Abstract
We offer a framework and exemplify how to integrate multiple community perspectives in research to develop breast cancer screening interventions among Latinas non-adherent to national guidelines. We leverage members of an academic institution’s community consultative service [community engagement advisory board (CEAB) members]; study team members [community health workers (CHWs)] and study-eligible individuals (non-adherent Latinas). First, we asked what was needed from CEAB members (N=17), CHWs (N=14) and non-adherent Latinas (N=20) in one-time semi-structured group consultations and focus groups. Second, we drafted materials. Third, we conducted group consultations and focus groups with a new set of CEAB members (N=13), CHWs (N=17) and non-adherent Latinas (N=16) to reflect on our initial analysis and draft materials. Fourth, we finalized interventions. Certain recommendations were shared across stakeholders and simple to integrate (e.g. costs → access to free services). Some recommendations varied, but complementary integration was possible (e.g. location versus recruitment → multiple recruitment in multiple community areas). Others were distinct across stakeholders and resulted in strategies to recognize participants’ agency and inform their choices about breast cancer screening (e.g. differences in preferred information about screening → personalized information and evidence about all screening options).
Introduction
There has been an exponential growth of approaches that incorporate community perspectives in health education research [1], especially those focused on minority health and health disparity populations. This increasing interest toward participatory-based approaches, which had been well-established but kept on the margins [2], partially reflects a frustration with the limited efficacy of traditional interventions; an interest in supporting populations’ capacity to handle the disproportionate burdens they face and a commitment to individuals’ rights to long, high-quality, healthy lives [3]. Approaches that incorporate community perspectives are not monolithic [4, 5] and vary in terms of goals of collaboration; research stage/duration of collaboration; type of stakeholder(s); associated nomenclature; type and magnitude of activities and other dimensions [6]. The optimal approach likely depends on the research problem and question [7, 8].
A growing body of work has begun to characterize and compare approaches [9]. To our knowledge, relatively few maps are available to guide researchers in terms of which approach may be optimal. This study offers two preliminary contributions toward the goal of achieving consensus on best collaborative practices: (i) we characterize three common types of community members and highlight their respective advantages as collaborators and (ii) we describe strategies toward integration of multiple community perspectives.
Characterizing multiple community perspectives
Table I provides one guide to summarize a substantial body of work [1–20] regarding diverse types of community stakeholders, including: (i) what they look like: (ii) how they generally function; (iii) what methods optimize collaboration and (iv) what factors should be considered when deciding which partner(s) are needed. Below, we describe each type of stakeholder, with a focus on the function and advantages of collaborating with them. This information highlights why and when different community stakeholders would be optimal for collaboration, based on the research question and goal (e.g. sustainability, feasibility and acceptability).
Table I.
Conceptualizations about multiple community stakeholders
| Type of community stakeholder | Collaboration function | Collaboration frequency | Collaboration methods | Advantages of collaboration |
|---|---|---|---|---|
| Advisory board members | *Overarching recommendations regarding study objectives and processes |
|
*Group consultation | Personal knowledge of historic and existing infrastructure for academic-practice-community partnerships |
| Knowledge based on academic, practice, and community wisdoms | ||||
| Study team members | *Specific recommendations to improve study processes logistically | *Regular interaction | *Study team meetings | Personal history with pragmatic issues in academic-practice-community partnerships |
| Knowledge of how to deliver information to community members, including academic and practice wisdoms | ||||
| Study-eligible individuals | *Specific recommendations to improve study acceptability | *One time interaction |
|
Personal history as a recipient of services by academic-practice-community partnerships |
| Knowledge of what is wanted by and would be appealing to self and individuals like self |
CAB = Community Advisory Board; CEAB = Community Engagement Advisory Board.
Researchers generally interact with advisory boards to obtain overarching or ‘big picture’ recommendations about study objectives and processes. Given this function, advisory boards often comprise lay community members, patient advocates, leaders of community organizations and/or researchers with significant experience in public health practice and research [18–21]. Individuals’ membership reflects their positions as gatekeepers and well-connected ‘hubs’ within local academic-community networks [22]. There are different types of advisory boards. Membership on steering committees and community advisory boards (CABs) is often a result of a direct invitation by researchers based on invitees’ personal, professional, leadership experience with specific communities, specific public health interests and researchers/academic partners [18]. Steering committees and CAB members generally collaborate with researchers via regular group meetings. Membership on community engagement advisory boards (CEABs) results from an invitation by an infrastructure grant/center within an academic institution based on invitees’ experience with public health practice and research within their local communities [19, 20]. CEAB members and consulting researchers share an interest in specific communities and significant experience in academic-practice-community partnerships, but do not necessarily share similar public health interests or have a pre-existing relationship. Researchers may only interact with CEAB members via group consultations once or twice during the project. Outside of these differences, advisory board members contribute information about historical and current context, including their roles and understanding of existing infrastructure and partnerships; and, recommendations that incorporate their personal history with researchers as advisors. Their position enables them to provide feedback regarding sustainable solutions.
Researchers generally interact with study team members to solicit logistic recommendations to optimize study processes and achieve study goals. Researchers and team members collaborate regularly via team meetings about the operational components of specific research projects. There are multiple types of team members, including community principal investigators (PIs) and workers who directly implement project activities [e.g. patient navigators, community health workers (CHWs)] [23]. Community PIs are as organizational leaders and peers to researchers. Other study team members are researchers’ employees. Overall, study team members leverage their shared sociocultural identities and spaces (e.g. as neighbors, congregational members) with study participants to facilitate project implementation and success [24]. Their regular, hands-on experiences with others’ lived experiences results in a rich understanding about local populations’ needs, experiences and perspectives [25, 26], including the optimal ‘mechanics’ of engaging communities of interest at the grass-roots, daily level [11]. Their daily interactions with other grass-roots stakeholders are helpful in terms of identifying what may be most useful for the project at hand. Study team members contribute information about their personal history as professionals within local contexts, including the pragmatic aspects of existing infrastructure; and, recommendations that incorporate their experiences as study team members delivering information directly to participants. Their position enables them to provide feedback regarding feasible solutions.
Researchers generally interact with study-eligible participants to obtain specific recommendations about the appeal of their research questions and projects. Study-eligible individuals are the community members who would be eligible to participate in the research study of interest. They generally collaborate once via participation in qualitative data collection (e.g. semi-structured interviews, focus groups). These individuals understand their contexts in terms of their personal, not professional, lived experiences. They are most salient of the personal connections that they have. Their own lived experience offers researchers nuanced insights regarding intrapersonal determinants (e.g. individual’s awareness of resources, factors contributing to using resources) that may not be consistently documented or understood by existing infrastructure and practitioners [12]. For example, they may clarify the unintended negative consequences of community-based study team members engaging them for research (e.g. pressure to participate, changed interpersonal dynamics). Study-eligible individuals contribute information about their personal history as individuals within local contexts; and, recommendations concerning what would be most appealing to them as recipients of health information. Their position enables them to provide feedback regarding acceptable solutions.
Obtainment and integration of multiple perspectives
Certain research questions may benefit from obtaining multiple perspectives. Intervention development is an excellent example. Researchers often hope to develop sustainable, feasible, and acceptable interventions. Toward that goal, a reasonable method involves leveraging the common overlap between different types of stakeholders. Specifically, community members may have multiple positions, including serving as advisory board members for an academic institution; as study team members for a specific project; and, as eligible participants for another project. Under these circumstances, researchers may solicit different perspectives by asking such community members to consider the same question through their different roles. Another reasonable method is to solicit individuals who are advisory board members, study team members or study-eligible participants. Of which we are aware, little work has clarified best practices for which of these methods of selection may be optimal. We posit that the optimal method will likely depend on the research question, including the degree to which exposure to researchers may change perspectives (e.g. if study eligibility includes non-adherent behaviors) and the degree to which there are multiple truths across academic, practice and community wisdoms (e.g. different possible responses to a public health problem).
Another gap in the literature concerns how to integrate multiple community perspectives. Three types of scenarios may result when obtaining multiple community perspectives. First, community stakeholders may share recommendations. This is likely to occur when an issue or factor is particularly salient for all stakeholders via shared personal experiences and perceptions. Under this scenario, it may be easy to ‘integrate’ recommendations, given there is a singular recommendation for study design. Second, community stakeholders may have complementary recommendations. This is likely to occur when there are differences in attributions and awareness between the different stakeholders. Under this scenario, integration may be possible via multi-faceted study design. Third, community stakeholders may provide distinct recommendations that are not easy to integrate. Under this scenario, researchers must attempt a balance in their study designs, including identifying which recommendations they prioritize [27], and/or considering multiple or ‘open’ methods/approaches [28] that incorporate all input.
Current study
The current study exemplifies the advantages of collaborating with three types of stakeholders and integrating their perspectives in the context of promoting breast cancer screening among Latinas who were not adherent to US Preventive Services Task Force (USPSTF) guidelines (52–74 years old; no mammogram in past 2 years) [29]. This is an ideal public health problem for understanding multiple perspectives, given the ongoing debates and associated shifts in clinical guidelines [30]; the availability of multiple perspectives to address the problem from past research and public health practice [31, 32]; and, the persistence of ethnic disparities in breast cancer screening uptake [33]. In our study, we describe: (i) how conceptualized advantages of collaborating with each type of stakeholder emerged and (ii) how we integrated perspectives into intervention design.
Materials and methods
Setting
This ancillary study is part of a larger NCI-funded project [34] that compares the effectiveness of different intervention approaches to promote mammography screening among non-adherent Latinas. The original study setting was a community neighborhood in West Chicago. Based on stakeholder recommendations, Phase 2 and the subsequent interventions added a new neighborhood in South Chicago. Both neighborhoods were predominantly Latino (≥65% of residents) and socioeconomically disadvantaged (≥30% of residents living within 100–199% of the federal poverty level). The study team included leaders from two community-based organizations and one academic institution. The Institutional Review Board at the University of Illinois at Chicago approved all procedures.
Current study design
The parent study used continuous stakeholder engagement for intervention development, an approach, which highlights the benefit of incorporating community perspectives at each stage of the research process [13]. During Phase 1, we identified the most relevant topics and procedures for promoting breast cancer screening from CEAB, CHWs and non-adherent Latinas through one-time group interactions. Subsequently, we analysed data and developed draft intervention protocols and materials. During Phase 2, we obtained feedback on our interpretation of Phase 1 recommendations as well as our draft protocols and materials from a new set of CEAB, CHWs and non-adherent Latinas through one-time group interactions. Table II highlights the timing for each phase; the frequency of collaboration; the eligibility criteria for each type of stakeholder and details regarding interactions with each community stakeholder type (type of group interaction; sample questions).
Table II.
Details concerning the purpose of interaction, year of interaction, eligibility criteria, procedures and questions
| Purpose of collaboration | Year of one-time group-based collaborationa | Eligibility criteria | Sample size | Focus group procedures & questions | CEAB consultation procedures & questions |
|---|---|---|---|---|---|
Phase 1 (‘Identify’)
|
2015 |
|
Questions and prompts solicited feedback
|
Questions and prompts solicited feedback
|
|
Phase 2 (‘Develop & Assess’)
|
2016 | Study team provided a ‘dress rehearsal’ of materials and asked questions after each segment of information (nine segments total)
|
CEAB received materials a week beforehand. Study team went through each segment and asked questions
|
CEAB = Community Engagement Advisory Board. CHW = Community Health Worker. Two non-adherent Latinas completed interviews during Phase 1, because of their preferences to interact with researchers alone.
One CEAB member was a part of Board A and Board B. All other stakeholders participated only once.
Although not a requirement, none of the CHWs who participated in this study had ever been part of a CEAB.
Data collection
To obtain CEAB perspectives, we scheduled two 45-min group consultations in English with two committees from a 15-year old CEAB hosted within the academic institution’s Clinical and Translational Science Awards (CTSA) program. CTSA coordinators recruited CEAB members. The CEAB included representatives who did not identify as Latino and who did not speak Spanish. The CTSA coordinators recorded anonymous, detailed notes, which were sent to the PI within a week of the consultation.
Focus groups were conducted to obtain perspectives from CHWs and non-adherent Latinas. The study team recruited targeted interventionists and participants through word-of-mouth and physical flyers posted at community venues. Interested individuals contacted the study team; were screened for eligibility (Table II); and, were scheduled to participate in a focus group (4–10 individuals) within community partners’ offices in targeted community areas. Focus groups were held separately for non-adherent Latinas and CHWs. Due to language preferences, focus groups were held mostly in Spanish. Individuals completed informed consent forms; completed a brief demographic questionnaire (Table III) and then participated in audio-recorded 1-h focus groups. Groups were led by bilingual/bicultural study team members with expertise in qualitative data collection. After participation, individuals received $40 for their time and effort.
Table III.
Study sample characteristics
| Non-adherent Latinas | CHWs | P-valuea | |
|---|---|---|---|
| (n=39) | (n=31) | ||
| n (%) | n (%) | ||
| Phase | 0.61 | ||
| 1 | 20 (51) | 14 (45) | |
| 2 | 19 (49) | 17 (55) | |
| Age | 0.41 | ||
| 52–64 years old | 31 (80) | 22 (71) | |
| 65–74 years old | 8 (20) | 9 (29) | |
| Country of birth | 0.47 | ||
| Mexico | 30 (77) | 26 (84) | |
| Other | 9 (23) | 5 (16) | |
| Language | 0.50 | ||
| Spanish only | 32 (82) | 28 (90) | |
| Bilingual | 7 (18) | 3 (10) | |
| Marital status | 0.09 | ||
| Married | 31 (80) | 19 (61) | |
| Other | 8 (21) | 12 (39) | |
| Insurance status | 0.26 | ||
| Private insurance | 17 (44) | 8 (26) | |
| Public insurance | 17 (44) | 16 (52) | |
| Uninsured | 5 (13) | 7 (23) | |
| Education | 0.24 | ||
| <Ninth grade | 23 (59) | 12 (39) | |
| High school | 13 (33) | 16 (52) | |
| >High school | 3 (8) | 3 (10) | |
| Household income | 0.77 | ||
| Missing | 5 (13) | 5 (16) | |
| <Median ($10 000) | 17 (44) | 14 (45) | |
| ≥Median ($10 000 | 17 (44) | 12 (39) | |
| Breast cancer survivorb | — | ||
| No | 39 | 21 (68) | |
| Yes | 0 | 10 (32) |
For analyses, wherein cell sizes were <5 individuals, Fisher’s exact test is reported.
To be an eligible, one could not have a personal history of breast cancer.
CHW = Community Health Worker.
Analysis
CEAB notes and verbatim focus group transcripts were first entered into Atlas.TI, version 7 (Berlin, Germany). Table III depicts how our qualitative analysis was utilized to inform intervention development. Specifically, we (i) analysed Phase 1 data to identify themes and assess different community stakeholders’ recommendations; (ii) developed intervention protocols and materials; (iii) analysed Phase 2 data to assess the believability of our Phase 1 analysis, the acceptability of our intervention materials as hypothetical participants receiving the intervention, and alternative content/materials and (iv) finalized intervention protocols and materials.
For Phase 1, we used multi-tiered content analysis with the principle of constant comparison [35, 36]. Three authors (Y.M., M.H.F. and T.B.G.) independently read each transcript, and identified meaning and themes from raw interview data. The team generated preliminary codes to capture the essence of each idea, compared notes, reviewed the data and clustered similar ideas together into categories representative of each emergent theme. All coders met to review the codes overall; assess intercoder agreement by examining code application within specific segments of text; and, resolve areas of disagreement [37]. The codebook was modified across multiple iterative coding cycles. Phase 1 codebook nodes were intervention procedures (example codes: recruitment, location, format) and content (example codes: barriers to screening, facilitators to screening).
For Phase 2 data collection, we used adaptive member-checking procedures for Phase 2 [38, 39], given member checking with original participants are challenging due to the aggregate presentation of qualitative data [40]. Accordingly, we presented draft intervention materials and procedures by introducing emergent themes from our Phase 1 analysis. For Phase 2 data analysis, we used the same multi-tiered content analysis approach as is described for Phase 1 [35]. Phase 2 codebook nodes included Phase 1 nodes; nodes concerning materials and procedures (example codes: minor changes, specific resources); and, nodes regarding consensus within stakeholder type (example codes: explicit agreement with Phase 1 stakeholders; similar recommendation as Phase 1 stakeholder).
For this study, after our final analysis of Phase 1 and Phase 2 data, we used peer debriefing strategies. The PI (Y.M.) requested the review of analyses, interpretations and framing from impartial colleagues who were not directly involved in the parent study (K.S.W., P.A.M.) [38, 40].
Results
Demographic characteristics for CHWs and non-adherent Latinas from both phases are reported in Table III. CHWs and non-adherent Latinas were largely between 52–64 years old, were born in Mexico, were Spanish monolingual, did not have private insurance, and had a high school education or less. The median annual household income was $10 000. CHWs and non-adherent Latinas were similar, except that non-adherent Latinas were slightly more likely to report being married/living with a partner than CHWs (80 versus 61%). Approximately a third of CHWs were breast cancer survivors.
Table IV depicts the Phase 1 recommendations provided by CEAB, CHWs and non-adherent Latinas according to the conceptual advantages of collaboration listed in Table I; type of recommendations (shared, complementary or distinct); intervention development based on Phase 1 data; and, Phase 2 feedback on Phase 1 recommendations and associated activities.
Table IV.
Qualitative results by community perspective, stage of research, conceptual advantages and emergent themes a
| Phase 1: identify recommendations for intervention development | ||||
|---|---|---|---|---|
| Themes | CEAB members | CHWs | Non-adherent Latinas | Qual analysis |
| Healthcare access & economic hardship | Personal knowledge of infrastructure: ‘Be mindful that they [priority population] often have barriers to health access.’ | Knowledge of how to deliver academic and practice wisdoms: [When asked what was most important to communicate in interventions] ‘That there is assistance and coverage – that it doesn’t matter if they don’t have insurance. There is always somewhere to go to get a mammogram.’ | Knowledge of what would be appealing: ‘When they [women] see free stuff, they’re there.’ | Recommendation shared across stakeholders |
| Proposed intervention location & recruitment |
|
|
|
Recommendation complementary across stakeholders |
| Breast cancer information | Knowledge based on academic, practice and community wisdoms: ‘It’s important to look at these [breast cancer] data.’ |
|
|
Recommendation Distinct between stakeholders |
| Phase 2: feedback on Phase 1 interpretation and draft intervention materials | ||||
| Phase 1 recommendations | Responsive solution | CEAB members | CHWs | Non-adherent Latinas |
|---|---|---|---|---|
| Address healthcare access & economic hardship | Navigation protocol to free/low-cost breast cancer screening for uninsured and underinsured women | Knowledge based on academic, practice and community wisdoms: [After reviewing the navigation protocol, appreciated information that] ‘there are places you can go for a mammogram without the high cost.’ | Personal knowledge of information delivery: I like it [navigation services] a lot because then women won’t feel these barriers so precisely and won’t say ‘I don’t have insurance.’ This happens a lot and they won’t go…I have known people who will even lend women the money [for the exam.] | Knowledge of what would be appealing: ‘Many women they don’t check themselves because they don’t have medical insurance – it’s the number one reason…I like would like to talk to them about this [information in intervention materials].’ |
Optimal intervention locations and recruitment strategies:
|
|
Knowledge based on academic, practice and community wisdoms: [After reviewing new location sites and strategies] [Emphasize the] ‘importance of understanding [the specific] community in one’s research.’ |
|
|
Optimal breast cancer information
|
Multimedia PowerPoint presentations that include:
|
|
|
|
Results are provided by type of community perspective, stage of research and emergent themes. Conceptual advantages are linked with exemplar quotes.
USPSTF = US Preventive Services Task Force. CEAB = Community Engagement Advisory Board. CHW = Community Health Worker.
Phase 1: identify recommendations for intervention development
Phase 1 was a first step to identify recommendations for intervention content and procedures. Our analyses revealed shared recommendations (need to address healthcare access/economic hardship); complementary recommendations (intervention location, recruitment strategies) and distinct recommendations (type of breast cancer information to provide).
A major recommendation shared across all stakeholders concerned addressing economic barriers to breast cancer screening. The lenses by which they discussed economic barriers aligned with different stakeholders’ conceptualized advantages of collaboration. CEAB members framed economic hardship as a well-known barrier, emphasizing populations’ collective lack of access. Their framing reflected their personal knowledge of infrastructure. CHWs discussed hardship in terms of emphasizing the availability of free/low-cost screening services when interacting with non-adherent Latinas. Their framing reflected their knowledge of how to deliver resources. Non-adherent Latinas discussed becoming aware of economic solutions would motivate women to obtain breast cancer screening. Their framing reflected what they believed would be appealing for women like them, based on their lived experience.
Recommendations concerning the proposed intervention location and recruitment strategies highlighted differences in how stakeholders experienced health information exchange. CEAB members indicated frustration that certain neighborhoods were saturated with breast cancer resources, whereas other neighborhoods were relatively untouched. However, non-adherent Latinas from those ‘resource rich’ neighborhoods indicated low awareness of such past efforts, potentially due to limited exposure to past efforts’ recruitment and outreach. Relatedly, the types of recruitment strategies highlighted the degree to which they interacted with the larger context of health infrastructure in Chicago. CEAB members discussed the importance of using resources available to the team’s academic institution, including established partners in priority neighborhoods. CHWs highlighted the importance of community venues, such as health fairs within the priority neighborhoods. Non-adherent Latinas emphasized the use of non-health resources, including community media, work places and other public settings.
The type of breast cancer information that was recommended differed across
CEAB members, CHWs and non-adherent Latinas. CEAB members focused on disseminating available research data during interventions (e.g. mammography, based on national guidelines). CHWs did not focus on the type of information to be given, but were more focused on how to deliver the information. Specifically, they emphasized the importance of discussing breast cancer as a community issue and the importance of breast cancer survivors who shared sociocultural identities within priority neighborhoods. Non-adherent Latinas, conversely, highlighted the importance of tailored information by a resource who knew them and an interest in information that they could highlight their autonomy.
Intervention development: integrating inputs
Phase 1 analysis resulted in three shifts in intervention development.
First, to address shared recommendations regarding economic hardship, we modified our protocol to incorporate navigation to low-cost/free mammography services at breast imaging centers of excellence. This navigation protocol leveraged existing academic-practice partnerships. We also incorporated content within intervention presentations that communicated clearly and repetitively the availability of navigation services.
Second, to address complementary recommendations regarding intervention location and recruitment options, we decided on a multi-faceted strategy. A new neighborhood was selected that was identified by CEAB members as underserved. The CEAB introduced the study team to an organization, who was ultimately invited to become a partner for the project. Multiple recruitment options were incorporated within the protocol, including use of community and practice partners affiliated with the academic institution; CHW-based recruitment at health fairs and other community venues; CHW interviews on community media and the use of peer referral, such that non-adherent Latina participants could refer their family and friends within non-health spaces.
Third, to address distinct recommendations about breast cancer information, we designed a multi-faceted, multimedia presentation that informed participants about their options and recognized their agency to choose. These presentations are led by CHW with experiences as breast cancer survivors or caregivers of breast cancer survivors. CHWs share their personal testimonials, highlight famous Latina breast cancer survivors, describe why breast cancer was a problem for Latino communities (incidence rates; disparities) and introduce different breast cancer screening methods (mammography, clinical breast exam, self-breast exam; CEAB). There are also activities that enable women to select their personal risk factors, perceived barriers and preferred solutions.
Phase 2: feedback on phase 1 interpretation and draft intervention materials
During Phase 2, we obtained feedback on our draft intervention content and protocols, which were introduced by first discussing Phase 1 results. Phase 2 was conducted with CHWs working within and non-adherent Latinas living in the original and new study location. With regard to our navigation program to address economic hardship, CEAB members emphasized existing infrastructure that was present. CHWs focused on how to disseminate information to motivate women to obtain screenings. Non-adherent Latinas were enthusiastic to share the message they believed women like them most wanted. With regard to intervention locations and recruitment options, CEABs appreciated the new location but emphasized important neighborhood differences and the need for in-depth familiarity in both areas. Non-adherent Latinas were particularly interested in peer referrals and use of community resources that did not necessarily pertain to health. With regard to our multimedia presentation, CEAB members emphasized that screening options which were not evidence-based should not be included, in line with their access to academic, practice and community wisdoms. CHWs were enthusiastic about how the information was provided (e.g. testimonials, multiple strategies for highlighting the importance of breast cancer). Non-adherent Latinas appreciated the activities that offered information directly pertinent to their personal circumstances.
Final intervention development and refinement
Phase 2 feedback confirmed that draft intervention content and protocols were relevant, important and useful to the targeted population. The final intervention made minor changes (e.g. wording, spelling, type of material for activities). Full details regarding intervention content and protocols are provided in detail elsewhere [34].
Discussion
We offer important contributions to the future of research that incorporates community perspectives. There is a diversity of collaborations that researchers and community members can have. We highlight that one size does not fit all. No specific stakeholder is likely able to give information about community across all of the different perspectives described above—nor should they be expected to do so. Our framework exemplifies this message by (i) offering one conceptual direction of the strengths and lenses of three types of stakeholders and (ii) describing how to integrate multiple perspectives. We posit that research questions should drive the type(s) of community perspective(s) that are used. We acknowledge that collaborating with multiple stakeholders, while offering a more nuanced perspective to public health problems and solutions, has unique challenges. As described below, sometimes integration may be relatively simple. Other times, it is likely to be more complex and need extensive consideration.
Considering multiple stakeholder perspectives
Our qualitative data overall supported the conceptualizations depicted in Table I. CEAB members’ feedback highlighted exposure to national evidence and research concerning breast cancer screening as well as a rich awareness of relevant local efforts in research and practice. The community-based study team members, CHWs, had feedback, which reflected experiences in delivering such programs to the local priority population. The women from the priority population, non-adherent Latinas, provided recommendations that reflected their lived experiences and what they believed women like them faced on a daily basis. These findings not only suggest the utility of our framework, but also highlight the kind of recommendations that researchers may receive from these distinct segments of the community.
Integration of multiple stakeholders’ recommendations
Our findings first exemplify how to integrate data when recommendations are shared across stakeholders. Specifically, we learned that we should incorporate navigation to low-cost/free mammography services to address the salient economic factors felt across different segments of the community. The consensus we observed across stakeholders parallels a growing consensus from multiple reviews and meta-analyses revealing economic factors contribute to ethnic disparities in breast cancer screening uptake [31, 41].
Second, our study exemplified how to integrate complementary inputs. Our example concerned recommendations about intervention location and recruitment options. We learned two truths. First, there are underserved geographic communities that have been prioritized by research and public health practice. Other communities that are similar have received relatively little attention from researchers and practitioners. Second, as importantly, geographic communities are not monolithic. There are likely individuals within these well-connected and well-studied communities that have not benefited from available resources. Altogether, these data enabled us to choose to implement interventions in multiple community areas and to use multi-method non-probability based sampling strategies to access individuals who may not be already connected to community resources.
Finally, our findings concerning breast cancer information exemplify how to integrate inputs when there are distinct recommendations and integration is challenging. This scenario exemplified perspectives reflecting what was acceptable by non-adherent Latinas, who clarified what they as consumers wanted; feasible by CHWs, who clarified what had been successful for them in the past; and, sustainable by CEAB members, who clarified the message in the context of national guidelines and evidence-based practice. These findings provided us with important information regarding our interventions. First, we needed to address the debate about mammography screening and growing popularity of risk-based screening decisions [42]. This is particularly important, given there is likely to be a fair amount of information being transmitted within women’s social environments in addition to our interventions that may reflect different scientific stances about breast cancer screening. There is an important balance we need to achieve between implementing evidence-based practices for which science and evaluation is ongoing while disseminating information that optimizes informed shared decision-making. In an effort toward that goal, we designed material to be comprehensive in terms of offering information about what is known nationally, acknowledges conflicting information that may be present and situates women’s decisions in terms of this information as well as their personal contexts.
Limitations
This study had several limitations. There were differences in how we interacted with different community stakeholders, which may have impacted the types of recommendations they provided. The CEAB was held entirely in English and did not include solely members of Latino communities. They also had long-standing relationships with researchers, although most met the study team only once for this project. In contrast, we interacted with CHWs and non-adherent Latinas mostly in Spanish and within an all-female group setting. Relatedly, CHWs were consulted via focus groups instead of via team meetings (Table I). This non-traditional method of collaboration may have impacted their recommendations. Second, our study used community-based, non-probability based sampling and was situated within the specific context of breast cancer among Latinas within specific Chicago-based neighborhoods. Other research is needed to assess transferability of themes and the potential of our conceptual framework across different contexts, populations, health conditions and interventions. Third, our questions and procedures for collecting qualitative data were not previously piloted or validated for use. Fourth, our pilot study’s design did not allow us to capture other factors that are relevant to collaboration between researchers and community stakeholders, including collaboration frequency and methods.
Conclusions
This study provided a framework for considering and integrating multiple stakeholder perspectives in terms of advisory board members, study team members and study-eligible participants. Our pilot study helped to clarify why understanding which stakeholder perspective should be considered in the context of the research question and what is needed from academic-community partnerships. Specifically, for intervention development such as ours, we needed to interact with multiple types of stakeholders. Our interaction with CEAB members, CHWs and non-adherent Latinas resulted in the development of breast health interventions for Latinas that implemented recommendations identified by all stakeholders (the need for low-cost/free mammography services); complementary recommendations (multiple community areas and multi-method recruitment) and distinct recommendations (showcasing all screening options and available evidence to optimize patient decision-making processes). We acknowledge that our exploratory study offers a few strategies; of course, these are not the only strategies. There is a need for future scholarship and intentional discussions to identify best practices for integration of inputs from different stakeholders as well as stakeholder solicitation strategies based on research questions.
Acknowledgements
The authors would like to thank the efforts and support of Wendy Choure, Raymond Ruiz, Marilyn Willis, Maria Medina, Juanita Arroyo, Irma Villa, Carmen Garcia, Marta Perea, Sylvia Gonzalez, Lizeth Tamayo, Carola Sanchez Diaz, Jessica Torres, Kryztal Pe�a, Casandra Robledo, Beti Thompson, Sarah Hohl, Genesis Rios, David Dubois, Alexis Grant, Genie Rizzo, Nadia Al-Amin and Ayo Olagoke.
Funding
The National Cancer Institute (K01CA193918); the UIC Institute for Race Research and Public Policy; the UIC Center for Research on Women and Gender and the UI Cancer Center, and the University of Illinois at Chicago (UIC) Center for Clinical and Translational Science (CCTS), which is supported by the National Center for Advancing Translational Sciences, National Institutes of Health, through Grant (UL1TR002003) funded this work.
Conflict of interest statement
None declared.
References
- 1. Wallerstein N, Duran B.. Community-based participatory research contributions to intervention research: the intersection of science and practice to improve health equity. Am J Public Health 2010; 100: S40–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Belone L, Lucero J, Duran B. et al. Community-based participatory research conceptual model: community partner consultation and face validity. Qual Health Res 2016; 26: 117–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Jagosh J, Macaulay A, Pluye P. et al. Uncovering the benefits of participatory research: implications of a realist review for health research and practice. Milbank Q 2012; 90: 311–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Israel B, Eng E, Schulz A. et al. Experiential learning in graduate education: development, delivery, and analysis of an evidence-based intervention. Sci Res 2005; 3: 649–57. [Google Scholar]
- 5. Israel B, Krieger J, Vlahov D. et al. Challenges and facilitating factors in sustaining community-based participatory research partnerships: lessons learned from the Detroit, New York City and Seattle Urban Research Centers. J Urban Health 2006; 83: 1022–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Sheridan S, Schrandt S, Forsythe L. et al. The PCORI Engagement Rubric: promising Practices for Partnering in Research. Ann Fam Med 2017; 15: 165–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Weerts D, Sandmann L.. Building a two-way street: challenges and opportunities for community engagement at research universities. Rev High Educ 2008; 32: 73–106. [Google Scholar]
- 8. Kenny A, Hyett N, Sawtell J. et al. Community participation in rural health: a scoping review. BMC Health Serv Res 2013; 13: 64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Hood N, Brewer T, Jackson R. et al. Survey of community engagement in NIH‐funded research. Clin Transl Sci 2010; 3: 19–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Wallerstein N. Empowerment to reduce health disparities. Scand J Public Health 2002; 30: 72–7. [PubMed] [Google Scholar]
- 11. Rhodes SD, Foley KL, Zometa CS. et al. Lay health advisor interventions among Hispanics/Latinos. Am J Prev Med 2007; 33: 418–27. [DOI] [PubMed] [Google Scholar]
- 12. Swider SM. Outcome effectiveness of community health workers: an integrative literature review. Public Health Nurs 2002; 19: 11–20. [DOI] [PubMed] [Google Scholar]
- 13. Mullins C, Abdulhalim A, Lavallee D.. Continuous patient engagement in comparative effectiveness research. JAMA 2012; 307: 1587–8. [DOI] [PubMed] [Google Scholar]
- 14. Minkler M. Linking science and policy through community-based participatory research to study and address health disparities. Am J Public Health 2010; 100: S81–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Minkler M, Vasquez V, Tajik M. et al. Promoting environmental justice through community-based participatory research: the role of community and partnership capacity. Health Educ Behav 2008; 35: 119–37. [DOI] [PubMed] [Google Scholar]
- 16. Minkler M, Wallerstein N.. Community-Based Participatory Research for Health: From Process to Outcomes. John Wiley & Sons, 2011. [Google Scholar]
- 17. Vaughn L, Jacquez F, Lindquist-Grantz R. et al. Immigrants as research partners: a review of immigrants in community-based participatory research (CBPR). J Immigr Minor Health 2017; 19:1457–68. [DOI] [PubMed] [Google Scholar]
- 18. Salsberg J, Parry D, Pluye P. et al. Successful strategies to engage research partners for translating evidence into action in community health: a critical review. J Environ Public Health 2015; 2015: 1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Matthews AK, Anderson EE, Willis M. et al. A Community Engagement Advisory Board as a strategy to improve research engagement and build institutional capacity for community-engaged research. J Clin Transl Sci 2018; 2: 66–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Matthews AK, Newman S, Anderson EE. et al. Development, implementation, and evaluation of a Community Engagement Advisory Board: strategies for maximizing success. J Clin Transl Sci 2018; 2: 8–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Quinn SC. Ethics in Public Health Research. Am J Public Health 2004; 94: 918–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. L SM, Allen F, C KD. et al. Lay patient navigator program implementation for equal access to cancer care and clinical trials. Cancer 2006; 107: 2669–77. [DOI] [PubMed] [Google Scholar]
- 23. O’Brien MJ, Halbert CH, Bixby R. et al. Community health worker intervention to decrease cervical cancer disparities in Hispanic women. J Gen Intern Med 2010; 25: 1186–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Kok M, Kane S, Tulloch O. et al. How does context influence performance of community health workers in low-and middle-income countries? Evidence from the literature. Health Res Policy Syst 2015; 13: 13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Saad-Harfouche FG, Jandorf L, Gage E. et al. Esperanza y Vida: training lay health advisors and cancer survivors to promote breast and cervical cancer screening in Latinas. J Community Health 2011; 36: 219–27. [DOI] [PubMed] [Google Scholar]
- 26. McQueen A, Kreuter MW, Kalesan B. et al. Understanding narrative effects: the impact of breast cancer survivor stories on message processing, attitudes, and beliefs among African American women. Health Psychol 2011; 30: 674–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Shippee N, Domecq Garces J, Prutsky Lopez G. et al. Patient and service user engagement in research: a systematic review and synthesized framework. Health Expect 2015; 18: 1151–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Schwartz LA, Brumley LD, Tuchman LK. et al. Stakeholder validation of a model of readiness for transition to adult care. JAMA Pediatr 2013; 167: 939–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Molina Y, San Miguel LG, Tamayo L. et al. The “Empowering Latinas to Obtain Breast Cancer Screenings” study: rationale and design. Contemp Clin Trials 2018; 71: 1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Wang AT, Fan J, Van Houten HK. et al. Impact of the 2009 US Preventive Services Task Force Guidelines on screening mammography rates on women in their 40s. PLoS One 2014; 9: e91399. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Molina Y, Thompson B, Espinoza N. et al. Breast cancer interventions serving US-based Latinas: current approaches and directions. Womens Health 2013; 9: 335–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Mona S, L IK, Linda C. et al. Cervical cancer screening and management practices among providers in the National Breast and Cervical Cancer Early Detection Program (NBCCEDP). Cancer 2007; 110: 1024–32. [DOI] [PubMed] [Google Scholar]
- 33. Harper S, Lynch J, Meersman SC. et al. Trends in area-socioeconomic and race-ethnic disparities in breast cancer incidence, stage at diagnosis, screening, mortality, and survival among women ages 50 years and over (1987-2005). Cancer Epidemiol Biomarkers Prev 2009; 18: 121–31. [DOI] [PubMed] [Google Scholar]
- 34. Molina Y, San Miguel LG, Tamayo L. et al. Empowering latinas to obtain breast cancer screenings: comparing intervention effects, part 2. Cancer Epidemiol Biomarkers Prev 2018; 27: 353–54. [Google Scholar]
- 35. Hsieh H-F, Shannon S.. Three approaches to qualitative content analysis. Qual Health Res 2005; 15: 1277–88. [DOI] [PubMed] [Google Scholar]
- 36. Bernard HR, Wutich A, Ryan GW.. Analyzing Qualitative Data: Systematic Approaches. Thousand Oaks, CA: SAGE Publications, 2016. [Google Scholar]
- 37. MacQueen KM, McLellan E, Kay K. et al. Codebook development for team-based qualitative analysis. Cult Anthropol Methods J 1998; 10: 31–6. [Google Scholar]
- 38. Creswell JW, Miller DL.. Determining validity in qualitative inquiry. Theory Pract 2000; 39: 124–30. [Google Scholar]
- 39. Guba EG, Lincoln YS.. Competing paradigms in qualitative research In: Denzin NK, Lincoln YS (eds). Handbook of Qualitative Research. Thousand Oaks, CA: SAGE, 1994, 105–17. [Google Scholar]
- 40. Morse JM, Barrett M, Mayan M. et al. Verification strategies for establishing reliability and validity in qualitative research. Int J Qual Methods 2002; 2: 13–22. [Google Scholar]
- 41. Freedman RA, Virgo KS, Yulei H. et al. The association of race/ethnicity, insurance status, and socioeconomic factors with breast cancer care. Cancer 2011; 117: 180–9. [DOI] [PubMed] [Google Scholar]
- 42. Talya S, Pamela SG, Olufunmilayo IO. et al. “Why take it if you don't have anything?” breast cancer risk perceptions and prevention choices at a public hospital. J Gen Intern Med 2006; 21: 779–85 [DOI] [PMC free article] [PubMed] [Google Scholar]