Background

Community engagement or community involvement in Aboriginal health research is a process that involves partnering, collaborating and involving Aboriginal and Torres Strait Islander people (respectfully Aboriginal people herewith, see Box 1) to empower them to have a say in how research with Aboriginal communities is conducted. It aims to protect and empower participating Aboriginal people and communities and is an ethical requirement in Aboriginal health research [1]. As with Indigenous peoples of other colonized countries such as Canada, the United States of America (USA) and New Zealand, Aboriginal people in Australia have endured significant health disparities as a result of the impact of colonization and associated race-based government policies [2, 3]. Although research provides an opportunity to address persistent health disparities, there remains concern that Indigenous peoples globally have been ‘researched to death’ without corresponding improvements in health [4]. One way to move forward is by embedding community engagement practices which involve Aboriginal people in all stages of health research from design to implementation and dissemination [5,6,7]. This has been described as an entry point to ‘decolonizing methodologies’ and in turn may ensure the ethical integrity and value of the research by shifting the focus away from researchers and to maximising community benefit [2, 6, 8].

Box 1 Terminology
Full size table

Based on the work of the Council for International Organizations of Medical Sciences, community engagement can be defined as ‘a process of engaging potential participants and communities in a meaningful participatory process that involves them in an early and sustained manner in the design, development, implementation and monitoring of research and in the dissemination of its results’ [5, 9]. Depending on the context, community engagement can include concepts such as community consultation, communication, education, participation, empowerment, collaboration and partnerships [10, 11]. Building partnerships with Aboriginal community representatives enables research which responds to community identified priorities, is guided by Aboriginal people and can facilitate decision making made in accordance with cultural values and beliefs. In addition to gaining insight to community identified priorities, researchers can gain a better understanding of the community under study which in turn informs and transforms the way in which we plan, develop and deliver health care. In a practical sense, engaging Aboriginal health stakeholders (i.e. anyone who has a ‘stake’ in the research, see Box 2) [12] such as Aboriginal health professionals, Aboriginal researchers and Aboriginal community members with lived experience of health conditions can together provide cultural brokerage and practical advice for non-Aboriginal researchers who can take on the role of facilitator [12, 13]. Involving stakeholders in research leads to research outcomes of greater quality and clinical relevance due to their being able to contribute unique perspectives and experiential expertise to the expertise of researchers [14,15,16,17]. This collaborative approach between non-Aboriginal researchers and Aboriginal researchers, services and communities is fundamental to informing, guiding and influencing how Aboriginal health care is designed and delivered and has the potential to help address persistent health disparities between Aboriginal and non-Aboriginal people [10, 18,19,20].

Box 2 Definitions
Full size table

One approach used to incorporate the key elements of community engagement and involve Aboriginal people in Aboriginal health research is to engage a community reference group to provide oversight, input and cultural guidance to research projects [5, 20]. Community reference groups formalize the academic-community partnership by providing a mechanism which emphasizes information and power-sharing, mutual respect and reciprocity between community members and researchers [5, 21]. The roles and responsibilities of community reference groups vary from project to project, yet activities conducted by the group may include reviewing project documents and study materials, participating in community liaison and communicating community concerns, advocating for the rights of research participants, identifying community priorities and providing advice on study design, implementation and dissemination [5,6,7].

While the purpose of a community reference group in Aboriginal health research is to strengthen partnerships, previous research highlights both operational and conceptual challenges in maintaining ongoing community engagement through this strategy [5]. Challenges include unclear power dynamics or a lack of decision-making influence of the group, financial constraints, mistrust of non-Aboriginal researchers, time commitment, limited capacity of members and issues facilitating group discussions in a way that enables Aboriginal community members a strong voice [5, 6]. Failing to address such challenges can lead to weakened partnerships, skepticism about the role of the community reference group or concerns from members that their involvement is ‘tokenistic’ [6, 22]. As a guiding body intended to represent the views of the community, there are also inherent challenges surrounding the notion of representation, particularly when working in diverse populations. The question of whom and how communities should be represented is complex.

Despite being endorsed in policies articulated by national research organizations in Australia and internationally, there remains a lack of Aboriginal community engagement in research. A recent systematic review investigating patterns of community engagement in arthritis studies in Canada, USA, Australia and New Zealand found that the majority of arthritis research projects published do not involve Indigenous peoples at meaningful levels, leading to minimal benefit to the participants and communities involved [3]. Meaningful levels of engagement were defined in this review based on a spectrum of community engagement which includes: informing, consulting, involving, collaborating with; and empowering communities of interest [2, 23]. Furthermore, it has been observed by Indigenous health researchers in Canada that although there is interest from Aboriginal and non-Aboriginal researchers to enact the principles of community engagement whilst conducting Aboriginal health research, there is a general uncertainty on how to do so [2]. This is supported by increasing contributions to literature which describe ideas for improved ethical research and engagement and practices [2, 24, 25]. However, there is a lack of primary research which describes recommendations or practical guidance for health researchers and practitioners on how to engage Aboriginal people and communities in community engagement practices. Therefore, the aim of this study was to provide a practical example of how to establish a community reference group and terms of reference within a Participatory Action Research (PAR) framework [26]. By demonstrating this process within a novel area of Aboriginal health research, we also aimed to present practical recommendations within this context to inform future research groups.

Project context and setting

The ECCO collaboration, addressing a novel Aboriginal health area

The Enhancing Equity, Collaboration and Culturally secure Osteoarthritis care for Aboriginal Australians (ECCO) collaboration is a national inter-professional team of Aboriginal and non-Aboriginal health practitioners, health service staff, and research leaders that was established in response to an unaddressed health care gap; the mismatch between the burden of osteoarthritis and access to appropriate care. Cultural security in health care occurs when services are offered by the health system in a way that will not compromise the cultural rights, values, beliefs, knowledge systems and expectations of Aboriginal people [27]. The objective of the ECCO collaboration is to build an evidence-based model of osteoarthritis care that addresses the needs of Aboriginal people. Establishing a community reference group was a foundational step to building partnerships within the ECCO collaboration as well as between ECCO and external health services and Aboriginal community members.

Research methods

Ethical considerations

This project follows the National Health and Medical Research Council (NHMRC) guidelines for ethical conduct for research involving Aboriginal and Torres Strait Islander People and Communities [1]. Aboriginal community engagement and consultation was at the core of this project. Each of the six values of spirit and integrity, cultural continuity, equity, reciprocity, respect, and responsibility have been addressed in ethics applications approved by St Vincent’s Hospital Melbourne Human Research Ethics Committee [HREC185/19].

Study design

This qualitative study was guided by the principle of cultural security (Box 2), which in the context of research refers to processes that ensure that research is conducted in a manner respectful of Aboriginal cultural values and beliefs [27]. This consideration is essential for ethical purposes [1], improves data quality and ensures that the interpretation incorporates an Aboriginal cultural lens [13, 27]. For example, culturally secure research methods such as yarning, an Aboriginal cultural form of conversation, were used as a data gathering tool [28]. Research yarning is acknowledged as being able to prioritize the lived experience and cultural context of Aboriginal participants [28]. Research yarning aligns with Aboriginal ways of knowing and doing, such as the use of storytelling. It ensures interviews and focus groups are informal, relaxed and requires the researcher to build a relationship that is accountable to Aboriginal people participating in the research [28].

The overarching theoretical framework was PAR [26]. Participatory Action Research is a framework for conducting research and generating knowledge which seeks to situate power within the research processes with those who are most affected by the research [29]. Participatory Action Research frameworks are considered particularly relevant for Indigenous peoples as the approach can help minimize the impact of ‘colonizing effects’ by shifting the power away from the dominant cultural perspective [26]. At the core of PAR is that power be deliberately shared between researchers and the researched by ensuring that those being researched are actively involved in the research process [26]. In this study, we were guided by the principals of PAR to establish a community reference group, who would then be engaged in an ongoing PAR research framework throughout all future Aboriginal health research activities conducted by the ECCO collaboration.

Aboriginal capacity building

Central to this project and the ECCO collaboration more broadly, is building clinical and Aboriginal research capacity in musculoskeletal health. At the outset of this project, we appointed an Aboriginal co-investigator (NR) with a background in nursing and an interest in health and medical research. With no prior research experience, NR worked alongside a female qualitative researcher (PO) with extensive experience conducting interviews with diverse populations and a background in social science. PO supported NR to develop research skills, whilst NR was able to provide cultural guidance on the research activities and a connection to members of the local Aboriginal community. By the conclusion of this study, the ECCO collaboration had appointed a second, male Aboriginal co-investigator (RP) to ensure gender balance within the research team. With a background in sociology and Indigenous studies, RP also participated in training and professional development including qualitative data collection and analysis for the purpose of this project. RP also provided extensive cultural guidance and support with intercultural communication throughout the project. By working collaboratively with Aboriginal co-investigators, the research team ensured mutual benefit and reciprocity between Aboriginal and non-Aboriginal researchers.

Participants and recruitment

To recruit members to our community reference group and to inform the development of the terms of reference, we started by interviewing participants from our target population. Our participants were key informants who represented different stakeholders. Stakeholders were defined as anyone who has a ‘stake’ in the research, in particular those with important knowledge, experiences, expertise or views that should be taken into account [12]. Eligible participants were Aboriginal and/or Torres Strait Islander adults residing in Victoria, Australia who spoke English and represented stakeholder perspectives in Aboriginal health and Aboriginal health research or perspectives of Aboriginal people who experience osteoarthritis. Initially, participants were purposively sampled by drawing on existing networks of project staff. This was augmented by snowball sampling, where enrolled participants recommended additional key informants from within their social, professional and family networks [30]. NR made initial contact with potential participants to gauge their interest in participating in a one-hour research yarn. Interested individuals were sent the study information and had the opportunity to ask questions before providing verbal informed consent to participate.

Data collection

Data were collected in two phases. Phase one involved semi-structured research yarns which aimed to explore stakeholder perspectives of establishing a representative Aboriginal community reference group and to inform the development of a draft terms of reference. Phase two involved a culturally secure consensus focus group with the established community reference group to ratify the terms of reference [31] (see Fig. 1). In phase one, data were collected by PO or PO and NR collaboratively, through semi-structured yarning interviews with participants between March 2020 and November 2020. Participants who expressed an interest in participating were offered the option of face-to-face, phone or teleconferencing yarns at a time convenient to them, however due to COVID-19 restrictions at the time of data collection, only two face-to-face research yarns were able to be conducted. While all participants were offered the option of having an Aboriginal co-investigator present at their interview, three participants declined this offer due to having an existing professional relationship with the first author, signaling that they felt comfortable to proceed one-to-one. The yarning schedule was designed in collaboration between Aboriginal and non-Aboriginal members of the research team and was guided by the NHMRC Model Framework on Community and Consumer Involvement in Research [14] as well as domains commonly covered in terms of reference documents (purpose, authority, decision making, roles and responsibilities, governance and function). These domains were also informed by examples of terms of reference for Aboriginal reference groups previously established by senior Aboriginal researchers within the ECCO collaboration (see example  semi-structured yarning schedule in Additional file 1). Each research yarn commenced with a social yarn, enabling trust to be fostered with each participant before moving into the research yarn [28].  Research yarns lasted between 35 and 90 min, with an average length of 53 min. All participants were given the opportunity to review their transcript upon their request to check that it accurately reflected their experiences. One participant reviewed their transcript before approving it to be included in the final sample.

Fig. 1
figure 1

Overarching methods

Full size image

In phase two, we used recommendations identified from participant research yarns about the composition of a representative community reference group to invite six Aboriginal people to become members of the ECCO community reference group (ECRG). Four out of six invited members participated in phase one of this study. All invited members consented to join the group and to participate in a culturally secure focus group during the first ECRG meeting. The purpose of the focus group was to come to agreement on, and formalize, the draft terms of reference document developed in phase one of this study. RP and PO both facilitated the focus group. The focus group was adapted from the consensus group technique outlined by List et al. [31], however we embedded strategies to enhance the cultural security of this process. Examples of these strategies included; allowing time for social yarning prior to and at the conclusion of the meeting, including an extended Acknowledgement of Country (see Box 2) and ensuring male and female Aboriginal co-investigator (NR and RP) were present and assisting or leading the group. We also emphasized the self-determination of the group. For example, being transparent about the process of the focus group whilst being flexible and reflexive to the input of group members and implementing ongoing changes to meet the needs of the group as they arose. Members were encouraged to attend face-to-face in a private and safe meeting room to allow the members to meet, socialize and build connections with one another. Light refreshments were provided from a local Aboriginal business and each member was reimbursed with a Visa gift card for their time. Following a brief social yarn and introductions, each domain or section heading of the draft terms of reference was read aloud for members to comment and provide feedback on. A process of tacit consent (i.e. implied agreement without being vocally stated) and vocal dissent was used to streamline this process. Any issues, feedback or suggestions raised by members were discussed as a group. Disagreement and dissent were managed by group negotiation, whereby any changes to be made to the terms of reference or function of the ECRG were adjusted until all members were happy to proceed. Pen and paper were also provided to allow members to convey issues in confidence if preferred. The formal section of the focus group ran for 1 h and 15 min, with social yarning continuing after the voice recorder was turned off. In the week after convening the focus group, members were followed up via phone by NR and encouraged to provide additional feedback on the process of the meeting. Research yarns and the focus group were recorded, transcribed verbatim by either PO or an external transcription service.

Data analysis

Data in phase one were analyzed in parallel with data collection using a modified framework approach [32, 33]. Framework analysis is a flexible qualitative analysis method used to sift, chart and organize data in accordance with key issues or themes [32,33,34,35]. In step one, one author (PO) became familiar with the transcripts through the transcription and editing process and by re-reading the transcripts. Transcripts were then uploaded into NVivo (QSR International Pty Ltd. Version 12). In step two, two researchers (PO and RP) coded the transcripts line by line using open coding to identify broad concepts relating to meaningful community engagement in the context of the ECCO research program as well as concepts related to establishing a community reference group. In step three an analytical framework was developed using pre-determined domains relating to developing a terms of reference document as well as concepts identified in step two. The pre-determined domains included; purpose, authority, decision making, roles and responsibilities, governance and function. Concepts identified in step two and included in the analytical framework were; recruitment, size of group, representation, reimbursement and engagement. In step four, one author (PO) applied the analytical framework to all transcripts charting the data into a framework matrix using Microsoft Excel. The framework matrix consisted of one row per participant and one column for each of the analytical framework domains. Using deductive coding [36], important concepts from within each transcript were extracted and mapped to the corresponding participant/domain in the matrix. Three authors (NR, RP and a third Aboriginal PhD student with a background in physiotherapy) applied the analytical framework to a subsection of the transcripts. The charting process allowed the research team to gain a better understanding of important concepts between and within each participant whilst noting similarities and differences. In step five, the data were further summarized and collapsed by describing the main categories within three overarching themes. In step six, the draft terms of reference were developed. The overarching themes, categories and analytical framework informed the section headings of the draft terms of reference. The draft terms of reference were presented to the authorship team comprised of both senior Aboriginal and non-Aboriginal researchers to allow components of the document to be refined and challenged.

In phase two, RP reviewed the focus group transcript to identify consensus outcomes in the focus group data. First, RP read through the focus group transcript to identify any feedback or recommendations that ECRG members raised about the draft terms of reference. For each suggestion identified in the focus group transcript, the main points of discussion and consensus outcomes were recorded. A consensus outcome was defined as the point at which all members agreed upon a change to be made to the draft terms of reference or function of the ECRG. These changes were then incorporated into an updated version of the terms of reference by RP and PO and distributed to the ECRG members to confirm that the changes made accurately represented their input.

Results

The final sample in phase one included 13 participants (eight female and four male), 12 who identified as Aboriginal and one whom identified as Aboriginal and Torres Strait Islander. One invited participant formally declined due to a lack of time. Participants represented a range of ages. While all were residing in Victoria, Aboriginal participants also identified as Wurundjeri, Nimanburru, Wiradjuri, Yamatji, Narrunga-Kaurna, Gooreng Gooreng, Kamillaroi, Wuthathi, Gunditjmara, Wotjobaluk, Ngarrindjeri and Taungurung and one Aboriginal and Torres Strait Islander participant identified as Mabuiag. Three out of 13 participants were considered Elders in their community. Participants came from diverse professional backgrounds including physiotherapy, nursing, general practice, health services management, hospital liaison, cultural safety education, health and medical research and the arts, and represented organizations such as Universities, Aboriginal health research organizations, Aboriginal Community Controlled Health Organizations (ACCHOs) and hospitals. Seven participants also identified as identified as having either lived experienced of osteoarthritis. These individuals experienced osteoarthritis or total joint replacement surgery, or were currently living with a family member who experienced osteoarthritis. Ten participants had previous experience as members of advisory committees or reference groups.

Results from phase one are presented in four tables which represent the three overarching themes identified (Fig. 2): Recruitment and Representation, Purpose, and Function and Logistics . Within each table we highlight supporting quotes for each category mapped to these themes as well as citing where we incorporated these recommendations in the draft terms of reference (Additional file 2). Supporting quotes are indexed by the participant identification number and their gender (e.g. Participant 1, F) with further participant characteristics being omitted due to the specific nature of our participant group and the risk of these features allowing participants to be identified.

Fig. 2
figure 2

Overarching themes and categories identified in phase one interviews

Full size image

We used the findings of phase one to invite six Aboriginal people to become members of the ECRG. Four out of six invited members were participants who we had interviewed in phase one of this study. An additional Aboriginal female with lived experience osteoarthritis and a manager of an Aboriginal health service were identified through networks of the research team. The final six ECRG members (four female, two male) had lived experience of osteoarthritis and joint replacement surgery, a family member of someone who experiences osteoarthritis, a male Elder, a senior Aboriginal health researcher, a physiotherapist and representatives from both regional and metropolitan health services.

Phase 1: Key informant yarning interviews—developing a draft terms of reference

Three themes were identified in the phase one interviews. These were Recruitment and Representation (Tables 1, 2); Purpose (Table 3); Function and Logistics (Table 4). These themes and the categories identified within them are represented in Fig. 2 and each of the three tables below.

Table 1 THEME 1—Recruitment and Representation
Full size table
Table 2 THEME 1—Recruitment and Representation, suggested members to be recruited to the ECRG
Full size table
Table 3 THEME 2—Purpose
Full size table
Table 4 THEME 3—Function and Logistics
Full size table

Phase 2: Consensus focus group—formalizing the terms of reference

Based on the key informant recommendations in phase one, six people from within established community networks were invited to become members of the ECRG and ratify the draft terms of reference were (Additional file 2). A summary of the changes made to the draft terms of reference through the focus group are outlined in Table 5. Following the focus group, participants were invited to review the final terms of reference to check that their recommendations had been accurately incorporated into document. This process revealed that individual follow-ups with ECRG members also provided us with valuable, constructive feedback on this process (as stated in Fig. 3) which was also used to further develop the terms of reference. The final version of the terms of reference is provided in Additional file 3.

Table 5 Changes made to draft terms of reference through the focus group
Full size table
Fig. 3
figure 3

Framework for establishing a community reference group in Aboriginal health research

Full size image

Discussion

This paper describes the ECCO research groups’ practical experiences in the process of selecting, engaging and developing an Aboriginal community reference group and terms of reference by adopting a PAR approach. Our process and recommendations are described in a 10 step-framework for establishing a community reference group in Aboriginal health research (Fig. 3). We demonstrate how calls for adopting community engagement practices by national research organizations [1, 6, 9, 11, 37] may be appropriately operationalized through establishing a community reference group to provide guidance on Aboriginal health research projects. Findings from this study concur with the literature which highlights that successfully building partnerships or community engagement strategies requires health researchers to work in ways which are collaborative, relationships focused and flexible in responding to the needs and priorities of Aboriginal people [2, 10]. Although we describe key practical recommendations that may assist in this process of establishing a community reference group, our framework (see Fig. 3) does not set out to replace the right of Aboriginal people to choose how Aboriginal health care is designed and delivered. Participants in our study often emphasized the importance of self-determination, ownership and decision-making power in establishing a community reference group. Therefore, our framework should instead be used as a blueprint, starting point or guide for research groups to engage Aboriginal health and research stakeholders and community members to shape as they see fit. For example, research groups seeking to establish a community reference group to guide Indigenous health research globally may input project and context-specific information for each of the 10-steps in this framework. This in turn will allow the community reference group to be informed by the values of the group under study and to be shaped to be health problem or be disease specific. This may be particularly relevant for research groups who wish to implement community engagement practices with Indigenous peoples of countries who share a common history of colonization [2]. Moreover, it should be noted that the process outlined in this study is just the beginning or the entry point of what should be an ongoing commitment to community engagement in Aboriginal health or Indigenous health research. Moving forward from the development and establishment phase, community reference groups should be engaged in an ongoing PAR framework [26, 38]. By establishing a community reference group, Aboriginal community members are able to participate in and provide guidance within the PAR cycle at all levels of the research. The PAR cycle revolves around four main steps: plan, act, observe, and reflect [26]. The ECCO research group aims to adopt this approach by planning each ECRG meeting around the PAR cycle which will ensure ongoing engagement of the ECRG into the future.

Undertaking the task of establishing a community reference group can be complex. Failing to plan for or address potential challenges may lead to weakened partnerships or skepticism about the role of the community reference group [6, 22]. Strategies cited in the literature for overcoming challenges to maintain an effective community reference group include recognising the importance of local Aboriginal knowledge and cultural values and beliefs, becoming familiar with local Aboriginal communities and ensuring that the benefits outweigh the time–cost of participating in such a group [10]. The PAR process we have outlined is both time and resource intensive, a challenge that appears to be common amongst community engagement approaches in Indigenous health research globally [2]. Researchers should not underestimate the time it can take to build meaningful relationships in this context. Having the financial means to employ Aboriginal co-investigators to broker relationships was paramount to the success and a strength of our study as it enabled us to build trust, build on existing relationships and receive cultural guidance throughout the project. In return, we were able to begin to build Aboriginal health research capacity in the field of musculoskeletal health, which was a further strength of the study. Research groups who plan to use this 10-step framework of community engagement should consider assigning research staff to a relationship building role, prioritize building the capabilities of Aboriginal health researchers and plan for these underlying costs when applying for funding. This should in turn demonstrate to funding bodies (and to Aboriginal communities and services) the researchers’ commitment to ongoing community engagement. We also support setting realistic and flexible project timelines. A true commitment to Aboriginal health will require research groups to stretch the boundaries of their usual systems and Western research paradigms to allow for flexible, organic approaches to relationship building and community engagement [2, 6].

Despite these potential challenges, the findings of this study suggest that our PAR approach to establishing a community reference group and terms of reference was effective in building the necessary connections required in the context of the ECCO project. In utilizing culturally secure research methods, for example by harnessing the power of ‘word of mouth’ or snowball sampling, we were able to build trust and relationships throughout our recruitment and data collection journey [27, 30]. These relationships enabled us to formally invite members to join the ECRG, many of whom we had interviewed in the process of conducting phase one of this study. We acknowledge that a limitation of this study was having the pre-conceived idea of establishing a community reference group and that barriers may exist for some Aboriginal community members participating in group-based activities, particularly as they require members to be available at a certain ‘place and time’ [39]. Additional modes of community engagement, for example, by engaging multiple community representatives in one-on-one discussions may enable a broader range of community members to participate in community engagement and involvement activities [39]. Additional methodological considerations were the small sample size in phase one and the use of tacit consent in the focus group in phase two. Although our sample size may be considered small, we believe that data collected through the 13 interviews was rich enough to answer our very practical research aims. We acknowledge that the use of tacit consent in the focus group may have influenced group members willingness to voice their opinions at the time. In the future, each participant could be called upon to speak to each of the focus group items in a round robin process. Despite this limitation, the research team were transparent about the process of the focus group while also being flexible and reflexive to the input of group members and implementing ongoing changes to meet the needs of the group as they arose. We also received valuable, constructive feedback from ECRG members in this process and recommend casual ad-hoc, one-on-one ‘check ins’ (as stated in Fig. 3) to give members an opportunity to privately raise any feedback or concerns and to ensure that members are satisfied with the conditions of their ongoing participation. Being flexible and adapting to the input of the ECRG members’ suggestions was an important modification aimed at enhancing the cultural security of our focus group technique.

Conclusion

Current literature suggests that Aboriginal community engagement is integral to any health policy, intervention or Aboriginal health research aimed at improving health care. We have described practical strategies that prioritize PAR, including Aboriginal input and voices in every step of establishing a community reference group and terms of reference. The 10-step framework presented may be especially relevant in guiding research groups who seek to explore novel Aboriginal health research areas by building a program of research from the ground up.