Sustaining CBPR Projects: Lessons Learned Developing Latina Community Groups

A significant challenge to community-based participatory research (CBPR) is establishing sustainable change as research projects and funding end. Building capacity among community members is one mechanism through which CBPR interventions can be made sustainable. This paper provides a case study reflecting on the development of two Latina community groups in two distinct neighborhoods located in suburban Charlotte, North Carolina, one of the country’s emerging 21st century immigrant gateways. Analyzing our process through the lens of four community group development stages, we critically explore our efforts to build capacity and ensure sustainability. We also assess how community members may or may not become ‘empowered’ through CBPR interventions to continue the efforts once the research ends. By identifying both the successes and failures of our work, we aim to provide a series of guidelines that other CBPR teams might adopt as they work to build capacity among similarly vulnerable populations and construct sustainable interventions at the neighborhood scale.


INTRODUCTION
Community-based participatory research (CBPR) aims to equitably involve academics, health professionals, and community members (those most impacted by the proposed projects) in the research process while recognizing the unique strengths that each brings. While CBPR has demonstrated success in the development and operationalization of research, significant challenges remain for CBPR studies that seek to affect sustainable change in local settings, impact policy, and address broader implications of research outcomes (Laverack & Wallerstein, 2001). In this paper, we offer a case study critically reflecting on the building of two Latina community health and wellness groups. Our contributions to CPBR literature include: a) developing a four-phase CBPR capacity-building process drawing on Arnstein's (1969) ladder of participation; b) expanding Mileski et al.'s (2014) methods of creating an engaging environment for community participation, and c) extending Beard's (2002) concept of a restrictive environment due to the hostile socio-political climate for (undocumented) Latina immigrants in new U.S. immigrant destinations. This paper is process-focused and, as such, does not formally measure the capacity building outcomes amongst research participants.
Establishing the two women's groups was part of a larger research study, funded by the National Institutes on Minority Health and Health Disparities (NIMHD), examining the social determinants of health among Hispanic immigrants in Charlotte-Mecklenburg, NC (for study aims, see Dulin et al., 2012). The research team, the Mecklenburg Area Partnership for Primary Care Research (MAPPR), 1 consists of social and health scientists at the University of North Carolina at Charlotte and researchers and medical providers at Atrium Health (formerly Carolinas HealthCare System). 2 A Community Advisory Board (CAB) with a range of organizational representatives reflective of the project's focus and goals informed all steps of the research and subsequent intervention development, implementation, and evaluation. The intervention planned and implemented for the overall research project consisted of nine neighborhood-based community events focused on health, community wellness and social capital building that engaged community members and stakeholders (Coffman et al., 2017). A critical component of social capital development was the establishment and nurturing of community health women's groups in two suburban neighborhoods heavily populated by Latinos (referred to as East and South in this paper; Figure 1). These neighborhoods were identified through the triangulated use of qualitative and quantitative data, and ground-truthing, including a Photovoice project with local youth (Schuch et al., 2014).
This case study examines our process and the outcomes of that process, focusing specifically on how we engaged a subset of research participants in capacity building activities, with efforts that were sustained into at least early 2021. Both individual and collective capacity were intentionally addressed to facilitate ongoing knowledge sharing and to escalate influence on the part of participants who, over the course of the intervention, moved from being service recipients to intervention volunteers and to a support and learning group, culminating in a community advocacy group independent of the research team. Partners in this part of our study were local Latina residents from the East and South target neighborhoods. As such, they were the intended beneficiaries of the larger project, which focused on identifying social determinants of health and understanding issues associated with lack of access to primary and preventive care. We believe our work adheres to the core value of CBPR research -recognizing and leveraging the unique strengths that each partner brings to the project. While research participants were not directly involved in the writing of this paper, their perspectives, experiences, and reflections, as shared through their close communication with the research team over a period of at least five years and as detailed through observation notes, verbatim transcripts and their own responses to questions and prompts, provide the foundation for our analysis and argument. In other words, their experiences and perspectives are centralized in our work.
CBPR practice calls for greater attention to be paid to the positionality of researchers and how that positionality might impact relationships with community partners, processes of research and engagement, as well as the lens through which findings are interpreted and understood (Muhammad et al., 2015). As longstanding community-engaged researchers trained and working across the disciplinary spectrum, we are keenly aware of the value different points of view and kinds of knowledge bring to both the structuring and analysis of research. Beyond this, the authors of this paper -all core members of the research team -share points of personal intersection with the Latina immigrant of the groups. Three are mothers, three are immigrants, two are Spanish speaking, and one is Latina and the child of immigrants. Each of us was positioned differently along the insider/outsider spectrum for this study and as a result experienced, informed, and were impacted by our work with the women's groups in different ways. Recognizing this openly and as a part of ongoing reflection not only allowed for "an enhanced cultural map for interpreting research data" (Muhammad et al., 2015, p.13) but also offered a way through which we could explore how involvement in the project was reshaping awareness of our personal identities and challenging traditional boundaries of professional practice.

APPLICATION OF PARTICIPATORY MODELS OF SUSTAINABILITY AND CAPACITY BUILDING
The sustainability of CBPR interventions and capacity building go hand-in-hand. It does not just build capacity frameworks and skill sets but also supports the community in using their own data to sustain the results of an intervention beyond the timeline of a project or involvement of the academic research team (Hacker, 2013). Capacity building within communities supports leadership development and acknowledges existing strengths that address problems in communities. Through the process of capacity building, community members develop skills and knowledge that contribute to maximization of continued leadership opportunities such as grant writing, community organizing, and accessing resources (Mileski, 2014;Moll & Gonzalez, 1997;Easterling, 2012  Note: Decennial Census data was used because it is more accurate and closer to reality on the ground at the time project selected target neighborhoods than annual ACS estimates. The mechanics of building capacity within immigrant communities in emerging immigrant gateways remains understudied (Smith et al., 2016), in particular how it relates to successful integration processes, which support the building and on-going health of economically strong, socially and culturally inclusive communities (Migration Policy Institute, n.d.). There is clear acknowledgement that if immigrant groups and the broader communities of which they are a part are to integrate successfully, capacity building and leadership must be developed and facilitated to emerge within immigrant communities themselves.
Community development and planning literature that focuses on participation as a process by which we ultimately create better and transformational solutions offers a potential way forward. Reflections on levels of engagement offered in Arnstein's (1969) work that discusses participatory processes as steps on a ladder, and later work by Rocha (1997) reflecting in a similar manner on a ladder of empowerment, are important foundational pieces from these fields of study that inform leadership and capacity building. Further, Beard (2003), using her research from Indonesia, discusses how citizens learn to engage in radical planning in restrictive socio-political environments. Important lessons from that context can be applied to the experiences of recent immigrants in the US, particularly for those settling in emerging gateways. Such immigrants find that they have little or no representation or voice in decisionmaking processes and are often the focus of fluid, and at times fractious, reception (Harden et al., 2015;McDaniel & Smith, 2017). Beard explains radical planning in the context of highly restrictive political environments as the outcome of social learning processes over an extended period that creates a sense of collective agency and action. This became our inspiration for imagining a model of building sustainability and capacity into our CBPR social determinants of health project. Drawing on the tenets of sustainable CBPR and capacity building -and the work of Arnstein (1969), Rocha (1997), and Beard (2003) -the Latina women's group had the following implementation phases: 1. Participation in/using a service provided 2. Joining a community interest group set up and facilitated by researchers 3. Transitioning to leadership roles within the group 4. Becoming independent of the research team

CHARLOTTE AND EMERGING IMMIGRANT GATEWAY CONTEXT
Charlotte, NC, is one of the nation's major emerging immigrant gateways, a city where, among other factors, rates of foreign-born growth outpace the national average by at least double and the dominant immigrant group is of Latin American origin (Singer, 2015). Prior to 1990, Charlotte had very few immigrants, but has experienced rapid foreign-born growth since that time (Smith & Furuseth, 2006). As of 2015, 15.7% of the city's residents were foreign-born, with 45.8% of whom identified as Hispanic or Latino (US Census). Just over half of all Latino residents in Charlotte are foreign-born (53.1%) and they are primarily Mexican (38.2%) or Central American (33.2%). About 28% record having an income below the poverty level and 41.2% of those over the age of 25 had less than a high school education.
As a relatively new gateway, Charlotte is at the nexus of local and national tensions about immigration. Embracing policies such as 287(g), which supports the collaboration of local law enforcement with the federal government to enforce federal immigration policy, are counterbalanced by a growing number of community-based services working to better understand and meet the needs of Hispanic newcomers (McDaniel & Smith, 2017;Schuch, 2020). The city is at once both hostile and welcoming -a context that produces fear and instability, particularly for immigrants who have only recently arrived or have yet to secure documentation. An additional characteristic of new gateways is that immigrant settlement skews towards the suburban, particularly in suburbs that are aging and experiencing socio-economic downgrading. MAPPR identified two such neighborhoods (one in East Charlotte and one in South Charlotte) on which to focus our intervention efforts (Figure 1).

METHODS
This paper presents a case study of a CBPR intervention study. A case study is defined as "an in-depth, multifaceted investigation, using qualitative research methods, of a single social phenomenon" (Feagin et al., 1991, p.2). Case studies can both reflect and inform broader phenomena. A key component of participatory research is critical reflection of the process (Baum et al., 2006). Below, we address methods specific to the establishment and development of the women's groups as well as our reflections at each stage of capacity building. Opportunities for reflection as researchers and partners occurred during weekly research team meetings and monthly community advisory board meetings. In between these meetings, there were opportunities to connect with community partners and discuss not only the progress of implementing the groups but also their own reflections on engaging with the project and community members more broadly. Research team meeting notes served as documentation of reflections.
Research activities prior to establishing the groups that inform this paper are outlined in other publications (Schuch et al., 2014;Schuch, 2017;Coffman et al., 2017;Simmons et al., 2018) and Figure 2. All research materials and protocol were approved by the Atrium Health Institutional Review Board (IRB). The East neighborhood (Figure 1) was the first to receive the health interventions, a series of events with free primary care and preventative services, and information about health and social services. At the East health events, we asked participating community members on their evaluation survey if they were interested in being part of a community group. Those who demonstrated interest were invited to meetings during the third, fourth, and fifth health events to discuss their topics of interest. In the South, recruitment for the community group started with an existing parent group at the school hosting the health events and continued at the health events. The different recruitment approaches of the two groups reflects our CBPR approach through which community partners helped identify the most appropriate and effective way to initiate group development in their respective areas. Focus topics were also identified in a participatory process among interested individuals. Initial meetings were open to all interested community members. Eventually, only women chose to participate.
As part of the process of ending our direct, 16-month engagements with each of the women's groups, we held focus groups to obtain feedback from participants on their experiences. The data used for this paper consists of participant observations (at approximately 40 meetings for each group), reflective team meeting notes, field notes, and focus group transcripts. The textual data were reviewed by each of the researchers separately and subsequently deliberated together. Following this discussion, we summarized and grouped our data into the four capacity building phases we established. Once categorized, we assessed as a group the extent to which the documented data aligned with how capacity building progresses. Table 1 offers an overview of the development of each group, organized by the four implementation phases.
During phase 1, community members were engaged in health events and interest groups. While the recruitment of members of the two groups varied, the remainder of the process was planned in similar ways. During phase 2, the research team was responsible for all roles, including facilitating meetings, putting together and printing meeting agendas, scheduling presenters, coordinating events, providing snacks and waters, and childcare. In a process facilitated by the research team, community group members decided on a group name and broadly defined goals. In phase 3, the mission of the groups became clearer and expanded beyond receiving services. Participants started taking on roles and tasks with the goal to build leadership and ownership. The intentions were for roles to transition through conversations about group independence and significant tasks delegation. The goal of phase 4 is for the research team to step away and the group to continue independently.

RESULTS
This section contains lessons learned from the two groups in each of the four capacity-building phases. The authors intentionally integrated results of the East and South groups -rather than presenting them separatelybecause we believe doing so helps us critically compare and contrast the developments of both groups.

PHASE 1: PARTICIPATION IN/USING A SERVICE PROVIDED
Although attendees had a chance to ask questions at the neighborhood-based health events, these were mostly a one-way flow of information and resources. To be successful with establishing a group, significant resources must be put towards this goal both by the research team and the community partner(s). However, in the earliest stages, that balance tilts towards the research staff whose time and project resources are focused on establishing partnership with community, recruitment and start up efforts in this initial phase. With that said, community partners play a critical role, especially in determining distinctive place-based approaches.
The engagement process was different between the neighborhoods which allowed for increased time in the South vs. the East, and greater opportunity for decision-making informed by pre-knowledge of the local community and its specific needs. The leadership from the school in the East connected the research team with the leadership from the school in the South five months prior to the implementation of the South intervention. This allowed time to build a more robust relationship and better understand the needs of the parents and school and integrate those into the health events developed in the South (see Figure 2). In the East, initial meetings took place during the third, fourth, and fifth health events, which was deemed convenient because participants were already going to be there. However, this limited the time available to complete research-related activities (i.e. survey completion and health screenings) and to interact with the social service providers that were in attendance. In the South, a core strategy that worked was to tie into existing programming at the school, for instance by attending the English as a Second Language (ESL) Breakfasts. There, MAPPR received names and phone numbers from parents interested in being a part of a community group. Meetings were scheduled both in the morning and afternoon to accommodate different schedules, and now with contact information, reminder calls could be made. Still, initial attendance was very limited. While the parent group at first was small (the first few months, only a few people showed up), our initial engagement created a snowball effect of more women being invited to join, both by researchers and active community members. The latter model -recruiting community group participants prior to the health events from an existing group and having members help recruit new members -proved more effective in securing long term engagement.

PHASE 2: JOINING A COMMUNITY INTEREST GROUP FACILITATED BY RESEARCHERS
In the East group, we attempted to connect participants to existing programs or services (e.g. at the neighborhood YMCA), but financial and transportation challenges proved

PHASES EAST GROUP, HISPANOS EN ACCIÓN (HEA) SOUTH GROUP, HISPANOS UNIDOS (HU)
1: Participation in/ using a service provided -Community members participated in health fairs.
-Interest meetings took place during the health fairs and were mostly attended by women in their early thirties.
-Interest groups were started at a local elementary school before the health events. These groups were with parents (mothers) whose children attended the school.

3: Transitioning to leadership roles within the group
-After a few months, HeA participants started to bring their own updates about things happening in the area. They also became more involved in gathering information (e.g. about spaces to meet) and reporting back to the group. In our regular meetings, we continued to reflect on what was going well, where we could make improvements, and how to transition leadership more to the group. -Three women expressed interest in being trained to become Zumba instructors and a MAPPR researcher started working with them on how to teach Zumba.
-HU quickly expressed interest to become involved with projects and events outside the group. In fact, part of their early established mission was to engage in community service. -During meetings, the group developed roles to facilitate meetings, including communications between group members, providing healthy snacks, note taking, childcare, inviting guest speakers, coordinating volunteer opportunities. They established roles such as the "meeting facilitator" and "note taker". The note taker also created her personal resource binder in which she files the pamphlets, contact cards of the presenters, etc.

4: Becoming independent of the research team
-There was interest and attempts to reach this stage by participants, but the group was unable to continue. Challenges included interpersonal dynamics among the women and uneven commitment.
-HU planned their own public event.
-HU continues to meet at the YMCA and have an active online platform for disseminating wellness-related events and information. barriers. Participants decided to start a walking group instead and, subsequently, a research team member offered weekly Zumba classes. Hiring a bilingual Hispanic undergraduate who lived in the neighborhood and previously participated in a Photovoice project conducted in earlier phases of this study (Schuch et al., 2014) was part of an intentional effort to make the groups more community driven. The women expressed they wanted the childcare to be more organized, that people should show up on time, and that we recruit more members. We asked who would take on roles, e.g. note taking, sending out reminders, bringing snacks and water, and putting together agendas. We offered training in case people were unsure how to do these tasks. The importance of establishing a shared understanding of expectations cannot be understated. In the East, the process of developing a Memorandum of Understanding (MoU) led to a document to which we could refer in times of disagreement among group members. Another critical lesson was to recognize the difficulty of maintaining a core group when all participants are volunteering to be there. We rationalized that the groups were for the benefit of the women, much like participating in an exercise class, and that compensation for this part of the work was not necessary (we compensate for more traditional research activities such as surveys and focus group participation with gift cards). Upon reflection, compensation would have been appropriate for core members who worked as hard as the research team to develop a model that would be useful. With "financial resources to pay its leaders reasonable honoraria for their time-consuming efforts" (Arnstein, 1969, p.221), we might have been able to develop a more consistent team. East participants often had other jobs or competing demands that restricted their participation in or beyond regular meetings.
In the South, the connection to the school and the prospects for having a positive impact in the life of their children appeared to be a clearer benefit of participation that motivated people to stay engaged. It was effective to have existing members distributing flyers to recruit new group members, e.g. at local stores, businesses, apartment complexes, neighbors, and to other community members and family. Having the Facebook communication platform (rather than relying on calling or texting every individual separately) which we had been doing in the East was also an effective strategy. From the mission, vision, and description, a MAPPR research assistant developed a pamphlet for the group to edit and revise. This pamphlet was utilized as an outreach method for further involvement with the community, for instance to recruit potential new members and to hand out at community events to inform people about the group.
Despite the successes outlined above, the establishment of Hispanos Unidos in the South was not without its difficulties. Changes in the meeting time and location (due to the school being closed during the summer) hindered members in attending all meetings. We also encountered a few personality conflicts. While developing the mission, vision, and description of their group, members had difficulty with the decision-making process; some were hesitant to share their opinion whereas others disagreed with one another about what the group should focus on. As certain individuals started to emerge as leaders, others were not always supportive. It was easier in the first phase to accept researchers as leaders than negotiating the peer process of sharing leadership responsibility. Some of these dynamics continue, but we have also addressed them directly with the group, which has helped us partly overcome them or prevent larger animosity.

PHASE 3: TRANSITIONING TO LEADERSHIP ROLES WITHIN THE GROUP
The anticipated progression to independence did not take place in the East despite efforts that were intense and similar to the efforts in the South where an independent group emerged. Participants appeared to be hesitant to take on roles, though there was always an expressed interest in becoming less dependent on MAPPR. We stressed that this would necessitate the group coming together and identifying their respective strengths and interests, but dividing tasks remained difficult. One of the women volunteered to send out mass text reminders, but then others said they did not receive them. People also signed up to take turns providing childcare, but this turned out to be difficult when people were late or did not show up. Though participation was not consistent enough to assign roles, some members reported noticing physical and mental health benefits to their involvement.
To understand these outcomes, we speculate on differences in the way the two groups were initiated with the initial anchor provided by the school in the South but not in the East. The infrastructure of support organizations that emerged in the South was largely absent in the East. That the intervention and establishment of the group in the South followed those in the East translated into greater experience on the part of the researchers to facilitate and mitigate conflicts (the East was an important learning experience), and social dynamics and leadership skills of the women involved. In both groups, MAPPR attempted to establish partnerships with outside organizations but this was more successful in the South, which was a key ingredient in the South group's sustainability. For example, the YMCA became involved with the group during the summer of 2014 and provided a meeting location while the school was closed, funded HU t-shirts, and helped set up YMCA memberships for HU members. A local church also offered a location for Hispanos Unidos events and HU members volunteered at the church (Figure 3). Moreover, we recognize the importance of participants taking advantage of outside trainings, workshops, community events, leadership opportunities. Four members of HU became certified Health Promoters (promotoras) through the American Heart Association. They subsequently presented about cholesterol and high blood pressure during regular HU meetings. The group also decided to volunteer at school, church, and YMCA events. We were unable to obtain the same degree of community participation from the East group.

PHASE 4: BECOMING INDEPENDENT OF THE RESEARCH TEAM
The goal from the beginning was to create sustainability through independence. Hispanos en Acción (HeA) did not make it to stage four. After meeting for over a year, one of the core participants suggested meeting weekly for about 30-45 minutes to chat with each other. This required little organization and could take place in different places (around the school, at a local restaurant, at someone's house) with the goal to offer women support and mental health benefits. They still wanted to meet to exercise outside the school in the park while their kids played in the playground. The women seemed excited about the new, more informal format. For us, it was a small success to see one of the group leaders come up with a new idea and 'pitch' it successfully to the group. With that said, subsequent check-ins throughout the year to see if members of the group had met indicated that they had not. Here, a lesson to share is for the research team to not only be critically reflective of their process throughout but also to accept when it might be time to step away. A group that is not functioning well despite repeated attempts is not serving the women, the community, or the goals of the research project. Indeed, continuation of such might introduce unintentional and negative impacts for the participants involved.
In contrast, between January 2014 and January 2015, 37 individuals had been to at least one HU meeting, with a contact list of 25 individuals and about 15 regular members. Meeting attendance fluctuated between 10-15 people. Having several core leaders who were able to put time and energy into the group, and take advantage of other training, community, and leadership opportunities outside the group, was an important component of the group's success and sustainability. As the group became more established, it gained visibility in the community. The group was approached by school leaders to become parent advocates and they also participated in city-wide events such as parades and 5K's. Several women in the group also signed up for English classes through the YMCA. In addition, Hispanos Unidos participated in the Village HeartBEAT (Building Education and Accountability Together) Challenge, a Mecklenburg County Health Department Initiative to enhance community health. Through this program, 11 HU members increased their physical activity, lost weight, and/or stopped smoking.
A major milestone took place on January 26, 2015. Hispanos Unidos held their first public event, in partnership with MAPPR and the YMCA. An immigration-themed presentation had already been listed on our calendar for June 2015 but Obama's November 2014 immigration executive order about the expansion of Deferred Action for Childhood Arrivals (DACA) and the introduction of Deferred Action for Parental Accountability (DAPA) sparked an interest to move this presentation up. The event took place in the school library and over 50 people attended. MAPPR supported the event by scheduling two immigration lawyers knowledgeable about the policy changes and recruiting childcare volunteers. The YMCA provided snacks and additional support for childcare. The HU women created a flyer and distributed it across the area at local businesses, apartment complexes, and even the Mexican consulate. Though MAPPR and the YMCA were present to help coordinate, HU members prepared the seating arrangements, welcomed and directed guests, and brought balloons and signs.
As of writing, HU continued to meet at the YMCA which served as the anchor institution for the group. Members have continued to develop leadership skills and confidence, and partnerships with the American Heart Association and health promoters. Per request, the YMCA Community Development Director put together a 16-meeting leadership development course for them. On September 23, 2016, they officially launched their revised mission, vision, values, focus, and logo. They created an organizational structure with clear roles for each member, including focus areas of health, education, and resources. Hispanos Unidos now has a Facebook page 3 which aims to inform the Charlotte Hispanic community about resources and events in the area with strong continued activity and participation.

DISCUSSION
The importance of engaging directly with immigrant community members in the context of an emerging immigrant gateway cannot be overstated. In Charlotte, for instance, immigrants, Hispanic and otherwise, are also not always welcomed by others in the broader community, leading to social distance and isolation. Compounding this is that we face a limited (but growing) service infrastructure for immigrants and non-English speakers. Resources catering to newly arrived immigrant groups are often scarce and there are fewer opportunities for connection with pre-existing co-ethnic communities. Many immigrants live in suburban areas that may not be close to or well-connected to needed resources (Singer, 2015).
We tried to address some of these challenges by forming groups within the context of suburban neighborhoods with high and growing Latino immigrant settlement and meeting within, and bringing resources to, those spaces. However, even then, barriers to connection and access may be substantial. Programs such as 287(g) enacted in Charlotte-Mecklenburg allow local officers to enforce federal immigration laws and expedite deportations (Mathema, 2012). This, along with raids (retenes), created a lot of fear among the participants and broader immigrant community. As such, even though Beard's framework for community engagement was conducted in Indonesia's "restrictive political environment" (Beard, 2002, p. 15), her work transfers across geographies because living in the US can also be a restrictive socio-political climate for Hispanic immigrants, particularly those who are undocumented. Our community groups tapped naturally into women's roles in their families and communities as advocates for their family's well-being and sharers of information. One of first-generation migrants' main motivations for coming to the US is typically to provide better opportunities for their children; consequently, we have benefitted from organizing around issues that benefit their children and seeking out family-friendly events.
A focus on capacity and leadership building were critical components of our successes in this project. Mileski and colleagues (2014) identified, "[f]ive methods that may create an open and engaging environment for local residents to serve as leaders in their community. First, provide opportunities for developing leadership skills. Second, create a kind and warm meeting space -an inclusive environment. Third, identify the common goal of the group while respecting diversity of individuals. Fourth, sincerely care about the members and their communities, both in the meetings and outside of the meetings. Finally, implement the decisions of the group members to the best of your ability." (Mileski et al., 2014.153) We followed these recommendations. For example, participants played a central role in devising the groups' name, mission, vision, and goals. The process of cocreation makes people more invested in the group. This may mean the group takes on topics or activities outside the original research topic. We demonstrated genuine interest in participants' wellbeing and supported them in decisions. This meant being flexible in our roles, which is central to relationship-building. Just as community members became co-researchers, our roles changed over time (from researcher to facilitator to Zumba instructor to child caretaker) depending on the developments and needs of the groups.

RECOMMENDATIONS
Our work suggests an expanded list of methods to ensure not only capacity and leadership building but also sustainability. First, build in time to reflect. The two researchers who were most directly involved with the groups debriefed after every meeting and the whole research team used regularly scheduled meetings as a place for on-going reflection. Each meeting was therefore informed by the previous one. In addition, we had monthly community advisory board meetings made up of key stakeholders/service providers who helped us reflect and steer the process. This collaborative and iterative process is central in CBPR.
Second, building on from our earlier comments about the importance of addressing researcher positionality, we argue that it helps to have researcher team members with whom participants can easily and comfortably connect. As we shared above, the four key research personnel engaged in this process are all women, three are mothers, three are immigrants and the fourth is a child of immigrants, two are Spanish-speaking, and one is Latina. These shared identities were points of relationship building between the research team members and participants. These connections offered opportunities to challenge traditional researcher-participant boundaries. For example, when needed, researchers served as child minders for Hispanic en Accion and Hispanos Unidos meetings and brought their own children to project events which contributed to a dynamic of shared motherhood. Over the course of the project, both the participants' and the researchers' positionalities shifted, moving towards each other, as participants developed expertise in running an organized group and lobbying for their needs, and researchers gained insight about the women's lives and conducting action research with a vulnerable immigrant population. Although we recognize that the blurring of boundaries has been critiqued for jeopardizing objectivity, as participatory researchers we recognize that all research is subjective and that valuable lessons can be derived from projects where positionalities and identities overlap and shift.
Third, address tensions early, as they arise. Conflict is an important part of the group process but needs to be dealt with sooner rather than later. There must be a desire to want to see each other and socialize with each other in order for people to be motivated to attend meetings and events. Modeling this behavior can help the group resolve issues by themselves. When participants are unhappy with the research team, this should also be addressed immediately and is an important learning moment.
Fourth, providing opportunities for leadership early on sets the tone for this to continue in the future but understanding that the process of fully transitioning leadership takes an extended period. Plan on this process taking at least a year and prepare accordingly in terms of budget and human resources.
Fifth, take the time and make the effort to incorporate the complexity of people's everyday lives and experiences into the project. In other words, work to understand the micro-scale dynamics and interplay of the social, spatial context and individual situations of the participants and, where possible, address issues and barriers revealed. This includes recognizing people's work and family situation (offering child care, meeting early evenings), economic and transportation circumstances (not everyone can afford to attend or drive to events across the city, so carefully consider locations and offer transportation options, participation incentives, remuneration for time). Not all participants will have a shared background, so talking about this as a group can make people more mindful of each other and offer their own solutions (e.g. carpooling).
Sixth, seek out at least one community partner, ideally physically located in the community, that is demonstrably supportive of the group by, e.g., offering space to meet or child care, showing up at meetings for encouragement, co-organizing events, going out of their way to make the group feel welcome and valued. Connecting with other agencies that will continue to support the community group is key to sustainability.
Seventh, offer to teach people how to facilitate meetings, organize events, etc., rather than assuming people already know.
Eighth, find a central way to communicate. Facebook played a large role in allowing us to stay in touch, share resources and events, communicate updates, post pictures, etc.
Ninth, be patient. Expect to see fluctuations in participation and 'success' over time. Capacity building and engagement is not necessarily a linear process. Include discussions about how to define and measure success in ways that are meaningful to the community participants, not (only) the research team.
Lastly, do not force success. It is possible that after all efforts by participants and the research team, the group does not continue after the research project ends. That is not always a 'failure'. Building capacity to the point of independence may not always be possible considering the context in which participants live. Perhaps the main issue around which the group organized is may no longer as pertinent, or Core members may move away. Individuals or parts of the group, however, might continue to use the skills and knowledge gained to form or participate in other groups or initiatives. They may continue to be involved in their community in new ways that build on what they learned through their research participation. For instance, HU took the lead in partnering with the local YMCA, which became the host of the group. They are not engaging with the YMCA as service recipients but as partners and they would not have been able to do that without their engagement in our research project.

CONCLUSIONS
This paper reflects on the development of two Latina health groups in two suburban neighborhoods located in Charlotte, NC. Analyzing our process through the lens of community group development stages, we consider how, through capacity building, community members may or may not become 'empowered', may or may not continue to engage in the project without the collaboration of external groups, or continue their group independently of the research team. With that said, without the formation of and capacity building within the women's groups, our work on this project would not have achieved its sustainability goals. While we would have successfully utilized CBPR to inform the building and implementation of a series of community wellness fairs, we would not have seen the lessons learned from the fairs (and the community grounded research that informed them) translated into longer term community engagement in health-related practice and advocacy. We would not have seen the women in those groups transition from community members and research participants to research project partners and become empowered as leaders of community health efforts beyond the scope and structure of the social determinants project. In addition to targeted measurement of capacity building, we recognize that future research should seek to better understand and evaluate how similar CBPR engagements impact access to services, health outcomes, and quality of life for participating community members, as well as outcomes for partner organizations supporting this work.