An evaluation of community engagement strategies to improve trust and vaccine confidence. : A review of the Primary Healthcare for Travellers Projects in Ireland.

(1)

Communication for Development Two-year master

15 Credits HT 2020

Supervisor: Josepha Wessels

An evaluation of community engagement

strategies to improve trust and vaccine

confidence

A review of the Primary Healthcare for Travellers Projects in

Ireland

Michael Buggle

Word count: 13,784 (includes in-text citations; excludes cover page, contents, and

(2)

Page 2 of 58

Abstract

Public trust in vaccines has shifted over time and at different rates based on vaccine and contextual factors. Despite the known benefit for public health, recent studies across many countries have indicated a perceived ‘hesitancy’ toward vaccines, namely those for the prevention of COVID-19. The decision to vaccinate can be complex, involving psychological and socio-cultural factors that can cause vaccination barriers. Trust or confidence, both in a vaccine and the people behind the vaccine, has been identified as a core element impacting people’s willingness to vaccinate, particularly if a vaccine is new.

Community engagement (CE) methodologies have been recommended in previous reports as strategies to improve trust among populations in health services such as vaccinations. The following study evaluates the applicability of CE strategies, such as the community health worker model, in addressing factors of vaccine hesitancy (VH) by leveraging trusted relationships. The study looks at the approach of a model used to address health

inequalities among the Traveller Population in Ireland as a case study of community engagement methodologies within an ethnic minority population.

Using the Primary Healthcare for Travellers Projects (PHCTP) as a case, this study evaluates how trust as a by-product of CE can play a role in improving vaccine confidence. The study examines elements of CE strategies that can be applied to the Irish Traveller context to address any potential COVID-19 vaccine confidence issues. Interviews with several stakeholders uncover perspectives on the PHTCP model and CE strategies generally and their impact on vaccine decision-making. These stakeholders include primary health

workers of the PHCTP teams in Dublin, and ‘expert’ stakeholders consisting of public health specialists with experience in the field of VH. The study shows that the PHCTPs have built up significant trust levels with the Traveller community and may be effective in improving vaccine confidence in a COVID-19 vaccine in the future.

(5)

Page 5 of 58

Introduction

Vaccine hesitancy

Vaccine hesitancy (VH) challenges the efficacy of immunisation programmes all over the world; however, some solutions are available. VH is defined by the SAGE working group (WG) on VH as ‘a delay in acceptance or refusal of vaccines despite availability of

vaccination services’ (MacDonald, 2015, p.4162). Despite widespread coverage, pockets of hesitant people remain across Europe, impacting vaccines’ ability to guarantee immunity within populations (Butler et al., 2015). Lack of trust has been reported as a factor of hesitancy among minority ethnic populations in the UK, where racism and discrimination have ‘eroded’ trust in the healthcare system (Razai et al., 2021, p.1). In overcoming the challenge, communication for development methodologies such as community engagement (CE) and social mobilisation have been described as having a role in encouraging uptake and ensuring necessary trust in vaccines and those providing vaccines (Goldstein et al., 2015). In an article on the effectiveness of health communication in immunisation strategies,

Goldstein et al. (2015) outline the importance of ‘who’ the messenger is and ‘where’ the message comes from in situations where trust is a key driver of vaccine hesitancy (p.4213). Unfortunately for health service providers, a one-size-fits-all approach to VH is argued to be ineffective. Many public health academics and practitioners argue for the importance of adapting communication and interventions to the given community’s local nuances (ibid.). In the absence of a panacea, and to provide a tool that could help with the process of adapting interventions to the needs of local communities, WHO Europe developed a VH diagnostic tool and intervention framework called Tailoring Immunisation Programmes (TIP) (Butler et al., 2015; World Health Organisation [WHO], 2019). The TIP framework seeks to identify subgroups of populations within regions with low vaccine uptake. The approach focuses on the community’s barriers and motivators to vaccines and recommends combining a multi-disciplinary approach to tailor interventions that leverage these

motivators and minimise the barriers to achieve the greatest uptake (WHO, 2019, p.4). This model has been adopted in several countries in Europe to date, addressing hesitancy in a range of different populations unified by varying characteristics. Some of the groups included those categorised by religion, ethnic background, immigration status, or even

(6)

Page 6 of 58 pregnancy. For example, among an Orthodox Jewish community in the UK, the TIP model was applied to improve vaccine coverage rates in the North London community (Letley et al., 2018).

Irish Travellers

One such subgroup with anticipated sub-optimal immunisation rates is the Irish Traveller (Traveller) community based in Ireland (All Ireland Traveller Health Survey [AITHS], 2010). Despite feeling that health services are generally accessible; Travellers typically engage less with preventive health services, including immunisation, compared with the settled

community in Ireland (ibid.). This lower immunisation coverage was underscored by a previous measles outbreak among the traveller community in 2009 (Gee et al., 2010). The reasons for avoiding preventive health services are evident from the All-Ireland Traveller Health Study (2010), a collaborative piece of research developed in partnership with many Traveller peer researchers and organisations across Ireland. In this study, it was shown that over half of travellers feel discriminated against in a health service setting, and less than half (41%) trust their general practitioners, a significant contrast to the majority population (83%) (AITHS, 2010, p.79).

The Primary Health Care for Travellers Projects (PCHTP) was set up in 1995 to improve the health inequalities faced by travellers, and to improve health service utilisation by

Travellers. The PHCTP developed a community health worker model of health promotion among the community to help engage Travellers in essential health services such as childhood immunisation (Keyes & McCabe, 2005). This primary healthcare model is discussed further in the background section and is referenced throughout as a CE case study.

COVID-19

The issue of VH is currently pertinent for many public health experts, policymakers, and members of the public around the world due to the ongoing COVID-19 pandemic. As many governments and health systems continue to roll out a COVID-19 vaccine in 2021, they will also be cognisant of addressing hesitancy towards uptake to ensure the efficacy of the

(7)

Page 7 of 58 inoculation. Herd immunity for COVID-19 has been estimated at 67% coverage, meaning that two-thirds of people will need to receive the vaccine for it to be effective within the population (Kwok, 2020). Compliance from the population to get vaccinated ‘relies on a personal risk-benefit perception’ that is vulnerable to ‘misinformation’ regarding safety or efficacy (Dror, 2020, p.775). This vulnerability poses a potential threat to the effectiveness of inoculation programmes. It should be noted that the primary concern with the COVID-19 vaccine reported in some surveys, which is consistent with other new vaccines, is the safety and the fear of potential side-effects (Dodd et al., 2021). Conversely, the primary reason for those who are motivated to vaccinate is ‘to protect themselves and others’ (ibid., p.162).

Research objectives and key questions

This study will evaluate elements of CE strategies, including the community-led model of PHCTP, to see if they may be useful in ensuring trust in future COVID-19 vaccines. These learnings may also be useful for other minority groups within Ireland and abroad that may have comparably lower trust levels in vaccination, as is the case in the UK (mentioned above). The study will seek to answer the following research questions:

Primary research question:

1. What elements of community engagement strategies are important in determining trust for vaccines within the Irish Traveller community?

Secondary research questions:

1. Which community engagement strategies are best-suited to improve confidence in a COVID-19 vaccine among Irish Travellers?

2. How might the existing Primary Health Care for Travellers Project (PHCTP) be useful in influencing acceptance of a COVID-19 vaccine for the Irish Traveller community? 3. How might these learnings be useful to COVID-19 vaccine immunisations strategies

for Travellers and other minority groups in Ireland?

To provide additional context, the PHCTP model will be explained further and how its community-led development model is relevant to the field of Communication for

(8)

Page 8 of 58 CE. This grounding in relevant literature on the core topics will provide the basis for the theoretical framework, which is presented as cross-cutting theory within the core fields (CE and VH). Theories of trust both interpersonally and the trustworthiness of communication are presented as models to underpin the analysis and discussion section. The research methodology is discussed including participant selection criteria, study sample, validity, and potential strengths and weaknesses of the method. Results from six semi-structured

interviews will be presented and interpreted in a discussion using the findings from the literature review and theoretical framework to situate the results within the broader field of study. Finally, the conclusion will offer a reflection on the research objective and pose a possible resolution or need for further research.

Background

Who are Irish Travellers?

Irish Travellers are an indigenous population that has been present in Irish culture for centuries. There are an estimated 36,224 Travellers currently in Ireland, spread across much of the country, with the majority living in urban areas near cities (Central Statistics Office (CSO), 2016; AITHS, 2010). Historically nomadic people moving from place to place, Travellers differentiate themselves from the so-called majority ‘settled’ population in Ireland (Relenthford & Crawford, 2013). In recent years, however, the majority (83%) of the Traveller population have moved to fixed residences, although despite this, they remain somewhat socially isolated from wider society in their tight-knit communities (CSO, 2016; Relenthford & Crawford, 2013).

Health inequalities among the Traveller population

Travellers across Ireland face stark health inequalities when compared with the majority settled population. Most notably, Travellers have a significantly lower life expectancy than the settled community, 70 vs 82 in women and 62 vs 77 in men, respectively (AITHS, 2010, p.96). Low life expectancy in males is partly due to high external causes of death, including a

(9)

Page 9 of 58 high suicide rate when compared with the settled community. The disproportionate impact of morbidity and mortality among the Traveller population is explained in the AITHS as being, in part, due to lifestyle risk factors such as smoking and poor diet explained as a coping strategy of hardship and disadvantage endured, as well as other psychosocial factors including distrust in others and general poor mental health (p.160).

Although Travellers self-report as attending health services e.g., general practitioners (GPs), roughly as much as the general population, their confidence or trust in healthcare

professionals is significantly lower (AITHS, 2010). Only 41% of Travellers have complete faith in their GP, compared with 83% of the general population (p.79). One explanation for this distrust in health services is the high level of discrimination faced by Travellers within health service settings and throughout society (p.150). Many advocates in the Traveller community see this discrimination from the broader society as the primary contributor to ill-health within the community (p.165). This lack of trust in healthcare professionals poses a potential communication challenge for the health promotion efforts within the community (AITHS, 2010).

Primary Healthcare for Travellers Project

The Primary Healthcare for Travellers Projects (PHCTP) began initially to serve the Traveller community’s unmet needs in terms of healthcare services and health literacy. In their report on Primary Health Care as a model of good practice, Keyes and McCabe (2005) outline a rationale for ‘special consideration’ for Traveller health solutions, namely because Travellers are a distinct ‘cultural group’ and share different perceptions of health to the general

population. They also recognised the different health issues and health behaviours present among the community, and thus the need for different and innovative approaches to solve them (Keyes & McCabe, 2005, p.15).

Along with capacity building and some advocacy efforts, one of the primary objectives of the

PHCTPs has always been to ‘liaise and assist in the dialogue between Travellers and Health service providers’. This emphasis on bridging the gap between the health service and the

(10)

Page 10 of 58 community has focused on aligning important communications to the community

preferences and capabilities. This includes acknowledging barriers such as low literacy levels and creating health promotion materials that are optimised for Travellers, often

incorporating symbols and shapes in printed materials to make them more accessible. Due to this, much of the work that PHCTPs do is based around interpersonal communication between Travellers and the Traveller Community Health Workers (Keyes & McCabe, 2005).

What do the projects do?

The PHCTPs coordinate and implement several health services within the community and take part in continuous health education training. They spend a portion of their time doing ‘fieldwork’, meeting with community members at their homes, collecting information on the status of the community’s health to tailor their future efforts (Keyes & McCabe, 2005). The PHCTPs are a valuable platform for data collection for external studies on community health and were leveraged during the AITHS in 2010. Since 2015, the PHCTPs have also prioritised improving uptake of childhood vaccinations with young mothers in the community. The PHCTPs ensure mothers have the latest information about immunisation schedules and emphasise the importance of vaccinating for the baby’s sake, but also the good of the community. The PHCTPs also work with the older Travellers in the community to promote the benefits of the influenza vaccine each flu season (Pavee Point, 2017).

Relevance to C4D

At its core, the PHCTP model is a community development project aimed at fostering social change among Travellers toward more positive health behaviours e.g., vaccinating new-born babies against measles, mumps, and rubella or attending a breast check. The PHCTP model is grounded in communication for development (C4D) style theories and principles. As outlined by Manyozo (2012), CE as a concept was adopted in development fields to ‘provide the relevant political economy in which grassroots participation’ can occur (p.154).

Grassroots participation is evident within the PHCTP model and the wider Traveller health structure more broadly. Several local, regional and national representative committees organise projects based on the community’s needs (AITHS, 2010). While CE does not directly

(11)

Page 11 of 58 equal communication for development, it is a core tenet at the centre of many C4D

strategies.

Literature review

Relevant literature on VH and CE have been reviewed to help situate the study within previous research. A series of systematic reviews with a focus on CE strategies in health interventions are assessed. Similarly, relevant findings of community health worker

approaches are discussed to present a well-rounded view of the methodology in relation to health services such as immunisation. Some gaps in the current research have been found and are also discussed.

Community engagement

The literature analysis focused on several syntheses of CE intervention studies that sought to improve a range of health-related outcomes to differing degrees. Given the breadth of CE research, systematic reviews were useful to provide a synopsised understanding of the methodologies in improving health behaviours. Individual public health and social science studies were also reviewed to help provide contextual examples of interventions using CE strategies concerning health service utilisation, vaccine acceptance and/or Traveller populations.

Overall, the literature showed that CE strategies can be effective in improving the success of health interventions, but there is a lack of evidence to show which strategies are most effective and how they achieve effectiveness (Cyril et al., 2015; Haldane et al., 2019; Milton et al., 2011; O'Mara-Eves et al., 2015).

In their analysis of CE studies in disadvantaged populations, Cyril et al. (2015) showed that CE models can be effective in improving health and health behaviours. Effectiveness was shown in interventions that were designed in collaboration with communities and implemented ‘through effective community consultation and participation’, while incorporating the ‘voice and agency’ of the community within the planning, design, and implementation phases of the intervention (p.2; p.7).

(12)

Page 12 of 58 In a multi-method and multi-phase study on CE and health service utilisation, McFadden et al (2018) examined the importance of trust between communities and health services. They analysed the impact of trust in the service provider and service user relationship and put forth a conceptual framework that suggested a strong correlation between trust in services and efficacy of interventions or usage. The study explained the importance of trust in ensuring service utilisation and as a result, improved health (McFadden et al., 2018). In a meta-analysis of public health interventions adopting CE strategies among minority populations, O’Mara-Eves et al. (2015) examined studies using a ‘theories of change’

framework to look at specific CE models as potential ‘moderators of effectiveness’ (p.2). The study shows that CE interventions have a positive impact on a range of health outcomes and behaviours. They found insufficient evidence to point to one strategy over another but identified several themes and provided recommendations based on these, which are discussed further below (O’Mara-Eves et al., 2015).

Best practices

Despite challenges regarding the conclusiveness of evidence favouring one CE method over another, several elements emerged from the literature that may be applied to improving vaccinations among an Irish Traveller population. One primary approach or strategy is the inclusion of community members in the delivery of an intervention. It should be noted that the design and planning phases also benefit from community involvement and collaboration particularly in assessing and understanding the needs of the community (Cyril et al., 2015). Carr et al. (2014) advocate for, in their systematic review of outreach programmes to

improve Traveller health, for the further prioritisation of outreach work for socially excluded populations such as Traveller communities (p.7). They noted that ‘outreach’ can be an effective strategy in reaching communities and defined outreach as ‘alleviating physical and ideological gaps between service and users’ (MacKenzie et al., 2005, p.2, as cited in Carr et al., 2014). Emphasis in this study is therefore placed on literature exploring the delivery of services to improve trust and engagement in health behaviours.

Types of community-led interventions

Community member-led or lay-delivered interventions have shown success in improving health behaviour outcomes by offering additional reliability and empathy to service users. In

(13)

Page 13 of 58 their review of CE strategies in reducing health inequalities for disadvantaged populations, O’Mara-Eves et al.’s meta-analysis of health behaviour, health consequences, self-efficacy, and social support outcomes looked at a variety of CE approaches and their impact on the outcomes of the study. In the study, there was no conclusion as to what type of CE had greatest effect as a moderator of efficacy, however it should be noted that the delivery of services or intervention through lay community members showed the greatest added improvement with some caveats. The reason for this increased improvement, O’Mara-Eves et al. note may be due to ‘credibility, expertise, or empathy that community members can bring’ (p.10). Further to this, they also noted that the studies included in the meta-analyses that were lay member focused had smaller sample sizes which may have resulted in

increased 1:1 time, which has been shown to be more effective at increasing engagement (ibid.).

Lay health educator

Research has shown that lay health worker models have been effective in promoting both child and adult immunisation (Lewin et al., 2010). Focusing on ‘outreach programmes for health improvement’ specifically, in Carr et al.’s (2014) multi-method systematic review into the efficacy of community outreach in improving the health of Travellers, a realist synthesis is developed to examine ‘explanations of how, for whom and in what circumstances

outreach works best’ (p.10). The ‘by whom’ and ‘how’ categories of analysis are particularly relevant for this study on trust as they look at the characteristics of the person delivering the intervention, as well as the nature of how the information or intervention is

disseminated. The impact of these characteristics on trust are discussed in the discussion section.

While lay health educator models have demonstrated to be effective due to added empathy shown, this approach has also indicated to be effective in ensuring cultural suitability of behaviour messages. Cyril et al.’s (2015) systematic review aims to understand which CE methodological approaches showed most effectiveness and which ‘components are

acceptable, feasible and effective with disadvantaged groups’ (p.1). In this study, it is shown that community health worker models can be effective in ‘fostering improved health

(14)

Page 14 of 58 feasibility of interventions as community members can overcome possible ‘cultural and access barriers’ (p.6-7).

It should be noted that through the available evidence in the study, Cyril et al.

recommended accompanying lay health advisors with other strategies to ensure success (2015). The systematic review also highlighted the effectiveness of including community members at the research phase of the intervention, and that this has added benefit in terms of focus group recruitment and research protocol design. Many of the studies included in the analysis adopted a community-based participatory research (CBPR) approach which seeks to connect research and practice through engaging the community in the research design and implementation. Combining community health worker models with CBPR was useful in several studies, with needs assessments being considered as adding positive outcomes (Cyril et al., 2015).

Community outreach and trust

The literature has shown that interventions adopting outreach strategies have success in improving immunisation behaviours among Traveller populations specifically for a variety of reasons (p.37; Lewin et al, 2005). As mentioned, the characteristics of the messenger are important in understanding the level of trust an outreach worker has in a given community. It is pointed out that those of similar ethnicity experience greater levels of trust due to close social ties within the community, however being a Traveller does not automatically earn the trust of the community (ibid.). Carr et al. put forth a suggested trust score typology (see below) based on the studies within their analysis. They suggest that outreach workers ‘already acquainted with families’ as well as professionals with ‘long standing relationships’ have high trust scores, whereas a professional with no previous relationship with

community members may have a low trust score (p.45). This shows that although ethnicity and being a member of the community is important, the quality of the interpersonal relationships are more important in determining the effectiveness of the health outreach intervention.

1. Outreach worker is a highly connected member of the Community, already acquainted with the extended family group – high trust

(15)

Page 15 of 58 2. Outreach worker is a Traveller with no immediate connection to the network –

neutral trust

3. Outreach worker is a professional or semi-professional with a long-standing relationship with the Traveller Communities targeted – high trust

4. Outreach worker is a professional without a prior relationship with the Traveller Communities targeted – low trust

5. Outreach worker is a settled lay person with no prior relationship to the Community – low trust.

Typology of trust in outreach workers working with Traveller populations (Carr et al., 2014, p.45)

In addition to the above, McFadden et al. further explore the relationship of trust between Traveller populations and health service utilisation, they find that trust can enhance CE strategy effectiveness in immunisation promotion (2018). It has been reported in the literature that Traveller communities can have levels of mistrust toward healthcare professionals due to previous bad experiences. This is both common in the UK and in Ireland, where, as mentioned, trust in GPs can be as low as 41% due to a several reasons including feeling of discrimination or no previous relationship (AITHS, P.76; McFadden et al., 2018). Mcfadden et al. (2018) aimed to examine how to increase trust between Travellers and health services and found that many solutions were on the service supply side e.g., ensuring Traveller patients have repeated exposure to the same GP to build a trusting relationship. However, they also found that working with community organisations, and in the PHCTP context, lay health worker projects, this could also help ensure engagement with services. Although the study is limited in detailing how CE can enhance trust specifically, it does advise that any intervention avoid being tokenistic, and ensure commitment to following through with results (McFadden et al., 2018).

Gaps in CE research

Although some efficacy for CE strategies in improving health interventions, such as vaccinations, has been shown, many of the systematic reviews raised concerns about the robustness of evidence across CE studies. In many instances, studies were unable to show

(16)

Page 16 of 58 the impact of the CE model or approach on the intervention itself, but merely that in some cases the interventions with CE elements were more successful (Cyril et al., 2015; Haldane et al., 2019; Milton et al., 2011; O’Mara-Eves et al., 2015; Swainston & Summerbell, 2008). A lack of longitudinal evidence is cited as a reason for lack of robustness in the studies, with many focusing on evaluation at a single point in time (Milton et al.; O’Mara-Eves et al.; Haldane et al.). Swainston & Summerbell’s meta-analysis points out that many studies in its review did not have control data which limits the ‘inferences that can be drawn about effectiveness’ (2008, p.5). Similarly, the research aims of CE health intervention studies did not focus on the specific impact of the CE element of the intervention, and therefore could not show with certainty how CE strategies improved the effectiveness (O’Mara-Eves et al.; Swainston & Summerbell).

In addition to data being non-robust, literature shows a challenge with the evaluation of CE interventions due their complex nature, as well as limitations of methods or tools for assessing the impacts of CE components specifically (Cyril et al., 2015; Haldane et al., 2019; Milton et al., 2011; O’Mara-Eves et al., 2015; Swainston & Summerbell, 2008). Health interventions with CE aspect often ‘operate on non-linear pathways’ meaning that the evaluation is less straightforward compared to ‘simple dose-response relationships’

(O’Mara-Eves, 2015, p.17). It is also said that CE and health processes are ‘influenced by an array of contextual factors’ and therefore it can be difficult to manage these variables (Haldane, 2019, p.21). Due to the complexity and multi-faceted nature, many authors call for more robust measurement ‘innovations’ to assess the efficacy of these interventions (Cyril et al; Swainston & Summerbell).

Vaccine hesitancy

Vaccine hesitancy terminology and discourse

VH research is a relatively new (10-15 years) multi-disciplinary field and discourse that seeks to understand and solve challenges to do with sub-standard vaccination coverage

(Habersaat & Jackson, 2020). The evidence gathered in the field comes from a range of disciplines including ‘psychology, sociology, medical anthropology, social and political

(17)

Page 17 of 58 science, and communication’ (ibid., p.36). Given the breadth of the field, there are

contested topics among interdisciplinary colleagues, including the use of the word

‘hesitancy’, as well as some of the core factors impacting VH. There are also elements that are largely accepted, such as the spectrum of vaccine acceptance and refusal (see fig.1), and the idea that VH is determined by both contextual and vaccine-related factors.

VH is a debated term primarily due to the semantics of the word ‘hesitancy’, as well as some of the core categories of contributing factors that were put forth by the SAGE WG. In a review of the SAGE WG on VH, MacDonald (2015) acknowledge the potential ‘negative’ connotations of the term in addressing a health challenge. When trying to define the term, vaccine ‘confidence’ is a term that was considered by the group and is used by many in the field to capture trust relationships between user and service provider, as well as perceived safety and efficacy of the vaccine (Larson, 2011). However, while ‘confidence’ is useful to explain some elements of lack of acceptance of vaccines, the Sage WG argued that it does not fully cover other factors such as individuals’ perceived risk of disease or issues relating to vaccine access (MacDonald, 2015). In their report the SAGE WG define VH factors as ‘complacency, convenience and confidence’ (see fig.2) (MacDonald, 2015, p.4163).

.

There is a consensus in the literature that VH is vaccine and ‘context specific’ meaning that different vaccines can impact hesitancy to different degrees and similarly, that different populations or communities can have different perspectives for different contextual reasons (Jackson & Habersaat, 2020; Karafillakis et al., 2018; MacDonald, 2015; Larson et al., 2014; Dubé et al., 2013;). This is a significant point for those developing interventions to minimise VH in that, it is necessary to understand the context as well as the specific vaccine

perceptions of a particular population. This is evident in the case of the HPV vaccine in countries across Europe, where hesitancy manifests among parents and young adults due to a variety of reasons. In their systematic review, Karafillakis et al. show that these

determinants of VH can be different depending on the location, e.g., in the UK ‘perceived insufficient and inadequate information’ was regarded as the most common contributor for hesitant populations. This differs in France, where fear of vaccine side-effects was the most common determinant. In Spanish studies, doubt in the effectiveness drove hesitancy (Karafillakis et al., 2018, p.4).

(18)

Page 18 of 58

Fig. 1 “Continuum of vaccine hesitancy between full acceptance and refusal” (MacDonald,

2015)

Vaccine hesitancy determinants and models

As discussed, VH is complex and therefore has a range of determinants that are influenced by vaccines and contexts. As part of their report, the SAGE WG endeavoured to categorise these determinants to aid future research and interventions around VH. The three Cs model (confidence, complacency, and convenience; see fig. 2) was conceptualised to provide a simple yet broad framework to better understand the determinants of VH. Confidence in the model is defined as trust in a) the effectiveness and safety of the vaccine, b) the system that delivers the vaccine, including in healthcare professionals and c) the government or policy makers who make decisions about vaccines. Complacency relates to the perception of benefit versus risk in deciding to vaccinate, with many additional sub-factors impacting that weighing up exercise. Paradoxically, the success of a vaccine can cause greater complacency among individuals, who feel that the threat of the disease has lowered and therefore may be less likely to decide to vaccinate (p.4162). Finally, convenience refers to physical access and ability to vaccinate, as well as less physical elements such as health literacy and vaccine offerings, including information being culturally sensitive or tailored to a group (MacDonald, 2015, p.4162).

(19)

Page 19 of 58 Another approach to categorise determinants, or ‘root causes’ as they are referred, of under vaccination, is the five A’s taxonomy developed by Thomson et al (2016). In response to varying existing models, Thomson et al. developed the approach to provide a more ‘practical’ framework that practitioners can use to identify and address causes of sub-optimal uptake. In their narrative review, they identify previous VH interventions as being ineffective, and acknowledge that the social and behavioural factors that influence VH can be complex. Their review argues that in many instances where VH is identified as the primary cause of sub-standard uptake, other reasons may be of greater importance,

including vaccine delivery. To move away from ‘conceptual’ models to a taxonomy that can be better translated into ‘practice’, the five A’s provides a ‘pragmatic methodology to

identify primary root causes of vaccination coverage gaps, and then to support development of strategies in closing these gaps’ (p.1018).

The above models offer useful parameters for the development of strategies to address sub-optimal vaccination coverage, or VH. In the context of this study, which seeks to look at how CE strategies can have a positive effect on trust of vaccination, it is pertinent to identify areas within these taxonomies that CE can have the greatest added impact for the Irish Traveller population. This will be explored further under the theme of trust in this literature review.

(20)

Page 20 of 58

Fig. 2 “Three Cs” model of vaccine hesitancy. (MacDonald & SAGE WG, 2015)

Vaccine hesitancy and Irish Travellers

Although there are no studies focusing on VH or sub-optimal uptake of vaccines among Irish Travellers in Ireland, some UK studies look at barriers and motivators of vaccine uptake, as well as health service utilisation generally (which includes immunisation). Although there is some evidence to say that immunisation rates can be low among Traveller populations (Dixon et al., 2017), an in-depth qualitative study with members of four Traveller

communities in the UK shows that many of the barriers and facilitators towards vaccines are comparable with the general public (Jackson et al., 2017). This study (Jackson et al.) noted as being the first of its kind, looked at vaccination uptake factors and Travellers. It highlighted that acceptance of vaccines and access to health services are primary motivators and barriers respectively (2017). The study showed that much of the negative or cautionary beliefs around vaccines such as whooping cough or influenza are ‘historical’ and are not present in the younger parents of the community (p.13). The primary challenge identified by respondents in the study was around registering with a GP without a fixed address and booking appointments with offices. The study also underlined the importance of ‘trust in health professionals, particularly GPs’ and other healthcare workers including health visitors (like community health worker model except not lay people). These relationships were viewed to be important influencing factors among Traveller communities but are also important in the wider population. It should be noted that these relationships may be particularly pertinent for the Traveller community as they have a history of not accessing preventive care and of experiencing discrimination in health settings (AITHS, 2010). This is also underscored by the close-knit communities that Travellers live in. The communities are characterised by trust and have a long-standing oral tradition, leaving the community potentially vulnerable to spreading of false or misleading information (Jackson et al., 2016, p.103).

Solutions to vaccine hesitancy

The literature on strategies or interventions to address VH is disparate but does show some core principles that should be considered for future interventions. While no one

(21)

Page 21 of 58 intervention is recommended over another, several best practices have emerged in the literature. Outlined by Habersaat & Jackson (2020), the use of community health workers has been shown to be effective to provide ‘clear and balanced information about

vaccination risks and benefits’ (p.36). In their review of vaccine demand and acceptance, Habersaat & Jackson also note the importance of the ‘face-to-face interaction’ between health workers and individuals on the decision-making process, and how that can be useful in building trust and providing reassurance (2020, p.37). In a ‘review of published reviews’ Dubé et al. (2015) present evidence on strategies to address VH and to improve vaccine acceptance. They note that there is no strong evidence to suggest which strategy is most effective, due to wide-ranging data and disparate studies. Despite this, they highlight the significance of the healthcare provider relationship as ‘the cornerstone of maintaining confidence in vaccination’ (Dubé et al., p.4201). This is echoed in the SAGE WG report (MacDonald, 2015) and in the article by Leask et al. (2014) that identifies ‘provider

interactions’ as being a primary influencing factor on VH, along with social norms (p.2601). They suggest ‘confident recommendations combined with respectful engagement,

narrative, and personalised approaches’ to help reassure parents or individuals when making vaccine decisions (ibid., p.2601). In the context of Irish Travellers, where trust in health care professionals may be lower due to discrimination faced or from previous poor experiences, bridging or improving that relationship is key, and therefore building or rebuilding trust in that dyad.

Theme of trust

Vaccine hesitancy and trust

As discussed, VH encompasses a broad range of factors that can contribute to indecision around, or refusal of, vaccines. Outlined as part of the three Cs model (MacDonald, 2015), confidence is one of the primary factors identified as impacting VH. As indicated in this literature review, the confidence subset of the model is characterised by trust. This can refer to individuals’ trust in the safety of the vaccine itself, their trust in the provider of the

vaccine, their trust in the messenger of vaccine information (which can include a health care professional or community health worker), as well as trust in the policy makers involved in

(22)

Page 22 of 58 delivering the vaccine programme. Low trust in these areas can negatively impact on

vaccine confidence. It should be noted that all factors within the three Cs model are important to consider with respect to VH. However, within the parameters of this study, confidence is particularly applicable given that CE methodologies have been considered as strategies to garner public trust in addressing potential factors of VH (MacDonald, 2015). Equally, within the Irish Traveller population, considering the goals of the PHCTP in bringing health services to the community, trust is an important vehicle for service delivery, including immunisation.

As mentioned, public trust in vaccines plays a significant role in ensuring the success of vaccine interventions. Several key determinants play a role in ensuring that trust is built or maintained. In an appeal to the vaccine community, Larson et al (2011) discuss the

‘confidence gap’ that has enlarged in years gone by due to waning trust in vaccinations. Although primarily concerning childhood immunisations, Larson and colleagues illustrate several possible explanations for this downturn in trust, including fear over vaccine safety, new media and the democratisation of media, and the range of vaccines (p.527). In

identifying determinants of public trust, Larson et al (2011) apply the learnings from a study of environmental-risk communication by Peters et al (1997), which provides an overview of factors that influence the ‘extent to which an individual or institution’ is trusted (Larson at al., 2011, p529.). To analyse perceptions of trust in risk communication for environmental issues, Peters et al (1997) showed strong evidential support for their hypothesis that trust, and credibility are influenced by: ‘perceptions of knowledge and expertise; perceptions of openness and honesty; and perceptions of concern and care’ (p.43). Addressing public trust deficits can be complex, however several studies have pointed to CE strategies for their potential in ‘garnering’ trust. It is also highlighted, as is consistent with all components of VH strategy, that they ‘need be locally tailored’ to the needs and concerns of the local

population (MacDonald, 2015; Larson et al., 2011, p.533).

Community engagement and trust

As mentioned in several studies above, CE can be a useful strategy to ensure public trust in health services including in vaccinations, and in those distributing or recommending them (Kallafinakis et al., 2018; McFadden et al., 2018; MacDonald, 2015). Trust has been

(23)

Page 23 of 58 associated with higher utilisation rates of health services, as well as ‘improved health

behaviours’ (Ozawa & Sripad, 2013, as cited in McFadden et al., 2016, p.2). CE strategies have shown promise in enhancing this trust between service and user, particularly in

marginalised populations such as Irish Travellers, where CE methods can ensure services are delivered in a culturally appropriate way (O’Mara-Eves, 2014; McFadden, 2016). In their multi-phase analysis of trust in health services among the Traveller population, McFadden et al. (2018) identify trust in healthcare professionals as a ‘facilitator’ of trust, and the

importance of this relationship in building confidence in the system.

It is also useful to explore CE and trust within the dyadic relationship as a potential solution to factors of VH. As outlined in the literature review, community health outreach strategies have also shown to be effective in garnering trust in health services including immunisation among Traveller populations (Carr et al., 2014). Trust within interpersonal relationships may have a positive impact in addressing factors of VH such as low confidence. This has been shown in the case of the healthcare professional and service user relationships, as well as with health workers generally (Dubé et al., 2015). Interventions adopting outreach

strategies had success particularly when the deliverer of the outreach was a person with a high trust score, as indicated in the typology by Carr et al (2014). This could be an ‘outreach worker [who] is a highly connected member of the Community’ or a professional with a ‘long-standing relationship with the community’ (p.45). In the context of the Irish Traveller population, the GP relationship can lack trust due to previously mentioned factors, including bad experiences and services being culturally insensitive (AITHS, 2010). In this regard, outreach workers or lay members of the community provide a trust bridge between the community members and important health services such as immunisation.

Conclusion

The literature has shown that CE strategies focused on community outreach have been reported to improve public trust in vaccinations by leveraging trusted sources and focusing on tailoring health promotion to needs of the communities. Several suited strategies have been identified as being particularly applicable to the case of improving trust in vaccines among Irish Travellers. Included in these strategies are those leveraging the interpersonal relationships of community health workers, including peer health educators and outreach

(24)

Page 24 of 58 workers. Overall, the literature shows that CE strategies demonstrate effectiveness in

improving the success of health interventions, but there is a lack of robust evidence to concretely show which strategies/approaches are most effective and how they achieve effectiveness (Cyril et al., 2015; Haldane et al., 2019; Milton et al., 2011; O'Mara-Eves et al., 2015).

The VH literature covers a range of disciplines including ‘psychology, sociology, medical anthropology, social and political science, and communication’ (Habersaat & Jackson, 2020, p.36). There are a variety of determinants of VH that are said to be unique to the context and the vaccine itself. The term ‘hesitancy’ is a debated term and is described by some as laden as it encompasses several broad topics including core determining factors as outlined by the SAGE WG on VH, confidence, convenience, and complacency. Alternative models have been suggested that aim to provide more practical assessments of the contributing factors sub-optimal vaccination, including Thomson’s five A’s taxonomy (Thomson, 2012). Models also exist to try address VH, including the previously mentioned TIP model which uses the COM-B model to assess individual and contextual factors of VH (Habersaat & Jackson, 2020).

Several cross-cutting themes are present in the literature presented on CE and VH. Most pertinent of these for this study on vaccine confidence among the Traveller community is the theme of trust. Combining how the two concepts intersect with trust can help explore how the PHCTP model can impact potential VH among the Traveller community toward a COVID-19 vaccine. From the above discussion, it is shown that trust can play an important role in ensuring vaccine confidence, and in particular the impact of key dyadic relationships on that trust, e.g., with lay healthcare workers or doctors. It is also understood that a primary concern of the COVID-19 vaccine specifically is around its safety and the fear of side-effects, therefore a potential lack of confidence in it. Couple this with the possible low trust levels of travellers in their GPs, as outlined in the AITHS (2010). For the reasons and evidence outlined, the subsequent theoretical framework for this study focuses on the Confidence element of the core theories underpinning it e.g., trust. The subsequent section provides a grounding theory in trust, borrowing from a widely cited model in the field of

(25)

Page 25 of 58 organisational management (Mayer et al., 1995), as well as similar interpretations of trust in the context of environmental risk communication (Peters et al., 1997).

Theoretical framework

To analyse the relationship between the two core concepts of this study, CE and VH, trust will be explored to provide a framework to guide the analysis of results in the discussion section. In this study, CE methodologies have been explored with an emphasis on peer- or lay person-led approaches. The concept and study of VH has also been discussed along with determinants of and strategies to address it. It has been illustrated that a lack of trust can contribute to VH in the form of low confidence as indicated by the three Cs model

(MacDonald, 2015). Conversely, leveraging trustful relationships or trusted sources can help address hesitancy and reassure hesitant people of concerns they may have (Larson et al., 2011). This section explores a theoretical model of trust, and subsequent studies on the topic and the most pertinent elements within trust for the specific context.

Trust as a theoretical framework

There are several trust frameworks that are useful to underpin the analysis of the empirical data in this thesis. Borrowing from the field of organisational management (Hurley, 2006; Mayer et al., 1995) and environmental risk communication (Peters, 1997) respectively, this section will outline useful frameworks with which to better understand the potential benefit of the PHCTP in ensuring trust in a COVID-19 vaccine.

Mayer et al.’s Integrative model of organizational trust (1995) will be outlined first. Although the study was intended for use in an organisational setting, the model has been applied in many varying fields, including communication, sociology, and healthcare (Schoorman et al., 2007). It is a well-cited and robust presentation of trust and trustworthiness, which the other two mentioned studies (Peters, 1997; Hurley, 2006) are drawn from. The purpose of Mayer et al.’s study was to present a ‘model of trust of one individual for another’, the authors clarify in a subsequent article, that the model is also intended for trust in groups or organisations (Schoorman et al., 2007). The integrative model of organisational trust differs

(26)

Page 26 of 58 from previous models, as it seeks to ‘to clarify the relationship’ between two parties and examines the reasons why a ‘trustor’ places trust in a ‘trustee’. They define trust as:

‘The willingness of a party to be vulnerable to the actions of another party based on the expectation that the other will perform a particular action important to the trustor, irrespective of the ability to monitor or control that other party’ (Mayer et al.,

1995).

In their model (see fig. 3 below), Mayer et al. outline key characteristics of the trustor and the trustee, and key ‘factors of trustworthiness’ which include 1) Ability, 2) Benevolence, and 3) Integrity. In this, the trustor is the person who is willing to be vulnerable, and the trustee is the person or organisation that trust is being placed in to carry out the act. In terms of the trustor, their ‘propensity to trust’ i.e., their openness to trusting, will impact the willingness to trust. This openness can be influenced by individuals’ character,

personality type and history with the trustee.

(27)

Page 27 of 58 Based on a thorough analysis of existing literature, Mayer et al. (1995) put forth several propositions within their model. Relevant propositions to this study are (note, these propositions are numbered differently when part of the full set of six in the model):

1. Trust for a trustee will be a function of the trustee’s perceived ability, benevolence, and integrity and of the trustor’s propensity to trust;

2. The effect of integrity on trust will be most salient early in the relationship prior to the development of meaningful benevolence data;

3. The effect of perceived benevolence on trust will increase over time as the relationship between the parties develops.

The first proposition outlined by Mayer et al. concerns the factors of trustworthiness of an individual or organisation. The others highlighted are useful to show how time and exposure can influence trustworthiness. Three core characteristics of a trustee are introduced as the primary factors in determining the trustworthiness of someone. ‘Ability’ refers to the ‘competencies’ or characteristics that make the person influential in a given context e.g., a doctor has training and experience with medical scenarios that leads to the perception that they are able or skilled. ‘Benevolence’ concerns the ‘extent to which a trustee is believed to want to do good to the trustor, aside from egocentric profit motive’ (p.718). This

characteristic centres around asking the true intentions of the trustee and their likeliness to be helpful. Finally, ‘integrity’ considers the principles of which the trustee adheres to, and whether those principles are aligned to those of the trustor. Mayer et al. highlight that these characteristics can influence each other but noted that they are separable. As expected, if someone exhibited strongly in mentioned characteristics, they would be perceived to be very trustworthy. However, the authors clarify that someone should not be considered ‘trustworthy or not trustworthy’, but that there is continuum of trustworthiness, as people may be perceived as benevolent but not able and so on (Mayer et al., 1995, p.721). This proposition is suited to the case of this study, as these trustee characteristics can be useful in identifying trusted parties among the target community. In evaluating outreach as a strategy for Traveller communities, Carr et al. identified ‘integrity’ and commitment to the outreach as important factors in gaining trust in the community (2015). For example, a GP may be perceived to be perfectly able or skilled in relation to advising on the topic of vaccination due to their expertise, however they may also be perceived as unkind or cold

(28)

Page 28 of 58 due to a previous bad experience. Similarly, a peer may be perceived to have great integrity within the community, but without having gone through any formal training may lack competency in providing health advice. These characteristics provide a useful basis for the analysis section as they help provide a picture for why the CE strategies such as the PHCTP can be successful in improving trust in vaccines.

Focusing on communication and perceptions of environmental-risk communication

specifically, Peters et al. (1997) posited three similar factors to those outlined by Mayer et al. In this empirical analysis the factors or characteristics discussed delineate from the individual and are focused on the communication content itself. To that end, the key determinants of trust and credibility are recognised as: perceptions of ‘knowledge and expertise’, perceptions of ‘openness and honesty’, and perceptions of ‘concern and care’ (p.43). There are evident parallels drawn with Mayer et al.’s model as indicated above, ‘knowledge and expertise’ matching up with ‘ability’, ‘openness and honesty’ aligning with Mayer’s ‘integrity’, and ‘concern and care’ comparable to ‘benevolence’. Larson et al. (2011) find the work of Peters particularly useful in uncovering determinants of trust in vaccines. They illustrate the importance of transparency or ‘openness and honesty’ from key vaccine decision makers, and the impact it can have on public trust, citing public PPV vaccine demonstrations in India as an example of ‘inadequate dialogue’ that was conducted

between government and citizens (p.529). Including ‘credibility’ as part of this synthesis on perceived trust is beneficial as it aligns more closely to the information or content of the communication rather than the character of the messenger. This can be useful in the context of vaccine communication, as credible evidence can be presented in favour of vaccination and arguably could supersede a messenger’s ‘trust score’.

Finally, in addition to the above theories and approaches, Hurley (2006) provides some added criteria or ‘mental calculations’ that an individual makes before trusting. Akin to the model put forward by Mayer et al., Hurley comes from a background in organisational management and his model is geared toward manager-subordinate relationships. Despite its epistemological roots, it has some applicable explanations into what influences a person’s willingness to trust. In developing a realist synthesis of how community health outreach works to improve health service utilisation among Travellers, Carr et al. (2014) adopted Hurley’s model as a framework to explain how trusts manifests between workers

(29)

Page 29 of 58 and communities. Hurley splits the model into two parts, the first part concerns so-called ‘decision factors’ which pertain to the trustor’s characteristics e.g., their tolerance to risk, their power position, and ‘level of adjustment’. Tolerance to risk is comparable to Mayer et al.’s ‘propensity to trust’ characteristic, both of which clarifying that not all the determinants of trust come from the messenger side or trustee (Hurley, 2006, p.56).

The second part of the model relates to ‘situational factors’ that are outside the control of the trustor and are predominantly focused on characteristics of the trustee. Hurley’s model borrows from existing literature and is consistent with the characteristics of the two

previous studies e.g., includes determinants such as ‘capability’, ‘predictability and

integrity’, ‘benevolent concern’, ‘alignment of interests’, all of which are comparable with the above models (Hurley, 2006, p.58-59). Some notable additions to this model include the ‘number of similarities’ between the trustor and the trustee (p.57). Hurley suggests that individuals can trust those ‘who appear similar to themselves’ more easily than those who don’t. These similarities can range from personality traits, religious beliefs, gender, or ethnic background. This trait in the ‘decision to trust’ is particularly relevant in the Irish Traveller community given a history of distrust toward members of settled community due to feelings of discrimination and marginalisation. Although Carr et al. show that the Traveller

community do not discriminate within their trust relationships, a Traveller is likely to score higher on the typology of trust if relationship history is not considered (2014).

As indicated in the above, there is a cross-cutting relationship between VH, CE, and trust, particularly within the context of the Irish Traveller community. The literature has shown the suitability of CE strategies in garnering trust, particularly those that emphasise peer-led approaches in delivery of information. Confidence in vaccines and vaccination has been shown to be heavily impacted by the trust between key messengers and decision makers relating to vaccination. Organisational models of trust offer a framework to analyse the propensity of Travellers to trust individual outreach workers implementing vaccine interventions as part of the PHCTP. Whereas Peters et al.’s (1997) model of trust and credibility offers an added element to the framework as it also considers the

trustworthiness of the content of the communication. The above theoretical framework provides a useful basis to evaluate the results of the empirical data collection.

(30)

Page 30 of 58

Methodology

Methodological approach

The following study aims to explore the elements of CE strategies that are important in influencing trust for vaccines within the Irish Traveller community, and how CE strategies such as the community health worker approach may offer use in ensuring trust in vaccines. In exploring the suitability of the PHCTP approach to a hypothetical COVID-19 vaccine acceptance intervention, an interpretive approach was taken to deduce how CE can have success in improving trust in health services such as vaccination. Given the limited existing research on the topic of Traveller health and adult vaccinations, this study is exploratory in its nature and therefore takes a qualitative approach (Merriam, 2015). Qualitative data in this study allows for a review of key themes and patterns across the topics of VH, CE strategies, trust, and COVID-19. As the field of VH is a new and multi-disciplinary field, this study will add to the existing body of research and ideally provide a starting point for further research on the topic of CE and trust in vaccines.

Research methods

The following study follows a single-method approach, completing semi-structured interviews with relevant stakeholders. The study aims to support an initial exploratory analysis into how community health worker programmes such as the PHCTP may work to promote acceptance of vaccinations among communities. The semi-structured interviews gather important perspectives from both PHCTP community health workers (referred to as ‘PHCTPs’) and from experienced researchers and practitioners in the field of vaccine acceptance/refusal1_{(referred to as ‘experts’). Findings from the literature are used to}

triangulate and substantiate findings from both interview groups, and first-hand insight into the priorities and actions of the PHCTPs and their suitability to vaccine acceptance

interventions.

(31)

Page 31 of 58

Semi-structured interviews

Semi-structured interviews were conducted to investigate key perspectives relating to the PHCTPs specifically and the perceived value of the model within the community, as well as VH interventions and elements of CE strategies more broadly. As a method, the

semi-structed interview is appraised for its application to a wide range of research goals (Galletta, 2013, p.45). The flexibility in the design and implementation of the interview method allows for both ‘open-ended and more theoretically driven questions’ which can produce data grounded in the perspective and experience of the participant (ibid., p.45). This was particularly helpful in adapting interview guides to tailor to the experiences of each participant. Unlike structured interviews, semi-structured interviews do not rely on a

prescriptive question guide and allow for more open and flowing discussion (Cook, 2008). To align discussion topics with the experience of the interviewee, four ‘discussion guides’ (see appendix for PHCTP guide and one sample expert guide) were developed to adapt to the experience or expertise of the participants. One template guide for the PHCTP community health workers was created, and three tailored guides for each expert stakeholder were developed, adapting the discussion topics to suit the experience or field of the participant. The sequence of the interviews meant that the interviews conducted with the PHCTP workers impacted the discussion with the expert stakeholders. All PHCTP Traveller

participants were interviewed before the expert stakeholders. This was due to challenges in recruiting PHCTP service-user interviewees i.e., lay Travellers, which is reflected on in the challenges section. The sequence of interviews meant that the experience of speaking with the PHCTPs influenced the discussion guides for the expert interviewees. The topic of trust emerged within the discussions with the PHCTPs, and this was explored further with the experts to understand other contextual examples where trust can play a role in community engagement and trust in vaccines.

Selection criteria and sampling

As mentioned, interviews focused on two primary stakeholder groups, Traveller health workers part of the PHCTPs, and VH clinical/academic researchers. Three participants were

(32)

Page 32 of 58 interviewed in both stakeholder groups (six interviews in total). The selection criteria for the PHCTP health workers was straightforward, interviews were carried out with individuals currently working with a PHCTP in Ireland who had experience of conducting outreach in their community. Experience among the PHCTPs workers varied between 5-15 years. Contact with the PHCTP participants was organised via staff from Pavee Point, a Traveller advocacy organisation based in Dublin. It was important to have Pavee Point as a

gatekeeper in this case as they were able to approach the PHCTP members on my behalf, allowing them to volunteer themselves to participate. This meant that the participants were less pressured into taking part. Pavee Point were also able to make sure the needs of the community were protected, advising against interviewing non-PHCTP staff due to interview fatigue.

In the case of the ‘expert’ stakeholders, desk research was carried out to identify individuals with expertise in the core topics of the study, e.g., CE and vaccine acceptance, Traveller health, and childhood immunisation (given that most vaccination interventions are focused on childhood immunisation). Several individuals were approached directly via email with an ask to participate, including authors from key papers that informed the study. All experts had experience in VH research/interventions, two had experience with the Traveller community via research studies carried out, and one had experience in CE and vaccine confidence in Global South contexts. The term ‘expert’ is recognised as an imperfect categorisation in this study and the connotations of the word were reflected on to avoid biasing the responses given by the non-PHCTP worker interviewees.

The sampling of the interviewees was reflected upon during the study design phase of the project. For the PHCTP workers, it could be argued that this sample was representative of the PHCTP workers in Ireland, given that most of the workers are women and operate in similar social networks. This prominence of female health workers is consistent globally2_.

The sampling of the expert participants cannot be deemed representative, as this stakeholder group as the categorisation of this group is broad and ill-defined.

2_{See WHO:}

(33)

Page 33 of 58

Conducting interviews

Six interviews were conducted in a variety of ways including Zoom video call (3), telephone call (2), and face-to-face meeting (1). It was my intention to carry out all interviews with the PHCTP workers in-person, as face-to-face conversations allow for more fluid and natural conversations (Irvine et al., 2013). However due to COVID-19 restrictions, telephone interviews were preferred, except for one face-to-face interview wearing personal

protective equipment. Telephone calls were preferred by two of the PHCTP workers as they felt more comfortable in that medium compared with a video call. All expert interviews were carried out over Zoom, which allowed me to speak with individuals based throughout the UK. This method was preferred for this group as it was a familiar mode of interview and it meant that visual expressions and cues could be picked up. Informed consent was

captured verbally for all participants bar one where written consent was collected.

Data analysis

As the research goal of the study is to explore how CE methodologies may improve vaccine confidence within the Irish Traveller population, a thematic analysis was carried out to identify key themes stemming from the in-depth interviews. Themes sought to support the suitability of the PHCTP approach to a COVID-19 vaccine acceptance intervention. The analysis took a primarily deductive approach. Outlined in the literature review and theoretical framework, core concepts or characteristics of interventions were identified pertaining to success of CE strategies in improving health service utilisation and vaccine confidence interventions. These have been included in the analysis of data to deduce whether these are substantiated in the context of the Irish Traveller population and COVID-19.

A strength of the thematic analysis is that it allows for triangulation across multiple data sets, and an identification of ‘similarities and differences’ within the data (Braun & Clarke, 2006, p.97). This is particularly useful here, as it allows the study to extract, and compare key themes across the two interview sets as well as with the existing literature. Braun and Clarke (2006) also note its flexibility as an analytical method, allowing for adaptation as the study progressed.

(34)

Page 34 of 58

Validity

In collecting the primary data for the study, in-depth interviews were organised to capture perspectives of stakeholders deemed to be relevant to the aim of the research. Initially, additional participants were identified: Traveller community members (PHCTP service users) and members of the Traveller Health Unit teams in the Health Service Executive (HSE). However due to research fatigue, and workload due to COVID-19 (respectively), they were unable / unwilling to participate. To counteract the impact of this on the validity of the data collected, expert stakeholder perspectives are included. They are triangulated with findings in the results and discussion section, along with evidence from the literature where

relevant.

Ethical or philosophical considerations

There were some ethical considerations of researching Traveller populations, particularly in seeking primary data during a pandemic. As an underserved group in Irish society, it was necessary to be cognisant of any potential biases I as a researcher may have toward

Travellers. It was also important to question whether the research served a purpose for the Irish Traveller community. It was expressed by members of the Pavee Point team that the community has reached a level of fatigue in participating in research studies (indicated through informal conversation with Pavee Point). This was also underlined by the frustration felt by community toward government inaction in response to the recommendations

outlined by the All-Ireland Health Survey (as outlined in the background section). For that reason, it was important that the study could be of value to work that Traveller advocacy organisations, such as Pavee Point, and Traveller specialised units of the HSE do.

Challenges

The initial research methodology design included semi-structured interviews with additional priority stakeholders, including service users of the PHCTPs (i.e., non-community health worker Travellers); and members of the Traveller Health Unit (THU), a specialist branch of the Irish health service that specialises in Traveller health. However, due to external factors, I was unable to capture these perspectives. In the case of the service users (Travellers), I was advised by a member of the Pavee Point team that research fatigue was present among the

An evaluation of community engagement strategies to improve trust and vaccine confidence. : A review of the Primary Healthcare for Travellers Projects in Ireland.

An evaluation of community engagement

strategies to improve trust and vaccine

confidence

A review of the Primary Healthcare for Travellers Projects in

Ireland

Michael Buggle

Table of Contents

Abstract

Introduction

Vaccine hesitancy

Irish Travellers

COVID-19

Research objectives and key questions

Background

Who are Irish Travellers?

Health inequalities among the Traveller population

Primary Healthcare for Travellers Project

Relevance to C4D

Literature review

Community engagement

Best practices

Types of community-led interventions

Gaps in CE research

Vaccine hesitancy

Vaccine hesitancy terminology and discourse

Vaccine hesitancy determinants and models

Vaccine hesitancy and Irish Travellers

Solutions to vaccine hesitancy

Theme of trust

Vaccine hesitancy and trust

Community engagement and trust

Conclusion

Theoretical framework

Trust as a theoretical framework

Methodology

Methodological approach

Research methods

Semi-structured interviews

Data analysis

Validity

Ethical or philosophical considerations

Challenges