Increasing the acceptability and rates of organ donation among minority ethnic groups: a programme of observational and evaluative research on Donation, Transplantation and Ethnicity (DonaTE)

Phase 1: systematic search and quality appraisal

Specific inclusion criteria for the review are shown in Box 2.

Systematic search: databases were searched during November and December 2010 (Box 3, name of host databases in bold). Relevant websites were also searched [i.e. Google (Google Inc., Mountain View, CA, USA), Department of Health website] and key journals searched by hand. Experts in the field were contacted regarding recently published or ongoing projects and grey literature.

It was not possible to obtain all relevant grey literature commissioned by NHSBT and undertaken by a range of private research companies. Although we attempted to get access to full reports of these reviews via senior staff at NHSBT, via the market research companies commissioned by NHSBT to conduct research and from the Central Office of Information, none of these organisations was prepared to share this information.

The search strategy was written in conjunction with an information specialist. Terms employed reflected previous and current phrasing associated with deceased donation (e.g. cadaveric and deceased donation included), and included a range of terms to identify the correct populations across the UK, USA and Canada. Specific ethnic categories were those attributed in the papers reviewed. Following the initial MEDLINE search, this was slightly modified to be applicable to other databases (see the medical subject heading search terms in Appendix 1).

The initial database and grey literature searches yielded a total of 2185 returns, reducing to 1461 following deduplication. These papers were initially screened for relevance based on the title and abstract, and those failing to meet the inclusion criteria were excluded. When it was not possible to determine relevance from the title or abstract, the full-text papers were retrieved and assessed. The main reasons for exclusion were that studies related only to children (students were included), were conducted in a country other than the UK, USA or Canada, focused on living donation/donors or on transplantation or treatment/follow-up rather than donor issues. Following relevance assessment 106 papers were retained (Figure 2).

FIGURE 2.

Flow diagram showing exclusion process for barriers review. 57

Phase 2: preliminary synthesis (grouping and clustering of studies)

Only five retained studies were published prior to 2000,33,34,38,41,53 reflecting the fairly recent growth in transplantation and deceased organ donation. Altogether, 15 studies were conducted in the USA,29–34,37,39,41,47–50,52,53 and these were primarily quantitative surveys to identify self-reported donation attitudes and behaviour among African Americans. The three Canadian studies were all qualitative and examined barriers relating to native, Chinese and Indo Canadians. 42–44 Seven of the eight UK studies were also qualitative (focus groups or semistructured interviews)35,36,38,40,46,51 and studied black African, black Caribbean and South Asian (Indian, Pakistani, Bangladeshi) ethnic groups, with just one quantitative survey in south London. 45

The papers focused on knowledge and attitudes, with a lack of attention to the effects of access to information or sources of registration, although such barriers have been identified as common in disadvantaged communities in the USA. 61

Phase 3: main synthesis and relationships

Phase 4: robustness of the synthesis

This synthesis is one of a small number of reviews that have integrated quantitative and qualitative research and thus bring together studies that differ in their design, measures and study populations. These differences limited precise comparisons, although the inclusion of both types of data enabled a more detailed understanding of beliefs and practices.

Retained studies were, however, mainly ‘thin’ in interpretive content, thus restricting the possibility of more complete explanatory models. This may partly reflect the particular challenges of research in this area, including the cultural taboos and fears surrounding death and notions of organ donation, as well as people not having previously engaged in discussion of this topic.

Despite acknowledged limitations of the research, there was evidence of commonalities in attitudes and beliefs across race/ethnic groups, particularly the similarities between African Americans beliefs and those of the black population in the UK, as well as evidence of some variations between ethnic groups.

Participant recruitment

Our research protocol identified the need to employ a recruitment company with experience of engaging people from minority ethnic groups in academic research, given the demands of recruitment and a limited timescale. We identified a company that had a network of local fieldworkers and was willing to undertake only the recruitment stage, with our team then conducting the groups and undertaking the analysis.

The recruitment strategy (Table 2) aimed to allow beliefs and attitudes to be examined both between and within ethnic groups. Separate focus groups were, therefore, conducted with people of Indian, Pakistani, Bangladeshi, Caribbean and Nigerian ethnic origin (the Nigerian and Nigerian British population was selected representing the largest proportion of the black African population in the UK). 65 These ethnic groups were also divided into younger (18–40 years) and older age groups (≥ 41 years). For the South Asian groups, separate groups were held for men and women to reflect cultural traditions and promote discussion among participants.

The recruitment company was tasked with recruiting participants in line with our stratified sampling and to organise suitable locations for holding the discussions. Initially we hoped to recruit some groups outside the London area; however, the logistics of recruitment resulted in our recruiting groups from the eight London boroughs with the greatest ethnic minority populations. With the exception of the African and Caribbean groups, recruitment of a particular minority group took place in single boroughs, where a significant proportion of the population was made up of the ethnic minority groups under study as identified by Ethnic Group Projections for 2010. 65 Table 2 shows which boroughs each ethnic group were recruited from.

Agency fieldworkers were of the same ethnicity, faith and gender as the population to be recruited, and often resided within in the recruitment borough, thus benefiting from local knowledge, recruited the participants. Recruitment occurred on a door-to-door basis, in the street, at places of worship and community organisation, and through snowballing.

The recruitment company was supplied with a brief information sheet for potential respondents, which outlined the purpose of the research, what to expect if participating, assurances of confidentiality and the promise of expenses (£25). Potential participants were invited to contact the researcher if they wished to discuss any aspect of the research before making up their mind to join the study.

The aim was to recruit 10 participants for each focus group to ensure that approximately eight participants attended. Altogether, just over two-thirds of those recruited attended the focus groups, with attrition between recruitment and attending the event being greater among Caribbean and Nigerian populations (Table 3).

Throughout the recruitment process the research team was in close contact with the company to provide feedback and to ensure that the high level of attendance was achieved, with participants drawn from our designated minority ethnic groups and age groups. The split between employing a specialist company for recruitment and facilitating the focus groups ourselves seemed to work well.

Non-English-language groups

Recognising that English was likely to be a second or third language for some members of the older generation, we employed bilingual fieldworkers with experience of conducting qualitative research to facilitate the discussion with older Pakistani groups in Urdu, older Bangladeshi groups in Sylheti, older Indian Hindu groups in Gujarati and Indian Sikh groups in Punjabi.

A training session was held to brief the fieldworkers and familiarise them with the topic guide. The DonaTE team also developed an annotated version of the focus group topic guide that explained the rationale for the questions being asked in each section and gave an indicative list of potential topics to probe further (see Appendix 4). The day also included training on the conduct and facilitation of focus groups. These sessions provided ‘top tips’ for effective facilitation and covered issues such as how to break the ice, managing dominant participants and dealing with the ‘group expert’. This phase of training also involved discussion of what areas to prompt and probe and how to do this effectively. The training day also allowed an opportunity to consider issues in translating the topic across languages when direct translations of words associated with organ donation are not available or the concepts around donation are unfamiliar. As the final stage of training, the fieldworkers were observed in facilitating practice focus groups involving other members of the group.

Topic guide and vignettes

The topic guide covered individual knowledge and attitude towards organ donation, wider social benefits, and religious and moral issues regarding donation (see Appendix 4).

Piloting the topic guide confirmed that participants knew little about deceased donation and had little direct experience of the situations in which organs are requested to be donated or required for transplantation. When initially asked about organ donation, they therefore mainly talked positively about the benefits of transplantation and any conversation about donation was mainly situated within narratives of doing a good deed and saving the life of another, reflecting prevailing narratives of organ donation/transplantation in the media.

In an attempt to prompt more detailed discussion, we revised the topic guide and included factual information about donation to introduce each section, including information on the disparities in donation and consent between the general population and minority groups (see Appendix 4). We also developed vignettes that required participants to consider how they might behave in a particular donation situation and how they would advise a friend faced with a similar situation (Figure 3).

FIGURE 3.

Vignette scenarios relating to registering as a donor. Reproduced with permission from Morgan M, Deedat S, Kenten C. ‘Nudging’ registration as an organ donor: Implications of changes in choice contexts for socio-cultural groups. Curr Sociol 2015:63:714–28. 66

The vignettes were informed by the literature and campaigns to promote organ donation. We also sought feedback from our lay advisory group to ensure that the situations depicted seemed plausible and comprehensible. Those attending the meeting felt that the topic guide and vignettes were appropriate and a few minor comments were incorporated. The decision to stratify by gender for the younger Indian, Bangladeshi and Pakistani groups was not regarded as necessary but at that stage it was too late to alter the recruitment strategy to reflect this sentiment.

Two vignettes are considered in this chapter and relate directly to registering as an organ donor. A third vignette focused on issues of family consent to donation. This was less successful in promoting discussion as participants were becoming tired and lacked understanding of the donation process within ICUs that required a new set of information. Moreover, their over-riding concern and almost sole focus of discussion was about whether or not you know if someone is really dead or may come back to life.

Conducting the focus groups

A total of 22 focus groups were held over 3 months (March–May 2010). Two researchers (CK and SD) facilitated 14 groups. The other eight community language groups were facilitated by bilingual fieldworkers, with a researcher also present to ensure that the groups ran smoothly and clarify any questions that the group might have about organ donation.

The focus groups were conducted in community spaces that were local to and often known by the participants and were also close to public transport. The groups were mainly held in the late afternoon/evenings or weekends, as this was the most convenient times for the participants. Each lasted 60–90 minutes. Participants received £25 in cash to cover expenses that they incurred attending the group. It was clear that, for some, this payment was an incentive to participation.

At the start of the group, the researcher provided an overview of the study and explained how the focus group would proceed. Written consent was obtained and a short questionnaire completed to provide summary sociodemographic information (see Appendix 5).

Data analysis

The bilingual fieldworkers interpreted and transcribed their focus group discussions into English. Other audio-recordings were transcribed verbatim by a professional transcriber and checked for precision by CK and SD. Listening to the transcripts and reading them for accuracy of transcription aided the identification of potentially relevant analytical concepts and patterns of response across groups.

To assist the analytical process, each transcript was imported into the qualitative software package NVivo (version 9). Analysis and interpretation occurred concurrently with data collection. Initial line by line coding was undertaken, with the responses grouped into categories that related to the five themes identified in the systematic review of barriers to donation. However, the analysis was not intended to be limited to these themes, with emphasis also given to the variations within and between groups, as well as to emerging thematic categories and interpretations.

Characteristics of participants

Table 4 summarises the characteristics of participants (see Appendix 6). Of the participants, students, retirees and those caring for family accounted for 29%. In general, the sample was biased towards the lower occupational and socioeconomic groups, with those employed (44%) mainly being in low-skilled or semi-skilled occupations, such as taxi drivers, security guards, and teaching and kitchen assistants. An exception was the group of younger Indian Hindu women, several of whom were in professional occupations, including an engineer and a general practitioner (GP).

The participants included a high proportion of recent migrants; of the 202 who stated their place of birth, 79% were born outside the UK. These characteristics of relatively low socioeconomic status and a high proportion of overseas born people are fairly typical of areas with a high multiethnic population.

Knowledge about organ donation

A small number of people attending each focus group knew someone who was waiting for a transplant or who had received a transplant; however, few people were sure they had definitely joined the ODR. An exception was four highly educated Hindu women who had signed up to the ODR following participation in a cord blood donation project at a London hospital that led to their knowledge and interest in donation. Another rather larger group were unsure if they had checked the relevant box agreeing to be registered as an organ donor when renewing their passport, applying for a driving licence or applying for a Boots Advantage Card.

With few exceptions, individuals tended to reflect positively and with some enthusiasm about the abstract notions of donation and transplantation, with responses frequently situated within narratives of doing a good deed and saving the life of another. However, when asked to discuss the first vignette and to share their opinions or what they knew about organ donation, their responses were limited:

I don’t have knowledge on it to be honest. I’ve never looked it up to be honest, nothing really – I don’t know anything about it, all I know is that they take things out of your body.

Bengali Muslim woman, 18–40 years

When asked about how they would go about registering on the ODR, most participants’ responses drew on what Schutz64 terms ‘cook book knowledge’. This refers to ‘recipes’ for dealing with routine matters encountered in daily life that can be drawn on by individuals when faced with new or unfamiliar situations. For example, given their lack of awareness and knowledge of how and where to register as a potential organ donor, participants suggested logical places where it would be assumed individuals should or could possibly register. These tended to be local hospitals or GP surgeries. As registration was perceived to be a medical matter and an important decision, registering in a ‘health space’ was regarded as appropriate and to have the advantage of the opportunity to ask questions about an unfamiliar topic:

And I think, you know, with the GP you’ve been, hopefully, with that GP for a long time, they know all your history . . . trust is built over time, you know, it’s, you go with all your personal issues, and so yes the trust is definitely there.’

Indian Sikh woman, 18–40 years

Awareness of the shortage of organs was also limited. Across the groups, a number of participants (both older and younger) were of the opinion that the NHS holds a store of organs within hospitals that would be available for transplant when required. Within this context there was little imperative to register as an organ donor as they perceived there to be abundant stores of organs available.

Oh, sorry, I don’t know much about it, but do they freeze these until someone needs them?

Nigerian Christian woman, 18–40 years

. . . my organs may be removed after my death and put in a freezer box for a while. Doctors may use these organs from the freezer box later . . .

Bengali Muslim man, ≥ 41 years

If they don’t find a person [suitable recipient], they will put it [donated organ], keep it in a freezer for a long time.

Bengali Muslim man, 18–40 years

Explaining low knowledge

The national ‘Prove it’ campaign focused on minority ethnic groups in 2010 and coincided with the focus group study (see Table 1). However, few respondents were aware of this or of any other campaign. We believe this lack of awareness is because organ donation has limited ‘topical relevance’ for the majority of participants. Schutz64 described the notion of ‘topical relevance’ as something that is imbued with meaning for an individual because it stands out for them (p. 125). Schutz described individuals as structuring their knowledge into zones of relevance, which decrease in degrees of clarity and precision as they move outwards from areas of personal concern. Within this context, the potential exists for information that is not perceived by an individual as ‘topically relevant’, in terms of being of close personal concern and relating to their existing ‘stock of knowledge’, to ‘pass by unnoticed’ rather than being actively rejected. This lack of personal relevance was often reflected in participants’ accounts:

. . . you see leaflets and cards in everywhere like surgeries, and you don’t really give it that much importance . . .

Bengali Muslim woman, 18–40 years

Yes exactly the thought [of registering as a donor] has been there when you’ve read a poster or something but as soon as you walk away from that poster . . . as soon it comes in it goes out . . .

Indian Sikh man, 18–40 years

Analysis of participants’ accounts identified several reasons why organ donation campaigns lacked topical relevance for them.

Stages on the donation pathway

The majority of focus group participants were at the early ‘pre-contemplation’ stage of Prochaska and Velicer’s67 transtheoretical model of behaviour change (TTM), with little knowledge of donation or its relationship with transplantation (Figure 4). In contrast, the small group of Indian Hindu women had achieved an ‘action’ stage and joined the ODR, having being prompted by cord blood donation and had discussed this with their spouse, thus providing further maintenance of their wishes. Another group appeared to be at a contemplation stage, but either had some unresolved worries or, as in the case of many younger Indian Hindu and Sikh participants, regarded other things in their lives as having greater priority. However, it appeared that this group would be likely to take action in response to a ‘prompt’, such as the availability of easy and immediate access to registration. In addition, some older members of the Asian community felt that trying to register online could be too difficult given a lack of familiarity with this and their limited English.

FIGURE 4.

Prochaska and Velicer’s five stages of behaviour change. 67

Stages of change adapted for organ donation

Pre-contemplation: have not considered registering as a donor.

Contemplation: considered organ donation but not taken steps towards registration or addressing unresolved issues.

Preparation: able to recognise the benefits of registration and have taken action to find out more.

Action: expressed their wishes about donation by joining the ODR or discussed wishes with immediate family/kin.

Maintenance: occasionally reinforce statement of wishes to family members/kin.

Explaining negative attitudes

The second vignette was designed to elicit views and attitudes towards organ donation by presenting a scenario that prompted discussion regarding the pros and cons of registering as an organ donor (see Figure 3). For those with little knowledge who were thus at a pre-contemplation stage, these were not attitudes that currently influenced their behaviour but had the potential to do so in the future.

It was notable that participants were generally positive about the benefits that organ donation has to deliver to others. However, there was a distinct variation across the groups in salient uncertainties viewed as barriers to joining the ODR. Thematic coding of the data revealed that the most pertinent uncertainties revolved around issues pertaining to religious permissibility (most commonly among Muslim participants) and issues of trust among Caribbean participants.

Trust and mistrust

Of particular salience among Afro-Caribbean participants were issues relating to trust. This partly related to the fair allocation of organs, with a common perception that white people would be privileged in terms of allocation:

I think the reason why the Afro-Caribbean not giving, because they don’t think that – it is fear because they think that even if we all come together and say we’re willing to give, we’re going to be the last one to get, right . . . Suppose 50 people have something wrong with their inside and they are 50 white, OK are they going to choose half of this and half of that? I think that they are going to go with the white one first and then whatever left, then the black one get it, then the Afro-Caribbean get it. That is what I think anyway, that’s the reason why they don’t go for it.

Caribbean Christian woman, ≥ 41 years

. . . I mean I am starting to turn the corner, because I have seen so many friends and family, you know, on dialysis and needing organs. But I mean, as Maureen just said, it does raise a kind of question mark in your mind, as to how it is allocated.

Caribbean Christian woman, ≥ 41 years

Additionally, Afro-Caribbean participants commonly vocalised a concern that donated organs may be used ‘without consent for other purposes like medical research’. In addition, they questioned whether doctors ‘will do all they can to save you’ or ‘just cut my body and take stuff’ (Caribbean Christian man, ≥ 41 years).

South Asian and Nigerian participants, in contrast, described generally trusting the UK medical system and having few worries about misuse of organs or unethical allocation practices in the NHS compared with systems back home.

. . . here [in the UK] I’d be a donor because the system is protected to a reasonable level. But If I go to Africa I will never be a donor, no way, because I don’t trust the system . . . In the UK I think I have the confidence in how it’s [organ retrieval, treatment of the body] being done especially in the NHS, I’m not saying its 100% but at least to a reasonable level, at least I’ll give them 85–90% if not more than that.

Nigerian Christian man, 18–40 years

Other beliefs and concerns

Participants identified a number of other beliefs and concerns that could enter into decisions regarding donation. For Muslims, a key issue was the need to be buried within 24 hours, which is recognised to present problems for donation and coroners’ inquests. General worries also included a dislike of the body being cut, as well as the notion that the dead body can feel pain and that this can even be more intensely felt by a dead person. A few people also voiced concerns about disfigurement and not looking normal if the body is displayed in an open casket, although most realised that donors would be presented so that they did not look different. Other worries related to general notions of tempting fate by joining the ODR.

There were, therefore, often a variety of influences that shaped views of donation and willingness to register as a donor, although for the small number of Muslims who believed donation was not permissible, this formed the single key concern.

Relevance and quality assessment

Following initial deduplication, 557 articles remained. This was followed by a review of relevance that was conducted by two researchers based on title and abstract, and, if necessary, review of the full article. This process led to the retention of 18 papers for synthesis (Figure 5). 29,69–85 These papers were reviewed for quality as a measure of robustness rather than for purposes of exclusion. Quality assessment was based on the Effective Public Health Practice Project tool that is applicable to randomised controlled trials, quasi-experimental studies and uncontrolled studies, and has established content and construct validity. 87,88

FIGURE 5.

Flow diagram showing exclusions for interventions review. Reproduced from Deedat S, Kenten C, Morgan M. What are effective approaches to increasing rates of organ donor registration among ethnic minority populations: a systematic review. BMJ Open 2013;3:e003453 under the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0). 86

Study characteristics

The 18 included studies29,69–85 are listed in Appendix 8 with their main characteristics summarised in Appendix 8, Table 17. These studies were all conducted between 1993 and 2012 and evaluated the effects of either an ethnically targeted mass media campaign or community-based education. All but one study85 was undertaken with ethnic minority populations in the USA, mainly African Americans or Hispanics. Most interventions published before 2009 adopted a pre- and post-evaluation design, whereas studies published since 2009 mainly employed a cluster randomised design. Outcomes assessed were verified registration, self-reported registration status or knowledge about organ donation (see Appendix 8, Table 17).

A detailed account of the findings is available86 and is briefly summarised here.

Educational interventions

A total of 11 papers reported educational interventions. 29,69,70,72–76,81,82,85 These interventions differed in their design and whether or not the outcome was verified enrolment on the ODR, changing knowledge or a shift towards greater readiness. However, shared characteristics of effective interventions were having a strong interpersonal element; being delivered by a range of individuals including recipient families, other trained lay individuals, or organ transplant and procurement staff; and being delivered in familiar settings, such as in a local church or a hairdresser salon, at community events or on a student campus.

The three main types of outcome evaluated in education interventions are considered below.

Mass media interventions

The seven media interventions attempted to detect changes in willingness or intention to join a donor register or sign a donor card. 71,77–80,82,83

Interventions that solely utilised mass media reported no significant change in intention or willingness to register an organ donor, despite focusing on a single ethnic group and undertaking formative research in the target population.

In contrast, a high level of success was achieved by one study78 that innovatively combined mass media techniques with interpersonal communication. This was conducted at vehicle licensing office in the USA, a setting in which individuals are able to join a state donor registry. To determine the effect of each component of the intervention on registrations, the campaign occurred in three stages, each lasting 3 months. The interpersonal component of the intervention was identified as accounting for the greatest increase in registrants, increasing from 444 in the year prior to 5588 in the post-campaign year. Overall, the combined effect of the intervention components increased donation by 700%.

Study sites

The study was based in the London region and the Midlands, with these regions selected because they have the greatest density of ethnic minorities in the UK. The study sites selected within these regions were ones with the larger ICU, where organ donation was expected to be a more frequent occurrence.

All study hospitals were teaching hospitals with on average 42 designated ICU beds (range 36–56). All the sites had general ICUs that cover a range of diagnoses (e.g. stroke, heart attack, head injuries, accidents) as well as a range of specialist ICUs; four sites had specialist cardiothoracic units and two had specialist neurological units for conditions such as stroke and brain haemorrhage, with other sites catering for neurological patients on general ICUs.

Data collection

This involved a mix of non-participant observation, semistructured interviews and informal discussions at five study sites.

The researchers observed the activities of the units in short time slots (approximately 30 minutes) by ‘hanging around’ nurses stations, which were usually positioned with a view to covering most of the unit, or watching from a short distance the work of one or two of the unit staff at the bedside. This assisted the researchers in becoming familiar with key issues and activities of the ICU.

The researchers also ‘shadowed’ SNODs over several hours when a potential donor was identified on the unit, and observed four requests for family consent to organ donation with the family’s permission. These family approaches were observed at three of the study sites. In just one instance, the approach for consent to donation was observed for a family from an ethnic minority group. This reflected the relatively small number of families approached for consent over a 3-week fieldwork period and particularly the small number of patients from minority ethnic groups on the ICUs. This is not surprising, given that only 237 eligible ethnic minority donors were approached about donation in the UK 2012/13. 3

Field notes were taken during observations or shortly after and contributed to the overall data set.

A series of topic guides were developed and covered five core areas for each group of staff, with additional questions applicable to a particular staff group. The core areas covered were:

• broad overview of work and responsibilities

• involvement in the organ donation process

• communication with relatives regarding futility/withdrawal of treatment, end-of-life care and approach to organ donation

• involvement with families of minority ethnic groups

• views of the roles/practices of the SNOD/CLOD.

All participants were provided with an information sheet and gave written consent prior to interview. Interviews ranged from about 10 minutes to 2 hours and were tape recorded with permission. Interviews were conducted on ICUs, in offices or a side room, or occasionally in the hospital cafe.

We originally envisaged undertaking 40 semistructured interviews with hospital staff. In practice, we conducted 107 interviews, although interviews with bedside nurses were mainly quite short. This increased number of interviews partly arose from the addition of an extra study site; however, it also reflected a shift in the balance of data collection, with a smaller emphasis given to the observational element, as the number of potential donors was quite small during the 3 weeks spent at each study site. We also increased the number of chaplains interviewed, as they appeared to be an under-researched group with potential for an increased contribution in this area (see Study 1: theme 3 – hospital chaplains and end-of-life care) (Table 7).

Analysis

Audio-recorded interviews were transcribed in full and transcript data were managed using NVivo version 9 software. Full transcripts were read and coded for themes and subthemes. These related to identifying potential donors, providing support to families and managing visitors with particular reference to minority ethnic groups, treatment withdrawal and the donation discussion, and the roles and relationships of different groups of staff in relation to the organ donation process.

Many issues identified by staff, such as difficulties that sometimes occur for families in understanding that their relative is dead and agreeing to withdrawal of treatment, occurred among all ethnic groups. However, the next sections focus on three themes that emerged as being of particular potential significance for family consent to donation among minority ethnic groups:

• theme 1: attitudes and practices of ICU staff in providing end-of-life care to minority ethnic groups

• theme 2: SNODs’ role and participation in consent discussions

• theme 3: hospital chaplains and end-of-life care.

Abstract

This theme focuses on the attitudes and practices of individual ICU staff (bedside nurses, SNODs and clinicians) in providing end-of-life care for minority ethnic groups. It focuses particularly on issues identified by staff as problematic, particularly cross-cultural communication, managing large families and visitors, responding to faith/cultural requests, the use of chaplaincy services and overcoming language difficulties. Some variations in the way practices addressed these issues were identified between sites, reflecting differences in facilities, priorities and individual preferences of the clinicians. It was also noted that some members of all staff groups acknowledged uncertainties about approaching families from different cultural groups, with junior nurses at each of the study sites most often describing feelings of hesitancy in approaching a different cultural group. They were also particularly positive about training to enhance their confidence and skills in this area.

Aims

This theme describes the practices employed in ICUs to meet faith or cultural needs and to manage large families and numbers of visitors in terms of both access to the bed space and involvement in discussions of treatment withdrawal. It also pays particular attention to bedside nurses’ perceptions of their own confidence and skills in communicating with different cultural groups.

Background

The successful redesign of donation services in most countries, including Spain and the USA, has involved the appointment of staff with specific responsibilities for organ donation, including discussing donation choices with families. In the UK, a workforce of SNODs were appointed following recommendation 9 made by the ODT:4

The current network of Donation and Transplant Co-ordinators (DTC’s) should be expanded and strengthened through central employment by a UK-wide Organ Donation Organisation. Additional co-ordinators, embedded within critical care areas, should be employed to ensure a comprehensive, highly skilled, specialised and robust service.

At the time of fieldwork, 250 SNODs were employed by NHSBT (the national organisation with responsibility for blood and organ donation). SNODs are mainly drawn from senior intensive care nurses and are embedded within hospital trusts, and are on-call for other ICUs in the same region. SNODs are regarded as providing ‘a comprehensive, highly skilled, specialist and robust service’, and have responsibilities that extend beyond those of the previous donor transplant co-ordinators. Their role involves overseeing the donation process, from identification of a potential donor to organising and managing donation, including offering organs to transplant centres and co-ordinating the retrieval process. They also provide early and continuing contact with families, collaborate in donation discussions and conduct the Potential Donor Audit (PDA) that records the number of patients who could be solid organ donors and identifies reasons for non-referral and non-consent.

The role of the SNOD in the UK is supported by CLODs, who are clinicians (generally intensivists) with a part-time responsibility as ‘champion’ for organ donation within a hospital trust. This involves an educational and facilitative role to achieve the goal of normalising organ donation so that it becomes usual practice within the trust.

Although SNODs and CLODs have special responsibility for organ donation, it is recognised that the collective and co-ordinated behaviour of the wider multidisciplinary team involved in caring for patients and families on ICUs, particularly the bedside nurses, is also of importance in promoting a culture of organ donation and ensuring appropriate communication and responses to the needs of different patient and family groups.

Methods

The analysis of practices undertaken in ICUs and responses to minority ethnic groups is based on interview and observational data as previously described (see Study 1: interview and observational study, Overall aims and methods).

Findings

This section discusses professional practices in supporting families of different ethnic, cultural and faith backgrounds and focuses on the themes of accommodating faith-based needs and requests, overcoming communication barriers and managing families at the bedside. These findings often provided examples of how staff met the needs of diverse patients and families despite having little formalised cultural competency training.

Conclusions

1. There is a lack of formal training in what is termed ‘cultural competency’, with individuals mainly relying on their ‘on-the-job’ experience to respond to people’s faith and culture in the provision of end-of-life care. However, junior nurses often had less experience, sometimes leading to what has been described as a ‘disabling hesitancy’ and inertia in their practice, particularly in situations where families are experiencing considerable emotional distress. 89 They thus welcomed training to increase their confidence in supporting families from minority ethnic groups and reduce stereotypical views.

2. Key aspects of care that were often regarded as problematic, and particularly in relation to minority ethnic groups, were managing large families and visitors in terms both of patient access and communication by clinicians, responding to language barriers and accommodating faith-based practices.

3. Variations in practices in terms of responding to the needs of ethnic/faith groups were often influenced by the hospital facilities and capacity as well as by established organisational practice and cost constraints. For example, the use of interpreters varied, with some hospitals more likely to use professional services, whereas some rely on staff first because of availability or cost, and others may use families because there is established precedent to do so. Similarly, being able to accommodate large family groups is influenced by the availability of side rooms off the main ICU ward and the size of family rooms. There were also differences between clinicians in their preferred practices, including their ways of managing large families, with some preferring to include a large family group so that all received information directly while others preferred a small group of close relatives.

4. It was common to offer chaplaincy services for withdrawal of treatment and formed an important example of how hospital staff aimed to meet the needs of people from different faith backgrounds. However, when feasible, staff also aimed to respond to particular requests that might be made by any families, reflecting the importance of achieving individualised care. Examples included families wishing to pray around the bed and requests for pieces of jewellery or other mementoes to be placed alongside the dying patient.

Abstract

The establishment of SNODs was a key aspect of service redesign that was introduced following the ODT report. Although SNODs receive specialist training in communication skills and aim to provide long contact with families, there is evidence that a significant number of donation requests continue to be undertaken by the clinician alone. This theme, therefore, examined the reasons for the lack of full implementation of the SNODs’ role and focused on both clinicians’ views and responses to the introduction of SNODs whose remit partially overlaps with the medical jurisdiction and the strategies employed by SNODs to embed their role within ICUs. This identified considerable variation in the degree of acceptance of SNODs and examines the varying personal, clinical and organisational factors that have led to the slow adoption of early referral to SNODs of potential donors and their participation in collaborative consent discussions.

Aims

To examine the SNODs’ role in relation to families including the constraints on early referral of a potential donor to SNODs and their opportunity to collaborate in donation discussions with families.

Background

Consultants were traditionally responsible for assessing patients’ suitability for donation and undertaking the donation discussion with families. However, it is now expected that the SNOD will participate in these activities. This involves being informed when brain stem testing for neurological death of a patient is planned or the futility of further treatment has been discussed with a patient’s next of kin. 4 The SNOD then checks the patients’ clinical suitability for donation and whether or not a patient is registered as a donor by consulting the ODR to which only the SNOD has access. Second, discussion of the possibility of organ donation with bereaved families was recommended by the ODT as ideally involving ‘collaboration’ between the consultant and SNOD, with this subsequently endorsed by the National Institute for Health and Care Excellence (NICE) guidelines, with the expectation that both the SNOD and the consultant would contribute and be present throughout the donation discussion. 89 Collaborative requesting has also been affirmed in local donation policies at trust level.

The routinely collected data support the positive effects of the involvement of SNODs, indicating that the consent rate is higher when a SNOD has been involved in a collaborative donation discussion with families than when this discussion is undertaken by the consultant alone. For example, in 2012/13, the consent rate was 65.8% when a SNOD was involved, compared with 36.7% when there was no SNOD involvement. 3 The difference was greatest for DCD donation. For example, in 2012/13, SNODs were involved in 71.3% of approaches, with this made up of 78.9% of DBD discussions and 66.7% of DCD consent discussions, reflecting a slight increase in involvement from earlier years. 3 However, as these figures indicate, a significant proportion of donation discussions continue to be undertaken by the clinician alone, although involvement of SNODs increased from 59.6% in 2010/11 to 71.3% in 2012/13.

The role of the SNOD impinges on the traditional medical jurisdiction, at least at its boundaries, and requires clinicians’ acceptance of the shift in responsibilities. However, consultants’ responses to this have not previously been explored.

Methods

This theme draws on the interview and observational data described in Study 1: theme 3 – hospital chaplains and end-of-life care.

Findings

The interviews indicated that all consultants welcomed the SNODs’ involvement in the administrative and nursing aspects of organ donation, including their role in organising the retrieval and offering process. However, they varied in their acceptance of those aspects of the SNODs’ role at the boundaries of their traditional medical jurisdiction. Views on this varied both within and between ICUs. This was reflected in differences in clinicians’ willingness to notify the SNOD of potential donors and to allow SNODs to participate in a collaborative donation discussion.

We distinguished three groups of consultants in terms of their response to SNODs and willingness to refer potential donors to the SNOD and engage in a collaborative consent discussion.

Explaining variations in clinicians’ responses

Each of the three types of responses occurred at each site and were also observed to occur within each ICU. More generally our analysis suggests that variations in consultants’ views and practices were influenced by a mix of personal, clinical and contextual factors, with these factors often interacting to produce a slow pace of change in what was a relatively infrequent event for individual consultants.

Responses by specialist nurses for organ donation

The SNODs were aware that consultants had the power to limit their claims to practice and of the varying responses of individual consultants. They also recognised that given the considerable autonomy of consultants in the NHS there was often little opportunity to achieve full engagement in their role through formal channels. SNODs, therefore, often relied on informal strategies to shift boundaries in their favour and as a way of ‘getting things accomplished’.

The informal strategies employed by SNODs included a range of non-verbal and positioning approaches to make their presence visible to ICU staff; this involved walking through the unit and talking to the bedside nurses and the clinical staff to make their presence felt and increase their opportunities for inclusion. They also valued having desk space on or close to the ICU to increase their visibility and accessibility, and sometimes used visibility strategically in situations when there was a potential donor on the ICU but they knew that the particular consultant might not automatically include a SNOD. In this situation:

we’ll be more visible. And we will have more discussions with them, just to make them aware that we actually here.

SNOD, site 5

The SNODs also often attended handovers as a way of enhancing their visibility and knowing if any patients were likely to be brain stem tested and whether or not organ donation would be considered.

The SNODs’ visibility and accessibility and their contact with other ICU staff sometimes led to ‘strategic alliances’, which were particularly valuable when ICU consultants were reluctant to refer a patient or to consider organ donation. In these circumstances a member of ICU staff, such as a bedside nurse or the nurse in charge, might informally inform the SNOD about a particular patient and the SNOD could then investigate further and talk to the consultant about the patient.

Negotiated approaches by SNODs sometimes benefited from support of the CLOD, who could use their peer-to-peer negotiations with their clinical colleagues to endorse the value of the SNOD and provide them with an entry point. For example, a CLOD described the value of ‘opportunistic chats’ with his consultant colleagues and explained that it was then up to the SNOD to take this forward and make further progress.

Conclusions

1. Implementation of the SNODs’ full role in relation to families is only gradually being achieved. This reflects a situation in which acceptance of the SNODs’ activity in areas that were traditionally within the medical jurisdiction varies between individual consultants with the lack of a fully ‘collective acceptance’ that is ICU or trust wide. There are also marked regional variations. Figures for 2012/13 indicated that collaborative involvement of SNODs in donation discussions varied from a high of 80.4% in South Central (London area) to a low of 40.5% in the South West. 3

2. None of the trust sites reflected a linear development of the SNODs’ involvement for the ICU as a whole. Instead clinicians categorised as ‘acceptors’, ’intermediate’ and ‘resistors’ appeared to occur concurrently on an ICU. This reflects a more general finding of the varying speed of uptake of innovations, which is influenced by the interests of consultants and other health professionals interacting with the organisational and cultural context. 93

3. Factors identified as contributing to lack of acceptance of the SNODs’ full involvement with families and their collaboration in consent discussions included the beliefs of clinicians in their personal competence in communicating with families; their experience of working with individual SNODs; perceptions of their own responsibilities to families; and the clinical and ethical issues surrounding controlled DCDs. There were also often perceptions of a conflict between the SNODs’ interests and role in conducting the PDA and what clinicians viewed as best practices for patients undergoing planned limitation or withdrawal of cardiorespiratory treatments.

4. Evidence that consent rates are higher when the donation discussion is collaborative indicates that increasing the proportion of family discussions conducted collaboratively would have a major impact on family consent rates.

Abstract

Hospital chaplains from different faith groups were interviewed as forming part of the multidisciplinary ICU team. Given the limited knowledge regarding the role of the hospital chaplains, we examined how they viewed their role in supporting patients and families from different faith groups. This identified the importance of generic chaplaincy work, offering spiritual support to patients and families in terms that are not informed by religion. All saw themselves as making a contribution to end-of-life care and most were supportive of deceased donation but had limited knowledge of its processes and practices. Scope was identified for greater involvement of hospital chaplains both within the hospital setting and in the community, with appropriate training and contractual requirements.

Aims

To explore the perceptions and practices of a diverse sample of hospital chaplains regarding end-of-life care and organ donation.

Background

NHS chaplaincy is emerging as an allied health profession that covers a range of religions and spiritual orientations. Hospital chaplains aspire to be relevant to the whole population of health-care users and have a potentially important contribution in supporting patients and families at the end of life as part of the multidisciplinary team. 94,95

Hospital chaplaincy teams in major British cities involve chaplains from different religious backgrounds and spiritual practices to reflect the local population, working in part-time and full-time capacities, both as volunteers and on a salaried basis. As with other allied health professions, NHS chaplains’ work is described and managed in secular rationalist and humane terms, without reference to religious doctrine.

In principle, a chaplain from any faith background can become the lead chaplain with responsibility for managing the chaplaincy team. In practice, it is likely that lead chaplains will continue to be from Christian traditions for the immediate future since salaried chaplains from minority faiths are recent and currently often part-time appointments.

Research from the USA suggests that the chaplains’ role in supporting patients’ decision-making around end-of-life care makes a measurable difference to outcomes. 96 It may also be expected that the chaplains’ role may, for some, influence donation decisions.

At a national level, all religious groups in the UK provide support for organ donation. 4 In contrast, a study of religious leaders identified some variations in individual’s views of the acceptability of deceased donation. 97 However, little is known about the views and practices of hospital chaplains from different faith groups.

Our research included chaplains as part of the broader category of ICU staff involved in the process of deceased organ donation. The key role that chaplains can play was emergent during our research and they were therefore analysed as a separate group. We were some way from reaching data saturation with these interviews, which limited the comparisons possible between faith leaders. Nevertheless, the data contributes to the very small literature in this area (for published paper from these data see Bradby et al. 98) and identifies hospital chaplains as a potentially important resource whose role could be expanded.

Methods

Semistructured interviews were held with 19 chaplains across five hospital trusts as part of the larger study of the study hospitals.

Findings

Recruitment

We initially contacted the CLOD and the SNOD to request for their assistance in contacting ICU staff and members of the trusts ODCs to participate in the EDGs. The CLOD usually circulated an e-mail on our behalf and when e-mail addresses were provided to the research team the general invitation was followed up with direct emails to consultants and other appropriate staff.

Discussion groups

A total of 10 groups were held, five with nurses and clinicians (or in one case nurses alone). Each group included people in junior and senior grades. The other five groups consisted of members of the ODC. Given the difficulties in arranging the EDGs and because we were only able to hold a doctors and nurses group at one of our original five sites and only an ODC members group at another site, we decided to include a sixth site where we held EDGs for both groups.

The numbers of people attending the groups were not known in advance as availability depended on the demands of the unit. EDGs were held in a meeting room close to the ICU. Groups with ODC staff were mainly held following an ODC meeting to facilitate attendance.

Conducting ethics discussion groups

The attendees were provided at the outset with a participant information sheet, a brief verbal introduction to the research programme and aims of the EDG. They were also asked to sign a consent form and to keep the proceedings confidential.

It is important to note that the EDGs were not focus groups as traditionally understood. Discussions were conducted in an open manner with the facilitator following and exploring themes that emerged, based on the approach of a philosophy seminar in which participants are asked to consider issues such as the value they attach to different concepts and choices. The EDGs also differed from interviews in that the facilitator (BF, a philosopher in our study team with long experience of facilitating similar groups addressing ethical issues in health-care settings) followed up contributions with probes informed by her philosophical orientation. Unusually, participants were asked to be prepared to explain and defend their positions and when appropriate to challenge those of the institution or other professionals in the group. Moreover, BF’s approach and focus was guided by reading transcripts of interviews already undertaken with staff at the study site.

It is a feature of this methodological approach that made the discussion schedule unique to each EDG. The topics for discussion were chosen by reading the interview transcripts of staff at the study site and identifying themes and issues for further exploration. In practice a number of themes were identified as common to all groups but there was also the opportunity to discuss issues of particular local interest and or novel points raised by individual interviewees. In the latter case care was taken not to reveal the source of the issue.

In the EDGs the discussion began with introductions and ground rules and then explored issues that the interview data suggested that participants would feel comfortable to be challenged on. As the discussion progressed more contentious issues were introduced and as more groups were interviewed there was also the possibility of using references to earlier discussions to take groups into new areas. By the end of the process all groups had covered a broadly similar range of issues but the emphasis, time spent and resulting discussion differed dependent on the interests of the group starting out, their prioritisation of issues, their willingness and ability to defend their views and challenge one another, and finally whether or not they felt a need or desire to reach agreement and/or consensus.

Nine groups were facilitated by a health-care ethicist (BF), one by a social scientist (CK) and eight were attended by the programme lead (MM). All discussions were recorded with participants’ permission. Groups lasted for about 90 minutes, with some people coming to join the group or leaving to fit with demands of their job.

Analysis

Both the interview and EDG recordings were fully transcribed and entered into NVivo version 9. For our analysis of the individual interview and EDG transcripts, we used thematic analysis following close readings of the transcripts. The themes reported here emerged from this analysis as the study proceeded and were discussed in our team meetings. All quotes are referred to using an EDG number or interview number.

Shortage and scarcity of organs

Underpinning these discussions was the construction of a ‘shortage’ of organs for transplant because more people will need to be recipients than the number willing (and able) to donate organs. Given the statistics available through NHSBT most staff took this claim to be relatively uncontroversial in pure numerical terms and most were comfortable to quote commonly available statistics without comment or criticism, including a specific and even more serious scarcity of organs available for members of BAME groups. There was a further tendency within the groups to move quickly on from the fact of shortage to discuss the barriers to donation from suitable donors within the same BAME groups.

The biggest group that we face difficulty with in terms of acceptance . . . was obviously amongst Middle Eastern and Asian minorities, but also amongst very strict Jewish faith. And pretty much the, the ‘no’ rate was 100%.

EDG site 1, doctors and nurses

I don’t know of any single Muslim family during the 6 months I was there, so maybe 10 to 15 times, the success rate was zero.

EDG site 1, doctors and nurses

The significance of the lack of BAME donors was connected to the need to find the best HLA tissue match for optimal transplant outcomes:

It’s like bone marrow, like all other things, how many of the people who are actually donating, agreeing to donation and how many of them are the recipients? So you would find that the South East Asian, you know, the tissue typing and stuff is quite specific. There’s better acceptance rates or recipients would have better acceptance rates if you get it from the same Asian matched or whatever group that you are in.

EDG site 3, ODC

However, there was a slight difference of opinion on the extent to which HLA tissue matching is essential to good transplant outcomes. For example, one group discussed how because of improvements to anti-rejection drugs the need to match donors and recipients ‘is probably not medically as true now as it was probably 15 years ago’ (EDG site 5, ODC).

This has been recognised to some extent in recent practice guidance,101 for example, in changes in HLA matching criteria and in acknowledging the impact of more effective immunosuppressant drugs, although these drugs are associated with high risks, including risks of cancer. 102

It is interesting to note that participants did not choose to interpret the scarcity of ‘suitable’ organs in anything other than a clinical/scientific sense. A suitable organ was seen as one with a chance of being a good match that is one which would work well when transplanted irrelevant of origin or destination. Donating to ‘one’s own group’ was seen as necessary or preferable only because of the clinically relevant issues raised by matching donors from different ethnic groups. There was no suggestion that prospective recipients were concerned about receiving organs from within their own group and there was no sense in which people were thought to be waiting for an ‘ethnically suitable match’. Unlike the discourse in fields such as sperm and embryo donation or xenotransplantation, there was no discussion of the potential unsuitability of donation across ethnic groups owing to concern for group identity or controversial issues such as ‘contamination’. An exception to this was a case reported in one of the EDGs:

I was speaking to one of my colleagues who is a very devout Muslim and he’s of the opinion that . . . a Muslim can only accept an organ from another Muslim. Whereas he can donate to anyone, but recipient wise, it must be from the right donor.

EDG site 5, ODC

By restricting the concept of ‘suitability’ to the very narrow clinical interpretation of the term, clinicians were at the same time able to acknowledge scarcity and the reasons for it without having to take on the full ethical implications raised by the possibility of there being a perceived ‘free rider problem’.

For as long as low rates of donation within a particular group disadvantage that group alone, there is the potential for the problem to be minimised or hidden, particularly if the groups’ interests are generally harder to highlight and defend. However, if it comes to be the case that the scarcity could be addressed by utilising the good offices of people outside the group, the nature of the issue changes. You could now have a group that does not do what it takes to help themselves, but rather relies on others to do it for them. There then emerges a negative rhetoric familiar within liberal politics (‘scroungers’ vs. ‘taxpayers’), which demands a response.

The assumption of scarcity was, therefore, an important element of the ethical context within which people saw themselves operating. However, the clinical limitations within which they were currently operating provided them with a way of avoiding some of the really difficult issues. The ‘groups’ in question could be understood in terms of differing clinical need (tissue type) as opposed to differing donation practices. Race, ethnicity and other forms of shared identity could be put to one side, as there was little or no evidence of people wanting to receive from only ‘their own’ and only a few notorious cases of people wishing to direct their donations on what could be seen as discriminatory grounds.

The shortage of suitable organs was seen as in large part attributed to a failure of sufficient members of BAME communities to donate compounded by a higher possibility of family veto within BAME communities. It was interesting to note that within the EDGs there was little willingness to ‘own’ the issue of scarcity or be the focus of a solution, with responsibility being given to others ‘outside the hospital’ to raise awareness, build community engagement and encourage donation. SNODs cited lack of time and resource, and, in some cases, expressed regret at what could be done if they only had time to ‘get out of the hospital’. Having said this, there was a clear acknowledgement of the need to approach BAME families and to do so in an appropriate manner to create the best possible chance of donation once the situation arose.

The clinical/scientific interpretation of scarcity is comfortable because it insulates the clinician from confronting the full extent of justice-related issues raised by the free rider problem. By stating that the problem of scarcity arises because people from within certain groups need to receive an organ of a particular tissue type and these organs are in short supply because of low rates of donation within that group, the problem is contained within the group, the solutions in terms of meeting the needs of individual patients reside with the group. If you accept that a possible response to this is to remove the clinical/scientific barrier to transplant by minimising the need for in-group donation, you allow for the possibility of members of the community benefiting from donations outside the community, when their own community has failed to respond.

Approaching black and minority ethnic families

There was a prevalent view that approaching BAME families might be seen as more challenging. Reference was made to the size of some family groups, the problem with identifying a key informant/decision-maker, language barriers and different patterns of bereavement, etc. Having acknowledged the problem of an initial approach, staff went on to consider the terms within which an approach might be framed. For some staff, there was ambivalence about whether families should be approached in ways that are informed by knowledge of different cultural beliefs and practices, or whether they should be approached in the same way as the families of all other patients:

I think you need to be able to know people’s cultures and the different cultures there are around, so then you know how you can approach the organ donation. You know – well actually we were saying that you approach everybody the same, but it would be good to know more about people’s cultures.

EDG site 5, nurses

Several participants echoed this feeling of not having correct or sufficient cultural and/or religious knowledge, as well as a fear of offending families:

The barrier is definitely not knowing their culture . . . Not knowing if you can approach that, if it’s appropriate or not . . . like do I discuss organ donation with these families, do they agree with it, do they believe in it, would it offend their religion or, you know . . .

EDG site 5, nurses

I’ve seen reluctance from senior consultants or even staff that they won’t even broach the subject because they don’t want to offend or . . . they don’t know if people are aware of the importance of it if they’ve come from a different culture, if they’re new to the country and things . . . they won’t necessarily know about it.

EDG site 5, nurses

A similar obstacle was the fear of being perceived as being discriminatory in their treatment of BAME families.

I think there are times when, especially with BAME communities, there’s a fear of how to engage, a fear of getting it wrong, and then, you know, the fear of being branded racist if something went wrong.

EDG site 5, ODC

On the whole, members of staff who participated in our programme of research were careful not to use stereotypes about particular groups or religions, and pointed out the importance of recognising diversity within such categories. To some staff, the extent of diversity even within particular ethnic or religious groups could work against the idea that culturally specific approaches could be formulated, since they could not be sure of being sufficiently appropriate for all families within a particular category. Some participants talked about all people being individuals and commented that it was not helpful to think in terms of ethnic groups having either different or shared group beliefs.

They also recognised that views often differed between generations, with examples mainly given of Asian families:

I do a lot of, we do a lot of open days and we have an awful lot of young Bengali, I would say sort of teenagers, early sort of – between sort of 13 to 17/18. And they’re all interested in organ donation, and some of them have come up to me and said, ‘Yes I’m very keen, but my family . . .’ – that’s the thing, you see, you’ve got a generational thing going on. So the young ones are getting it, they are getting it.

EDG site 5, nurses

Discussions around the approach to families raised a number of ethical issues, some of which staff clearly found difficult to discuss openly. As in previous studies that have adopted the EDG approach, there was a shared sense among many of the health-care professionals that religious- or faith-based objections hold a special force in a health-care setting. This, in turn, meant that the default response might be to accept an individual or family’s view without question or challenge if it is presented in terms of a faith-based objection or demand. This may be even more prevalent when the health-care professional is unfamiliar with the tenets of the belief and/or when the group in question is perceived as particularly vulnerable to misunderstanding and/or prejudicial views.

The wish to neither stereotype nor make assumptions based on ethnicity or group membership was clearly stated, yet it is possible to move beyond a crude and reductionist form of profiling to acknowledging that cultural and religious factors will combine with age, class, education, etc., to provide some clues to a family’s preferred way of doing things. Although no two families of any type are the same, and all should be treated appropriately, it was clear that a wish not to discriminate could potentially translate into a tendency not to pay due regard to morally relevant differences across and within groups. Another interesting finding from the groups was the fact that health-care professionals from BAME groups were not necessarily feeling any better equipped to approach the families in question.

Reciprocity

As previously mentioned, when there appears to be inequity in the provision and consumption of social goods there is the potential for injustice specifically through the notion of free riders (people who benefit without contributing even though they could). At the time of the groups, there was a growing interest in the concept of reciprocity as a potential response to this issue both within the professional literature and in the media as a result of policy decisions in Israel.

Some staff were aware of the use of officially produced materials relying on reciprocity-based scenarios.

I worked at [x hospital] recently and . . . I noticed there was a lot of posters up with different ethnic minorities with a, you know, ‘I’m on the transplant list, if something happened to me I would want, I’m waiting for kidneys, if I were well I would donate mine,’ or something, you know, something. And it was quite clearly, they had three posters, one each from different ethnic group . . . some of them were those posters where you had the person well and then you had the person who is ill . . . it felt very targeted actually, the ethnic minority and the ‘here I am well,’ and the reciprocal, you know, ‘if I’m on the register, I expect that I would give anything I could’.

EDG site 1, doctors and nurses

However, there was a caution expressed in relation to the use of publicity material with several members of staff reporting that they had observed the removal of advertisements targeting BAME patients and families in waiting rooms. This was linked to anecdotal claims that this approach could be counterproductive:

So there’s some evidence to show that currently in these, in the communities, the advertising potentially is counterproductive because people, you know, have an instinctive reaction to it and then reinforce their prejudices.

EDG site 6, ODC

A member of an ODC put the problem in economic terms:

There is a business case for it . . . one of the things I want to do, you know, my involvement with this committee, is to take this message out to communities and just say, ‘Look it’s in your interests that you are involved in this.’ And there was one meeting I went to last year and I actually said to a lot of BAME communities there that, ‘Look if you don’t wise up, people in your community are just going to die, it’s as simple as that’.

EDG site 4, ODC

Some of those in favour of making special efforts to encourage organ donation amongst BAME families commented that families might be more willing to consider donation if they believed someone they knew or at least someone from their own ethnic group could benefit. So, for example, one strategy could be:

educating people to say that if your relative does donate organs, it’s not just going to an English person, it may go to another person who you know or somebody else who has renal failure.

EDG site 5, nurses

People will respond to that, it’s our people who want organs. It’s kind of something about your being mine, you being us, and therefore it will help you and therefore it will help one of our own.

EDG site 5, ODC

These comments suggest that an appeal to solidarity with people from the same ethnic background, or people with the same disease, could be the basis on which to make an approach to BAME families for donations. However, participants acknowledged that it might not always be possible to trump fundamental resistance. One consultant recounted how he had tried this approach with the family of a patient in ‘resus’ (resuscitation) with no success:

I specifically used the, the little hook that, you know, ‘her kidneys will usually match with people who are similar to you.’. . . But obviously that was a refusal as well . . . that was a Muslim family . . . that was the first I’d tried the little, you know, ‘It’s like her, because of her genetics, her organs would go to help someone who’s likely to be of a similar race to you’. . . but it didn’t really – so I don’t know the ideal way to approach really. I suppose, watch this space.

Site 3, consultant interview

Within the EDGs it was seen that the notion of reciprocity allowed for an acknowledgement of the scarcity problem while still containing it in the manner achieved by the clinical criteria-based discussion above. Thus, an appeal can be made to community solidarity that can be respectably and legitimately defended, because the clinical facts mean that your donation is likely to go to someone within your community. This moral convenience is unlikely to persist, however, as it is also in the interests of the community to conduct research and introduce new practices to minimise the impact of tissue types on the success rates of transplanted organs. The question then arises whether or not it is morally defensible to seek to raise organ donation within BAME groups by appealing to solidarity and the wish to benefit one’s own if that would only be possible through the introduction of some form of directed donation. There was no support for this step within the EDGs, but there was some sympathy for the wider case for societal rather than group reciprocity, such that if you do not give you should not receive, with the acknowledgement that this could further disadvantage BAME groups in the absence of an increase in donation rates.

Unwillingness to donate

The EDGs were keen to speculate on why for members of some groups there was a reluctance to give organs despite there being a willingness to receive them. A suggestion was that an asymmetry in giving and receiving organs could be explained in relation to Muslim patients by the belief among this group that the body should not be ‘cut up’, which overrides the consideration of benefiting other people; and on the other hand the belief that transplants are made possible without the need for personal agency because:

Allah has given them an organ to save their lives.

EDG site 5, nurses

One of the consultants, talking about a Muslim family who had declined donation, commented on how far the concept of belonging to a wider society might apply for such families:

I think that there’s not the same sense of community. In the same way as I’ve described a fiduciary relationship, which begins with the patient, moves to the family and goes out into the wider society, I don’t think they have that same sense of wider society. I think their view of society is the family, and I think that’s where the walls stop.’

Site 3, consultant interview

This view chimes with the reported experience of some members of staff who had been involved in the field of live organ donation; anecdotally they reported that while some African-Caribbean families will participate in live organ donation within the family it was rare, if not unknown, for members of this community to come forward to donate beyond the family.

Although these are clearly not robust data on which to build conclusions, it is important to understand the shared understandings and ‘stories’ that circulate within medical teams. An assumption that people from certain communities ‘don’t donate unless to their own’ in the narrowest of senses might well have an impact on the willingness of staff to approach such families. This might also have an impact at a policy level with the problem being seen as structural and insurmountable, as well as culturally and religiously sensitive. Similarly, another consultant thought that in some very:

. . . isolated cultural communities, they would either want organ donation, only to have them if it’s one of their own, or they can’t see the benefit because they’re very locked in communities, they don’t perceive the rest of communities outside their own, to be given a gift. And they very much feel that, you know, they’re the ones who need everything. And they’ve not prepared to move beyond that.

Site 5, consultant interview

Staff felt that a further reason for questioning targeted approaches was that this strategy could be perceived as being associated with directed donation (i.e. donation solely for the benefit of specific groups or identified individuals), which is prohibited by UK regulations. These regulations are specifically designed to ensure fairness in the allocation of organs and to ensure that morally irrelevant factors such as race or even the football team one supports do not become factors in limiting the scope of a person’s donation.

Some staff warned against ‘tiptoeing in the direction of selected donation’ and commented that it was important not to move away from the idea of pooled donation. In this view, donations should be seen in terms of contributing to the greater society, not just for the benefit of a particular group. In turn, there was a feeling among some practitioners that even the suggestion of direction to a community member by dint of tissue matching should be avoided.

Taking but not giving organs

Staff did not avoid the conclusion that the lack of available organs was due, in significant part, to low donation rates within BAME groups and they were also prepared to invite members of the communities, including those on waiting lists for transplant, to reflect on this fact. Having rejected any notion of explicit directed donation within communities and in the absence of a united call for the introduction of a reciprocity model, such as that introduced in Israel,23 there were several calls for better community-based education specifically highlighting the unmet need and utilising the possibility for community-based altruism.

As one consultant described the situation:

I believe a large population of those waiting for a right match are from the ethnic minority. And they are waiting the longest, and that’s because a right match is not found, and that’s because the donation pool is not as big. I think the community should know those figures rather than just saying, ‘We are doing good for mankind . . . And then we could then give the community a real perspective of what the actual scale of the problem is . . . But I feel that that message is not conveyed yet . . . If we came out with a figure to say, ‘Look guys, organ donation is difficult in this community because it has to be appropriately matched and your, your sort of ethnic minority are the longest waiting, and there is X many thousands of patients waiting.’ I think that is when it hits home.

Site 3, consultant interview

Similarly, other participants thought that BAME families should be informed of the situation and invited to think about reciprocal benefits when the possibility of donation arises, for example by saying:

‘If it is one of your relatives on the waiting list, they will wait longer because your community are less likely to donate.’ That sounds like putting the guilt factor in, but it’s the truth.

EDG site 6, ODC

However, two consultants, while generally supportive of increasing understanding at a community level, were not happy with using this approach in the ICU:

It’s not the time when you’re discussing the issue about organ donation with the relatives. You can’t say, ‘Oh did you know so many people are waiting?’ because it’s completely not the right time to do it. But yes, it could be done in outpatients . . . So that those who are on dialysis or in end-stage renal failure, their relatives can learn to say, ‘Actually yes,’ you know, yes I think there is a role for hospitals to play as part in community education.

Site 3, consultant interview

Participants varied in terms of their propensity to judge those who would receive but not donate an organ. In an interview with a consultant the term ‘double standards’ was used to describe a potential recipient being willing to be saved by transplanted organs, but not being willing to agree to donation:

I think that people born and bred in this country probably have thought at some stage, ‘Well, you know, if I needed a heart, I’d have one, if I needed a kidney, I’d have one, I wouldn’t refuse to be a recipient, just on the basis that I’m so and so religion.’ And, you know, if they’ve been through that thought process, then when it comes to saying, ‘Well I’m happy to be an organ donor,’ they would be I aware, I think, of the double standards involved in that.

Site 4, consultant interview

Some staff commented on why reciprocity perspectives might not sit well with BAME families. For example, there was some anecdotal evidence that some people from BAME families have a belief that one day they will be returning ‘home’, so they do not want to donate in the UK. One member of staff commented that this was a view he could relate to:

I think the majority of them don’t want to know, don’t want to get involved because I sense that these people are here temporarily. I think, you know, minority ethnic minorities are here . . . some of them are in transition. So they come here to work for how many years and they have in mind that they will retire and go home. So it’s nothing that they would like to get involved, like ‘Why would I want to register as a donor here, I’m going home.’. . . I see from my own perspective, I would donate anywhere obviously, but in my head one day I’ll move and I won’t be here. So if I just had that in mind I’d go, ‘No’, if I had to tick a box because I’m going home.

EDG site 5, nurses

This could be read as an expression of alienation or a sense of not ‘belonging’, and, therefore, not feeling that there is a commitment from the wider society to people from BAME groups as citizens that they would wish to reciprocate. A consultant made a similar comment:

They need to build from the ground up, the idea that transplantation is a benefit to everyone, because maybe . . . if you’re in any minority group and you can see the establishment as being not there for you really, even if you’re using it, it can be still a cultural perception that it’s more for other people than it is for you.

Site 2, consultant interview

So although many of the staff considered issues that could be conceptualised in terms of solidarity and reciprocity in relation to the value of donating and receiving organs, there was acknowledgement from their experiences and knowledge of their patients and their families that these were not necessarily priority values in the minds and belief systems of those families. They were also prepared to situate a consideration of societal reciprocity (as the ethically defensible model) alongside issues of community identity and membership, bodily integrity, and historical and future affiliations, which came together to provide a complex picture of the ethically appropriate context for donation.

Study sites

The initial application implied that only bereaved families at five NHS trusts in the London region who met the inclusion criteria would be invited to take part in the study. However, in 2012, only 56 of the 1212 donors in the UK came from black, Asian or other minority ethnic groups and thus constituted a very small group. Owing to a small sample and an expected low participation rate, recruitment was subsequently extended to include the top 20 English hospital trusts in terms of numbers of approaches to BAME families for organ donation and we also approached the five remaining London trusts (of which all but two were able to participate). This involved 23 additional sites, giving 28 sites in total. This required that each of these trusts became a patient identification centre (PIC), which permits clinicians or clinical units to identify and refer potential participants to external research teams without requiring a full governance review.

Subsequently, it was clear that 28 sites were unlikely to provide sufficient participants and NHSBT provided data that identified a further 18 trusts as having one or more consenting kin from an ethnic minority group; of these, 12 trusts participated and PIC applications were completed. A total of 40 trusts were therefore involved.

Identifying potential participants

The most comprehensive way to identify and contact potential participants was via NHSBT, which retains records of all organ donors and next of kin contact details, unless no further communication was requested.

The intention was to initially employ a retrospective approach and invite the next of kin or family member who had or had not consented to organ donation to take part in an interview, if they fulfilled the following inclusion criteria:

• adult donor (aged ≥ 18 years)

• died on ICU [or accident and emergency (A&E)]

• recorded as being from a minority ethnic group or mixed background or family

• next of kin did or did not consent to deceased organ donation.

The expected number of participants at the time of submitting the grant was 20–24, ideally split between consenting and non-consenting families.

It was proposed that both consenting and non-consenting families would be retrospectively recruited via NHSBT. However, NHSBT’s governance requirements mean that it does not centrally retain the details of families who were approached but declined organ donation and asking hospitals to search their records was not regarded as feasible owing to the length of time this would take and the finite length of the programme. Considerable contact and negotiation occurred with NHSBT to try to overcome this problem; however, this did not prove feasible. Instead, an alternative recruitment strategy was suggested that involved the prospective recruitment of non-consenting families. The two recruitment strategies are described below.

Recruitment delays

Recruitment for this study was very time-consuming with considerable researcher time involved. This partly reflected the substantially expanded number of study sites (see Methods, Study sites), together with the need to include a prospective component, which required a substantial amendment to be submitted to the Research Ethics Committee. Recruitment began in September 2011, with the final recruitment packs sent out in September 2012.

Interviews

Additional training was undertaken by the researchers to prepare them to interview the bereaved, a group with whom neither had previously conducted research. They attended a training course, ‘It’s hard to talk about it: taking the risk of opening a can of worms’, at Sobell House, Oxford. This focused on difficult conversations about the end of life and death, primarily aimed to prepare those who work in hospices.

All participants preferred to be interviewed at home. The topic guide was structured chronologically to guide the participants through their experience (see Appendix 8). It covered:

• background information about the donor and events surrounding their hospital admission

• impressions of the hospital, diagnosis and communication

• potential treatment withdrawal/breaking bad news

• approach for consent for organ donation including how this was explained, discussions with family and decision-making

• DBD- and DCD-specific questions

• end-of-life practices

• consent paperwork

• reflections on the donation decision and taking part in the research.

All participants received a participant information sheet and signed a consent form before the start of the interview. The interviews were all conducted at home as preferred by participants. They were audio-recorded (with permission) and designed to last 60–90 minutes.

It was made clear to the participants that they could have a break during the interview or could stop the interview at any time if they found this too distressing. The researchers took information from local bereavement counselling services and ensured that the participant was all right before they left.

After each interview the participant was sent a thank you letter together with a brief post-interview feedback questionnaire that was returned to the chief investigator. These feedback responses were extremely positive, with the participants regarding the researcher as most understanding and feeling very comfortable talking with them. They also reported that they were able to cope with the length of the interview and, although recounting a particularly distressing time, they generally found it helpful to have talked about their experiences.

Characteristics of participants

There were a total of 17 participants, with three interviews involving more than one person. Altogether four participants were a spouse of the deceased, seven a parent (mainly mother), three a sibling and three a child of the deceased. Participants were predominantly of Indian origin and mainly Hindu and Christian faith (Table 10).

Initial communication

The events leading to their relative’s hospital admission were unexpected and the result of, for example, heart attacks, aneurysms or accidents. The initial contact and communication from the staff varied. Those whose relatives were admitted to A&E and remained there reported waiting a considerable time before being informed about their relative’s condition and were not happy with this situation:

The doctor who was in charge, you know . . . I kept asking because I wasn’t sure what was happening. And didn’t say anything until about 3 hours, that that’s what’s happened. But they had told me that he’s not well.

Respondent 2: wife, Indian, Jain

Participants were aware that an admission to intensive care implied that the prognosis was unlikely to be good, but their initial contact with and impressions of the hospital were generally positive and often supported by existing knowledge of the hospital’s reputation:

I knew that for head injuries, he was in very good hands at [name of hospital]. So from the medical point of view, I was, there wasn’t a better [place] that he could be. So we knew that he was at the right hospital with the right teams presumably looking after him on that evening.

Respondent 1: brother, Indian, Hindu

Relatively soon after admission to an ICU the participants reported meeting with an ICU consultant who updated them on their relative’s condition and they appreciated the honest communication that they received:

I don’t know, he was a surgeon, he was still in his scrubs, he was quite matter of fact. Honest, I mean that’s what you want, I mean you don’t want to be given false hope, do you really, you know.

Respondent 10: sister, Indian, Hindu

Subsequent communication from the ICU staff appeared to demonstrate good practice that was characterised by talking to the family in a private space (away from the communal waiting room), explaining the patient’s condition clearly, assuring them they were doing all they could and ensuring the relatives were comfortable to ask any questions they may have:

At the [hospital] they was so, it just seems that no stone was left unturned. Everything was done to clearly let us know and to assure us there was nothing.

Respondent 3: mother, white British, Catholic

Being near to the patient and being able to spend time at their bedside was important for all of the participants and other family members and was usually facilitated by the ICU staff. However, when the family was updated about the patient’s condition this often took place in a small family room that often could not accommodate everyone the family wished to be present, although the efforts of staff to identify and allocate another larger room were appreciated.

Breaking bad news

During the patient’s time on ICU the participants reported conversations with the consultant to receive updates about their relative’s condition and prognosis. The participants appreciated receiving accurate and honest information but reported disbelief about the seriousness of the patient’s condition particularly if their physical appearance looked ‘normal’.

The communication of the ICU staff communication with the families was reported to be clear and usually delivered in a way that the family understood, with a space provided for the family to ask additional questions that the staff were happy to answer. Respondents also commented favourably on the efforts to facilitate end-of-life care in a way that was sympathetic and empathetic to the family and their situation:

So it’s like everybody done everything from their heart. So there is no complaints about it. There’s no mistakes or something. So whatever they’re doing is very good.

Respondent 12: son, Sri Lankan, Hindu

Diagnosis of death

While participants’ specific recollection of what they were told about brain stem tests varied, most reported understanding the concept of brain stem death and knowing that a ventilator was ‘breathing’ for the patient. The same level of explanation may not have been circulated to wider family and friends, who perceived the patient was alive:

Well some of our family relations didn’t believe she’s dead because they said, ‘There’s breath,’ so I said, ‘Yes it is artificial breath because they don’t want to upset you straightaway, that’s why they kept her breathing.’ But some of the people say, ‘No she’s not dead, there is still breath.’ I said, ‘Yes I know there’s a breath . . . because it’s artificial breath.’

Respondent 12: husband, Sri Lankan, Hindu

One inconsistency is whether or not families are invited to observe the brain stem test, which was often a way of assuring themselves that their relative was dead.

From the sample, three patients were DCD donors and all ‘died in time’ for donation to proceed. The participants reported receiving a clear explanation about how death would occur including the body’s physical reactions and the need to move swiftly to theatre after the patient had been declared dead for organ retrieval to proceed.

Approach for consent

After families had accepted the death of their relative, at an appropriate juncture, organ donation was raised, usually with the next of kin. The approach was often led by the SNOD, although for at least two families in this sample, they were initially asked about organ donation by a consultant. Regardless of who made the approach the families felt this was done sensitively. Differences occurred in how organ donation was seen to be presented to the families who reported, for example, organ donation presented as a way for their relative to live on or to save or improve the lives of others. Three families distinctly remembered being told that if they donated their relative’s organs these were likely to go to someone of the same ethnic background. This was not reported as influencing their donation decisions. However, one family felt they might have been slightly misled when they learnt that the recipient of a donated organ had gone to someone from a different ethnicity.

One family reported feeling under pressure by the ICU staff to consent to donation but other families described and appreciated being given time to think about organ donation and discuss this with their family, particularly when their initial reaction was one of uncertainty. Although a few families initial reaction to organ donation was ‘no’, after being given time to think about organ donation and usually having discussed this with other family members, they decided to consent. The families welcomed the lack of pressure to make a decision, so that they could make a decision that was right for them and their relative.

No there was no pressure. We could step back at any stage. However, she explained that, ‘Once we go into the theatre, that’s it, it’s their right after that stage, but up until that door opens and we wheel her inside there, you can step down at any stage.’ And that’s what I liked about the whole process with them, you know, just the respect, the way they spoke about her, the way things were done.

Respondent 8: mother, black Caribbean, Christian

Making the donation decision

Two patients were on the ODR, with their wishes therefore known and respected. However, other patients had not discussed organ donation or their wishes with their families, with the decision, therefore, left with the next of kin. In this situation there was little discord reported between family members when organ donation was suggested and if initially discord occurred, donation was eventually supported with the final decision usually taken by just one person, often the (legal) next of kin. However, this person sought the views of other family and friends who were broadly in favour and made it clear that they would support the final choice the next of kin made.

In considering organ donation and coming to a decision, the participants thought about what the patient would have done and often involved reference to what the donor was like as a person:

And then what I had to do was put myself in my daughter’s position, think of my child. And I based it upon what would [daughter’s name] have done? And knowing [daughter’s name] she would give away everything, [daughter’s name] would give away everything. And that’s what I based it on.

Respondent 8: mother, black Caribbean, Christian

Perhaps unexpectedly, several families raised organ donation themselves with the ICU staff. This arose as either they knew of someone needing or having received a transplant or they had a wider awareness of the need for transplants via the media. For example, one family described seeing an organ donation leaflet in the family waiting room. Underlying their support for organ donation was a desire to help others live and a belief that it is what their relative would have wanted or supported, even if they did not know their wishes:

I think it’s just when you hear a lot of people needs kidney and everything, you know, you hear so many things on the telly. I usually watch the programmes all the time, the hospital programmes and things like that, and I see how many people, you know, the people who need all these organs, and how difficult it is in here. So I would have thought that that would have been something that I’ve done for other people’s life.

Respondent 2: wife, Indian, Hindu

After consenting to organ donation, family and friends usually spent time with the patient at the bedside to say their goodbyes. This was often facilitated by the SNOD in negotiation with the ICU and supported by the bedside nurse. In addition, some families reported wanting to carry out particular faith or cultural practices, for example, praying at the bedside, which was relatively easy to facilitate for other families, compromises were found so they could fulfil practices without compromising the unit or the care of other patients. For example, for a Malaysian Buddhist family the burning of a talisman had to take place in a designated area of the hospitals’ car park rather than at the bedside. If a compromise was required, the participants were happy to go along with this.

As well as meeting practical, emotional, faith and cultural needs, the families were offered a memento of their relative. These ‘mementos’ vary by hospital and are arranged by the SNOD. The most common mementos are handprints or a lock of hair; with less common items including photographs of tattoos, footprints and for a paediatric donor casts of their feet.

Post donation

It is usual for the SNOD to telephone the next of kin or a nominated contact to tell them that the organ retrieval is complete and later to send a letter outlining which organs were suitable for transplants with a brief descriptor about the recipient. The families were pleased to learn that their relatives’ organs had saved or improved the lives of others, albeit strangers who they would never meet.

On the first anniversary of the patient’s donation, the SNOD usually sends an anniversary card to the next of kin, again thanking them and updating them on the progress of recipient(s). However, in practice, not all the participants had received further contact from the SNOD, and would have generally have appreciated an update just to know that the recipients of their donated organs were healthy and getting on with their lives.

Strengths and limitations

This is the first UK study to ascertain the experiences of minority ethnic groups in relation to deceased donation. With just 13 participants who consented to donation it is not possible to argue that saturation was reached in the analysis of this group, although there are consistencies across the data reflecting good communication practices and experiences despite the tragic circumstances. The most significant limitation was being unable to retrospectively contact non-consenting next of kin. This group forms the majority of the minority ethnic groups who have been approached for consent for organ donation, and without access to this group we were unable to identify the reasons for not consenting to organ donation. It is not known if commonly cited reasons for not consenting such as faith or cultural practices hold true or if other reasons influence this decision. Without this knowledge barriers cannot be addressed and these may continue to prohibit organ donation.

Families’ positive views of experiences

Among the group of consenting families the quality of the communication between the ICU staff and the families was reported to be of a high standard. Families appreciated clear, honest and accurate information about their relative’s condition after admission and subsequent updates, often in a private space. They felt comfortable, were able to ask questions at any point and recognised the time consultants and nursing staff spent with them. Consultants broke bad news and explained how the patient would die or be declared dead with sensitivity and clarity often after earlier conversations about a poor prognosis, which had prepared the family for bad news.

Families’ perceptions of gaps in services

Families would have liked the offer of observing brain stem death tests when these were carried out, but this did not always occur. There was room for improving communication with families who came via or remained in A&E when it appeared to take some time before the family were spoken to by clinicians. This is likely to be a significant future issue with A&E departments, regarded as a further source of potential organ donors. Families often heard nothing about the organ recipients without asking and would like to be automatically updated on how they were doing unless they had opted out of further contact.

Methods

The one-page feedback questionnaire comprised six structured questions with 7-point Likert scales and two open answer questions, plus questions on the respondents’ experience of organ donation and job category (see Appendix 12). Responses on the Likert rating scales were categorised as positive (rating 5, 6 or 7), intermediate (rating 4) and negative (rating 1, 2 or 3) (the scale was flipped around for questions 3, 4 or 5 to take account of the difference in wording).

A total of 66 participants completed feedback forms at the end of the session. Respondents comprised a mix of bedside nurses of various grades (bands 3–7), student nurses, SNODs and doctors (a mix of consultants, specialist registrars and training grades). Apart from the SNODs and consultants, few had any experience of organ donation.

Findings

Altogether, 83% of responses to the structured questions were positive, 11% intermediate and 6% negative. The only question with < 82% positive responses related to the length of the session, with only 65% regarding this as ‘about right’ and 31% as too long (Table 11).

Examples of free answer comments given by nurses to the question of ‘What aspects of the training were well done?’ were:

the video was very interesting, especially the look at different faiths.

Nurse

Religious faiths – made me think about utilising faith leaders for advice.

Nurse

Covered several actual problems I have encountered re [regarding] next of kin and end of life/communication

Student nurse

As a student nurse we come across a lot about end of life care and been left to deal with patients’ relatives fears on our own sometimes don’t know what to do just observing.

Student nurse

I would love to be added to a mailing list for future training/or conferences regarding organ donation. I have enjoyed today’s session – thank you.

Student nurse

Specialist nurses for organ donation comments mainly related to the value for training purposes:

Clear and concise. Perfect for those new to ICU.

Very clear with good parts.

Clear, useful, inoffensive film.

Doctors also noted that the video provided a good introduction and covered important aspects simply and clearly. Particularly positive aspects identified were the ‘broad range of opinions/professional participants’, ‘dealing with sensitivity of handling the family’, ‘the belief of different faith groups’, and ‘the video of donor recipients’.

Negative comments mainly related to the length of time required to complete the baseline TPB questionnaire and what was seen by some as the confusing mix of positive and negative statements in this questionnaire. However, some nurses indicated that they would have liked a longer training session, which is envisaged when the full package is employed.

A further indicator of the quality of the presentation was the nomination of the DVD for a national Learning on Screen award, 2014, in which it was highly commended.

In summary, feedback ratings were high for relevance, usefulness and clear presentation by all professional groups. Nurses also commented that it filled an important gap for them, and both SNODs and doctors commented on its value for training.

Methods

Development of the TPB questionnaire for the evaluation followed the manual for TPB questionnaire construction. 120 This requires that the four constructs of attitudes, subjective norms, perceived behavioural change and behavioural intentions are populated with questions of relevance for the study, with each question using a 7-item Likert scale (Appendix 12).

Analysis for the TPB questionnaire comprised five stages: (1) description of the characteristics of participants; (2) exploration of the psychometric properties of the scale; (3) descriptive analysis of TPB variables; (4) exploration of change in TPB variables; and (5) exploration of relationships between TPB variables.

Findings

Exploration of psychometric properties of the theory of planned behaviour scale

The TPB questionnaire was designed to encompass six subscales reflecting the constructs in the TPB as follows:

• attitudes towards organ donation

• subjective norms – normative beliefs about subjective norms

• subjective norms – motivation to comply with subjective norms

• perceived behavioural change – control beliefs

• perceived behavioural change – influence of control beliefs

• behavioural intentions.

An item analysis of each subscale was conducted for the data from the pre-intervention completion of the scale (n = 99) according to the procedure suggested by Anastasi. 121 Specifically, for each subscale, items that were, at the design phase, identified as reverse scored were scored in that way. Corrected item-total correlations were then calculated for items within the scale in order to identify items that appeared to be inconsistent with the scale (defined as a corrected item-total correlation < 0.3). Cronbach’s alpha for the scale including the remaining items was calculated to determine the internal consistency of the scales. Scale scores were calculated by summing the items that made up each scale and dividing by the number of items in the scale.

All subscales achieved a satisfactory level of internal consistency (see Appendix 13). However, lower levels of internal consistency were found for the subscale assessing perceived beliefs about subjective norms and beliefs about the individual’s ability to control the behaviour of talking about organ donation. Overall, the subscales were found to have satisfactory levels of internal consistency and were used as the basis for analysis of the impact of the intervention.

Pre-intervention subscale scores of participants by professional group

The pre-intervention scores for participants in three occupational groups were compared (see Appendix 14). Significant differences between the groups emerged for four subscales:

1. SNODs had the most positive attitudes towards organ donation among individuals from minority ethnic communities, whereas the other nurses were the least positive.

2. All three groups had similar levels of belief about the subjective norms in regard to seeking organ donation from minority ethnic communities. However, both SNODs and nurses expressed high levels of willingness to comply with such norms, with doctors being less willing overall.

3. All three groups expressed high levels of understanding of their role in seeking consent for donation but the SNODs felt more positive that those beliefs would influence their behaviour, probably reflecting their specialist role in family discussions.

4. Intentions to seek consent for organ donation from individuals from minority ethnic communities were highest among the SNODs and nurses and significantly lower among doctors.

Pre- and post-intervention responses to the theory of planned behaviour questionnaire

The follow-up questionnaires were distributed electronically [via SurveyMonkey^® (www.surveymonkey.com)] 3 months following the initial session and sent out in March–September 2013.

The 3-month follow-up responses were received from only 21 of the 99 intervention participants (21%). They comprised 12 nurses of the 77 SNODs (21%) and 9 doctors of the 22 (41%). No SNODs completed the follow-up questionnaire. Scores for initial and follow-up questionnaires for TPB sets of factors are shown for the nurses (see Appendix 15) and doctors (see Appendix 16).

For both bedside nurses and doctors, the only significant change following training was in attitudes towards seeking organ donation from individuals from minority ethnic communities, with both professional groups being more positive in their attitudes following training.

There are several possible reasons for why only attitudes changed following the intervention:

1. The intervention was designed as a three-part package with the DVD designed to address attitudes and knowledge, whereas the interactive elements and planning components were designed to change perceptions of behavioural control and subjective norms. Unfortunately, given the time pressure on staff and the length of the DVD, it often proved impossible to include the more interactive elements and planning components in the time allocated for the training.

2. It is highly likely that the participants had no opportunities to change the way that they thought about and practised seeking organ donation over the 3 months, as seeking organ donation is a rare event and even rarer where consent is sought from an individual from a minority ethnic community.

3. The lack of impact may also partly reflect awareness on the part of participants of the structural barriers to practising what they learned from the intervention.

Correlation of theory of planned behaviour constructs with behavioural intention

To explore the relationship between the components of the TPB and behavioural intentions, Spearman’s correlations were calculated across the participants at the pre-intervention stage (Table 13).

Attitudes towards organ donation and the subjective norm components were predictive of intention to seek consent for organ donation, which further suggests that the impact on practices may increase over time.

Outcomes for consent rates

The PDA, which is completed by the SNODs at each hospital to record the numbers and outcomes for eligible donors, was used to examine the impact of the intervention on family consent rates.

Methods

Results

The numbers of eligible donors and families approached decreased slightly between the pre- and the post- intervention periods for both the white and BAME populations, whereas the consent rate among those approached increased, with the increase being particularly large for the BAME population [(11 percentage points increase in consent compared with nearly 7 percentage points among the white population (Table 14)].

For white families approached, there was an improvement in consent rates from pre to post intervention in both the intervention (three percentage points increase) and the control hospitals (8 percentage points increase). The increase in rates was more marked in the BAME families, with an increase of 12 percentage points (intervention) and 10 percentage points (control). Neither the increase in the intervention nor the increase in the control hospitals was statistically significant (p = 0.15 and 0.11, respectively) (see Appendix 17).

We believe the observed similarity of consent rates in the intervention and control groups among the BAME might have been partly explained by contamination of the control sites, given the considerable movement of SNODs between study sites and our presenting the DonaTE project at regional meetings of SNODs. In addition, a number of local initiatives focusing on increasing donation rates among the BAME community were taking place at hospitals across the study regions.

A post-hoc analysis was therefore conducted to pool the intervention and control hospitals. This showed that the pooled change from pre to post intervention was statistically significant (pooled difference 11%, 95% CI 1.2% to 21%; p = 0.03). However, this must be interpreted cautiously as this is a subgroup analysis and does not allow for either the variations between regions or, importantly, regression to the mean (i.e. the possibility that rates would have increased with or without the intervention because of some other factor).

When the approach rate was modelled, this showed no firm evidence for a difference between the intervention and the control areas (difference in percentages 4.5%, 95% CI –5% to 14%) after adjusting for the time period and ethnicity (Table 15). Overall, the approach rates were lower after the intervention when other factors were allowed for and this was borderline significant (difference –8.4%, 95% CI –17% to 0.2%). The analysis confirmed the difference in approach rates between the white and the BAME donors (difference –8.4%, 95% CI –17% to 0.2%).

In the model for the consent rate, the period effect was quite large, 7.2 percentage points (95% CI –4.0% to 18.5%), but this was not statistically significant after allowing for the variations in regions, the difference between the intervention and the control hospitals, and ethnicity. The interaction effect for the period effect by ethnicity was tested but was not significant (p = 0.54). There was, therefore, no statistical evidence that the period effect varied by ethnic group and so further subgroup analysis is not permitted. The analysis again showed the large effect of ethnicity with –21 percentage points difference in mean rates (95% CI –4.0% to 18.5%; p < 0.001). The multivariable analysis, therefore, while not conclusive, is suggestive of an intervention effect of approximately 7 percentage points with an upper confidence limit of 19%.

Knowledge

Both the systematic review of barriers to organ donor registration among minority ethnic groups in the UK and USA and the focus group study identified low knowledge about organ donation and registering as a donor as a key barrier. This occurred despite recent campaigns in the UK that aimed specifically to promote organ donor registration among minority ethnic groups. People often remained at an initial ‘pre-contemplation stage’, with organ donation not being something they had thought about. However, our research went beyond merely reporting on this situation, with the focus group study leading to a conceptually and empirically grounded explanation of why poor knowledge about organ donation and organ donor registration persists despite a number of targeted campaigns.

Drawing on Schutz’s64 theory of relevance we argued that the existence of large numbers of people at a pre-contemplation stage reflected a situation in which national campaigns to increase knowledge and promote registration as a donor among minority ethnic groups were not perceived as ‘relevant’ as they did not link with individuals’ stock of knowledge or priorities. The messages therefore often ‘passed them by’ as part the general filtering information that we all engage in, rather than being actively rejected. Reasons for a lack of perceived relevance were identified as including a lack of familiarity with organ donation in their home country, a belief particularly among black African men that their organs were ‘strong’ and that organ donation and transplantation were therefore not of relevance to their group, and a general view that transplantation and the need for organs was depicted in the media as largely a problem for white people.

Faith and attitudinal barriers

The focus groups confirmed and extended our understanding of attitudinal barriers to registering as a donor by identifying the significance of particular barriers for different ethnic/faith and age groups. For example, people of Islamic faith were most likely to identify uncertainties, or occasionally prohibitions, arising from their faith. However, there was also some evidence among the younger generations of their awareness of different schools of thought and the confusion that exists between religion and culture, thus identifying the heterogeneity and fluidity of ethnic groups.

Whereas ‘faith’ is often regarded as a general barrier to donation, few focus group participants from Christian, Sikh or Punjabi groups regarded their faith beliefs as a barrier. Indeed their faith often appeared to form a positive force encouraging an altruistic approach that extended to deceased donation, particularly among Sikh and Hindu participants.

Personal autonomy

A theme running through accounts by the younger generation was the increasing emphasis on individual autonomy and choice that was not constrained by religion or by the views of their elders, although this was also often accompanied by a concern to avoid family conflict and upset. These beliefs correspond with the current policy emphasis, which suggests that the family should not be able to overrule donation if the patient is registered on the ODR and has thus made their own wishes known. 15 It also identifies the fluidity and changing beliefs among ethnic groups with attitudes and beliefs of younger people associated with fewer faith/cultural barriers to deceased donation.

Effective interventions

The focus groups identified the importance of identifying the target groups’ stage of change (stage on the donation pathway) and designing appropriate interventions in relation to their circumstances.

For those at an early pre-contemplation stage on the donation pathway, the synthesis identified educational interventions (or combined educational and mass media approaches) as most effective in increasing knowledge and achieving a shift along the donation pathway, with important characteristics being a strong interpersonal element, delivered by trained members of the lay community, and including personal case studies of both donors and recipients from their own communities which address key concerns.

Other groups comprised those who were aware of the message regarding the importance of organ donation and of joining the ODR but had not acted, either because of worries and concerns or because their level of motivation was not sufficiently high to make this a priority. For example, younger men of Indian descent often explained that they had not joined the ODR because they were occupied with work and other interests and that these took priority in their lives. For this group, a small ‘nudge’, including immediate access to registration, may prompt action.

Organisational practices

Intensive care units generally described similar challenges in relation to different ethnic/faith groups, including managing large numbers of family members and visitors, responding to language difficulties and to varying forms of emotional expression and meeting faith-based needs. Local resources, including the opportunity for engaging interpreters, the availability of facilities to cater for the wider family and visitors and the availability of hospital chaplains from different faiths, led to variations between sites in how particular needs were met.

Junior nurses

A widely held view was of a lack of training and preparation to enable them to meet the needs of families from different cultural groups. This was acknowledged to sometimes lead to hesitancy in communicating with families from different cultural groups, especially at a time of considerable emotional distress. Junior nurses also had little familiarity with the organ donation process, as this is not a common occurrence, and they welcomed more knowledge and training in these areas.

Specialist nurses for organ donation

This professional group was established following recommendations of the ODT4 and is experienced in responding to the needs of families from different cultural backgrounds reflecting its specialised role. However, its training does not specifically cover issues of cross-cultural communication or of ethnicity and donation rates/needs for transplantation and there is therefore scope to include training in these aspects.

Specialist nurses for organ donation differed in the opportunities that they had to fulfil their full role. This arose from differences in whether or not an early referral of potential donors was made to SNODs, thus allowing them to begin to establish contact with families and undertake preliminary work relating to the donation process. Such referrals often depended on consultants’ views of the appropriate role and responsibilities of SNODs and, thus, their willingness to involve them at an early stage.

There are known to be differences in whether or not consultants involve SNODs in consent discussions with families; the present study examined the reasons for this in detail. The responses and practices of some consultants indicated that they were very positive about the SNODs and collaborated at all stages (‘acceptors’). In contrast some consultants (‘resistors’) were reluctant to accept the SNODs’ involvement outside traditional nursing and administrative tasks. A third group of consultants (‘intermediate’) were identified as probably in a transitional situation, as they often accepted the involvement of a particular SNOD but not the enhanced role of SNODs in general.

The reluctance of some consultants to involve SNODs partly reflected their own personal characteristics, which included a general resistance to change, particularly among the older consultants, together with a lack of knowledge and appreciation of the training and expertise of SNODs. Other influences related to aspects of the local environment, including the leadership provided by CLODs and the general culture of the ICU. There were also differences associated with the type of donation, with consultants generally being much more receptive to SNODs’ involvement in relation to DBDs following conclusive brain stem tests. In contrast, there was more tension with the role of the SNOD with uncontrolled DCD, when the decision to withdraw cardiorespiratory treatments and the timing of death are less certain and may involve a balance between the needs of patients and families, and those of a potential recipient. In this situation, clinicians often preferred not to involve a SNOD and emphasised their priorities and responsibilities in acting in the interests of the patient and family.

Hospital chaplains

This refers to chaplains of all faith groups with a full- or part-time hospital appointment. Chaplains provide support to patients and families and were often valued by ICU staff in communicating with families but had little knowledge specifically about organ donation.

Hospital chaplains comprise a group in the process of developing as a professional specialty, with discussions taking place regarding specialised training in hospital work. They have potential for an expanded role, particularly in relation to families from minority ethnic groups, with this involving both supporting families during end-of-life care and also promoting awareness of organ donation and addressing perceived barriers within faith groups at a community level.

Bereaved families

The small group of bereaved consenting families from minority ethnic groups who were interviewed were very positive regarding the kindness and consideration shown by ICU staff and the quality of care their relative experienced. Such positive views and experiences are likely to have impacted favourably on their consent to donation. The only critical comments related to the initial time spent by some in A&E departments and limited communication that occurred when this was the entry point. Families also described varying practices regarding brain stem testing, with some families wanting the opportunity to attend but this was not offered. Some families would also have valued follow-up information about how the recipient was getting on being provided automatically for the first few years.

Public engagement

Our primary engagement activity focused on involving patients and the public as a lay advisory group in each of the main phases of our research, with meetings held to review materials and receive comments. In addition, a member of Donor Family Network was a co-investigator and fully involved as a member of the project group, contributing particularly to the bereaved family interviews and training film. However, we went beyond these forms of consultation and collaboration, with the lay advisory group providing the impetus for two public engagement events.

This section describes the setting up and activities of the lay advisory group and the two public engagement events that were held.

Public engagement events

We held two public engagement events that stemmed from the interests of members of our lay advisory group.

Public/policy dissemination

Academic dissemination