Do patients need to know they are terminally ill? NoBMJ 2013; 346 doi: http://dx.doi.org/10.1136/bmj.f2560 (Published 24 April 2013) Cite this as: BMJ 2013;346:f2560
- Leslie J Blackhall, section head
Patients do not need to be told that they are terminally ill. However, this does not mean we should pretend we can cure them of incurable illnesses or that we should withhold prognostic information from those who want it. But insisting on prognostic disclosure to facilitate “patient choice” about end of life care is a failed model for medical decision making that creates more suffering than it relieves.
What does terminal mean?
The choice model has outlived its usefulness for two main reasons. Firstly, disclosing a “terminal” prognosis is impossible because we do not have a clear definition. Even palliative care physicians do not have a universally accepted definition for this, or for related terms such as dying or end of life.1
Does telling someone that they are terminally ill mean telling them how long they have to live (hard to know for any individual)?2 Does it mean telling them that they will eventually die (true for all of us)? Does it mean telling them there is “nothing we can do” (never true)? Are patients terminally ill when they first develop incurable cancer, even if they might live many years before dying of that illness? Are they terminally ill only when their functional status starts to decline and they develop more symptoms, or only when they are bedbound and unconscious? This lack of precision is not just a semantic problem; it indicates an underlying failure of the medical profession and wider culture at large to consider how the fact of human mortality should be accounted for in the practice of medicine.
The second problem with the case for prognostic disclosure is the claim that patients need this information to make choices about their care, especially artificial life support. The way these choices are usually framed, however, further reveals our confusion about the role of medicine in caring for patients with life limiting (meaning progressive, incurable) illnesses. The underlying assumption is that patients have a choice between therapy that will prolong life and therapy directed at comfort.
But this assumption is largely an artefact of the unusual cases (mostly of patients in a persistent vegetative state) that have captured the attention of the media, bioethicists, and the legal system in the decades since the development of cardiopulmonary resuscitation.3 4 Most adults die of chronic life limiting illnesses such as cancer, end stage heart and lung disease, and complications of dementia. In these illnesses the choice between comfort and prolongation of life is usually a false one. Patients with dementia do not die faster without feeding tubes.5 Patients with advanced cancer do not die faster if they choose hospice or palliative care, and may in fact live longer, even though they are less likely to receive artificial life support.6 7 8
Focus on best care
The real question is not whether patients should be told that they are “terminally ill” but how can we provide excellent care to patients with incurable, progressive illnesses? Patients do not need to know that they are terminally ill (whatever that means) but what we can and cannot do for them, and what the trajectory of their illness is likely to be.9 We cannot cure widely metastatic lung cancer, so no one can “choose” to die from it.
Treatments may improve the quality and quantity of life, but these will eventually stop working. Although we cannot say how long any individual patient with cancer may live, we do know that as the illness progresses these patients become weaker and have more symptoms, eventually needing help with all the activities of daily living. Pronounced functional decline is a prognostic marker and defines a portion of the illness trajectory where chemotherapy and artificial life support are increasingly unlikely to provide benefit.10
In this situation, framing end of life care around the notion of patient choice often creates a tortured situation. We insist that patients be told that they are dying and choose to forgo treatment even when their doctors think that treatments will do more harm than good. Patients think that they are being asked to accept that they are dying and to choose death instead of treatment. Not surprisingly, they often resist both the label and the choice.
Refusal to accept the label “terminally ill” does not improve prognosis, however, and the result is confusion about direction of care, moral distress in clinicians, and anxiety and trauma to patients and their families.11 12 13 Patients with advanced cancer and poor functional status do not need to know that they are terminally ill so that they can “refuse” chemotherapy or cardiopulmonary resuscitation. In most cases they should not be given these treatments exactly because they are terminally ill.
Although patients have a right to the best prognostic information available, they also have a right to ignore that information.14 To make decisions about care, patients with life limiting illnesses do not necessarily need to know how long they have to live or be informed when they pass some ill defined threshold of “terminal illness,” or choose to accept death. Instead, they need to understand the limitations of disease modifying therapy for their condition; what medical care can do for this disease; what side effects treatments might have; and what may happen as the illness progresses—or to delegate that task to a trusted surrogate.
This is not an argument for deceiving patients, or for reverting to a paternalistic mode of care. On the contrary, it is an argument for honesty about the efficacy of various types of medical care throughout the spectrum of life limiting illness.
Cite this as: BMJ 2013;346:f2560
Competing interests: I have read and understood the BMJ Group policy on declaration of interests and have no relevant interests to declare.
Read Emily Collis and Katherine Sleeman’s side of the debate at doi:10.1136/bmj.f2589
Provenance and peer review: Commissioned; not externally peer reviewed.