Development of skills for end-of-life discussions ‘a lifelong process’

The American Cancer Society estimates more than 589,000 Americans will die of cancer this year, making end-of-life conversations an inescapable reality for oncologists and other providers on their care teams.

These often are the most difficult discussions clinicians have with patients and their caregivers due to the uncertainty about when they should be initiated, the range of complex issues that must be addressed, and the delicate balance that must be struck between clearly explaining medical evidence and providing emotional support.

Medical societies and education programs offer guidance, but standard parameters may only go so far. As with cancer treatment, a personalized approach often yields the greatest benefit, according to Jean M. Youngwerth, MD, assistant professor in the department of medicine and internal medicine and director of palliative care at University of Colorado.

In each conversation, clinicians must strive to identify patients’ priorities — Do they prefer more time vs. better quality of life? Do they wish to prolong their life by any means necessary or spend as little time as possible in the hospital? — and establish what they are most worried about, experts suggest.

“We want to learn each patient’s story,” Youngwerth told HemOnc Today. “That way, we can understand who this person is as opposed to just understanding the cancer. We want to learn about the values in their life and how they make decisions. What are their hopes and worries and how they have drawn strength in their life? This helps us to understand how the illness has affected them, how we can best address their needs and what is most important to them.”

HemOnc Today spoke with oncologists and palliative care specialists about the ideal timing of end-of-life conversations, strategies to help clinicians navigate sensitive topics to address the needs and desires of patients and caregivers, the importance that providers manage their emotions while providing clinical compassion, and the need for more training to help clinicians communicate more effectively with patients who are in the final stages of life.

‘No such thing as 100% certainty’

The first challenge associated with end-of-life conversations is determining the appropriate time to initiate the discussion.

National guidelines recommend advance care planning for terminally ill patients with less than 1 year to live, but study results reveal gaps with regard to when clinicians and patients discuss these issues.

Nancy L. Keating, MD, MPH, professor of health care policy at Harvard Medical School and associate physician at Brigham and Women’s Hospital, and colleagues conducted a national survey of more than 4,000 physicians who care for patients with cancer to assess when they speak with terminally ill patients about several factors, including prognosis, use of hospice, do not resuscitate (DNR) status and preferred location of death.

Researchers asked physicians to imagine how they would interact with a patient who had 4 to 6 months left to live but was asymptomatic and still feeling well.

The results — published in Cancer — revealed 65% of physicians would discuss prognosis immediately, but fewer would initiate immediate conversations about DNR status (44%), hospice (26%) or preferred site of death (21%). The remainder indicated they preferred to wait until patients felt worse or had no other treatment options.

The researchers did not ask participants to explain the reasons for their responses, but physician disagreement with the guidelines is one possibility.

“Or they may not be comfortable discussing it,” Keating said in a press release. “These conversations are time-consuming and difficult. Some doctors may feel patients will lose hope. It’s easier to say, ‘Let’s try another round of chemotherapy’ instead of having a heart-to-heart discussion.”

That approach creates the potential for problems, including unrealistic expectations about outcomes and a lack of emotional preparedness on the parts of patients and their families, experts said.

“One thing is for sure: Even if a disease is curable, there is no such thing as 100% certainty of it being cured,” Rizwan Romee, MD, assistant professor in the bone marrow transplantation and leukemia section at Washington University School of Medicine in St. Louis, told HemOnc Today. “From the very beginning — the day of diagnosis — you should start to have this conversation about at what point you are going to call it quits.”

The family element

The decision to have end-of-life conversations can be more difficult if caregivers or other family members push patients to exhaust all treatment options.

Arif H. Kamal, MD, director of quality and outcomes at Duke Cancer Institute, described a recent appointment during which a 75-year old military veteran with advanced lung cancer sat silently in his office.

The patient’s wife did all of the talking. Despite her husband’s poor prognosis, she pushed for further treatment in hopes of improving his chances for survival.

These scenarios are becoming increasingly common, Kamal said.

“Family members are often the ones who will request therapies or push for interventions that the patient may not be ready for or may not want to go through,” Kamal said. “It’s important to identify and understand that the family member is advocating for their loved one. But, it’s also important to understand the patient’s own hopes and fears.”

If this disconnect is left unresolved, tension can build between patients, their loved ones and members of the care team.

The best way to address this is to ask caregivers three questions, Kamal said. They are:

What is your actual understanding of what is happening with this illness?

When you think of the future, what do you hope for your loved one?

When you think of the future, what are you most worried about?

“Often times, when you have communication simultaneously with caregivers and patients, you’ll find that you are introducing a conversation that they’ve never actually had,” Kamal said.

Although these conversations often are necessary, they may lead to more emotional conflict, especially because they often occur in sterile or uncomfortable settings such as hospitals or physicians’ offices.

“It’s a spectrum,” Romee told HemOnc Today. “If the patient is taking that news about the diseases well, sometimes the family does not. It goes through all those phases of denial — being angry and being upset. Ultimately most patients and families accept it, but it is difficult to hear … and people take the news differently.”

Clinicians are taught how to identify indictors that suggest treatment has been unsuccessful and few options are left to treat the disease, but patients and their families are not.

“It’s tough across the board, and it is one of the more frustrating things for a [medical] team,” Mark Speyer, MD, MPH, assistant professor, associate medical director for home care/hospice and medical director of palliative care at Loyola University Hospital in Chicago, told HemOnc Today. “What is the underlying reason these patients or caregivers are making these demands? Are they doing it because they want to live long enough to see their daughter graduate, or some other life milestone? You don’t know until you ask.

“Or maybe they’re really afraid,” Speyer said. “Existential suffering is something that needs to be dealt with, just like physical or psychological pain.”

Physician compassion

Many clinicians who develop a good relationship with their patients, or those who want to appear more compassionate — and, in turn, be viewed by their patients as providing a higher quality of care — may overestimate their survival projections for patients as they near the end of life.

Clément-Duchêne and colleagues evaluated data from clinicians who treated patients with advanced lung cancer, and results appeared in The Oncologist. The researchers determined the physician consultants predicted median survival of 25.7 weeks (standard deviation, 25.9), whereas the actual median survival of patients was 11.7 weeks (standard deviation, 26.8).

A study by Christakis and colleagues, published in The Western Journal of Medicine, evaluated the survival predictions of 343 physicians for 468 terminally ill patients. Results showed only 20% of the predictions were accurate — defined as coming within 33% of the actual survival time — and 63% of predictions were overly optimistic.

“Studies show us repeatedly that we are overly optimistic, and I think it’s because we want people to do well and live longer,” Youngwerth said. “We overestimate. We know that. Sometimes, if you have a close relationship with your patient, you need to consciously be more aware of your tendency to overestimate, as we may be clouded by hoping they will do better than we really believe. Clinicians with more experience tend to be more accurate — we teach learners to think about prognosis for all of their patients. Physicians don’t want to be wrong for many reasons, but we know predictions are only estimates and we acknowledge we’re likely to be wrong, and need to be comfortable with that.”

A study by Tanco and colleagues — published in February in JAMA Oncology — indicated that patients associated physician optimism with a higher level of compassion and preferred physicians who were optimistic (P < .001 for both). Further, patients considered physicians who were optimistic as more trustworthy (P = .03).

“It is concerning that studies are showing that doctors tend to be optimistic, but patients interpret what doctors tell them differently sometimes,” Romee said. “They are so desperate and want to latch on to whatever hope they can until the very end. That can create a gap between what doctors are telling them and what they and their families are interpreting.

“It’s a glass-half-full way of looking at it,” Romee added. “If I tell a patient that their chance of 5-year survival is 60%, they will latch on to that 60% until the very end, even if we know after 3 months the patient is not responding to treatment and is unlikely to be in that 60% group. What we all should be doing is having this conversation time and again, and not just in that initial setting. We should be updating our predictions because things change over time.”

Yet, others suggest that some overly optimistic predictions are offered intentionally because physicians know their patients well enough to realize that is what they want to hear, and it may make them more comfortable as end of life approaches.

“When patients are coming to an oncologist, they often are looking for hope,” Christopher Pietras, MD, an assistant professor and palliative care specialist in the department of medicine at the University of California, Los Angeles, said in an interview. “Oncologists are aware of that, but what happens when the curative portion is no longer possible? How can we redefine hope? It’s very difficult to bring a realistic viewpoint to a patient who may not want to hear it because it usually feels pessimistic. … Trying to force a realistic viewpoint onto someone who doesn’t want to hear it isn’t beneficial to anyone.

“I would argue that, when a patient hears an overly optimistic viewpoint, it may be just what they want to hear from their provider, and the provider knows from discussions with their patient that they do not want to hear something else,” Pietras said. “But we shouldn’t stop there. It is our responsibility to help ease our patients into these difficult discussions, so that their understanding of their illness and the negotiated plan of care is meaningful.”

A study conducted by Shin and colleagues — published in February in Psycho-Oncology — indicated patients with cancer and their family caregivers may again differ in their expectations for end-of-life discussions. Results showed patients preferred disclosure, direct disclosure by a physician and palliative care options (P < .001 for all). Further, family caregiver predictions of patient preferences and actual patient preferences were comparable for terminal disclosure but significantly differed with regard to family involvement in the disclosure process and end-of-life choices (P < .001).

Patient and family caregiver perceptions on this process may be clouded by other factors, Kamal said.

He believed the patient he previously described — the veteran with advanced lung cancer — was beginning to understand the reality of his illness and that he did not have much time left. Still, the man’s decision to agree with his wife may have been for her benefit.

The man told Kamal and his wife that his biggest concern was about who was going to take care of his wife after he died.

“She broke down crying because this was the first time this very stoic veteran had ever said anything like that to her, and I was really just the conduit for them to have their conversation,” Kamal said.

When Kamal spoke with the man’s wife, he discovered her sole concern was about her husband’s pain and suffering. Length of survival never entered the equation.

In Kamal’s view, physicians must help place more focus on the needs of patients and their families, and less on how much time someone has to live.

“We find that we can meet those goals with other tools that patients and caregivers aren’t familiar with, and it’s our responsibility to tell them about them,” Kamal said. “I told them third-line chemotherapy is not going to help with those things. We decided it may also be advantageous to introduce the concept of hospice because they can be my eyes and ears at home and help with pain management as an extension of my team when it does not make sense for him to come back into the office.

“In that one conversation, which was 30 to 45 minutes, my goal was not to talk him away from chemotherapy. If that’s what he wanted, it would have been perfectly fine with me,” Kamal said. “But, I wanted to make sure everybody in the room heard what their priorities and goals were. Once they did, the rest of it flowed naturally, and the decisions made sense to them.”

Physicians’ emotions

End-of-life conversations take an emotional toll on patients and their families, but they also are difficult for physicians.

Burnout is an inherent risk in the medical profession, especially among clinicians who deal with death regularly.

Asai and colleagues evaluated data from 697 physicians engaged in end-of-life care of patients with cancer. Results of that analysis — published in Psycho-Oncology — showed 22% of respondents had a high level of emotional exhaustion, 11% had a high level of depersonalization, 62% had low levels of personal accomplishment and 20% had psychiatric morbidity. Further, clinicians who did not feel they had sufficient time to communicate with patients demonstrated greater likelihood for all aspects of burnout.

“It’s important to have a wall; otherwise, your clinical decision-making would be badly affected by those kinds of emotional factors,” Romee said. “However, it is hard. As a cancer specialist, I have cried with families. I do shed tears — it’s not uncommon. When we know the patients well and we reach a point where we are quitting the treatment or they die even without quitting the treatment, it’s hard because you have known this patient and their family for months and sometimes even years. It’s one of the most challenging aspects of being a cancer doctor.”

Time management — striking a balance between time spent working vs. at home — is key for clinicians to avoid becoming overwhelmed by loss, Romee said.

Shanafelt and colleagues surveyed 241 oncologists about their well-being and personal wellness promotion. The results — published in Oncology — indicated the development of an approach or philosophy to dealing with death and end-of-life care was one of the strongest factors associated with improved quality of life among physicians.

At UCLA, there are multiple options in place for clinicians to make sure they are not overtly affected by loss of a patient or the challenges associated with having end-of-life conversations.

Physicians routinely practice team meditation as a reminder of how sacred and difficult certain conversations are for some patients, Pietras said. Other support services are available, too.

“Clinicians often want to make sure they’re making the right decisions, and we have them talk through the difficulties they’re having,” Pietras said. “It’s reassuring to know that their decisions are the right ones, or to get a second thought about whether something else can be tried to help the patient. Overall, I think that’s one of the ways we can improve as a specialty — adding another uniform layer of support for the clinicians during each case.”

The University of Colorado offers clinicians the opportunity to express their emotions — whether they be disappointment, frustration, happiness, anger or sadness — during meetings with other professionals to discuss the loss of a patient that may have affected them on a deeper level.

“We debrief after every encounter in the hospital just to get off our shoulders how that encounter made us feel,” Youngwerth said. “When you let it out, it doesn’t weigh as heavy on you anymore. If you attend to your needs and recognize when things are too much or hitting too close to home with something going on in your personal life, being able to recognize that allows us to take the appropriate steps. We have to recognize that doctors grieve, and we need the space to do that.”

Communication skills

Christakis and colleagues showed that clinicians in the top quartile of experience made more accurate end-of-life predictions than less experienced physicians.

Still, the clinicians interviewed for this article agreed more emphasis must be placed on educating physicians about how to communicate with patients.

“It’s a little better than it was 10 years ago, but it’s still considered a taboo topic,” Speyer said. “Providers tend to worry that having to talk to their patients about end of life means that they have failed them.”

Although these conversations can be difficult, patients often are appreciative when providers take time to have this dialogue, Speyer said.

“There needs to be more education,” Speyer said. “The education system is improving and communication skills are starting to become part of the curriculum, but too often students think that they don’t need to learn any more communication skills as they go for their advanced degrees. Well, I’m still learning. Learning is a lifelong process. We should take refresher courses at our annual conferences. Communication is not natural to all people, but it definitely is something that can be learned.”

Large medical societies should support improvement in these areas, Romee said.

“Despite the fact that we all think we handle these situations well, there’s still room for improvement, especially on the training level,” he said. “Even though during training we are exposed to these situations under the supervision of our mentors or the attending physician with whom we are working, ASCO and ASH need to push for sessions on end-of-life care.”

These also can be incorporated into fellowship programs, Romee said.

“During training, a lot of us would take this as an elective to work in palliative care and hospice services,” Romee said. “But making end-of-life communication more compulsory so that most of us in training would have that kind of exposure and experience is going to be critical so we’re exposed to these kinds of situations before we start working independently.” – by Anthony SanFilippo

References:

Asai M, et al. Psychooncology. 2007;16:421-428.

Christakis NA, Lamont EB. West J Med. 2000;172(5):310-313.

Clément-Duchêne C, et al. Oncologist. 2010;15:782-789.

Keating NL. Cancer. 2010;doi:10.1002/cncr.24761.

Shanafelt TD, et al. Oncology. 2005;68:23-32.

Shin DW, et al. Psychooncology. 2015;doi:10.1002/pon.3631.

Tanco K, et al. JAMA Oncol. 2015;doi:10.1001/jamaoncol.2014.297.

For more information:

Arif. H. Kamal, MD, can be reached at Duke University School of Medicine, 8 Searle Center Drive, Durham, NC 27710; email: arif.kamal@duke.edu.

Christopher Pietras, MD, can be reached at UCLA Hospitalist Division, 1250 16th St. Suite A454, Santa Monica, CA 90404; email: cpietras@mednet.ucla.edu.

Rizwan Romee, MD, can be reached at Washington University School of Medicine in St. Louis, Division of Oncology, Campus Box 8007, Washington University Medical School, 660 South Euclid Ave., St. Louis, MO 63110; email: rromee@dom.wustl.edu.

Mark A. Speyer, MD, MPH, can be reached at Loyola University Hospital, 2160 S. First Ave., Maywood, IL 60153.

Jean M. Youngwerth, MD, can be reached at University of Colorado Hospital, 12605 E. 16th Ave., Aurora, CO 80045; email: jean.youngwerth@ucdenver.edu.

Disclosure: Kamal, Pietras, Romee, Speyer and Youngwerth report no relevant financial disclosures.

 

 

Are physicians responsible for initiating end-of-life conversations with patients with cancer?

Physicians should be able to broach this sensitive subject, but they need more training and education on how to do so.

A few months ago, my mother-in-law shared an email she wrote to a good friend who was diagnosed with glioblastoma multiforme. This friend was soliciting advice about facing a life-threatening illness. My mother-in-law shared her experience from her 18-month journey with her husband who died of leukemia. She gave these recommendations: first, spend more time with your family; second, talk about the hard stuff.

As she reflected on their experience, she remembered having many concerns and questions about end-of-life decisions, but because she did not want her husband to perceive a lack of optimism, they were never shared. She wished physicians would have incorporated these conversations early in his care when things were going well. She believes it would have normalized the conversation and allowed both of them to simultaneously hope for the best while planning for the worst. It may have eased later regrets about whether she made the right choices at the time of his death, simplified financial decisions after his death and supported her family through their grief.

A report by Hagerty and colleagues showed the majority of patients feel similarly. People trust that their physician will give them a clear signal when it is time to talk about dying. Though we know there are inherent challenges to prognostication, our knowledge and experience put us at a significant advantage to see our patients’ fates before they can, often at the time of diagnosis. The important question is not who should start these discussions, because it seems clear this is the responsibility of physicians. Rather, it is: Why are we still not preparing physicians for this immense responsibility?

I am not surprised that most physicians feel unprepared to talk about “the hard stuff.” Most of our medical education focuses on disease diagnosis and management, despite an increasing awareness of the importance of effective communication. We are not trained to identify and respond to our patients’ diverse sociocultural, emotional and spiritual needs, so we often avoid these uncomfortable situations. It is not for lack of caring about our patients. In fact, I would argue it is our strong commitment to our patients that presents one of the greatest challenges. We care deeply for our patients and worry about causing emotional harm. We want to avoid cultural missteps and struggle with how to respond to questions of spirituality. We also don’t know how to process our reactions when a patient’s personal values or choices differ from our own.

Our medical education system does not systematically train physicians in these complex communication skills. Research by Temel and colleagues showed people with training in communication around serious illness can improve patient outcomes. Yet, a robust communication curriculum for serious illness is still largely absent from most graduate-level training.

I wish every patient had someone like my mother-in-law to give them sound advice from real-life experience. I also hope more patients are empowered to raise these concerns with their physicians because many will not and cannot wait for our educational system to make this paradigm shift. However, we owe it to our future patients, especially those who may be too exhausted, overwhelmed, alone or vulnerable to be better prepared when the time comes. In doing so, we may find that sometimes our greatest healing occurs even when we cannot cure.

References:

Hagerty RG, et al. Ann Oncol. 2005;16:1005-1053.

Temel JS, et al. N Engl J Med. 2010;363:733-742.

Caroline J. Hurd, MD, is clinical assistant professor of palliative care and hospitalist services and the director of education operations at the Palliative Care Center of Excellence at University of Washington Harborview Medical Center. She can be reached at churd@uw.edu. Disclosure: Hurd reports no relevant financial disclosures.

Patients should initiate the end-of-life conversation when they feel comfortable enough to speak about it with their doctors.

Most Americans prefer to die at home, minimizing the burden to their loved ones. Data support that patients want to discuss their end-of-life wishes with their doctors. The doctors just need to conduct end-of-life conversations with patients and provide care accordingly.

That should be an easy problem to solve, right? Tens of thousands of dollars invested in numerous years of painstaking medical training enables doctors to do the most risky, sophisticated medical procedures consistently and effectively. Extrapolating that, we can assume that doctors will skillfully and easily conduct end-of-life conversations. After all, how hard can it be to just talk to your patients? Or so we thought when we set out to investigate if doctors encountered barriers to conducting end-of-life conversations.

We asked 1,040 doctors from the 11 common medico-surgical sub-specialties, during their last year of training in two large hospitals, if they encountered barriers to conducting effective end-of-life conversations with seriously ill patients and their families. If so, we asked to what extent conducting these conversations is challenging.

We found that 99.9% of the doctors reported barriers to conducting end-of-life conversations and 86% rated them as very challenging. Our participants struggled to talk with patients whose ethnicity differed from their own, and those with low health literacy and limited English proficiency. When patients’ cultural and religio-spiritual background impacted their end-of-life decision-making, doctors often did not know how to navigate these cultural conundrums. When ethnic families forbade the doctor to talk to the patient about her/his diagnosis, the doctors were frustrated.

If 99.9% of any group is having difficulty with a particular task, asking them to lead the task may not the most prudent solution. Patients and families need to know their doctors are likely neither trained nor rewarded to have end-of-life conversations with them. Thus, doctors may never broach this topic; if they do, they likely will stop it cold at the first sign of resistance.

Patients and families need to lead this dance. End-of-life conversations are nuanced and fraught with emotion, and patients are in a good position to initiate the conversation when they are ready. Common triggers for the conversation could be life milestones: when you become eligible to vote; when you get your first job; when you get married; when you have your first child; when your children graduate high school; or when you retire.

Admittedly, it is awkward and challenging to initiate this sensitive conversation with a doctor. We recommend patients use our simple letter format to elucidate the key issues and take the completed letter to their doctor. Next, they should work with the doctor to ensure that their wishes are documented in medical records. And most importantly, they must remember that — when it comes to discussing end-of-life issues — they may have to lead and support their doctor to have a candid discussion.

 

Reference:

Periyakoil VS, et al. PLoS One. 2015;doi:10.1371/journal.pone.0122321.

VJ Periyakoil, MD, is clinical associate professor of medicine at Stanford University School of Medicine and director of the Stanford palliative care education and training program. She can be reached at periyakoil@stanford.edu. Disclosure: Periyakoil reports no relevant financial disclosures.