Trust and distrust: Identifying recruitment targets for ethnic minority blood donors

Abstract Background We explore the role of trust, distrust, and the prevailing socio‐political context to better understand why people from ethnic minority communities are less likely to be blood donors compared to people from White communities. Recruiting more ethnic minority donors will enhance representativeness, reduce inequality, and help meet the clinical need to increase the proportion of blood with Ro Kell antigen to treat Sickle Cell Disease (SCD). Study design and methods A 2 (donor‐status: current donor; non‐donors) by 4 (ethnicity: People from Asian, Black, Mixed and White ethnic backgrounds) quasi‐experiment (N = 981) was conducted to examine perceptions of trust/distrust and their influence on willingness to donate blood, within the socio‐political context of the Windrush scandal and Brexit. Results We identified five domains of trust (‘National Health Service [NHS] and staff,’ ‘NHS Blood and Transplant,’ ‘outgroups,’ ‘individuals’ and ‘politics’), and a single domain of conditional distrust domain. Trust across all the domains was lower, and ‘conditional distrust’ higher for ethnic minorities. Trust in ‘individuals’ and ‘NHSBT’ predicted willingness to donate in non‐donors from ethnic minorities and White non‐donors, respectively. Concerns about the Windrush scandal were related to lower political trust. Viewing Brexit as ‘positive for the UK’ was related to lower trust across domains and reduced willingness to donate in White non‐donors through its influence on reduced trust in NHSBT. Conclusion Distinct domains of trust and distrust are identified, and targeting ‘trust in others’ through conditional cooperation is recommended as a strategy to increase donor numbers from ethnic minority communities.

treatment of sickle cell disease [SCD]) benefits. [1][2][3]7 For example, better outcomes for SCD are observed with donor-recipient matching on Ro Kell antigens, which are more common in Black (52%) than White (2%) 1-3 people. However, while demand for Ro antigen blood increases by 10%-15% each year, only 2% of blood donors in England have Ro antigens. 7 Thus, a better understanding of why people from minority communities are less likely to donate blood will inform recruitment strategies that will help realise these potential benefits. 4,[8][9] To address this issue, this article focuses on the one key dimension known to influence interactions with healthcare in minority communities: trust. [4][5][6] 2 | TRUST, ETHNICITY, HEALTHCARE AND BLOOD DONATION While many barriers and motivators for donating blood are similar between minority and non-minority donors and non-donors, [10][11][12][13][14] lower levels of trust in healthcare and donation services could partly explain the lower donation rates in ethnic minority communities. 4,[8][9][10][11][12][13][14][15][16][17][18][19][20][21] A lack of trust in medicine is also a demotivating factor for engaging with healthcare generally, [18][19] specifically for people from ethnic minority communities. 4,[15][16][20][21] Thus, a broader understanding of the role of trust in the context of blood donation should help to uncover new insights and inform recruitment strategies. 18,22 3 | DOMAINS OF TRUST Trust operates across many different domains in life. 23 For example, people express varying degrees of trust in strangers (individuals), [24][25][26] diverse communities, nationalities, and faiths (outgroups), [27][28] physicians, 18,29 and organisations of various types, including healthcare providers and the apparatus-of-states (e.g., police, judiciary, Government). 28 These domains are all potentially important when individuals are considering donating blood. For example, blood donation is a public good, where a few donate blood to benefit all. 30 A significant predictor of public good giving is trust in the generosity of individuals and members of other groups. 31 Furthermore, historical betrayals of ethnic minority groups (e.g., Tuskegee, Windrush) reduce trust in the state (e.g., Government, lawenforcement), 14,[32][33] which may undermine donation decisions, especially if the state and healthcare systems are perceived as linked. 14 However, at present, the existing research on trust and blood donation has focused on a narrow set of domains, specifically trust in healthcare or physicians. 4,[8][9][10][15][16][17] To fully appreciate how trust impacts decisions about blood donation, we need to understand how trust (including trust outside the domain of healthcare) varies by ethnicity and donor status.

| TRUST AND DISTRUST
It is essential to recognise that trust and distrust are separate constructs. While both function to reduce social complexity, 23,34-35 trust creates positive expectations with desirable acts perceived with certainty. [25][26][36][37][38] In contrast, distrust is not just a lack of trust but is linked to feelings that others are active harmful agents who cannot be relied on, leading to distrust, suspicion and alienation. 23 36 Therefore, questions concerning decisions to donate blood need to be commensurate with the stage of the donor career being studied. 37 For people who are inexperienced in a particular domain (e.g., blood donation), decisions are based on behavioural willingness (i.e., an individual's openness to behavioural opportunities and willingness to consider a behaviour); however, as the person becomes more experienced, decisions based on intentions become more important. 38 As a primary focus of this article is to explore the predictive power of trust in non-donors, behavioural willingness is assessed as the most appropriate decision-making index.

| SOCIO-POLITICAL CONTEXT
Perceptions of trust and distrust are influenced by the contemporary cultural and political landscapes. 3 However, previous work on trust and blood donation has not considered the influence of the broader socio-political context. To account for the political context at the time of the study, we examined how perceptions of Brexit and the Windrush scandal influence trust in donors and non-donors.
Brexit concerns the UK's exit from the European Union (EU) following a national referendum on the 23 rd of June 2016. This issue has dominated the political landscape in the UK since, leading to divided public opinion. 39 The Windrush Scandal emerged in 2017, when hundreds of Black Commonwealth citizens, who came to the UK between 1948 and 1973 on their parent's passport, were erroneously classed as 'illegal' immigrants because the relevant documentation was lost. They were denied legal rights, detained, and deported. 34 We test the conjecture that reduced trust in the political establishment is linked to perceiving leaving the EU as 'a positive step for the UK.' [40][41]     where higher scores equate to greater trust, except for trust in individuals, 25 which was responded to on a 4-point scale.

| Willingness to donate
Participants were asked, 'Would you consider donating blood in the future?' yes (1) or no (0).

| Socio-politicalcontext
In terms of perceptions of Brexit, participants were asked: "Do you think Brexit is a positive or negative step for the future of the UK?"

| The structure and dimensionality of trust
Results from the exploratory factor analysis are shown in Table 2  safety. The third factor, 'Conditional Distrust,' represented a belief that the NHS experiments on patients without their knowledge, that wealthy patients receive better care than poor patients, and that people from their ethnic community cannot trust NHS staff. This is combined with a general lack of trust in the police and judiciary. We term this 'distrust' as it reflects perceptions that others will actively harm the patient or person based on their ethnicity and social status (wealth) and, therefore, cannot be relied on. 21 Text S7 for   the full margin table relating to Table 3 Table 4 details two exploratory logistic regression models that examine predictors of willingness to donate in non-donors. The first (col- show that younger non-donors were more willing to donate and that overall, trust in NHSBT predicted greater willingness to donate.

| Predicting donation willingness in non-donors
There were two significant moderating effects of ethnicity on trust, one for NHSBT and one for trust in individuals. The margins for these interactions are in Tables S8 and S9 in Supplementary File S10.
These show that greater 'trust in individuals' predicts willingness to donate for people from ethnic minority backgrounds (Table S8) and that trust in NHSBT predicts willingness to donate in people from White communities (Table S9).  Conditional cooperation occurs when people are aware that other people are cooperating, which motivates them to cooperate. 54 As such, conditional cooperation is a powerful phenomenon that could be harnessed to increase cooperative behaviour, such as blood donation. 55 One way to achieve this is via social media status updates such as-'I have just donated blood' or a blood donation status icon on Facebook, WhatsApp or Instagram, which would inform people that the individual has just donated blood and thereby encourage others to consider donating blood. This approach is effective in increasing opt-in organ donor registrations. 56 Thus, conditional cooperation may be particularly effective at recruiting non-donors as it is a strong social force when free-riding is high, which is the case for blood donation. 57

| Caveats
We showed that 'Trust in Individuals,' not trust in healthcare, predicts willingness to donate in non-donors from ethnic minority communities. However, we must acknowledge that we grouped ethnicity into broad categories, minimising any effect of heterogeneity and wider diversity. Furthermore, the sample sizes for the analyses supporting the moderation and mediation analyses are small, and as such underpowered. 58 Thus, while this work offers a starting point, it needs to be refined to explore trust and concomitant interventions in different ethnic communities and replicated in larger samples and crossvalidated with other methods.

ACKNOWLEDGEMENT
We would like to thank all donors and members of the public and community groups who provided data.

SUPPORTING INFORMATION
Additional supporting information may be found in the online version of the article at the publisher's website.