A Protocol for a Systematic Review of the Effectiveness of Interprofessional Education Interventions in Pre-Licensure Health Professions Education

Background: A growing body of research demonstrates that interprofessional education (IPE) has the potential to lead to successful interprofessional collaborative practice (IPCP). In turn, IPCP has potential to mitigate the negative effects associated with an underfunded and shrinking healthcare workforce and population growth. Nonetheless, over the past decade, there has been a lack of systematic reviews that have examined the effectiveness of IPE in preparing pre-licensure health professions students for IPCP. The purpose of this paper is to provide a protocol for a systematic review that will explore and identify (1) whether IPE interventions in the past decade have addressed the shortcomings of past interventions, (2) the effectiveness of recent IPE in preparing students for IPCP, and (3) current knowledge gaps within the eld. Methods: We used the Cochrane Collaboration and the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) guidelines to articulate this protocol. Furthermore, we used the Population, Intervention, Comparison, Outcomes, and Study Design (PICOS) framework to formulate our research questions, “What are the major trends and ndings of IPE implementation studies conducted over the last decade (2010–2020)?” and “How do the trends and ndings of IPE studies conducted over the last decade compare with IPE studies in the decade prior?” Discussion: The systematic review is ongoing. We expect to complete it by September 2021. This systematic review will provide a useful synthesis of best practices in the implementation of IPE between 2010 and 2020 as well as their effectiveness in preparing health professions students for IPCP. Systematic review registration: We submitted this protocol with PROSPERO on January 19 th , 2021.


Background
According to the World Health Organization (WHO, 2006), healthcare systems worldwide have been struggling to adequately meet the needs of their patient populations, mainly due to underfunded and shrinking healthcare workforce (WHO, 2006(WHO, , 2010 and population growth (Van Bavel, 2013).
Interprofessional education (IPE) and interprofessional collaborative practice (IPCP) have been proven to mitigate these effects with aim of providing high quality healthcare and best possible individual health outcomes (Hammick et al., 2007;Lemieux-Charles & McGuire, 2006;Zwarenstein et al., 2005). As such, the WHO (2010) has been promoting IPE, which occurs when "two or more [health] professions learn about, from and with each other to enable effective collaboration and improve health outcomes," (p. 13), whereby the implementation of IPE is "a key step in moving health systems from fragmentation to a position of strength" (p. 10).
Several previously published systematic reviews (e.g., Abu-Rish et al., 2012;Lapkin et al., 2013;Zhang et al., 2011) have examined and provided valuable syntheses of IPE implementation from a total of 117 studies that were collectively conducted over more than a decade; between 1999 and 2011. These reviews, which are discussed in more detail below, outlined several shortcomings as well as offered explicit recommendations for future research and practice. One decade later, many questions regarding the status quo of IPE implementation remain unanswered, including but not limited to: What are the IPErelevant best practices in preparing pre-licensure health professions students for IPCP upon graduation?
What theoretical and conceptual frameworks have been employed to guide the implementation of IPE?
Have IPE facilitators developed interventions that mandate participation of students from various health professions whereby long-term outcomes are assessed? Have IPE interventions been enhanced to include various health professions? In addition to the widely investigated short-term student learning outcomes, have IPE interventions been evolved to emphasize long-term programmatic, patient-oriented, and facilitator-oriented outcomes, as well?
The purpose of this paper is to provide a protocol for a forthcoming systematic review of relevant literature published between January 1st, 2010 and December 31st, 2020. In so doing, we will attempt to seek answers to the questions presented above. Answering these questions through research-informed undertakings is important to understand the effectiveness of IPE and identifying the best practices that lead to IPCP, and eventually to improved health systems and individual health outcomes (Hammick et al., 2007;Lemieux-Charles & McGuire, 2006;Zwarenstein et al., 2005).

Methods
This protocol is informed by the elements of the Cochrane Collaboration (Higgins et al., 2019), which divides the general methods for conducting systematic reviews into three main stages: (1) Preparatory Steps; (2) Search of the Literature; and (3) Analysis. Furthermore, we consulted the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) checklist (Moher et al., 2015) and its supplementary pedagogical "elaboration and explanation" document (Shamseer et al., 2015), which describe 17 items that should be included in any protocol (see Additional File 1 for the completed PRISMA-P checklist). The use of these resources in articulating this protocol ensures that we provide the rationale for the forthcoming systematic review (Moher et al., 2015) and fully report key ndings consistently (Shamseer et al., 2015), establishes the review as a legitimate, independent research study, and reduces author bias (Gupta et al., 2018).

Preparatory Steps
Developing the Research Question. A well-formulated research question is critical for performing an effective literature search, de ning the scope of the systematic review, and identifying the population of researchers, educators, and students to which the ndings are most relevant (Gupta et al., 2018). Hence, using the parameters delineated in the "PICOS framework" , we developed the following research questions, "What are the major trends and ndings of IPE implementation studies conducted over the last decade (2010-2020)?" and "How do the trends and ndings of IPE studies conducted over the last decade compare with IPE studies in the decade prior?" Evaluating the Quality of the Systematic Review. Evaluating the quality of our review is two-fold (Gupta et al., 2018). First, the feasibility of the review must be assessed. As such, we con rm that the required resources and expertise to conduct this are available, including the dedicated time commitment by the authors as well as the funding required for successful completion.
Lack of Theoretical Underpinnings. Most interventions prior to 2010 were not widely guided by neither guided nor supported by theoretical and conceptual frameworks. Furthermore, even when some researchers employed some form of framework to guide their interventions, they were only ambiguously described. The integration of theory and practice enables the effective development of IPE interventions, identi es the speci c learning objectives, and facilitates the assessment of learning outcomes (Clark, 2006). Hence, it is crucial that IPE interventions are guided through theoretical or conceptual frameworks.
These frameworks are usually instituted within social constructivist perspectives, through which researchers can interpret how student learning occurs within one's social environment, where social interactions during IPE interventions help shape each student's attitudes, beliefs, and skills, (Vygotsky, 1978), ultimately leading to students making meaning of their experiences (Creswell & Poth, 2018).
Lack of Research Rigour. Most interventions lacked rigour of their study design and methodology. One example is that many researchers relied solely on quantitative methods to implement their interventions and report their ndings. Researchers within the IPE eld are increasingly encouraged to employ qualitative or mixed methods approaches when implementing their interventions to enable them to report a wider array of critical ndings. The fact that qualitative and quantitative approaches to research are inherently different indicates that each brings distinctive qualities to the research process; thus, we believe that both approaches can complement one another. While quantitative methods may be interested in the generalizability and outcomes of meaning-making by examining the functional relationships of operational variables, qualitative methods delineate the meaning-making process to provide rich, thick descriptions of how and why people make meaning of their experiences (Merriam & Tisdell, 2016). Indeed, "no single method can grasp the subtle variations in ongoing human experience," (Denzin & Lincoln, 2011, p. 12), where the interplay which results from the integration of both qualitative and quantitative approaches "can bring accounts of social phenomena to progressively greater levels of clarity," (Cupchik, 2001, para. 33). Other examples of inadequate research rigour include the use of small samples of convenience and the employ of post-test-only surveys.
Unrepresentative of All Health Professions. Most IPE interventions largely included medical and nursing students as major participants, with minimal participation from other health professions. For instance, Abu-Rish et al. (2012) reported that approximately two-thirds of the interventions they reviewed involved students from nursing and one other profession, where medicine dominates over all other professions. Furthermore, Lapkin et al. (2013) and Zhang et al. (2011) also reported that in the interventions they each reviewed, approximately half included nursing and medical students alone. These ndings indicate that, at the very least, the outcomes of more than half of all the interventions can neither necessarily be generalized to nor representative of all health professions.
Short-term Interventions with Potentially Temporary Effects. Interventions were predominantly shortlasting and, therefore, lacked evidence of the effectiveness of IPE in the long-term. For instance, Abu-Rish et al. (2012) demonstrated that approximately three-fths of interventions were offered once, whereby the vast majority lasted for durations between 6 hours and one 12 weeks. These tendencies restrict the development of best practices that provide evidence for the long-term effectiveness of IPE in preparing health professions students for successful IPCP, leading to improved health outcomes. Thus, researchers are encouraged to implement longer-term interventions and to evaluate the outcomes of such interventions longitudinally (Lawn, 2016), even after participants have graduated and entered the healthcare workforce.
Exclusive Emphasis on Student Learning Outcomes. Interventions generally reported student learning outcomes including increased positive attitudes towards IPE, student satisfaction with IPE, and gains in pro ciency of IPE-relevant skills and competencies. Nonetheless, most interventions failed to report outcomes other than those of student learning, such as programmatic, facilitator-oriented, and patientoriented outcomes. Assessing these outcomes provides understandings of the perceptions of not just student participants, but of all the main stakeholders as well; thus, leading to enhanced design, delivery, and implementation of future IPE interventions.
Other shortcomings include the lack of administrative support, insu cient funding, inadequate faculty development, and logistical di culties when scheduling interventions that include numerous health professions. Our review is relevant to the subject matter to the extent that it builds on these existing systematic reviews. We will attempt to reveal whether recent studies (2010-2020) have enhanced the implementation of IPE interventions, in part, by addressing the above-mentioned shortcomings. In so doing, we will determine how effective such interventions are in preparing health professions students for IPCP as well as identify potential best practices.
Establishing the Inclusion and Exclusion Criteria. Through working with an expert member of the Canadian Interprofessional Health Collaborative (CIHC), we established our inclusion and exclusion criteria. For feasibility purposes, we will exclude search results that were not peer-reviewed, were written in a language other than English, or were published prior to 2010.
The rst of three inclusion criteria mandates that the identi ed study must involve the health professions in a health education setting. In this study, we will examine the implementation of IPE worldwide; therefore, we adopted a universal de nition for "health professionals," which the WHO (2006) describes as "all people primarily engaged in actions with the primary intent of enhancing health" (p. xvi). This de nition encompasses not only those professions specialized in physical health, but those specialized in mental and social health, dental health, and ocular health, as well.
Furthermore, we will focus on pre-licensure education, as opposed to both pre-licensure and postlicensure education. Therefore, the second inclusion criterion is that the study must involve pre-licensure health professions students. In North America, pre-licensure education in the health professions occurs at both the undergraduate level (e.g., as is in medicine, nursing, and social work) and the graduate level (e.g., physical therapy, occupational therapy, and pharmacy).
Lastly, the study must describe an interprofessional intervention. IPE interventions may have diverse methodological approaches; as such, no restrictions for inclusion apply for either the type of study design or the presence (or lack thereof) of comparison groups. Furthermore, in accordance with the WHO's (2010) de nition of "interprofessional education," an intervention must involve students from "two or more [health] professions," whereby the outcomes of students learning "about, from and with each other" (p. 13) are reported. Hence, all types of systematic reviews (including scoping and meta-analyses), articles describing IPE-relevant curricula, abstracts, and instrument validation studies will be excluded.
Registering the Review. Registration of the review mitigates unnecessary duplication of studies, increases transparency, and reduces bias (Gupta et al., 2018). This protocol has also been submitted for registration with the International Prospective Register of Systematic Reviews (PROSPERO) on January 19th, 2021. Search of the Literature Selecting the Databases. We consulted with two academic librarians with expertise in systematic review searching from the Teaching and Learning Team at our institution to help us identify the databases for the search processes. With their guidance, we selected four electronic databases: (1) Cumulative Index to Nursing and Allied Health Literature (CINAHL); (2) the Education Resources Information Center (ERIC); (3) PubMed; and (4) Scopus. These databases were chosen because they offer a wide array of studies in health and health education.
Conducting and Evaluating the Quality of the Search. With the librarians' guidance, we generated the appropriate search terms using the established inclusion criteria in preparation for completing the search of the electronic databases. The generic search strategy is presented in Table 1 (see Additional File 2 for the search strategy on CINAHL). We consulted with the librarians and successfully conducted an evaluation of the search strategy. We have yet to lter the studies-the next stage in our systematic review. Table 1 General Search Terms for Search of Databases 1. ("interprofession*" OR "inter-profession*" OR "health profession*" OR "health occupat*" OR "healthcare profession*" OR "health care profession*" OR "health and social care profession*") AND 2. ("education" OR "collaborat*" OR "collaborat* practice" OR "teamwork*" OR "practice") AND 3. ("implement*" OR "simulat*" OR "workshop*" OR "interven*") Filtering the Studies. Following the database search, identi ed articles will be uploaded onto Covidence (Veritas Health Innovation Ltd, 2021) to be ltered according to the modi ed PRISMA owchart (Moher et al., 2009) illustrated in Fig. 1. Once duplicates are removed, we will independently lter the studies according to the inclusion and exclusion criteria. Initial screening and evaluation of the titles and abstracts will be followed by an evaluation of the full texts of the remaining studies. Finally, we will compare our results and resolve any disagreements by discussion until consensus is reached and if necessary, have a third researcher adjudicate the nal judgement.

Analysis
Extracting the Data. We will extract and tabulate the data on Covidence (Veritas Health Innovation Ltd, 2021) in preparation for data analysis. We developed a data extraction checklist for documenting the major elements, attributes, and procedures reported in the included studies. Our checklist is comprised of a total of 44 items that we categorized into the following seven domains: (1) Details of Publication; (2) Study Pro le; (3) Characteristics of Intervention; (4) Demographics of Students/Participants; (5) Demographics of Educators/Facilitators; (6) Methods and Procedures; and (7) Outcome Measures (Table 2). Evaluating Quality of Included Studies. We will utilize the Mixed Methods Appraisal Tool (MMAT; Hong et al., 2018) to independently to evaluate the quality of the included studies. The MMAT is useful for assessing the quality of all research approaches including completely quantitative and qualitative studies as well as both randomized and non-randomized studies. The tool has been well-established, veri ed for reliability and content validity, and used in dozens of systematic reviews.
Evaluating the quality of each study using the MMAT involves several stages. Initially, we will respond with either "Yes" or "No" to two screening questions, (1) Are there clear research questions?" and (2) "Do the collected data allow to address the research questions?" Then, we will respond to one of several sets of questions, depending on the type of research approach. If incomplete or discrepant data were reported for any given question, we will respond with "Can't tell" and either consult the study protocol (if publicly available) or contact the study investigators for clari cation using open-ended questions. We will resolve any disagreements by discussion until consensus is reached and if necessary, have a third researcher adjudicate the nal judgement. Then, we will provide a detailed presentation of our evaluation, possibly perform a sensitivity analysis, and report ndings in publication. Lastly, we will assess the strength of the body of evidence using the Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) approach (Brozek et al., 2009).
Analysis and Synthesis. We expect the extracted studies to be heterogeneous and have methodological diversity in interventions, research approaches, instrumentation, and data collection and analysis. As such, undertaking a meta-analysis of the extracted studies is neither appropriate nor feasible within our review. Alternatively, guided by Ryan (2013) and Popay et al. (2006), we will use narrative synthesis to organize and present the data obtained from our review in responding to our research question. Our narrative synthesis will go beyond providing summaries and primary characteristics to include a comparative analysis of differences and similarities of extracted studies and the emergence of salient themes and patterns (e.g., design of the study, nature of study participants, characteristics of intervention, outcomes sought or measured, etc.). The narrative synthesis will also include an analysis of relationships within and between studies as well as their outcomes and the factors affecting those outcomes.

Discussion
The systematic review is ongoing. We are now in the process of conducting the search of electronic databases and expect to complete the review by September 2021. The key ndings from this study will allow us to identify best practices of IPE implementation, as well as its effectiveness in health professions education programs and its impacts on IPCP. We will also scrutinize the strengths, weaknesses, and challenges of the included studies as well as the limitations of our systematic review itself. Furthermore, we will examine and compare our results with those of previously published reviews to identify whether included studies have addressed the IPE-relevant shortcomings since. Finally, we will discuss potential contradictory ndings, recognize gaps in the literature, and offer recommendations for future research and practice.

Consent for Publication
Not applicable.

Availability of Data and Materials
The search strategies of individual electronic databases for the current study are available from the corresponding author upon request.