Evaluating Facilitators’ Experience Delivering a Stress-Reducing Intervention for Indigenous Women with and without HIV

Indigenous women encounter increased stressful life experiences such as socioeconomic insecurities as well as inequities in health services and outcomes. Tese stress inequities, which stem from the historical and ongoing efects of settler colonialism, also worsen health outcomes for those women living with HIV. As a part of a broader research project on the impacts of stress-reducing interventions for indigenous women, this study examines the experiences of the women who facilitated the intervention. Tis research was conducted to evaluate the impacts of a biweekly stress-reducing intervention conducted in Tunder Bay, Ontario. Te facilitators of the intervention participated in a focus group in 2019 where they detailed the reach, efectiveness, adoption, and implementation of the intervention. Te results suggest that efective stress-reducing interventions should strive to be accessible and inclusive and that doing so can increase program engagement with the service organization hosting the intervention. Further, the results highlight the strengths and challenges of the intervention, including how it helped foster leadership skills and increased indigenous cultural learning among facilitators. Tese fndings demonstrate the strength of community-led interventions and subsequent opportunities for facilitators to grow as leaders. Further, the study highlights how this style of intervention can also encourage participants to engage in other health and wellbeing programs ofered by the community partner hosting the intervention. Tese fndings suggest that interventions aimed at reducing stress among indigenous women facing socioeconomic insecurities, including those living with HIV, are both feasible and benefcial for participants, facilitators, and the nonproft service organizations delivering them.


Introduction
Te ongoing efects of settler colonialism and the embedded systemic discrimination therein contribute to inequities in health service and outcomes among indigenous people living in Canada. Indigenous people also encounter increased health-related stressful life experiences such as socioeconomic insecurities related to housing, food, income, employment, and education [1]. Tese insecurities can translate to a higher incidence of disease, such that one study evaluating mortality among First Nations adults in Canada and nonindigenous showed avoidable causes (e.g., diabetes, substance disorders, and unintentional injuries) contributed to two times higher mortality rates among First Nations persons compared to nonindigenous persons [2]. For indigenous women, these health inequities are often exacerbated by gender-related impacts of intergenerational trauma, where indigenous women experience higher rates of childhood sexual, physical, and emotional abuse including adverse parenting experiences (e.g., neglect, parental substance use, foster care, and the trauma of residential school) [3][4][5][6].
Tese inequities and experiences are also linked to increased rates of HIV and the overrepresentation of indigenous women among those with the disease [2,7,8]. For example, 11.3% of all new HIV infections are among indigenous people despite the fact that they make up less than 5% of the national population [9]. Further, a study of those living with HIV in Ontario found that people living with HIV were more likely to be female or trans women, younger, have lower educational attainment, unemployed, homeless or unstably housed, and experienced housing-related discrimination [10]. Te overrepresentation of indigenous women among those living with HIV is compounded by their increased likelihood to experience stressful life events. Te interplay between HIV and stress can contribute to the progression of mental illness and other chronic diseases [11][12][13][14][15][16] while stress-reducing interventions have been shown to reduce anxiety, distress, and cortisol levels while improving mood among participants [17][18][19].
Given the connection between stress, mental health, and HIV progression [10,20] and the related outcomes among indigenous women, this research seeks to understand how a stress-reducing intervention can be implemented in community settings. As a part of the broader Indigenous Women's Stress Study (IWSS), this research examined the feasibility and utility of a culturally inclusive intervention aimed at reducing stress levels and improving the overall wellbeing of indigenous women in Ontario, including but not limited to those living with HIV. Looking to the experiences of the women who facilitated the intervention over a six-month period, this analysis highlights their perspectives.
While the broader IWSS aimed to describe indigenous women's life stressors and measure changes in stress through salivary biomarkers and questionnaires, this aspect of the project looked to evaluate this culturally inclusive stressreducing intervention. Following the implementation of the twice-weekly intervention in Tunder Bay, Ontario, our team conducted a focus group to elicit feedback from the facilitators. Te objective was to evaluate the intervention to better understand the experiences of facilitators, how they described the program's strengths and challenges, and the extent to which the intervention was seen as being feasible and useful in a real-world setting as a part of community programming through a nonproft service organization. To achieve this, the study broadly employed the RE-AIM (reach, efectiveness, adoption, implementation, and maintenance) Implementation and Evaluation framework [21]. Tis framework is used to evaluate research interventions, particularly prevention and health behaviour change programs and provide internal and external validity perspectives [21].

Study Design.
Te Indigenous Women's Stress Study was a single-arm intervention trial that was delivered biweekly to indigenous women living in Tunder Bay, Canada, over six months. Participants of the intervention were led through the program by a group of facilitators who are themselves the focus of this portion of the analysis. Tese facilitations led the intervention, which was comprised of three components. Te frst component was a 15-minute relaxation technique known as autogenic training which consisted of repeating a set of visualization statements focused on the sensation of feeling calm, warm, or heavy. Following this, participants engaged in the second component, which was an educational and strength-based activity. Examples of the 13 activities included reclaiming ancestral foods, emotional self-regulation and attachment, healthy sexuality and ceremonies, traditional dancing, drumming, the efects of stress, stress management, learning Cree syllabics, and naloxone training. Te educational and strength-based activities were developed alongside community partners, and many were rooted in cultural identity. Tis particular aspect of the intervention is to account for the positive role that cultural continuity can play in health outcomes for indigenous people [22]. Facilitators of the intervention included Indigenous Cultural and Healing Practitioners, Knowledge Carriers including Elders and Helpers, Clinical Psychologist (a guest speaker), and Research Assistants with dual roles as Outreach Workers. Te fnal component of each biweekly session was a 20-45minute guided imagery relaxation exercise which was designed and guided by elders or indigenous healing practitioners. Te exercise included stimulating or recreating perception of sights, sounds, tastes, smell, movement, and/or touch to conjure pleasant images of past or new experiences with the occasional use of cultural tools (e.g., drum or rain stick) to induce relaxation.

Recruitment.
Focus group participants were recruited through two community partners, nonproft service organizations in Tunder Bay. Te indigenous community partner's ofce is located within the nonindigenous organization. Eligible participants for this aspect of the study were facilitators from one of the four periods when the intervention was delivered between May 2017 and May 2019. Eligibility to act as facilitators included local community members who have worked within the organization, which itself worked directly with indigenous women living with HIV. Facilitators of the intervention included staf from the community partners hosting the intervention: a community developer and an elder as well as the lead research assistant. Two former intervention participants were also recruited; one as a facilitator and helper to the elder to foster mentorship and ownership over the intervention, and the other as a research assistant who subsequently began working as an Outreach Worker for the community partners. Te research assistants provided feedback on the design of the intervention and administered the intervention, inviting local guests to facilitate some sessions and collect data from study participants. Te facilitators' roles included designing and running the cultural activities, leading the guided imagery and relaxation sessions, and ensuring group cohesion. As the intervention itself was in indigenous culture for indigenous women, all facilitators were indigenous women themselves, one of which was living with HIV. Tis community-led component is critical to the intervention and has been highlighted in related literature on community programming [22][23][24]. All of these facilitators were invited to participate in the focus group portion of the study, with fve facilitators agreeing to take part.

Measures.
As mentioned above, measuring evaluators' experience highlights the organization-level changes as made clear through the RE-AIM Framework [21]. Typically, "Reach" and "Efectiveness" evaluate dimensions at the individual level, such as who the participants are in the intervention (i.e., reach) and whether the intervention reduced their stress levels (i.e., efectiveness). Tese measures are intended to be collected from participants recruited into the intervention itself. For this facilitator-focused aspect of the study, however, reach is examined at the organizational level and is characterized as the (re)engagement of participants at the study site (the nonproft service organization) beyond the intervention and perceived efectiveness of the intervention to reduce stress among the facilitators' participants. Te "Adoption" (e.g., interest of the intervention by external groups) and "Implementation" (e.g., changes from research to real-world settings such as delivery, use of resources, and costs) components are measured at the organizational level. "Maintenance" refers to more long-term implications, so this aspect of evaluation is not captured in the design of the focus group questionnaire or in the data analysis.

Data Collection.
Data collection occurred through a 1hour focus group in June 2019 with all fve facilitators who led the intervention throughout the 2017-2019 study. Te focus group guide was grounded in the RE-AIM framework. "Reach" questions focused on understanding who participated in the intervention, and how facilitators observed retention and drop-of rates. Te "Efectiveness" aspect included questions relating to facilitators' views on changes in stress for their participants and changes in leadership capacity for facilitators and participants alike. Te "Adoption" component was covered by questions that asked about how the intervention was used by the nonproft organization and whether the facilitators saw any further adoption of some of the activities among their colleagues. Tis mirrored the "Implementation" questions that focused on the actual intervention of activities, how they were facilitated, costs and consistency in delivering the intervention, and any challenges associated with specifc aspects of the intervention. As noted above, "Maintenance" refers to more longterm implications, so this aspect of evaluation is not captured in the design of the questionnaire or in the data analysis.

Data Analysis.
All data gathered during the focus groups were audio recorded and transcribed verbatim. Consistent with recommendations by Onwuegbuzie et al. [25] for analyzing focus group data, we employed Strauss and Glazer's [26] constant comparison and followed their three stages of analysis. In the frst stage, open coding was employed and referred to "the part of analysis that pertains specifcally to the naming or categorizing of phenomena through close examination of data" ( [26], p. 62). In this frst step, we grouped smaller units of our data that expressed similar properties and assigned a code (e.g., friendship, accessibility, and stigma reduction). In the second stage of coding-axial coding-we grouped the codes "back together in new ways by making connections between a category and its subcategories" ( [26], p. 97); italics in original). In the fnal stage-selective coding-a theme was developed to highlight the content of each group. Selective coding is the process of "selecting the core category, relating it to other categories, validating those relationships, and flling in categories that need further refnement and development" ( [26], p. 116). Tis three-stage process also included a team-based analysis where individual codes were collected from coauthors to assess for consistency in categorizations. To further ensure the trustworthiness of the fndings, as the paper was fnalized, we sought out focus group participants who were also coauthors to assess the validity of the analysis. Tis helped to account for the nature of data collected in a group setting, where focus group participants may echo one another or not be as forthcoming with negative responses. Te analysis was also done keeping in mind repeat comments or when focus group participants had either shared or discrepant views.

Results
Guided by the semistructured focus group guide, the coding and thematic analysis examined subthemes across the three main groupings: (1) reach, (2) efectiveness, and (3) implementation and adoption of the intervention. In keeping with the RE-AIM framework, the results highlight the extent to which the facilitators felt that the intervention encouraged growth at the nonproft service organization and re-engagement with participants. Further, the analysis shed light on how the intervention was implemented, the degree to which it could be adapted to real-world community settings, and how indigenous culture was embedded in programming. Finally, the coding demonstrated the impacts of the intervention on the nonproft service organization and the skill development of the facilitators. A summary of the results with select quotes is displayed in Table 1. 3.1. Demographics. Tese facilitators included local women, including a community developer, two research assistants, an elder, and a helper. Te demographic characteristics of the facilitators varied by age and socioeconomic status, but all fve of the facilitators were previously engaged with the nonproft community organization and four were hired as staf. One woman is living with HIV, all the women were caregivers of children aged <18 years. Te in-person focus group was led by a student research assistant, who was not involved in the design and delivery of the intervention or participant-related research activities in Tunder Bay.

3.2.
Reach. Tis aspect of the results' highlights the reach of the intervention in terms of how the nonproft service organization developed further ways to incorporate indigenous culture and re-engagement with participants.

Organizational Community Building.
For the community partner that hosted the biweekly intervention in Tunder Bay, the sessions drove engagement and built trust within the community. Te community organization had also been actively looking to implement changes in response to the health-related calls to action in the Truth and Reconciliation Commission's Final Report (2015). Te organization also saw participants returning for the site's other program oferings and services, including health clinic appointments. Te intervention has also informed the content of other programming and research activities at the organization. For example, the facilitators described including intervention content in the programs they are responsible for delivering. Overall, facilitators echoed each other in confrming that the reach of the intervention resulted in participants joining other programs or accessing other services at the nonproft, something that has been noted as a priority in other organizations ofering indigenous cultural programming [27].

Connections.
Te facilitators also drew attention to the friendships and connections built during the intervention, noting many intervention participants looked forward to seeing other participants both inside and outside of the regular sessions. Facilitators even noted how participants would arrive early and discuss spending time with other participants outside of the intervention.
"We [a facilitator who was a former participant] looked forward to coming to the meetings every other Tursday and coming early just so I can connect with them and see them all."

Stigma Reduction.
Analysis of the focus group transcript also reveals the extent to which the intervention helped to reduce HIV-related stigma among both participants and facilitators. For the facilitator living with HIV, they discussed that leading the intervention and "seeing others (with HIV) living well" was a positive takeaway. All facilitators also observed reduced changes in perceived levels of stigma among their participants without HIV and shifts in comfort among those participants living with HIV. Tese results highlight the benefts of the group dynamic and the connection and trust built amongst community members throughout the course of the intervention.

Reach
Organizational Community Building "I've also seen another facilitator here, (name), use autogenic training or mindfulness for groups, I've seen the crafts come out, I've seen a lot of diferent things." Connections "We looked forward to coming to the meetings every other Tursday and coming early just so I can connect with them and see them all." Stigma reduction "Like it was, it actually eliminated the stigma for the (HIV) positive women, actually, and probably the women who aren't living with HIV."

Efectiveness
Facilitator development "I didn't think (cries) that I could be in that role (of facilitator). So, it was empowering to know that I can do this, and people trust me." Participant development "But there's always a leader (participant) in the group that would show somebody else how to have techniques or (laughing) make fun of me (a facilitator) for not having some." Knowledge carrier development "I would get some of the participants, when we were doing the smudging, to get them to do the smudging on their own." Adoption + implementation Accessibility "Sometimes activities maybe weren't a good ft, or the skill level was, the expectation was a little too high, so we would modify to makes things a little bit easier." Connecting to indigenous culture "I enjoyed doing the outings, getting connected with the land and nature and, for me it makes me feel good when other people are enjoying themselves." Challenges for the nonproft "Um, to be really forthcoming I struggled with the honorarium. . . it was not enough money, and it was also too much. And it's just, it's just how it goes with research." Health & Social Care in the

Knowledge Carrier
Development. An elder, acting as one of the facilitators, also discussed the degree to which the intervention helped with the development of their own leadership skills. Beyond leading activities and ceremonies, they also highlighted their growth as "mentors" and opportunities to pull the intervention participants out of their comfort zone or help to encourage them as leaders in their own right.
"I would get some of the participants, when we were doing the smudging, to get them to do the smudging on their own."

Implementation and Adoption.
Tis fnal aspect of the results speaks to how the intervention was implemented, the degree to which it could be adopted in a real-world community setting, and how indigenous culture was embedded in programming.

Accessibility.
Te facilitators highlighted the accessibility of the intervention and how relatively low barriers made for increased participation. Within this subtheme, facilitators and an elder highlighted a focus on "meeting people where they are" and providing meals and covering any costs associated with activities. Tis included ensuring that the intervention was adaptable, including considerations to physical limitations and skill levels, and looking to participants for direction on activities and any changes needed as challenges arose.

Implementation Challenges.
Facilitators were asked to highlight any difculties or tensions that surfaced while implementing the six-month intervention. Te group noted that the payment provided to intervention participants in the form of an honorarium and the research study setting made for stricter participation rules than typical nonproft program settings.
"Um, to be really forthcoming I struggled with the honorarium. . . It was not enough money, and it was also too much. And it's just, it's just how it goes with research." Te challenges the facilitators faced were typically the result of the research setting. For example, some facilitators noted that the honorarium (e.g., payment to participants) could make group cohesion more difcult because of different levels of engagement. Additionally, some of the external activity leaders like ftness instructors were sometimes uncomfortable working with individuals living with HIV and made this clear to facilitators.

Discussion
Te analysis speaks not only to the strengths and challenges that facilitators highlighted but to the concept of community-led health programming more broadly. Tese results are in keeping with other studies that have relied on facilitator perspectives and the usefulness of gathering observations from individuals that represent both the community and the intervention lens [28,29]. Tis aspect of the evaluation contributes to the broader community-led health and HIV literature. For example, while this study did not use the same framework as Brown et al., [23], it can be examined alongside the "What Works and Why" (W3) framework that Brown et al., establish for peer-led program success, namely: an ability to demonstrate the credibility of their peer and community insights; adaptability to changing context and priorities; and the maintenance of infuence in both community and policy systems (p. 8 [23]).
Looking to the successes of the intervention and how future iterations might adapt it, the facilitators noted how the feasibility of the intervention largely depended on the accessibility of the programming. Te adaptability of the program is in keeping with literature that recommends "meeting people [participants] where they are," to the extent that programming can be made as barrier-free as possible [30]. Specifcally, the implementation should be accessible and culturally safe, both of which were strengths highlighted by the focus group participants. Tese results also highlight how knowledge might be shared between participating service organizations such that adoption challenges are made easier. Further, the community-led element, which was frequently noted as a strength by focus group participants, may also lend itself to increased adoption by other service organizations and partner communities.
Te results also speak to the impacts of the intervention in terms of how the programming afected facilitators, participants, and knowledge carriers. Focus group participants noted how each role, be it leading or participating in the intervention, led to increased leadership capacity and knowledge of Indigenous culture. Te connection between facilitating culturally inclusive interventions or connecting with others through shared culture and positive self-identity and wellbeing has been documented in other work [31][32][33] and speaks to the strengths of the intervention more broadly. Te focus group results also point to the reach of the intervention, both in terms of how the study drove reengagement at the nonproft service organization and how it helped to reduce stigma more broadly at the organization and within the community of participants. Again, this supports earlier studies that highlight the strength of communityled programs and what is needed for them to successfully engage with the community in a sustainable way, including where fexibility in programming and building of existing partnerships is critical [23,28,29,33].
Several recommendations stem from this evaluation. First, certain modifcations were made to accommodate individuals accessing the intervention with diferent skill levels and abilities. Organizations can consider an assessment of abilities upon intake of clients to understand their needs to better plan and develop programming to be inclusive at the onset of implementation. Compared to rigid programming that does not adapt to the needs of participants, future interventions should consider fexibility in programming, including adaptable activities and creating opportunities for participants to provide direct feedback to facilitators throughout the course of the intervention program.
Second, it was observed that in Tunder Bay, study participants could take on leadership roles. Organizations may want to consider including career and training opportunities within programming and research studies for their clients and study participants, respectively. In our study, a facilitator described being recruited into a research study, later taking on leadership roles in the same study, and then working as a research assistant for an intervention in which she had not been recruited as a study participant. Eventually, the person was hired in a permanent position as an outreach worker by the organization. Opportunities such as these may contribute to the early phases of building generational wealth, setting participants on a potential job or career path, or at the very least building a community to support life and career goals.
Finally, future interventions should consider the challenges for facilitators, not just in administering intervention but any impacts on them as members of the community, especially in cases where they may be subject to further marginalization or stigma related to HIV [34,35]. Given the ongoing stigma attached to HIV, organizations ofering these types of interventions to those living with HIV should be cognisant of working with outside contractors, such as ftness instructors or guides of any kind. As noted in our results, an outside contractor made insensitive comments to the group's facilitators upon learning that some of those in the group were living with HIV. Maintaining group safety and comfort should be of the utmost importance, so organizations should consider working with experienced contractors or those with established connections to the nonproft service organization and broader community. Alongside the earlier recommendations, this suggestion can help to ensure that interventions are implemented with care for both participants and facilitators.

Limitations.
Tis study has several limitations. First, the focus group only included facilitators who led the Tunder Bay intervention. As facilitators, the research assistants delivered the intervention four times at the same site leading to greater opportunity for adjustments at each delivery and increased engagement of facilitators. Te intervention was also implemented at three sites in Toronto, though a focus group was not conducted. Perspectives from the Toronto-based facilitators may have yielded diferent results. However, conducting a focus group specifcally for Tunder Bay facilitators was intended to avoid pan-indigenous fndings or overgeneralizing the results. In addition, the facilitators, specifcally the research assistants, were those that recruited study participants, collected data, prepared each session as well as invited, and greeted guest speakers. Teir perspective in the evaluation is critical. In Toronto, there was a high turnover of research assistants as there were more opportunities to move on to permanent positions over the course of the intervention. In fact, this also applied to some participants who were, on occasion, hired at certain study sites and thus could no longer participate in the research study as requested by the sites. Tis contrasted with Tunder Bay where study participants could be hired or volunteer in the study and where facilitators held positions external to the study. Te Principal Investigator was consistently present for all study activities in Toronto, and to establish continuity additional research assistants were not hired. Tus, an evaluation was not conducted in Toronto from the perspectives of facilitators. Lastly, with only fve Tunder Bay facilitators participating in the focus group, we worked with a relatively small sample size.
Another limitation of this study is that one RE-AIM measure is not described. Maintenance of the intervention at several time points beyond the study period was not collected since components of the intervention itself were integrated into regular programming following the completion of the 6-month intervention. It is also worth noting that while we are not presenting the individual-level measures which would provide a more comprehensive evaluation of the intervention; this has been reported separately in the participants-focused portion of the study, which was recently published as [36].

Conclusion
By bringing together the facilitators, including those who had previously acted as participants earlier in the intervention series, this study highlights the strengths of a stress-reducing intervention for indigenous women in Ontario. Tese fndings may be of use to clinicians, nonproft groups, community organizers, and others looking to support indigenous women. Tese fndings suggest how the facilitators navigated the intervention successfully and ways in which future interventions or programs could be designed to expand on the strengths of the Tunder Bay intervention. Te application of these results could help to improve the feasibility and usefulness of future interventions for indigenous women experiencing chronic stress, including for those living HIV.

Data Availability
Te focus group data used to support the fndings of this study are available from the corresponding author upon reasonable request. Te research goals and questions must align with the community's health priorities for whom the original research is intended to beneft.

Additional Points
What is known about this topic. (i) Indigenous women face higher instances of stressful life experiences such as socioeconomic insecurities. (ii) For those women living with HIV, stress can negatively impact their health and HIV progression. (iii) Interventions grounded in ethnic and cultural identity have been shown to reduce stress and improve health outcomes among women living with HIV. What this paper adds. (i) Our fndings demonstrate that interventions aimed at reducing stress among indigenous women facing socioeconomic insecurity, including those living with HIV, are feasible in real-world settings. (ii) Focus group analysis shows that facilitators who deliver the intervention grow as leaders, and participants of the intervention are likely to engage in other health and wellbeing programs ofered by the community partner hosting the intervention.

Ethical Approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Te study was approved by the Women's College Hospital Research Ethics Board and the University of Toronto Research Ethics Board. Te protocol was also reviewed by the research ethics board of coinvestigators on the project which included Lakehead University, Dalhousie University, McMaster University, and at the University Health Network. Informed consent was obtained from all individual participants included in the study.

Conflicts of Interest
Te authors declare that they have no conficts of interest.