Introduction: dying in intensive care

Since the discoveries of sulfonamides and penicillin, medicine has had remarkable success in producing cures for diseases which were previously fatal. Technological advancement has led to the development of complex healthcare systems which have at their foundations a curative focus.1 This is typified by hospitals and intensive care units (ICUs) and can lead to an unquestioned assumption that the goal of modern medicine is to keep patients ‘alive at all costs’.2 However, these advances in medical technology have also resulted in prolongation of the dying phase for many patients and so made death more uncertain.3 Dying more slowly has obscured the distinction between being sick and dying,4 which can make it difficult for patients to know when their death is imminent.5

The curative focus of modern healthcare context tends to encourage a fight against death, rather than an informed preparation for death. This creates a culture where the main focus of medicine is to extend life, which can result in prolonged and painful deaths for patients undergoing burdensome treatments with the goal of prolonging life.6 For example, one in five deaths in the USA occurs in critical care7 and 56% of patients had life-sustaining treatments in the last 3 days of life.4 Half of the patients who die in hospitals or in ICUs have significant pain.8 Many patients die prolonged and painful deaths, receiving invasive and unwanted care.9 Overall, 3%–17% of patients receive life-sustaining treatments despite indicating that they did not want such treatment.4 There is significant variability in end-of-life practices in ICU, both within the USA,10 and across Europe.11

Many treatments at the end-of-life are invasive, costly, unnecessary and unwanted.12 This is in contrast to how people actually want to die. Overall, 31%–87% of patients and 49%–70% of the general public want to die at home.13 Despite this, many people do not die where they want to die,14 with 58% of people dying in hospital, whereas only 18% of people die at home and 4% in hospices (with 17% of people dying in care homes and 3% elsewhere).15

When survival is the primary goal of medicine, considerations of dying well can be relegated in clinical decision-making, especially in the context of intensive care medicine. Traditionally, the goal of ICUs has not been to provide patients with a good death16 and intensive care medicine’s focus on cure can obscure the need for palliation.17 Resuscitative interventions can end up being the default approach in life-threatening illness on the ICU, regardless of patients’ prehospital status.18 Although there is diversity in what people consider to be a good death,9 ICUs are not generally considered to provide a suitable environment for dying well, because they lack privacy, are often noisy and can be brightly lit.16 17 Dying in ICU (and hospital) is associated with a worse quality of life at the end of life and with increased physical and emotional distress, compared with dying in a hospice or at home.2 Therefore, dying in ICU is likely to be a significantly negative experience.19

A further concern about the experience of dying while receiving intensive care treatments arises because patients on ICUs may not be considered to be dying until it is decided that ICU treatment is failing and should be withdrawn, rather than at the start of the terminal disease process. If dying is defined by the withdrawal of ICU treatments, then it is not possible for ICU treatments to cause a negative dying experience, as their use defines a patient who is still being actively treated and so is not considered to be dying. However, this argument rests entirely on an arbitrary distinction regarding the clinical goals of treatment,20 and takes no account of the perspective of the patient. Simply defining dying as the withdrawal of ICU treatment does not, however, change the fact that, when a patient dies on ICU, the last part of their life will have been spent experiencing some of the most burdensome treatments in modern medicine. Therefore, for a patient who dies on an ICU, the whole of the period of intensive care treatment will be part of the dying process they experience.20 Furthermore, precluding palliative care provision until after active treatment in the ICU is withdrawn typically limits such provision to a very short time frame: between 71% and 83% of adult patients die within 2 hours of withdrawal of intensive care treatment.21 While ICU patients are likely to receive sedative and analgesic medication during their treatment, these do not prevent the negative experiences reported by ICU survivors and so cannot be considered to be an adequate symptom-control substitute for palliative care.

This paper will argue that decision-making regarding the provision of burdensome ICU treatments should be guided by Aristotle’s concept of human flourishing, which involves both living a long life and dying well. Making the promotion of human flourishing in the context of health the goal of medicine creates an argument for the integration of palliative care considerations into intensive care decision-making and for advanced care planning with healthy patients.

The goals of medicine

Consideration of the goals of medicine is important as they provide vital guiding concepts and ethical restraint regarding how medicine is used.22 However, the goals of medicine are in constant negotiation within the social context in which medicine is practised.23 Analysis of how the goals of medicine are changing through the processes of social construction and negotiation should be a vital concern for medical ethics, because as the goals of medicine change, the nature of the activity that medicine is also changes as a result.22 Therefore, it is vital that any gradual changes to the goals of medicine are analysed ethically, because without clear goals medicine risks becoming the ‘hireling’ of society, becoming put to whatever purpose that society decrees.24

The technological success of modern medicine in treating diseases has given rise to a ‘diagnose and treat’ model of medicine, where the goal of medicine is simply understood as the treatment, and ideally cure, of disease, which has the potential to result in the reduction of medicine to a technical response to disease processes.24 For example, Kass, in his monistic account of the goal of medicine, emphasised healing, the production of healthy humans, as the sole goal of medicine.25

Pluralistic accounts, such as four goals produced by the Hastings Centre of “the prevention of disease and the promotion of health, the relief of pain and suffering, the treatment of disease and the care of those who cannot be cured, and avoidance of a premature death and the promotion of a peaceful death”24 capture the tension that can exist between pursuing survival and providing a peaceful death in medicine. However, in this framework, these goals risk being understood as competing, rather than as co-operating components of the larger holistic goal of human flourishing. This competition is especially clear in the language of the final goal of Brody and Miller’s goals of their internal morality of medicine: ‘When all else fails, helping the patient to die with dignity and peace’.26 (my italics) This understanding of the goals of medicine relegates consideration of a peaceful death to an afterthought, only considered once any hope of achieving curative outcomes has been exhausted.

The virtue ethics concept of human flourishing

Aristotle’s virtue ethics relied on a guiding concept, eudaimonia, which can be translated as happiness or human flourishing. For Aristotle, the concept of eudaimonia was the goal towards which all human action aimed and which defined the good which informed all of his discussion of morality.27 The concept of human flourishing has been argued to be foundational for the development of the ideas of healing and health which act as traditional goals for medicine.28 Therefore, the concept of human flourishing is vital for ethical considerations of life as a mortal being, especially when facing the inevitability of death.29

Aristotle argued that the goal of human flourishing was ‘activity of the soul exhibiting virtue… in a complete life’.27 (my italics) This theme has been picked up by some other virtue ethicists, such as MacIntyre, who argued that happiness is the achievement of a completed or fulfilled life.30 However, translating eudaimonia as happiness, can suggest that eudaimonia is a human experiential state, rather than a fulfilled and completed human life. Aristotle’s conception of eudaimonia is not limited to an emotional experience, therefore, for the rest of this essay eudaimonia will be referred to instead as human flourishing, to capture the sense of a complete, fulfilled life inherent in this concept.

Aristotle argued that the person who ‘has ended wretchedly no one calls happy (eudaimonia)’.31 Therefore, his conception of human flourishing unites the goals of avoidance of a premature death with the promotion of a good death, as both are required to achieve human flourishing when an individual’s life is considered as a whole. Aristotle still considered non-peaceful deaths in battle to be good deaths if they displayed courage, facing death, and suffering for a good cause. However, in the context of modern medicine, courage to accept the reality of death can allow patients to accept palliation and symptom control measures to reduce their suffering. Alternatively, some patients may prefer to endure a degree of suffering in preference to the dulling effects of opiates—that too may manifest courage at the unpeaceful end of a patient’s life and yet contribute to that patient’s eudaimonia.1 These approaches stop death representing a failure of medicine as the technical treatment of disease, and instead allows medicine to proactively support patients to courageously create contexts where they can experience a peaceful death. When medicine is understood as the pursuit of human flourishing, death itself is not the enemy, only death at the wrong time, death for the wrong reason or death in the wrong way.24 Death can be considered premature when it prevents a person’s capacity for flourishing, in terms of the individual’s whole life cycle.24 Therefore, not all deaths will be premature.

Understanding the goal of medicine to be the promotion of human flourishing in the context of health suggests that human life should be considered an instrumental good, in that the intrinsic good of human flourishing cannot occur without it. The goodness of prolonging life in medical practice would, therefore, be understood to derive from the human flourishing that resulted from the survival of the patient. However, human life is not by itself sufficient for human flourishing and contexts may occur when prolonging life through burdensome medical treatments, such as intensive care treatment, may actually detract from human flourishing by exposing patients to prolonged and burdensome dying experiences.19 In contexts where extending a patient’s life no longer promotes their flourishing, the link between prolonging life and human flourishing is broken, and so prolonging life ceases to derive goodness from the concept of human flourishing. Therefore, the use of life-prolonging treatments in such contexts could be considered to be outside of the goals of medicine.

Human flourishing is something that can apply to both individuals and societies, so there is a negotiation necessary between an individual’s desire to flourish in a particular way and a communal understanding of flourishing. The importance of respecting an individual’s concept of human flourishing provides a reason against coercing individuals into treatments that they do not think will help them flourish. However, respecting an individual’s concept of human flourishing also creates a reason against compelling individuals to provide treatments that they do not think will promote human flourishing.

When there is disagreement between clinicians and patients or their proxies, for example, regarding whether extending the life a patient who will not survive once ICU treatment is withdrawn adds to their flourishing, then ideally a processes of dialogue and compromise32 will lead to a treatment plan acceptable to all parties. However, just as patients should not be compelled to accept treatment, clinicians should not be compelled to provide treatment that they judge will hinder the flourishing of their patient.

Human flourishing and patient autonomy

If medicine aimed to promote human flourishing in the context of health, this could imply that promoting patient autonomy should also be considered an instrumental good, that derives its goodness from the concept of human flourishing. Autonomy has been understood to mean different things, including self-expression, taking responsibility for adherence to principles of obligation towards other agents, or a right to act on one’s own judgement about matters regarding one’s own life without interference of others.33 For this argument, autonomy is taken to mean the self-governance whereby humans make and actualise choices.34 A human being’s autonomy over their body needs to be respected and promoted if they are to flourish, and the flourishing of competent patients can be undermined by the violation of their bodily integrity. Furthermore, the harm inherent in forcing competent patients to accept treatments they do not want means that respecting refusals of treatment from competent patients is fundamental to allowing them to flourish. Therefore, only when a patient is not competent to engage in medical decision-making will clinician enforced treatment to promote a patient’s flourishing (eg, forcible psychiatric treatment) have a chance of promoting their flourishing.

What makes the promotion of a patient’s autonomy good is the flourishing of the patient that it produces, and this is what justifies the use of medical resources to promote the patient’s autonomy. Therefore, clinicians promote the autonomy of their patients by informing them of different treatment options that have the potential to benefit them and facilitating a shared decision-making process among these options. However, clinicians do not help their patients to flourish by informing them of treatments which offer no benefit and would only expose them to harm, even if this increases the range of options available to their patient. Similarly, in situations where a patient autonomously requests a treatment that would inhibit their flourishing, such as burdensome ICU treatment when there is little or no chance of survival, promoting their autonomy may become detached from promoting their flourishing.

Therefore, the fact that a patient’s request for treatment has been made autonomously does not by itself mean that the request will produce flourishing for that patient, or that medicine should facilitate that request. Autonomous requests for treatments which will detract from a patient’s flourishing sever the link to human flourishing that normally makes promoting autonomy a good that medicine should strive for. Therefore, complying with a patient’s autonomous request for a treatment that will detract from their flourishing could be considered to be outside of the goals of medicine. An example of this is a patient or their family, who out of fear or a lack of acceptance of the inevitability of death (eg, due to advanced motor neuron disease) request intensive care treatment which has no chance of providing the goal of survival which they are desperately seeking.35 The sad reality is that such treatment will only prevent the little flourishing that is still available to the patient through experiencing a peaceful death.36 In such contexts, if a process of dialogue and compromise32 fails to lead to a treatment plan acceptable to all parties, then clinicians should not be compelled to provide treatment that they judge will hinder the flourishing of their patient. If fulfilling patients’ requests becomes the goal of medicine, then medicine may simply become a blind and directionless technical service in a consumer culture, for patients to use however and for whatever goals they see fit.

A similar criticism is discussed by Mathison and Davis regarding their argument that value promotion should be a goal of medicine.37 The goal of promoting a patient’s values has the potential to be criticised for turning doctors into ‘scalpels for hire’, despite Mathison and Davis’s insistence that in their account clinicians will not be obligated to fulfil patient’s requests and should instead actually sometimes act to prevent patients from embarking on courses of action that are counter to the patients’ values.37 The goal of promoting human flourishing in the context of health has the potential to add some guidance to clinicians deciding whether to promote a patient’s values by fulfilling their requests or by protecting them from treatments that run counter to the patient’s values. When promoting human flourishing in the context of health is the goal of medicine, value promotion could also be understood as an instrumental good, deriving its goodness from the concept of human flourishing, which it will normally tend to promote. However, when the patient has values and requests which tend away from their flourishing, the link with human flourishing that makes value promotion a good for medicine to strive for is broken, and so promoting such values could be considered to fall outside of the goals of medicine.

Palliative care in intensive care decision-making

Palliative care, both in hospices and through facilitating the wishes of patients to die at home,2 involves a teleological shift in the goals of medicine from avoiding death by striving for a cure, to achieving a good death through the acceptance of the reality of death. By focusing on death as a process that is a natural part of the human experience, how to best go through that process becomes the main focus.38 This reflects the importance that quality of dying has to patients and their families.39 Palliative care’s importance as a vital public health priority in response to large scale humanitarian crises was highlighted by the COVID-19 pandemic.40 However, palliative care has often been treated as a separate system within medicine, rather than as integral to the whole philosophy of healthcare.41 This can mean that focusing on the goal of facilitating a good death is forgotten until all curative options for patients are exhausted. By then it may not be possible to ascertain or action a patient’s wishes for their end-of-life care.

Making the goal of medicine to promote human flourishing in the context of health has the potential to provide medicine with a moral resource to navigate the tension between pursuing survival and providing a good death. For Aristotle, a long life and a good death are two sides of the same coin and so the pursuit of one without consideration of the other is to risk missing a vital element of human flourishing. That is not to say that medicine will always be able to provide patients with a long life and a good death. Disease and resource limitation have potential to limit human flourishing in ways that medicine cannot prevent. However, by uniting survival and dying well in the concept of human flourishing in the context of health, as the goal of medical decision-making, the risks of causing a negative dying experience will not be overlooked in the risk–benefit analysis of burdensome medical treatments which aim to prolong life. With diminishing chances of survival from burdensome treatments there may come a point at which the risk of causing negative dying experiences outweighs the survival benefit. Research with patients who have undergone such treatments, and know what they are like, is vital to explore the chance of survival that would make them willing to undergo such treatments again. This will provide helpful insights for patients, their proxies and their doctors, although it is important to remember that individual patient’s own values may conflict with aggregate norms.

In ICUs, pursuing the goal of promoting human flourishing in the context of health could require greater involvement of palliative care specialists in ICUs. Palliative care is possible on ICUs7 42 but there has been variation regarding changes to end-of-life practices in ICUs across Europe.11 Increased education about palliative care in ICU allows earlier decisions to be made about the inevitably dying, reducing delays in shifting the goal of treatment to comfort measures. These interventions decrease the average length of stay on ICU, so decreasing the non-beneficial treatment time, without increasing mortality.18 Deciding to limit life-prolonging ICU treatments before end-of-life situations occur, means that the treatment burdens faced by dying patients can be reduced.11 Treatment-limiting decisions make the death of ICU patients more acceptable than the sudden death of ICU patients, by providing some control over the timing of death and allowing death rituals to occur, such as family visiting.43 It is occasionally possible (although difficult) to discharge patients home from ICU to die.17 Therefore, it has been suggested that the aim of intensive care medicine should be to maintain a ‘meaningful tension’ between preserving life and pursuing a peaceful death, that is, both outcomes should be treated with equal seriousness rather than always presuming in favour of using technology to extend life.44

Another important practical outworking of the goal of promoting human flourishing in the context of health is early discussions about end-of-life wishes and advanced care planning. Overall, 75% of Americans45 and 95% of British people46 have not communicated their advanced care wishes, which means that they cannot be actioned by medical staff. However, appropriate advanced care planning can prevent unnecessary suffering from unwanted treatments and facilitate a good death.7 Actions to prevent negative dying experiences resulting from the inappropriate use of intensive care treatment will not be effective if such considerations only occur at the doors of the ICU.

The impersonal nature of modern healthcare makes judgements about the balance between pursuing survival and accepting death difficult, as acute healthcare providers rarely know patients for whom they have to make time-critical life-or-death decisions. A baseline assumption to prolong life (eg, when an ambulance is called) acts as a safeguard to ensure all preventable deaths are prevented. However, the cost of this is that assumptions to act to preserve life could deny many patients from having a good dying experience. Advance care planning represents an important intervention that can occur in the community with healthy patients to allow their autonomy and values regarding their end-of-life wishes to be considered in any future medical decision-making that has to balance a chance of survival with a risk of a negative dying experience.

Conclusion

The technological success of curative medicine has led to a gradual renegotiation of the goals of medicine to the technical elimination of disease and the prolongation of life. Aristotle’s concept of human flourishing, incorporating both a long life and a good death, has the potential to act as a goal for medicine that will prevent palliative care considerations being relegated to an afterthought in medical decision-making, once all curative treatments are exhausted. Making the promotion of human flourishing in the context of health the guiding goal for medicine creates an argument for the greater incorporation of palliative care considerations in intensive care decision-making, when the benefits of burdensome treatments that offer some chance of survival must be weighed against the harms of exposing patients to negative dying experiences. It also provides an argument for advanced care planning to ensure that patient’s values and autonomous end-of-life choices are known so that the importance of a good death, as part of what it means to flourish as a human, is not forgotten.

Data availability statement

Data sharing not applicable as no datasets generated and/or analysed for this study.

Ethics statementsPatient consent for publication

Not applicable.