Introduction
Timely implementation of evidence-based practices is critical to improve health care delivery and health outcomes across many disciplines and specialties, yet the time lag between research inception and translation of useful findings to common practice can be lengthy, and many worthwhile innovations are never fully implemented. Additionally, research intended to generate evidence may remain unpublished, rely upon outdated policies and practices, or not align with health system policies and priorities [1,2]. Dissemination and implementation (D&I) science provides a collection of research designs and methodologies to help clinicians and researchers translate evidence-based practices, interventions, and policies into real-world, sustained practice [1]. D&I science holds promise for reducing research waste by accelerating the uptake of new, evidence-based solutions to complex health and healthcare delivery problems [Reference Wuchty, Jones and Uzzi3,Reference Aarons, Reeder, Miller and Stadnick4,Reference Hall, Olster, Stipelman and Vogel5]. As a result, federal and nonfederal funders across many disciplines increasingly require the use of D&I science in studies and projects, thus increasing the demand for D&I experts who can support these undertakings through collaboration or consultation [Reference Tabak, Bauman and Holtrop6].
The skills and responsibilities of a research collaborator or project/study consultant differ from those of a principal investigator. The current demand for seasoned D&I collaborators exceeds supply, despite a growing number of training programs to support D&I science competency development [Reference Chambers, Proctor, Brownson and Straus7,Reference Schultes, Aijaz, Klug and Fixsen8,Reference Brownson, Colditz, Dobbins, Emmons, Kerner and Padek9,Reference Viglione, Stadnick and Birenbaum10,Reference Luke, Baumann, Carothers, Landsverk and Proctor11,Reference Stamatakis, Norton, Stirman, Melvin and Brownson12]. These programs are primarily focused on training individuals as independent investigators capable of applying for investigator-initiated D&I funding. Given the current demand, early career researchers (ECRs; i.e., individuals within 10 years of terminal degree or postdoctoral fellowship completion) are often approached to serve as D&I consultants and collaborators, despite potentially lacking skills or training specific to research collaborations, particularly with principal investigators in other fields or disciplines.
The paucity of guidance on building D&I-focused research collaborations is surprising given the breadth of challenges to interdisciplinary research [Reference Koorts, Naylor, Laws, Love, Maple and van Nassau13,Reference Guerrero, Hahn, Khachikian, Chuang and Brown14] that have been characterized in the D&I literature. Recognizing the unmet need for collaborative research support, Tabak et al [Reference Tabak, Bauman and Holtrop6]. outlined hypothetical scenarios that D&I researchers are likely to encounter when working with scientists unfamiliar with D&I concepts and methods, and offered practical suggestions on the appropriateness of study team roles (e.g., when the expected scope of D&I contributions indicates the need for a consultant versus a co-investigator role).
Our project was prompted by a guided discussion of the Tabak et al article during an interdisciplinary D&I journal club supported by a training grant (K12HL138030). This journal club is led by a senior D&I scientist and largely attended by ECRs, with representation of multiple academic institutions, healthcare systems, and academic disciplines. Prompted by the following four questions, attendees discussed how the collaboration scenarios and recommendations outlined by Tabak et al., might apply to ECRs:
-
1. What are the marketable skills of an early career D&I researcher?
-
2. What questions should an early career D&I researcher ask when approached about becoming a research collaborator?
-
3. What are the responsibilities of a D&I collaborator during the grant-writing process?
-
4. What range of responsibilities should a D&I collaborator be willing to accept if the grant obtains funding?
Journal club attendees responded to these prompts in real time using Google’s virtual Jamboard platform. ECRs reported that they desired guidance on how to better communicate their D&I skillset to scientists in other disciplines, how to advocate for themselves to serve as funded co-investigators on grants, and how to negotiate for the resources needed for successful execution of D&I research activities as part of larger grants. This discussion confirmed a gap in the D&I literature pertaining to guidance for cultivating and sustaining effective interdisciplinary research collaborations and affirmed that ECRs viewed engagement in collaborative research as essential for developing their broader D&I expertise.
The purpose of this study, therefore, was to identify a consensus-derived set of collaboration recommendations for early career D&I researchers and to develop guidance that can help ECRs more effectively collaborate with scientists across a range of research disciplines.
Materials and methods
This iterative, systematic, multi-phased study (Figure 1) used a consensus-building approach previously applied in the D&I literature [Reference Lewinski, Sullivan, Allen, Crowley, Gierisch and Goldstein15,Reference Powell, Waltz and Chinman16,Reference Hull, Goulding and Khadjesari17] to identify recommendations for research collaborations specifically for ECRs specializing in D&I. All study activities were reviewed and determined exempt by the Duke University Health System Institutional Review Board (Pro00108634). Supplementary File 1 details our decision process across phases. Supplementary File 2 includes the survey instruments and Supplementary File 3 includes guiding slides and script for our focus groups.
Study flow diagram. Legend: ECR = early career researcher; D&I = dissemination and implementation.
Phase 1: Development of D&I collaboration domains
Four authors of this manuscript (HL, SDA, JS, AAL – all ECRs) reviewed and summarized the general comments raised during the journal club, identifying three distinct areas of attention for ECRs as they establish their D&I career [1]: “Marketing yourself” (i.e., describing your value to potential non-D&I research collaborators in order to locate and maintain engagement with collaborators) [2], “Collaboration considerations” (i.e., clarifying expected contributions of the D&I collaborator during proposal development), and [Reference Wuchty, Jones and Uzzi3] “Responsibilities once project is initiated” (i.e., defining your role during project execution). With guidance from a D&I expert (HBB), the four authors thematically grouped journal club responses into 12 domains across these three areas.
Phase 2: Refinement of D&I collaboration domains via consensus-building
Phase 2 consisted of building consensus among other D&I scientists on the 12 initial domains and constructing preliminary recommendations through a survey and series of focus groups.
Survey 1
Overview
We designed a descriptive survey (Survey 1; Supplementary File 2) based on prior studies [Reference Forrest, Chesley, Tregear and Mistry18,Reference Waltz, Powell, Fernández, Abadie and Damschroder19] to determine the clarity of the initial domains and their relative importance to interdisciplinary research collaborations. The authors distributed this survey to several D&I colleagues to test its wording and clarity. We then pilot-tested the survey with input from several D&I experts before deploying it via Qualtrics in early summer 2021.
Recruitment
We sought to recruit a mix of ECRs and senior researchers with varied experiences in D&I (i.e., principal investigators, collaborators, and mentors). We used convenience and snowball sampling to recruit participants via listservs and emails to colleagues. All survey responses were anonymous. On completion of the survey, participants could provide their contact information to be entered to win a $50 gift card.
Survey items
The survey consisted of 24 items that solicited demographic information and participants’ reactions to the domains identified in Phase 1. Participants were asked to (a) rate the importance of each associated domain (n = 12) on a 3-point scale (1 = not that important to 3 = very important) to effective research collaborations with investigators from other disciplines, and (b) describe via open-ended text any needed clarifications of the existing domains. Participants could also propose new domains.
Analysis plan
We used descriptive statistics (e.g., frequencies, means) to describe ratings, and two team members independently reviewed, summarized, and analyzed the open-ended responses. The authors met as a group to review findings and achieve consensus on which domains required further development (Supplementary File 1).
Focus groups
Overview
We conducted virtual focus groups to further explore Survey 1 findings and generate additional feedback on the initial domains. Three focus groups intending to last approximately 45 minutes occurred via Zoom sequentially, approximately one week apart, in late summer 2021.
Recruitment
Survey 1 participants were asked to indicate their interest in focus group participation. Those who expressed interest were contacted via email to confirm participation. Participants received $10 gift cards.
Structure
A moderator and two notetakers from the authorship team conducted each focus group. The moderator established ground rules for discussion and obtained verbal consent to record. Focus groups were audio recorded, but the recordings were not transcribed; notetakers summarized key discussion points in real-time. The focus group interview guide was developed based on the survey findings (Supplementary File 3). The focus groups were iterative, with each discussion informed by prior group findings.
Analysis Plan
We analyzed data using rapid qualitative analysis with purposeful data reduction techniques [Reference Lewinski, Crowley and Miller20,Reference Lewinski, Shapiro, Bosworth, Crowley, McCant and Howard21,Reference Gonzales, Okusaga, Reuteman-Fowler, Oakes, Brown and Moore22]. At the end of each focus group, the moderator and notetakers held a debriefing session to discuss findings and to identify topics for discussion in subsequent focus groups. Debriefing sessions were audio recorded but not transcribed. We created a summary of findings for each domain based on our detailed focus group notes and summaries of the debriefing discussions. Summaries were integrated with the survey results after completion of the three focus groups (Supplementary File 1).
We presented select findings to the D&I journal club attendees, who provided feedback for further refinement. By the end of Phase 2, the team compiled 29 activities within 9 refined domains (reduced from the 12 initial domains). Within each domain, we also began to identify specific ECR activities or behaviors that could promote more successful research collaborations (Supplementary File 1).
Phase 3: Refinement of D&I collaboration domains and activities
Overview
Phase 3 consisted of a second survey (Survey 2; Supplementary File 2) that was designed to further clarify domains and transition proposed activities into specific recommendations. The survey was conducted via Qualtrics in early fall 2021.
Recruitment
As with Survey 1, we sought to achieve a mix of senior researchers and ECRs with varied experiences in D&I. We used convenience and snowball sampling to recruit participants via listservs, emails disseminated through colleagues, and social media. Participants who completed Survey 1 were eligible to complete Survey 2. At the end of the survey, participants could provide their contact information to win one of several $50 gift cards.
Survey items
The survey consisted of 39 items pertaining to (a) demographic information, and (b) reactions to the 9 refined domains and 29 proposed activities. We randomized the order in which the areas were presented in order to minimize participant burden and increase the likelihood of obtaining adequate responses for each area. For each domain, we asked respondents to “Indicate if you recommend changes and, if so, describe what is needed to provide additional clarity.” Respondents could then suggest additional domains or activities they considered essential to effective D&I research collaborations.
Analysis plan
We used the same processes described in Survey 1 to analyze the quantitative and qualitative data, including assessing recommendations for changes to content and clarity within each domain. As part of Phase 3 analysis, we presented select findings to an implementation science workgroup (consisting of D&I scientists, practitioners, and research staff) to ensure the findings’ sufficiency to guide additional changes to our domains and associated activities. Phase 3 analysis is described in Supplementary File 1.
Phase 4: Development of final list of domains and recommendations
In Phase 4, we further refined each domain and drafted relevant recommendations related to each activity using Bloom’s taxonomy [23] to ensure that our recommendations could ultimately be translated to achievable learning outcomes. The list of recommendations underwent 3 iterative rounds of review by members of the authorship team (HL, SA, JS, AA) to eliminate duplicative language, refine wording, confirm fit within the underlying domain, and assess coverage of the 3 original areas of focus. Additionally, during this review, we developed data-driven guidance for actions or activities related to each recommendation. Finally, we engaged two D&I experts (AB, HBB) to provide feedback on the domains, recommendations, and guidance. In particular, AB provided perspective on intersections of D&I and health equity research that emerged from our recommendations, and HBB provided perspective on mentoring ECRs in health services research and implementation science.
Results
Herein we describe findings from Phases 2–4 of the study. The final list of domains and recommendations are described in Figures 2–4 and Tables 1–3. A more detailed description of our results across phases is described in Supplementary File 1.
“Marketing Yourself” domains and recommendations. Legend: D&I = dissemination and implementation.
“Collaboration Considerations” domains and recommendations. Legend: D&I = dissemination and implementation.
“Responsibilities Once Project is Initiated” domains and recommendations. Legend: D&I = dissemination and implementation.
“Marketing Yourself” guidance and examples by domain
Legend: ECR = Early Career Researcher; D&I: Dissemination and Implementation; VA: Veteran’s Administration’; NIH: National Institutes of Health.
“Collaboration considerations” guidance and examples by domain
Legend: D&I = Dissemination and Implementation; FTE = Full-Time Equivalent.
“Responsibilities Once Project is Initiated” guidance and examples by domain
Legend: D&I = Dissemination and Implementation; FTE = Full-Time Equivalent.
Sample
Survey 1 received 47 responses of which 44 (94%) were sufficiently complete to include in the analysis. Most respondents self-identified as cis-women (82%), white (84%), and non-Hispanic (95%); and half (50%) identified as ECRs. Most (87%) identified as a D&I researcher, and 74% reported having D&I training or experience. Forty-three percent reported having served as principal investigator on a D&I project, 69% as having collaborated on a D&I project, and 48% as having mentored D&I ECRs. Of the 44 respondents, 16 indicated interest in attending a focus group, and 12 attended one of the three focus groups (range of attendees n = 3–5; average length = 47 minutes). Focus group participants included a mix of ECRs and experienced researchers.
Survey 2 received 49 responses, 29 (59%) of which were sufficiently complete to include in this analysis. Of the 29 participants, 69% identified as cis-women, 90% as white, and 90% as non-Hispanic. All identified as D&I researchers, 62% as ECRs. Fifty-two percent had served as principal investigator on a D&I project, 52% had collaborated on a D&I project, and 34% had served as a consultant on a D&I project.
Refining domains
Herein we briefly describe the refinement process across the three areas. Supplementary File 1 provides additional details.
Marketing yourself
In Survey 1, most respondents rated domains in this area as “very important”; however, the domain describing specific D&I training was rated “very important” by only slightly over half of respondents. In open-ended text, respondents suggested adding the following domains [1]: access to D&I resources and mentors [2], experience working with operational/community partners, and [Reference Wuchty, Jones and Uzzi3] communication skills related to D&I terminology.
In focus groups, the relative importance of describing specific D&I training sparked considerable debate. Several participants noted that although training is important, it may be less accessible than applied experience, particularly for under-represented ECRs, and therefore should not be viewed as more valuable than experience. Focus groups also suggested the need for recommendations within an additional domain: describing one’s communication-related skills and experiences with both research teams and operational partners.
In Survey 2, participants provided suggestions for clarity and/or content modifications for all 10 activities within this domain. Suggested recommendations pertained to knowledge of health equity, knowledge of international D&I landscape, and experience as D&I consultants and/or trainers.
Collaboration considerations
In Survey 1, nearly all respondents (98%) ranked defining your role, as “very important.” In open-ended text, respondents [1] indicated confusion and overlap regarding the principal investigator’s commitment to the project [2]; emphasized the need to assess how the D&I components of a project will be appropriately resourced (e.g., staff time and effort, support for implementation activities and partnership-building); and [Reference Wuchty, Jones and Uzzi3] identified that most domains in this area were important for collaboration in general but not specific to the D&I field.
In focus groups, the importance of clarifying roles and team dynamics dominated the discussion. Participants outlined many functions that a D&I collaborator could potentially serve (e.g., a “go between” person connecting study team with operational partners; delivery agent; instrument developer; data collector on a project, grant, or study). Participants emphasized the importance for ECRs of avoiding common collaborator “traps” (e.g., over-involvement in data collection, not requesting sufficient resources to complete the D&I work, being seen as a team member rather than team leader) that could impede career advancement. In addition, several participants noted the need for frequent communication and “expectation management” between ECR D&I collaborators and primary study principal investigators who do not have D&I expertise.
In Survey 2, clarity and content edits were suggested for eight of the nine activities. Suggestions included aligning one’s role/effort with your title, garnering peer support for D&I work, troubleshooting challenges, and communicating one’s contribution to non-D&I experts.
Responsibilities once project is initiated
In Survey 1, most respondents scored working well with others as “very important” but suggested in open-ended text that this was not specific to D&I collaborations. Open-ended text also indicated that setting expectations was more appropriate for the Collaboration Considerations area and that having access to mentors and resources, strong partnerships, and continuous learning was important across all stages of collaboration.
Focus groups suggested revision of existing domains related to project and resource management, and incorporation of an explicit domain focused on mentoring, partnerships, and learning. Participants indicated the importance of a priori communication with the principal investigator about both the iterative nature of D&I research and the expectation that refinement of the scope of the D&I component may be inevitable. Although attendees agreed on the importance of mentorship and teaching for ECRs across the research continuum, they had mixed opinions regarding whether, when, and how ECRs should engage as mentors and teachers. It was noted that because few institutions have a “deep bench” of D&I scientists, such responsibilities may inappropriately fall to ECRs.
In Survey 2, recommendations for clarity and/or content edits were made for all nine activities. Suggested additions included time and resource management related to strategic partner involvement, networking and mentoring, and science communication.
Findings across 3 areas
Several interdependent topics spanned the three areas, including building and sustaining partnerships, navigating team dynamics, and communicating clearly and effectively to multiple types of partners. The importance of being able to articulate the “value added” of a D&I skillset for a given project was also noted, as was the need to consider health equity concepts. A particularly salient cross-cutting topic was when and how to seek mentorship from more senior D&I experts during the collaboration process; participants viewed this understanding as critically important for ECRs attempting to implement our collaboration recommendations. These topics were incorporated into various recommendations to reflect their importance across research stages.
Finalizing the set of recommendations
The collaborative consensus-building process elicited D&I collaboration behaviors that align with the 3 ECR-identified areas of interest from journal club comments. The 25 total recommendations, as well as the guidance to achieve them based on our data, are detailed in the accompanying figures and tables. The first area, Marketing yourself, contains 3 domains and 9 recommendations (Figure 2 , Table 1). The second area, Collaboration considerations, consists of 3 domains and 9 recommendations (Figure 3 , Table 2). The third area, Responsibilities once project is initiated, consists of 3 domains and 7 recommendations (Figure 4 , Table 3).
Discussion
Translational research collaborations involving D&I scientists, both within and outside their disciplines, are becoming increasingly common and form a critical component of an ECR’s career development and advancement [Reference Tabak, Bauman and Holtrop6]. Through an iterative, multi-phase consensus-building process, we identified specific recommendations that can help ECRs build productive research collaborations to advance D&I science. Our data support that (a) interdisciplinary research collaborations are critical and likely to occur throughout a D&I scientist’s career, (b) such collaborations hold a unique set of challenges due to the interdisciplinary nature of D&I, and (c) mentorship is critical to obtaining the skills needed to collaborate effectively as a D&I scientist.
Interdisciplinary research collaborations are critical and likely throughout a D&I scientist’s career
Invitations to participate in collaborative D&I research can greatly benefit ECRs professionally but may come at a time of transitioning from pre- or postdoctoral training to formalized research appointments [Reference Tabak, Bauman and Holtrop6]. Critically, there remains limited guidance for ECRs on cultivating productive research collaborations with scientists who are unfamiliar with D&I concepts or methods [Reference Tabak, Bauman and Holtrop6,Reference Chambers, Proctor, Brownson and Straus7]. The lack of guidance may be due in part to the underrepresentation of recommendations for collaboration among researchers in existing D&I competency frameworks. To date, various competencies for D&I areas have been published: D&I science [Reference Brownson, Colditz, Dobbins, Emmons, Kerner and Padek9,Reference Gonzales, Handley, Ackerman and OʼSullivan24,Reference Padek, Mir and Jacob25,Reference Shea, Young, Powell, Rohweder, Enga and Scott26], implementation practice [Reference Schultes, Aijaz, Klug and Fixsen8,Reference Metz, Albers and Burke27,Reference Tabak, Padek and Kerner28], knowledge translation [Reference Mallidou, Atherton, Chan, Frisch, Glegg and Scarrow29,Reference Straus, Brouwers and Johnson30], and learning health systems [Reference Forrest, Chesley, Tregear and Mistry18]. Knowledge and skills related to research collaborations have been previously identified as obligate competencies for D&I researchers [Reference Schultes, Aijaz, Klug and Fixsen8]; however, in most D&I competency frameworks, the primary collaboration-related focus is on fostering research-practice partnerships and community partner engagement in D&I research [Reference Gonzales, Handley, Ackerman and OʼSullivan24,Reference Shea, Young, Powell, Rohweder, Enga and Scott26,Reference Tabak, Padek and Kerner28,Reference Mallidou, Atherton, Chan, Frisch, Glegg and Scarrow29]. Findings from this study suggest that the skills needed for ECRs to engage in and sustain research collaborations with non-D&I scientists may differ, at least in part, from the knowledge, skills, and other abilities needed to engage in and sustain relationships with practice partners. Such skills should be incorporated into D&I training and mentoring.
Collaboration has benefits but also poses challenges for D&I ECRs
Our findings suggest that ECRs embarking on a research career in a team-based and interdisciplinary field such as D&I can expect challenges and benefits. Benefits of the highly interdisciplinary nature of D&I work include a breadth of literature to guide research and practice, senior mentors willing and able to usher in a new generation of D&I scientists, and funding to add D&I scientists to grants and projects. In both the Collaboration considerations and Responsibilities areas, study participants identified myriad roles and functions a D&I collaborator could potentially serve (and roles they should not serve) to benefit themselves and the project. Additionally, the nature of D&I research enables nontraditional products and dissemination channels that can expedite and expand the impact of research findings. We identified several cross-cutting recommendations across the three areas, including building and sustaining partnerships, navigating team dynamics, and communicating clearly and effectively to researchers with limited D&I expertise. Similar to Tabak et al., we recommend being upfront in defining and establishing role expectations and ensuring alignment with the ECR’s skillset, available resources, and career trajectory [Reference Tabak, Bauman and Holtrop6].
Challenges exist that should be acknowledged. For example, ECRs should ensure that the scope of the D&I work is realistic given the proposed D&I aims and allocated resources and effort. D&I research is complex; thus our recommendations provide guidance on how to approach the principal investigator in the context of the project and the ECR’s professional advancement, craft reasonable research questions, and advocate for adequate resources during a project. Early in the research collaboration, ECRs should emphasize the potential for, and importance of, nontraditional career metrics that may differ from the home institution’s promotion metrics but are normalized in the D&I field (e.g., publishing methodological approaches early in studies, developing dissemination products to operational partners) and could expand the reach of the study’s findings [Reference Stamatakis, Norton, Stirman, Melvin and Brownson12]. An additional challenge is the time, effort, and cost involved in collaborative D&I research approaches, particularly in instances in which essential tasks (e.g., relationship building, hiring staff with unique expertise, ensuring fair and just compensation for partners, promoting the willingness of the team to engage in collaboration throughout the project) may have implications for researchers’ promotion and tenure. As a field, more work is needed to incentivize implementation manuscripts – while findings from successful clinical trials may be published in high-impact journals, findings that inform implementation, sustainability or scalability of an efficacious program or intervention may have a broader societal impact. However, individual ECRs may need to weigh their institution’s internal promotion metrics against the value of nontraditional publications and transdisciplinary research as they contemplate collaborations within and outside of their institution [Reference Maddox, Phan and Byeon31]. Our recommendations prompt the ECR to be explicit about the challenges and potential benefits of implementation-focused projects and products with principal investigators, research teams, and mentors.
Training and mentorship are critical throughout D&I collaborations
We identified several considerations for the field regarding the preparation of ECRs as D&I collaborators. First, there was some debate within our study sample about the importance of formal D&I training given the nascency of the field; however, as the field expands, literature proliferates, and demand for expertise increases, it is likely that credentialed learning opportunities (e.g., certificate programs) will become more essential to collaboration success [Reference Viglione, Stadnick and Birenbaum10]. Formal training opportunities that span disciplines are becoming more widespread – a 2023 review identified 165 D&I capacity-building programs, with activities ranging from providing resources/tools and webinars to coursework and formal mentorship [Reference Viglione, Stadnick and Birenbaum10]. From a health equity perspective, our study participants noted that D&I training may be less accessible to under-represented ECRs and that senior mentors may be hesitant to collaborate with ECRs from different backgrounds. It is important that the field (a) more explicitly acknowledge the complexities of equity in D&I training [Reference Maddox, Phan and Byeon31], (b) ensure that equity is incorporated into training expectations for ECRs, and (c) focus on expanding access to D&I mentoring for under-represented ECRs, including offering more training in cultural competence for senior mentors as well as investing in efforts to recruit under-represented D&I experts to faculty positions.
An ECR’s career development and future success can be enhanced by interactions with traditional mentors as well as sponsors and people connected to various relevant networks. A particularly salient finding in our study that aligns with prior work [Reference Stamatakis, Norton, Stirman, Melvin and Brownson12,Reference Jacob, Gacad and Pfund32] was the importance of access to mentors with established D&I and non-D&I networks who could provide opportunities, offer methodological insight, act as a champion, and serve as resources should the ECR have collaboration or scientific questions. Study participants viewed such mentors as crucial to ECRs’ success in meeting key career benchmarks. As Tabak et al. note, having a network can lend credibility to an ECR who has less experience [Reference Tabak, Bauman and Holtrop6]; however, given the nascency of the field, there are increasingly few “experienced” D&I researchers to serve as a mentor for ECRs. Peer mentoring units, such as the one from which this paper was developed, can help fill training and knowledge gaps for ECRs if more senior mentorship is unavailable, as long as roles and responsibilities are shared across the group [Reference Dickson, Glass, Barnett, Graham, Powell and Stadnick33]. Peer mentoring units can also offer an avenue for establishing new interdisciplinary collaborations. As noted by Luke et al, peers participating in a structured D&I training program reported new projects, grants, and writing collaborations [Reference Luke, Baumann, Carothers, Landsverk and Proctor11]. Our work can serve to guide discussions among ECRs and both their D&I and non-D&I mentors because it incorporates recommendations regarding points in one’s career at which additional mentorship may be needed, how to access mentors (e.g., through training programs), types of mentorships which may be helpful, and areas in which a mentor can provide specific guidance.
ECRs should consider how health equity and D&I science intersect in collaborations
Throughout our study, participants described the importance of incorporating health equity into our recommendations. Participants suggested that a health equity-informed D&I ECR should be able to articulate ways in which D&I research approaches can advance health equity; demonstrate familiarity with applying health equity frameworks to D&I; and describe their experiences of having collaborated and/or engaged with diverse team members and collaborators. Although progress at the intersection of health equity research and D&I science is continuously evolving, we offer suggestions to support ECR in this regard: First, ECRs with only a fundamental knowledge of health equity science or scholarship should seek opportunities to partner with and co-learn from individuals who have acquired health equity expertise and/or individuals with lived experience of health inequities and/or systematic discrimination, rather than participating in “health equity tourism” [Reference Lett, Adekunle and McMurray34]. ECRs should seek to learn from the vast historical literature in health equity, and stay abreast of more current literature from experts who intersect health equity and D&I science in order to establish guiding principles for their collaborative research (e.g., framework selection and instrument development and adaptation). Second, ECRs should practice reflexivity – acknowledging and articulating their individual and professional equity-related strengths and limitations, and how those strengths and limitations may influence their role in a given study or within a given discipline. For example, an ECR who is very familiar with organizational metrics developed for D&I may still be unequipped to systematically adapt those metrics for use within historically minoritized populations. Third, in considering whether to collaborate on a given study, D&I ECRs should consider the study principal investigator and research team’s stated commitment to addressing inequities (e.g., budget for equity consultant, time built in for relationships-building), familiarize themselves with the role and reputation of the organization(s) involved in the project within the community, and encourage community engagement and co-learning from the outset of the study [Reference Baumann and Long35,Reference Kerkhoff, Farrand, Marquez, Cattamanchi and Handley36].
Limitations and strengths
Our study has several limitations and strengths which should be acknowledged. First, our study sample is small and homogenous (sample primarily self-identified as cis-women and/or white). Future efforts are needed to refine recommendations and establish competencies through a broader set of perspectives, including both researchers and practitioners belonging to marginalized groups, in various other settings in which D&I work is conducted (e.g., community locations, nonacademic health care systems), and in other geographical locations (USA, Europe, Australia, Latin America, Africa). Second, we recruited through convenience and snowball sampling in our networks, thus some individuals may have participated in both surveys, which may have limited the number of participants or the reach of this study. Third, although we made efforts to decrease the amount of time it took to complete the survey, about 40% of individuals who started the second survey did not complete it, which may have been due to the number of text-heavy and open-ended questions.
Despite these limitations, our study has several strengths: First, during our study conceptualization, data collection, analysis, and interpretation, we leveraged the expertise of many D&I researchers and practitioners of different career lengths who represented various practice locations and backgrounds and diverse areas of content and expertise. Second, the domains and recommendations were iteratively developed across several phases, which enabled us to discuss and refine the domains and recommendations. Third, we sought and obtained feedback from a diverse group of individuals who engage in D&I projects and studies, including ECR mentees and senior D&I research mentors, who provided valuable insights into career levels for D&I work.
Conclusion
Productive, interdisciplinary research collaborations are essential to advance clinical and translational research. ECRs trained in D&I may require additional support to enhance their collaborative research practice, but the need for D&I-specific collaboration competencies remains open for consideration. Overall, our list of domains and recommendations can assist ECRs in promoting and leveraging their diverse skillset within D&I and non-D&I-focused projects. Future work is needed to integrate these recommendations as key competencies in fundamental D&I trainings as well as for the D&I mentoring workforce. Findings lay the groundwork for future studies to expand recommendations (e.g., by eliciting practitioners’ perspectives) and empirically assess the extent to which the recommendations impact both the quality and quantity of D&I ECRs’ scientific collaborations.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/cts.2024.684.
Acknowledgments
The authors thank Donnalee Frega, PhD for editorial assistance. The authors thank the attendees of the Implementation Science Journal Club at Duke University, particularly Lindsay Ballengee and Kate Sheahan, and the participants across our study phases.
Author contributions
HL, SDA, JS, AAL: conceptualized study, developed study materials, collected data, analyzed and interpreted data, and wrote the manuscript. LB, KS: conceptualized study and assisted in manuscript editing. HBB: conceptualized study, reviewed study materials and data collection practices, interpreted data, and assisted in manuscript editing; AB: interpreted data and assisted in manuscript editing.
Funding statement
The research reported in this publication was supported by the following: Implementation Science Research Collaborative at the Department of Population Health Sciences at the Duke University School of Medicine (INTERACT); Durham Center of Innovation to Accelerate Discovery and Practice Transformation grant #CIN 13–410 and Veterans Health Affairs Health Services Research & Development grants #18–234 (to AAL); National Heart, Lung and Blood Institute grant ID K12HL138030 (to HL and HBB); and National Cancer Institute grant ID P50CA244431 and National Institute of Mental Health grant ID P50MH122351 (to AAB). The content is solely the responsibility of the authors and does not reflect the position or policy of Duke University, the U.S. Department of Veterans Affairs, or the U.S. Government.
Competing interests
Dr Lewinski reports receiving funds from PhRMA Foundation and Otsuka, unrelated to this study. Dr Bosworth reports research funding through his institution from BeBetter Therapeutics, Boehringer Ingelheim, Esperion, Improved Patient Outcomes, Merck, NHLBI, Novo Nordisk, Otsuka, Sanofi, VA. Elton John Foundation, Hilton Foundation, Pfizer. He provides consulting services for Abbott, Esperion, Imatar, Novartis, Sanofi, Vidya, Walmart, and Webmed. He has served on the board of directors of Preventric Diagnostics. The remaining authors have no competing interests to declare.
Ethics approval
The Duke University Health System Institutional Review Board reviewed and approved all study activities and provided a Declaration of Exemption (Pro00108634).
Open access
