End-of-life care for older patients dying in an acute general hospital can we do better?

SIR—The successful provision of a high standard of care to the dying in our population represents a significant healthcare challenge, a challenge influenced by patient age, disease process, place of care, and by the expertise of the professionals caring for patients at the end of life. Recent estimates predict that by the year 2020, 26% of the UK population will be aged 65 years and older [ 1 , 2 ]. It is also known that the majority of patients who die in acute hospital care are over 65 years of age [ 3 ]. Though it is recognised that excellent end-of-life care is provided in the hospice setting [ 4 ], there is less evidence relating to the provision of this care in hospitals. In anticipation of these predicted demographic changes, statutory and specialist bodies [ 1 , 2 , 5–9 ] have produced guidelines and recommendations suggesting how better end-of-life care should be provided to older people in all care settings. These documents highlight the current inequity of access to good end-of-life care, the societal, professional, organisational and statutory obstacles to improvement in this area, and to outline how these inequities should be remedied. One such document, the Liverpool Care Pathway for the Dying Patient (LCP) [ 4 , 10 ] provides an evidence-based framework for the delivery of care to dying patients, promotes excellent documentation of all aspects of the care provided and lends itself easily to the audit process. Originally designed to enable transfer of the hospice model of care to other care settings, the LCP is used increasingly in the hospital sector. The purpose of this study is 3-fold. We wish to review the extent to which evidence of the care being provided to older patients dying in an acute hospital was documented, to ascertain whether or not the future implementation of the LCP on the wards of the Department of Medicine for the Elderly (DME) would benefit patients and staff, and to ascertain whether any of the admissions to hospital of patients who subsequently died there might have been avoided and these patients more appropriately managed within the community. This study is a retrospective case-note review of the medical and nursing notes and drug charts of those patients who died under the care of the DME at University Hospital Aintree in a selected month. The cohort of patients was identified and their case-notes retrieved by the hospital medical records department. The chief investigator (FT) used a proforma developed by the authors to retrieve the required data from the case-notes relating to each patient's final admission only. Data retrieved included basic demographic information, evidence of family and professional support in the community prior to admission, the reason for admission, diagnosis(es), and detailed information relating to the care provided to patients as they approached death, as documented in the case-notes. Twenty-five patients died during the study period and were included in the study. Their demographic data are presented in Table 1 . Evidence of domestic and community support was poorly documented. Patients had a range of primary diagnoses while several had multiple active disease processes. Only one patient had a malignant diagnosis. Demographic data ( n = 25) a Adenocarcinoma of sigmoid colon (1), old age (1), general deterioration (1). Demographic data ( n = 25) a Adenocarcinoma of sigmoid colon (1), old age (1), general deterioration (1). The details of care provided at the end of life are presented in Table 2 . Recognition that a patient's death was imminent, occurred most frequently between 24 and 48 h prior to death, and the likelihood that the death of their relative was approaching was discussed with family members in the majority of cases. Of note, there was no documentation of such a discussion being held with any patient. ‘Do Not Actively Resuscitate’ orders were made at a range of times prior to death with over half the orders put in place in the last 72 h of life. Details of end-of-life care ( n = 25) a Two patients died unexpectedly. Details of end-of-life care ( n = 25) a Two patients died unexpectedly. The symptoms of agitation, dyspnoea, nausea or vomiting, pain and pharyngeal secretions were documented to be present in the patients in our study during the last 3 days of life. Despite the recognition and documentation of these symptoms widely recognised to occur in the later stages of life, many patients were not prescribed either regular, or as required, treatment to alleviate them. In only two cases was the advice of the specialist palliative care team sought. On review of the intensity of care provided to the patients studied, it was felt that admission to an acute hospital had not been necessary in six (24%) cases and could have been avoided. Five of these patients, those admitted from care homes, survived a short period following admission (combined length of survival 24 days) and all five died as a result of a not unexpected progression or complication of known long-standing illnesses. This study is limited by its size, the retrospective approach employed, and in particular by its reliance on the presence of written documentation as the sole evidence of care provided to the patients studied. Despite the body of evidence confirming that healthcare professionals have difficulty in accurately predicting likely survival in patients with advanced malignant [ 11 , 12–14 ] and non-malignant disease [ 10–12 ], our study shows that medical and nursing staff were able to recognise symptoms associated with dying and when death was imminent for most patients. In spite of this, they often failed to discuss their concerns with patients and frequently neglected to prescribe appropriate symptomatic relief for them. It is known that elderly people are less likely to receive hospice care or to receive appropriate pain control than those younger than them [ 8 ]. Suggested reasons as to why this might be the case include the difficulty of survival prediction in non-malignant disease, a reluctance to acknowledge treatment failure or to relinquish end-of-life patient care to other specialists, a reluctance to engage in conversations about death and dying with patients and their families and/or a lack of generic palliative care skills amongst healthcare professionals [ 11 , 12 , 15 , 16 ]. The fact that palliative care advisory support was sought for two of the patients in our study only may give credence to these views. On the other hand, Watson et al . [ 17 ] illustrate how the LCP implementation process can lead to the removal of some of these barriers. The NHS End of Life Care Programme (EoLCP) [ 18 ], established in 2004 has as its three pillars, the Liverpool Care Pathway, the Gold Standards Framework [ 19 , 20 ] and Preferred Place of Care [ 21 ]. It states that ‘the care of all dying patients must improve to the level of the best’ and describes the need to give people greater choice in their place of care and death, to reduce the number of emergency admissions to acute care for those who wish to remain at home and to reduce the number of patients transferred from care homes to acute care in the last weeks of life as key tenets of its strategy. The findings of this study demonstrate that the implementation of the EoLCP would benefit older patients dying in hospitals by enabling those who care for them to do so better. Despite its limitations, our study highlights that it is possible to improve the standard of care we deliver to older hospital inpatients at the end of their lives and strongly suggests that there are improvements to be made in our documentation of the care provided. The implementation of the LCP would bring improvements in these areas at least. Evidence exists to support the benefits of LCP implementation in the nursing home and acute hospital settings [ 22 , 23 ] and in relation to control of end-of-life symptoms [ 24 , 25 ]. In addition, work is ongoing in the United Kingdom to assess carers' views of the end-of-life care received by their relatives, and in Holland to investigate the impact of LCP implementation on carers and health professionals by comparing experiences in settings where the LCP is and is not in use. The results of these ongoing evaluative studies will inform the need for further evaluation of the LCP and the other components of the EoLCP and the form(s) that such evaluation should take. The increased dissemination and implementation of the NHS End of Life Care Programme and its three key tools will enable those caring for older patients dying in hospitals, hospices or in the community to do so to the highest level and importantly, allow patients to indicate their preferences for place of care at the end of their lives. Symptoms commonly occurring at the end of life may be recognised, but are not always treated. Admissions of dying older patients from care homes to acute hospitals may be inappropriate and unnecessary. Recognition and acceptance of the dying process are essential if we are to provide the highest level of end-of-life care to older patients dying in acute hospitals. Mechanisms by which excellent end-of-life care can be provided have been proposed by national and international governmental and specialist bodies. These should be embraced and put into practice in all healthcare settings. None.