Organizational Knowledge Translation Strategies for Allied Health Professionals in Traumatology Settings: A Realist Review Protocol

Background: Knowledge translation (KT) is an important means of improving health service quality. Most research on the effectiveness of KT strategies has focused on individual strategies, i.e., those targeting directly the modication of allied health professionals’ knowledge, attitudes, and behaviours, for example. In general, these strategies are moderately effective in changing practices (maximum 10% change). Effecting change in organizational contexts (e.g., change readiness, general and specic organizational capacity, organizational routines) is part of a promising new avenue to service quality improvement through the implementation of evidence-based practices. Methods: A realist review will be conducted to explore the various organizational KT strategy action mechanisms (how, why, for whom, to what extent and under what circumstances) with a view to facilitating the choice of effective strategies for sustainable implementation of evidence-based practice for allied health professionals in traumatology settings. The review will begin by presenting initial theories developed by the research team, followed by the search for evidence, the selection of literature on context-mechanism-outcome congurations related to organizational KT strategies, and end with a renement of initial theories and a synthesis. Discussion: Using a systematic and rigorous method, this review will help guide decision makers and researchers in choosing the best organizational strategies to optimize the implementation of evidence-based practices.


Background
Knowledge translation (KT) (1) is an important means of improving health service quality (2). The science of KT is currently booming (1). However, most research to date has focused on the upstream stages of the process (e.g., assessing facilitators and barriers or monitoring the use of interventions) and only a limited number of studies have examined the effectiveness of KT strategies (3,4). KT effectiveness refers to the ability of the different strategies employed to produce results for users, stakeholders, and the health system (3,4). Research on KT strategy effectiveness has primarily dealt with individual strategies, i.e., strategies that directly target the modi cation of health workers' knowledge, attitudes, and behaviours (5,6). In general, these strategies are moderately effective in changing practices (maximum 10% change) (5). In this context, it is imperative to consider other KT avenues than those focused on individuals alone in order to continue to improve health service quality.
A promising new avenue for supporting service quality improvement through the translation of evidencebased practices targets changes in organizational contexts (e.g., change readiness, general and speci c organizational capacity, organizational routines) (7)(8)(9)(10)(11). Increased emphasis (12)(13)(14) is being put on the decisive in uence that organizational KT strategies could have on evidence-based practice implementation. This in uence is now included in the majority of theoretical and conceptual models explaining this phenomenon (15)(16)(17). Organizational strategies are de ned by Wensing et al. (5) as "planned rearrangements of one or more aspects of the organization of patient care." In contrast to individual strategies, organizational KT strategies focus on changes in the care and service contexts of organizations, such as rede ning professional roles, changing meeting structures, and care protocols. These strategies, in addition to those focused on the individual, have the potential to signi cantly catalyze health and social service quality improvement. However, their mechanisms of action and effectiveness have been poorly investigated and remain unclear (18). Recent studies conducted in the Quebec trauma rehabilitation context have highlighted the value of organizational implementation strategies to promote the effective and sustainable implementation of evidence-based practices (19)(20)(21)(22). For example, changes in disciplinary roles, interprofessional meeting structures, and physical changes to the work environment were reported as facilitating implementation in this particular context. In 2008, Wensing et al. conducted a structured review of systematic reviews of the effectiveness of organizational strategies to improve professional practices and health outcomes for users; their results were equivocal, as none of the documented strategies produced consistent results. In addition, only the effectiveness of these strategies was analyzed, without any attempt to open the "black box" of their functioning. The action mechanisms of these strategies have not been documented and it is therefore di cult to understand how they may or may not contribute to changing the behaviours of health professionals.
Furthermore, in 2012, Flogren et al. conducted a systematic review of the effectiveness of organizational structures in promoting the use of evidence-based practice in nursing; they found only one study with a su ciently rigorous design that did not conclude on the effectiveness of the strategy used for three months post implementation (23). As regards our rehabilitation practice context speci cally, it is characterized by certain particularities in terms of KT due to the interdisciplinary nature of interventions, on the one hand, and the evolutionary nature of patient pathologies and hence their impact on patients' capacity and lifestyle, on the other.
A realist review is required to assess the different action mechanisms of organizational KT strategies (how, why, for whom, to what extent, and under what circumstances) and to facilitate the choice of effective strategies that will enable effective and sustainable implementation of evidence-based practice for allied health professionals in traumatology settings. Realist synthesis (or review) is a scienti c literature review approach that provides an explanatory analysis of how theories work (or fail to work) in different contexts (24). It is used to develop an understanding of the implementation chain of interventions by detailing the prerequisites to achieving positive results and avoiding negative outcomes (24). No research has been done to date on the functioning and effectiveness of organizational KT strategies with allied health professionals in traumatology settings. Focusing on a new organizational target can improve KT effectiveness and service quality.

Study Design and Objectives
We will conduct a realist review with the aims to identify why, how, and under what conditions, organizational KT strategies have been shown to be effective or ineffective in changing the a) knowledge, b) attitudes, and c) clinical behaviours of allied health professionals in traumatology settings.
Data synthesis will take into account the circumstances in which these organizational KT strategies were used, including the characteristics of the targeted evidence-based practices, as well as internal, external, and personal contexts of individuals involved, in addition to any other information that supports the understanding of the context. We report this realist review protocol following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses for protocols (PRISMA-P) (25)

Stages of the Realist Review
The realist review consists of four main stages: 1) Formulation of initial theories, 2) Search for evidence, 3) Knowledge extraction and synthesis, and 4) Recommendations (24).

1) Formulation of initial theories
In terms speci c to the realist review, we will put forward initial theories to explain the contextmechanism-outcome system (CMO). From a methodological point of view, researchers begin by extracting from the literature the main ideas that relate to a class of interventions (the initial theories).
These initial theories explain how and why a class of interventions (in this case, organizational KT strategies) work and generate the outcome(s) of interest (in this case, change in the knowledge, attitudes, and clinical behaviours of allied health professionals in traumatology settings).
We will use the Consolidated Framework for Implementation Research (CFIR) (15) to guide the creation of the initial theories. The CFIR is a metamodel that groups constructs used in 19 different KT theories. This model distinguishes ve main areas of factors in uencing the achievement of KT strategies: 1) the characteristics of the evidence-based practices to be implemented; 2) the external context; 3) the internal context; 4) the individuals involved; and 5) the implementation techniques. CFIR is widely used in the study of complex implementation projects where interactions between factor categories are expected (27).
Based on CFIR and team members' own clinical and organizational experience, we will develop initial theories explaining why, how and in what context organizational KT strategies can work. Speci cally, we will hold a 1,5 hour, audio-recorded, group discussion to develop a rst-draft theory answering the following question: how (context and mechanism) an organizational KT strategy (intervention) has been shown to be effective or ineffective in changing the a) knowledge, b) attitudes, and c) clinical behaviours of allied health professionals (outcomes). We will make an initial schematic representation based off the discussion and re ne it through revision of the audio-recorded material. This representation will then be tested against data from studies included in the review.

2) Search for evidence Information Source and Search Strategy
The search will be conducted in the Pubmed, Embase, CINHAL, Cochrane Library, and Conference Proceedings Citation Index -Science databases and may be extended to other databases of a more social or sociological nature as required. It should be noted that, depending on the realist review methodology, subsequent iterative research may be conducted to generate additional data on a particular aspect of the research question.
Two knowledge management specialists will perform the primary article search in the various databases. Subsequently, the research team will use three different mechanisms: 1) the list of cited references will be reviewed to extract articles of interest; 2) the " nd similar articles" function of the databases will be used to check if other articles of interest are related to the primary articles; 3) authors who have published more than two articles deemed relevant will be contacted by e-mail to request a list of relevant references, published or not, on our subject. All articles found during the secondary search will in turn be put through the three secondary search mechanisms to complete the search. Finally, Grey Matters, a practical tool for searching health-related grey literature (28) will be used to complement the secondary research. The list of articles will be recorded in a specialized reference management software (Endnote) and checked to eliminate duplicates.

Eligibility criteria
We will consider French-and English-language studies relating to organizational KT strategies published between 1990 and the study end date. We believe that the last thirty years are su cient to identify organizational KT strategies that correspond to the reality of the current health network. We will include studies conducted with allied health professionals (occupational therapists, physiotherapists, social workers, psychologists, speech-language pathologists, sexologists) in traumatology settings. The outcomes considered will be the objectively measured effects on the clinical knowledge, attitudes, and behaviours of allied health professionals.

Study selection
First (by title and abstract) and second (full text review) screenings will be carried out independently by two reviewers (KL, MEL) using Covidence (29). We will pilot the screening form across a random sample of 20 titles and abstracts. In the event of a difference of opinion for a given article, a third researcher will take a position on the eligibility of the study. The same method will be applied to full-text reviews. The evaluation of the articles will focus on (a) Relevance, i.e., does the study create at least one CMO con guration about organizational KT strategies and (b) Rigour, i.e., whether this CMO con guration has su cient weight to make a methodologically credible contribution (24). This will lead to the development of the nal list of included articles that will be retained for the data collection process. It is possible that the choice of articles may raise questions about the focus of the review and, as a result, iteratively re ect on the research question and the selection of articles (24). These questions and re ections will be put to the research team.

Data Collection Process
We will carry out a thorough reading of the selected articles followed by the extraction of relevant data. The extraction will be also performed in parallel by two research team members (MEL, JD) using a standard extraction form to reduce variability and bias (30). The form will be based on the CFIR theoretical model (15) and will support the use of realist logic to answer the review question(s), e.g., data on context, mechanisms, and outcome con gurations, demi-regularities, middle-range and/or program theories. The extraction form will be piloted with ve studies. Here again, data extraction can lead to questions about the focus of the synthesis and, consequently, to iterative re ection on the research question and selection of articles. These questions and re ections will also be referred to the research team.

4) Recommendations
Based on the emerging results of the studies, the research team will be able to validate (con rm or refute) the initial theories on the effectiveness of organizational KT strategies through group discussion. The team will perform a comparative analysis between the initial theories and the emerging results of the review as well as a search for both contradictory and consensual results (24,26). Speci cally, they will meet for a 1,5 hour, audio-recorded group meeting. From this discussion, several proposals for CMO combinations will emerge to better specify which organizational KT strategies are most effective in in uencing professionals' knowledge, attitudes, and behaviours. According to the data, it will be possible to identify patterns of CMO and to identify a nal theory that will enable a theoretical generalization of the observed results in other contexts (31). This in turn will "[p]rovide guidance on what policy-makers or practitioners could put in place to change the context or provide resources in a way that most likely triggers the appropriate mechanism(s) to produce the desired outcome" (26). We will test the nal theory against audio-recording to con rm our ndings.

Discussion
Using a systematic and rigorous method, this review will help guide decision makers and researchers in choosing the best organizational strategies to optimize the implementation of evidence-based practices in trauma eld.

Anticipated Challenges and Limitations
Researchers have explained some of the challenges that are speci c to realist reviews. Among other things, making the distinction between mechanism and context can be di cult (32)(33)(34). Robert and Ride explain that the mechanism (M) in one CMO con guration can be the context (C) of another C-M-O con guration (34). Moreover, Pawson and al. mention there is a limit to what can be understood through the available information. Less tangible information such as relationships of in uence and power, for example, may not be described in an article (35); hence the importance of opening up the search for information to documents other than primary studies, but this also makes it more di cult to assess the quality of the information. In this context, quality assessment is not necessarily based on formal grids, but mainly on the judgment of the authors of the synthesis with regard to data credibility and its impact on theories (35). Finally, the recommendations based on the synthesis may also be limited. It will be di cult for the authors of the synthesis to claim that they have grasped all the opportunities and constraints associated with interventions and to predict the circumstances under which subsequent projects will be implemented (35). This limits the scope of the recommendations. It is crucial that future users of the knowledge produced through realist synthesis understand the usefulness, but also the limitations of this approach. This underscores the importance of their participation in the review from the outset.