What is known
There is evidence that supporting relatives’ involvement with the dying person is an important aspect of end-of-life care – for both patients and their families. A systematic review of experiences of relatives in end-of-life care identified a large body of research using a range of methodologies, but there were no randomised or intervention studies. However the evidence identified was consistent. Satisfaction of relatives depends on how the family experiences communication and support from the healthcare team. [1-2]
Clinicians should understand the challenges of meeting patients’ and families’ needs for information that is honest but can also sustain hopefulness. Caregivers can learn communication strategies that foster realistic hope. 
Two systematic reviews of family members acting as surrogate decision makers suggest that although family members are not completely accurate when they try to predict the dying person’s wishes (they were accurate in about two thirds of the studied situations) nonetheless they are more accurate than healthcare professionals. [4-5]
There are considerable psychological and emotional burdens for family members taking on the role of surrogate decision maker. Advance care plans, when they exist, reduce the distress of surrogate decision makers. Many families have a preference for consensus decision making, ideally involving the whole family. 
Ethnic, religious and cultural issues can have a very significant impact on decision making processes, and on how the healthcare team relates to the family. Some cultures do not share Australian norms in which the dominant idea is that of individual autonomy. Attitudes to specific aspects of care (such as the discussion of dying or resuscitation, withdrawal of life support or other active interventions, analgesia and sedation, or fluids and feeding) can be affected by religious and cultural differences.  Language and cultural differences may be a significant barrier to care. 
The issue of patients who request hastened death has been studied.  The evidence base is not extensive, and relies mostly on either professional assessments of the patient who made the request, or family assessments. Both of these approaches have significant methodological problems. In the small number of studies where patients themselves were included, it appears that existential and psychosocial factors were common, and fear of pain rather than uncontrolled physical symptoms was also a factor. As the domains were poorly defined in these studies it is not possible to draw more definite conclusions, and all studies would be classified as descriptive research.
A systematic review has explored the problem of patients who believe that they are a burden to others at the end of life. The literature that was identified in this review was small, but suggests that this problem can significantly affect the patient’s experiences at the end of life, as well as having an impact on choices that they may make. There appears to be an association between self-perceived burden and requests for hastened death. 
What it means in practice
- The support and care of families as well as patients is a crucial part of end-of-life care. This is best delivered by the involvement of a multidisciplinary team, to ensure access to psychosocial support, pastoral care and bereavement support. 
- End-of-life family conferences allow for consensus decision making by families, and can support families to identify realistic goals of care and contribute to decision making, whilst also assisting them to deal with their own distress. [11-13] Family conferences require well developed communication skills within the healthcare team.
- Careful communication, and access to health interpreters, are required to identify and address differences in cultural perspectives in end-of-life care, and to respond respectfully to these differences.