Search tips
Search criteria 


Logo of jrsocmedLink to Publisher's site
J R Soc Med. 2002 November; 95(11): 536–538.
PMCID: PMC1279248

What happens when elderly people die?

Death, previously a taboo subject, is now a topic of ordinary conversation: not long ago, a person was shown on British television in the actual dying process. There is a widespread view that death can be managed, and that an ‘undignified’ death signifies deficient care. This is erroneous. In 1963, when Hinton1 looked at deaths in general wards, he found widespread distress. Thirty-four years later the SUPPORT study in the USA yielded very similar findings2: despite great advances in palliative care in the intervening period, many dying patients were still troubled by pain and other symptoms. Death and bereavement are linked, and a perception that care has been deficient can disturb bereavement reactions. This article questions some current assumptions, with particular reference to dying in old age.


The classic model of the dying process described by Kubler-Ross3, involving phases of denial, anger, rationalization and acceptance, is shadowed by a similar model of the bereavement process described by Parkes4. Both of these models originated in observations of aware patients with aware relatives dying of specific, often malignant, diseases. In elderly people, the processes of dying tend to be less clear-cut than in younger people with malignancy. The ‘dying trajectory’ may last longer, and awareness may be clouded by cognitive dysfunction. Exton-Smith found that 41% of the elderly patients in his study of dying people were delirious at the time of death5. Just as death may be different, so too may bereavement. The profound distress exhibited by many relatives of critically ill patients in general hospitals is often difficult to understand in the context of traditional models of bereavement. This may suggest that, when old people die, a more complex process is going on than bereavement when younger people die.


Sherwin Nuland, an American surgeon, has made the point that death in old age is often a protracted affair, rather than a clear-cut process that can allow patients and those bereaved to go through the classic stages6. He quotes an elderly patient as saying ‘Death keeps taking little bits of me’. Her physician commented: ‘She saw that with each attack of dizziness or fainting or confusion she became a little older, a little weaker and a little more tired. She knew that for ten years or more, she had been moving step by step towards the grave.’ Nuland quotes Osler as saying: ‘These people take as long to die as they did to grow up.’ Nuland also observed that fewer than one in five people can have a peaceful end, since ‘dying is a messy business’ for which relatives are unprepared. He continues: ‘Too often, patients and their families cherish expectations that cannot be met, with the result that death is made all the more difficult by frustration and disappointment with a medical community that may be able to do no better.’


Lynn7 noted that function and symptoms at the end of life generally follow one of three trajectories—(1) a short period of obvious decline at the end, typical of cancer; (2) long-term disability with periodic exacerbations and unpredictable timing of death that characterize dying with chronic organ or system failures (some cancers that respond to treatment and then relapse come into this category); (3) self-care deficits and a slowly dwindling course to death from dementia. The Kubler-Ross concept of ‘aware death’ can apply only to the first trajectory. Relatives who expect aware deaths may become angry and turn their anger onto doctors and nurses when death takes other forms. Dying people often need psychosocial support, but the potential for introducing this occurs only when the dying phase is identified8. This is not always possible in trajectories 2 and 3. Psychological stress, through lack of this support, can spread to relatives and carers.


In a study of the dying process in a hospice in the UK, Lawton, a sociologist, observed that physical degeneration—which happens commonly when elderly people die—causes an inevitable destruction of the self9. A central goal of the hospice movement is to enable patients to retain control of their lives until death, ‘to live until you die’10. Destruction of the self makes this goal very difficult to achieve in practice since the apathy and withdrawal seemed to stem from declining physical function rather than knowledge of impending death10. Many elderly people view functional decline as a loss of self. Another goal of good end-of-life care is dying with ‘dignity’. A desire to avoid loss of dignity, which may mean different things to different people, is a major reason given by patients in the Netherlands for requesting euthanasia11. Death with dignity is not always attainable even with the best possible care12. Lawton has suggested that the hospice movement has given the impression that dying can always be dignified because of its successful conquest of terminal pain9. Many dying cancer patients, and many more people dying from the end-result of multiple pathology in old age, have symptoms other than pain—such as delirium, urinary and faecal incontinence, sores and discharges. These can be successfully managed in fit elderly people but become less controllable as the body disintegrates. Lawton continues ‘As worthwhile as the new methods of pain control have been, the removal of one problem often leads to new problems surfacing in its place, and these latter problems may not be so readily resolved by palliative care... it may well be that, in retrospect, the pioneers of the modern hospice movement were a little too overzealous in their critique of the care of the dying in general hospitals.’

Calls for the principles of palliative care to be applied in acute hospitals are proper13. Harris has noted that ‘the disproportionate concentration of care on those dying of cancer has created an underclass of dying people’14. Nevertheless, it should be understood that patients in hospices and acute hospitals may be very different with different outcomes, however expert and caring the medical and nursing management. The likely reason is that pain is not a predominant part of their dying process. In 1999, the authors of the final report from the Debate of the Age Health and Care Study Group, under the auspices of Age Concern, proposed twelve principles of a good death15,16. Five of these principles are about control and choice by the person who is dying. It may not be possible to implement all of these principles when an elderly person, who has had little control and choice in her life for some years because of the burden of multiple pathology, finally succumbs to pneumonia in an acute ward. Yet dissemination of principles such as these can raise relatives' expectations and increase their distress at a vulnerable time if these expectations are not met.


In a patient with malignant disease, although the exact day of death may be unpredictable, one can normally identify a terminal phase in which he or she has not long to live. In the USA, Medicare funds hospice care (both domiciliary and inpatient) when death is likely within six months. The cost to Medicare of end-of-life care for people aged 85 and over is one-third lower than for people aged 65-75. This may be because more people are dying at older ages after lengthy chronic illnesses and long periods of functional decline17; and it may be the absence of a definable terminal stage that excludes many older people from hospice care. The median estimate of six months' survival of patients with congestive heart failure was 54% even within three days of death18 (this study was published after ACE inhibitors had been found to prolong survival in heart failure but before the increase in the use of beta-blockers). When patients have cancer, the impact of death on relatives can be lessened by working through a stage of ‘anticipated bereavement’19. This is not always possible when elderly people die.


Seale, a sociologist, has questioned accepted views about the needs of dying people and their carers, on the basis of findings in a nationally representative sample of 639 adults who died in the UK in 198720,21. A smaller parallel study compared reports given by some people with terminal illness with reports from relatives after their death. He noted that an important function of bereavement processes is to repair the damage that a death causes to social cohesion. He referred to other work that describes the maintenance of the human social bond as the most crucial human motive. Dying is a severance of this bond. He also noted that many bereaved people who were interviewed wanted to portray their actions in a moral light, expressing concerns about their moral reputations as part of re-establishing themselves in the social order. In some circumstances they used the research interview to help them in this process. When Lawton has spoken publicly about her research, she has often been approached by people who felt a sense of failure because their relative had not died in the way he or she had expected and hoped for9. Seale noted the importance of cleanliness and food as part of the social order of life. People who feel that social order is threatened by death may use a concern with cleanliness and food to emphasize their attachment to ‘a bounded secure sense of self’. Clearly those who do not feel such a sense of self will suffer and often become very agitated. Those who observe the effects of debilitating disease in people to whom they are close may feel a disintegrating sense of self which can release ‘a flood of elemental anxieties in a collapse of security about being in the world. A return to childhood sensations may be experienced, or imposed as carers engage in infantilisation of the elderly’. Seale points out that many of his respondents were using the Kubler-Ross concept of aware death as a ‘cultural script’ to understand their personal experience of the death of a loved one. However, this script was most appropriate for those with a specific terminal disease. If death was sudden, or associated with delirium, it could not be ‘written into’ this script. Seale found that relatives whom he interviewed for his research used the interview to assert their relationship to the moral order. They may use complaints in a similar way. In a study of complaints to a department of medicine for elderly people, relatives were more likely to complain about aspects of care relating to their own moral adequacy, if the patient had died22.


The approach in this paper should not be misunderstood as suggesting that, in old age, a dignified death is impossible or that relatives who complain are not pointing out real issues of care. Nevertheless, the findings do suggest that the issues outlined above have relevance to deaths of patients in general hospitals and the bereavement processes of their relatives. By understanding these processes, those working in such wards will become more sensitive to the feelings of relatives.

Integrated care pathways for dying people—a system aimed at improving care for people dying in general hospitals, based on dying processes of people with cancer—may not be totally relevant to people dying from other conditions23. Even in cancer, modern psychosocial support provides a broader view than that of Kubler-Ross, emphasizing intra-individual and inter-individual variation and cultural differences (such as issues relevant to gay men or black communities)8.

A common notion is that much of the distress of relatives is due to guilt for not doing enough for their parents or living up to their parents' aspirations. It is easy for staff to collude in this and so think less of relatives; and these thoughts may be unwittingly communicated to the relatives, provoking antagonism. The idea that bereavement is a disruption of the social order is similar but involves no criticism. Strategies to sympathize with relatives who are going through such internal struggles may help relieve distress. These strategies could include emphasizing the inevitability of death in old age as a normal part of human existence and the importance of registration procedures and funeral arrangements in retaining bereaved people within the social order. In Jewish tradition, following on from a funeral as soon as possible after death, there is a week of mourning when there is open house at the home of one of the immediate family. Family, friends and neighbours congregate in the house, particularly in the evenings when a short religious service takes place. The atmosphere, while serious, is rarely filled with gloom unless there has been the shock of a sudden death. The event is an accepted part of the social and moral order, allowing mourners to feel that they are part of society, not cut off from it. It helps to fulfil one of the crucial tasks of mourning as described by Seale16—‘to organise the human environment in the face of an otherwise disorderly invasion by death’. In Hindu contexts, the death of the person is believed to regenerate the entire cosmic order20.

Hanson and colleagues interviewed bereaved relatives to find out their perceptions of what could be wrong with end-of-life care24. They concluded that the primary concern of bereaved relatives was communication failure; discussions that focus on specific treatment decisions may not satisfy the real needs of dying patients and their families. The approach suggested here may help communication by explaining some of these real needs. Acute hospitals must do all they can to maintain the dignity of their patients; but at the same time they should not excessively raise expectations by implying that relief is available for every aspect of human suffering.


1. Hinton JM. The physical and mental distress of the dying. Quart J Med 1963;32: 1-21 [PubMed]
2. Lynn J, Teno JM, Phillips RJ, et al. Perception by family members of the dying experience of older and seriously ill patients. Ann Intern Med 1997;126: 97-106 [PubMed]
3. Kubler-Ross E. On Death and Dying. New York: Macmillan, 1970
4. Parkes CM. Bereavement: Studies of Grief in Adult Life. Harmondsworth: Penguin, 1986
5. Exton-Smith AN. Terminal illness in the aged. Lancet 1961;ii: 305-8 [PubMed]
6. Nuland SB. How we Die. London: Chatto and Windus, 1994
7. Lynn J. Perspectives on care at the close of life. JAMA 2001;285: 925-32 [PubMed]
8. Shipman C, Levenson R, Gillam S. Psychosocial Support for Dying People. London: King's Fund, 2002
9. Lawton J. The Dying Process. London: Routledge, 2000
10. Saunders C. The evolution of palliative care. J R Soc Med 2001;94: 430-2 [PMC free article] [PubMed]
11. Haverkate I, Onwuteaka-Philipsen BD, van der Heide A, Kostense PJ, van der Wal G, van der Maas PJ. Refused and granted requests for euthanasia and assisted suicide in the Netherlands: interview study with structured questionnaire. BMJ 2000;321: 865-6 [PMC free article] [PubMed]
12. McNamara B, Waddel D, Colvin M. The institutionalisation of the good death. Soc Sci Med 1994;11: 1501-8 [PubMed]
13. Brown M. Principles of palliative care are yet to be applied in acute hospitals. BMJ 2000;320: 1206 [PubMed]
14. Harris L. The disadvantaged dying. Nurs Times 1990;86: 26-8
15. Debate of the Age Health and Care Study Group. The Future of Health and Care of Older People: the Best is Yet to Come. London: Age Concern, 1999
16. Smith R. A good death. BMJ 2000;320: 129-30 [PMC free article] [PubMed]
17. Raphael C, Ahrens J, Fowler N. Financing end-of-life care in the USA. J R Soc Med 2001;94: 458-61 [PMC free article] [PubMed]
18. Levenson JW, McCarthy EP, Lynn J, Davis RB, Phillips RS. The last six months of life for patients with congestive heart failure. J Am Geriatr Soc 2000;48(suppl): S101-9 [PubMed]
19. Zlatiev-Scocard N. Bereavement follow-up; a preventative programme. In: Michel J-P, Hof PR, ed. Management of Ageing. Basel: Karger, 1999: 199-207
20. Seale C. Constructing Death: the Sociology of Dying and Bereavement. Cambridge: Cambridge University Press, 1998
21. Seale C. Theories and studying the care of dying people. BMJ 1998;317: 1518-20 [PMC free article] [PubMed]
22. Kafetz K. New ideas about death and bereavement associated with complaints by relatives. J Am Geriat Soc 2002;50: 978-9 [PubMed]
23. Ellershaw J, Coackley A. An integrated care pathway for the dying. Geriatr Med 2002;32: 45-9
24. Hanson LC, Danis M, Garrett J. What is wrong with end-of-life care? Opinions of bereaved family members. J Am Geriatr Soc 1997;45: 1339-44 [PubMed]

Articles from Journal of the Royal Society of Medicine are provided here courtesy of Royal Society of Medicine Press