Can versus should: Public expectation and the high-intervention death – A practical guide for clinicians

Abstract

Escalation of medical intervention near the end of life has become increasingly common, often driven by public misperceptions of medical capabilities, cultural narratives of ‘fighting’ disease, and clinicians’ discomfort with discussing limitations of more extreme treatment regimens. The result is the high-intervention death: intensive, invasive and frequently misaligned with patient values.

This article examines the dynamics underpinning these decisions from the perspectives of emergency medicine, intensive care and palliative medicine. We describe the consequences of late acceptance of dying, the absence of advance and future care planning, and societal unpreparedness for death. We propose practical strategies to support better decision-making, including early goals-of-care discussions, compassionate communication, and a cultural shift towards acknowledging death as a realistic and acceptable outcome. Restoring humane, value-aligned care which still includes a plethora of treatments, albeit less invasive ones, requires clinicians to balance medical capability with compassion, clarity and courage.

Introduction

Balancing the level of clinical intervention with the uniquely personal better interests of a patient during a serious and progressive illness is one of the most demanding aspects of medicine. It requires careful, compassionate communication to enable a patient and those close to them to understand and make decisions in the light of their deteriorating health and increasingly limited medical/surgical options for meaningful intervention. Many members of the public are, unsurprisingly, unprepared and unprimed for such conversations; some have lost confidence in the medical profession altogether.¹ This mistrust was fanned during and after the COVID-19 pandemic when social media criticism and scepticism about health services, about the safety of vaccines, and even about whether there was a pandemic at all, was commonplace.² In addition, there were, and remain, unrealistic expectations of the power of medical interventions to save lives beyond the point of no return; beliefs fuelled by stories from popular culture, for example, high success rates of soap opera cardiopulmonary resuscitation (CPR).³

Some patients and families prefer to take a fatalistic approach; others may have approaches relating to their faith. An individualised approach, together with an understanding of a person’s viewpoint, demands practitioners’ skills in listening, empathy, language, patience, advocacy and negotiation. This also takes time.

In our practices of emergency medicine, intensive care and palliative medicine, the authors commonly meet patients in a crisis or during a period of deterioration. Not uncommonly, these patients are already well known to our colleagues in other specialties. We offer here a distillation of our experiences that may help readers in deciding when to shift gear, and how quickly to do so, for instance, beginning to adjust their patients’ expectations over time as their medical conditions progress, or their vulnerability to dying increases. This may not avert crises in their clinical condition, but it might reduce the impact and surprise for a patient and family, when a conversation turns towards care at the end of life, rather than towards yet more escalating medical interventions.

Decision-making in a medical crisis

In the rush to evade death, clinicians may be faced with unrealistic requests for escalating medical interventions. These requests may come either from patients (‘I don’t care how hard it is, I want to try it’), from colleagues, or from family members or significant others who may exhort the beloved patient to ‘try just one more surgery’ or to tolerate ‘one more round of chemo’, or who demand we ‘do everything!’ as their relative’s life is approaching its inescapable end (see BOX 1).

BOX 1. Kate’s story.

Kate had arrived by emergency ambulance and been brought straight through to the resuscitation room during another busy afternoon in the emergency department (ED). She was in her late 60s and had collapsed at home earlier today. The paramedic who brought her filled the ED doctor in on the background: Kate had advanced uterine cancer with multiple metastases throughout her abdomen, pelvis and spine. She had recently started a third course of chemotherapy, from which she was experiencing significant side effects. She had developed acute severe abdominal pain that afternoon and then fainted at home. Her family had called 999.

Kate was conscious if a little drowsy, but clearly in pain. She was pale, had a low blood pressure and her abdomen was distended and tender. After pain relief was administered, a CT scan report from the previous week was reviewed; it confirmed progressive widespread metastatic cancer, including a tumour that was spreading in the vicinity of the inferior vena cava and another that was breaching part of the small bowel. It was apparent that one of these structures had ruptured.

Kate’s three daughters were waiting anxiously in the relatives’ room for an update on their mum. The ED doctor was aware of the heavy atmosphere in that room when she entered: an air of fear and uncertainty. She shared with them what had been established so far and gently explained the severity of the situation.

‘You must do everything!’ one daughter implored.

‘Mum is a fighter!’ said another.

‘Will she be going straight to surgery?’ asked the third daughter.

Alt-text: Unlabelled box dummy alt text

Each crisis represents a crossroads. Some of the paths forward will increase the possibility of a high-intervention death, characterised by maintaining unrealistic hope, ever-escalating treatments and machines taking up so much space that there is little room for human touch and proximity. Some will think that this is the way they would wish to die, but for others, the risk of an outcome that involves an undignified, unpeaceful death weighs too heavily. The consequence of public misunderstanding of dying is this: a belief that all patients coming to hospital can be kept alive, all of the time, if the clinicians just try hard enough; and that ‘being alive’ in any form and at any cost is always better than death. Sadly, when this belief inevitably clashes with biological reality, the outcome can prolong an established dying process rather than reclaim a life, however short, that the patient would consider worthwhile.

Media reports, news bulletins and casual conversations about serious illness often adopt bellicose vocabulary, with the disease as an enemy to be fought; a patient is a warrior, remission or cure ‘a victory’, and death means ‘losing the battle’. Patient opinions about this war metaphor vary widely;^4,5 but many who accept that their death is approaching are adamant that they should not be seen as ‘losing’. Yet if families describe their loved one as a fighter, then their expectation is that clinicians must be seen to fight too – and any suggestion of a move away from intervention risks being interpreted as a medical failure of willing, rather than as a patient-centred, graceful bowing to the inevitable. This language of fighting crops up in emergency departments, intensive care units and cancer clinics, in primary care, in medical specialty outpatient conversations, in medical units and palliative care consultations. It supports some patients in their endeavour to ‘fight’ for what matters to them, even when they are no longer fighting to live longer. It is an example of the delicacy and sensitivity that we need to bring to any discussion that seeks to help patients and families understand and work in the reality of a difficult situation: it is not always ‘fighting’ that is the problem, but what people want to fight for.

Levels of intervention

Early on in hospital admissions, it may be necessary to review a patient’s likelihood of responding to different interventions on offer, and to clarify how realistic it is that they will improve with different interventions. For example, a clinical team may decide that fluid resuscitation would be of benefit to a person living with severe frailty, and also that the same person would not survive a cardiac arrest. They may then wish to clarify the patient’s views on this clinical assessment, tell them which treatments are likely to be most helpful (blood transfusions, electrolytes, intravenous antibiotics) and those that will fail or leave a person with significant organ damage (for example prolonged CPR). The patient and family may then understand that treatments are not just geared at prolonging life at all costs, but can reduce their symptoms and optimise their wellbeing. Managing nausea is just as much a treatment approach as requesting an angiogram. Understanding that treatments can be aimed at comfort rather than cure enables discussion of the benefits and costs of specific interventions. In this context, discussion of CPR becomes an explanation of its limited effectiveness and many unwanted side effects when a person is approaching their death.

Too often though, natural dying is conflated with cardiac arrest, and is managed with full-blown interventions, unless there is a written exemption: a Do Not Attempt CPR (DNACPR) form. Coronial cases, and reviews that criticise those who do not initiate CPR immediately, can serve to make clinicians more wary of accepting the limits of medical intervention.⁶

Failure to recognise anticipated dying, and responding with CPR rather than with supportive care and compassionate explanation, is one of the more common examples of doing what we can: an intervention, rather than doing what we should. Ideally, we should be offering (earlier on, and preferably in collaboration with primary care colleagues) a discussion about dying, a re-setting of expectations, an explanation of the progression of dying which will not be reversed by CPR, and the writing of a DNACPR form to protect the patient from unhelpful, aggressive treatment in their final moments of life: a high-intervention death.

This ‘default to intervention’ path often ends at the intensive care unit, initially at the point of referral, as it becomes increasingly difficult for colleagues in other specialties to know which patients they should refer to ICU. Once admitted to a critical care setting, the limits of what can realistically be achieved for a particular patient become impossible to ignore. Intensive care clinicians operate in a perpetual dualism of detailed physiological support, mixed with holistic, ‘big picture’ thinking. By far the easiest route would be to maximally support every patient in need indiscriminately, but it is accepted by most intensivists that this would not be appropriate. The myriad high-tech possibilities in an ICU can create the ultimate ‘can versus should’ dilemma, raising family expectations. This is an issue on which clinicians spend a significant proportion of ICU family communication time, helping families to consider how best to proceed when multi-organ support becomes unable to save the life of that patient to a standard that would be acceptable to them. By this late point, however, patients and families are often entrenched in their ‘fight’ and have become accustomed to there always being another treatment to stave off the unthinkable. Helping them to U-turn from this coping strategy is sometimes very challenging for us as medical professionals and often very painful for them (see BOX 2).

BOX 2. Kamal’s story.

Kamal was a devoted husband of 40 plus years to his wife Jenni, and he sat alongside her in the ICU all day, every day for 4 weeks following her admission with advanced liver disease. Initially, it was hoped that a period of ICU support would allow Jenni to recover, but after some initial signs of stabilisation in the first 48 h, her deterioration continued. The ICU team repeatedly explained to Kamal that his wife was becoming more unwell, and that she had become sick enough to die. He resisted telling their young adult children, who lived far from the hospital, because he believed that Jenni would recover.

The ICU and palliative care teams worked to communicate with Kamal about Jenni’s short life expectancy, and they explained that in the event of a further cardiac arrest there would be no benefit to Jenni from attempts to restart her heart. They advised him to send for the family, who arrived the next day and were shocked to discover how close to death Jenni was.

Alt-text: Unlabelled box dummy alt text

Even in less high-tech clinical areas, the change of direction from, for example, cure-seeking treatment to ‘stability-for-a-while’-seeking treatment is a common scenario. Palliative physicians frequently support the care of people with cancers or other life-limiting conditions who may have tolerated highly toxic treatments in the early course of their illness in the hope of a cure or significant remission. As their condition progresses and their physiological reserves are diminished, both the possible gains from high-intervention treatments and the patient’s ability to withstand the side-effects of treatment shrink, too. Palliative care input may reduce the symptoms caused by the illness itself, or the toxicity burden of the treatment, or both: in fact, some studies involving early introduction of palliative care show not just improved quality of life, but also a survival advantage for the group randomised to receive palliative care alongside their disease-modifying treatment.^7,8

Introduction of palliative care is too often seen as a proxy decision for ‘no active treatment’, while ‘treatment’ is often misinterpreted as only that which is curative, rather than the myriad interventions which prioritise wellbeing and quality of life. Some colleagues conflate the ‘treat more conservatively’ narrative with not checking even the most basic parameters. For example, a nauseated cancer patient who is no longer deemed fit for chemotherapy may find that their serum calcium is not checked for fear of burdening them with blood tests. But hypercalcaemia of malignancy is usually easy to reverse, allowing this person to live weeks or months without the burdens of nausea, delirium, constipation and accelerated muscular weakness. What is an unnecessary burden to an individual patient depends on understanding their perspective and what that patient considers ‘high intervention’.

Contrary to the still-prevalent ‘do nothing’ misunderstanding of palliative care, hospices and palliative care services offer many treatments. They deploy intravenous antibiotics, transfusion of blood products, interventional pain-management strategies that include nerve blocks and spinal analgesia: the equation is not binary ‘everything or nothing’, or ‘treatment versus no treatment’. In all areas of medicine we can, and sometimes should, offer interventions that may be intrusive to buy better quality of life, even when they will not prolong life. A pneumonia can be incredibly unpleasant, from a symptom-control point of view, as can an abscess, even in the last days or weeks of life. Antibiotics, surgical drainage? A grey area for exploration between the binary extremes.

Exploratory conversations

The better we explore our patients’ values, the more we understand what matters most to them (see RCGP and partners’ website https://www.whatmattersconversations.org/), the better we can offer treatment options not based on everything that medicine could do, but on those options that are a ‘best fit’ for each individual’s priorities.

This applies as much to withdrawal of life support begun in hope but no longer supporting any chance of recovery, as to re-evaluating the benefits of haemodialysis that was life-enhancing in this patient’s thirties, yet is now a constant burden decades later. Conversations that acknowledge the limits of treatment, explore this patient’s own views of what levels of intervention would be deemed unacceptable, and begin the process of advance and/or future care planning can help each patient and their loved ones be better prepared and less shocked when their condition deteriorates in the future.⁹ Future care planning also helps the colleagues who meet the patient later to understand both their condition and the patient’s understanding of it, providing a bridge towards compassionate conversations about what medical interventions may still be appropriate, adjusting expectations and reducing the pull towards high-intensity, low-value treatments (see BOX 3). Sadly, these conversations are not taking place: the UK’s 2024 national audit of care at the end-of-life (NACEL) showed that fewer than one in five people who died in hospital had any record of previous anticipatory care planning on admission, even though the majority died of progressive, life-limiting conditions rather than during an unanticipatable crisis.¹⁰ No wonder patients expect ‘more treatment’ and families demand miracles. We are not, as a profession, stepping up.

BOX 3. Linda’s story.

Linda’s wife Wendy was admitted to neuro ICU following a planned neurosurgical procedure to biopsy a mass in her left frontal lobe. Wendy had been experiencing headache, right-sided weakness and some word-finding difficulties before her biopsy, but postoperatively she was confused, apathetic and suffered multiple seizures until she was stabilised on anticonvulsant medication. The biopsy confirmed the MRI scan suggestion of high-grade glioblastoma multiforme. After a further sudden deterioration in Wendy’s condition, the neurosurgeons offered a debulking procedure to reduce intracranial pressure. Wendy was not sufficiently conscious to consent for surgery, and Linda explained to the clinicians that she and Wendy had already talked about how far Wendy would be prepared to tolerate incapacity caused by her glioma or its treatment. ‘She would rather die than lose her personality or her speech’, Wendy reported. ‘But she wants to be an organ donor if any parts of her can be used to make somebody else well’.

Those prior discussions between the spouses ensured that the ICU team could respect Wendy’s wishes, care for her as she deteriorated rapidly over the next few days and honour her wishes to be a tissue donor by donating her corneas after she died. This last act of generosity was a great consolation to her widow, Linda.

Alt-text: Unlabelled box dummy alt text

From the authors’ shared experience of counselling bewildered families through medical decisions towards the end of life, we have some pointers for colleagues across the spectrum of medical practice: BOX 4. Likewise, BOX 5 offers some considerations about the way we reach and communicate treatment-limiting decisions.

BOX 4. Communicating uncertainty, limited possibility for treatment, and likelihood of dying.

Begin by listening to understand what patient and family know, assume, fear and expect. Interrupt as little as possible. Notice but don’t yet correct any inaccuracies in their account. Ask clarifying questions. Let them tell their story.

Pick up their story and narrate it back to them, clarifying and reorienting any points they had missed or misunderstood. Summarise and check your understanding of what you have understood of their ideas, concerns and expectations. Show them that you have heard their story and understood their concerns.

Centre on compassion: the bad news and limited possibilities for intervention are not your fault. You are telling an unwelcome truth that should be explored gently and clearly. Use simple language. Avoid euphemisms.

The three Ws are helpful when patient/family expectations are very different from reality: ‘I wish…, I worry…, I wonder…’

‘I wish we had another treatment option for your cancer as much as you do. I worry that the options now are not a good match for your type of cancer, and that they will give you lots of side effects without reducing the cancer. I wonder what you would choose to do with the next part of your life if you stopped treatment to focus on the things that matter most to you’.

‘I wish what you hope for could happen, but I worry that your brother is very much sicker than last time he was in hospital. I wonder what we can do that makes sure he is comfortable, even if we can’t save his life’.

‘Your mother is so sick we believe she is dying’. ‘Your body has become too weak to survive’. ‘We think it’s possible that X will die in the next few days’.

Allow silence while they think. Resist the impulse to ‘fix’ or to rush towards decisions. Time invested now will help patient and family to anticipate the next part of their story.

Respect a patient’s right not to know: ask them who to talk to instead.

Name any disagreement without judgement. Acknowledge uncertainty. Model being clear, kind and human.

‘We don’t all agree about what is the best thing to do for Y, but we all agree that we want to do the best we can’.

Alt-text: Unlabelled box dummy alt text

BOX 5. Making decisions to limit unhelpful medical interventions.

Walk two roads¹¹ for patients who present sick enough to die, by putting plans in place for comfort care/palliation alongside exploring any possibilities for helpful interventions or pursuing early resuscitative measures. Parallel planning like this is common in ED.

‘In case Z can’t survive this, what should we do today that we might wish we had done tomorrow?’ This may include cultural or religious observations, sending for family, opportunities for important family conversations, etc.

We may put limits of interventions, but we never put limits on care. Avoid expressions like ‘ceilings of care’ or ‘withdrawal of care’. Explain ‘ceilings of intervention’ or ‘withdrawal of unhelpful treatments’. Every setting can offer intensive caring, even when we are no longer pursuing intensive treatments.

Best interests decision-making (making decisions for a patient’s benefit in Scotland) is a consultative process, not a medical practitioner’s decision without consultation except in an emergency, to save a life. This is both the law (Mental Capacity Act 2005¹²; Adults with Incapacity (Scotland) Act 2000¹³) and GMC regulation.¹⁴ Respect the process and record how a decision is reached, including the contributions of staff and family (and patient) consulted.

Make space for, recognise and address moral distress in staff and/or family members who disagree with the decisions and plans. Not all members of an ICU, ED, palliative care team, ward team or community team take the same view, yet we must work collegiately and care for each other.

Alt-text: Unlabelled box dummy alt text

Conclusion

In our roles in emergency medicine, intensive care and palliative medicine, we meet patients and families unprepared for the unwelcome news that we often need to break. This paper acknowledges that humans are mortal; that longitudinal teamwork between primary care and hospital specialists must lay the groundwork for patients and families to have realistic expectations about the course of a person’s long-term conditions; that compassionate communication to consider the possibility of dying is necessary when we meet patients in crisis; and this combined effort might gradually move the dial towards doing what we should do: care with compassion and expertise to support people who are so sick that their death is imminent. Whenever we meet a patient facing the end of their life and yet still reach blindly for the high-tech answers medicine can offer, we tip the sickest people, at their most vulnerable time, towards a high-intensity, low humanity death. We can’t always save everyone’s life. Being able to save their death can be an important and compassionate alternative.

Further reading

Royal College of General Practitioners and partners’ website https://www.whatmattersconversations.org/

MCA (2005) Code of Practice: Department for Constitutional Affairs (2007) Mental Capacity Act 2005: Code of Practice. London: TSO

https://www.gov.uk/government/publications/mental-capacity-act-code-of-practice

Faculty of Intensive Care Medicine

https://www.ficm.ac.uk/standardssafetyguidelinescriticalfutures/care-at-the-end-of-life

Royal College of Emergency Medicine

https://rcem.ac.uk/wp-content/uploads/2025/05/RCEM-End-of-Life-Care-Toolkit-2025-Update-FINALV2.pdf

General Medical Council

https://www.gmc-uk.org/professional-standards/the-professional-standards/treatment-and-care-towards-the-end-of-life/guidance

TalkCPR

Better conversations about cardiopulmonary resuscitation: www.TalkCPR.com

NICE

https://www.nice.org.uk/guidance/ng142

CRediT authorship contribution statement

Sophia Henderson: Writing – original draft, Project administration. Calvin Lightbody: Writing – original draft. Mark Taubert: Writing – original draft.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.