Introduction

Our learning over 35 months from patients and families from a diversity of ethnic, faith and sociocultural backgrounds who were living with advanced disease and with the likelihood of deterioration and dying over the coming months has been described in the previous chapters. These accounts paint a rich and diverse picture of how people live with the prospect of their death and whether or not and how they consider their future.

We acknowledge the difficulties in summarising participant experiences and perspectives because these relate to many clinical encounters in many different settings and are characterised by great diversity.¹³² However, despite specific areas of dissatisfaction, most participants were broadly positive about their experience of care for the patient. A few had serious complaints and some were extremely satisfied. Dissatisfaction tended to relate to general shortcomings in care and communication, which have been widely reported previously.^24,72 A few complaints were specific to a patient’s ethnicity and culture, but, overall, this was not a strong theme in the data.

Our findings highlight the unique nature of every patient and family, and the differences between individuals within the same family and ethnic group, as well as between them. The findings also reveal many similarities in the experiences of EOLC and communication between patients and FCGs from ethnically diverse groups and from the wider UK population, as well as patient and FCG responses to EOLC and communication.^69,88 Previous work by co-applicant Kristian Pollock with white British participants concerning ACP discussions has found that prognostic uncertainties made such discussions challenging and that many HCPs found it difficult to raise the topic and recognise when patients were ready to talk about the future. Most patients preferred to leave such discussions until they had become severely unwell, and the findings challenged the policy assumptions about ACP that all patients wish for, or would benefit from, an open discussion about death.⁷²

Furthermore, a systematic review of the ACP experiences of patients with life-limiting illnesses identified diversity of experiences and individual preferences in engaging in discussions that aligned to three themes, which can be seen as a continuum along which patient receptiveness and desire for ACP rest, but which may also fluctuate, bidirectionally, during their illness: ambivalence to the balance of benefits and burdens of thinking ahead about future deterioration, ‘readiness’ as a prerequisite for the benefits, and a degree of openness or comfort in discussing preferences for the future.¹³³ These themes interlink as ‘two sides of the coin’ with respect to barriers to and enablers of both discussions about future deterioration and the development of personalised advance care plans that seek to enable self-determined preferences in that future.

Our findings affirm that these underpinning themes present barriers and facilitators for people from a diversity of ethnic and religious backgrounds. In addition, they indicate that there are distinctive and highly nuanced aspects of the experience and circumstances of patients and FCGs from ethnically diverse communities that compound the complexities alluded to in other studies of ethnically diverse and palliative care the UK.^23,134,135 The specific issues on which we focus here relate to the findings concerning understanding and engaging with EOLCP; prognosis; awareness of dying and thinking ahead; language, communication and decision-making; models of care; and navigating a complex system of health care.

Understanding and engaging with end-of-life care planning

Our findings indicate that there is some confusion, especially among patients and FCGs, about what constitutes EOLC and when discussions concerning this should be triggered. This serves to confound the current policy that it is valuable for people to have such discussions if they are to be enabled to express preferences and plan future care and obtain appropriate services at the end of life.¹³⁶ What constituted discussion about ‘thinking ahead’ and whether EOLCP had occurred was often ambiguous and hard to recognise.

In their accounts of thinking ahead, participants tended to focus their attention on death and what happened afterwards, such as funerals and wills, rather than the preceding and uncertain process of dying.^136,137 Some patient participants described their efforts to make arrangements in advance to help relatives cope with loss and administrative burden, to leave a legacy of care and support after their death. Several participants had completed lasting power of attorney documents, for financial affairs, and a few expressed a wish for their ashes to be repatriated to their native countries. Concern with funerals and disposal of the body reflects what was of most immediate importance to people and was how they tended to interpret questions about EOLCP.

Decision-making around EOLC was not so pertinent to participants, perhaps because they concerned experiences that were unknown, unwelcome and could not be easily imagined.^136–138 Consequently, reference in the accounts of participants to anticipatory questions about, or discussion of, what was likely to happen during the process of dying was unusual.¹³⁷ There was little detailed consideration in the data of the content and formal documentation of EOLC plans, although they were valued by the few families who reported having completed them. However, participants did often recognise specific topics that were mentioned by researchers during the interview (e.g. resuscitation, preferred place of care and death, limits to care and treatment) and described their experiences of talking about these. Although ACP should involve a holistic discussion of ‘what matters to me’ at the end of life, our participants’ experiences highlight the extent to which this consisted largely of a simplified, disaggregated, task-based focus on specific, documentable items such as decisions about DNACPR, and place of death.^{3,10,33,72,139}

Offering opportunity for conversations with people in the last year of life about future deterioration is supposed to fall within the remit of all HCPs, regardless of role.^140,141 However, in the Thinking Ahead study, many HCPs did not consider discussion of EOLCP to be their responsibility, suggesting this was the remit of palliative care services. They did not consider that it was relevant or appropriate to raise EOLC topics when the patient was still receiving active treatment and was not perceived to be close to death. They reported rarely initiating discussion of the topic, and not before the patient was clearly gravely ill.^142,143 Some positive accounts were given of consultations being scheduled to discuss EOLCP within the context of an established relationship with the patient’s GP. As widely reported in other studies, HCPs were concerned that talk about end of life, especially if mistimed, risked damaging their relationship with patients, and the trust they had established, as well as destroying hope.^72,143–145 Thus, there was often an alignment of motivation between patients, FCGs and HCPs to maintain a positive recovery orientation and avoid the topic of end of life.^72,146 Not knowing too much about the prognosis could be a prerequisite for maintaining hope, optimism and engagement.

Professionals emphasised the importance of identifying ‘readiness’ in patients and family members and the need to be alert to cues that they were receptive to discussion of future planning.^72,144,147 Policy drivers towards open and transparent communication between patients and professionals, and the encouragement of partnership and shared decision-making in medical consultations, have been promoted for decades.^148,149 In the face of this, the persistent entrenchment of patterns of communication in which ambiguity, and even denial, may be preferred over clarity points to the functional and adaptive capacity of the positive recovery orientation for many patients.^69,148 Although a positive recovery orientation may suit many patients and FCGs, there are some, a minority in our study, who have a strong desire to be fully informed about their illness, prognosis and future care to be aware of their ending and prepare for it.^150,151 In these cases, the positive recovery orientation might work to their disadvantage, much as previous studies have reported over a period of more than half a century.^{148,152–154}

Individuals who sought to know the details of their illness and prognosis became extremely frustrated by what they perceived to be professional reluctance to address these issues, and by their evasion and indeterminacy, whether this stemmed from deliberate evasion or lack of expertise. Some participants appreciated that professionals themselves were subject to the intrinsic uncertainty of prognosis. Their experience evokes the distinction originally made between ‘clinical’ and ‘functional’ uncertainty.¹⁵⁵ ‘Clinical’ uncertainty refers to the real unpredictability of illness progression and outcome, whereas ‘functional’ uncertainty is a strategy professionals adopt to manage their engagement with patients and families and avoid being drawn into difficult and protracted conversations. Some family members in the study adopted a similar strategy in withholding information from relatives, especially patients, to avoid distress and maintain hope.⁶⁹ This effort to protect patients from awareness of their poor prognosis has been widely reported among ethnically diverse populations.^{99,136,156–158}

In contrast to the circumspection with which professionals often approached EOLC discussion during scheduled or routine encounters, FCGs described a much blunter approach prompted by patients’ urgent hospital admissions following a crisis or rapid deterioration in their condition. The families described distressing experiences when professionals sought rapid answers to questions, notably about resuscitation, for which they were not prepared. These often occurred during times of great distress, when the patient was acutely ill, and relatives were struggling to come to terms with the gravity of the situation and the uncertainty of recovery. Consequently, the setting and context within which discussion of advance decisions occurred was an important determinant of patient and FCG experience and willingness to engage with the topic.¹⁴⁸

It might be proposed that it is precisely the purpose of advance discussion and documentation of wishes to pre-empt the occurrence of such situations. However, as Mol¹³⁸ and others^{136,159–162} have argued, decisions by patients and FCGs are always hampered by the intrinsic uncertainty and incomplete nature of the information they have at their disposal about what is going to happen in the future and what is the best way to respond. Patients and FCGs (from any culture) tend to adopt a cautious, provisional and pragmatic approach to preferences for, and anticipation of, their future end of life, anticipating that these may change, as their illness progresses and their circumstances decline.¹⁶² The current focus on ACP in the UK and other Western nations is underpinned by an ideological commitment to freedom of choice as a condition of individual autonomy.¹⁵⁴ However, the relevance of individual autonomy to the way most people live their lives, and in particular in isolation of their relationships with other people, is being called into question: the consumerist notion of autonomous ‘choice’ has limited relevance, particularly in relation to patients from ethnically diverse groups for whom individual autonomy may not be the predominant underpinning to decisions.^{72,136,138,156,158,163,164}

The limited and ethnocentric stance of current policy is underlined by consideration of the alternative approaches that have been identified in research from ethnically and culturally diverse groups. Respect for persons is manifest in the desire to protect the patient from an unwelcome truth, in the desire for the family or elders to assume responsibility for making decisions about treatment and in the commitment to care by family members, particularly of adult children for their parents.^{99,106,136,156–158,165} A natural orientation to such approaches was apparent among many of the study cases. Alongside this was a strong preference by many patients to protect their family members from the burden of care and for their choices for treatment and care to be considered carefully in relation to their anticipated consequences for others.^72,138

Patient desire to reduce the burden their illness and death imposed on their significant others was a recurring theme, and was one reason for the very specific focus on making preparatory arrangements for after their death, especially the funeral. Thus, some patients expressed a preference to move from home to hospice once the demands of home care had become excessive, and to withhold information about their illness to protect individuals from distress. Just as HCPs search for a window of opportunity and assess the ‘readiness’ of patients to discuss EOLCP, patients and FCGs also exercise judgement and caution about what information about their future they wish to share and discuss with others, including HCPs. This prioritising of the interests of others can be viewed as a manifestation of agency and is also a reciprocation of the love, care and support bestowed by others on the patient.

Who to talk to about illness and when was influenced by perceptions of an individual’s situation. Information might be deliberately not shared with people who were themselves ill, who had experienced bereavement or who were thought to be in some way vulnerable. Patient awareness of the family suffering as a result of their illness was itself a source of suffering for the patient. The selection and management of information could be a means of patients asserting personal agency through protecting others from the distress and anxiety caused by terminal illness. Within informal networks it could be considered a form of reciprocal gift exchange, cementing ties of affection, obligation and responsibility. It could also be a source of conflict and tension, whereby individuals disagreed in their views about what information should be shared, especially with and by the patient.

In some families, EOLCP may not be something that members wish to discuss with HCPs. The family, or specific individuals within the family, may be considered the natural locus of discussion and decision-making in such matters. Thus, it is important for HCPs to be aware, and take account of, the nexus of relationships within which the patient is embedded, and to involve family members, as well as the patient, in planning and undertaking EOLC discussions.¹⁶⁶ In tension with this preference for fostering of optimism present in the majority of our case studies is the testimony of some bereaved FCG participants who, in retrospect, expressed regret that they had missed the significance of the final days or weeks of the patient’s life and perhaps even the chance to say goodbye. Regardless of what they had been told or understood, participants often seemed to have been taken by surprise by the speed with which the patient’s final decline and death occurred.

Prognosis, awareness of dying and reluctance to think ahead

Staff from disease specialist services such as oncology and cardiology largely expected colleagues from district nursing and palliative care services to initiate discussion of EOLCP with patients once they had exhausted all active treatment options and been referred to community services for ongoing care. However, prognostication, especially in relation to patients affected by conditions other than cancer, is notoriously difficult and prone to error.^167–170 The positive recovery orientation encouraged patients and families to maintain hope in the possibility of an extended future and to distance themselves from the imminence of death.^{154,159,171–173} Having pulled through against the odds, on past occasions, FCGs proposed that the patient was likely to do so again, or at least they should be treated as if this was likely to be the case. There could be a lot of ambivalence about prognosis and how much people wanted to know about this and their future options. Moreover, preferences were not necessarily linear or stable.^138,162 Many participants preferred to maintain an overtly positive stance and to focus on making the most of the present, rather than an uncertain and probably unpleasant future.^72,164 Patients, FCGs and HCPs in many of the case studies converged in their view that there was no benefit, and could be a risk, in thinking too far ahead until the patient had become so gravely ill that death was clearly imminent.

Health-care professionals rarely expressed any awareness, or curiosity, about patients’ religious or other beliefs, their significance for EOLCP, how they should be factored into the discussion or how they might influence decisions about treatment and future goals of care. Patients and FCGs who held strong religious convictions about divine intervention regarded EOLCP as irrelevant and inappropriate. Indeed, when all matters of life and death were believed to be determined by a god, future planning could be considered presumptuous and indicative of insufficient faith. However, there was considerable variation and individuality in attitudes towards EOLCP and how participants positioned themselves in relation to religious or cultural affiliation. Preferences for resuscitation and continuation of active treatment, for example, could not be reliably predicted from knowledge of an individual’s religion or ethnic identity. Some families rejected DNACPR as being incompatible with religious faith, specifically the belief that life and death are subject to divine determination. Others had a purely secular desire to continue living for as long as possible, because of a hopefulness about the future and what this might deliver in terms of natural recovery or advances in treatment.

Language, communication and decision-making

Previously, we suggested that the optimistic recovery orientation often involved a degree of collusion between patients, FCGs and HCPs. This was an important mechanism for protecting others from damaging revelations and confrontation with unpleasant alternative truths, but it could also result in misunderstanding and false optimism among patients and family members, who failed to realise how seriously ill the patient was and how close to the end of life. Although this approach suited many participants, at least for a time, this vagueness and indeterminacy was regarded as highly unsatisfactory and intensified anxiety among patients and families who wished for clear and specific information about their illness and prognosis. Faced with very uncertain patient futures, HCPs were unable, and possibly reluctant, to pin down exactly when treatment might fail or deterioration might occur. However, HCPs had considerable power to frame information to influence patient awareness of the illness and the ‘choices’ with which they were confronted.^72,160,174 This was a much more general phenomenon than the blunt imperative for decisions to be made during acute hospital admissions, as described previously.

Regardless of the setting, HCPs raising the topic of resuscitation was a powerful signifier of the gravity of a patient’s illness and anticipated mortality: after discussion of DNACPR, the future abruptly felt shorter. It was clear that participants often did not have a clear or accurate idea of what resuscitation involved or the circumstances under which it would be attempted, or that it did not mean that treatment would be withheld for treatable or reversible conditions. However, several FCGs described situations when they felt that hospital staff had indicated that no further treatment of any kind would be given to their relative.

In discussions of resuscitation, professional input can be critical, in giving information, time, space, and repeated opportunities for patients and family members to reflect and discuss. Accounts of participants in this study identified their confusion and inconsistency in professional judgement and advice that had been distressing and, they felt, had resulted in poor care. It was evident, also, that some families disputed professional judgements about a patient’s quality of life being necessarily unacceptable following a resuscitation attempt, and disputed their capacity to predict prognosis. Others found the decision about resuscitation to be extremely difficult and one they were reluctant to accept.

Several participants described feeling under pressure from HCPs to agree to DNACPR orders and both lay and professional participants gave accounts of patients and FCGs changing their minds. It was not clear if this was following the chance for reflection and to obtain greater information and understanding, or if patients may have come under professional, family or other pressures to accept or make specific decisions. Especially when translators were required, it was not clear how much patients understood about future planning and decisions. Accounts of patient involvement in decision-making varied widely, from being actively in charge, through shared discussion, to family members taking full responsibility.

Some patients, especially those who lacked fluency in English, were reported never to see a HCP on their own, but always accompanied by at least one relative. Patients valued being able to communicate with HCPs directly in the patient’s first language. There were a few accounts of consultations in secondary care involving a HCP who could converse in the patient’s first language. These were off-chance encounters, following which it was unlikely that the same HCP would be seen again. If they cannot communicate directly with the patient, there is the potential that professionals may wrongly assume the locus of decision-making to be the family, especially in the case of older patients who rely on family members as translators and advocates.

Translators may act as gatekeepers, whether these are family members or professionals, framing and shaping patient understanding and choice.^56–58 Thus, patients who require translation may struggle to have their voice heard and HCPs may feel uneasy about a decision being in keeping with the patient’s wishes and best interests. Disclosing bad news requires skilled, empathetic communication, which is particularly difficult to achieve through third-party translation. Some relatives felt that HCPs had put on them the responsibility of translating on behalf of their parents, and recoiled from the decision-making responsibility, particularly about difficult decisions such as resuscitation, which they felt amounted to ‘signing a death warrant’ for their loved one. Thus, for some FCGs, being asked to act as an interpreter added to their emotional distress and burden of care.

Previous studies have reported that HCPs are dissatisfied with family members acting as translators because they suspect that they may be selective in filtering information given to the patient, and also will also find it difficult to explain technical details of illness and treatment options about which they lack knowledge and understanding. However, relatives are often reluctant to use translation services, from a desire to control information being conveyed to the patient, and also a concern about disclosing confidential information to strangers. It is evident that the need for translation continues to constitute a barrier to skilled and sensitive EOLC, particularly in a context of complex, challenging and difficult experiences relating to end of life in which, effectively, communication is care.¹⁷⁵ Wilkinson et al.¹⁷⁵ reported that understanding and effective communication were enabled by the availability of staff who could converse in a patient’s first language. The use of bilingual HCPs should be sought whenever possible.^56–58

However, even when consultations are conducted in English, it may be that patients and FCGs lack the fluency or cultural acclimatisation required to understand technical terms or pick up on the nuances of professional communication. In consultations involving participants from ethnically diverse communities, it is particularly important that professionals exhibit curiosity about patient and FCG understanding of the situation and interpretation of what they have been told, and elicit their concerns and expectations of illness progression and options for future care. This information must be placed within the context of the wider ecology of professional and informal care within which the patient is embedded and within which care is enacted.^166,176

Uncertainty is intrinsic to the experience and treatment of serious illness, and is a fundamental existential issue for patients and their family members.^{138,150,162,177} Communication and decision-making occur against a backdrop of uncertainty through evolving prognosis linked to trial of treatments and progression of illness over time.¹³⁸ The centrality of uncertainty needs to be factored into all contacts and communication in health service consultations. The management of uncertainty has a powerful impact on patient and FCG experiences of illness and orientation to planning future care. For some patients, uncertainty can provide a justification for optimism and they can embrace the positive recovery orientation. For others, however, uncertainty can be profoundly unsettling.¹⁵⁰ The complex relationship between hope and uncertainty is dynamic.^128,162,178

A key message of the study findings is for staff to be aware of and responsive to patient and family experiences of uncertainty and to use this awareness as the basis for engaging in discussion of patient goals and values, rather than to evade the issue or seek to impose a formulaic agenda relating to ACP that derives from a narrow stratum of Western values centring on personal autonomy and decision-making. The findings also point to the importance of professional understanding of a patient’s informal network of support and awareness of the relational nature of decision-making.

Effective communication is pivotal to good care, and its centrality is reflected in the summary of our findings, ‘being known’ (termed with the Zulu word ‘sawubona’ in our key study outputs), presented in Figure 5. There is much evidence about what constitutes effective communication with patients about future illness progress and end of life and, specifically, in engaging people in an ‘end-of-life talk’.^179,180 However, there is scant literature on the additional attributes of such conversations with people when English is not their first language, when translation is required and when there may be disparities in conceptual understanding of health, illness and care provision. This study has highlighted the importance of professional awareness of these elements as an area for further research.

FIGURE 5

Enablers of ‘knowing’ ethnically diverse patients and families to optimise EOLC.

Models of care

UK health policy is based on the idea of the patient as an autonomous decision-maker. The realisation of individual choice in EOLC, especially regarding place of death and preferences for or against treatment, is regarded as a marker of quality care.^{10,158,161,181} However, recent research and associated critiques have cast doubt on the extent to which these policy assumptions represent the lived experience and concerns of many patients in the general population.^182–184 Instead, patients are understood to operate within a field of relational autonomy, where the decisions and choices are made with reference to, and mindful of the effects on, significant others within a social network.^{137,138,159,185–187} This collectivist model of decision-making is recognised to be particularly relevant to patients from culturally and ethnically diverse backgrounds.^{60,132,136,156–158,163} In these contexts, the locus of decision-making is often a joint responsibility, vested in the family group, and sometimes requiring concurrence from elders or respected non-family community or faith leaders, rather than the individual.

Degrees of individual family commitment to patient care very much affect their attitude and receptivity to accepting support from sources outside the family. In the Thinking Ahead study, varying levels of commitment appeared to be driven by profound personal attachment and close family ties, as well as by a sense of family, cultural and religious duty, and responsibility. This may include, for some individuals, a desire to accrue religious merit in line with their faith beliefs and a desire to avoid censure from the wider family or community. In some cases, adult children expressed concern about being criticised by their wider family or community if they were seen to fall short of expected commitment to care for their parents. This consequential stigma appears to be an important determinant of the family’s willingness to seek out and accept external help, especially regarding personal care and social support or care in a hospice or care home.

There was considerable variation between individuals and families in relation to closed or open awareness of a patient’s illness and prognosis and the extent of discussion and sharing of information between the patient, FCG and other relatives.^148,154,188 Preferences varied widely regarding what individuals wanted to know about illness and prognosis, even within the same family, generation and cultural or religious group.¹⁸⁹ The extent and composition of individual family networks also shaped how and with whom information was shared. There was also uncertainty about what different family members and the patient knew about the illness prognosis, who had talked to different professionals about this, and whether or not and how information had been relayed to others, within professional and informal networks. Some decisions were prolonged and distributed, seeming to require consensus among patient and family members, but the locus of decision-making could be configured differently across cases. This was largely determined by individual circumstances, of family composition; location; personality; religion; and cultural norms about family responsibility, engagement and authority. It was not the case that all patients had access to extended family support, and even when they did, some families struggled to provide this. Small, dense networks tended to be self-contained and seemed less likely to have established extensive contacts with health and social services.

This accords with Granovetter’s¹⁹⁰ thesis that ‘weak’ ties connecting individuals within a network to external persons and agencies are more likely to provide access to information and resources than a strongly embedded tie.¹⁹¹ Thus, although our data can be suggestive only, it may be that families that are insular, including those who have access to a GP who speaks their first language, and typically with elderly parents who do not want to accept outside help, are more likely to be disadvantaged in terms of the information and support they could effectively access. Conversely, within families that include a strong advocate, typically an adult child, who takes an active stance in relation to contacts with health and social services, and is willing and able to challenge these and follow through requests for support and information (e.g. through direct contact with different HCPs and services, who/which themselves are linked to wider points of contact and referral within the professional network), that advocate is an agent for extending the network of care around the patient.

Navigating a complex system of health care

Patients and families confronting the end of life enter an alien world, full of uncertainty and the need to engage with, usually, many different professionals and services.^138,150,177 May et al.¹⁹² and Mair and May¹⁹³ have drawn attention to the ‘burden of care’ imposed on patients and their families in the course of receiving treatment and services from a complex system of care. Several FCGs were exasperated by the demands imposed by an ongoing series of referrals and appointments for which they could identify no purpose or helpful outcome. Other participants described confusion, misunderstanding and lack of information during their experience of care. The difficulties posed by system complexity have been widely reported in the literature.¹⁹⁴ Stacey et al.¹⁴⁸ highlight the significance of institutional and organisational factors in shaping how resources such as time, expertise and resources are able to flow through complex systems of care. They identify the increased specialisation of professional roles and the consequent fragmentation of patient care as important factors contributing to the lack of effective communication and awareness of patient prognosis and future illness trajectories.

In the Thinking Ahead study, some patients and, more usually, their family members were skilled and effective mediators and advocates, with the ability and resources to find the information they needed. These skills developed with experience, sometimes in relation to the illness and death of more than one relative. In addition, some participants were, or had been, employed within the NHS in some capacity. However, the situation was very different for patients and FCGs who were unfamiliar with the system and lacked the confidence or fluency in English to communicate effectively with staff or achieve an understanding of their entitlements, especially if they lacked family support and translation.

A few patients described long-term relationships with specialist nurses, but the nurses, by their own admission, rarely engaged in discussion of planning of EOLC and did not, from the accounts available in the study, have a deep knowledge or understanding of the patients’ wider circumstances and concerns. Several HCP participants expressed surprise that they had been nominated to take part in the study because they had only brief or superficial contact with the patient. Few participants described a relationship with a HCP who could be described as occupying the role of ‘key worker’, a ‘go-to’ person who helped them navigate the complexities of care and whose sense of being available enabled the patient and FCGs to feel ‘safe’ within the system.¹⁹⁴ The experience of uncertainty, unfamiliarity and lack of information prompted a stance of vigilance on the part of relatives. A particular issue was the concern to ensure that cultural and religious practices relating to diet, hygiene, ritual purity and dignity should be observed during interactions with health and social care staff. For some families this was a motivation for refusing or limiting outside help. Hospitalisation was a particularly challenging experience, especially for older patients and those who were not fluent in speaking English. This is a known issue within the research and policy literature and should be on the agenda of things to address of all services.

Participants’ experiences of EOLC was shaped by the organisation of services relating to a patient’s illness and treatment needs and how these related to each other. This created subtle forms of disadvantage. Many different services and HCPs could be involved in each case and as the patient’s needs changed over time. However, some services, notably those for patients with MND, enabled a degree of continuity of care that was not available to most patients with cancer and other illnesses. In relation to other conditions, particularly cancer, patients tended to move through a process of referrals and to be seen by many different services and HCPs.¹⁴⁸ Participants described complex processes of referral to different types of test and treatment, and discontinuity of care and poor communication between staff from different services. It was not clear to individual patients, FCGs or HCPs what different individuals knew or had been told and had discussed with others. In addition, the boundaries of responsibility for discussing EOLCP were not clearly demarcated.

These are likely to be contributory factors underlying the reported vagueness of professional communication. In addition, patients and FCGs described their frustration at constantly having to tell and retell their story to the different HCPs they encountered during consultations and their sense of not ‘being known’ within the system, or the system ‘not knowing’ where patients were at any point in their process of care and referral. A great deal could hinge on chance encounters with individual HCPs, at any point in the treatment process, who proved to be particularly forthcoming and effective communicators. However, patients had no means of maintaining contact with such helpful individuals, who were likely to be replaced by other colleagues during subsequent appointments and consultations.

The centrality of effective care co-ordination in easing the experience of families negotiating their way through a complex system of health care was evident in the specialist MND nurse’s account of her role as a key worker accompanying patients from the moment of referral after diagnosis to the end of life. The study findings also highlighted the variability and ambiguity of the FCG’s role as advocate and decision-maker for the patient, especially when they also take on, or are co-opted into, the role of translators in medical consultations.

Being known: summary of the findings of workstreams 1 and 2

Family caregivers and HCPs often oriented to a positive recovery orientation unless and until the patient’s condition deteriorated to an extent that the imminence of death was clearly apparent. For some patients, not knowing their poor prognosis was a prerequisite for maintaining hope, optimism and engagement. Even in specialist community and hospice settings, professionals described not raising discussion of future goals of care or EOLCP until the illness was at an advanced stage. However, it could be hard for the minority of patients who did want to be fully informed about their illness to establish a clear prognosis. Moreover, there is a risk that, in leaving discussion of EOLC until patients are ‘ready’, or at least very close to death, professionals may never initiate a conversation that would have been of value, as some BFCGs subsequently reflected to be the case. Thus, there is a tension between the need to respect patient preferences, the negative consequences of revealing information about the future for which the patient and FCGs may not be ready, and the adverse consequences of families being confronted by crisis situations for which they are not prepared.

The clinical setting had a significant impact on the experience and content of thinking-ahead discussions. Families found being put under pressure to make decisions about whether or not their acutely ill relatives should undergo resuscitation following the stress of unscheduled hospital admission very difficult, and described professionals being brusque and lacking empathy. In contrast, when professionals in community settings, GPs or nurses took on board the need for the patient and family to engage in EOLCP and made time specifically for this purpose, participants reported a more positive experience.

A key issue is the very varying experiences of care reported within a complex system, including the type of illness, the services referred to, the relationship with individual professionals, experience of continuity or fragmentation of care, and the nature and availability of family support. In the absence of any clear demarcation of responsibility for staff to raise the topic of EOLCP, including in community and palliative care services and hospice, staff may be wary of raising the issue. This is possibly for good reason, since their judgement seems to be in accord with the preferences of many patients. However, some individuals were frustrated or felt disadvantaged by not being able to find a professional who would tell them what they wanted to know. The outcome may be (a possibly needless) anxiety or being misled into a false sense of security.

Many participants oriented to a collectivist or shared model of decision-making and responsibility regarding EOLC. This gave rise to tensions within some families and highlights the different agendas and values with which patients and family members may be operating. It contrasts with the foregrounding of individual autonomy and best interests embedded in consent and decision-making processes in health care, including EOLCP. Patients’ primary motivation in thinking ahead was often to protect their significant others by lessening the burden of care and responsibility resulting from their illness. These responses apply to all sections of the population, but have been highlighted in this study of patients and FCGs from ethnically diverse populations.

In contrast to ‘the logic of choice’, Mol¹³⁸ has proposed that health care be informed by ‘the logic of care’. This underlines the limitations and precarity of forward planning based on formulaic tasks and oversimplified decisions. Instead, professionals need to engage with uncertainty, their own as well as that of patients’, and to exercise curiosity about patient agendas and their perspectives and goals regarding knowledge of their prognosis and future care.¹³⁸ Decisions may be pragmatic, provisional, tentative and changing, as patients learn how they respond to situations that may have been beyond their imagination previously. The professional’s role is to accompany the family in the moment and support their readiness to deal with challenges and the decisions that need to be made as these arise.^72,195

However, the current configuration of services makes it difficult for patients to experience sufficient continuity and consistency of care and for professionals to develop sufficient knowledge of the person to generate the trust required for patients to feel confident in their support. ‘Being known’ and ‘being seen’ are key desires of patients and are enablers of good care, including personalisation of discussions about deterioration and planning care for the end of life. Our model of the determinants of ‘being known’, and the enablers of this, is derived from the study data (see Figure 5).

We have suggested that an allocated key worker could be a valuable resource in providing continuity and helping patients, especially those experiencing additional vulnerabilities and inequities; navigating the complexities of health care; and translating between language and culture. It turns out, unsurprisingly, that, at a macro level, good care looks largely the same, regardless of ethnicity: personalised, compassionate, holistic and safe. Understanding participants’ experiences and aspirations for care in the Thinking Ahead study provides a powerful lens through which to view and calibrate these components and the nuances of care for all patients confronting the end of life.

Stakeholders’ views on the barriers to and enablers of ethnically diverse patients, family caregivers and health-care professionals engaging in end-of-life care planning

Strengths and limitations of the study

The diversity of participants in the study is both a strength (in terms of including a wide range of experiences and perspectives) and a limitation. We have tried to include a broad ethnic diversity in the sample, but acknowledge that we could not include all groups; for example, Jewish, Gypsy, Roma and Irish Traveller communities were not included. Alongside ethnicity and faith, our sample included diversity in family and social circumstances, and in illness type, including two case studies of patients who lacked capacity; five participants were enabled to contribute through a professional interpreter and one by a bilingual researcher. The intention of the sampling frame was to explore the broadest diversity of experiences in the constraints of the project. However, this inevitably compromises, to some extent, the depth of understanding of specific experiences and issues related to discrete groupings of faith, ethnicity or culture. The study was pragmatically limited in its geographical recruitment to include two areas of the East Midlands. Each has somewhat differing ethnic demographics.

Most participants were recruited via three hospice services, which means their accounts may reflect experiences of a distinctive type of specialist care, available to a minority of terminally ill patients.¹⁵⁰ The majority of cases were identified and first approached by HCPs with whom the researchers had discussed the study eligibility criteria. However, several patients did not seem to consider themselves, nor were they considered by their nominated HCP, as being close to the end of life. Although these data may be outside the study’s original remit, they serve to highlight the variable and uncertain nature of prognosis in relation to terminal illness and the real-world challenges in integrating discussions about potential future deterioration alongside care of patients in stable and active treatment contexts.

Serial interviews by the same researcher in WS1 built rapport with patients and FCGs, allowing increasing depth of disclosure about participants’ experiences and feelings, and allowing a view, over a period of time, of the unfolding nature of their lived experience. We were also able to triangulate different perspectives from patients, FCGs and HCPs relating to experience and involvement in the same case. The richness of the data and the range of themes that emerged are strengths. The ‘messiness’ of life and the considerable labour in living with advanced illness is reflected in the accounts and in the logistical arrangements of undertaking the interviews. These are not sanitised accounts and give confidence in the veracity of the data, providing insight into the participants’ reality.

Many patients and families found it quite difficult to identify a HCP to nominate who they felt was to be closely involved in their care. All 11 of the nominated HCPs in WS1 were nurses. Consequently, the data do not encompass the perspectives of the wider clinical care team for an individual patient, including those of professionals, especially doctors, who may have a different approach to offering conversations about preparing for future deterioration and end of life. The need to stop data collection involving HCPs from March 2020 because of the COVID-19 pandemic meant that seven out of 18 cases did not contain HCP perspectives. Similarly, clinical notes were available for only 13 out of 18 patients. However, there were only three cases for which there was little or no additional information or representation of health-care perspectives outside the family.

The setting in which participant accounts were elicited is an important factor influencing their content, for example whether the interview was held in private or with one or more individuals, including an interpreter. Patient and FCG interviews were mostly held in the participant’s home. The use of translators enabled recruitment of patients who are normally excluded from research, which is a strength of the study. Issues with translation (in four case studies and one BFCG interview) and how these were handled are discussed in Chapter 2, Workstream 1: longitudinal patient case studies. However, as discussed previously in relation to clinical consultations, some degree of detail and nuanced meaning is inevitably lost in translation.

During case study interviews, most participants were asked if the experience of taking part in the research had influenced them in any way. In three cases, participants explicitly said that they had been prompted to reflect and to think a little more about what the future might entail (this included the idea of thinking ahead about end of life). In a couple of cases, participants’ reflections had led them to consideration of further practical help they might seek. At least two HCPs expressed possible changes in their practice, in particular related to knowing more about their patients and their potentially diverse needs. Some participants described interviews as being emotionally burdensome, causing anxiety, whereas others reported them as positive or even cathartic. Distress management of participants required careful consideration, particularly when using interpreters. Researchers also found interviews to be emotionally challenging. A pyramid of support needs for participants was in place, and preparatory and reflective resilience work, alongside regular support meetings, were provided for researchers.

Implications for health professional practice and training

Health professionals in the case studies and stakeholder workshops described approaches to the complexities and uncertainties involved in initiating EOLC discussions with patients and FCGs from ethnically diverse groups. HCPs have the advantage, but also the responsibility, of anticipation, from past experience and the literature, as to what may lie ahead for patients and the imperative as the expert guide to balance the present well-being of the patient and family with the potential regret in future. There was considerable reflection on the need to recognise and wait for when a patient may be ready for ‘end-of-life talk’ as their illness worsened. There was scant detail, however, on what the skills and attributes of this required and looked like in practice.

End-of-life care planning documentation is seen in policy as the predominant vehicle for enabling good EOLC, but this may be in tension with the preferences and values of patients. In particular, collectivist models or responsibility for care and decision-making is directly at odds with the commitment to patient autonomy and choice that underpins professional policy and practice. How HCPs assess and navigate the potential for harm in this policy context needs to be widely discussed and clarified. There is currently a risk that emphasising difference results in marginalising and stereotyping patients and overlooking both the heterogeneity of individuals within diverse groups and what is common to the experience of good EOLC for all patients, regardless of ethnicity. Good practice currently orients to encouraging patients to conform to UK policy and practice, so that diversity is seen to constitute a barrier, rather than a manifestation of alternative values and preferences. In addition, the predominant stance of patients was one of hopefulness, and for some it was against their religion to engage in planning for the end of their lives. Current policy aims for patients and FCGs to change their perspective, and reluctance to plan ahead is constituted as a problem. Within cultures oriented to a collectivist and relational approach to decision-making, or religious beliefs about the absolute nature of divine intervention, it is the assertion of autonomy that may be considered unsatisfactory.

The findings from the Thinking Ahead study are supported by those of previous research in reporting that many HCPs do not perceive engaging in EOLCP to be within the remit of their role. Among those who do, there is often a reluctance to raise the issue. Professionals wait for an indication of patient ‘readiness’ and tend to postpone discussion until the patient’s health has deteriorated to the point that death is clearly imminent. Our findings suggest that this approach is broadly aligned to the preferences of many patients. The desirability of otherwise routinely initiating such discussions pre-emptively, at a much earlier stage of a patient’s illness, may be debated. However, a significant minority of patients want to be informed and aware of their prognosis and are disadvantaged by the deferment of discussion about their options for future care and how this will affect the plans they make for the life they have remaining. A key task of HCPs involved in supporting patients through serious and terminal illness is to elicit patient preferences for future planning and respond appropriately.

Thus, professionals require high-level skills to navigate complex, sensitive communication and interpersonal relationships that foster appropriate discussions and planning for anticipated deterioration, when patients and FCGs wish to discuss this. In relation to ethnically diverse patients, this could be fostered by an integrated team approach that includes members with skills in language and cultural bridging or advocacy. Enabling patients to communicate confidently in their preferred language is an important means of reducing disadvantage. However, systemic support needs to be more effectively attuned to the values and sociocultural contexts of patients and families regardless of ethnicity, and to provide relational and service continuity to deliver sensitive and personalised care.

It is important to provide HCPs with the training they need to support skilled and appropriate discussion of EOLCP, including through translators. However, it is also important to acknowledge the systemic barriers to such discussions. These include the ways that health care is organised and staffed, particularly the fragmented nature of services, which block the transfer of information and co-ordination of care across a complex network. In addition, the intractable nature and pervasive influence of prognostic uncertainty and the influence of spiritual values on hope and value of life should be acknowledged as major barriers to the discussion of EOLC and as existential issues with which patients and professionals have to engage.

Implications for research

Our findings suggest prioritising the following research questions to benefit patients. Such research must include participants from ethnically diverse communities, including those who would want to contribute in a language other than English.

•

How can health professionals identify if/when a patient is ‘ready’ for anticipatory discussions of deterioration and dying?

•

How can discussions about uncertain recovery and the need for decisions about ceilings of care, especially resuscitation, be most effectively conducted in a crisis?

•

How can professionals recognise and respond both to the diversity of faith and cultural practices among different ethnic groups and to the heterogeneity of beliefs and preferences relating to EOLC between individuals from different ethnic groups?

•

What are the implications of an understanding of collectivist, relational and faith-led models of care and decision-making for current UK health policy and practice and their focus on individual choice, autonomy and anticipatory planning?

•

How can conversations be most effectively conducted when translation is required to enhance patient understanding?

Conclusion

The findings highlight the unique nature and experiences of every patient and family and the differences between individuals within the same family and ethnic group, as well as between ethnic and faith groupings. The overwhelming impression of these accounts is their diversity and real-world complexities, the ‘messiness’ and challenges of life when someone is progressively approaching death. For each individual patient and family, the specific details of how, when and if they want to engage in thinking about the future and develop anticipatory plans are singular and emergent. They are rooted in uncertainty and coloured by past experiences of the deaths of family and friends; by the intersectionality of ethnicity, educational attainment, socioeconomic status, available health care and other factors; by values and beliefs; by information and understanding; by trust in organisations and individual professionals; and by experiences of an often unco-ordinated and complex health-care system.

The data reveal many similarities in the experiences of EOLC and communication between patients and FCGs from ethnically diverse groups and those from the wider UK population, and in their responses to EOLC and communication. People from ethnically diverse backgrounds want compassionate care that is personal to them. This is no different from people from white British backgrounds, as is well described in the literature, and supports the use of physical, social, spiritual and psychological care frameworks used in palliative care to address the individual needs of patients and families. However, HCPs need additional skills to navigate complex, sensitive communication and enquire about spiritual values and aspects of people’s lives that may be unfamiliar. The data show that there are clearly distinctive aspects of the experiences and circumstances of patients and FCGs from ethnically diverse communities that compound their difficulties in achieving good care. Deficits in communication, especially involving translation, misunderstanding and lack of information result in patients falling through the cracks in services and suboptimal experiences of EOLC.

Personalising care requires HCPs to get to ‘know’ the person, and the development of confidence and skills in discussing EOLC with persons from ethnically diverse backgrounds is vital for enabling tailored responsive discussions that are of real value to patients and their families. Training resources, using stories based on our interviews, were seen as an effective way to support this.

Finally, the data reveal the rather marginal position of professional input in people’s lives. Most of our participants orientated their lives to getting on with the challenges of living in the present, rather than dwelling on the future: being positive, holding to the essence of faith and all things being in god’s hands, and fear and avoidance of future talk. A minority did seek to think and plan ahead and sought information and discussion to support them in this. Participants had difficulty in nominating a HCP who they considered had a significant role or any sense of accompaniment in their illness. HCPs came into their lives at times, but there was little sense of continuity of care or of professional support to accompany and think with them as their health deteriorated towards the end of life. Patient and FCG participants described an experience of care that was often fragmented and unco-ordinated. Workshop stakeholders proposed more effective sharing of information between multidisciplinary professional teams as a prerequisite to enabling more tailored and effective care. Effective communication and trusted individuals sharing information in an easier way was seen by stakeholders as important in providing better care.