-Approximately half of all patients who die do so in hospital. Despite the advent of palliative care in the UK, there is evidence that the care that many patients receive in the final phase of their illness in hospital is poor. Building on a study of bereaved relatives' views of the information provided by an inner city hospital trust during an admission in which a patient died, this article explores the factors that may contribute to sub-optimal care for patients dying in hospital.
IntroductionApproximately half of all patients who die in the UK do so in hospital. There is evidence to suggest that these patients have multiple symptoms, 1 and that the care many of these patients receive in the dying phase can be poor. Observational studies have shown that many dying patients do not receive basic comfort care (such as mouth care), are not given assistance eating and drinking, and that contact between healthcare professionals and patients is minimal. 2,3 Costello's recent observational study of patients dying on care of the elderly wards showed that nurses tended to focus on meeting patients' physical needs, while avoiding open communication about prognosis. In consequence, they rarely addressed important aspects of psychological and spiritual care. 3 Despite numerous studies that have highlighted the benefits of open communication at the end of life among cancer patients, 4 these needs are often unmet. [5][6][7] Recent studies using the views of bereaved carers (informants) to evaluate the care given to cancer patients in the last year of life confirm that poor communication at the end of life remains commonplace. 8,9Experience of dying in an inner London teaching hospitalIn 1998, a post-bereavement survey was conducted as part of a programme to identify and address the needs of patients dying in an inner London teaching hospital trust. This had arisen out of concerns regarding the care of this patient group. The methodology for this survey has previously been reported. 10 Of the 182 informants invited to participate in the study, 78 (43%) returned completed questionnaires. Thirty-six (20%) of those invited expressed a wish not to participate and 68 (37%) did not respond. There were no differences between the responders and non-responders. The majority (95%) of informants classified themselves as white; 54% were a spouse and 27% a son or daughter of the deceased. The most common cause of death was cardiovascular diseases (36%, including cerebrovasular disease); a further 20% had died from infective causes, predominantly bronchopneumonia; and in 13% the cause of death was cancer. There are several potential limitations to this study, including the low response rate and the lack of representation in informants from ethnic minority groups. In an area of high socio-economic deprivation, the views of informants who were unable to read or write in English may have been under-represented. Some doubts have been expressed about the reliability of such proxy reporting. 11-13 However, the research evidence suggest...