Engagement of community stakeholders to develop a framework to guide research dissemination to communities

Abstract Background Dissemination of research findings to past study participants and the community‐at‐large is important. Yet, a standardized process for research dissemination is needed to report results to the community. Objective We developed a framework and strategies to guide community‐academic partnerships in community‐targeted, dissemination efforts. Methods From 2017 to 2019, a community‐academic partnership was formed in Nashville, Tennessee, and iteratively developed a framework and strategies for research dissemination using cognitive interviews. A deductive, constant comparative analysis was conducted on interview responses to examine framework and strategy content. Feedback was used to finalize the framework and strategies for the evaluation. Using existing data, the framework's utility was evaluated in seven town hall meetings (n = 117). Bivariate analyses determined its effect on community members’ trust and willingness to participate in research using pre‐ and post‐surveys. Evaluation results were used to finalize the framework. Results The Community‐Engaged Research Dissemination (CERD) framework has two phases. Phase one is a preliminary planning phase with two steps, and phase two is the four‐step dissemination process. There are five standards to be upheld conducting these phases. We provide competencies for each component. Three feasible, culturally adapted strategies were developed as exemplars to disseminate research findings. Using pre‐ and post‐surveys for intervention evaluation, there was a significant difference in trust in medical research and researchers (P = .006) and willingness to participate in research (P = .013). Discussion and Conclusion The CERD framework can potentially standardize the process and compare the effect of dissemination efforts on the community's trust and willingness to participate in research.


| INTRODUC TI ON
Dissemination of research findings in translational research is necessary to facilitate uptake and adoption of interventions to improve health outcomes. 1 It involves the distribution of research findings to key stakeholders (e.g., providers, research participants, community-based organizations [CBOs] and community members), maximizing reach and benefit of findings for target communities. 2 However, traditional approaches to dissemination (e.g., academic publications) have limited reach and utility by patients, their families, and the community-at-large. 3,4 Schroter et al. (2019) found that only 27% of researchers of clinical trials disseminated results to the participants. 5 In a study reporting dissemination in community-based participatory research (CBPR), only 23% of researchers disseminated results to the public and 26% disseminated results to community participants at each stage of the research process (Chen et al 2010). 6 Lack of communication about study results to these groups, particularly minority communities, strains the community-academic partnership 7 and contributes to mistrust in medical research(ers) and the health-care system. 8 Subsequently, this affects CBOs' and community members' understanding and willingness to engage in research opportunities. 9,10 This delays the identification of new evidence-based knowledge, questions the generalizability of findings, and lessens use of research to improve health outcomes. 11 Stakeholders on multiple levels increasingly recognize the value in research dissemination to study participants and the community-at-large. Funders (eg Patient-Centered Outcomes Research Institute, National Institutes of Health and Agency for Healthcare Research Quality) now consistently request dissemination plans for the broader community and may request engagement of stakeholders in plan development to increase the use of findings. [12][13][14] Recognizing an ethical obligation to provide findings to the community, 14,15 researchers-particularly those engaged in CBPR-are increasingly looking for effective ways to do this. 14,16,17 Last, community members advocate for 'community-friendly' approaches to research dissemination. [18][19][20] While these strategies extend dissemination efforts to inform communities, 14,16,20 there is a need for a standardized, community-guided process for researchers across the translational research continuum to return research results to study participants and the community-at-large. 6 Many conceptual frameworks exist for dissemination relating to intervention applicability in health-care practice at community and clinical levels (i.e., dissemination and implementation science). [21][22][23] For example, Harris et al (2012)

developed the Health Promotion
Research Center Dissemination framework to promote uptake of evidence-based practices among user organization(s). 24 In this model, researchers collaborate with a CBO to refine the practice and approach to dissemination using the principles of social marketing. Adoption, implementation and maintenance are steps taken by disseminating organizations for a successful outcome of change in organizational practice and personal behaviours that lead to increased productivity and improved health. These frameworks offer important concepts on dissemination, and some further demonstrate the importance of stakeholder engagement in this process. 16,25 However, a practical framework is needed that provides a stepwise, community-guided approach to return individual research findings directly to research participants and the community-at-large. 6 This is important, as few studies report providing findings to communities throughout the research process. 18,26,27 Additionally, studies engaging community partners may vary in application of CBPR principles in dissemination efforts. 6 Developing a guide for researchers to return results to the community could: (a) expand our understanding of processes to determine the findings to be disseminated; (b) identify methods to develop dissemination strategies; (c) understand the relevance and importance of the dissemination process and methods to community; and (d) potentially impact trust, willingness and participation 2 as it relates to medical research.
The objective of this paper is to describe the development and evaluation of a practical framework and strategies to guide community-academic partnerships in dissemination efforts with CBOs and community members participating in all phases of research.
Specifically, we describe the development of a novel, community-driven framework that was used to provide research evidence to past research participants and the community-at-large using community-engaged research (CEnR) principles. This research identified the needs, priorities, and recommendations for underrepresented groups to participate in research. 18 This work was conducted through a community-academic partnership (one academic partner and two community organizations).

| ME THODS
To develop an effective, systematic process to improve academiccommunity partners' engagement in research dissemination to communities, we: (a) conducted a literature review to identify best practices and current frameworks; (b) developed an initial framework; (c) elicited feedback on the framework by conducting cognitive interviews with researchers, community leaders and members; (d) evaluated the framework, and (e) finalized the framework using study results. Collectively, we had expertise in clinical and translational research, community engagement, research dissemination and qualitative data analysis. This work was approved by the Institutional Review Boards of Meharry Medical College and Vanderbilt University.

| Phase I literature review
We conducted a systematic literature review applying a 'purposive' search and article selection to gather dissemination concepts. 28 This is not the traditional approach of exhausting the literature on the topic. 29 We conducted the review in PubMed, CINAHL and PsycINFO from September 2018 to May 2019 using key words 'dissemination', 'implementation', 'conceptual framework', 'research results', 'engagement', 'evidence-based' and 'partnership'. The 6177 selected articles identified guiding principles, recommendations, models, frameworks and interventions within the context of research dissemination to communities. Using these criteria, we identified a subset of 66 articles in which we cite a few examples here. 20,22,23 We completed the search once no new characteristics of dissemination emerged (i.e. saturation).

| Phase II development of initial research dissemination framework
We used the articles to identify key concepts of research dissemination. Specifically, we coded the articles and then linked the codes using a deductive-inductive approach. Then, we developed an initial framework depicting the process for dissemination to research participants and community-at-large. We further refined the framework based on our research experiences and community partner feedback.

| Phase III stakeholder review of initial framework
We purposefully recruited two leaders from our community partners, two Latino community leaders, 12 community members and 10 researchers to evaluate the initial framework. Recruitment sites were CBOs (e.g., community health centres, colleges and churches) representative of original study participants and the community-at-large. Eligibility criteria included: (a) being over 18 and (b) had received or would like to receive research findings.
One of our community partners was unavailable to participate at the time. However, we incorporated their feedback from the initial planning phase in framework development and while developing Trained in qualitative data collection and analysis, members of our partnership (i.e., two researchers and a graduate assistant) transcribed the interview data. Using a line-by-line coding technique, we coded the text with the a priori codes 'keep', 'remove', 'add' or 'clarify' as defined in the codebook. Using a deductive approach, we used a constant comparison analysis to iteratively compare codes to determine which steps and its process in the research dissemination framework should be kept, removed, added or clarified.
For both community and researcher participants, we identified all characteristics of dissemination and improved cultural appropriateness using peripheral, evidential and linguistic strategies (i.e., saturation). 31 Participants who agreed to follow-up were contacted for final comments, a qualitative verification procedure known as member checking. 32

| Phase IV evaluation of the framework's utility
To evaluate the proposed framework, we used a previously published 18 study to provide the research findings to study participants and the community-at-large. Using 11 community listening sessions, the study identified the research concerns and barriers to research participation among underrepresented populations (i.e., low socio-economic status, African American, Latino, deaf and hard of hearing). We postulated that disseminating research findings using the targeted, dissemination strategies described in the framework would increase trust in research and willingness to participate in research opportunities.

| Development and targeting of the dissemination strategies for framework evaluation
In order to disseminate research findings to past study participants and their communities, we iteratively develop three, culturallytargeted strategies: a one-pager, videos , and town hall meetings.
Nine additional participants (three from each community partner and three from the Latino community) were purposefully recruited for 60-minute cognitive interviews to determine cultural appropriateness of the strategies using constituent-involving, sociocultural, linguistic, evidential and peripheral strategies. 31 These participants were in the original study or were members of communities underrepresented in research. We asked participants about content, design, concerns and suggested revisions for each strategy. They were compensated with a $25 gift card. Interviews were audio-recorded. After being transcribed, members of our partnership (i.e., two researchers) coded the data line by line. They then analysed the data using a deductive, constant comparative approach to edit the dissemination strategies iteratively. This approach was comparable to that used in framework development.
We finalized the strategies.

| Data collection
We implemented seven town hall meetings throughout Nashville, Tennessee, to disseminate the research findings. We invited past study participants who had not participated in previous interviews and community members-at-large to participate. Our goal was 20 participants per town hall. We recruited using word of mouth, CBOs and flyers. On the day of the town hall meeting, participants viewed the PowerPoint presentation, video, and one-pager. We

| Statistical analyses
We used version 23 of the IBM Statistical Package for the Social Sciences (SPSS) to analyse the data. Frequencies identified participant demographics. The chi-squared test examined the bivariate association between willingness to participate in research pre-and post-town hall. A t test evaluated the bivariate association between trust in medical research and researchers pre-and post-town hall.
We set the alpha level at P < .05.

| Phase V framework finalization
We combined text excerpts from each post-survey and analysed the data using the codebook and comparable data analysis used in framework and strategy development. Using post-survey data and our experience, we finalized the content and structure of the framework. Specifically, we merged the information into the phases, steps and processes of the framework on research dissemination.

| Initial framework
Using the results of our literature search and experiences of the researchers, CBO leaders and community members, we iteratively created the initial framework. First, we identified and combined similar concepts related to the return of findings to the community.
Then, we identified initial steps for research dissemination and then mapped the concepts to those steps. Using those concepts, we created processes for the steps within the dissemination process.  Table 1 provides characteristics of community members, CBO leaders and researchers. All participants agreed the framework was necessary and included main components for research dissemination.

| Cognitive interviews: framework development
Positive feedback from both researchers and community included having community involvement and the ability to adapt to all stakeholders' needs; however, they differed in distinct ways. For example, a community member emphasized the importance of tailoring the dissemination strategy to meet their needs, while a researcher recommended adding ethics as a core value in the framework. Table 2 describes the interview results of researchers and the community along with how their input was integrated into the framework.

| Cognitive interviews: dissemination strategy development for framework evaluation
There was an even distribution among race [Caucasian (n = 3); African American (n = 3); and Hispanic (n = 3)]; however, the majority (67%) were female. The mean age was 47. Cognitive interview participants suggested modifying content of one-pager, video, and town hall meeting agenda to increase cultural appropriateness.
Based on the data, we made changes to each strategy (see Table 2).
There were no suggestions made to change the framework after showing the final framework to five participants who agreed to review post-interview and could be reached via the contact information provided. Below is a brief description of the strategies.

| Evaluation results
Response rates for the seven town hall meetings ranged from 11 to  Tables 3   and 4). Table 3 describes participants' willingness to take part in research pre-and post-town halls of dissemination of research findings. Table 4

| Final dissemination framework
The final Community-Engaged Research Dissemination (CERD) framework is a two-phase process for disseminating research in the community where phase one involves planning for dissemination effort and phase two involves conducting the dissemination process (see Figure 1). In phase one, step one involves the devel-

| D ISCUSS I ON
Researchers lack effective, evidence-based processes for research dissemination beyond traditional academic methods to 'end-users'.
In response, we provide one of the first frameworks and strategies

| Practical application of the framework
There are implications at the individual, interpersonal, community, organizational, policy and methodological levels when conducting CEnR dissemination. Dissemination strategies may vary at each level during framework application. Using this framework at the individual level is the most fundamental with the greatest potential for an individual to engage in a behaviour change to improve public health outcomes. 35 Engaging the community as partners in dissemination

| Limitations
This framework was developed using community member and researcher feedback. Insight on additional stakeholder (e.g., policy makers, grant funders) perspectives came solely from the literature.
Furthermore, representation of community and researchers could have been more diverse. However, this was an initial framework developed using CEnR and partnership principles to describe a dissemination process. Also, there was only one researcher who intended to disseminate results; however, she shared similar perspectives related to the dissemination process. Future studies should explore ways for sample representativeness for framework evolution. Next, while this framework was developed based on one dissemination project, (a) we gauged the views of researchers who had engaged in community research dissemination and community members who would like to receive or have received study findings, and (b) our implementation process appeared to result in an increase in willingness to participate in clinical research and trust in medical research and researchers on the part of community members. Future studies are needed to confirm these results.

| CON CLUS IONS
Researchers can apply the CERD framework to partner with communities and implement the proposed strategies to disseminate Ultimately, framework application could improve implementation of research dissemination efforts, creating a foundation for robust dissemination efforts. Implementing this process could build rapport and trust in communities, particularly those underrepresented in research, influencing research participation and public health outcomes.

ACK N OWLED G EM ENTS
The authors thank Neighbor 2 Neighbor, Bridges, and other CBOs and community members of Nashville, Tennessee, for assisting in the development, implementation and evaluation of this dissemination study. We also would like to thank the researchers who assisted in the development of the framework.

CO N FLI C T O F I NTE R E S T
None reported.

Ethical approval has been granted exemption by Meharry Medical
College Institutional Review Board (Protocol #18-07-844) for interviews conducted with researchers and community stakeholders to better understand the dissemination process while applying principles of community engagement.

DATA AVA I L A B I L I T Y S TAT E M E N T
Research data are not shared due to ethical restrictions.