Trainee-led research using an integrated knowledge translation or other research partnership approaches: a scoping reviews

Background There are increasing expectations for researchers and knowledge users in the health system to use a research partnership approach, such as integrated knowledge translation, to increase the relevance and use of research findings in health practice, programmes and policies. However, little is known about how health research trainees engage in research partnership approaches such as IKT. In response, the purpose of this scoping review was to map and characterize the evidence related to using an IKT or other research partnership approach from the perspective of health research trainees in thesis and/or postdoctoral work. Methods We conducted this scoping review following the Joanna Briggs Institute methodology and Arksey and O’Malley’s framework. We searched the following databases in June 2020: MEDLINE, Embase, CINAHL and PsycINFO. We also searched sources of unpublished studies and grey literature. We reported our findings in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews. Results We included 74 records that described trainees’ experiences using an IKT or other research partnership approach to health research. The majority of studies involved collaboration with knowledge users in the research question development, recruitment and data collection stages of the research process. Intersecting barriers to IKT or other research partnerships at the individual, interpersonal and organizational levels were reported, including lack of skills in partnership research, competing priorities and trainees’ “outsider” status. We also identified studies that evaluated their IKT approach and reported impacts on partnership formation, such as valuing different perspectives, and enhanced relevance of research. Conclusion Our review provides insights for trainees interested in IKT or other research partnership approaches and offers guidance on how to apply an IKT approach to their research. The review findings can serve as a basis for future reviews and primary research focused on IKT principles, strategies and evaluation. The findings can also inform IKT training efforts such as guideline development and academic programme development. Supplementary Information The online version contains supplementary material available at 10.1186/s12961-021-00784-0.


Introduction
Collaborative research approaches, such as coproduction, codesign, engaged scholarship and integrated knowledge translation (IKT) [1], aim to produce relevant Open Access *Correspondence: ccassidy@dal.ca 1 School of Nursing, Dalhousie University, Halifax, NS, Canada Full list of author information is available at the end of the article research findings to address healthcare issues. IKT, specifically, focuses on making research more useful through research partnerships. IKT is defined as "a model of collaborative research, where researchers work with knowledge users who identify a problem and have the authority to implement the research recommendations" [2]. Studies have shown that use of an IKT approach improves the quality of research [3], enhances value for research among decision-makers [4], increases capacity among decision-makers for engaging in research [4][5][6], and creates more impactful and useful research findings [3,7,8]. Research partnerships have been shown to be critical during the COVID-19 pandemic, as patients, citizens, healthcare providers, researchers, policy-makers and health system leaders from around world have come together to collectively address this global crisis [9].
Despite the value of IKT and other research partnership approaches, studies report many challenges in establishing and maintaining research partnerships with knowledge users [3,4,10]. Significant time is needed to develop trusting, authentic relationships, and there may be insufficient resources to support partnership development and maintenance [10]. Further, differing needs and priorities among researchers and health system decision-makers [3,4,10], as well as unclear goals, roles and expectations, can hinder research partnerships [4].
Health system leaders have identified a lack of researcher preparation for engaging in collaborative partnerships as a significant barrier to successful research partnerships [11]. This can lead to ineffective researcher behaviour (e.g., mismatch of researcher interests and organizational needs, lack of researcher understanding of health system context, lack of respect) and affect the development of positive, mutually beneficial research partnerships [11]. Other studies have shown that researchers require specific knowledge and skills for working in partnership with health system decisionmakers [12]. However, researchers often do not have the opportunity to learn how to establish effective research partnerships with knowledge users in the health system [10].
Specific training in IKT or other research partnerships is needed to promote collaborative health research moving forward. Most graduate students do not receive formal training in collaborative health research approaches [13,14]. Efforts are needed to support trainees, defined as graduate students (master's, doctoral) and postdoctoral fellows, in developing an understanding of health system context and skills to engage in collaborative research partnerships [11,15]. Such training efforts are essential to support trainees in building trusting, effective relationships with knowledge users to foster meaningful, ethical research with relevant outcomes [16]. Currently, the IKT or research partnership literature describes strategies, barriers and facilitators to research partnerships from the perspective of researchers and knowledge users [10,11,[17][18][19]. However, little is known about how trainees engage in research partnership approaches such as IKT. This is needed as a first step to inform research trainees on the use of IKT or other research partnership approaches and future academic training modernization efforts. As such, the purpose of this scoping review was to map and characterize the evidence related to using an IKT or other research partnership approach from the perspective of health research trainees in thesis and/or postdoctoral work.

Methods
We conducted this scoping review following the Joanna Briggs Institute (JBI) methodology [20,21] and Arksey and O'Malley's framework [22]. Our a priori protocol has been published previously [23]. This full report followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis Extension for Scoping Reviews (PRISMA-ScR) checklist [24] (Additional File 2).

Stage 1: Identifying the research question
We conducted a scoping review to map and characterize the available evidence related to using an IKT approach or other research partnership approach from the perspectives of trainees in thesis and/or postdoctoral work. Specifically, we answered the following research question: 1. How have IKT or other research partnership approaches been applied in thesis and/or postdoctoral health research?
Additional research objectives included: a. Identifying IKT/research partnership principles, strategies and/or tools used in trainee-led health research b. Identifying barriers and facilitators to using IKT or other research partnership approaches in trainee-led health research c. Identifying if/how outcomes were reported and evaluated in trainee-led health research using IKT or other research partnership approaches.

Stage 2: Identifying relevant studies Participant
This review considered literature for which health research trainees (i.e., graduate students [master's, doctoral] and postdoctoral fellows) were the primary author/researcher of the paper. Postdoctoral fellows were described as postdoctoral researchers or postdoctoral research associates. Students/fellows in the position of a trainee meant that the included studies were related to the student's or fellow's thesis/programme project/dissertation/fellowship projects.

Concept
This review considered studies that explored IKT or other research partnership approaches in trainee-led health research. We included studies that described the trainee's experience with IKT research or related partnership approaches (i.e., manuscript or thesis/dissertation chapter that provides a reflection or description of the approach, text or opinion paper describing how the partnership approach was used). Included papers explained how the research partnerships approach was used, including principles, strategies and/or tools. Studies that stated the use of a research partnership approach but did not describe how it was used were excluded. Studies that described barriers and facilitators to using an IKT or other research partnership approach were included. Studies that evaluated the IKT approach were also included. Papers that did not describe a research study were excluded as wrong design. For this review, we used the following operational terms and definitions (Table 1).

Context
We considered literature focused on trainee-led health research. For the purpose of this review, health research referred to research that aimed to "increase our knowledge of health, disease, and health services, and to then apply that knowledge to help people lead healthier lives" [25]. It also included "biomedical research, epidemiological studies, and health services research, as well as studies of behavioral, social, and economic factors that affect health" [26].

Search strategy
In collaboration with a health science librarian, we developed a search strategy to locate both published and unpublished primary studies, reviews, and text and opinion papers. Hoekstra and colleagues' [27] comprehensive search strategy for synthesizing the research partnership literature was used to inform our search strategy. We followed the three-step process in accordance with the JBI Scoping Review Methodology. First, we conducted an initial limited search of MEDLINE and CINAHL to identify articles on the topic. Second, from the selected articles, we derived relevant text words and index terms to develop a full search strategy. Third, the search strategy, including all identified keywords and index terms, was adapted for all included information sources. We searched the following databases on 24 June 2020: MED-LINE (Ovid), Embase (Elsevier), CINAHL (EBSCO) and PsycINFO (EBSCO). The final search strategy for each database can be found in Additional file 1. We included peer-reviewed studies, editorials and commentaries. We also searched sources of unpublished studies and grey literature (not empirical studies) including ProQuest Dissertation & Theses Global databases (ProQuest) and the first 50 pages of Google Scholar during the period of September-November 2020. Only papers published in English were included. No date limits were applied, to allow for exploration of the use of IKT or other research partnership approaches in trainee-led research over time.
We also used websites of research and academic institutions and health system organizations, together with concept papers, reports and blog posts, that reported non-peer-reviewed literature. Research partnerships "Individuals, groups, or organizations engaged in collaborative research activity involving at least one researcher (e.g., individual affiliated with an academic institution) and any stakeholder actively engaged in any part of the research process (e.g., decision or policy-maker, health care administrator or leader, community agency, charities, network, patients, lived experience advisor, etc.)" [144] Approaches The IKT and research partnership activities that comprise or promote collaboration in the research process Barrier "A circumstance or obstacle that keeps people or things apart or prevents communication or progress" [145] Facilitator "Someone or something that facilitates (to make easier)" [146] for Leadership in Applied Health Research and Care (CLAHRCs), and KT Canada, and we also contacted individuals or groups for additional material (i.e., Twitter, IKTRN members, relevant KT listservs). We used our professional networks in the area of IKT and collaborative health research to email relevant content experts to identify additional sources that met the inclusion criteria. Reference chaining was conducted with all included papers.

Stage 3: Study selection
All identified citations were collated and uploaded into Covidence [28], and duplicates were automatically removed. A pair of reviewers (CC, HDS, JL, CL, ER, KM, MDV) independently screened and assessed titles and abstracts against the inclusion criteria. Next, fulltext articles were retrieved for potentially relevant studies. After screening titles and abstracts, two independent reviewers (CC, HDS, JL, CL, ER, KM, MDV) assessed the full text of relevant studies in detail against the inclusion criteria. Any discrepancies between the reviewers at each stage of the study selection process were resolved through discussion or by a third reviewer. We sought out the trainee status of each primary author during the fulltext assessment stage. When papers did not mention a primary author's status, we searched the author's name and the indicated affiliation in Google. If the primary author was found to be a trainee, we included the paper. We also made use of available LinkedIn profiles of the authors to identify their trainee status.

Stage 4: Charting data
We developed a data extraction tool to capture information on the general characteristics of the included paper, trainees' characteristics, IKT approaches, the trainee-reported barriers and facilitators to knowledge user engagement, and outcomes. Two reviewers (CC, HDS) first pilot-tested the extraction tool on three studies to identify any discrepancies and ensure consistency of data extraction. Then reviewers (CC, HDS, JL, TN, ER, AB, AC, RD, LD, MDV) were paired and independently extracted data using Covidence. Conflicts between reviewers regarding data extracted were resolved through discussion or by a third reviewer. We used several frameworks to analyse the extracted data. First, we mapped included papers onto the seven phases of the knowledge-to-action (KTA) cycle [29] based on the reported research purpose and objectives. Second, we categorized the knowledge users' engagement into the five levels of public participation of the International Association for Public Participation (IAP2) spectrum to help conceptualize the data and offer a structure for reporting the results [30]. Third, we identified the research stages [31] that knowledge users were engaged in, including (1) development of research question, (2) development of research proposal, (3) pre-study launch administration, (4) recruitment and data collection, (5) data analysis, and (6) dissemination and implementation. Fourth, we used the Workgroup for Intervention Development and Evaluation Research (WIDER) [32] reporting checklist to describe details about the IKT approach, including (1) content (nature and goal of the study and/or IKT partnership); (2) mode of delivery (specific types of IKT activities in which knowledge users were involved); (3) duration and/or frequency (timing of IKT activities); (4) participants (who was involved in specific IKT activities); and (5) personnel (who coordinated or led IKT activities) (review objective A). Fifth, we used the Capability, Opportunity, and Motivation-Behaviour (COM-B) model [33] and Theoretical Domains Framework (TDF) [34] to describe trainee-reported barriers and facilitators to knowledge user engagement (review objective B). We first coded data using the COM-B and TDF, and then generated themes inductively. Next, we used McLeroy's social-ecological model [35] to map where the reported barriers and facilitators existed within the trainee's research ecosystem (i.e., individual, interpersonal and/or organizational levels). Lastly, we developed three categories informed by the IKT outcomes reported by Gagliardi [4,36] to organize reported outcomes and impact of the IKT approach in the included studies (review objective C), if the studies reported outcomes. We coded reported outcomes into the following three categories: (1) immediate outcome of partnership formation (e.g., mutual understanding of language, work style, needs and constraints, and appreciation for the collaborative process are established [4]); (2) intermediate outcome of partnership that occurred during the preparation and research process (e.g., identification of research questions, and conduct of research [36]); and (3) long-term outcome of partnership at the completion of the research process and post-study (e.g., scale-up/spread of research, and use of research in practice and policy [4,36].) One reviewer (HDS) initially coded data using the coding schemes for the first five papers. This coding was verified by a second reviewer (CC) to identify any discrepancies and ensure consistency in coding. After verifying the coding strategy, one reviewer (HDS) coded the remaining data, and the second reviewer (CC) verified the coded data. Findings were presented and reviewed with all team members to interpret key findings.

Stage 5: Collating, summarizing and reporting results
We charted the data in a tabular form to align with the review objectives. In addition to the tables, we created a figure of the barriers and facilitators. We also produced descriptive numerical summaries of the quantitative data (i.e., frequency counts). Lastly, we provided a narrative summary to accompany these presentations and described how the findings addressed the review's question and objectives.

Results
Our database searches resulted in 3237 citations. We identified an additional 23 relevant papers through other information sources (e.g., IKTRN, Google Search), for a total of 3260 citations. After duplicate removal, 2895 citations remained for assessment against the inclusion criteria. After screening titles and abstracts, 343 citations remained for full-text review, and 74 citations, describing 72 studies, were included. See Fig. 1 for the PRISMA flow chart.

Characteristics of included studies and trainees
Of the 74 papers, 34 were peer-reviewed studies, 28 were theses, 9 were web-based sources and 3 were other published grey literature. The 72 studies were a mix of participatory action research (n = 41), mixed or multi-methods (n = 12), codesign (n = 1), scoping or systematic review (n = 2) or other qualitative designs (n = 12). Web-based sources included casebooks from the IKTRN and text interviews of trainees who used an IKT approach in their research. Most grey literature sources were theses (i.e., primary research) and non-peer-reviewed journal articles.

Knowledge user engagement
Knowledge users were engaged in multiple stages of research: (1)  Four studies did not report the level of knowledge user engagement. See Table 3 for more detail.

IKT strategies
Overall, there was a lack of reporting on the use of IKT strategies. As detailed in Table 3, meetings were the most common mode of engagement with knowledge users (n = 27). Written communication, such as handouts or letters, was also used (n = 11), as was email (n = 7) and phone communication (n = 6). Table 4 summarizes the key barrier themes identified from the included papers. Trainees' lack of knowledge and skills posed challenges for IKT research [48,71,72,74,79,84,97,109]. IKT knowledge and skills were related to knowing how to apply an IKT approach and specific procedural techniques related to IKT research processes. Also, trainees reported that their "outsider" status from the health system made the knowledge user engagement challenging [44,51,52,67]. In terms of the IKT process, trainees reported that competing priorities among trainees (e.g., coursework, thesis/project completion), knowledge users (e.g., patient care, service delivery, resource use) and university institutions (e.g., timely thesis completion, resource use) posed challenges to the IKT approach [62,65,70,72,78,79,94]. Furthermore, knowledge users can have full schedules, making it difficult to find a common time to meet as a team [47,89]. It was also difficult for trainees to define the scope of the research project that fulfilled knowledge user needs, aligned with health system priorities, and adhered to institutional guidelines (i.e., thesis requirements) [44,63,67,74,83]. The time-limited nature of graduate programmes also posed challenges to trainees [64,81,99]. Trainees also experienced power dynamics within the academic and health system institutions, which further contributed to the difficulties of knowledge user engagement [44,51,56,72,93]. Lastly, it was challenging to meet knowledge users in person when they lived too far and had to travel a long distance [39,47,52]. Table 5 summarizes the key facilitator themes identified from the included papers. Some trainees gained the necessary knowledge and skills to undertake IKT research in specialized graduate courses focused on applied health research [69,82]. Similarly, trainees viewed themselves as facilitators [39,55,65,110] and reported that facilitation skills helped with knowledge user engagement [38,53,65,80,91,102]. For example, trainees aimed to use clear and common language to facilitate knowledge user engagement activities [80]. Additionally, trainees' flexibility and problem-solving skills were helpful when engagement challenges required quick adaptations [45,53,91]. Trainees also reported that being an "insider" was a facilitating factor for doing IKT research [63,76,88,110]. A pre-existing relationship with knowledge users or previous experience in the related field helped with building partnerships [70,76]. Trainees also reported that trusting relationships and a safe place for partnership development were facilitating factors for knowledge user engagement [38,50,65,66,72,80,83]. This includes promoting a team culture that respects diverse perspectives. Trainees described using agendas and written agreements, such as a memorandum of understanding, to help set common goals early in the IKT process [65,83,98]. Supervisors played an important role in facilitating trainees' IKT involvement [69,70,80,100]. Trainees reported that having supervisors or committee members with expertise in IKT was helpful to guide them through the research partnership. Figure 2 represents a modified illustration of McLeroy's social-ecological model [35] to help visualize where the reported barriers and facilitators exist within the trainee's research ecosystem (i.e., individual, interpersonal and organizational levels). As shown in Tables 4 and 5, barriers and facilitators have been categorized to corresponding level(s) to guide interpretation of the data. First, the individual level represents trainees themselves. Second, the interpersonal level represents people that trainees interact with, such as supervisors and knowledge users. Lastly, the organizational level represents institutions, including universities and health systems, as well as communities. Identified barriers and facilitators are not static or absolute; as represented with connected lines, some of these barriers and facilitators exist across the three                         levels. Further, the three levels demonstrate the complex nature of individual trainees within a broad micro-, meso-, macro-system. Some of the barriers are the direct opposite of facilitators; for example, lack of funding and financial resources are reciprocal barriers and facilitators.

IKT impact and outcomes
As shown in Table 6, reported IKT impact and outcomes were related to immediate (n = 38), intermediate (n = 40) and/or long-term (n = 15) outcomes of partnership formation. Twenty studies did not report outcomes related to IKT or other research partnerships. See Table 6 for more detail.

Discussion
This scoping review describes how IKT or other research partnership approaches have been applied in thesis and/ or postdoctoral health research. We identified 74 papers from published and grey literature sources to be included in the review. Overall, study findings provide insights for trainees interested in IKT or other research partnership approaches, and offer guidance on how to apply an IKT approach to their research. Further, the review highlights the important role that academic supervisors, knowledge users, and academic and health system institutions play in providing support and infrastructure to facilitate IKT or other research partnerships in trainee-led health research.

IKT/research partnership principles, strategies and/or tools
The majority of studies involved collaboration with knowledge users in the research question development, recruitment and data collection stages of the research process. We used the IAP2 Spectrum [30] as a way to categorize the levels of knowledge user engagement. It is important to note that, as reviewers, we had to make inferences on the level of engagement due to a lack of reporting on IKT and research partnership principles and strategies. We categorized the majority of studies (n = 43) in the collaborate stage, which indicates a high level of Organizational partnership in the decision-making process. This is not surprising, as IKT and research partnerships are rooted in deliberate partnership and shared decision-making throughout the research partnership from start to finish [112]. There are distinct differences between research partnerships that empower partners to be active participants in the shared decision-making process and lesser levels of engagement, such as communication, information-sharing and consultation [113]. The included papers described the former, where trainees made deliberate efforts to support research collaboration with knowledge users and described this partnership in detail. While the IAP2 Spectrum was a useful framework for categorizing engagement, its origins are in broad public participation [111], and it does provide a comprehensive representation of the types of knowledge users involved in health research (i.e., patient, caregivers, healthcare providers, health system leadership and policy). Efforts are needed to develop a clearer set of IKT principles and strategies for knowledge user engagement in the health context.

Barriers and facilitators to using IKT or other research partnership approaches in trainee-led health research
This review illustrates the intersecting barriers and facilitators for IKT and research partnership approaches at the individual, interpersonal and organizational levels (Fig. 2). Many of the reported facilitators were reciprocal to the barriers and have the potential to support IKT or research partnerships, including partnership skill development, co-creation of common goals and leveraging research programme partnerships. We used the COM-B model of behaviour [33] to categorize reported barriers and facilitators related to trainees' capability, opportunity and motivation for using an IKT or research partnership approach.

Capability
Our review identified several capability-related barriers and facilitators at the individual trainee level for engaging in IKT research, including IKT knowledge and skills, and flexibility in the research approach. These findings align with recent research highlighting the lack of researcher preparation in research partnership approaches [11,114]. Building on this known gap in the literature, our review offers a clearer understanding of what types of knowledge and skills are needed to support researcher preparation for collaborative health research, including the development of facilitation skills. Facilitation is defined as "the process of enabling (making easier) the implementation of evidence into practice" [115]. It supports a purposeful two-way process of change that focuses on building trusting relationships and establishing common goals between the facilitator and those engaged in making the change [116,117]. More formal training efforts are needed to support facilitation skill development for health research trainees. Dogherty and colleagues developed a taxonomy of 53 facilitation strategies, ranging from providing assistance with certain tasks, to a more holistic process that empowers change in individuals' attitudes and ways of thinking and working. This taxonomy may be a helpful starting point to develop trainees' skills as facilitators in the research process, which to date has not been a primary focus in traditional academic programmes. Previous research has identified core competencies that trainees need for applied health research and knowledge translation [118,119]. A baseline self-assessment on these competencies may be helpful for trainees interested in developing their IKT or other research partnership skills [120]. Further, several training programmes exist to provide trainees with experiential learning opportunities in IKT or other research partnerships. For example, the Canadian Institutes of Health Research (CIHR) Health System Impact (HSI) Fellowship was designed to modernize doctoral and postdoctoral training to better equip researchers with the professional and research skills needed to address complex health system challenges [121]. HSI doctoral and postdoctoral research fellows are embedded in health system organizations to develop Canadian Health Services and Policy Research Alliance's 9 enriched core competencies, understand the intricacies of health system delivery, and partner with members of the health system to support relevant research efforts [121]. Similarly, in the United States, AcademyHealth developed the Delivery System Science Fellowship to provide experiential learning and professional development opportunities for postdoctoral trainees [122]. These types of training programmes offer formalized experiences focused on development of professional skills not currently emphasized in health services graduate training. Trainees may wish to consider these types of formalized training programmes to further their IKT skills development and better support the use of a research partnership approach in their graduate or postdoctoral research.

Opportunity and motivation
Successful research partnerships require more than individual trainees having the right skills and willingness to engage with knowledge users. The majority of barriers and facilitators identified in this review were categorized in the interpersonal and organizational levels, specific to intersecting opportunity and motivation determinants that either support or hinder research partnerships for trainees.
Trainees reported that their status as an insider or outsider to the organization/community was a key facilitator or barrier for building partnerships. This is consistent with Dwyer and Buckle's [123] previous work; having an insider membership typically gives an advantage to researchers, as it creates a sense of trust and openness from the participants. However, researchers are never completely insiders or outsiders, and their status does not make for a better or worse researcher [123][124][125]. Dwyer and Buckle [123] challenge the dichotomous insider and outsider identity of researchers and propose that "space between" is critical. In this "space between, " [123], researcher's positionality is complex and fluid in nature [126] and can impact the research topic, epistemology, ontology and methodology [127]. Therefore, when trainees engage in IKT, it is more important that they practice self-reflexivity to understand their positionality in the study rather than focus on whether they are insiders or outsiders. Moreover, this is consistent with our findings on trainees reporting reflexivity as a facilitator for engaging in IKT research. We encourage trainees to reflect on the commonalities and differences they share with their research partners and how their positionality in the "space between" might affect research relationships. From there, trainees may have a better sense of how to Changes in clinic processes and increased clinicians self-efficacy navigate relationship-building with different knowledge user groups. Academic supervisors play an important role in supporting trainees to navigate the "space between" and develop effective research partnerships with knowledge users. Our findings reinforce previous research on leveraging pre-existing research relationships to support IKT [18,128] and highlight the value of research programme partnerships to facilitate trainee IKT research. Research programme partnerships and longer-term relationships, as opposed to one-off project partnerships, offer time and space for clear articulation of expectations, which allows for effective collaboration and mutual gain [129]. If research programme partnerships exist, it is easier for trainees to align their project within their supervisors' larger research programme, which includes clearly developed practice linkages and existing research partnership supports [128]. Further, research programme partnerships ensure continuity of relationships between academia and the health system; when a trainee finishes their training and moves on to their next career stage, the supervisor can maintain the research partnership with the knowledge user. There may also be instances where trainees bring their own knowledge user partnerships into their academic training. For example, several papers described an existing role that trainees had within a community or organization, in which they used their partnerships to build a research study [76,88]. In this scenario, supervisors should respect this pre-existing relationship and offer support on the academic research process to further develop the existing relationship into an effective research partnership.
Our findings also highlight how interpersonal influences can help and/or hinder the research partnership. The interpersonal and group dynamic barriers and facilitators to research partnerships illustrate the inherent relationship focus of IKT. Previous reviews of IKT-related literature have highlighted similar relationship-focused barriers and enablers to partnership research [4,5,17,130]. Power dynamics are often at the core of research partnerships [131]. Several included studies highlight the influence of power hierarchies on the trainee experience with IKT, including pressure from both the academic and health institution to meet certain deliverables and timelines. This is particularly relevant for trainees who are already challenged with power hierarchies within the academy and may be entering into additional power differentials with knowledge users. Establishing mutually respectful relationships requires trust-building and being trustworthy to promote power-sharing and co-create knowledge [130]. Humility is an emerging area of interest in the field of IKT for developing meaningful and trustworthy research partnerships [132]. Humility stresses the need for self-assessment of one's intellectual strengths and limitations, but also an appreciation of the contributions of others [133,134]. Key strategies for practicing humility include clarity of thinking, open-mindedness, a commitment to mutual learning and ensuring that researchers do not bring a view of themselves as superior to the research team [132]. It is important for trainees to practice humility during research partnerships by acknowledging their positionality and taking the time to learn about the context and culture of an organization or community [132,135]. These efforts will support collaboration, mutual learning and responsiveness among researchers, communities and health organizations.
Several organizational barriers and facilitators influence trainees' opportunity and motivation for engaging in IKT or other research partnerships. Included papers reported a number of competing priorities among trainees (e.g., thesis/project completion), knowledge users (e.g., patient care, service delivery, resource use) and university institutions (e.g., timely thesis completion, resource use). Similarly, previous IKT research has illustrated how competing priorities can hinder successful research partnerships [4,27,136]. However, this review details specific challenges that trainees encounter with IKT. Individual trainees may have the motivation to engage in research partnerships, but there is often a lack of structures and resources in place to support the IKT process (e.g., funding to support knowledge user engagement). Our findings build on previous IKT recommendations including co-creating value-driven work with a shared purpose, and thereby fostering mutual gain [129,137]. Further, institutional resources (e.g., funding) and policies are needed to support research partnership approaches for trainees, including flexibility in thesis guidelines, memoranda of understanding between academic and health system institutions, and tangible resources (i.e.,guidelines, worksheets) to support co-creation of common goals.

IKT or other research partnership outcomes
IKT and other research partnership approaches suggest that active engagement of knowledge users throughout the research process increases the use of research in practice and policy [2,4]. Despite the positive intent of IKT and research partnerships, there has been limited empirical evidence of the effects of IKT on evidence use to inform policy and practice [4]. Building on Gagliardi and colleagues' scoping review on IKT [4], we organized our outcome findings into three categories of partnership formation. Most papers reported on intermediate outcomes of partnership formation, such as enhanced relevance of research (n = 18), and partnership formation outcomes, such as valuing different perspectives (n = 13).
However, only nine studies reported the use of research in practice/policy as an outcome, which is the primary goal of IKT. Further, over a quarter (n = 20/72) did not report outcomes at all. These findings highlight the lack of studies evaluating the impact of the IKT beyond the partnership formation and relevance of research. Efforts are needed to advance the science by moving the field beyond partnership formation studies and understanding the impact of IKT on conceptual, symbolic and instrumental use.
Our findings align with recent calls to advance IKT evaluation, including Kreindler's [138] work on using a realist evaluation approach to examine how IKT influences the tangible use of evidence in decision-making related to practice and policy. Realist evaluation investigates "what is it about an intervention that works, for whom, and under what conditions' [139] and is a valuable approach for understand the complexity and nuances of IKT partnerships. Quantitative research methods are also warranted in IKT research, including the application of valid and reliable measurement tools of knowledge user engagement. Several tools exist for patient and public engagement in research that may be useful in examining broader stakeholder engagement (i.e., patients, public, healthcare providers, health system leaders) [140][141][142]. Trainees involved in IKT research are encouraged to apply these qualitative and quantitative methods to evaluate their IKT approach and contribute to the growing body of evidence on IKT impacts and outcomes [143].

Summary of implications
Findings from our review address two of the four main goals of scoping reviews [22]: (1) examine the extent, range and nature of the research activity; and (2) identify research gaps in the existing literature. First, this study mapped what is known about trainee-led IKT and partnership research in the literature. Our findings highlight that more formal training efforts may be needed to support skill development for research trainees involved in partnership research. Further, our findings build on previous IKT work highlighting the need for organizational resources and policies to support research partnership approaches.
Second, this review identified research gaps in the existing IKT literature. To advance our understanding of IKT principles and strategies and how trainees partner with knowledge users, we recommend clearly articulating specific activities used in each level of engagement. These details would help trainees to plan knowledge user engagement during their research projects, including the feasibility and utility of different levels of engagement, and would help to understand differences among types of knowledge users. Further, while the included studies described IKT and research partnership approaches, it was not clear why trainees embark on IKT. Additional empirical research is needed to understand the trainees' motivations for using an IKT approach in their research. This type of research should also aim to understand the unintended consequences of IKT and the power dynamics experienced by trainees and knowledge users, and to identify strategies to promote authentic collaboration, power-sharing and co-creation of knowledge. Lastly, our findings highlight the need for more evaluation of IKT and partnership approaches. We recommend that trainees evaluate their IKT approach and contribute to the growing body of evidence on IKT impacts and outcomes.

Limitations
Several limitations may affect the interpretation and use of our review findings. First, although we built our search strategy from an umbrella review of IKT and research partnership reviews [144], partnership research is a broad phenomenon. We narrowed our focus to IKT approaches or similar research partnerships that focused on working with knowledge users throughout the health research process. However, IKT is predominantly a Canadian term, and we may have missed papers that described research partnerships differently or in a language other than English. This may be why no papers were identified from lower-and middle-income countries. Second, we only included papers that were written by trainees. Although we conducted a followup search of the author's information to identify trainee status, it is possible that we may have missed papers and did not capture all trainees who have used an IKT or other research partnership approaches in their work. Third, many papers lacked detail on the IKT principles and strategies, which made it challenging to analyse the IKT approach using the WIDER [32] reporting checklist. A specific reporting guideline for this type of work is needed to improve reporting and understanding of the IKT approach, including level-of-knowledge user engagement and IKT strategies. Fourth, we categorized the barriers and facilitators into three levels (individual, interpersonal, organizational) to aid in interpreting how the individual trainee is situated within the academichealth system. These categories are not absolute; many barriers and facilitators exist across the three levels and could be categorized differently depending on the context. Lastly, we found that a quarter of included papers did not evaluate their IKT approach outcome of partnership formation. However, it is important to note that our focus was on the IKT trainee experience and not specifically on evaluation of IKT. These types of evaluation studies may have been published separately and at a later date, as the impact of IKT projects can take years to be fully realized and evaluated.

Conclusions
This scoping review of 74 papers describes how health research trainees have used an IKT or other research partnership approach in their research. Most trainees have collaborated with knowledge users in the research question development, recruitment and data collection research stages. We identified key barriers and facilitators related to individual, interpersonal and organizational factors. Findings illustrate the importance of having specific knowledge and skills to engage in research partnerships, interpersonal influences of supervisors and knowledge users, and institutional support for a different way of working in graduate and postgraduate work. We also identified enhanced relevance of research findings as an important outcome of IKT trainee research. These review findings can serve as a basis for future reviews and primary research focused on IKT principles, strategies and evaluation. Further, we expect these findings can be used to inform IKT training efforts including trainee guideline development and academic-health system research supports.