How do you tell someone that they are dying?

I ask several doctors to share their experiences and the patients that stick in their minds when they tell someone that the are dying or seriously ill. One remembers a woman who came in soon after Christmas. She’d been in and out of the doctor’s over the last nine months or so, with general symptoms that are all too easy to ignore: feeling tired, swelling. Then she had suddenly become jaundiced and short of breath, so a relative brought her into the ED.

It was one of those situations, the doctor says, where you pull the curtain back and immediately think, “This is not good.” “On very few occasions do you touch something and say, ‘This is cancer.” When she examined the patient’s abdomen, it felt “rock hard.”

“She kept saying to me, ‘It’s going to be fine, isn’t it?’ And I’m saying, ‘We’ll do everything we can; let’s just do a few tests and figure out what’s going on.’ At that stage in my mind, I knew it was bad, but I still had to figure out exactly what flavor of bad it was.”

The woman was anxious to be home on New Year’s Eve to make a call to family overseas. But blood tests confirmed that she’d need to stay.“

She said to me, ‘Tell me the worst-case scenario.’ I looked at her. She looked at me. And in my mind, I was thinking, ‘She’s not ready for this diagnosis.’ Then her relative stepped in, and she said, ‘No, no, she means what’s the worst-case scenario in terms of how long does she have to stay in hospital?’

“At that moment, you realize that we all know exactly what we’re talking about, but we’re all accepting it to different degrees.”

Compartmentalisation seems to be important in coping when having to tell someone that they are dying – she mentions several times that bad news, death, is part of the job. “You have to be strong for the family. I can easily go back into the staffroom and cry my eyes out, but at that moment, I have to be there; I have to be the hand to hold or the shoulder for the patient to cry on.”

The doctor speaks of the first patient she cared for, a man with metastatic prostate cancer. She was called into his room by his wife, and as she arrived, the man took his last breath. His wife broke down onto the floor in front of her. “At that moment, you have to say, ‘I’m sorry, he’s gone.’ And you kind of just has to suck it up and be there for them at that moment because that moment is everlasting for them.”

The scenarios are given out. The students’ eyes hurriedly scan down their sheets of paper. Cringes. Sharp intakes of breath. One scenario is based around telling a parent that their son has suspected schizophrenia. 

Another is that a person’s relative has unexpectedly died in hospital and they need to tell the family that the person has died. One peer over his neighbor’s shoulder. “You got the short straw,” he says, shaking his head.

When the time comes, his neighbor heads for the door. “You might want to take in some tissues for this one,” the facilitator tells him. “Yeah, for me,” the young man replies.

The rest of the group watched on a live video link, seated in plastic chairs in a half-moon in front of the screen. They touch their faces self-consciously, arms crossed protectively across their bodies. The speaker hisses.

In the room sit a couple. The woman fiddles with her handbag, refusing to accept the news. The man stares silently at his hands, fists clenched around his mobile phone. “Is that true, or just something you say?” he says aggressively to the young doctor. He’s just been told that his baby, born at just 26 weeks, has severe brain damage and is unlikely to survive.

The students’ eyes dart to each other across the room. A shake of the head, a smile, a wince, sensing their colleague’s pain – the pain they are all going to feel for themselves sooner or later. 

Perhaps unsurprisingly, researchers at Aristotle University in Greece, when investigating how doctors tell patients they have cancer they found that telling a (pretend) patient that they had cancer was more stressful for a doctor than concealing the diagnosis. Doctors who don’t tell the truth, they suggest, may be doing so to keep control of the situation and avoid their and their patients’ emotional reactions.

The stress of having ‘bad news’ conversations can lead some doctors to put them off, or to deliver news in a less-than-optimal way, says Dr. Laura-Jane Smith, a respiratory registrar in training who works in London.

Finding the right time and place to have awkward conversations about things such as the progression of the disease can be challenging, and patients react in all sorts of ways. “Some people want to have that conversation when they realize they’re unwell. Some people will strongly say, ‘I’ve been in hospital before, there’s no way you’re sending me to [intensive care],” says Smith. Some just don’t want to know.

There’s also the danger of shooting the messenger. Katherine Sleeman, medical doctor and lecturer at King’s College London’s Cicely Saunders Institute quotes from a study that found that patients perceived doctors as better communicators when they gave a more optimistic view of palliative (non-curative) chemotherapy. “It seemed as though you can inform patients that a disease is incurable but at the expense of the relationship with them, which is fascinating,” she says.

The demands that patients and families put on doctors – to find a balance between honesty, truth, and hope, to be human, yet not too human, to know everything, even the unknowable – add to the stress.

“I think we find it difficult to admit that we don’t know,” says Dr. Stephen Barclay, Senior Lecturer in General Practice and Palliative Care at the University of Cambridge, “because patients come to us, and we look to ourselves to be people who investigate, make decisions, make a diagnosis and have an action plan.”

He thinks doctors find it quite emotionally difficult to acknowledge uncertainty – something born of not the doctor’s incompetence but more the unpredictability and uncertainty of medicine, particularly the later stages of many diseases. “It is frightening. No one ever enjoys having these sort of conversations,” he says.

It takes a lot of conversations to find someone who can tell me – in full-color, human, non-clinical terms – what it’s actually like to have to, on a daily basis, tell people that they’re seriously ill. Finding doctors to talk to isn’t the problem. Our conversations start promisingly enough. But somewhere along the line, everything comes through a professional filter. 

They become less clear, less direct, obscured in medical language, cloaked in the self-preserving bubble of the passive voice, or generalized to just any doctor’s experience. “You can become upset by it, but….”

Read more: Withholding the truth from dying patients

For my sister, a doctor for eight years, it isn’t so much telling the bad news that stays with her but the small, seemingly insignificant things that went along with it: noticing a pristine newspaper, unread, on the bedside locker of a patient who had just died, despite attempts at cardiopulmonary resuscitation. Finding a cheque waiting to be paid in the wallet of a man who’d had a fatal accident.

These things – the physical manifestations of the nearly-done, not-done, never-to-be-done – seem to resonate. Unopened birthday presents, canceled holidays, unworn clothes: all symbols of a life ending prematurely, of potential diminishing, of a future fading. They’re what remains after the practicalities of dealing with a patient, and their relatives are long forgotten.

You do what you can to process it, one doctor says. Review what has happened from a medical management point of view: analyze, rationalize, and conclude. Did we do everything we could? Would we do anything differently next time? Have a cup of tea, splash your face with cold water, have a cigarette, and get on with the next patient.

Then your shift ends.

Self-medication takes a variety of forms. For some doctors, it’s straight to the pub. One (teetotal) doctor’s prescription: “Go home, order a pizza, eat lots of ice cream, sit in front of the TV and watch trash.” Another has a friend who’s also a doctor, and they’ve agreed that either of them can call the other at any time and vent.

Annabel Price, Consultant in Liaison Psychiatry at Addenbrooke’s Hospital, Cambridge, says that some doctors are more vulnerable to being affected emotionally by breaking bad news than others. This may be because they are struggling with bereavement or mental health issues in their own lives. Or it may happen if there’s a patient or medical case they relate to closely.

Doctors need to be resilient, but so do the organizations that they work in. “You would hope that if a situation is very difficult for an individual… the team then would help them to manage it, either by allowing them to step back or by providing them with extra support to be able to do that,” says Price.

“I would be very optimistic if I said that that works 100 percent well every time and that all teams function in that way, but that’s the idea that I think we should be working to recognizing that doctors are people, just as much people as our patients are, fallible humans who have our strengths and weaknesses, and weak points and struggles, just like anybody else.”

Dedicated services do exist for doctors to get help with their mental health, for example, but are those in need willing or able to find and use them?

One paper advising doctors on how to look after their mental health identifies three challenges for doctors seeking medical help. The first is stigma, including the fear that seeking help will lead to their fitness to practice medicine being challenged. The second is the idea that doctors can feel that they are somehow “letting the side down” if they need time off. 

The last are barriers to care. “Doctors do not make good patients,” the authors write. “Typically, they do not follow their health care advice, self-diagnose and self-medicate, and present late after corridor consultations. Once unwell, doctors are often reluctant to consult a general practitioner (even if registered) or to take time off work.”

Moreover, healthcare systems are far from perfect. Resources – including the time, space, and supportive colleagues conducive to productive, healthy working lives – are often limited.

I asked Twitter if doctors need, and can get, support when breaking bad news. An Australian doctor said, “Can’t even get time at work to go to the toilet while working in [the emergency department], let alone support when breaking bad news.”

When you are giving bad news about a life-limiting illness, many patients and their families want to know all they can about the condition: treatments, cures, and how long patients survive. But the data aren’t always available. Studies are often small or dated or in populations of patients so different that findings are hard to extrapolate to the patient sitting in front of you. Everybody and every body are different – so certainty can be in short supply at the very time when people want it most.

“How long have I got?” is an especially dreaded question. No clinician I speak to says that they give patients actual numbers, often preferring to talk about whether it is a matter of years, months, weeks, or days. This is because prognostication – judging the course of a disease or condition – is notoriously hard. While there are ways to assess imminent death in, for example, a patient with terminal cancer, it can be extremely difficult to estimate how long a patient with frailty, dementia, or a chronic lung condition will live.

Research shows that doctors tend to overestimate the survival of terminally ill patients, predicting them to survive more than five times as long as they end up doing.

Stephen Barclay never gives numbers to patients asking about survival. He reinforces this practice to his medical students: “I teach the students very firmly not to give numbers because often the information isn’t there,” he says. “And if the information is there, it’s average survival, and by definition of an average, 50 percent of patients live longer, and 50 percent live shorter.”

Giving people a ‘sell-by’ date is not only virtually impossible but can be harmful. “I’ve certainly had a number of patients who have been told, ‘you’ve got six months, and when six months comes, they assume they’re going to die today, and that can be really difficult if they’re clearly living longer,” Barclay says. Conversely, if things are clearly moving on a lot more quickly, then having a particular date in mind can encourage people to have unrealistic hope.

Laura-Jane Smith agrees that you have to choose your words carefully. “I have realized from having spoken to patients that they never forget that conversation, and quite often they don’t forget the specific words that you use,” she says. 

Read more: “How long do I have left to live after treatment?”

“You fall into the trap of wanting to dress it up, and wanting not to use words like ‘cancer’ and ‘incurable’ and ‘life-threatening,’” she says. “The more that I do it, the more I try and find ways to say things as clearly as possible without being blunt, and I think that’s much more effective.”

Individual words matter. Professor Elena Semino and colleagues at Lancaster University have been conducting a study of how certain kinds of language are used in communication about the end of life. They’ve created a set of over 1.5 million words, collected from interviews and online forums, where patients, carers, or healthcare professionals meet to talk with their peers.

Violence or war metaphors (“battling my disease,” “keep up the fight!”) can be disempowering or disheartening for people with cancer, potentially demanding constant effort or implying that a turn for the worse is a personal failure. But in other contexts, they can empower people, helping someone express determination or solidarity or bringing a sense of meaning, pride, and identity.

“You don’t need to be a linguist to realize what metaphors a patient’s using,” says Semino. Doctors should ask: are those metaphors working for the patient at that point? Are they helpful, giving them a sense of meaning, identity, and purpose? Or are they increasing anxiety?

While patients and their relatives may want to delay or avoid conversations that discuss death directly, this isn’t a benign act of self-preservation. A study of over 1,200 patients with incurable cancer had shown that those who had early conversations about the end of life (in this case, defined as before the last 30 days of life) were less likely to receive “aggressive care” in their last days and weeks. This included things like chemotherapy in their last two weeks and acute care in a hospital or intensive care unit in their last month.

Are doctors obligated to someone that the are dying? Are doctors obligated to give this kind of information to patients?  The UK General Medical Council guidance "is that you should tell the patient all he or she wants to know; you should be honest; you should disclose as much as you can about what’s going on,” says Deborah Bowman, Professor of Bioethics, Clinical Ethics and Medical Law at St George’s, University of London.

“The way contemporary ethics is taught, learned, understood, it’s more about different types of knowledge and different types of expert,” Bowman says. “You may well be an expert on radiotherapy, but the patient is the expert on his or her own life, preferences, values, etc.”

While patients have the right to know, they also have the right not to know. Stephen Barclay and his research group looked at the timing of conversations about the end of life with patients who had conditions such as heart failure, dementia, and the lung disorder chronic obstructive pulmonary disease. “There are a significant proportion of patients who appear not to want to have early open conversations, and some never wish to have conversations at all,” he says.

Barclay cautions that the urge for so-called “professional tidiness” – getting these end-of-life conversations out in the open – can lead doctors to place themselves rather than their patients at the center of the care. There’s no one-size-fits-all approach, he says. It’s about the offer of a conversation rather than the conversation necessarily happening.

For Katherine Sleeman, what – and how you tell someone that they are dying is incredibly important for patients with life-limiting illnesses. She believes that having open, honest conversations allows doctors and nurses to establish the patient’s preferences: everything from where they’d like to die to the level of medical intervention they’d want if they stopped breathing or suffered a cardiac arrest.

But it’s not just about medical care. The last thing somebody does may be as small as writing a letter or transferring money from one bank account to another to cover their funeral costs. But if they’re not told they’re dying, they’re being denied the last opportunity to exert control over their own life.

Good communication can even increase hope. Sleeman cites a small study of patients with end-stage renal disease. The researchers found that being given more information earlier in an illness could increase a patient’s hope rather than extinguish it. 

“With the provision of prognostic information, new threats will be perceived, but rather than annihilate hope, it provides an opportunity to reshape hopes, making them more consistent with the future,” they wrote.

“The great fear is that we’ll destroy hope by having these conversations,” says Barclay. “There is a really quite good evidence base that actually sensitive and appropriate patient-led conversations might destroy unrealistic hope, but they do generate realistic hope.”

He says that unrealistic hope is ultimately unhelpful because it’s never fulfilled. He remembers a patient with advanced cancer who said his family was planning to take him on a luxury beach holiday in six months. Barclay’s response: could they go a bit sooner, have the holiday in the UK? “They clocked what I was saying, had a holiday in this country the next month – had a lovely time.” Two months later, the patient died.

Though my Dad’s terminal cancer diagnosis came nearly 21 years ago, my mum remembers his reaction clearly. “Dad turned to me and said, ‘Well, I’ve had a good life, a very happy life.” He was reluctant to have the operation that might prolong – but not save – his life.

I remember him coming home from the hospital soon after to rest before the operation. We’d bought him an apple turnover, his favorite cake, but it remained uneaten. Dad was propped up in bed, writing a list of things to do. He phoned a neighbor and invited him to take the tools he wanted from our garage. He read an old friend's letter posted through the door, triggering the only tears my mum remembers.

A couple of days later, earlier than planned, he was taken back to the hospital. He didn’t come home again. But the short time that he lived knowing he was terminally ill gave him the chance to say goodbye. And us the same.