Article Text

Download PDFPDF
Quality of dying—how are we doing?
  1. Roderick D MacLeod
  1. Professor Roderick D MacLeod, Goodfellow Unit, Department of General Practice and Primary Health Care, School of Population Health, University of Auckland, Auckland, New Zealand; rd.macleod{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Most people, when given a choice, would choose to die at home, however, the reality is that many of us will die in hospital. How that death will be, is often a cause for debate and, for some, concern.

The Study to Understand Progress and Preferences for Outcome and Risks of Treatment (SUPPORT)1 is one of many studies that showed much was at fault with the way end-of-life care was provided in hospitals. Of the hospitalised patients in that study, 50% experienced moderate to severe pain during the last 3 days of life; fewer than half of doctors discussed cardiopulmonary resuscitation with their patients, and almost half of the do not resuscitate (DNR) orders were written within 2 days of death.

It may be that the studies referred to above were undertaken some time ago, but there is still evidence to suggest that more needs to be done. Pain and symptom management have been reported as main concerns for the dying patient and their relatives.23 Patients report psychological symptoms4 and difficulties in communication.5 Even when treatment options are discussed, often patients do not understand their options.67

It is especially heartening therefore, to see a study in this issue of Quality and Safety in Health Care(see page 244) that shows improvements in quality of care for people dying in hospital.8 This retrospective observational study in a major New Zealand hospital showed that end-of-life discussions took place in 82% of cases, 74% of those who died had a DNR order in place, and, remarkably, 96% of those in whom pain status was documented (70%) were pain free. Although there are limitations regarding the conclusions that can be drawn from this study it is undoubtedly good news. Compared with similar studies, the signs are positive. Pain status was identified in a majority of nursing notes and pain was found to be absent (sadly other symptoms were not identified in such detail). Invasive procedures still occurred during the last admission in almost half of the patients studied, but communication and planning are two aspects that seem from the data to be improving considerably on studies in the past.

So how could this be explained and what does this imply for how we move forward?

There is certainly a growing awareness of the need to seek a “good enough death” wherever people die.9 The establishment of palliative care teams in hospitals, although not new,10 is gathering pace and is undoubtedly influencing care. The development of clinical pathways may have considerably improved end-of-life care where they have been instituted,11 and when used as an aid to auditing care of the terminally ill, such pathways can provide a model for furthering improvements in care.12 Initiatives to improve care in the community have also been developed over the past decade1315 and have undoubtedly had a spin-off effect on inpatient care. But none of these innovations were in place in the hospital when Glasgow and colleagues undertook their study.

Certainly, improvements in undergraduate education in end-of-life care and communication skills may sensitise some to and inform others of the needs of people who are dying and this may be the case here. Generally, there is wide variation in undergraduate education and reviews suggest that there is a lack of consistency in what undergraduates are taught about palliative care; what teaching there is tends to be fragmented and ad hoc and lacks coordination.16 The presence of acute pain services, as Glasgow et al8 suggest, may also influence care but it may just be that people who are sick and their families are expecting more than they were a decade ago.

Jocelyn Clark, writing in the BMJ17 identified that modern dying involved a “struggle for control”. Incorporating patients’ views into the development of their care may well have influenced care provided in that hospital and therefore the results of this study. It would be cheering to think that the notion of “doctors fear[ing] failure when they cannot keep their patients alive” may be a fading memory. As long ago as the 1960s, Glaser and Strauss18 showed how the doctor’s diagnosis of dying shaped the interactions of staff, family and the patient. More effective communication, including the imparting of a terminal prognosis, as identified by some of the results reported here, has possibly enabled patients and families to take more control of their dying, a desire that has been identified in earlier work.19 The historical lack of openness in society (and in hospitals) about death and dying may be diminishing; the fact that over three-quarters of the people who died had the opportunity to discuss their dying must have had some impact. Glasgow et al8 could demonstrate that open communication can assist in improving the quality of dying. Clark’s call for more research17 was answered by Kendall et al,20 who cited a number of reasons why research may be tricky to undertake in this group of people. But they were not insurmountable reasons; nothing is impossible. By educating the profession and the public that death is a normal and inevitable consequence of living which should be seen as a normal part of life perhaps we will advance our understanding of the important issues of end-of-life care. Ethics committees and clinical staff are understandably concerned for the well-being of people who are dying, that is their function, but it is as well to remember that if they are dying then these people must be alive and often want to contribute to research to help to improve things.

Our job, whether in hospital or the community, is to help those people live right until the end of their life with as good a quality of life as they can achieve. Quality of death should become as important an indicator in healthcare as quality of life. The results from New Zealand suggest that things may well be changing, for the better. It behoves us to remember, as Sherwin Nuland21 wrote:

“The greatest dignity to be found in death is the dignity in the life that preceded it.”



  • Competing interests: None.

Linked Articles

  • Quality lines
    David P Stevens
  • Original research
    J L Glasgow S R McLennan K J High L A G Celi