Connecting knowledge: First-year health care students’ learning in early interprofessional tutorials

ABSTRACT Collaboration across professional boundaries is an essential aspect of health care, and interprofessional education (IPE) is a common way to help increase students’ collaborative abilities. Research on how and when IPE should be arranged in a curriculum remains, however, inconclusive. How students actually develop interprofessional competencies have been difficult to demonstrate and is still an under-researched area. Studying IPE in context is therefore important to understand its full complexity. This paper examines how students work with scenarios from professional health care contexts when learning together in interprofessional problem-based learning tutorials during the first year of undergraduate education. The data are video-recorded tutorials of students from medicine, nursing, occupational therapy, speech and language pathology, and physiotherapy programmes. The analysis focuses on students’ discussing their readings of the literature. Drawing on “Communities of Practice,” findings show that students discuss and connect professional knowledge, with “brokers” (the tutors) and “boundary objects” (scenarios) supporting the emergence of students’ professional knowledge. The scenarios, as boundary objects, also enabled the students to turn into brokers themselves. The paper contributes to research on interprofessional learning and offers support for implementing IPE in the early stages of undergraduate education.


Introduction
Learning a profession implies developing a complex set of competences and skills, including conceptual knowledge, generic abilities, professional language, and ways of behaving in a specific context (Hean et al., 2009). In other words, learning occurs through participating in so-called "Communities of Practice" (CoP) (Wenger, 1998). Within healthcare professions, effective working relationships are built on a broad understanding of the fields of expertise of colleagues within related professions. Interprofessional Education (IPE) has therefore become an important aspect of healthcare education, aiming to provide students with the opportunity to "learn about, from, and with each other" so that they will be able to collaborate across professional boundaries in their future careers (World Health Organization [WHO], 2010). Due to the variation of IPEarrangements across universities worldwide, however, researchers have struggled to reach a comprehensive understanding of how to effectively organize IPE (Abu-Rish et al., 2012;Fox et al., 2018). It has also been suggested that IPE research should have a more practice-oriented approach to shift the focus from selfreported opinions to analyses of student practices within interprofessional learning activities. In this way, research could enable an understanding of IPE within its own context (Dahlgren et al., 2012). Furthermore, there has been a call for more theoretically-driven empirical research to help unpack the complexity of IPE (Reeves et al., 2011;Xyrichis, 2020). This paper therefore attends to such concerns through empirical observations of first-year students' as they work together in IPE tutorials using the theoretical lens of Communities of Practice (CoP).

Background
Interprofessional Education and Collaborative Practice (IPECP) has been highlighted as an important route to ensuring patient safety for some time. As IPECP focuses on team collaboration and harnessing each member's knowledge in relation to the patient's needs, it offers a more efficient use of competence and allows healthcare services to optimize resources (Frenk et al., 2010;WHO, 2010). Working in interprofessional teams implies sharing values and ethics with colleagues from different professions, being able to communicate with people who have different conceptual understandings, understanding and respecting different roles and responsibilities, and knowing how to work in teams (Interprofessional Education Collaborative Expert Panel [IPEC], 2016). The aim of IPE is therefore to provide students with opportunities to develop these competences (WHO, 2010). It relates to a sociocultural perspective on learning, with a focus on collaborative learning across different student programs. Some forms of IPE also make use of Problem-based learning (PBL) as pedagogical approach, in which students work with cases to guide their learning, and this further adds situated learning as a theoretical foundation (Dahlgren, 2009).
Despite the noted importance of IPE for developing professional competencies, research in this area remains inconclusive as to how and when, for example, students should participate in IPE (Reeves et al., 2012). In a literature review of IPEactivities, Abu-Rish et al. (2012) report numerous inconsistencies in what is offered across different programs, making it difficult to make comparisons. IPE-activities vary regarding aspects such as course duration, number of participating professions, whether they are singular or recurring activities, and if they are compulsory or not. While this heterogenous sample could offer enormous potential in providing advice on how to implement IPE, there is insufficient information about the learning outcomes of these programmes and concrete conclusions cannot yet be drawn (Abu-Rish et al. (2012). Similar results were noted by Fox et al. (2018) who were unable to provide a "best method to teach IPE teamwork to undergraduate medical students" (p. 133) due to the heterogeneity of the IPE-activities included in their scoping review.
Despite the challenges of drawing clear conclusions in research to date, examining IPE-activities offers considerable potential for understanding how students learn to work collaboratively during interprofessional learning activities (Spaulding et al. (2019). The practicalities of collaborative learning and course design are complex; requiring not only learning with fellow students but also structuring that learning within the confines of existing curricula across one or more programs. Grace (2020) noted, for instance, that most IPE is embedded within existing courses and only a few of the IPE models studied were designed as extra-curricular or partially integrated activities. There is thus a pressing need for more knowledge about the complexities of IPE and how their arrangement within the curriculum might affect students' learning (Reeves et al.'s, 2011). This includes a consideration of where IPE-activities might be located within the early or later stages of undergraduate courses, when students become increasingly knowledgeable and embedded within their own professions. It could be questioned as to whether students would have enough knowledge about their own profession to be able to benefit from IPE, given that a successful interprofessional exchange requires a certain level of expertise (Reeves et al., 2012).
Research which investigates not only the outcomes of IPE but also its practices is thus needed in order to better understand how to integrate IPE into existing curricula. The aim of this study was therefore to examine how students work with scenarios from professional healthcare contexts when learning together in problem-based interprofessional tutorials during the early stages of their undergraduate education. The theoretical lens of CoP is used to inform the analysis, thus providing a more robust contribution to knowledge in this area.

Communities of practice
The concept of CoP (Wenger, 1998), which has its roots in situated learning, proposes that learning can be understood as a collective rather than an individual enterprise. Groups of people with a common interest learn through engaging in shared practices, whether in a formal or informal arrangement. For example, a shared practice can be a study group in which students have chosen to work together or have been allocated to a group by their teacher. The group commits to each other and participates in what is referred to as legitimate peripheral participation (Lave & Wenger, 1991), i.e., activities that successively support the development of the CoP as well as members' individual learning. Over the course of conversations and discussions, the group develops a shared repertoire of discourses and actions that are specific to the group's interests. As a result, there are boundaries between different CoP's, marking the distinction between the types of knowing contained within each CoP. These boundaries do not imply that an area of knowledge is exclusive to a CoP nor that the boundaries are themselves clearly defined. Instead, boundaries are continually being revised and negotiated as new knowledge is gained and shared within the group. As Star (2010) points out, the boundary should be considered as a shared space rather than an edge, with no precise distinction between each CoP. Instead, connections are created across CoPs through participation and reification; crossing boundaries therefore holds unique opportunities for learning. Wenger (1998) explains participation as the processes through which group members interact and connect with one another in the group and reification as the production of documents or objects through which meaning is produced.
Just as is within other domains, students in healthcare engage in multiple CoP's simultaneously as they have several interests, both of a personal and a professional concern. Regarding the latter, no CoP alone holds the complex set of competences and skills referred to as the body of knowledge of a profession. Rather, this body of knowledge is understood as a nexus of knowing and competences, composed of a combination of multiple CoP's closely related to each other (Wenger, 1998). This is of special interest when examining domains of professional knowing, both of their own profession and to some extent that of other professions. Moreover, when working together they are also expected to develop innovative solutions to joint problems. Students must therefore cross and navigate boundaries between their CoP's on a regular basis.
The process of navigating CoPs and connecting knowledge is facilitated by what Wenger (1998) refers to as "brokers:" people who are themselves a member of the CoP as well as other closely connected CoP's. Through their multi-membership they can support fellow members to learn from other CoP's by interconnecting their knowledge and understandings. It is not unusual for a CoP to focus on its own enterprise without considering possible connections with other CoP's, so brokers play an important role in counteracting this tendency through monitoring boundary crossing (Wenger, 1998). In addition to brokers, there are also "boundary objects" which facilitate learning at the boundaries between two or more CoPs. The concept of boundary objects, first introduced by Star and Griesemer (1989), explains how objects, such as documents and concepts, can serve as coordinating mechanisms by both connecting and disconnecting a CoP with the world outside. In this way, boundary objects enable the participation and reification processes noted earlier. Furthermore, by being of relevance to multiple CoP's, boundary objects enable coordination between different interests and support learners in negotiating their knowledge across different CoP's (Wenger, 1998). According to Star (2010), boundary objects are the "stuff of action" that form boundaries.
As IPE seeks to develop students' collaborative competence by learning "about, from and with each other to enable effective collaboration and improve health outcomes." (WHO, 2010, p. 7), CoP thus has the theoretical potential to explain the mechanisms and practical arrangements of IPE.

Research context
The study was conducted at the medical faculty of a Swedish university. First-year students studying medicine, nursing, occupational therapy, physiotherapy, and speech and language pathology participated in a six-week-long IPE module. The IPE module is part of an overarching interprofessional curriculum consisting of three modules, of which the first was videorecorded and used in this study. The other two modules take place in the middle and final stages of the students' undergraduate program. During this first module, students were involved in IPE in parallel with their profession-specific education. More precisely, during four out of six weeks, students were involved in IPE half the time, and profession-specific education for the other half. The remaining two weeks of the module consisted solely of IPE, one week at the start and one at the end of the module. The intention of this arrangement was to support students' learning of professional distinctiveness as well as commonalities across professions. By extension, students should also develop an understanding of how interprofessional competence is intertwined with profession-specific knowledge rather than separate from each other (Lindh Falk et al., 2015).
The students in this IPE module are therefore in the process of becoming part of an educational CoP within their specific medical faculty through legitimate peripheral participation. This is a CoP whose members share an interest in healthcare, who participate in numerous learning activities (both theoretical and practical, profession-specific and interprofessional), and develop a problem-based practice by discussing different learning objectives. In other words, the students can be seen as members of a CoP whose shared practice and interests focus on becoming health care professionals through academic studies. This implies learning both profession-specific and interprofessional knowledge and is thus at the core of the community of professional health care practices. The professional CoP and the academic CoP are differently situated and enacted and, hence, not identical practices, but they are related to each other, as well as linked to practices and CoP:s elsewhere in sharing standards, knowledge and conventions for what is good practice in the profession (Nerland, 2018).
PBL is the pedagogical approach used within this curriculum and in many ways has similar characteristics to CoP. For instance, PBL is based on situated learning theories as well as sociocultural learning, requiring students to work and learn together in small tutorial groups by discussing different scenarios. PBL is especially suitable when doing IPE as it enables students to learn about, with, and from each other while discussing commonalities and differences across each professions' area of expertise (Dahlgren, 2009). In the current study, students participated in both interprofessional tutorial groups and profession-specific tutorial groups. Students were therefore working with different learning objectives simultaneously, moving back and forth between the two throughout the IPE module. The PBL scenarios were designed for parallel use in both IPE and program-specific tutorials to support students' understanding of how interprofessional competence is relevant for profession-specific knowledge. Each week the students were given a new scenario, first used in the interprofessional tutorials, and thereafter in the profession-specific groups. Respective scenarios shared the same basic information about the person in focus, i.e., age, socioeconomic background, and main health issue. Subsequently, details of interest related to the specific learning objectives of the IPE-module and each program were added, serving the possibility of complementing each other. University teachers with different backgrounds (either healthcare or a related profession) and experiences tutored the IPE groups.

Methods
This is an observational study analyzing video-recordings of interprofessional PBL-tutorials. The study design is inspired by Heath et al.'s (2010) descriptions of video analysis, as well as Higginbottom et al.'s (2013) explanations of focused ethnography. Focused ethnography is a relatively pragmatic and effective form of observational study, primarily used in the medical field where the intention is to gain in-depth knowledge on a specific issue or phenomenon to further develop an organization. This also implies that the phenomenon being studied, as well as the specific context and participants, is usually selected in advance, based on a specific problem or research question. This method of conducting ethnography can be described as what Hammersley (2018) calls thin and non-participant. In turn, video offers opportunities to repeatedly and in detail analyze activities as they occur in their normal setting (Heath et al., 2010).
The researchers had prior knowledge about the IPE-module before commencing the study, being well-informed of the concept, design, and learning objectives. This prior knowledge includes the profession-specific studies being conducted parallel to the IPE-module.

Data collection
Data comprises video-recorded PBL-tutorials held over a sixweek period of an IPE module, in which students from different professional programmes worked together on specifically designed cases. We also had access to the scenarios used during the IPE-module. In total, four tutorial groups consisting of 7-9 students (see Table 1) and 1 tutor were video-recorded on seven occasions over the six-week period. Each tutorial lasted between 90-110 minutes, resulting in approximately 50 hours of video-recordings in total. The analysis presented in this paper is based on around 30 hours of the full corpus, during which the students discussed what they had learned from their reading of the literature since the previous tutorial.
The data collection involved setting up small video cameras in each tutorial room, prior to the participants entering the room. 1 Two or three cameras were used in each room to capture different angles and make it possible to observe faces, capture all voices, and observe the interaction between all participants. The use of multiple cameras also allowed for the possibility of technical failures. Once the cameras were in position and switched on, there was no researcher present in the room for the duration of the tutorials. Cameras were switched off and removed only once the tutorial had finished and all participants had vacated the room. All video data was then saved onto password-protected hard-drives and stored according to ethical procedures.

Data analysis
Video recordings were first watched in full, focusing on those sections in which students discuss what they have learned after having conducted their individual searching and reading of the literature. During this viewing stage, observational notes were simultaneously written down, initially concentrating on situations where students discussed their professionspecific knowledge and then related discussions. Analysis was conducted through an iterative process as suggested by Srivastava and Hopwood (2009), continually revisiting the video recordings at all stages of the analytical process. Throughout the analysis we were guided by the questions "What are the data telling me?," "What is it I want to know?" and "What is the dialectical relationship between what the data are telling me and what I want to know?" (Srivastava & Hopwood, 2009, p. 78).
Following this initial analytical stage, video clips were selected by the first author to discuss in data sessions with the other authors. These analytical discussions involved observation of, and reflection on, the content and form of behavior of participants, such as seating arrangements, what was being discussed, who initiated talking, and how the group work was managed. These group data sessions enabled cross-checking of analyses and the comparison of notes about the data. Following these sessions, the first author re-watched the recordings, contrasting and reflecting upon similar situations from the different tutorial groups to find patterns within, and across, the groups. This iterative approach to analysis enabled a datadriven perspective, with the aim serving as a guiding framework. Additional situations of interest to the aim were also noted. Focus remained on those situations in which students discussed profession-specific knowledge, with the addition of what seemed to support them in their discussions. Data extracts were selected on the basis that they highlighted patterns that were observed across the data corpus. Special attention was paid to how documents and technology were being used, how they connected related knowledge and experiences, as well as interactions between students and tutors. Analytical memos (see Hammersley & Atkinson, 2019, Chapter 8) were written down as the analysis progressed and once specific sequences from the video data were selected and transcribed verbatim, we continued analyzing by adding a theoretical framework, CoP, to help us better understand the findings (see Hammond, 2018;Swedberg, 2012).

Ethical considerations
The study was approved by the Regional Ethics Committee in Linköping, Sweden, 2016/439-31. All participants were given written information about the study in advance of recordings and gave their full written, informed consent to be recorded and for sections of the transcript and anonymized images to be used in academic reports. All transcripts and written documents based on the video data have been anonymized and all video files are kept confidential and secure.

Findings
From the analysis, we were able to observe how the tutors' ways of introducing the first tutorial and presenting themselves to the group seemed to create a group norm as to whether and how professional knowledge was discussed, maintained, and recognized by the group when working with a scenario, or whether a more personal focus was applied. We were also able to see how the construction of the scenarios influenced what focus was applied. Our findings show how the discussions varied when the story included a health care unit and/or health issue, or rather featured a personal story. In relation to the theory of CoP our findings indicate that the tutor serves as a broker and the scenarios as boundary objects. In some instances, the students themselves also served as brokers when discussing the scenarios.

Ways of working with a scenario
Throughout the IPE-module, the primary emphasis during the groups' discussions was either a professional or personal focus. A professional focus included discussions concerning objective facts from textbooks and other sources, and knowledge about health status and physical or mental well-being provided in the scenarios. For instance, those groups emphasizing a professional focus discerned aspects such as medicine, treatment options, and the different professionals' area of knowledge when discussing the facts and theories they had learned. By contrast, a personal focus emphasized the individual portrayed in the scenarios as a person. The students discussed knowledge related to the personal story presented in the scenarios, such as the family situation, personal desires, and so on.
Excerpts 1 and 2 portray examples of these patterns when students from groups B and D discussed the scenario called "Hans, 69 years old." This was the fourth scenario for the module and tells a story about Hans being interviewed by a local journalist on his upcoming 70 th birthday. He talks about his upbringing, finding love, and then getting married. He also talks about his experiences of being hospitalized in the 1980's. The excerpts show how both groups conduct a kind of inventory of what the profession-specific part of the Hansscenario contained when they returned to the interprofessional tutorial after having worked with profession-specific learning activities for the previous couple of days. In group B, this conversation takes place just before they begin the tutorial, when most students are in place and waiting for the tutor to arrive. In contrast, group D held their conversation with the tutor at the start of the tutorial session. Both excerpts have been translated into English from the original Swedish. As can be seen in Excerpt 1, the students in group B discern personal information related to Hans, such as ice hockey, how he plays Pokémon-go, has a grandchild, and so on. In group D there is very little of that. Instead, the students list more professionally related facts such as the skin, its structure and how it ages, metabolism, and related processes. On closer inspection, it can also be noted how the two physiotherapists, one in each group, both mention information/facts that suggests how their physio-scenario has been about a situation where Hans had to remove his clothes for an examination. They present this, however, with different emphases. The physiotherapy student in group B talks about it from a personal perspective (to undress in front of strangers), while the physiotherapy student in group D talks about it from a professional perspective (sexual health and how to physically approach a patient). In both groups, these patterns were established in their first tutorial together. Group B applied a personal focus from the start, while group D applied a professional focus. When considering this using the theoretical lens of CoP, these conversations suggest that the students are connecting knowledge from their profession-specific tutorials to their interprofessional tutorials, which can be understood as a form of boundary-crossing between different practices within the more overarching CoP. The students' contributions to the discussion show us how they have picked up on facts relevant for everyone to know together as a group. More in-depth analysis showed how this was supported by the tutors acting as brokers and the scenario acting as boundary objects.

The tutors acting as brokers
The next excerpt illustrates an example of how tutors shed light on what facts are important when discussing professional knowledge. Our findings show that tutors served as rolemodels for the students and thereby supported the students' legitimate peripheral participation in the IPE community. We noted several examples of how students imitated their tutors during the discussions following the introductions in the initial IPE tutorial. Each of the tutors focused on different topicswhether professional experiences or personal interest -and the students often followed suit. In excerpt 3, the tutor for group A tells the students about her career, what responsibilities she has had, and career turns she has taken. A more personal aspect of herself is mentioned in only one sentence. The students then continue and mirror this professional focus when introducing themselves by stating their work experiences and planned moves in their upcoming career. When analyzing group A throughout the rest of the IPE module, they continued to embrace this professional focus and maintain it during their discussions in subsequent tutorials, thus maintaining their shared repertoire. Personal factors were also acknowledged but were not so prominent as the professional focus. For example, the students acknowledged and invited each other's area of expertise when discussing a certain matter, recognizing how different professional knowledge are important in terms of helping patients with different issues. The tutor also sustained a professional focus throughout the module, by commentating on her experience of working with the subject in focus. When students posed questions to the tutor, they focused on professional aspects such as "How do you reason when writing charts?". The tutor thus served as a broker for the students' legitimate peripheral participation, supporting them to connect knowledge within not only the academic CoP they are becoming members of, but also with those who they are likely to engage with in their future careers. A similar pattern occurred in those groups where the tutors focused on personal factors when introducing themselves, such as personal interests and preferences. The students mirrored this in their own discussions, and as we saw in Excerpt 1, group B focused on Hans as a person, on his hobbies and family connections. Another example of a personal focus is seen in group C, where the students organized their discussions by focusing on each member's personal learning question rather than the group's common learning questions. Due to limited discussion time, questions had to be prioritized. One student had a recurring question; "What is each profession's take on this subject?", though this was never prioritized by the group. Instead, they held more personally-oriented discussions triggered by other types of question.
To summarize, tutors served as brokers supporting students to connect knowledge known from their profession-specific tutorial groups to the interprofessional tutorial group as a CoP, through emphasizing a professional or personal focus on the scenario.

The scenarios acting as boundary objects
Findings also demonstrate how the students became brokers themselves during the discussions as the scenarios served as boundary objects. The scenarios enabled the students to coordinate and incorporate their knowledge and experience gained from profession-specific tutorials to the interprofessional discussions. As the scenarios built on each other in both interprofessional and profession-specific tutorials, the students seemed able to connect their respective knowing's.
When analyzing all tutorials throughout the IPE-module, we could see how the story in the different scenarios generated discussions of a varying character. If the scenario included a professional focus, i.e., by including a health-care unit or specific health issue, students discussed knowledge they had learnt in their profession-specific tutorials. This was especially prominent when the health-care unit was central to the story being told. Scenarios which featured a portrait of a person rather than a professional setting, on the other hand, resulted in more general discussions about the students' literature searches and findings. Excerpt 4 below illustrates a more professionally oriented scenario that encouraged students to discuss profession-specific knowledge.

Excerpt 4: from the fifth scenario
After seeing the doctor at the district health care center, she is referred to a multimodal rehabilitation program [MMR]. They say she will not be able to get rid of the pain, instead she must learn to live with it. It felt as a punch in the face at first. Really, learn to live with it? She wants to get rid of it so that she can continue as usual and was expecting an increased dose of pain relief. She tries to convince the doctor to insert more Tramadol (a morphine-like drug with a high risk of dependence) without success.
In group A, the scenario prompted the students to invite, acknowledge, and connect to each other's professional knowledge. During the discussions, the students reasoned about medication and how to deal with patients who want more medicine than professional judgment permits. In the transcript below we can also observe how one of the nursing students connects profession-specific knowledge mentioned by the occupational therapy student (diary) with profession-specific knowledge of her own (self-care theory). And there you could also learn perhaps to, how an everyday life should look like for me to reduce my pain.
NURSE2: It has a little too with Orem's self-care theory, that you have functional disabilities so you won't be able to do the same things as before, and you have to accept that and change to what you can do. And then she has children and a husband to, they could probably have been involved in some way that will make her more motivated, don't you think? Others: Mm IPL6. A.1a, [00:17:05.13] -> [00:18:58.14] In this example, the scenario proves to be a boundary object as the OT-student makes a comment relating to what she has learned as profession-specific knowledge in her OT-tutorial: how OT's work with activity diaries during an MMR-program. It is not explicitly mentioned, however, that it is in fact profession-specific knowledge of an occupational therapist. Regardless, the nursing student picks up on this by relating to what the OT-student had said by adding some professionspecific knowledge of her own, she thus recognizes the similarities in each profession's knowledge. In this way, the two students are connecting knowledge from different tutorial groups and even though it is not explicitly displayed, they consequently navigate boundaries between their respective practices within the larger CoP.

Discussion
As our background demonstrated, IPE is complex in its nature and there is a need to understand this complexity. As Hean et al. (2018), Reeves's et al. (2011) and Xyrichis (2020) point out, research should use theory as a way forward and, as Hammond (2018) argues, as a way to understand data more comprehensively. Hean et al. (2009) discuss different theoretical orientations suitable for interprofessional learning, such as CoP, arguing that the cultural and contextual focus of CoP is a natural way of understanding IPECP. By conceptualizing learning as situated practices, as seen in both CoP and PBL, we can elaborate our understanding of how knowledge is socially contained within the community and the relations between people participating in joint activities. A recent study shows how final year students establish a boundary-crossing CoP when working together in interprofessional teams sharing responsibility for a patient (Jentoft, 2020). The study showed how students negotiated professional knowledge and thereby crossed boundaries, which resulted in different professional actions taken in relation to the patient. Our research contributes to that work by showing not only that students' cross boundaries, but also what supports them to do so. Our aim was to examine how students work with scenarios from professional health care contexts when learning together in problem-based interprofessional tutorials during the early stages of the students' undergraduate education. By studying what students discuss during their tutorial groups we hope to contribute to understandings of how IPE can be arranged in order to ensure that students develop interprofessional competencies as described by (IPEC, 2016). Our observations showed how the discussions were influenced by the tutors, the scenarios, and the students themselves. This is important as it enables educators to consciously plan IPE in ways that enable students to learn about, from, and with each other.
Through the lens of CoP (Wenger, 1998) it can be argued that tutors served as brokers during the IPE-tutorials. By setting an example of what findings from the literature are important to discuss, and maintaining this focus, they support students to discuss and connect professional knowledge between different practices within the larger CoP. The tutors supported the students to connect knowledge gained in one profession-specific educational context to another and how it can be of relevance for students from other programmes. As well as how their respective knowledge and perspective relates to and supplement each other. Considering this from the perspective of interprofessional competencies (IPEC, 2016), tutors support students' process of coming to understand roles and responsibilities. During the discussions, students also learn how to communicate in an interprofessional team since they need to formulate their knowledge and understandings for others.
Furthermore, our findings show how the tutors shed light on both professional and personal aspects about themselves and/or in the scenarios. This resulted in patterns where the groups of students followed suit, focusing on one or the other. It should be clarified, however, that our findings are not to be considered as dichotomous, or that one way is better than the other. The students varied in focus when discerning facts from textbooks and other sources of knowledge, leading us to understand how the tutor can consciously guide students' learning by supporting them to focus on the knowledge that is essential in that context.
Likewise, the scenarios served as boundary objects, setting a focus by which the students were supported to discuss and acknowledge professional knowledge. In relation to what Star (2010) writes about boundary objects and interpretive flexibility, these findings indicate that students act toward and with the scenarios in a way that enable them to learn about, from, and with each other. The scenarios served as a roadmap that the students interpreted differently, thus taking various routes toward the same learning objectives. This is of particular interest as it shows how educators, with careful considerations, can design scenarios that enable students to discover and reflect on different aspects that are of importance when working together in interprofessional teams. This journey of discovery is also aligned with the situated learning perspective used in PBL.
The scenarios also turned the students into brokers themselves, connecting knowledge between their different groups. Relating to Wenger's (1998) way of thinking, the scenarios as a reification of health care seem to be important for enabling participation in interprofessional knowledge exchange. The scenarios enabled them to coordinate their knowledge and experience from related CoP's and bring that into the discussions during the interprofessional tutorials. Our findings are therefore to be considered as a potential contribution to what we have known about boundary objects and how it relates to brokering.
Another interpretation of the findings could be that the students turning into brokers themselves is an indication of the dynamics involved when working with boundary objects without obvious consensus in the group. As we could see in our data, students were connecting different knowledge together. Even though they did not explicitly always found consensus, they kept collaborating and discussing. Considering Star's (2010) input on this, a boundary object seen as a vague common object can be tailored by groups to make it work in a local setting within a CoP. As she also points out, groups can tack back-and-forth between different forms of the boundary object. This reasoning is interesting from an interprofessional perspective as it helps us understand how common knowledge is dealt with when there are also profession-specific nuances to consider.
Analyzing empirical data through the lens of CoP, as suggested by Hean et al. (2009), has therefore offered some insights on how to arrange IPE. Observing video-recordings enabled us to study what happens during IPE-tutorials in their specific contexts. We thereby contribute to collective knowledge on IPE by showing how tutors can use themselves in a way that will influence the students to have a certain focus while discussing a scenario. As well as designing scenarios in a way that will mediate brokering and support the students to connect knowledge known from different scenarios.

Limitations
As with any qualitative study, there are limitations in the extent to which findings are transferable to a different context. We are aware that our findings are related to the specific educational setting and limited to our research questions, and as such, similar studies could arrive at different conclusions. However, the use of CoP as a theoretical perspective allowed us to elevate specific and contextual findings to an broader level. We believe this will enable other researchers and teachers to reflect upon their context and how the concepts of brokering and boundary objects could support their students.
When working with a large set of video recordings and using a qualitative analysis it is impossible to incorporate all aspects observed in the data corpus. Therefore, the findings are necessarily selective and focus on depth rather than breadth. The use of video recordings alone means that the analysis is based on observation rather than an examination of participant perspectives, though this is consistent with the methodological approach taken. Our aim was to provide insight into student practices within IPE and the analysis is therefore limited to addressing how learning can be supported within student discussions in IPE tutorials.

Conclusion
In conclusion, interprofessional education can be arranged in a way that allows students to work with scenarios and to connect knowledge between different practices within an overarching and mutual CoP. Not only can students connect knowledge, they can also serve as brokers when given a boundary object -such as a scenario -through which they can coordinate their professional knowledge. Tutors can also serve as brokers by adding knowledge and experience from their own CoP's. These practices show that it is possible for students to begin assimilating the concept of IPECP as they connect professional knowledge when working together with scenarios. Our findings do not show exact learning outcomes per se, however, nor do they show to what extent the students are aware of profession-specific knowledge being discussed during their discussions. Further research should therefore investigate these areas.

Note
1. For the first tutorial, cameras were only switched on once all the participants were in the room and had signed and returned the consent forms. Thereafter, cameras were already switched on prior to each tutorial meeting.