Perceived Barriers and Facilitators to Using Knowledge Brokers in Canadian Rehabilitation Settings: A Qualitative Study

Background: Knowledge translation experts advocate for employing knowledge brokers (KBs) to promote the uptake of research evidence in health care settings. However, little is known about factors inuencing the utilization of KBs, thereby limiting their employment within healthcare organizations. This research aimed to identify factors likely to hinder or promote the optimal use of KBs within rehabilitation settings in Canada. Methods: Qualitative study using semi-structured telephone interviews with individuals performing KB activities in rehabilitation settings across Canada. The interview topic guide was informed by the Consolidated Framework for Implementation Research (CFIR) and consisted of 20 questions covering ve domains (characteristics of individuals, inner setting, process, outer settings, and innovation characteristics). All interviews lasted 60 to 90 minutes, were digitally recorded, and transcribed verbatim. We conducted qualitative descriptive analysis combining deductive coding guided by the CFIR. Two independent analysts coded and rated all interviews, then met to review, deliberate and modify the codes as appropriate. A matrix was created by listing the salient codes for each CFIR construct to identify factors (facilitators and barriers) at the individual, organisational, and process level most likely to impact the KB’s success/failure. Results: Twenty-three participants, from ve Canadian provinces were interviewed. At the individual level, the majority of participants reported having strong communication skills, being condent about performing KBs activities, and possessing solid clinical experience and prior research skills. At the organizational level, most respondents indicated constantly networking and engaging with clinical teams and different stakeholders, and having an acceptable level of guidance from their managers. Very few participants felt that they received sucient organizational support (i.e., clerical support and IT support). At the process level, all participants indicated needing evaluation tools to better gauge their performance, and the majority mentioned that they would benet from having additional training tailored to their roles as KBs. Conclusions: Individual, organisational and process level factors likely to hinder or promote the optimal use of KBs within Canadian rehabilitation settings include skillsets and networking abilities; culture, resources, and leadership support; and the need for specic training for KBs and for evaluation tools to monitor their performance.

Knowledge brokers need specialized training in brokering activities and to evaluate their own performance.

Background
Clinical practice often fails to be optimally informed by research evidence [1]. For example, despite available clinical guidelines to inform practice in rehabilitation [2][3][4][5][6][7][8], there are persistent gaps between knowledge generation and its use in practice [9][10][11][12][13][14][15][16][17][18]. Such gaps have a negative impact on the health outcomes of individuals and communities [19] and can lead to ine cient use of limited health care resources [1,12]. There is a growing interest in the concept of knowledge translation (KT) as a means to promote the use of research evidence into clinical practices [20].
(Additional le 1 presents each role, with de nitions and examples of related tasks) One of the underlying features of a KB is being an insider. Several studies have indicated that KBs are clinicians who are typically embedded in their organization and are performing the additional role of broker in order to in uence peers [21,[57][58][59]. However, a recent study provided new insights on the challenges and tensions experienced by KBs that can impact the effectiveness of the brokering process [60]. For example, the tensions between different aspects of brokering (i.e. collecting information, sharing information, and adopting information) and those resulting from being positioned between individuals with different perspectives (i.e., between clinicians and researchers). In Canada and elsewhere the interest in the utilization of KBs as a promising strategy is growing, knowledge on how the brokering role is mediated by different facilitators and barriers is limited [40][41][42][43][44][45][46][47]61]. To date, no previous research has identi ed potential barriers associated with using KBs as a means to promote the uptake of research evidence in rehabilitation settings. In the absence of such knowledge, the ability of rehabilitation organizations to utilize KBs within rehabilitation settings remains limited [11,62]. This research aimed to identify the factors likely to promote or hinder the optimal use of KBs within rehabilitation settings.

Research Design
The study consisted of a qualitative descriptive design [63]. Semi-structured telephone interviews were conducted with individuals who perform brokering activities in rehabilitation settings across Canada. The Checklist for qualitative studies: Standards for Reporting Qualitative Research (SRQR) is available in Additional le 2.

Participants and setting
KBs who promote the uptake of research evidence in clinical practice for rehabilitation practitioners, were invited to participate in the study, regardless of whether they worked in clinical, educational or research institutions, across Canada.

Eligibility Criteria
Eligible participants who had already participated in a related study [59] (n = 100) were invited to participate in this current study. They received an invitation email including an information sheet describing the study context and objectives. Adopting a convenient sample approach, we planned to interview all interested participants. Participants were asked to follow a hyperlink attached to the invitation email to complete a consent form and to provide their availability for an interview using Microsoft Webform. A reminder was sent by e-mail every 2 weeks for 6 weeks. The recruitment process ended 2 weeks after the last reminder (i.e. at 8 weeks).

Recruitment strategies
Eligible participants who had already participated in a related study [59] (n = 100) were invited to participate in this current study. They received an invitation email including an information sheet describing the study context and objectives. Adopting a convenient sample approach, we planned to interview all interested participants. Participants were asked to follow a hyperlink attached to the invitation email to complete a consent form and to provide their availability for an interview using Microsoft Webform. A reminder was sent by e-mail every 2 weeks for 6 weeks. The recruitment process ended 2 weeks after the last reminder (i.e. at 8 weeks).

Instrument
The interview topic guide was informed by the Consolidated Framework for Implementation Research (CFIR), which provides a pragmatic structure for determining potential factors related to the implementation process [64]. In our case, the CFIR was useful to guide various questions that cover all domains related to the KBs roles and work environment. The CFIR is composed of ve major domains: characteristics of individuals, inner setting, process, outer settings, and innovation characteristics. The interview guide consisted of 20 questions covering the ve CFIR domains (see Additional le 3).
The interview topic guide was developed jointly by the rst author (DG) and three KT experts (AB, SA, AT) familiar with the CFIR. The interview guide was translated into French and revised for content and face validity by three English-and three French-speaking KBs working in the rehabilitation eld, and revisions were made accordingly. The two interview guides (English and French) were pre-tested with the rst six recruited KBs (three English and three French-speaking KBs) before starting the data collection. Minor changes were made based on the feedback to generate the nal version.

Procedures
Semi-structured individual interviews were conducted between February and May 2019. The rst author conducted all English-language interviews, while another investigator (DZ) led the French-language interviews. A 2-hour training session with both interviewers was deigned to discuss the content of the interview guide and the structure of the interviews. Interviewers had no prior relationships with any of the participants. Informed consent was obtained before each interview. All interviews lasted between 60 to 90 minutes, were digitally recorded using the Zoom meeting platform [65], and transcribed verbatim.

Analysis
Our qualitative descriptive analysis [66] consisted of deductive coding guided by the Consolidated Framework for Implementation Research (CFIR) [64,67] and then inductive coding to identify subthemes within CFIR domains. The process involved three steps:

Data coding
Two team members independently coded and analyzed each interview [68,69] using a Microsoft Excel sheet to facilitate data organization, management, and coding. In order to maximize the rigour of the coding process, the study team had several meetings to discuss and review the coding scheme. The two coders compared their coding on a rst transcript, resolved discrepancies, and reached consensus through discussion. The coding scheme sheet was further tested with four additional interviews. Minor modi cations were made by adding pre-speci ed sub-codes to four questions to facilitate coding. Coders then met periodically to compare and adjudicate coding differences and achieve consensus. Three experts in qualitative research (AB, SA, AT) provided a critique of the analysis and interrogated the coding to ensure a robust and defensible coding of the data. Lastly, the coders met to review, deliberate and modify the codes as appropriate.

Code rating
As per CFIR rating rules, the rating process was used to help elucidate the relative importance of each construct across all interviews [70]. The rating was performed for two dimensions: valence and magnitude. "Valence" refers to the construct's in uence (positive [+], negative [-], no impact [0]), valence was considered to be positive (facilitated KBs roles), negative (hindered KBs roles), or have no impact (not affecting KBs roles). "Magnitude" refers to the extent to which the constructs were discussed. Magnitude was determined based on the level of agreement among participants, which was reached by calculating the proportion of participants who mentioned each code (e.g., few = 0%-25%, some = 26-50%, many = 51%-75%, most = 76-100%). Only codes that were described by two or more participants were tabulated [71] and only codes that were rated with 25% or higher were included in the study results.
Salient codes were those discussed by the majority of respondents (more than 50%) [72,73]. Two raters independently rated the codes, then met and compared their rating until consensus on all ratings was achieved.
Generate matrix and identify key constructs: Finally, a matrix identifying the factors that appeared to positively (facilitators) or negatively (barriers) impact the KBs was created by listing the salient codes for each CFIR construct.

Results
Of the 100 individuals invited to participate in the study, 23 KBs from ve Canadian provinces (Quebec "QC", Ontario "ON", Alberta "AL", British Columbia "BC", Manitoba "MB"), agreed to be interviewed.
Demographic characteristics of the participants are displayed in Table 1. Tables 2 and 3 present the salient facilitators and barriers, as per the CFIR themes. Figure 1 presents the salient barriers and facilitators, reported by more than 75% of participants. Additional les 3 and 4 present the descriptive analysis based on CFIR domains together with illustrative quotes.

I. Characteristics of Individuals
A. Knowledge about KBs Roles: Many participants (70%) reported that their role was mainly to seek, adapt, and share evidence within their local context. More than half (57%) of the respondents indicated that linking different groups of stakeholders was a key role, whereas others (39%) stated that implementing new practices by building individual capacities and addressing barriers for clinical practice change was an important aspect of their role.
"It's helping people access the right evidence at the right time in the right amount to help them address their questions and or to have supporting evidence to move forward" (MB5) B. Self-e cacy: Most participants (83%) felt con dent about their ability to perform their KBs roles, and one quarter of participants believed that they have the skills needed to perform KBs roles, which promoted their self-con dence.
"I feel con dent … I am an occupational therapist …for almost 14 years... I have a really good understanding of the clinical environment, the frontline care.... I've also spent almost 12 years being actively engaged in research activities ...so having my feet in both worlds I think gives me a lot more con dence" (AB7) C. Individual Identi cation with the Organization: About half of the participants' job evolved to include KBs roles over time (52%). In contrast, several other participants (39%) applied for a KB position.
"I was not identi ed... I created the role for myself I think it became self-identi ed ...I was successful in being able to sort of advocate for the importance of having a role like this" (AB7) D. Personal Attributes: 1. Clinical Experience: 74% of interviewees stated that they had clinical experience and of those, 10 (43%) reported that an in-depth understanding of clinical topics helps them better address the needs of their peers while performing KBs roles.
"A broker… somebody who is somewhat connected to the topic (clinical topic) right and understands the real-life context so that's one thing" (AB6) 2. Research skills: Similarly, 74% of participants had formal research training (e.g., Master degree) or had taken part in research activities (65%).
"My training as a master student is a facilitator because I've been exposed to research so looking for info in database is easier for me than it is for a clinician" (QC9 ) 3. Communication Skills: Nearly all participants (91%) stated that good communications and networking skills were essential to perform their job.
"I think that communication skills are probably one of my strengths" (QC12) 4. Interpersonal attributes: Other attributes perceived as helpful to perform their KBs roles included being interested and motivated to implement the latest evidence and able to motivate others (52%), being exible (48%), having emotional intelligence (43%), leadership skills (39%), and being a life-long leaner (26%).
"I had already volunteered, I was always the rst one to put my hand up to be involved in a new initiative or a new project or be the chair of a city, so they had had lots of opportunities to kind of see me in action" (ON1)

II. Inner Setting
A. Networks and Communications: Most participants (91%) had consistent networking and engagement with clinical teams and different stakeholder committees (e.g. clinical teams, professional groups and provincial groups). 61% regularly shared information of potential interest with team members (e.g., upcoming training or funding opportunities). Mostly, networking activities were performed remotely for almost all participants (91%) through email exchanges, phone calls and online meetings (Skype, Zoom, WebEx) especially if a participant was responsible for a large organization. In-person meetings were also very common (78%).
"If there's a workshop coming up or a webinar that people might be interested in, a grant funding, call for research or for program development, then I would email that to everyone in our organization" (MB11) Almost two thirds of participants (65%) reported that they need more communication with their stakeholders, and one quarter of participants suggested using online platforms to improve communication.
"One thing that would improve my ability to do the KB role it certainly is more and better networking. I still nd that communication from kind of provincial groups getting that information to frontline is still a barrier" (AB17) B. Needs of Those Served in the Organization: Almost all participants (91%) were made aware of their stakeholders' needs by questions and concerns raised by their staff. Needs were also identi ed through informal engagement with peers (91%), during regular staff meetings (70%), by questioning the stakeholders (43%), through receiving stakeholders' feedback and complaints (30%), or through conducting needs assessments (30%). 5. Organizational support: One third of interviewees (35%) received administrative support such as graphic design and clerical help, IT support, and digital media, and adequate time to perform KBs activities. Eight interviewees had time to perform KBs activities, and the remaining seven had an information sharing system. However, most participants reported the lack of nancial support (78%) (i.e., nancial support to attend training opportunities), lack of time (78%) (i.e., mot liberating KBs or clinicians to participate in activities), and lack of administrative support (61%) were barriers.
"Limited budget that you have access to. …You know the common phrases, we have no money" (MB4) D. Readiness for Implementation 1. Leadership Engagement: Two-thirds of participants (65%) reported receiving guidance from their managers. Several also mentioned that managers are accessible (52%), supportive (48%), and open to discussion (39%).
"My manager was great… Very supportive" (MB3) Six participants praised their managers for liberating them to attend training opportunities, believing in KBs activities, and allowing for more KBs autonomy. Some participants (30%) complained of the lack of managers' accessibility and availability.
"I wish I had more access to her [my manager] sometimes she's a very busy woman" (AB7) 2. Available Resources: All participants had access to computers, many had o ce space (57%) and access to software programs (Telemedicine Skype, Zoom, SharePoint, Adobe connect, OneNote) (52%), and or conference rooms (35%).
"We have persuaded many teams to use zoom as a way to communicate so zoom has increased our capacity to reach out to certain clinicians even patients and physicians" (QC18) 3. Access to Knowledge & Information: Most participants (78%) reported that networking with colleagues, experts, or other stakeholders (i.e., patients), and social media helped them gain information. Many also access different sources such as organizations newsletters (74%), the library databases (70%), and other online searching (52%).
"We're also involved in various communities of practice which shares information latest research clinical practice guidelines" (BC2) Nonetheless, several expressed the need to access information resources (i.e., databases) (52%).
"It's quite hard to access evidence-based because our library services is not great…. we don't have much access to data back databases" (QC18)

III. Process
A. Planning: Eighteen participants shared that they did not receive any training on their KBs roles before starting their job. Nearly half of the participants (43%) relied on self-learning activities and searched for educational training opportunities that could help them perform the KBs role. One third of participants (35%) mentioned that their organization provided ongoing training opportunities at work, and the remaining participants (30%) said they gained their KBs knowledge and skills over time with work experience. Few participants (26%) received formal KT training (Master degree or a certi cate).
"There was the opportunity to do this knowledge translation certi cate at SickKids [hospital] ....I did the one through Guelph… that course in knowledge translation open my eyes" (ON19) Most participants (87%) expressed the need for additional training to improve their skills in communication, research, managing people and projects, as well as change and con ict management.
"I think I could be more effective if I add more training… training for myself in terms of hopping my skills" B. Engaging: Interviewees reported several factors likely to encourage their peers' involvement in KT activities, including KBs' credibility, building trust, and being seen as a source of information (48%); participants' attitude toward teammates and mutual respect (39%); providing clear explanations and justi cations when implementing new evidence, favouring shared decision-making (35%); being insiders, engaged within teams, and aware of the local context needs (35%); being interested in their peers (35%), and avoiding being seen as "giving orders" (35%).
"Yes I would say... it has to do with my credibility authority and come relational interpreter relational competencies" (QC18) C. Re ecting & Evaluating: Nearly all participants monitored their performance through different strategies, including having ongoing follow-up with their managers (96%), presenting regular reports (83%), receiving feedback from their managers (48%), tracking productivity and meeting stakeholders' needs (35%), and meeting goals and deadlines (30%).
"There are reports, periodic performance reports, I think every six months" (ON8) Nonetheless, all participants expressed the lack of formal evaluation of their knowledge brokering performance (or an evaluation framework) and some participants (57%) agreed that a valid evaluation tool to gage their performance would be useful "I feel like there must be a better way to measure. I'm just not sure what it is…. momentum plan where we said like three months six months or one year type goals and it has anything to do with knowledge brokering" (ON16)

IV. Outer Setting
A. Cosmopolitanism: Half of the participants (52%) were connected to professional support groups (community of practices "CoP") or provincial committees (35%) which kept KBs up-to-date.
"We have our community of practice and things like that that we discuss you know best practice and what is going on and what people are experiencing at their sites and work together as a team" (BC2)

B. Peer Pressure (Peer Support)
Half of the participants (61%) reported a need for a CoP for individuals who perform KBs activities. Nearly half of participants (43%) stated that they sometime contacted other individuals who perform KBs activities, and one third of participants didn't contact any KBs at all.
"It is really important to the people, the KB community of practice… that mentorship, having other knowledge brokers to talk to, like the librarian, and just having some of those structures that are in place and the support" (MB5)

V. Innovation Characteristics
A. Innovation Source: The majority of participants (78%) reported that their organizations believe in the importance of keeping clinicians up-to-date and to support them, to ensure the highest standard of care.

Discussion
This study aimed to identify factors likely to hinder or promote the optimal use of KBs within Canadian rehabilitation settings. Our ndings showed that factors common to the ve different Canadian provinces likely to in uence KBs roles are mainly associated with three levels: individual, organizational, and process level.

Individual level
Each broker in the present study was unique in terms of their personal attributes and the particular skills required for their position and their local context. Prior research has suggested exploring which of these attributes and skills are most likely to support and enhance KBs efforts in knowledge translation [74,75].
Our ndings address this gap by showing that having certain attributes and skillsets (i.e. clinical experiences, understanding of local context demands, communication and research skills, and involvement in research activities) was viewed as favourably impacting the performance of KBs.
Cultivating these features may help to ensure the success of the KT process. These ndings are consistent with those from a realist review [58] and a national survey [59] undertaken by our team, showing that KBs are often clinicians embedded within the organization with over 15 years of clinical experience. Likewise, previous research reported that positive traits of KBs include professional competencies [76][77][78], experiential knowledge [76], and communication skills [79]. This emphasized that KBs success does not only lie in what they do, but also in who they are [80].
Our nding also indicated that personal attributes that are common traits of KBs included motivation and exibility, having emotional intelligence and leadership skills, intellectual curiosity and analytic skills. This is also in line with previous research suggesting that KBs are enthusiastic, agreeable, friendly, exible, positive, persuasive, entrepreneurial, proactive, comfortable working in a dynamic environment, and openminded [81,82]. Interestingly, our ndings showed that many participants were self-motivated to get involved in brokering activities voluntarily; these ndings are also consistent with another study [59], showing that many KBs volunteered to perform this role. For several participating KBs, and consistent with the scoping review by Bonawitz et al [81], the feeling of satisfaction was a su cient motivator and reward for continuing to perform these types of activities, even in the absence of organizational incentives. A recent mixed-methods study also showed that ownership, persuasiveness, and grit may all contribute to the one's ability to drive the KT process [80].
Together, those skills and personal attributes appear to provide KBs with the required self-con dence and credibility among their teams, thereby reducing resistance to behavioural change. This is supported by prior research indicating that KBs tend to be trusted, accountable, respected individuals who have credibility among their teams [31,75,83]. These individuals appear to be in uential among various stakeholder groups [84,85] because of their positive attitude which facilitates the knowledge sharing process and drives behavior change within an organization. These identi ed skills and attributes need to be considered by employers and KT researchers when selecting individuals to play the KB roles within their organizations.
Another preferred feature of KBs was that of insider (i.e. working in the same setting as team members) as it appeared to facilitate networking and engagement in brokering activities and increase the KBs awareness of the local context needs and the desired change. Our ndings are consistent with research indicating that brokering activities are highly responsive to the context in which they occur [75], and that KBs should work within the given clinical setting [21,57] as this raises awareness of their peers' needs, schedules, clinical roles, caseloads, current practices, and past experiences [84][85][86]. Findings from this study showed that working in a different building or city was perceived as a barrier to constant involvement. Bonawitz et al. [80] have indicated that the physical presence at the point of change may contribute to an individual's ability to drive the desired change. Applying multiple facilitation methods (i.e., interactive discussion combined with online resources and multiple technology methods) [56,85,87,88] was reasonable compensation for KBs working remotely to promote the success of the KBs roles. In general, our ndings at the individual level can be used to select more effective KBs to enhance the KT process.

Organizational level
Recognizing that knowledge brokering involves interactions between various types of stakeholders, it is important to consider the social determinants of brokering activities [74]. Our ndings showed that networking and engagement with different stakeholders was seen as an essential element of the brokering activities since constant networking helped KBs to be aware of stakeholders' needs. Also, ongoing conversations among KBs and their peers seem to have a number of advantages at the inner setting level when communicating with clinical teams and managers (e.g., decreasing the resistance to change and providing the informal evaluation for KBs activities) and at the outer settings level when communicating with professional and provincial groups (e.g., facilitating the accessibility to information). The existing literature [89][90][91][92][93] and a recent mixed-method study [94] showed that by building relationships with knowledge users Canadian KBs' help their peers to gain access to research evidence that can inform or improve their practices. Previous research also emphasized the importance of interpersonal communication as a substantial element in knowledge brokering; communication acts as a foundation to build relationships of trust between KBs and their working teams [10,54,76,90,95].
Emphasising strong communication skills and networking abilities needs to be highlighted as essential elements in KBs' job requirement in the future.
Organizational support (access to resources, administrative, technical, and nancial support) can positively in uence the initiation and sustainability of the KBs role within an organization [31,74,81,[84][85][86]96]. Indeed, our ndings suggest that several forms of organizational support can impact the success of KBs roles. For instance, providing access to a library and databases and subscribing to relevant newsletters, providing IT support, clerical support, virtual communication tools, physical space "o ces and conference rooms"), and allowing adequate time for KBs activities (i.e., liberating individuals to perform KBs activities and liberating clinicians to participate in KBs activities) were deemed important by participants.
Our ndings also highlighted a lack of nancial support for KBs activities (i.e., budget for developing KBs resources and attending conferences). Previous research has reported that dedicating nancial support for brokering activities clearly facilitates these activities [84-86, 96, 97]. One solution might be to work in collaborations with researchers and graduate students, [59] taking advantage of funded research investigating brokering activities [97]. Prior research also emphasized that organizations should value and prioritise brokering activities as well as enhance the awareness of the KBs role to positively impact their function [97]. This was re ected in the views of our participants who raised issues related to the organizations' limited awareness of KBs, unclear or poorly de ned KBs roles, a lack of prioritizing of brokering activities, absence of initial or ongoing training for KBs, and the need for a reward system for brokering activities.
Our ndings highlight that the impact of knowledge brokering activities is a shared responsibility between KBs and their organizations, as devoting time, allowing facilities for communication, and dedicating nancial support all seem to positively impact the KBs' roles.

Brokering process
Our ndings highlighted the lack of training for KBs, which has also been reported elsewhere [58,59]. Despite this, KBs seem to be keenly aware of expected roles and targeted goals (e.g., supporting the implementation of research evidence, keeping clinicians up-to-date, and networking with different stakeholders), suggesting that KBs activities are highly responsive to the local context [75]. Previous research also indicated that cumulated experience for KBs may balance the lack of KBs-related training [59]. One important nding, not previously reported in the literature is the lack of awareness of existing KBs-related training opportunities; this may explain why most KBs depended on self-directed learning as well as on-job learning, despite their need for formal KB training [59]. There is a need to increase the KBs' awareness of the existing training opportunities, and increase the accessibility of those training through integrating virtual learning approaches to access a greater number of KBs. This research exposed substantial needs for strengthening the KB role and its impact on practice change and research. First, the need for standard evaluation tools to evaluate KBs performance; this was consistent with ndings from Newman et al. [94], that have emphasized the lack of evaluation for KBs practices, and if happened, it was informal evaluation. Although the literature on program evaluation has grown substantially in the past decade [98][99][100], that has not been adapted and adopted for evaluating KBs performance and practices. Dobbins et al. [101] recently suggested that KT researchers need to develop concrete and actionable indicators and tools to measure KBs practices. To this end, the outcomes-focused knowledge translation framework [102] may be adapted to evaluate KBs performance. This framework was proposed as a means to conceptualize how knowledge seekers can access and utilize information while receiving real-time feedback data about the outcomes [102]. The real-time feedback component ts well with the nature of the brokering process and constant networking of KBs with stakeholders. We encourage KBs to integrate evaluative frameworks into their practices in order to assess their impact by appropriate outcome measures. Determining objectives, activities, and outcomes speci c to knowledge brokering might help in evaluating the effectiveness of knowledge brokering roles [94]. Second, participants suggested creating a provincial or national community of practice (CoP) for KBs to promote networking and information exchange among KBs and avoid work duplications. A CoP provides a vehicle to connect a group of individuals, with a shared concern, who might not otherwise have the opportunity to interact, share knowledge, identify solutions to common problems [103,104]. Recent technological and social networking advancements facilitates the creation of numerous virtual CoPs, which allow connecting individuals from varying disciplines, contexts, and geographical locations [104].

Future research
As personal attributes (characteristics and skills) seems to be essential to the success of KBs role [74], well-designed studies quantifying the impact of those attributes on KBs performance would be useful [74,81]. Second, interventions to help improve on these attributes and skills should be developed and tested [81]. Third, an environmental scan that identi es and describe the existing educational training opportunities for KBs would be bene cial. Fourth, there is a need for developing an evaluation framework and tool to monitor KBs performance. Lastly, establishing a national COP for KBs working in rehabilitation and evaluating its impact may be a way to help KBs network and stay abreast of the latest development in their eld.

Strengths And Limitations
This research provides new insights into the brokering activities from a range of rehabilitation settings in Canada, and into the facilitators and barriers they encounter when performing brokering activities. The data analysed were consistently blinded during coding and applying ratings to constructs, which increase the trustworthiness of the study's nding. Moreover, quantifying the CFIR domains to determine the magnitude of each theme across participants increases the trustworthy of our interpretation. In addition, our sample size was consistent with previous studies employing the CFIR [105][106][107][108], and participants came from diverse professions, educational levels, and age groups. Nonetheless, our study is not without limitations. First, our participants were from ve Canadian provinces only and nearly half were from Quebec. Further research exploring barriers and facilitators among KBs in other provinces would be needed. Second, this study was restricted to rehabilitation professionals, limiting the generalizability of our ndings to other healthcare sectors. Third, we did not reach data saturation for all the CFIR domains (tension to change, relative priority, learning climate), and few of the CFIR domains (evidence strength and quality, adaptability, trialability, complexity) were not explored as they were deemed to be of low pertinence to the KBs roles.

Conclusion
The novelty of this study centers around capturing potential barriers and facilitators to the optimal use of KBs within rehabilitation settings in Canada. Key individual determinants identi ed by participants included communication skills, clinical experience, and research skills. Organizational determinants included allowing a consistent networking and engagement with relevant stakeholders to promote the awareness of local needs, and enhancing the accessibility to physical (i.e. computers) and informational resources (i.e., latest research evidence). Strategies aiming to overcome barriers such as limited time and nancial support to perform KBs roles should be considered. Key process level determinants were providing KBs training and utilizing evaluative tools for KBs performance. These nding may be useful to the organizations currently employing KBs to help improve their work productivity.

Declarations
Some data generated or analyzed during this study are included in this published article. Additional data (generated and analyzed) are available from the corresponding author on reasonable request.

Ethics approval and consent to participate
Ethical approval was obtained from the McGill University Institutional Review Board (Study Number A05-E25-18B). Participants were informed of the purpose of the research and their rights as participants to voluntary participation, anonymity, and con dentiality. Written consent was obtained from all participants.

Consent for publication
Not applicable Figure 1 delineates salient barriers and facilitators (more than 75%) according to themes back to the CFIR.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.