Intended for healthcare professionals


Embracing inherent uncertainty in advanced illness

BMJ 2016; 354 doi: (Published 18 July 2016) Cite this as: BMJ 2016;354:i3802
  1. B Kimbell, researcher,
  2. S A Murray, professor of primary palliative care,
  3. S Macpherson, carer support group representative,
  4. K Boyd, consultant in palliative medicine
  1. 1Primary Palliative Care Research Group, Usher Institute of Population Health Sciences and Informatics, University of Edinburgh, Edinburgh EH8 9AG, UK
  1. Correspondence to: K Boyd kirsty.boyd{at}

B Kimbell and colleagues argue that we should be helping people plan for possible future needs rather than trying to provide certainty

Everyone dies, but uncertainty about how and when that will happen is inevitable.1 Uncertainty can be seen as an “enemy” to be constrained by medical science. Work continues on tools that try to provide a reliable prognosis for people with advanced health conditions as far as six or 12 months ahead on the grounds that meaningful choices require robust evidence of risk of death. However, such tools are not reliable when applied to individuals.2 Uncertainty also drives fears about future debility, loss of control, and either inadequate or excessive healthcare.

We question the perception that uncertainty is a barrier to appropriate person centred care towards the end of life. Uncertainty may in fact be more friend than foe. An unpredictable but evident risk of deteriorating and dying should be a trigger for planning care with all people who have an advanced illness and in all care settings. It offers a prime opportunity to accept and manage the inherent uncertainties of living and dying well with progressive, advanced conditions.

Challenge for people with advanced conditions

Most people have a poor understanding of the underlying causes, severity, and future course of many long term conditions. Many struggle to make sense of events and experiences or to predict likely outcomes. They have difficulty finding the resources they need to cope with them. This in turn reduces their capacity for self management and puts patients and carers at risk of anxiety, depression, and poor outcomes. Professionals need to manage uncertainty, coordinate care, and help people to sustain their identity.3

When we consider the experiences of people with different advanced conditions, contrasting illness narratives and perceptions of uncertainty emerge among patients, carers, and professionals (table 1).4 5 6 7 8 9 10 11 The diversity and variability of people’s experiences of deteriorating health and dying belie attempts to specify what constitutes a “good” death. We can clarify some core principles but need to be ready to adapt our treatment and care plans in line with people’s changing circumstances and evolving priorities.

Table 1

Experiences and perceptions of uncertainty in advanced conditions

View this table:

A degree of uncertainty can have benefits too and may provide a way for people to live well with advanced illnesses. Disavowal (choosing to live in the present) provides a coping strategy—people understand the threat to their life situation but seek hope through positively reconstructing this threat to avoid keeping it at the forefront of their minds.12

Challenge for professionals and policy makers

For professionals, uncertainty has long been cited as a major barrier to identifying when treatment goals should be reviewed and the focus of care shifted towards prioritising quality of life and a comfortable death.13 14 Treatment advances, public expectations, fear of litigation, and a plethora of evidence based guidelines combine with established beliefs about professional roles, responsibilities, and values to make “changing gear” and accepting the reality of death and dying challenging for professionals.15 Primary care teams also face uncertainty because of perceived limitations in disease specific expertise, difficulties with judging prognosis if people do not have cancer, a lack of clarity about role boundaries across care settings, and conflicting information about the benefits of further treatments.16 17 18

Prognostic paralysis

A widely adopted strategy for addressing prognostic uncertainty in advancing illness involves making predictions about when a person is likely to die so that end-of-life care can be started. However, clinicians’ estimates of prognosis are notoriously unreliable.19 Despite this, more recent iterations of the well known “surprise question” (Would you be surprised if this patient died within the next year?) instruct clinicians to identify people who might die instead of using it, as originally intended, to support broader judgments about a risk of deteriorating and dying.20 Strenuous efforts to develop more robust mortality risk scores and tools have not helped very much either.1 Predicting how long an individual will live remains an inexact science in all diseases and is nigh on impossible when people have multiple advanced conditions.2 21

Professionals and policy makers continue to differ in their interpretation and definition of the end of life. Most have adopted the last year of life as a working definition, although the recent NHS England action plan acknowledged that for some people end-of-life care refers to the last few years of life, whereas for others this could be a matter of months, weeks, days, or hours.22 This is consistent with the World Health Organization definition of palliative care, which focuses on improving the quality of life of people living with life threatening illnesses and their families at all stages of their illness journey.

Planning for an uncertain future

In the US, advance care planning had its roots in an imperative to reduce medicalisation of dying and promote individual autonomy. Advance care planning does not control inherent uncertainty and is most effective when the main focus is on preparing people for shared decision making.23 24 Respect for autonomy and preservation of personal choice when dying have become established professional and service responsibilities in many countries. However, emphasising choice above other considerations can be inequitable or cause unintended harm if we do not listen to people’s evolving ideas and personal stories and then review the available options as their health, personal circumstances, and priorities change.25 26

People welcome opportunities to discuss their future health and care with professionals but will struggle if they are expected to participate in making explicit advance care plans for situations they cannot relate to or about unknown future circumstances.27 When people are told that their health is deteriorating, they become more distressed if their fears about how to handle an uncertain future are not addressed during discussions about treatment and care.28 29 Calls for more realistic medicine aim to reduce burdensome interventions that are of little benefit and not in line with people’s preferences. This depends on professionals, people living with advanced illnesses, and the wider community developing new ways of exchanging knowledge, perspectives, and information and embracing a degree of uncertainty.30

Delivering person centred care in context of uncertainty

Uncertainty presents major challenges for people living with advanced illnesses and for the professionals caring for them. If signs of deteriorating health are identified in timely ways, uncertainty offers valuable opportunities for conversations that explore people’s concerns, wishes, and priorities for care in the short and longer term. Professionals need to adopt effective and sensitive ways of broaching unpredictable decline and a risk of dying. This should be accompanied by public campaigns that promote more open dialogue about what matters to people whose health is deteriorating.

A gradual approach to increasing prognostic awareness is of proved benefit in advanced cancer.29 Observational evidence of doctor-patient communication shows that hypothetical talk about what might happen “if the person became less well” is particularly effective in engaging people in discussions about planning ahead. Likewise, linking our hopes with our concerns about possible deterioration helps to open a supportive dialogue about the future.29 31 Promoting the idea of “hoping for the best, but planning just in case” can enable people to become active and informed decision makers in their ongoing care without causing undue distress. This includes acknowledging the reality and challenges of living with uncertainty and talking about what this person and their family would find helpful as they manage its effects on their lives.32 Finding out about what matters to each person fosters a positive approach to care and supports hope.33 This lets us integrate holistic supportive and palliative care with appropriate disease modifying treatments while also making effective and responsive plans for the future with patients and their families.

What next?

Uncertainty offers a trigger for starting conversations with people that explore their concerns, values, and priorities and inform shared decision making about their current and future care. People with a new diagnosis of any life limiting condition or who have deteriorating health due to multiple illnesses or indicators such as an unplanned hospital admission or increasing care needs can all benefit. These conversations should not focus exclusively on breaking bad news about approaching the end of life or only be about clinically driven priorities such as cardiopulmonary resuscitation or escalating treatment. They should encompass and respond to diverse individual worries about admission to hospital or a care home, what to do if a carer becomes ill, support for family members, important future events, or a potential loss of capacity (box). This will enable people to have informed roles in determining the quantity and quality of their remaining time.

Ways of talking about deteriorating health and uncertainty

  • Find out about what the person knows and expects:

    • When people have a change in their health, it helps to talk about what is happening and what might happen next…

    • Can you tell me what you know about your health problems….?

  • Share information about what might happen linked to their understanding:

    • If you did get more unwell again, what do you think about going to hospital?

    • I hope you will get better with this treatment, but I am worried that at some point… It might be a good idea to talk about that now…

  • Ask about the things that matter to this person now and in the future:

    • What things are important to you that we should know about?

    • Do you have a family member or a close friend we should speak to as well?

  • Talk about what they think would help; then possible options and plans for the future:

    • What would be the best way for us to help you stay at home?

    • Can we talk about how we might cope with not knowing quite what will happen or when?

    • It is a good idea to talk about what we can do and how to get help if you get more unwell …

Key messages

  • Uncertainty characterises most advanced illnesses and is a challenge for patients, families, professionals, and health services

  • Current focus is on determining when to trigger end-of-life care

  • Instead we should plan for the future with people at risk of deteriorating health

  • Acknowledging uncertainty is the basis of effective shared decision making about treatment and care options as a person’s health declines


  • Contributors and sources: We have drawn on a synthesis of over 1000 in-depth serial interviews with people who had diverse life limiting conditions and their carers from studies carried out in the past 15 years by our research group. We have integrated these data with a review of the wider literature concerning uncertainty and care planning in advanced illness, and recent policy documents and reports about palliative and end-of-life care. The authors comprise a carer with extensive experience of representing service users, a social scientist, a general practitioner, and a palliative medicine consultant.

  • Competing interests: We have read and understood BMJ policy on declaration of interests and have no relevant interests to declare.

  • Provenance and peer review: Not commissioned; externally peer reviewed.


View Abstract