Background

Interprofessional collaboration (IPC) in healthcare is regarded as important by healthcare professionals and increases quality of care while decreasing costs (Schmitz et al., 2017; Wei et al., 2020, 2022). It involves two or more health or social care professions providing care and can positively impact several domains, including healthcare access for users, patient satisfaction, and length of hospital stay (Reeves et al., 2017; Wei et al., 2022). For providers, IPC can contribute to facilitating information flow, fostering professional exchange between professions and increasing workplace satisfaction (Reeves et al., 2017; Wei et al., 2022). IPC is often defined as the coordinated effort of various healthcare professionals to deliver comprehensive, quality care to patients and communities across multiple settings (WHO, 2010). The IPC construct is conceptually proposed to comprise four interrelated dimensions: Interprofessional education (IPE), organizational requirements, practical interprofessional collaboration, and the effects of such collaboration (Wagner et al., 2019). IPE is commonly described as consisting of situations that allow students or professionals to “learn with, from and about each other to improve collaboration and the quality of care” (CAIPE, 2017, p. 14).

With IPE as a prerequisite of IPC (Spaulding et al., 2021; Wagner et al., 2019), educational institutions training healthcare professionals have a unique opportunity to cultivate IPC in their future workforce. IPE has been found to influence learners’ attitudes, knowledge, and skills of collaboration (Campion-Smith et al., 2011; Makowsky et al., 2009; Reeves et al., 2013; Sargeant et al., 2011), which are plausible vectors through which IPE affects IPC. Learning is often defined in terms of long-term behavior changes that are preceded by changes in learner attitudes (De Houwer et al., 2013). This points to attitudes as a feasible target for educational efforts and is supported by theoretical propositions in Ajzen’s (1991) Theory of Planned Behavior. As a form of academic learning, IPE may favorably affect learners’ attitudes toward members of other professional groups and attitudes towards IPC. Indeed, IPE explicitly aims to improve learners’ knowledge, skills, and attitudes (Stephens & Ormandy, 2018).

Despite the theory-based expectation that IPE affects attitudes towards IPC, the contextual factors and underlying mechanisms by which IPE affects attitudes are not always clearly delineated in the literature. Much of the empirical research focuses on predictor variables that are statistically associated with interprofessional attitude score improvements after conducting IPE (e.g., Biehle et al., 2019; Bloomfield et al., 2021; Fusco & Foltz-Ramos, 2018; Lockeman et al., 2017). However, literature syntheses show mixed results of IPE’s impact on attitudes (Berger-Estilita et al., 2020a, 2020b; Spaulding et al., 2021). The contradictions are partially attributed to the variety of different IPE initiatives and the lack of sufficiently sophisticated evaluation methodology (Thistlethwaite, 2012). There also appear to be only a few reviews summarizing the evidence of IPE’s impact on attitudes applying a conceptual model (e.g., Hammick et al., 2007; Reeves et al., 2016; Spaulding et al., 2021). These shortcomings highlight the need for realist approaches (Wong et al., 2012), which consider the contextual factors in which attitude change occurs, as well as the mechanisms associated with positive attitude change. By developing a more fine-grained understanding of how IPE works to develop positive attitudes, we can support the development of more targeted IPE curricula and more conducive learning environments.

As a theory-driven method, realist synthesis enriches data analysis with existing theory (Jagosh, 2019). At least four branches of IPE literature can be discerned which address theory, theory development, or theory application. The first branch consists of competency frameworks, such as of the Interprofessional Education Collaborative (IPEC, 2016) and the Canadian Interprofessional Health Collaborative (CIHC, 2010), which describe some of the theoretical assumptions of how IPE works. The second branch of literature consists of empirical studies which are framed or interpreted from a specific theoretical perspective, for instance the Theory of Planned Behavior (e.g., Keshmiri et al., 2020; Przymuszala et al., 2023) or Social Learning Theory (e.g., Chen et al., 2022; Wu et al., 2022). These middle-range theories (MRTs) are frameworks that organize hypotheses and facilitate the development of empirically testable propositions (Boudon, 1991; Merton, 1949). A third literature branch evaluates the contribution of theory to IPE curriculum design, delivery, and evaluation (e.g., Anderson et al., 2016; Hean et al., 2018). A fourth branch consists of literature reviews which contribute to theory development by summarizing the effects of IPE on learning and other outcomes (e.g., Reeves et al., 2017; Spaulding et al., 2021).

The application of these various forms of theory can guide analysis and fill evidence gaps concerning mechanisms with theoretical assumptions (Jagosh, 2019). Drawing on various sources of data, the realist heuristic of context-mechanism-outcome configurations (CMOCs) can be applied to uncover hidden causal mechanisms producing positive attitude outcomes and the conducive contexts that trigger them (Jagosh, 2019, 2020; Pawson & Tilley, 2004).

Even though realist synthesis provides a useful heuristic for theory development (Jagosh, 2019), theory inevitably remains undertermined by evidence (Turnbull, 2018). That is, often multiple theories are equally well suited to explain a specific set of data, a fact that has also been echoed in the realist synthesis literature (Wong et al., 2010). This circumstance requires one to consider a range of criteria when developing IPE theory or applying an existing one to explain IPE phenomena. These criteria may also serve as additional heuristics to guide theory development. For instance, one might consider the context of theory application, i.e., whether the theory is to serve in IPE curriculum design, delivery, or evaluation (Hean et al., 2018). Hean et al. (2016) have also proposed considering aspects of theoretical quality such as parsimony, testability, operational adequacy, empirical adequacy, and quality of theory application. Theory selection may also depend on dimensions such as which group of stakeholders a theory is intended to serve (e.g., curriculum developers, facilitators), when in the IPE intervention a theory might be applicable (i.e., pre- or post-registration), the learning environment, the utility of a theory in helping achieve an intervention goal (Hean et al., 2012), or its level of analysis (i.e., individual, group, or system level) (Hean et al., 2018). Ultimately, theory provides IPE curriculum developers with the theoretical foundations to support the IPE activities being implemented (Hean et al., 2018) and to help reflect on existing practices (Hean et al., 2012). Thus, it may be useful to take an instrumental approach (Suppes, 2014) or tool box approach (Hean et al., 2012) to selecting a theory, in order to pick one that is best suited to describe, explain, or predict phenomena of interest, make the links explicit between an intervention and an outcome (Hean et al., 2016), and inform the social processes underlying IPE curricula (Hean et al., 2018). It is also necessary to ensure that developed IPE theory is useful for practioners, for instance in curriculum development, IPE delivery, or evaluation, and the links between theory and practice clearly put forward (Hean et al., 2016).

Of the criteria described, we focus our theory development and MRT selection on their ability to describe and explain positive attitude development, ensuring explicit links to IPE, and providing useful recommendations for IPE practitioners. The overarching research questions we investigated were: How does IPE work to positively influence student attitudes toward IPC, IPE and professionals from other healthcare fields? For which learners is IPE effective? Under what circumstances does IPE work?

Methods

We conducted a rapid realist synthesis to identify CMOCs that link IPE to learner attitudes towards IPE, IPC, and professionals from other disciplines. As a method that provides the tools to investigate “what works, for whom, how, and in what circumstances” (Pawson et al., 2005, p. 32), realist synthesis is particularly suited to addressing the IPE literature’s shortcoming by using the CMOC heuristic to uncover the hidden mechanisms (Jagosh, 2019) through which IPE affects learner attitudes. Applying this form of theorizing called retroduction, we began with the outcomes and analytically worked our way backward to the conditions required to produce them (Jagosh, 2020, p. 129). With changes in attitudes as our outcome, we looked for potential mechanisms that could be generating changes in attitudes. We also paid attention to intermediate outcomes that might be linked to attitude change. We were mindful that outcomes cannot always be predicted by deterministic theory. Instead, contexts can be expected to shape decisions and behavior in ways that are partially predictable because people are likely to make similar decisions or behave similarly given a certain set of circumstances (Wong et al., 2010). Thus, we analyzed the data looking for descriptions of contexts, as a context provides the trigger for every mechanism (Dalkin et al., 2015; Greenhalgh & Manzano, 2022) that leads to changes in attitudes. We used multiple sources of evidence, as realist synthesis does not privilege specific sources of evidence over others (Duddy & Roberts, 2022; Wong, 2018), combining information gained from qualitative interviews with IPE subject matter experts (SMEs) and published studies. Finally, we looked for MRTs to support our proposed mechanisms and used them to explain our findings, building on their theoretical propositions for how and why specific mechanisms are necessary to yield intended outcomes. The Theory of Planned Behavior (Ajzen, 1991) is one example of an MRT we used.

Endeavoring to complete the review within approximately one year, we set fixed dates to complete the various project phases. This limited the length of time available for analysis iterations, group discussions, and developing versions of CMOCs and program theories. This paper, a synthesis of literature and SME input, follows the RAMESES publication standards for realist reviews (Wong et al., 2013).

Data collection

Data collection comprised three elements: 1) a preliminary literature search, 2) qualitative interviews with subject matter experts, and 3) a realist synthesis of documents contributing to development and refinement of a PT and CMOCs.

Preliminary literature review

The preliminary literature review targeted existing systematic reviews on the effects of IPE on attitudes, as this body of literature provides a comprehensive overview of the empirical literature and and may point to potential mechanisms by which IPE affects attitudes. This literature supported initial program theory (IPT) development of how IPE works to affect students’ attitudes towards IPE, IPC, and professionals outside of one’s own profession. This preliminary review was also used to develop an interview guide for subject matter experts (SMEs) and search terms for the realist review.

Qualitative interviews

The interview guide included questions about the components of IPE to better understand the contextual factors (e.g., teaching environment and conditions, student background and characteristics) and to explore potential underlying mechanisms associated with attitude change in students. SME interviews were conducted in Swiss German with healthcare professionals in Switzerland who have backgrounds in nursing (2 SMEs), medicine (1 SME), physiotherapy and neuropsychology (1 SME). While three SMEs are involved in interprofessional curriculum development and teaching, one SME heads an interprofessional clinic and serves on the board of directors of a large university hospital. All persons interviewed are mid- to late-career professionals with extensive experience in interprofessional practice. Three SMEs were actively involved in interprofessional teaching at the time of the interview. The interviews took place between November 2021 and July 2022. Two interviews took place online using the MS Teams video platform, and two took place in the respective offices of the SMEs.

Realist synthesis

Based on the preliminary literature review and SME interviews, a realist synthesis search strategy was developed. A Population-Intervention-Comparison-Outcome (PICO) framework was used to organize eligibility criteria (Table 1) for a systematic database search of MEDLINE (PubMed), CINAHL, PsycINFO, and Social Science Citation Index. The target population included pre- and post-registration learners so as to provide insights into IPE impacts on attitudes in the formative stages of professional development, as well as insights into IPE’s potential impact on attitude maintenance after entry into practice. IPE and multiprofessional education were included as interventions to be studied, under the theoretical assumption that they may share some similar mechanisms and some unique mechanisms that contribute to positive attitudes. The broader selection criteria chosen may allow studies to be found which permit a comparative analysis of group outcomes, a broader generalization of findings, and the potential discovery of IPE mechanisms of effect common to both contexts. The search strategy was developed with the support of a medical librarian at a medical university library. Four groups of search terms were combined: terms related to 1) education, 2) attitude, 3) practice, and 4) impact (Table 2). The full search strategy is included in Supplementary File 1. Various types of publications and all study designs (quantitative, qualitative, mixed methods, review) were included if published between January 1, 2010 and June 13, 2022. Publications were assessed based on the realist synthesis criteria of relevance and rigor (Dada et al., 2023; Wong et al., 2013). Relevance refers to how well the study provides answers to the research question. Rigor refers to the trustworthiness of the research method used to arrive at the study conclusions. To perform these assessments, authors individually reviewed assigned papers for content that could contribute to the research question and for potential methodical issues that could cast doubt on a study’s findings. Critical appraisal tools were not applied, as we prioritized the inclusion of substantive content.

Table 1 Eligibility criteria
Full size table
Table 2 Configuration of search terms
Full size table

The study selection process is illustrated in Fig. 1.

Fig. 1
figure 1

Study selection process

Full size image

Data extraction and organization

A data extraction file in the form of an Excel table was created to document study characteristics (e.g., study sample, setting, study objectives, putative contexts, mechanisms, and outcomes) (Table 4). Another table was used to collect and annotate verbatim text segments from studies that appeared to contain contexts, mechanisms, or outcomes and potential linkages between them, for instance a context-mechanism, a mechanism-outcome, a context-outcome, or a full context-mechanism-outcome linkage. MRTs explicitly cited in the studies were noted separately for later reference. The verbatim text segments were thematically coded, which formed our conceptual “buckets”, for instance as “IPE characteristics” or “IPE learner development/transformation” (Supplementary File 2). The conceptual buckets served as analytic containers for grouping text segments that appeared to discuss the same phenomenon. After all studies and all their informative text segments had been thematically coded, they were sorted according to the conceptual buckets, allowing all text segments of a conceptual bucket to be shown together.

Evidence synthesis

Having grouped several text segments by conceptual bucket and similarity, we aimed to analyze text segments looking for context-mechanism, mechanism-outcome, context-outcome, or context-mechanism-outcome linkages implied within the text, while also paying attention to potentially repeating patterns when looking at several similar text segments. We added additional columns in the analysis table to serve as analytic steps towards developing CMOCs. A simplified version is presented in Table 3 to illustrate the procedure for a central CMOC in our study. We examined the text segments for concepts and explored how the concepts were being related to each other within the text. This resulted in a note of the concept and its function within that particular text segment, i.e., whether a concept appeared to function as a context, mechanism, or outcome (Column 3). For example, in Quote 41, the concept “Formal education arrangement, informal spaces, and opportunities to socialize” appeared to be a context (C). Another concept was “Know each other on a personal level”, which could be interpreted as a mechanism (M). This particular text segment provided a potential context-mechanism linkage. Text segments with similar content were coded in the same manner. The next step (Column 4) was to further abstract the extracted concepts and potential CMOC components from Column 3. For instance, “Formal education arrangement, informal spaces, and opportunities to socialize” became “Time and space to socialize”. In a next step, we tried to fit the various abstracted concepts and their linkages into a coherent CMOC, using the original text segments as aids into a CMOC template statement in the form of: “If/when [context], then [outcome], because [mechanism].” In this process, we allowed ourselves to also be guided by the information gained from the SME interviews and MRTs extracted from the included studies and other MRTs that the research team judged potentially relevant. For instance, the idea of interprofessionally collaborating professionals liking each other being a contributing factor was indicated in one SME interview. As an MRT, Contact Hypothesis was judged to be relevant already prior to data analysis. Some CMOCs, such as the one presented here, were strongly informed by prior theory and interview data, and the data in the studies included supported them. Other CMOCs were more strongly data-driven. Having formulated a candidate CMOC, we compared its fit with other text segments. In some CMOCs, amendments were made to the structure of the CMOC, e.g. what appeared to be an outcome was sometimes discovered to be more suitably interpreted as, for instance, a mechnanism for our research question. For example, “knowing each other professionally and personally” might be considered an outcome of IPE, but it seemed more appropriate to view it as a mechanism which leads to the outcome “trust, respect, and liking among learners”. When we were satisfied with the fit between candidate CMOC and validation text segments, we considered them as pre-final CMOCs. To validate the CMOCs, MRTs from psychology were explored that could explain the underlying mechanisms we extracted from the data, as MRTs offer a more generalized theoretical explanation than a CMOC (Wong et al., 2010). We selected theories that were capable of explaining various phenomena related to positive attitude development and which helped to make explicit theoretical links (Hean et al., 2016) between IPE, intermediate variables, and attitudinal outcomes. Where possible, we selected theories that were previously applied in IPE research, as previous use provides additional evidence of the theory’s applicability for IPE research. The refined PT and CMOCs based on data from included review documents were shared with SMEs for feedback, to ensure rigor and relevance for IP educators and practitioners, and to garner recommendations for enhancements to current IPE.

Table 3 CMOC development
Full size table

We constructed six CMOCs to reflect the outcomes described in the studies and to describe the theoretical links proposed by the developed PT model, the development of which is described in the following chapter.

Model building

As part of IPT and PT development we built a model to describe the components and processes of a learner’s interaction with IPE and their development within it. Its basic structure is based on Biggs’s (1987, 1993) 3P model, which features a three-phase model of student learning, comprising the phases of presage, process and product. It was also used as analytical tool in two systematic reviews on the effects of IPE (Hammick et al., 2007; Reeves et al., 2016), demonstrating its utility in modeling IPE as a process and describing its components. In an analysis iteration we relabeled the phases to “pre-program”, “program”, and “post-program” to better align with presenting IPE as a program of teaching and learning that extends beyond any one course or educational or healthcare institution. The model is a schematic representation in which we integrated theoretical concepts and propositions derived from the preliminary literature review, SME qualitative interviews and conceptualizations of identity, socialization, and development as they are commonly applied in sociological, psychological, and education literature (e.g., Haller & Müller, 2008; Hurrelmann, 2002; Oerter & Montada, 2002). The model helped inform our realist synthesis search strategy and data extraction (e.g., conceptual buckets). We revised the model iteratively as indicated by new data and analyses from the systematic literature review and CMOC development. The final model is featured in Fig. 2.

Fig. 2
figure 2

Program theory

Full size image

Results

Our study results are presented in the sequence of realist theory development, beginning with the IPT and followed by the refined PT. We present the characteristics of the 23 studies included in the realist review in Table 4. Realist review exemplar quotes are provided in Supplementary File 2.

Table 4 Extracted data from included studies
Full size table

Initial program theory

The development of the IPT combined findings from the preliminary literature review and SME qualitative interviews. We used Biggs’s (1987, 1993) Presage-Process–Product (3P) framework (1987, 1993) cited in two systematic reviews on IPE (Hammick et al., 2007; Reeves et al., 2016). We renamed the 3Ps as into the phases “Pre-Program”, “Program”, and “Post-Program”, which together comprise the overarching structure of the PT model (see Fig. 2). In the Biggs framework, presage and process factors are hypothesized to interact to produce the outcome. “Presage” encompasses the conditions prior to IPE and include the characteristics of teachers and learners as well as the “sociopolitical context” (Reeves et al., 2016, p. 658). These provide the context within which the mechanisms of IPE operate (Hammick et al., 2007; Pawson & Tilley, 1997). “Process” describes the IPE approaches to teaching and learning that were applied (Hammick et al., 2007; Reeves et al., 2016) which are factors hypothesized to interact to affect attitudes.

In addition to the preliminary literature, interviews provided some additional key concepts to include as conceptual buckets during data extraction. “Trust,” “respect,” and “liking” were repeatedly used by SMEs to indicate positive conditions for IPE and IPC. For example:

I think interprofessional collaboration is very much based on knowing (…) who has what skills. (…)that requires that there is sufficient contact between the different professions already during training(…) so that you have sufficient trust (…) (Quote 46, Interview 2).

This means that if people like each other, then everything works well on its own. (…) And when people know each other as friends and know personal things [about each other], that creates trust. (Quote 12, Interview 4)

SMEs also noted how some professions may be less receptive to IPE and IPC:

We have been able to determine a clear, clear difference among the students (…) that [interprofessional] collaboration is also viewed as more important by nurses. So, there are differences early on. Quote 33, Interview 3)

What has always come up as a difficult professional category for interprofessional collaboration (…) [is] with physicians. (…) we have seen what the setting can do. (…) [Settings] where [students] have to master everyday situations together. And [students] realise that you really have [to work] together. (Quote 32, Interview 3)

Based on our preliminary search and SME interviews, we summarize the IPT as follows: IPE needs to provide the circumstances that allow enough contact between learners, so that they get to know each other as professionals and also personally. Through IPE, trust, respect, and liking among learners generate positive attitudes when  they realize that they share common values and goals, need each other to solve common practice problems, and must work together to successfully deliver care in complex healthcare settings. Even though there may be differences among learners aspiring to different healthcare professions, developing positive attitudes can be achieved among all professions by fostering sufficient contact and providing a learning setting where learners experience interdependency and collaboration among different professions.

Refined program theory and causal context-mechanism-outcome explanations

Our refined program theory is schematically presented in Fig. 2, beginning before IPE (pre-program), continuing through IPE (program) and concluding with the completion of the program (post-program). This is an adaptation of the Presage-Process–Product or 3-P model (Biggs, 1987, 1993; Hammick et al., 2007; Reeves et al., 2016). The following sections provide overviews of each phase of our developed model in relation to evidence from our realist synthesis.

Table 4 summarizes key data extracted from the 23 papers included in our realist synthesis.

Table 5 summarizes the CMOCs associated with the refined PT (Fig. 2) and a few exemplar quotes are provided for each CMOC. Supplementary File 2 provides additional supporting quotes for the CMOCs.

Table 5 CMOCs and associated quotes and MRTs
Full size table

Pre-Program

CMOC 1: When learners have initial, discipline-specific attitude differences (C), it will influence IPE’s impact on learners’ positive attitude development (O) because learners’ existing attitudes act as anchors against which new attitudes being taught are compared (M).

Even before entering a healthcare programme and attending IPE, learners will differ in terms of their personal identity, e.g., gender, age, social background, and aspired profession (Fig. 2). At this time, they may already have a rudimentary profession-specific identity of their aspired profession (Roberts et al., 2018) as well as prejudices towards other professions (Berger-Estilita et al., 2020a, 2020b). This profession-specific identity can be described in terms of its norms, values, attitudes, stereotypes, and behaviors (Friman et al., 2017; Stephens & Ormandy, 2018). As attitudes may already be partly formed during education (Friman et al., 2017), these attitudes will shape how IPE is experienced and also impact IPE’s effectiveness (Berger-Estilita et al., 2020a, b; Lockeman et al., 2017). As Berger-Estilita, Fuchs, et al. (2020b, p. 14) note:

(…) students’ educational backgrounds, as well as attitudes, expectations and stereotypes, may vary considerably between institutions and countries and may influence how the IPE interventions are experienced. This probably accounts for many differences in effectiveness of IPE activities in different settings.

Several studies have shown a differantial impact of IPE on attitudes by profession (Berger-Estilita et al., 2020a, 2020b; King & Violato, 2021; Lockeman et al., 2017; Matulewicz et al., 2020; Snyman & Donald, 2019; Thompson et al., 2020). This may be partly due to individual attitude differences already present prior to training (Snyman & Donald, 2019) that select for and channel prospective students into different professions; and partly due to different professional cultures inculcated from the start of training (Friman et al., 2017). For instance, students of medicine compared to students of other healthcare professions may be more inclined to be attracted by their chosen profession’s prestige and status (Biehle et al., 2019; Friman et al., 2017) and be more individualistic, competitive, and less team-oriented (Friman et al., 2017). Similarly, students of more technically oriented professions are sometimes perceived as having poorer interpersonal skills (Smith et al., 2020). More generally, attitudes and values already present at the beginning of training will likely affect the extent to which interprofessional education can modify them (Lockeman et al., 2017).

In their study of undergraduate medical and allied health students, Snyman and Donald (2019, p. 332) conclude:

The findings suggest that students’ attitudes toward IPCP were influenced by their profession, with some professions showing less enthusiasm for IPCP than others.

Program

CMOC 2: When IPE provides time and facilities for formal and informal interactions (C), it fosters trust, respect, and liking among the learners and helps to reduce professional stereotypes (O) because it allows them to get to know each other professionally and personally (M).

IPE is described as a process of interprofessional socialization (Matulewicz et al., 2020), as it provides resources in terms of time and facilities for learners aspiring to different healthcare professions to come into contact in formal and informal interactions (Berger-Estilita et al., 2020a, 2020b; Friman et al., 2017). Friman et al., (2017, p. 624) described their IPE intervention as follows:

The formal education arrangements also created informal spaces and opportunities to socialise and get to know each other on a personal level.

This allows learners to learn about the skills, competencies, and knowledge other professions have to offer (Friman et al., 2017) and to get to know each other professionally and personally (Berger-Estilita et al., 2020a, Snyman & Donald, 2019). IPE aims to achieve IPC that is egalitarian, respectful (Bloomfield et al., 2021), and appreciative of each healthcare profession’s contributions to patient-centred care by building upon a common curriculum delivered early in a healthcare professional’s training (Berger-Estilita et al., 2020a, 2020b; Biehle et al., 2019; Fusco & Foltz-Ramos, 2018). The interpersonal relations and mutual respect developed help the learners to communicate and function better as a healthcare team (Skolka et al., 2020).

Such interventions allow for exchange of knowledge or skills and sharing of different experiences, which improves understanding and communication between groups, and builds trust (Berger-Estilita et al., 2020a, p. 7).

Getting to know the the other learners means having the opportunity to better understand what each professional’s role in care is (Fusco & Foltz-Ramos, 2018), thereby also supporting professional identity development (Stephens & Ormandy, 2018). At the same time, learning together allows professional stereotypes to be deconstructed (Bloomfield et al., 2021; Matulewicz et al., 2020) and be replaced by actual knowledge of what competencies other professionals offer and their added value in healthcare (Smith et al., 2020).

The social component of IPE was mentioned as a goal and as an advantage. Students considered the networking beneficial, and by engaging on interprofessional relationships on a personal level, they could learn about each other’s curricula in informal settings and even foster friendships. (Berger-Estilita et al., 2020a, p. 15)

Some of the advantages of IPE lie in the knowledge gained of the other professions as well as the social and relational aspects, which allow mutual trust and respect to develop. The literature’s suggestion that personal relationships and friendships have a positive impact on attitudes and that IPE may foster the development of friendships suggests liking the other as an important factor in developing positive attitudes towards IPC and other professionals:

Students suggested that being in relationship with people from other health professions outside of professional context – be it family members, friendships, romantic relationships or shared living spaces – had a valuable influence on their increased knowledge and positive perception of other health professionals. (Snyman & Donald, 2019, p. 331)

CMOC 3: When learners are put in settings where they need to work together to overcome everyday practice problems (C), they develop an interprofessional identity and learn to respect and trust each other (O) because they are able to observe how one profession can help the other, and they come to realize that they are all dependent on each other (M).

Creating situations for students to professionally collaborate on complex patient cases can raise learners’ appreciation for why input from other disciplines is relevant and necessary (Ng et al., 2021; Roberts et al., 2018).

Their experiences at the clinic consolidated an appreciation that collaboration across professional boundaries can amplify any one profession’s capacity to care for patients with complex needs (Ng et al., 2021).

Learning and working together in situations where professionals are interdependent for effective problem-solving may increase respect (Friman et al., 2017; Smith et al., 2020), trust (Berger-Estilita et al., 2020a, b), valuing of collaboration, and result in greater appreciation for other professionals’ skills and contributions to patient care (Ng et al., 2021; Smith et al., 2020; Stephens & Ormandy, 2018). The experienced mutual interdependence promotes the development of a dual identity (Lockeman et al., 2017), that is, a profession-specific identity (Bloomfield et al., 2021; McGregor et al., 2018) and an interprofessional identity (Bloomfield et al., 2021; Lockeman et al., 2017; Matulewicz et al., 2020; Seaman et al., 2018). This is summarized by Lockeman et al., (2017, p. 33) as follows:

Practitioners develop an interprofessional professional identity as a collaborator that complements each individual’s profession-specific professional identity (…). Studies among practitioners have shown that interprofessional education helps to redefine professional identities consistent with the dual identity model (…).

Sharing responsibility for complex patient care may also reduce hierarchy by highlighting how different team members make valuable contributions for optimal patient care delivery (Berger-Estilita et al., 2020a). The collaborative experiences in IPE also results in transformative learning, that is, learning that allows previous beliefs and assumptions of the learner to be reshaped (Muzyk et al., 2019; Ng et al., 2021).

(…) our research showed that the clinical experience in an SRFC helped facilitate a shift in attitudes, knowledge, and comfort working with other healthcare professions.(Ng et al., 2021, p. 707)

CMOC 4: When IPE facilitators serve as role models to students by being experts in their field, reflective of their own practice, and cultivating feelings of equality (C), they contribute to positive attitude development (O) by displaying good behavior to emulate, eliciting positive affect in their learners, and positively shaping interprofessional relationships (M).

As learners are socialized into their roles and develop their professional and interprofessional identity (Friman et al., 2017; Lockeman et al., 2017) through social learning (Lockeman et al., 2017), IPE facilitators serve as role models for learners’ development.

Professionals form identities through a process of socialization (…). The process of socialization is influenced by multiple factors, including (…) role models and mentors, as well as formal teaching with faculty and self-assessment. (…) interactions and experiences can be developed by educators to help shape positive interprofessional relationships (…). (Lockeman et al., 2017, p. 33)

Having facilitators who are competent role models, who facilitate reflections (Berger-Estilita et al., 2020a), and cultivate feelings of equality between facilitator and learner (Stephens & Ormandy, 2018) may elicit positive affect during IPE and support positive attitude development by producing positive interactions and experiences and positively shaping interprofessional relationships (Lockeman et al., 2017). On the value of positive interactions, Stephens and Ormandy (2018) note that:

Students commented on the positive dynamics within the groups (…). Students across groups agreed that they felt a sense of equality between the group facilitators and the group members (…) (p. 352)

(…) the positive group dynamics had a significant effect on the student’s values, attitudes, and beliefs about each other (…). (p. 354)

A similar perspective on the value of facilitators as role models is described by Berger-Estilita et al., (2020a, p. 14):

(…) small-group reflections, facilitated by adequate role models, may allow students to remodel their own professional and personal attitude (…).

Conversely, less competent facilitators negatively influence learner perceptions, resulting in a decline in positive attitudes towards that facilitator’s profession. For example, when medical students were taught by non-physicians they perceived as less competent, they may express a lack of confidence in their IPE facilitator and their IPE instruction (Berger-Estilita et al., 2020a).


CMOC 5: When the IPE curriculum is perceived to be career relevant, boosts confidence, and increases learners’ comfort in working with other professions in patient care delivery (C), it leads to improved attitudes (O) because learners come to expect positive experiences from IP interactions and come to value these interactions more (M).

The attributes of IPE can also contribute to developing positive attitudes among learners, particularly when IPE is perceived as relevant to one’s career (Roberts et al., 2018). The design of the IPE curriculum can contribute to positive group dynamics (Stephens & Ormandy, 2018), leading to feelings of enjoyment and positive affect (Filies & Frantz, 2021; Mowat et al., 2017; Muzyk et al., 2019; Ng et al., 2021; Squires et al., 2021; Thompson et al., 2020). Eploring the experiences of nursing students in IPE, Thompson et al., (2020, p. 5) reported:

(…) they found the group to be very welcoming and respectful, and the session to be very relaxed. The results of this study also suggest that the nursing students became more confident as a result of the teaching; with some indicating that they would be happier to approach a doctor in the future (…).

IPE may contribute to positive affect by increasing professionals’ comfort in working with other professionals (Seaman et al., 2018), boosting confidence, and reducing feelings of inferiority (Thompson et al., 2020). Whereas having only limited interactions with other professionals may lead to feeling intimidated by them (Smith et al., 2020). The importance of curriculum relevance and the need to boost confidence in order to positively affect attitudes is summarized by Roberts et al., (2018, p. 39):

First, in order to maintain or increase positive attitudes towards IPE in introductory programs that span professions, the curriculum needs to be designed to demonstrate relevance to the future careers of participating students from all professions. Second, as IPE may be particularly challenging for students who do not have confidence in their abilities to communicate and work effectively in teams, educators may need to focus on building these skills to decrease negative attitudes.

Whereas high quality IPE may boost positive attitudes, poor quality IPE can lead to boredom, frustration, negative attitudes, and students finding IPE unnecessary (Berger-Estilita et al., 2020a). A further IPE characteristic that can contribute to more positive attitudes is when learners are motivated to improve their own attitudes, such as when they become IPE instructors in their organization, thereby becoming advocates for interprofessional collaboration (Yang et al., 2017, p. 9):

When trying to improve each health professional’s IPC attitude with limited resources, including the time needed to carry out the training, the number of faculty members needed to run the training and the facilities needed for the training, each newly-trained participant should act as a seed instructor within their team. In other words, successful training of seed instructors can result in profession-wide IPC promotion and attitude remodelling.

Post-Program

CMOC 6: When there is organizational support for IPC and healthcare team members participate in IPE on an ongoing basis (C), sustained positive attitudes towards IPC are more likely (O) because the attitudes and values expected in IPC are continually positively reinforced, and are eventually integrated into the learners’ personal value system.

When the goal of developing positive attitudes in learners is achieved, the question of long-term sustainability is important to address post-program. The literature evidence suggests that positive IPC attitudes can be temporary without ongoing organizational efforts to reinforce them (Berger-Estilita et al., 2020a; Mowat et al., 2017; Snyman & Donald, 2019). Negative organizational culture towards IPC can dampen positive IPC attitudes and be passed on to next generations of staff (Friman et al., 2017). As noted by Squires et al. (2021, p. 197):

It is also important to note that even if students are prepared to be collaborative when they graduate, if the organizational culture does not support IPC and operates under traditional hierarchical, non-team friendly models, then the sustainability of graduates implementing the lessons learned about IPC from these programs is threatened.

Maintaining positive attitudes is also supported by continued IPE attendance (Mowat et al., 2017), which is more likely to occur if IPE is experienced as positive (Squires et al., 2021), relevant (Roberts et al., 2018), comfortable and open for attendees from different disciplines (Matulewicz et al., 2020). Sustained attitude change is more likely when attitudes are internalized and correspond to learners’ own values (Stephens & Ormandy, 2018, p. 349):

The first two stages of development (compliance and identification) are types of conformity and can revert to previously held attitudes and values, as they are both extrinsically motivated and require constant reinforcement. However, the third-stage internalisation is when a student embraces the new values and they become part of their belief system (…).

Answering the realist research question

Building upon CMOCs 1–6, we answer the realist question as follows: IPE appears to work by providing formal and informal space for learners to socialize, which fosters trust, respect, and liking among the learners because it allows them to get to know each other professionally and personally. IPE appears to work positively for all learners but will differentially affect learners in different fields of study because they are likely to have different backgrounds and different attitudes and values prior to beginning their studies. Medical students have been noted to be more difficult to positively influence, whereas other professions, notably nursing students, may more easily develop positive IPE and IPC attitudes. When learning conditions and IPE facilitators produce positive affect in their learners and when learners experience that they need each other to solve practice problems, they are more likely to develop positive attitudes towards each other and towards collaboration.

Discussion

This study developed a realist theory of how IPE affects learner attitudes towards IPE, IPC, and members of other healthcare professions than one’s own. This was achieved by conducting a realist synthesis, which identified requisite contextual factors necessary to trigger mechanisms resulting in intended outcomes. The analysis was iterative and built a successively more complex theory of how IPE affects learner attitudes, helping to understand “what works, for whom, how, and in what circumstances” (Pawson et al., 2005, p. 32).

Explaining our findings with middle-range theories

We located several MRTs associated with our CMOCs. We present these associations by phase of IPE (See Table 5). The following sections discuss key MRTs associated with CMOC mechanisms.

Pre-Program

CMOC 1 suggests that attitudes and values already present before beginning university studies may contribute to IPE having less impact on some students, while having more impact on others. This proposition is supported by Social Judgment Theory (Sherif & Hovland, 1961), which suggests that existing attitudes may serve as anchors against which new information is judged. Existing attitudes will impact the probability that new ideas presented in IPE will be accepted or rejected. Especially when an idea being judged is closely tied to a learner’s self-concept and values, the range of ideas that would be considered acceptable becomes narrower and the higher the chances are that a new idea will be rejected (Park et al., 2007).

There are several likely factors related to personal identity that explain why there might be already pre-existing attitudes formed before learners begin their healthcare studies. Some of these factors are presented in the PT figure (Fig. 2), such as gender, age, social background. For example, people develop a preference for specific work activities that are associated with their attitudes and values, and they prefer work environments congruent with their personality (Chen & Simpson, 2015; Holland, 1985). Some disciplines, such as medicine, may be best aligned with a personality that favors hierarchical differentiation. Social and cultural values can also influence individuals’ preference for careers based on status, prestige and social dominance orientation (Pratto et al., 1994; Sidanius & Pratto, 2001). Some research evidence suggests that preference for hierarchical differentiation among disciplines is inversely related to positive attitudes towards IPC (Ballard, 2016). This might explain why medical students who are aspiring to a high-status profession may have more negative attitudes towards interprofessional collaboration compared to students from other professions. In contrast, having a pro-social personality was found to influence job self-selection such that pro-social nurses tended to choose posts with more difficult working conditions (Lagarde & Blaauw, 2014).

Program

CMOC 2 proposes that when IPE provides time and facilities for formal and informal interactions, learners get to know each other professionally and personally, resulting in the intermediate factors mutual trust, respect and liking which are linked to positive attitudes towards IPC. An MRT that supports the importance of regular contact is Allport’s (1954) Contact Theory, also known as Intergroup Contact Theory. This theory suggests that, given the right conditions, contact between groups can affect a reduction in prejudice, conflict, and lead to an improvement in attitudes. Contact is more likely to lead to improved attitudes when there is equal status and respect among group members and when they work towards a common goal (Ballard, 2016; Bridges & Tomkowiak, 2010).

CMOC 3 suggests that when learners are put in settings where they need to work together to overcome problems, they develop an interprofessional identity and learn to respect and trust each other because they are able to observe how they can help each other in patient care and they come to realize that they all depend on each other. This proposition is given support by Social Identity Theory (Tajfel et al., 1979), which suggests that individuals construct their identity, at least in part, from their social group (Ballard, 2016). This can lead to in-group preference, wherein healthcare professionals favor members of their own profession over members of other professions (Biehle et al., 2019). However, through IPE, the development of a dual social identity is facilitated, which is composed of a professional identity and an interprofessional identity. When a social identity is developed as part of the interprofessional group, it is an example of decategorisation as part of the out-group and recategorisation as part of the in-group, as predicted by Social Identity Theory (Tajfel et al., 1979). These processes help in deemphasizing the salience of group distinction (Ballard, 2016). We believe this deemphasis of group distinction may be supported by learners experiencing their mutual interependence among each other to solve problems in everyday practice (Maddock et al., 2023). Deemphasis of group distinction and emphasis of joint group identity as an interprofessional team may also help learners to attribute positive characteristics to members of the other profession and to like them more, as they are now considered members of one’s own interprofessional group.

CMOC 4 recognizes the importance of IPE facilitator role modeling. As proposed by Social Learning Theory (Bandura & Walters, 1977), learning takes place through observing and modeling the attitudes and behaviors of others. Thus, when learners observe good role models who are reflective, cultivate feelings of equality, and contribute to positive interactions within the group, learners are likely to emulate them.

CMOC 5 acknowledges how well-designed IPE curricula must be perceived as relevant to learners’ careers as well as valuable and achievable. According to Expectancy-Value Theory (Wigfield & Eccles, 2000), when tasks such as IPE or IPC are deemed relevant and valuable to learners, and learners expect that they can achieve the objectives, they are more likely to engage in the activity and develop positive attitudes towards it. Indeed, Expectancy-Value Theory formed part of the theoretical foundation for the development of a questionnaire instrument measuring behavioral confidence to undertake interprofessional education activities (Blumenthal et al., 2022).

Post-Program

Dealing with the post-program mechanism of IPE, CMOC 6 proposes that when IPE receives organizational support and is attended on a continuing basis, team members are more apt to sustain their positive attitudes towards IPC. The Affective Domain Development Model developed by Epstein (1977) lends support to this proposition. The model developed with nursing students suggests a three-stage development process of values, attitudes, and behavior. In the first stage of development, learners merely assume or conform to the expected attitudes and behavior. However, in the last stage, the stage of internalization, the new attitudes and values are embraced for their intrinsic value because they have become part of the learners’ value system. This model was recently applied to another study on the development of attitudes among nurses, which used a validated assessment tool to track nurses’ attitude formation over time (Stephens & Ormandy, 2019).

The elaborated links between our developed CMOCs and MRTs provide additional theoretical validation for the underlying proposed mechanisms of IPE in producing positive attitude outcomes.

Implications for practice

The developed CMOCs highlight the manifold social processes embedded in IPE teaching (Hean et al., 2018), which can be shaped by IPE curriculum developers and IPE facilitators. A key finding of our research indicates the importance of trust, respect, and mutual liking in developing positive attitudes. IPE that facilitates formal and informal interactions through group projects, joint clinical placements, and social events are likely to provide such suitable settings. Our finding that IPE needs to be perceived as relevant for one’s career suggests that the alignment of course content with real-world healthcare scenarios is particularly valuable for engendering positive attitudes. Creating IPE experiences where multiple professions have to collaborate to solve complex patient cases and where learners can experience the mutual dependence of the various professions may be especially valuable in emphasizing the relevance of IPC. IPE facilitators need to be mindful that they shape how a course is experienced by learners and how they themselves are perceived. If they are perceived as competent in their roles and are able to elicit positive group interactions and positive affect in their learners, it can contribute to learners having a more positive attitude towards IPE and towards their fellow learners. Finally, adopting a more theory-driven approach towards curriculum development and IPE facilitation may more likely yield positive attitude outcomes, as one can benefit from previous knowledge of what is likely to have the desired effect, under which conditions, and for which learners. Our CMOCs can provide guidance as to which contexts are conducive to triggering the mechanisms that may increase the effectiveness of IPE in influencing sustained positive attitudes.

Strengths and limitations

This rapid realist review synthesizes 12 years of studies on IPE and positive attitude development. Applying realist synthesis, a key strength of this approach was its iterative analysis of data and validation by SMEs. The refined PT and the six CMOCs are testable explanations for what works, for whom, how and in what circumstances to influence positive attitude development towards IPE and IPC. The PT is a placeholder for contextual factors to consider in the progression from one professional identity to a profession-specific and interprofessional dual identity. The CMOCs are hypotheses that provide plausible explanations for what works in the real world. They also can be used for theory-building, considering the number of existing MRTs that may be associated with the mechanisms in our six CMOCs.

Because this was a rapid realist synthesis, we uncovered some important contextual factors, such as gender, age, social/cultural background, aspired profession and personality type. We have included them as factors to consider in the program theory (Fig. 2), but we were unable to delve into all related literature and develop CMOCs for this diversity of contextual factors. A further study limitation lies in the possibility of subjectivity in the interpretation of textual data, although the research team included IPE SMEs and realist methodologists. Finally, given the short period of time for this project, the PT and CMOCs may not capture all potential intermediate causal mechanisms. For example, there may be other mechanisms and theory associated with discipline-specific attitudes and the transition to a dual identity. We believe this paper provides further theoretical advancement of how IPE can result in sustained positive IPC attitudes.

Conclusion

This realist synthesis sets the stage for appreciating contextual factors and associated mechanisms resulting in positive IPE and IPC attitudes. Key mechanisms of positive attitude development include getting to know the other learners professionally and personally, experiencing positive affect during IPE, and learners experiencing mutual dependence. IPE that facilitates formal and informal interactions, is led by competent facilitators, and is perceived as career-relevant may provide the conducive contexts to trigger such mechanisms.