Table 1 Commonly identified concerns of patients and families in ICU care, in progressive illness: data from literature, phase I interviews and observation

Symptom management, with distressing symptoms sometimes not addressed [5–16, 31]

There was a general perception among family members that symptoms were well controlled, and that timely action was taken when the patient appeared to be distressed. There was varying documentation of assessment of symptoms and symptom assessment tools changed frequently, for example, from one nursing shift to the next

Most interviewees thought that end-of-life care was well provided by the medical and nursing workforce, which was considered very experienced

Communication issues, in particular communicating the changing circumstances to patients and families; [17, 31–39]

The importance of being able to ask questions and have them answered was emphasized by family members, but there was also a sense that family members felt they were responsible for getting information by asking the right questions of the right people. Particular words stood out for some people, reinforcing the need for careful choice of the words in this context. Factors described by family members as affecting communication included: their poor memory of the details of a discussion; shock affecting their ability to take in information; and their poor knowledge of medical issues making it hard for them to understand all the information given. There was an overall sense that the information given was complete and honest, that family members found it helpful if bad news could be tempered with good news, and that the uncertainty of outcome was explained well

Improved information for families including what to do in the event of a death; better privacy/side rooms for patients and family members; private space to talk to families; a position on reception for someone to greet and support family members; more support for family members and provision of on-site accommodation

Dealing with prognostic uncertainty and decision-making, which may lead to prolonged dying [25, 40], or conversely, very rapid deterioration of the patient once a decision to withdraw treatment is made, with families sometimes feeling abandoned [31, 33, 35, 36, 41–44]

The greatest factor influencing involvement in decision-making was patient capacity. Preferences of family members for involvement varied; generally most of them wanted information about the process but not greater input (some preferring less). Some types of decision were more likely to be influenced by family members (place of care, aim of care, some interventions for example, tracheotomy) whereas others were more often directed by the clinician (for example, resuscitation status)

Decision-making towards the end-of-life in ICU is complex and multifactorial. Although not everyone was aware of it, the LCP was reported to be used particularly on the medical ICU. It was thought to be useful for patients on a longer dying trajectory and for those discharged to the wards, to ensure continuity of care. There were concerns about its length, but also some support because it included components of basic nursing care. It was felt by some to be unnecessary in contexts where there was one-to-one nursing care and where many patients died too quickly to benefit

Meeting individual wishes, expectations and spiritual needs that vary between different social and cultural groups [21–24]. Resource use, organ donation, and surrogate decision-making, all of which may involve conflict [21, 22, 33, 45, 46]

Family members identified factors such as financial and legal concerns; parking and transport; information needs; religious, cultural and spiritual needs assessment; and documentation and management. The importance of understanding family structures was emphasized by some family members, along with ensuring that information about the patient was given to the correct family members. When this was done well it was praised by those interviewed, but there were instances of family members feeling that their relationship was not respected by staff or that information was being given to someone other than a person designated to receive this information, and these situations caused distress

Greater support for staff and debriefing in difficult cases, although there were contrasting views about the usefulness of debriefing; a support group for nurses, particularly for those new and less experienced; better support for staff, for example the opportunity of access to an external counseling service; and more educational opportunities

Supporting dignity, respect and peace of patients and supporting the family [19]

Family members often described not knowing when discussions with staff would take place, and a few suggested that having a plan for future communications would be beneficial

Staff suggested changes to practice including: a ‘home to die’ service; more spiritual support; more hands-on nursing, massage, washing, support for family members; overt recognition of the importance of caring for the dying so that junior staff do not see it as a failure; reducing noise levels where possible; a bereavement service; and improved communication with other teams, including palliative care