Background
While there are some recent systematic reviews regarding strategies to change health care practitioner behaviour [
1‐
3], there are currently no definitive answers of how best to move toward 'evidence-informed' public health decision making. It is believed however, that the incorporation of the best available evidence into health policy and practice decisions would result in optimal patient and population health outcomes [
4]. Currently, the evidence demonstrates that traditional one-way passive strategies used alone are relatively ineffective [
5,
6]. Strategies that are more interactive and involve face-to-face contact show promising results [
5,
7‐
11], and involvement of decision makers in the research process is associated with a higher degree of research uptake [
12,
13]. One hypothesis emerging from the literature is that a combination of strategies, such as an interactive KTE approach that reinforces relationships between researchers and users, and reaches potential users on multiple levels interacting face-to-face, may be most effective in achieving evidence-informed decision making [
14,
15].
A KB is a popular emerging KTE strategy to promote interaction between researchers and end users, as well as to develop capacity for evidence-informed decision making (EIDM). Although the health care literature is sparse with evaluations of KB impact [
16], there is considerable evidence in other fields, particularly the business and agricultural sectors [
17‐
23].
A KB provides a link between research producers and end users by developing a mutual understanding of goals and cultures, collaborates with end users to identify issues and problems for which solutions are required [
24], and facilitates the identification, access, assessment, interpretation, and translation of research evidence into local policy and practice [
16,
17,
25‐
27]. KBs also facilitate knowledge exchange, build rapport with target audiences, forge new connections across domains [
28‐
31], and assess end users, whether they be individuals or organizations, to identify their strengths, knowledge, and capacity for evidence-informed decision making [
32], in order to better tailor KB interventions to their specific needs. Knowledge brokering can be carried out by individuals [
16,
20,
27,
33], groups and/or organizations [
4,
23,
29], and entire countries [
34]. In each case, the KB is linked with a group of end users and focuses on promoting the integration of the best available evidence into policy and practice-related decisions. A key attribute of the KBs is their skill in the interpretation and application of research.
The KB also synthesizes local community and patient data with general and specific research knowledge to assist users in translating the evidence into locally relevant recommendations for policy and practice. An important component related to the success of this activity is the KB's ability to tailor the key messages from research evidence to the local/regional perspective, while also ensuring the 'language' used is meaningful for different end users [
4,
8,
29,
35,
36]. Another key component is the KB's ability to develop a trusting and positive relationship with end users and to assist them to incorporate research evidence in their policy and practice decisions [
17,
37‐
39], while at the same time promoting exchange of knowledge such that researchers and users become more appreciative of the context of each other's work.
In order to incorporate appropriate forms of knowledge at the appropriate times, KBs need to be attuned to their audience as well as their audience's environment. KBs then work to facilitate organizational change [
24,
31], eliminate environmental barriers to evidence-informed decision making (EIDM) [
40], and promote an organizational culture that values the use of the best available evidence in policy and practice [
17,
25,
41]. Political and infrastructure support for EIDM are seen as important precursors for the incorporation of research evidence into decision making [
21,
25], and hence the KB must focus on ensuring adequate support for EIDM to be achieved. Finally, creating networks of people with common interests is a key KB activity [
17,
20,
32,
41,
42], and has been shown to be an integral [
43,
44] and effective [
45] component of knowledge brokering.
The KB role is a unique and challenging one, and few people currently possess the skills necessary to be effective in this position. It is also unknown to what extent these skills and attributes can be taught. However, to be successful KBs require superior interpersonal skills [
26,
46,
47] communication skills [
16,
31,
32,
41,
47], and motivational skills [
32], and should possess expertise from both end users' and researchers' domains [
12,
17,
41,
47,
48]. Furthermore, a KB requires expertise in gathering evidence, critically appraising evidence, synthesizing information, and interpreting the information in terms of the bigger picture. In terms of personality attributes, a KB should be someone who is a skilled mediator and team builder while being flexible and diplomatic with excellent business and communication skills [
16].
Anecdotal evidence suggests that knowledge brokering can be effective in improving the quality and use of evidence in healthcare decision making [
25,
41]. While the number of published papers discussing knowledge brokering has grown dramatically; few have studied the impact of KBs on EIDM using scientific approaches [
26]. The purpose of this paper is to describe in detail the KB intervention that comprised one of three KTE interventions evaluated in a randomized controlled trial (RCT) and to reflect on the future development of the role in public health as well as other health care settings. While the overall finding from the RCT demonstrated that tailored messaging was more effective, under certain circumstances, compared to knowledge brokering or access to an online registry of synthesized evidence, there was evidence that knowledge brokering had a significant positive effective for public health departments that perceived their organization did not value the use of research evidence in decision making. The results of the RCT have been submitted for publication elsewhere (Dobbins M, Robeson P, Ciliska D, Hanna S, Manske S, Cameron R, Mercer S, O'Mara L, DeCorby K., A randomized controlled trial evaluating the impact of knowledge translation and exchange strategies, submitted).
Methods
A stratified RCT was conducted among Canadian public health departments. Public health departments in Canada are responsible for promoting the health of the population, preventing disease, and providing medical care to treat communicable diseases. They provide services that focus on promoting prenatal, newborn, and parent health, as well as health promotion within schools and worksites, nutritional counselling, physical activity promotion, injury prevention, development of community strengths to promote and improve health, and the promotion of healthy environments [
49]. All provinces and territories in Canada have recommendations in place requiring public health departments to develop and implement strategies to promote healthy body weight in children. Despite these recommendations there is limited capacity (
i.e., time, skill, access) among public health decision makers and limited resources to utilize the best available research evidence with which to plan and implement effective healthy body weight programs and services.
The KTE interventions, implemented for one year in 2005, focused on promoting the uptake of effective public health strategies for promoting healthy body weight in children. One decision maker from each participating local or regional public health department was randomized to three intervention groups with progressively more active KTE strategies: access to an online registry of effectiveness evidence
http://www.health-evidence.ca; registry access and targeted messages; and registry access, targeted messages, and interaction with a KB. These decision maker participants were directly responsible for making decisions related to program planning or health policy for healthy body weight promotion in children in their public health department. In Ontario, relevant titles included program managers and/or coordinators, and in the rest of Canada program directors.
Following ethics approval and recruitment, organizations were stratified into three strata according to size of population served, and randomly allocated to one of the three groups using a computer generated random numbers table by a statistician external to the study. The primary unit of analysis was public health departments. The KB kept a daily journal in which all interactions were documented and reflections of the impact of these activities were noted. The journal provided the data used for describing the KB role in this paper. The primary investigator and KB reviewed the journal separately and came to consensus on the major themes identified in implementing the role.
Discussion
KBs represent an emerging human resource in the health sector. However, the evidence regarding their effectiveness in promoting EIDM is lacking. While there are many commonalities across activities of those in formalized KB positions, no one job description comprehensively defines the role, and the required qualifications may differ significantly, depending on the target audience. Furthermore, there is some evidence linking KB attributes (
i.e., personality characteristics) to impact, drawing into question the generalizability of interventions and outcomes to other settings or KBs [
41,
65,
66]. Yet, knowledge brokering is considered to be adaptable to different contexts [
31,
47], and KBs have been shown to be instrumental in facilitating and improving communication and knowledge sharing between key stakeholders [
32]. They are also associated with facilitating learning [
17,
67‐
69]; building capacity to locate, appraise, and translate evidence into the local context [
17,
38,
47]; improving the quality of evidence used in decision making [
41]; and increasing interpretation of research findings and implications for action [
40].
Lessons learned
In this section, lessons learned by the KB herself, as well as the research team in implementing the year-long KB intervention, will be highlighted. First is the importance of conducting an in-depth assessment of both the participant and the organization as early in the project as possible. Optimally, this assessment should be conducted face-to-face, although the telephone can be used when resources are limited. Early one-to-one contact was instrumental in facilitating the development of the KB/participant relationship, and in essence, set the stage for all activities to follow. For example, the one-third of participants in the RCT who had very early contact with the KB appeared to become more engaged in the EIDM process, and utilized the KB services to a greater extent than those who did not 'meet' the KB until later in the study. A further 30% either did not engage with the KB at all, or to a very limited extent. There did not appear to be any differences between those who engaged early with the KB and those who didn't on their level of capacity for EIDM. Not every participant responded to KB communication right away, meaning some did not meet the KB until two to three months following initiation of the intervention. The in-depth assessments also allowed for tailoring of the KB services over the full duration of the study by identifying at baseline the knowledge, skill, resource, support, and organizational change needs among the public health decision makers.
A second key lesson was the importance of putting in place a mechanism (
e.g., network) to promote interaction and knowledge sharing among participants and with the KB. The KB recognized that public health decision makers across Canada were struggling with similar issues related to healthy body weight promotion in children, requiring similar knowledge and research evidence. Upon reflection, the KB believed that a facilitated network supported by electronic means such as teleconferencing, webinars, or groupware enhancements (
e.g., discussion forum, shared workspaces) would optimize limited time and resources to more efficiently address participants' needs. Through a facilitated network, literature searches could more easily be shared with multiple participants; critical appraisal of the evidence could be done collaboratively online; and interpretation and implications of the research evidence could be discussed. A networking forum provided participants with the opportunity to share their experiences in using the evidence, the activities in which they were engaged, and their impact on local program planning and on changing organizational culture. Similar ideas are reported in the literature [
70], particularly from a systematic review [
46] that reports that social networks and formal networking approaches enhance EIDM efforts.
A third key lesson relates to time. It became apparent during the RCT that knowledge brokering is even more complex than we expected (e.g., it takes longer to develop collaborative, trusting relationships; much more capacity development was necessary than anticipated), and that the process of developing capacity for EIDM among public health decision makers and health departments takes considerable amounts of time. While the time it took any given participant and health department to move from one step of the EIDM process to the next varied, what became evident was each step took longer than we anticipated (e.g., we estimated capacity development would require two to three months of the intervention rather than six months). In hindsight, it is more likely that a multi-year KB intervention is needed to adequately impact on organizations' capacity for EIDM and would require a longer-term commitment of financial and human resources.
The final key lesson relates to the KB interaction and style. It is believed that a greater degree of face-to-face interaction between the KB and the participants would have been useful for developing the relationship, tailoring interventions, and promoting EIDM capacity. Effective strategies are required to facilitate partnership development and encourage individuals to work collaboratively with KBs. In addition, it is believed that several participants from each health department should have been involved in the KB intervention, thereby creating a critical mass in the organization with the skills and capacity for EIDM. Lastly, the KB must be cognisant of many factors that may affect success, such as political and organizational changes, issues of confidentiality, competing interests and priorities, and turf issues within and between organizations.
To where from here?
While several important lessons were learned along the way in regard to the implementation of the KB role, a number of recommendations for future research were also identified. Most importantly, studies are needed to evaluate the effectiveness of KBs in different settings and among different health care professionals. In addition, research is needed to explore the optimal preparation and training of KBs, as well as the identification of the KB characteristics most closely associated with KB effectiveness. Finally, much work is needed to better understand which combination of KB activities are associated with optimal EIDM outcomes, and whether the combination changes in different settings and among different health care decision makers. Other important questions that need to be addressed include:
1.
Is there an optimal dose for knowledge brokering?
2.
What are effective strategies to promote participant engagement?
3.
Is there a critical level of engagement between the organization and the KB that is associated with changing organizational culture?
4.
Would KB facilitation of a network of public health decision makers improve the use of evidence in decision making, capacity development, and organizational change?
5.
How important are KB attributes to the success of KB interventions?
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
MD conceived of the study, participated in the analysis and drafted the manuscript. PR provided the intervention and assisted in draft of the manuscript. DC, SH, RC, LO, KD, SM, and SH consulted on the intervention as it was designed and provided, and participated in review of the manuscript. All authors read and approved the final manuscript.