate matters. (Studies show that African-Americans do not entirely trust white doctors and nurses -they fear that white health care providers don’t value the lives of their gradparents or parents as highly as they value the lives of elderly white people, and so many be less inclined to “do everything possible” to save them.
    Pat S.– Well put.
    I would add only that Gundersen is an integrated health system–it’s both a health plan (an insurer) and the provider.
    This seems to work very well–I think of Geisinger. (Some of Geisinger’s patients are insured by other insurers, but many are insured by Geisinger.)
    Again, it’s a matter of the provider and the payer collaborating–and realizing that the least expensive care is usually the highest quality care. You don’t aim directly at saving money (see my comment to Barry above and how the Gundersen nurse doing palliative counseling encourages the patient to consider dialysis); if you aim at quality–patient centered, collaborative care– the savings follow.
    Good to hear from you.
    Yes, Gundersen is an interesting example, I hadn’t heard about them before. See my comments to Pat and Barry.

  4. Maggie,
    I did read the whole article. The part on Gundersen was toward the end, in fact. I think I’ve learned enough over the last couple of years to understand what palliative care is about and that living wills aren’t enough. They do get the conversation started, however. Also, while some people may need a number of sessions with a palliative care team to figure out what they want and don’t want and what’s important to them “when time is short,” others, including myself, may not need much time at all because we’ve already given the issue considerable thought. In La Cross and at Gundersen, at the end of the day, end of life care costs a bit more than half of the national average. That’s a big deal, I think, especially since what they do could be easily replicated if doctors were inclined to move in that direction and payers were willing to pay for the consults. Pat says these conversations are routine in other countries. They should be here as well and long before serious illness strikes. At the very least, even if insurers don’t routinely pay for the consults now, case managers could probably authorize paying for them for those who become seriously ill even if standard Medicare continues to not cover them.

  5. Another outstanding piece from Dr. Gawande. This guy (and maybe the New Yorker) should be nominated for something really great.
    As a non-medical caregiver in my post-retirement life I have seen a range of end of life issues. I can affirm your assessment of Southern health care and attitudes.
    ►”The medical culture in the South is dominated, to a fair degree, by for-profit-hospitals which, by and large, are not leaders in palliative care. In the South religious beliefs and racial gulfs also complicate matters.”
    The whole article is a rich read, but this jumped out at me:
    ►”Like many people, I had believed that hospice care hastens death, because patients forgo hospital treatments and are allowed high-dose narcotics to combat pain. But studies suggest otherwise. In one, researchers followed 4,493 Medicare patients with either terminal cancer or congestive heart failure. They found no difference in survival time between hospice and non-hospice patients with breast cancer, prostate cancer, and colon cancer. Curiously, hospice care seemed to extend survival for some patients; those with pancreatic cancer gained an average of three weeks, those with lung cancer gained six weeks, and those with congestive heart failure gained three months. The lesson seems almost Zen: you live longer only when you stop trying to live longer. When Cox was transferred to hospice care, her doctors thought that she wouldn’t live much longer than a few weeks. With the supportive hospice therapy she received, she had already lived for a year.”
    Too little is known and understood about hospice. Even in professional medical circles it’s sometimes called “the H-word.” There is a widespread misconception that hospice is only for the last few days or hours of life. But I have been told that there are many reasons someone may be in hospice (palliative) care, and they don’t all die. I heard that upwards of thirty percent of hospice patients do not die but are discharged, having “recovered” from whatever condition put them into palliative care. (This sounds suspicious, but it was passed on in a matter of fact manner.) Maybe you can find out more accurate stats, but I recall reading somewhere that hospice patients are sometimes taken off the list thanks to “recovery,” not death.

  6. I don’t want to imply that living wills are the solution to this issue. As the article makes it clear, at Gunderson the living will is used as a starting point for a discussion, not an end point.
    Here is a brief quote from the article:
    “It became routine for all patients admitted to a hospital, nursing home, or assisted-living facility (run by the Gunderson system) to complete a multiple-choice form that boiled down to four crucial questions. At this moment in your life, the form asked:
    1. Do you want to be resuscitated if your heart stops?
    2. Do you want aggressive treatments such as intubation and mechanical ventilation?
    3. Do you want antibiotics?
    4. Do you want tube or intravenous feeding if you can’t eat on your own?”
    As the subsequent discussion in the article makes clear, this is just used as a foot in the door to begin a discussion of terminal care issues, but it does have the effect of getting the discussion started. Because it is routine for all patients, it is asked of everyone. Because the answers are then in the hospital chart, it can then serve as an opening for further discussion.

  7. Regarding the end of life care at the Gunderson Clinic, it is also important to note that the served patient population has smoking and obesity rates in line with the national average. Yet, their life expectancy is one year longer than the national average. And, as a result of the extensive use of living wills and palliative care, end of life care costs just over half of the national average even though anyone who wants all the heroic treatment, along with the accompanying side effects, can have it. This is the clearest proof that I’ve ever seen that more care is not better care at the end of life even though the conventional wisdom was always that people who choose hospice care were likely to die sooner than those that opted for aggressive treatment.
    Dr. Gawande’s article, especially the section about the approach used at the Gunderson Clinic, should be required reading for every physician in the country. Doctors in the Boston – DC corridor, South FL, LA, TX and Southern CA should take particular note. The folks at Gunderson figured out how to bring true shared decision making to end of life care. Their approach should be widely copied and regional differences in end of life medical outcomes should be widely publicized.

  8. “You got to know when to hold ’em.
    Know when to fold ’em.
    Know when to walk away,
    Know when to run …”
    K Rogers

  9. stop smoking help, Barry, Pat S.
    Yes, I would like to see more television programming about hospice and palliative care.
    And, just as in your case, that third person who is not a family member can really help prevent potentially painful disagreements within the family.
    Just one caution about end-of-life directives.
    You and I may think we have given the issue of end-of-life care a great deal of thought–and we have.
    But that doesn’t mean that our thinking might not change when actually faced with the imminent possibiilty of death. People tell me that a survival instinct kicks in.
    I think much depends on whether you’re in pain, how much time a possible treatment might give you (2 months or 2 years?), family circumstances (an spouse who is not dying, but is chronically ill and depends on you), a grandchild about to be born . . .
    I could go on.
    So while persuading people to sign end-of-life direcives while they are still healthy could start a very useful conversation, we have to realize that when they become sick, those directives may change.
    And once someone is very ill, handing them a directive to fill out seems to me a pretty abrupt way to start the conversation . . .
    Finally, I definitely agree that the Gunderson model shows that end-of-life care CAN be handeled in a compassionate and rational way, giving the patient a chance to share in the decison-making. We don’t have to “ration” end of life care, but we do need to make hospice and pallative care available to everyone who is seriously, perhaps terminally ill. We save money and spare patients and their famlies needless suffering. What more could we wish for?
    That said, as discussed earlier, the medical culture is rigid in many parts of the country. Palliative care and hospice are gaining acceptance, but change will take time . .
    If we don’t recognize that, we will become frustrated and feel that reform is “failing.”
    I hope to see real structural change in how care is delivered–and how we pay for care– over the next ten years. . .
    Pat. S.-
    Would Gunderson hand a patient a questionaire if he came in gasping for breath, or a half hour after having suffered what appears to be a debilitating stroke?
    I’m assuming the answer is “no” . . . (We don’t want to scare patients to death).

  10. “And once someone is very ill, handing them a directive to fill out seems to me a pretty abrupt way to start the conversation . . .”
    In thinking back over the history of our exchanges on this and other healthcare issues, my perception is that I tend to think in terms of how best to deal with patient populations and the overall healthcare system while you focus more on the individual and, often, the most extreme cases at that.
    With respect to end of life care and the Gunderson model of encouraging the execution of living wills and starting conversations about care choices early, consider the following:
    1. A diagnosis of dementia or Alzheimer’s is likely to come years before the conditions take their ultimate toll. Early discussions about what care the patient and family want and don’t want when the quality of life deteriorates and the patient can no longer communicate will likely prove valuable when that point is reached.
    2. A cancer diagnosis at an early stage or often even at stage four can leave sufficient time for the execution of a living will and for palliative care specialists to discuss treatment options and the quality of life implications of each.
    3. Patients can live with congestive heart failure for several years or more. Again that leaves plenty of time for exploration of treatment options, end of life care discussions and the execution of a living will.
    4. The same is true for end stage renal disease (ESRD) and dialysis.
    So, while it may not be appropriate to start an end of life discussion with someone who shows up at the ER with a severe stroke, the Gunderson model can work perfectly well in the overwhelming majority of end of life cases, most of which involve elderly people.
    As for the rigidity of medical cultures, I recommend transparency around the combination of medical outcomes and resource utilization. Perhaps financial incentives can be refined to encourage more providers to adopt the Gunderson approach. In this context, I think the sunshine of transparency is likely to prove to be the best disinfectant.

  11. Maggie –
    No, they would not hand the patient a questionnaire if the patient was in clear distress.
    The whole point of the questions – and from what I read in the article it sounds like they are not a questionnaire but rather a guideline for providers to initiate the discussion of life support management – is that that information is on the chart BEFORE the patient faces serious illness.
    As Gawande goes on to point out in the article, the fact that the information is already on the chart and can serve as the cornerstone for discussion of end of life issues is the main value of holding these discussions in advance. The article features a section in which a Gunderson intensive care specialist talks about how the presence of the information elicited in discussions that occur before a crisis occurs allows much better management at the time of crisis, and in particular creates a setting in which the doctor, the patient, and the family all have an entry point for serious discussion of end of life strategies.
    This works well at Gunderson partly because Gunderson is a partly closed system – as a major provider in a small city, a large number of the patients who arrive at the hospital in distress are patients who have been seen by the system before. Although Gawande does not discuss this part of the picture, I think that this provides an important illustration of the value of “medical homes” in creating a setting in which the patient has caregivers who are very familiar with their needs and wishes.
    I will say again that I do not believe that “living wills” are the be all and end all of management of end of life care. Rather, as Gunderson has shown, they can serve as a valuable jumping off point to initiate a discussion that is difficult for providers, patients, and families. That is why the early versions of the Federal Health Care Plan contained a provision for holding exactly the same sort of discussion that Gunderson does. Unfortunately, opponents used that provision to vilify the plan, calling it “death panels.”
    Gunderson has shown that by requiring providers to hold a discussion of end of life issues before a crisis occurs, and by recording the results of the discussion in what they choose to call a living will, that they can improve the process of management of severe illness and death, actually have superior medical results as measured by outcomes, and incidentally lower costs very significantly.

  12. Pat S.
    I agree with everything you’re saying.
    My point was only that, these days, when someone is admitted to the hospital, as an in-patient, they are probably very sick.
    And so, as you say, this would not be the ideal time to initiate the discussion.
    It would be best if doctors begin talking to patients about options before they fall ill–recording what they say on their chart. And since Gunderson is a partially closed system, that is likely to happen.
    In NYC, by contrast, if I have a heart attack and an ambulance takes me to the hospital, no one will know what my PCP might have recorded about my last wishes.
    Unless I carry my “living will” with me (or my husband has the presence of mind to bring it along the hospital won’t have a clue.
    In that case, my best hope is that the hospital has a palliative care team and lets me (or at least my husband) talk to them once I’m stablized.
    Ultimately, I think reform legislation should be revised to
    1) hike pay for palliative care teams (if they were paid more, hospitals would be more likely to offer palliative care)
    2)require that hospitals of a certain size have palliative care, and 3) require that med students take a course in death and dying.
    I’d also be happy to see PCPs talk to patients, while they are still healthy, about their wishes. But at this point in time, I don’t think most PCPs are trained to do this well (If Gawande can’t find the words . . )
    I also think that what a patient says when he is healthy may not coincide with what he will want if he is seriously ill. I’m told that when you’re facing death, a survival instinct often kicks in.
    (Or you may simply change your mind. For instance, my “Living Will says” I don’t want a feeing tube. Since then I have heard that patients who refused the feeding tube, and survived, report that they felt they were starving to death.
    I’m not so sure that I want to experience the pangs of starvation while dying.
    Finally, if a PCP isn’t trained to draw a patient out, I can imagine a conversation where a PCP asks a healthy 60-year-old — “Would you want the doctor or hosptial to take extraordinary measures to keep you alive?”
    The 60-year-old says “No–I wouldnt’ want to be hooked up to a lot of machines. ”
    The doctor nods (as if to say, “right answer”)
    –and probes no further.
    That’s why I think that a conversation with a PCP cannot replace counseling with a palliative care specilaist. Probably the most important healthcare I will ever receive in my life is the care I receive when I am dying.

  13. John Ballard, Barry
    John–Thanks– and thanks for confirming what I had read and heard about the South from doctors down there.
    On people recovering while in hospice– I haven’t read that. Some people live longer than expected but the vast majority die. (A prerequisite for being admitted to hospice is to be suffering from a terminal disease, and certain enough that you are dying that you are willing to stop all treatment except treatment to keep you comfortable and out of pain.)
    I suspect that what you read was referring to palliative care (or confusing hospice and palliative care.) Palliative care is for patients who are very ill and might–or might not–be dying. They are usually in enough pain that they need pain management as well as counseling to help them decide what options they want to pursue if it turns out that they are dying–and to help them face that possibility.
    Many of these patients will survive, and go home.
    Barry-I agree that what Gunderson does would work in many places.
    On rigid medical cultures, you write: “I recommend transparency around the combination of medical outcomes and resource utilization. Perhaps financial incentives can be refined to encourage more providers to adopt the Gunderson approach. In this context, I think the sunshine of transparency is likely to prove to be the best disinfectant.”
    Many doctors in places like Manhattan are not concerned about resource utilization. If you tell them that their hospital spends much more per-patient, they will say: “That is becuase we give better care. And our patients demand the best.”
    (I’m sure you’ve read quotes from Manhattan doctors in the NYT saying just that.)
    These doctors are concerned with the one patient they are treating-not the ramifications for society as a whole.
    If you tell them: this nation just can’t affod to spend 20% of GDP on health care, they will say: “What is more important than your health? Sure, health care is expensive. People keep complaining about it, but then they take their kids to Disney World! And they think I shouldn’t be paid what I’m worth (say $500,000 a year.)
    I have had this conversation with more than one Park Avenue specialist. They are not willing to give up their belief that “You can never spend too much on healthcare” –especially because it supports their lifestyles
    And try talking to them about overtreatment. They flatly deny it.
    Finally, when someone is earning that much (which is not unusual in the most rigid medical cultures– Boston to D.C., Florida, Southern California, parts of Texas– it is hard to come up with financial incentives that are large enough to persuade them to change how they practice medicine.
    They also are not easily pressured by hospital administrators. Hospitals need the well-insured patients that these specialists bring in.
    In some places, we may have to wait for a generation of doctors to retire. They are not going to join large medical organizations; they are going to remain in small private practices doing things exactly the way they think best–the way they have always done them.
    Their patients have great faith in them–partially because they charge so much.
    I do agree that, when it comes to end-of-life issues, I tend to focus on the individual rather than the larger cost to society. I believe that if you focus on giving the highest quality of care, the savings will follow.
    And I believe that dying is one of the most important parts of life. It’s a time to be as “patient-centered” as possible, letting each person die in his or her own way. For most of us, that would mean dying at home with palliative or hospice care.
    But if you don’t think of your home as a “safe haven”, you might prefer to be in a hospital–or even in an ICU. (Imagine a person who has been marrked for 50 years, and over the years he and his spouse have fought constantly. Familiarity has bred great contempt. His wife tells the doctor: “I don’t want him back in my house.”(I know of a case where this happened. Attempting palliative care at home probably not a good idea.

  14. Maggie,
    I hear what you’re saying about the NYC doctors. I also hear from a number of my colleagues who live in Manhattan that lots of doctors, especially surgeons, don’t take any insurance at all. Their fee is their fee and whatever insurance pays, it pays. They expect the patient to cover the difference. The mother of one colleague who recently moved to NYC from NJ called eight different doctors in an attempt to find a primary care doctor who accepted Medicare. All eight either didn’t take Medicare at all or weren’t accepting new patients.
    I think one approach we should consider is to have beneficiaries pay a Part B premium that varies by the county that they live in. Let them pay 25% of the actual cost of Part B services in their county of residence with a circuit breaker provision that would limit their premium to some reasonable percentage of their income. Everything else we buy, including all other types of insurance, varies by where we live and other factors. If NYC residents had to pay twice as much for their Part B benefits as people in La Crosse, WI, they might start to ask their doctors why costs here are so much higher when outcomes are no better and in some ways worse.
    Insurers who sell Medicare Advantage plans are already paid different rates in each county they do business in, along with appropriate risk adjustment payments. It makes no sense conceptually for people across the country to pay the same nominal dollar premium for Part B benefits when medical input costs are different, practice patterns vary, and beneficiaries’ previous wages and salaries varied for similar work as well. People in high cost locales earn higher nominal dollar wages and ultimately receive higher nominal dollar social security benefits than those who did similar work in lower cost regions. Part B premiums should reflect that differential.

  15. The part of Gawende’s piece that struck me is the distinction he makes is between what we should do (but too often don’t) and what we do that may not be necessary. A lot has been said about the high cost of end of life care, and a lot of emphasis has been placed on the cost savings that could result in forgoing that care. But the fears that efforts to reduce costs will lead to “rationing” are real, weather you agree with the use of that term or not, and a focus on increased access to palliative care seems like the better approach. If the evidence he points to is true, then increasing access to palliative care improves outcomes while decreasing costs. That seems like a no-brainer and begs the question why we still have such hurdles in the way of hospice/palliative care like poor reimbursement and a requirement that patients stop “traditional” care in order to enroll in hospice.

  16. Barry, Word
    Barry–People in New York City, California, Florida, Texas, etc. will never pay higher premiums for Medicare than people in Iowa, Wisconsin, etc. for one simple reason: Senators from N.Y. California,etc. have far more power in Congress.
    Also brand-name hospitals and specialty groups in these states have great power–very strong lobbies.
    Logically, your argument makes sense, but politically it would never fly.
    If Medicare law had stipulated that people in states where both the cost of health care and wages are higher would have to pay higher premiums, Medicare legislation would never have passed.
    It’s really that simple.
    On docs in NYC accepting Medicare: the vast majority do take Medicare.
    But the vast majority of primary care docs are not taking new patients period–whatever insurance you have. Go to their waiting rooms–they have more patients than they can handle. You may wait an hour or more.
    Specialists, on the other hand, are taking new patients. And brand-name
    specialists continue to take Medicare. They need the business. (I find waiting times are usually quite short in the offices of Park Avenue specialists.)
    The problem is not Medicare, but a shortage of primary care doctors.
    Yes, expanding palliative care does seem like a no brainer–from the point of view of the patient.
    He or she is in control; the patient makes the final decisions; the patient is not in pain. Often, patients who undergo palliative or hospice care live longer than those who stay in the hospital or ICU.
    What’s not to like? Many doctors and hospitals fear that they will lose revenue. Someone is paid for those extreme, often very lucrative treatments during the last two weeks of life: chemo, another surgery, whatever. A hospital needs to keep its ICU beds full in order to make money on the ICU.
    Also, some doctors truly believe that palliative care stands in the way of letting them “save” the patient. There are surgeons and oncologists out there who tell palliative care docs: stay away from “my” patient.
    Often patients don’t know that palliative care is available.
    And finally, conservatives have played on our fear of death by presenting palliative care teams as “death panels”who are trying to hasten the patients’ death in order to save money.
    That said, I am hopeful that more and more hospitals and doctors are going to make sure that patients are aware that palliative care is available. As bloggers and jouranlists write more about palliative care, the public becomes more aware, and both patients and relatives are more likely to ask about it.
    My guess is that Gawande’s article will have a real impact.

  17. “Senators from N.Y. California, etc. have far more power in Congress.”
    CA and NY have no more senators (two) than any other state though they do have more House members.
    Difficult times call for difficult measures. If you look at what’s happening with state and local finances, you see everything from furloughs to pay cuts to serious pension reforms including at least one reform in CO that affects CURRENT retirees – a lower than promised cost of living increase. In NJ, we’re seeing all sorts of successful initiatives from the Christie Administration that was considered politically impossible in the past. It may take a crisis, fiscal or otherwise, but substantive reform of entitlement programs is both possible and likely in the fairly near term, in my opinion. Fiscal trends that can’t continue because they’re unsustainable won’t continue indefinitely.

  18. Barry–
    Yes, every state has 2 senators.
    But Hillary Clinton (who respresnens N.Y.) has far more political power than the two senators from Arkansas combined.
    I’m sure you realize that the Senate is not about one man (or woman) one vote.
    As for fiscal trends that are unaffordable–they have continued for a very long time–and may well continue, taking the country down with them.
    (See the history of other once-prosperous empires.)

  19. Barry– P.S.
    I should add, I’m not comparing the U.S. to Rome or Greece, but to the U.K.
    Many historians suggest that the U.S. is going to wind up like the U.K. after WW Ii.
    We’ll still be a world power, but not #1 or #2 . . .
    The standard of living here will be lower than it has been over the past 30 years.
    We won’t have the same clout in global finance.
    But I wasn’t suggesting
    that the U.S. will windup
    in rubble–just that we’re coming to the end of an era, and, by and large, are blind to that fact.

  20. Maggie,
    I wasn’t going to comment any further on this thread but I wanted to mention the very informative article about the Swiss healthcare system in the most recent issue of Health Affairs which I received today. I found the following points noteworthy:
    1. Insurance premiums are charged per person in the family with one rate for adults and one for children. Thus, large families pay significantly more than small families and childless couples. Deductibles can vary within certain limits.
    2. About 40% of the population qualifies for a subsidy to help pay the insurance premium while part of the cost of operating hospitals is paid by general taxation.
    3. Overall, about one-third of Swiss healthcare costs are covered by health insurance premiums paid by the person or the family, a bit more than one-third is paid for with general taxes which finance health insurance subsidies for lower income people and a portion of hospital operating costs, and 28% is paid out-of-pocket by patients – deductibles, co-pays and services not covered by insurance including dental fees and long term care costs.
    4. People who live in the more expensive cantons pay higher insurance premiums for the same coverage than people who live in lower cost cantons.
    5. The risk equalization mechanism is far from perfect but attempts to at least partly compensate insurers who wind up with sicker, higher cost members.
    6. Price differences per service, test, procedure and drug are the primary factor that accounts for higher healthcare costs in the U.S. vs. Switzerland. According to the Swiss expert, many of their hospitals resemble five star hotels and the overall structural healthcare infrastructure is quite high in quality.
    7. There are 84 insurance companies in Switzerland with the six largest accounting for about 80% of the market.

  21. I’m 62 years old and my right knee joint has been in chronic, unremitting pain for months now. Due to the usual Health Insurance madness I can’t even get a network provider in my own state [CA]. Made an appointment at a big deal clinic in AZ, but suspect they may, contrary to what they told me, not qualify. I suspect another factory farm health care institution.
    I am ready for palliative care RIGHT NOW! It is well established that joint replacement surgery is serious, and does nothing for function. It is only for PAIN. Yet the rigid healthcare establishment is prejudiced against opiate derived medication [which works] and niggardly about prescribing it. I will have to travel out of state anyway. Can anyone recommend a sympathetic doc? Email:

  22. Good discussion, Maggie and commenters. In the palliative care community, it’s believed that the relative homogeneity of LaCrosse contributes to the effectiveness of advance care planning there.
    Also, efforts to instigate timely ACP conversations can now focus on Physician Orders for Life Sustaining Treatment (POLST, known in NY as Medical Orders for Life Sustaining Treatment or MOLST). Developed for use by people with advanced illness or frailty, POLST addresses the limitations of DNRs and other traditional practices for communicating treatment preferences. And it works! A recent study found that “POLST orders restricting medical interventions were associated with less use of life sustaining treatments” for dying patients. See Journal of the American Geriatric Society 58:1241-1248, 2010 for details.

  23. If we help patients, families, and caregivers to ask better questions we’ll all get better answers.

  24. Pingback: Angry Bear » Breast Cancer: Catching Up With Amy Berman, a Woman Who Chose Life Over Longevity