A Community-Academic Partnership to Plan and Implement an Evidence-Based Lay Health Advisor Program for Promoting Breast Cancer Screening

Abstract

Despite a growing body of evidence concerning effective approaches to increasing breast cancer screening, the gap between research and practice continues. The North Carolina Breast Cancer Screening Program (NC-BCSP) is an example of an evidence-based intervention that uses trained lay health advisors (LHA) to promote breast cancer screening. Partnerships that link academic researchers knowledgeable about specific evidence-based programs with community-based practitioners offer a model for increasing their use. This article describes a partnership between CrossWorks, Inc., a community-based organization, and the UNC-CH Lineberger Comprehensive Cancer Center in planning and implementing an evidence-based program for promoting breast cancer screening among older African American women in rural eastern North Carolina communities. We used in-depth interviews to explore the relationship of the partnership to the activities that were undertaken to launch the evidence-based program.

Breast cancer is the leading cause of cancer death in U.S. women¹ and while White women have higher breast cancer incidence than other racial or ethnic groups, Black women have higher rates of late-stage breast cancer and breast cancer mortality.^2,3 For women ages 50–69 and at average risk for breast cancer, biennial screening mammograms can detect breast cancer early and reduce mortality through earlier treatment. Starting at age 40, women should discuss mammography screening with their health care providers to determine if individual risk and patient preference warrant earlier initiation of screening mammography or continuing screening after age 69.⁴ Although research has tested different approaches to increasing mammography use in diverse populations, yielding a growing body of evidence-based programs (EBPs),^5,6 the knowledge and skills needed to find and use these EBPs is low,⁷ particularly among public health practitioners without graduate or post-graduate training.⁸

Providing practitioners with information on EBP is generally recognized as a necessary but insufficient step in closing the gap between research knowledge and clinical or community practice.⁹ Practitioners may also need support with selecting and adapting EBP to the specific contexts of their populations and settings, and with implementing and sustaining these programs.¹⁰

Relationship-building partnerships that join researchers knowledgeable about EBPs with practitioners working in communities can help propel research-based evidence into practice.^11-15 In these partnerships, practitioners, as community partners, contribute in-depth knowledge of needs, preferences, and resources in the intended population. Researchers, as academic partners, contribute expertise in selecting and adapting EBP.^16,17 Partnerships between academic researchers and community practitioners have potential to increase community members’ and organizations’ capacity to implement EBP.^18-19 Despite the prevalence and potential of community-academic partnerships, only minimal research has been done to evaluate how these partnerships work.²⁰

Study aim.

The aim of this study is to identify the beneficial or supportive factors, as well as the challenges, or sources of tension, in a community-academic partnership aimed to implement an EBP for increasing breast cancer screening among African American women.

Methods

Research setting.

CrossWorks is a faith-based community organization that works to reduce chronic diseases and health disparities in two rural eastern North Carolina counties where African Americans constitute 57% and 37% of the population.²¹ CrossWorks began partnering with the University of North Carolina at Chapel Hill’s Lineberger Comprehensive Cancer Center (UNC Lineberger) in 2007. In 2008, CrossWorks submitted an application in response to a competitive Request For Applications issued by the Center for Disease Control and Prevention via the Southeastern US Collaborative Center of Excellence for the Elimination of Disparities (SUCCEED), a cooperative effort led by the Morehouse School of Medicine Prevention Research Center.

CrossWorks considered several EBPs before selecting the North Carolina Breast Cancer Screening Program (NC-BCSP),²² a lay health advisor program that had increased mammography use by older African American women in five nearby counties with socio-demographic characteristics similar to those of the CrossWorks’ service area.²³ The NC-BCSP intentionally recruited women who appeared to be natural helpers to serve as lay health advisors (LHAs). “Natural helper” characteristics include being locally recognized and respected for trustworthiness, having good listening skills, being responsive to the needs of others, and being in control of their own life circumstances.²⁴ Prior to selecting NC-BCSP, the community and academic partners met for a half-day retreat to review and pilot various parts of the EBP. CrossWorks confirmed that the curriculum’s content and mission were consistent with the needs, skills, and wishes of its staff and volunteers. One month later, in June, 2009, UNC Lineberger and CrossWorks co-led a three-session, 12-hour LHA training using the NC-BCSP curriculum.

Three CrossWorks staff and four UNC Lineberger staff co-led the NC-BCSP lay health advisor training. Five were African American and two White. All trainers had at least a college education; two had a master’s degree and one had a doctorate degree. The trainers were all between the ages of 35–65. In comparison, the LHA volunteers were eight African American women and one Latina. All LHA volunteers had at least a high school education; several were college educated. The LHA volunteers were all between the ages of 30–65, with most of them being in their 40s. Although, there were demographic differences between the trainers and LHAs in terms of education and race, differences were not especially pronounced.

Research design.

The community and academic partners jointly designed the evaluation plan. Multiple meetings occurred in person and by phone to clarify evaluation goals, develop data collection and analysis protocols, and define each partner’s role in the data collection, analysis, interpretation, and reporting. Three co-authors also were interview participants. Their role in reporting and interpreting results was limited to giving feedback on language clarity and affirming the sequence of activities. The first author conducted the interviews, analyzed the data, and is responsible for interpreting the findings. The Institutional Review Board at the University of North Carolina at Chapel Hill approved this study.

Interview guide.

Staff of UNC Lineberger led the development of the interview guide. CrossWorks staff reviewed and edited its various iterations and approved the final instrument. The interview guide included open-ended questions about the partnership; activities that influenced selection, adaptation, and implementation of NC-BCSP; and activities that could be improved. The first interview was used to pilot the interview guide. The first interviewee affirmed the approach and suggested clearer wording for some questions.

Interview participants.

Interviews with 10 individuals took place from December 2009 to February 2010. Interviewees were three CrossWorks staff members, four UNC Lineberger staff members who provided training and technical assistance to CrossWorks, and three UNC Lineberger program leaders who ensured financial support and academic oversight for this project. All gave written consent for their participation. Interviews lasted 60–90 minutes, were audio-recorded, and were transcribed verbatim.

Data coding and analysis.

We used qualitative content analysis to identify, code, and categorize our data.²⁵ The interviewer and primary coder (RT) was not involved in the partnership to plan or implement NC-BCSP. He is a trained qualitative researcher with content analysis experience. He first listened to each interview while simultaneously reading the transcripts (interviews were transcribed by an outside contractor) to ensure their accuracy and to generate an initial set of codes. After generating the initial set of codes, the primary coder used the qualitative software management program, ATLAS.ti (ATLAS.ti Scientific Software Development GmbH, Hardenbergstr. 7, D-10623 Berlin), to code the data. During the coding process, some codes were added, others changed, and some were collapsed into more inclusive codes. Using ATLAS.ti and codes developed by the primary coder, a research assistant then independently coded all transcripts. After they reconciled inconsistencies through discussion and consensus, the primary coder reassembled the data based on assigned codes, allowing analytic concepts to be connected and descriptive themes to emerge.²⁶

Results

Findings that describe the beneficial or supportive factors that contributed to perceptions of the community-academic partnership’s overall success are presented first, followed by a description of challenges or sources of tension in the partnership. Beneficial or supportive factors include a history of a positive relationship between the partners, leaders who recognize the long-term benefits of community-academic partnerships, responsiveness to the community partner’s capacity-building interests, shared responsibility for planning and implementing activities, and academic partners with knowledge and skills specific to the EBP. Sources of tension in the partnership include lack of common definitions for discussing evidence-based practice, disagreement within the partnership about adhering to or modifying elements of the evidence-based program, and attempting to sustain high-quality services while conforming to external pressures.

Factors that contributed to the partnership’s success.

History of a positive relationship between the community and academic partners.

All three categories of interviewees indicated their joint effort to plan and implement an EBP directly benefited from having worked together in the past. One participant explained,

There were these prior interactions and a relationship process that started years ago, so there are some stumbles with our words, but because you’ve interacted with each other enough to know … okay, they misspoke, maybe they meant it, maybe they didn’t mean it, but I’m going to work through the process to reach a better understanding.

Academic interviewees expressed confidence in CrossWorks’ capacity to implement an EBP because its leader and staff had served as advisors to university studies, participated in workshops and conferences, and helped implement studies as a sub-contractor. Similarly, a community-based interviewee noted that UNC Lineberger collaborators had made valuable contributions to her organization and community, “We’ve had very successful relationships with the UNC community because we’ve seen, for the most part, that you value our time and respect our opinions, asking for input from us on these various projects. Not just coming in with this air of we’re the researchers, this is how it needs to be done.”

Leaders who recognize the long-term benefits of community-academic partnerships.

There was widespread agreement among interview participants that the leaders from both organizations were crucial to establishing trust and fostering the opportunity for collaboration. CrossWorks’ executive director is a registered dietitian with years of professional experience and a solid reputation for forging partnerships and mobilizing resources to improve public health infrastructure and outreach. Correspondingly, leadership at UNC Lineberger encourages its faculty and staff to engage in community-based participatory research and public service activities.

Leaders in both organizations acknowledged that community-academic partnerships can be synergistic, simultaneously enhancing the partners’ reach and long-term impact. This is reflected in the following reflection from one of the academic interviewees:

We’d like to sort of shorten the distance between better health outcomes and our research and education mission. And, CrossWorks is really about improving health right now. So, we have these different missions, but I think wherever we can sort of add some rungs between our organizations in order to build some opportunities to reach those better health outcomes faster is great.

Responsiveness to the community partner’s capacity-building interests.

Both academic and community partners noted that funders increasingly expect (or require) grantees to use EBPs. Working with academic researchers is one way to meet those expectations. A community partner interviewee explained that teaming up with an academic institution to carry out a targeted activity, such as implementing an evidence-based LHA training, builds organizational capacities that transfer to other initiatives:

… because all of our programs overlap so much, I would say that [the NC-BCSP lay health advisor training] has spilled over into the Komen program because it’s just helped our lay leaders to be more abreast of what they are supposed to do and just understand their role better. They’ve got more resources under their belt. Again, we have to do the cancer clinical trials outreach and recruitment. And, there’s overlap. You’re learning the skills of a lay advisor, you learn about confidentiality … .

Shared responsibility for planning and implementing activities.

Community and academic partners described their shared and distinct roles in deciding which EBP to adopt, how to adapt and implement it, and how to evaluate the collaboration. A UNC Lineberger leader explained, “I think we have to deal with partnerships in a bipartisan way, where the partners are not only interested in the question but help frame the question and maybe even help frame the intervention.” Most comments about shared responsibility, however, focused on joint planning and leadership of the NC-BCSP lay health advisor training, “Each person just brought their own personality and their own character and expertise to the training and it was just a wonderful collaborative,” one participant summarized. As previously noted, CrossWorks identified and recruited women who had reputations for being natural helpers and who (as a group) were culturally diverse to ensure the EBP would reach vulnerable populations. CrossWorks also selected and set up the LHA training location, recruited local speakers, sent reminders to training participants, and coordinated refreshments for all training sessions. Participants agreed that the training site, a local hospital, was ideal because trainees could find it easily and it increased the training’s credibility. One participant stated,

The facility was lovely. … It adds value to what you’re doing if you’re able to do a training in a facility that’s seen by the community as having power and being respectful. So it was excellent that it was in the hospital … a church basement wouldn’t have been bad but the hospital just brought the diverse populations together with a Muslim woman and a Latino woman. I mean it was neutral but it was highly respected.

CrossWorks and UNC Lineberger decided together whether to alter or delete LHA training content and who would lead each training presentation or activity. The academic partner updated training content to reflect the most current breast cancer incidence and mortality statistics and screening recommendations, prepared most of the training materials, and used feedback from the community partner to re-organize the order in which some information was presented. Academic partners also presented most of the curriculum content.

Academic partners with knowledge and skills specific to the evidence-based LHA program.

Two UNC Lineberger staff who contributed to this project were previously members of the NC-BCSP research team (between 1992–2002). They had helped develop the NC-BCSP LHA curriculum and had led numerous LHA training sessions. Interviewees appreciated these academic partners’ ability to contribute in-depth knowledge of the selected EBP. One trainer said, “Having [DB—a trainer who had worked on the NC-BCSP study] as a part of the team that actually went out to Crossworks was tremendous. She is amazing and knows this stuff; she’s done it a million times in different areas within North Carolina and outside North Carolina … .”

While the other trainers did not have prior experience with NC-BCSP, they spent time rehearsing training sections they had volunteered to lead. An interviewee recalled, “Everybody knew what they were supposed to do and everything just flowed in sync.”

Sources of tension in the partnership.

Lack of common definitions for discussing evidence-based practice.

Partners lacked a shared vocabulary to talk about one of the central purposes of the collaboration, which was translating an EBP into community practice. Academic partners applied inconsistent definitions and vague terms, such as evidence-based approach, evidence-based intervention, evidence-informed practice, fidelity, and so on. An academic partner voiced frustration, “… everybody thinks that they know what [evidence-based] means and that other people don’t.” Community interviewees defined evidence-based more loosely than their academic counterparts as illustrated by this comment:

… we started working with the lay advisors then and we saw how that worked well. And, then again, when we applied for the breast cancer funding, we continued with that model. And, then we started learning that it was evidence-based. We didn’t really know back then that it was evidenced-based. But, it was something that we were introduced to through the folic acid education.

Disagreement within the partnership about adhering to or modifying elements of the EBP.

Opinions differed on whether and how to modify the EBP. CrossWorks staff thought the 12-hour training spread over several weeks was too long for unpaid, mostly rural volunteers travelling long distances. UNC Lineberger partners wanted to “honor the wisdom” of the community partners, but were concerned that deleting training content, a core element of NC-BCSP, would reduce its effectiveness. As one trainer put it,

I was thinking that we were not really being as true to the original concept of the lay health advisor because if we’re doing something from five to eight tonight and you show up tomorrow morning at nine o’clock, you’re basically going home, grabbing dinner and sleeping. There’s no homework time. There’s no digestion time … . So I had to wrap my head around that and say okay look if this is the way it’s going to be then we have to figure out how to fit it all in because it’s not for my benefit. It’s for the people who are coming so what can we do, how can we adapt it to ensure that the fit and fidelity is there for the learners?

CrossWorks also noted that the NC-BCSP was designed and tested in a community of African American women who were slightly older, more rural, and more culturally homogeneous than the women in CrossWorks’ service area. CrossWorks updated language in the NC-BCSP curriculum that it knew to be incongruous with the spiritual and cultural traditions of its future lay health advisors.

There was also a discussion on whether or not to financially compensate the trainees. The NC-BCSP LHA program is designed to train volunteers to integrate breast cancer information into their daily conversations with friends, neighbors, coworkers and family, and attend monthly or bi-monthly meetings with other LHAs. For these types of activities, compensation was not offered in the original NC-BCSP or by CrossWorks. However, in addition to informal one-on-one advising, some CrossWorks’ LHAs planned to participate in formal education and outreach events. For helping with organized activities, such as staffing a health fair booth or speaking to a church group, LHAs were compensated $30 for each four-hour shift.

Attempting to sustain high quality services while conforming to external pressures.

As a community-based organization that relies on small, often one-year grants for operational support, CrossWorks continuously must strike a balance between having to follow the money (as it were) and sustaining programs launched with earlier grants. A CrossWorks partner explained the problem:

I guess the determining factor of where we spend most of our time is the funding. Because when we get funding that says ‘breast cancer’ and it’s a sizeable enough pot that we can spend a good deal of time on that, then we spend more time on breast cancer. But then when we get another bigger pot of money and it’s [another grant], then we’re pulled over here and start doing some diabetes work.

An academic interviewee echoed concerns from CrossWorks about possibly undue influence from external pressures. A well-meaning academic partner’s financial and technical assistance to focus on breast cancer screening might push a community partner to neglect other priority health issues that CrossWorks would like to address: “We’re on this breast cancer path with [CrossWorks] and I think that there are a lot of other health issues that are high priority … I worry that we’re defining the relationship too narrowly.”

Expectations of community and academic partners were sometimes at odds with each other. “When we’re looking at our academic research partnerships, I’m always trying to be true and faithful to my community partners and ensuring that they’re getting what they want and what they need … that’s hard sometimes to keep balanced,” said one interviewee. A community partner interviewee explained that while community members are enthused about lay health advising, closely replicating an EBP, such as NC-BCSP, is a lower priority: “I’ve got community people here who don’t give a hoot about [research] deliverables … they want something that’s meaningful to them and we’ve always been about quality.”

Discussion

Community-academic partnerships offer a model for integrating both research and practice knowledge to develop public health programs that are both culturally-sensitive and evidence-based.²⁷ We identified several factors that help guide these partnerships to select and successfully begin using an evidence-based program like NC-BCSP. As has been found in prior research, we found that strong community-academic research partnerships develop over time as partners build a trusting relationship and establish clear roles and shared responsibilities.^27,28 Not unexpectedly, the expertise and skills of the academic partners who directly interact with communities are critical.^27,28 Both community and academic interviewees valued the contributions of individuals who were well-versed in the evidence-based NC-BCSP. Wandersman et al. note that adoption and implementation of evidence require both “innovation-specific capacities” (such as NC-BCSP-specific knowledge) and “general capacities,” such as supportive communication throughout the partnership and during the LHA training.^10[p.175] In addition, both the academic and community partners were willing to evaluate and modify practices when warranted.

Differing perspectives can strain community-academic partnerships.^29-31 Urban and Trochim noted difficulty reconciling practitioners’ desire to act quickly to address immediate concerns with researchers’ orientation toward long term knowledge acquisition.³² Lasting community-academic partnerships must have the capacity to take note of and share information about resource limitations, develop solutions together, and adjust.³³ In our evaluation project, community and academic partners initially held conflicting views on how much time to spend at the beginning to adapt the NC-BCSP intervention to local conditions and plan for long-term sustainability. However, the partnership was not undermined by such tension, partly due to the continuous and open communication among partners and the willingness to find common ground. Roles and responsibilities associated with the LHA training were carefully discussed and outlined, taking into consideration the strengths and limitations of both partners. A joint decision was made for the academic partner to lead training sessions and the community partner to assume a co-leader role (while developing skills for delivering this particular training in the future). In-depth discussions about whether and how to fully implement the NC-BCSP were postponed until after the LHA training.

Maintaining acceptable levels of fidelity to the original design of an evidence-based program and adapting it to fit with local needs and resources are well-documented challenges.^34,35 The present study contributes to our understanding of this challenge, highlighting how important it is for academic partners to demonstrate knowledge of implementing the EBP equal to what they demonstrated while planning and co-leading the LHA training. Additional implementation aspects to consider are systems for supervising and mentoring LHAs, defining and coordinating LHA activities, roles and responsibilities, and evaluating program impact.

Use of academic language is a widely recognized barrier to effective communication between community and academic partners.²¹ It is important for partners to agree on clear definitions for evidence-based public health. The absence of a common language has broad implications for how these partnerships define and measure their success in improving the quality of breast cancer screening programs.

Through this partnership, CrossWorks strengthened its capacity for community outreach and education to increase breast cancer screening and leveraged new knowledge and skills to win bigger grants and begin implementing new programs. At the same time, the partnership has provided UNC Lineberger faculty and staff hands-on experience and a better understanding of the needs and resources of a community organization.

The women who graduated from the training continue to identify as LHAs and volunteer for CrossWorks; however, due to loss of funding for its breast health initiatives and subsequent staffing constraints, CrossWorks is not fully implementing its LHA program. CrossWorks meets regularly with two other organizations that also have LHA programs. Program leaders from these three organizations share information and support each other’s cancer prevention and control activities. Following a recent discussion about challenges to more fully implementing and institutionalizing their LHA programs, a member of this group contacted the UNC research team to request consultation and technical assistance that will help all three agencies build on the skills and resources of their trained LHAs.

Strengths and limitations.

Qualitative research emphasizes depth over breadth and insight over generality; it involves an element of subjectivity. We used several qualitative research approaches to increase the credibility of our findings, including the use of an experienced interviewer who was not involved in the partnership, audio-recording, verbatim transcription of interviews, software to code transcripts and document coding decisions, and duplicate coding. While we feel confident in our application of sound qualitative methods, we cannot rule out the possibility that researcher bias influenced the interpretation of our evaluation findings.

Summary.

In summary, this evaluation yields several recommendations for developing effective community-academic partnerships aimed at promoting the dissemination and implementation of an evidence-based program, such as NC-BCSP. Partners should commit to long-term relationships that extend beyond the life of any single project. Academic partners should take care in selecting the individuals who will directly interact with community partners to ensure that they have the necessary knowledge and skills. As consultants on evidence-based practice, academic partners should thoroughly know the EBP they plan to implement and be prepared to model how to most effectively implement it. Partners should work together to negotiate plans that ensure that roles are clear, responsibilities are shared, and project time lines meet the needs of both communities and academic partners. Lastly, establishing a shared vocabulary to define the goals of a partnership for implementing an evidence-based intervention, such as NC-BCSP, is critical to ensuring that interventions are adapted to match community capacities and expectations while also being implemented as fully as possible.