Researching a good death

BMJ 2007; 334 doi: https://doi.org/10.1136/bmj.39141.417454.80 (Published 08 March 2007) Cite this as: BMJ 2007;334:485
  1. Stephen Workman, associate professor
  1. Department of Medicine, Dalhousie University, 1278 Tower Road Site, Halifax, NS, Canada B3H 3S9
  1. stephen.workman{at}gmail.com

    Raises difficult issues, but many patients are keen to participate

    In this week's BMJ, a qualitative study by Kendall and colleagues assesses the challenges in conducting research in people nearing the end of life.1 The study—conducted in researchers, people with cancer, and carers—provides a landmark in a challenging area, as well as offering encouragement for future researchers. It finds that many patients do wish to participate in research, and that researchers, while appreciating the challenges of conducting research in this area, think that it is no more demanding than in other areas. The study also offers potential solutions to the barriers to carrying out such research. These take the form of a useful checklist to be consulted before designing any study intended to research a good death.

    A central moral point of the study is that patients must not be paternalistically excluded from researching a good death, because research can enrich the lives of participants. This perspective reflects the work of Harvey Chochinov, who has developed an empirically based psychotherapeutic interview that places illness within a life context and can enhance the dignity of patients. Ninety one of 100 patients who received the interview were satisfied with the results.2 Integrating the enhancement of dignity into end of life research could improve the research experience of patients who are terminally ill. That some seriously ill patients have limited capacity to consent3 must be considered when developing recruitment methods, however.

    Kendall and colleagues acknowledge that their results relate mainly to people with cancer, who have a somewhat predictable illness trajectory. However, in Canada (www40.statcan.ca/l01/cst01/health36.htm), as in other wealthy nations, only about a quarter of all deaths are due to cancer. People with other illnesses also need innovative approaches to end of life care. For example, one third of patients with heart failure will die suddenly.4 Models of care where curative efforts are withdrawn as palliative care is escalated5 cannot easily be applied to meet the needs of these and many other patients.

    The first three barriers to end of life care listed by Kendall and colleagues are unstable definitions of end of life care and terminal care, inability to make an accurate prognosis, and variability in awareness of diagnosis and prognosis in patients and carers. These barriers hinge around uncertainty. Such uncertainty especially confounds researching a good death for patients expected to die of non-malignant disease. These patients have an uncertain prognosis and often die in hospital during a therapeutic trial.

    The extreme variation in intensity of treatments provided at the end of life for patients in hospital6 suggests that the reality of death is frequently overlooked. The tendency of doctors to overestimate survival probably contributes to a failure to focus upon providing a good death.7 Many doctors' actions suggest that death is viewed as a failure in care rather than being inevitable.8 Ways to overcome barriers to a good death in hospital require administrators and doctors to make end of life care a priority, as the necessary resources are readily available to most patients.

    Problems at the end of life are common for family members as well as patients. Over half of all complaints to the National Health Service are about care around the time of death.9 Harms to the family may be severe and long lasting. One study found that listening attentively to family members before the death of a patient in an intensive care unit reduced post-traumatic stress from 70% to 45% at three months.10 Although this reduction is dramatic, a 45% rate of post-traumatic stress is hardly ideal. Future research that incorporates the challenges to researching a good death while focusing on the needs of family members would be welcome.

    Patients will always have complex and conflicting needs for end of life care. A Canadian study of patients with an expected mortality rate of 50% at six months found that the most important aspect of care was having trust in their doctor. Not having the dying process unduly prolonged and having open and honest communication with the doctor were the second and third most important aspects of care.11

    If the disparate needs of patients are to be met, if hope and trust are to be maintained while death is honestly conceived, prognostic uncertainty must not simply be managed but should be embraced. We all live with the knowledge that death is inevitable, yet our lives can be rich and full of meaning. “Dying” patients and their families deserve the same opportunity (a recently developed frailty index12 could help identify patients at risk of dying). Ideally, communication about death and dying should enhance the dignity of patients. Research creating ways to facilitate such communication could underpin future research studies intended to help ensure a good death for all.


    • Competing interests: None declared.

    • Provenance and peer review: Commissioned; not externally peer reviewed.


    View Abstract

    Sign in

    Log in through your institution