Variations by age in symptoms and dependency levels experienced by people in the last year of life, as reported by surviving family, fiends and officials

Age and Ageing, March, 1998 by Julia Addington-Hall, Dan Altmann, Mark McCarthy

Introduction

Although most newly-diagnosed cancer patients are over the age of 65 [1], until recently little attention has been paid to cancer care in elderly patients. Cancer patients over the age of 70 have been excluded from some clinical trials and there is evidence of under-treatment of elderly cancer patients [1]. Little is known about the quality of palliative care received by these patients. Cartwright and Seale's 1987 survey of a random sample of deaths from all causes found that people dying when they were 85 or older were less likely than younger patients to be admitted to a hospital or hospice and, despite evidence of greater need, did not get more care from general practitioners and community nurses [2]. This suggests an age bias in care for the dying. Further support comes from reports of older cancer patients receiving less adequate analgesia than younger patients [3] and having more unmet needs for help with personal care [4].

It was recently suggested that the hospice movement is prejudiced against elderly people, spending most of its resources on relatively young people [5]: this has been strongly denied [6] and results of a survey of UK inpatient hospices which reported that 65% of new inpatients admitted in the period 1994-1995 were aged 65 years or above appears to support the rebuttal [7]. However, fewer than half the units were able to provide information on age, and the survey does not report the percentage of patients aged 80 and above--the age group shown in an Australian study to be particularly unlikely to receive hospice care [8].

Even if there were definitive evidence that fewer than expected older cancer patients are admitted to hospices, this would not in itself demonstrate an age bias: it would also be necessary to demonstrate that older cancer patients are as likely as younger patients to need the specialist skills of hospices in symptom control, open communication and psychological support.

Cartwright and Scale looked at the association between symptoms and age and concluded that "the main finding is a lack of difference between people dying at different ages": there were no age differences in the proportion reported to have experienced 15 symptoms; mental confusion, incontinence, difficulty hearing and seeing and dizziness were more common in elderly subjects and only a persistent cough was more common in those dying at younger ages [2]. However, the symptoms were reported to have been more long-lasting in elderly patients and were less likely to be reported as having been `very distressing'. They suggest that this may have been due to "the different relationship between respondents and the people who died", "to different perceptions of distress among older and younger people, to a more stoical attitude among older people, to an adjustment to persistent and long-lasting symptoms or to a real difference in the intensity of symptoms and the resulting level of distress".

Some evidence for the latter interpretation comes from the finding that newly-diagnosed older cancer patients reported less symptom distress than younger patients [9]. If older cancer patients have less distressing symptoms then this would suggest that they have less need of specialist palliative care than younger patients. However, a comparison between cancer and non-cancer deaths showed that people dying from cancer tended to have more distressing symptoms which were of relatively short duration while people who died from causes other than cancer had less distressing but more long-lasting symptoms [2]. Cancer deaths were on the whole at a younger age than non-cancer deaths. This suggests that the finding that older people had fewer and less distressing symptoms than younger people may be an artefact resulting from the larger proportion of non-cancer deaths in this group or the larger proportion of cancer deaths in the younger group.

In this paper we use data from the Regional Study of Care for the Dying (RSCD) [10], a retrospective survey of deaths in 1990 in 20 English health districts, to look at the relationship between age and reported symptoms in the last year of life in those who died of cancer and in those who died of other causes. We also explore variations by age in the level of dependency in daily activities for cancer and non-cancer deaths.

Methods

All district health authorities in England were invited to take part in the RSCD. Twenty districts from a variety of inner-city, outer urban and rural settings did so. Although self-selected, they were nationally representative in terms of their socio-demographic characteristics and on many indicators of health service provision and usage [11].

In each district 270 deaths were randomly selected from those in the last quarter of 1990. Cancer deaths were over-sampled (54%) to provide sufficient numbers to evaluate specialist palliative care services. Following a letter describing the study, trained interviewers wrote to the deceased's last known address (as given on the death certificate) to begin their search for the best informant about the deceased's last 12 months of life. They then conducted a semi-structured interview. Topics covered included symptoms experienced in the last year of life, dependency levels, use of and unmet need for health and social care and the respondents' satisfaction with these services and experience of bereavement. The median time between the interview and the deceased's death was 10 months, with an interquartile range of 9-12 months. A 69% response rate was achieved (3696/53751). The study methods were based on those used by Cartwright and Scale [2] and are described fully elsewhere [10].