Realist approach to evaluating an interprofessional education program for medical students in clinical practice at a community hospital

Abstract

Introduction: We examined the interrelationships between context, mechanism, and outcome using a realist approach following the introduction of interprofessional education (IPE) to clinical practice for medical students in the community.

Methods: Through participant observation and interviews, a working hypothesis was developed. To evaluate IPE in clinical practice, medical students’ reports were thematically analyzed, and configuration on contexts, mechanisms, and outcomes were identified using a realist approach.

Results: Influential contexts were medical students’ experience of clinical practice and learning characteristics, the capacity of other professionals, interprofessional relationships, and characteristics of the community hospital. One key mechanism was observational learning. Others were self-regulated learning, legitimate peripheral participation, experiential learning, contact hypotheses, awareness of social structure, and cognitive empathy. As faculties supported these key mechanisms, medical students became aware of the legitimacy of community-oriented primary care, noting the roles of physicians who support patients’ and/or their family’s life in collaboration with other professionals, and reflecting the necessity of shifting from physician-centered perspectives. As a result, medical students deepened their empathic understanding for other professionals.

Conclusion: Faculties should develop IPE programs in clinical practice based on the ‘mechanism’, ‘context’, ‘outcome’ pattern and ‘context-mechanism-outcome’ configuration in primary care settings.

Introduction

In clinical practice, interprofessional collaboration is essential for improving the efficiency of healthcare systems and health outcomes. Interprofessional education (IPE) is necessary to the acquisition of the competencies of interprofessional collaboration (Gilbert et al. 2010; Reeves et al. 2013). The World Health Organization recommends that medical students have knowledge of community-oriented medicine and widen their perspective as medical care providers through interprofessional training (Gilbert et al. 2010; Celletti et al. 2011). The Lancet Commission for Health Professionals for the twenty-first Century states that those graduating from medical universities must be able to provide team-based healthcare services that meet society’s needs (Gilbert et al. 2010). Policy makers, implementers and educators are aware of the need for medical cooperation for better health outcomes and healthcare quality and medical safety (Jackson et al. 2013; Institute of Medicine 2015). This need for IPE has led to its introduction to medical education in several countries. In the US, IPE courses and clinical IPE training were provided in 92% and 89% of academic health centers in 2010, respectively (Greer et al. 2014). In Canada, IPE is provided in 12 of 17 medical schools(You et al. 2017). In Japan, a 2016 survey targeting all 81 medical school in Japan (response rate 79.0%) showed that IPE courses were provided in 46, of which 41 were compulsory. Most of these courses were designed for first-year students, and fewer were implemented for higher-grade students (41% for first-year students, 11% for second-year students, 13% for third-year students, 15% for fourth-year students, 9% for fifth-year students, and 3% for sixth-year students) (Maeno et al. 2019). It is preferable for medical students to receive IPE continuously throughout their school years (Reeves 2016). For instance, students’ motivation for collaborative learning reportedly diminishes with time (Hayashi et al. 2012). To enhance readiness for interprofessional collaboration, the clinical encounter approach is favorable (Reeves et al. 2013). Due to fragmented practice, scheduling issues, and a lack of faculties, however, this approach is difficult to implement in university hospitals as a further specialized place where medical students mainly experience clinical practice (Ogawa et al. 2015; Maeno et al. 2019). In comparison, the primary care field is suitable for the implementation of IPE because healthcare services need to be changed from a basis on traditional hierarchical relationships to a basis on interprofessional partnership and community approaches (Adler et al. 2008; Baker and Durham Fowler 2013). Therefore, implementation of IPE for medical students in primary care settings, where most healthcare professionals value interprofessional collaboration based on flat relationships, may dramatically improve the benefits of IPE.

Practice points

• An IPE program in clinical practice at a community hospital was evaluated using a realist approach for complex invention assessment.

• Shadowing other professionals enabled the medical students to compare contexts, understand their community (learning environment) and the community’s comprehensive roles/contributions of other professions, and reflect on their own role.

• Key mechanisms included context learning, self-managed learning, legitimate peripheral participation, empirical learning, contact hypotheses, observational learning, awareness of social structure, and cognitive empathy.

• With support from other professionals, medical students considered the legitimacy of community practice.

A few IPE programs in the primary care field have been provided. In one study, students from various departments took part in interprofessional team projects at community clinics to improve their understanding of various professions (Hosny et al. 2013). In another, medical students, nursing students, medical social work students and pharmacy students participated in clinical IPE training for half a day, once a week for 8 months to learn about the social determinants of health (Suiter et al. 2015). The Centers of Excellence in Primary Care Education (CoEPCE) in the US reported the short-term learning effects of implementing IPE in clinical practice in the primary care field (Harada et al. 2018). Nevertheless, conducting interprofessional collaborative learning and practice in the primary care field is complex and challenging (Brienza et al. 2014). While the establishment of methods to evaluate complex IPE programs in clinical practice is expected to contribute to the diffusion of IPE, few reports have described how to evaluate IPE programs. To provide a more explicit and in-depth understanding of what works, for whom and in what circumstances in the evaluation of complex healthcare interventions, realist evaluation has been recommended for the evaluation of integrated care interventions (Byng et al. 2005; Salter and Kothari 2014).

Here, we aimed to investigate the interrelationships between the conditions of clinical practice, what and how medical students learn, and the outcomes of IPE in clinical practice in the primary care field using the realist approach.

Methods

Study design

The realist approach is designed to answer questions such as “what works, for whom and under what circumstances?” based on a paradigm of realism between positivism and social constructivism (Pawson and Tilley 1997). Using a realist approach, evaluators can better develop complex intervention or education programs and refine middle-range theories based on them. Sequential phases of a realist approach are as follows. First, a working hypothesis or scope of research regarding complex interventions and education programs is established. Second, based on a purposefully collected set of quantitative or qualitative data, the contexts, interventions (including opportunities or resources), mechanisms, and explainable outcomes are detected, verified and refined. This procedure is repeated to refine the middle-range theories based on interrelationships. Relationship between contexts, mechanisms and outcomes (CMOs) is expressed as “contexts + mechanisms = outcomes” (Pawson and Tilley 1997). There are no limitations on CMOs (=interrelationships as working hypothesis) in complex interventions or education programs. CMOs are refined according to a set of data that were collected based on the working hypotheses, and are verified for consistency and integrity (Ogrinc and Batalden 2009). The relationships between CMOs are determined based on a set of collected data; additional data is collected to verify the CMOs; the integrity of the CMOs is refined; and middle-range theories that are conditionally applicable are established. Middle-range theories refer to thoughts that are common in certain conditions (Smith and Liehr 2013; Bolander Laksov et al. 2017). The middle-range theories are explanatory models that fill the gap between empirical research based on small-scale working hypotheses and general theories, which integrate theories and empirical research to explain patterns of social behavior and outcomes in a particular social setting, including limitations on situations and conditions (Merton and Merton 1968). The realist approach enables the establishment of middle-range theories by predicting variations in context related to variation in mechanism under certain conditions and the detection/verification/refinement of variations in related outcomes. Middle-range theories improve understanding of the relationship between the real world and various interpretations of the real world and the characteristics of CMOs.

Medical education curriculum in Japan

In the US, university graduates receive medical education for 4 years. In Japan, high school graduates receive medical education for 6 years, generally comprising liberal arts courses in the first and second years, lectures and training on basic medical concepts including anatomy and physiology in the second and third years, and more advanced medical concepts including internal medicine and surgical medicine in the third and fourth years. Medical students are given the title of student doctor once they pass computer-based testing (CBT) and an objective structured clinical examination (OSCE) before starting clinical practice. In the University in the present study, clinical practice is conducted for 78 weeks between October in the fourth year and June in the sixth year. In the present study, we introduced IPE to a part of the compulsory clinical practice in the department of general medicine in the fifth and sixth years. Fourteen to sixteen medical students participated in the clinical practice every 4 weeks, and received the clinical practice in community hospitals either individually or in groups of two to four members. In this study, we evaluated one program which introduced an IPE at a community hospital.

An IPE program at a community hospital

The community hospital at which the first author developed the IPE program has 30 beds and is located in a city with a population of 70,000 at the center of the prefecture. The community hospital provides primary healthcare services that meet the needs of the community while the hospital has cooperated with an advanced treatment hospital, which is located 10 min away by car, other clinics and welfare institutions. As of April 2017, the community hospital employed four full-time physicians, one senior resident physician, 14 hospital nurses, three home nurses, two physiotherapists, one occupational therapist, one speech therapist, one medical social worker, two pharmacists, and one registered dietician. In March 2014, the hospital initiated a “community-based medical education promoting station” under agreement with the University. Against this background, the first author established an IPE program for medical students in clinical practice at the community hospital. The first and second authors, who are both general physicians, have been working at the community hospital since 2015 and 2014, respectively. The first author developed an interprofessional competency framework and was engaged in IPE pre- and post-graduation in Japan.

Learning goal

We developed three goals for medical students in the IPE program. It was difficult for medical students to become involved in establishing treatments and care planning with other professionals in the short period of 1-2 weeks. Therefore, we utilized the strengths of primary care settings that allow observation of interprofessional partnerships and set three IPE learning goals that focused on the roles of multi-professionals. First, they can understand the roles of other professions. Second, they learn about the role of physician as expected by other professionals. Third, they verbally explain the roles of physicians as the member of a team.

Learning strategy/methods

We adopted observation through shadowing as a learning strategy based on three steps for learners to achieve an effective observation experience (Boud et al. 1985).

1. Preparation before events

2. Active observation

3. Reflection during the observation and afterward

Medical students, through shadowing of other professionals, observed the activities of the nurses, physiotherapists, occupational therapist, speech therapist, registered dieticians and pharmacists at the community hospital, community general support centers and community health centers. Faculty indicated the aims of observation and assessment report format which described the points of interprofessional learning with, from, and about other professions using the following report format:

1. Why did you choose the profession when you describe the learning points?

2. What are the roles of the profession?

3. What does the profession think of patients, their families and community?

4. How does the profession work in the work place?

5. What did you learn from other professionals in clinical practice?

6. How do you think you will work with the profession as a physician in clinical practice after graduation?

Clinical practice schedule

Depending on the rotation of students, clinical practice was conducted for 1 or 2 weeks. The Table 1 below shows the 1-week clinical practice schedule for one student.

Table 1. One-week clinical practice schedule in a course of IPE program in a community hospital.

	Monday	Tuesday	Wednesday	Thursday	Friday
Morning	-Orientation -S: Nurse and rehabilitation therapist in hospital wards	-O: Medical services at a special elderly nursing home	-O: Home rehabilitation services	-O: Outpatient medical services	-Presentation on the student’s assigned patient -S: Pharmacist
Lunch time	-S: Registered dietician
Afternoon	-O: Home medical services	-S: Home nursing services	-S: Medical social worker (MSW)	-Writing report	Reflection on training

The student attended case conferences held for patients with multiple professions when they had time.

S: Shadowing; O: Observation.

Assessment

A summative assessment was conducted based the observational assessments by hospital staff and faculty members. For formative assessment, feedback was provided as necessary throughout the 360-degree assessment from the perspective of many professions and a reflection on the last day of the clinical practice.

Study participants

The study participants were fifth- or sixth-year medical students of the University who were randomly allocated to the community hospital for clinical practice between October 2015 and June 2018.

Realist approach

The realist approach was based on the four-step procedure established by Pawson and Tilley (1997).

Establishment of working hypotheses or theories

Hypotheses and theories regarding complex interventions or education programs can be formulated using deductive methods, inductive methods or mental models of relevant persons (Funnell and Rogers 2011). The first author visited the work places of several of the healthcare professionals who participated in the clinical practice before 2015, and conducted needs assessment and shadowing of healthcare professions. Within the shadowing, the first author also shared the medical students’ readiness for interprofessional experience with healthcare staff in the hospital, and wrote their professional performance down as a field note within a few days after the shadowing. Based on this information, the first author developed the field training programs and established the working hypotheses.

Hypotheses based on CMOs

Contexts indicate the conditions and background in which programs are introduced. It is the complex interventions and education programs are dependent on the context for their proper operation. Mechanisms indicate the processes by which complex interventions or education programs are interpreted. A key to the successful implementation of the IPE is to investigate the effects of mechanisms and mechanism-related changes. First, hypotheses about possible mechanisms must be established. A previous study reported that it is important to differentiate mechanisms from the activities and interventions of programs and resources (Astbury and Leeuw 2010). For example, outcomes are attributable to increases in students’ knowledge and trust rather than to educational interventions. Mechanisms are usually hidden, sensitive to variations in context, and produce effects (Pawson and Tilley 2004). It is important to investigate relationships between interventions and context, as well as relationships between interventions and outcomes. Dalkin et al. classified the mechanisms or assumed mechanisms of programs/interventions/resources and, based on the equation described by Pawson et al., “Contexts + Mechanisms = Outcomes” (Pawson and Tilley 1997), established the following equation: “Contexts + Mechanisms (programs/interventions/resources) → Mechanisms (assumption) = Outcomes” (Dalkin et al. 2015). This equation includes real and assumed mechanisms. Outcomes are classified as those that are and are not expected from programs. Outcomes vary according to the contexts and mechanisms of programs.

Programs introduce various contexts. The mechanisms (i.e., reasoning and reactions of human agents) to be activated and outcomes to be obtained vary according to the contexts. CMOs indicate how various elements of programs are harmonized/integrated. To determine this, we referred to feedback from various professionals and reviewed communications between students and faculties, reports submitted after the end of the clinical practice, and 360-degree assessments from the perspective of many professions.

Observation and verification

Hypotheses and theories were verified based on a set of quantitative or qualitative data on CMOs. Programs were verified on the basis of feedback provided in field notes by various professionals and students that reviewed the fiscal year, students’ reports after the end of clinical practice, and 360-degree assessment from the perspective of many professions. The authors discussed whether or not middle-range theories on CMOs based on the hypotheses could comprehensively explain the results verifying the use of IPE in clinical practice. To evaluate consistency and integrity, the question of whether or not theories and hypotheses were valid for various students and in relationships between CMOs was investigated without paying particular attention to any single result.

Establishment of theories

Through the verification and refinement process, middle-range theories regarding relationships between contexts, mechanisms to explain unknown complex processes, and related outcomes were established (Pawson and Manzano-Santaella 2012). The theories were verified and refined based on sets of data obtained in 2015, 2016 and 2017.

Results

Working hypothesis

Qualitative research was conducted, including ethnography and interviews. Between April 2015 and July 2015, the lead author shadowed and interviewed a home physician for 2 h, a hospital nurse for 2 h, a home nurse for 3 h, a registered dietician for 1 h, a pharmacist for 1 h, an occupational therapist for 3 h, a speech therapist for 2 h, a health center for 1 h, and a community general support center for 1 h, for a total of 16 h.

The ethnography and interviews suggest that these professionals of the community hospital were proud of their professions and were contributing by properly playing the roles of their professions (Table 2). Shadowing programs were developed to allow medical students to observe these professionals in their work environment. With regard to readiness to teach medical students, some staff members hesitated to do so based on their concern that the intelligence of medical students was such that their teaching would not benefit the students.’ In addition, healthcare professionals were afraid of medical students who did not know how to behave in front of patients or their family members and those who were not motivated. However, many staff members were cooperative regarding clinical practice. As an introduction to IPE, free discussions titled “How should you teach medical students?” were held for 30 min in the morning. Various issues were shared including the method of shadowing and the readiness of medical students (including their insufficient medical knowledge and immaturity as members of society). Through these discussions, healthcare staff became convinced that they could teach as professionals and had the opportunity to give essential advice for future doctors with whom they might one day work. Following further discussion between the researchers (JH and YY), the IPE program was developed for medical students. The researchers hypothesized that programs that allow medical students to shadow other professionals at a community hospital as part of their clinical practice will enable them to review their physician-centered thinking and appreciate the importance of interprofessional collaboration.

Table 2. Representative data used to develop the working hypothesis.

Confident professional role

Occupational therapist: “I always try to determine the reasoning behind patients’ movements and activities through communication with the patients. For example, a lady had difficulties walking after developing cerebral infarction. However, she wanted to bring a plateful of offerings to the family altar every morning. For rehabilitation, I suggested that she walk holding an empty plate. I think that occupational therapists have to identify the reasoning behind patients’ activities and help them attain their purpose.” Pharmacist: “Pharmacists’ work is not only to fill prescriptions. I try to learn what patients think of drugs and how they use drugs through ordinary conversations with them. I am enjoying working in the primary care field.” The pharmacist provided an explanation to a patient about the prescribed drugs and asked about the patient’s daily life. Both the pharmacist and patient spoke with smiles on their faces. Nurse: “I think patients look down on nurses. But it’s ok. Our work is to help patients live their lives. I feel I am helping patients when I see they are increasingly able to do more things or are discharged from the hospital.” Despite receiving many calls, the nurse was well-organized and capably cared for each patient.

Readiness to teaching medical students

Speech therapist: “I think that people who become physicians have special brains. They are different from us. I do not know what I can teach them.” Nurse: “I recently noticed that some students did not neatly line up their shoes at the home entrance when they visited patients. I advised the students to arrange their shoes tidily because their behavior was rude to the patients and their families.” Registered dietician: “I notice when students’ interest is low. They are reluctant even to thicken soup, acting like they are being forced to do it.” Nurse: “We have to look after medical students because there is a possibility that we might work with them in the future.” Community general support center: “We want medical students to come to our center and observe the work of care managers and other community healthcare staff.

Table 3. Four patterns of configuration of CMOs.

CMOs	Context	Mechanism (intervention)	Mechanism (assumption)	Outcome
I. Comparisons between the university hospital and community hospital and self-managed learning based on the comparisons	Field training at the university hospital where medical students have fewer chances to meet professionals other than physicians	Shadowing of other professionals at the community hospital and preparation of reports	Learning of contexts due to structural unreasonableness	Awareness of the roles of various professions for life healthcare services
			Self-managed learning	Awareness of physician-centered thinking
II. Legitimate peripheral participation, empirical learning and contact hypotheses based on the learning characteristics of medical students and contexts of the community hospital	Medical students are likely to prioritize efficiency and perform intergroup comparisons (learning characteristics)	Participatory clinical practice and hospital-stay clinical practice, review and reporting	Legitimate peripheral participation	Legitimacy of community-oriented primary care and the roles of other professions
	Support for field training from skilled and knowledgeable professionals		Empirical learning
	Patient-centered information sharing and cooperation based on close interprofessional relationships		Contact hypothesis/observational learning	Increased affinity with various professionals
III. Observational learning among other professionals and awareness of the social structure to deepen understanding of the contexts of observational learning	Roles of other professions at the community hospital, which had limited resources	Shadowing of many professionals	Observational learning	Understanding of the special roles and comprehensive patient-centered roles of each profession
	Environments in which healthcare professionals have greater knowledge about community healthcare services	Field training outside the community hospital	Awareness of the social structure (Freire)	Understanding of the roles of each profession in supporting patients
IV. Learning effects of free discussions with other professionals that increased cognitive empathy in medical students	Preparation and free discussions with other professionals	Other professionals’ expectations of medical students and advice regarding the roles of physicians	Cognitive empathy	Awareness of the roles of physicians and motivations for contribution

Table 4. Representative data extracted CMOs based on the medical students’ reports, various professionals’ 360-degree assessment reports, and field notes.

I. Comparisons between the university hospital and community hospital and self-managed learning based on the comparisons	Report from Student A: “During field training at the university hospital, some physicians instructed the staff members to prepare soupy or paste-like meals for patients who had difficulties swallowing. I tried these meals. The taste and flavor were totally different from ordinary meals. I have to try to remember the reason behind the need for soupy or paste-like meals when I give such instructions.” Report from Student B: “During the clinical practice at the community, I observed comprehensive and holistic medical services that differed from medical services provided at the university hospital. Because I am studying diseases, I am more likely to pay attention to diseases than to patients. This tendency increases when I am busy working. I will never forget the experiences I had during the clinical practice and want to provide patients with holistic medical services.”
II. Legitimate peripheral participation, empirical learning and contact hypotheses based on the learning characteristics of medical students and contexts of the community hospital	Report from Student C: “Efficiency is important to me. I thought the clinical practice would be a waste of time and useless if it was unclear to me what I would learn. For the first three days of the community clinical practice, I thought the training was a waste of time and useless, and I wanted it to stop. However, on the last day, I realized the significance of the training. Upon review, I realized that there was something I could learn from it. I continued the training because the staff had a sense of responsibility as humans rather than as professionals and were gentle and kind to me. If I had discontinued the clinical practice, I would have felt bad. I don’t think you would want to attend clinical practice if you were being ignored.” Report from Student D: “I thought that the occupational therapist and speech therapist were mentally mature and were always thinking of the patients and their families. When the occupational therapist and I visited a patient for rehabilitation, the occupational therapist checked whether or not the patient had bedsores or rashes by referring to the visiting histories and materials of various professionals. This taught me about interprofessional cooperation.” Statement from Student E: “Because the community hospital was small, I had the chance to meet other professionals. I enjoyed having friendly conversations with them at the cafeteria and established close relationships with them: when one staff member referred to “that person,” all the staff members knew who “that person” was.” This student spoke about the atmosphere of the hospital with smile. 360-degree assessment: “He asked many questions about my profession (OT), and he seemed to understand the role of therapists” “Please continue to be a doctor who is humble and collaborates with nurses.”
III. Observational learning among other professionals and awareness of the social structure to deepen understanding of the contexts of observational learning	Report from Student F: “The speech therapist massaged the salivary glands and mimetic muscles of patients with global aphasia and the four limbs of patients with hemispatial neglect. The speech therapist prepared rehabilitation programs from the broad perspective of various professions. During home rehabilitation, the speech therapist tried to resolve every one of the patients’ problems: when a patient with COPD said they felt sick when lying on the bed, the speech therapist suggested that the patient use cushions to adjust the height or to change the position of the cushions.” Report from Student G: “I received training in home healthcare, nursing and rehabilitation services. When I noticed that the medical staff were providing support to the patients as humans rather than as professionals, I vaguely understood the significance of home medical services. Considering the demographics of this region, I think home medical services are necessary in this region.”
IV. Learning effects of free discussions with other professionals that increased cognitive empathy in medical students	Report from Student H: “The clinical practice taught me that I should write as much information on patients as possible and the reasons for selecting drugs in the notes section of a prescription. I want to establish close relationships with other professionals so that they will feel free to ask me any questions.” Report from Student I: “Because the physiotherapist, occupational therapist and speech therapist were always positive about helping patients regain the life they had before, the patients felt free to express their emotions. The physiotherapist, occupational therapist and speech therapist accepted them as they were. Therefore, the physiotherapist, occupational therapist and speech therapist seemed to have knowledge that only they knew. I want to be a physician who can communicate well with other professionals and take advantage of information from other professionals for treatment.” Report from Student J: “Assessment, improvement and maintenance of swallowing function and activities of daily life are essential for treatment during hospital admission. I want to exchange opinions and ideas with occupational therapists, speech therapists and other professionals to determine the most optimal treatment. I want to create an environment/relationship whereby all professionals understand each other and can work smoothly together.” Report from Student K: “It is concerning that other professionals hesitate to speak to physicians.” Report from Student L: “Until this clinical practice, we thought that doctors only had doctors’ specialties; we had a narrow perspective of a doctor. However, in the same way, we thought nurses had nurses’ specialties and rehabilitation therapists had specialties that doctors did not have. If rehabilitation therapists are involved with my patients, I would love to listen to their opinions and refer to their evaluation to provide higher quality medical care.”

Verification and refinement of hypotheses on CMOs

A total of 58 medical students participated in clinical practice in the present study: 19 participated between October 2015 and June 2016; 25 between October 2016 and June 2017; and 14 between October 2017 and June 2018. Two researchers performed data analyses and extracted CMOs based on the medical students’ reports, various professionals’ 360-degree assessment reports, and the field notes. The medical students prepared reports on other professions as follows: 20 reports were prepared in 2015, namely three on medical social workers, three on occupational therapists, five on speech therapists, eight on registered dieticians, and one on nurses; 32 reports were prepared in 2016, namely five on physiotherapists, seven on occupational therapists, eight on speech therapists, 11 on registered dieticians, and one on nurses; and 13 reports were prepared in 2017, namely one on physiotherapists, seven on occupational therapists, two on speech therapists, and three on registered dieticians. These reports were used for theme analyses (Braun and Clarke 2006) to extract CMOs.

The following seven contexts were extracted: (1) medical students’ experience at the university hospital; (2) medical students’ learning characteristics; (3) support for clinical practice from other professionals; (4) close interprofessional relationships between community hospitals; (5) the unique and comprehensive roles/contributions of other professions; (6) the learning environments of the community hospital that help the medical students learn of community needs; and (7) free discussions on the stereotypes held by other professionals of medical students. The following seven educational interventions were extracted: 1) shadowing of other professions; (2) writing reports on the roles of other professions; (3) review; (4) participatory clinical practice; (5) clinical practice involving a hospital stay; (6) field training outside of the community hospital; and (7) other professionals’ expectations of medical students as they develop into physicians and advice regarding the roles of physicians.

The following eight assumed mechanisms were extracted: (1) learning of contexts; (2) self-managed learning; (3) legitimate peripheral participation; (4) empirical learning; (5) contact hypothesis; (6) observational learning; (7) awareness of social structure; and (8) learning process for cognitive empathy.

Patterns of configuration of CMOs (Tables 3 and 4)

Four patterns of configuration of CMOs emerged based on the above process.

1. Comparisons between the university hospital and community hospital and self-managed learning based on the comparisons

2. Legitimate peripheral participation, empirical learning and contact hypotheses based on the learning characteristics of medical students and contexts of the community hospital

3. Observational learning among other professionals and awareness of the social structure to deepen understanding of the contexts of observational learning

4. Learning effects of free discussions with other professionals that increased cognitive empathy in medical students

When these CMOs functioned, the medical students thought clinical practice at the community hospital was legitimate. They felt an affinity with the other professionals and deepened their understanding of the roles of physicians in cooperation with other professionals. They reviewed their physician-centered thinking and became aware of the importance of interprofessional cooperation.

I．Comparisons between the university hospital and community hospital and self-managed learning based on the comparisons

For the medical students, clinical practice at the university hospital was their first experience in clinical practice. They learned about other professions at the community hospital. They understood the differences in clinical practice between the university hospital and community hospital and compared the structure of clinical practice by reviewing the roles of other professions and preparing reports. They also compared the roles in the community hospital with the functionally divided roles in the university hospital and found that the hospital staff who were engaged in patient care at the community hospital had a much better understanding of each other’s roles, despite their different professions. At the university hospital, the physicians were more likely to pay attention to diseases than to patients’ illness and life, and had little knowledge about healthcare services. They thought that the clinical practice at the community hospital was valuable in their journey to becoming physicians.

II. Legitimate peripheral participation, empirical learning and contact hypotheses based on the learning characteristics of medical students and contexts of the community hospital

Some medical students could not understand the significance of shadowing other professionals at the beginning of the clinical practice at the community hospital. However, they observed the work of skilled and knowledgeable hospital staff and thought it was significant to work with healthcare professionals for community healthcare services. Empirical learning such as active participation and reflection by reports and reviews allowed the medical students to develop more interest in other healthcare professionals. They became used to the concept of interprofessional collaboration by observing the close relationships which existed among the professionals in the community.

III. Observational learning among other professionals and awareness of the social structure to deepen understanding of the contexts of observational learning

The medical students observed the roles of other professions they had little knowledge of and learned the unique and comprehensive roles of other professions. Field training outside of the community hospital helped them understand the status of society and themselves: they became aware of the social structure and understood the roles of medical providers in supporting patients.

IV. Learning effects of free discussions with other professionals that increased cognitive empathy in medical students

Many professionals demonstrated their daily work to the medical students and occasionally performed medical interventions with them. This helped the medical students improve their cognitive empathy by understanding medical services from the perspectives of professionals other than physicians. Healthcare professionals showed their expectations of the medical students and advised them on the roles of physicians in the future. This allowed the medical students to learn what other professionals thought of physicians and motivated them to contribute to the roles of physicians.

Discussion

We established an IPE program in clinical practice for medical students at a community hospital that provides primary care services. The IPE program included shadowing other professionals, and reviewing and writing reports on the roles of other professions. The following contexts were identified: comparison of training between the university hospital and the community hospital; medical students’ learning characteristics; cooperation with skilled and knowledgeable professionals; close interprofessional relationships; the expertise and comprehensive roles/contributions of other professions; and learning environments that taught the students about the world outside of the hospital. Certain mechanisms successfully functioned in clinical practice; these mechanisms required the medical students to stay at the community hospital, and included context learning, self-managed learning, legitimate peripheral participation, empirical learning, contact hypothesis, observational learning, awareness of social structure and cognitive empathy. The medical students thought that the clinical practice at the community hospital was legitimate, became aware of their physician-centered thinking and the roles of physicians in communicating with other professionals, and deepened their sympathetic understanding towards other professions.

The medical students became aware of similarities and differences between the clinical practice they received at the university and community. This is rated a learning II (complexity; learning to learn) by Bateson and is attributable to field training in a new and different learning environment (Tosey 2006). Medical students often participate in field training in groups. Japanese medical students are likely to compare their capacity with that of other group members and avoid deviation from group norms (Matsuyama et al. 2018). In the clinical practice, it is likely that a medical student who shadowed another professional observed and reflected on the other professional’s performance based on his/her learning experience without concern about the thoughts of other group members, and thereby deepened his/her understanding of various other professionals in the community. Because medical students in East Asian countries have few chances for self-managed learning (Matsuyama et al. 2016, 2018), shadowing of other professionals on their own may be one way to enhance self-managed learning.

Clinical practice at a community hospital is categorized as community-based medical education (CBME), which is characterized by participation in patient management through family-related contexts and community contexts (Brooks et al. 2018; Schrewe et al. 2018; Rodríguez et al. 2019). From CBME, medical students learn about community-based healthcare services rather than hospital-based healthcare services (Henderson et al. 2018). In our present study, medical students observed other professionals providing healthcare services to patients with disabilities, such as patients following a stroke whose difficulty in physical movement hinders their return to society. In this case, medical students noticed that physicians whose toolkits were limited to medicines were mostly helpless. This active observation and experience improved the legitimacy of the roles of other professions in supporting patients’ daily life. Learning mechanisms based on two CBME-based theories, namely legitimate peripheral participation and empirical learning, affected the medical students’ learning (Kelly et al. 2014).

Contact hypotheses are learning theories that reduce interprofessional psychological distance (Mohaupt et al. 2012), and observational learning is one of the learning methods used to understand the roles of other professions (Fatemeh et al. 2015). The medical students had many chances to contact and shadow other professionals in the community hospital and thereby increased their affinity with these professionals. The medical students were able to understand that the physicians and the various professionals of the community hospital had the same goal. This observation likely helped the medical students learn the value of interprofessional communication. Additionally, flat relationships without hierarchy between the medical students and other professionals helped the medical students become aware of the social structure in the community, as reported by Freire (Paulo Freire 2013), and learn the professionalism of physicians (DasGupta et al. 2006). IPE in clinical practice at a community hospital could help medical students reflect on their identity as physicians. IPE in the community as part of clinical practice may be useful in improving medical students’ understanding of the roles of physicians in interprofessional collaboration. Further investigations are needed to examine the changes in cognitive empathy towards other professionals.

The advantages of IPE in clinical practice at a community hospital include the absence of scheduling issues with non-medical students and the lack of any burden on faculty to prepare courses and instruction. Apart from explaining the goal of observation and shadowing, and review and report preparation on the first day of clinical practice, the program is mostly self-directed with help from the other professionals, with medical students shadowing other professionals and developing close interprofessional relationships and cooperating in free discussions. Additionally, IPE in clinical practice can be conducted by taking advantage of learning environments such as a CBME-based community. Contact hypotheses that shorten psychological distance and learning strategies that increase cognitive empathy to deepen interprofessional understanding are useful for IPE in clinical practice at community hospitals.

A limitation of the present study is that data were collected based on the researchers’ hypotheses and were therefore likely biased, although two researchers analyzed the data to minimize the bias. Further investigations of COMs are needed using other clinical practice environments. Moreover, since the realist approach is a relatively new research method, diffusion of the realist approach as a study method is also needed for complex healthcare interventions.

We used the realist approach to evaluate IPE in clinical practice programs at a community hospital as a complex intervention. The findings obtained in the present study provide suggestions for how to introduce clinical practice in combination with current medical education programs and to evaluate educational program. These approaches are expected to be useful for IPE around the world.

Conclusion

The interrelationship between IPE, mechanisms and outcomes in clinical practice at a community hospital were investigated using the realist approach. Findings from the present study may aid in the introduction, implementation and evaluation of IPE in clinical practice at community hospitals, and contribute to the diffusion of IPE in the medical education field.

Acknowledgments

We thank the healthcare staff in the community hospital for supporting the practical training of medical students.

Disclosure statement

The authors report no conflicts of interest. JH and YY contributed to the design and implementation of the program and the research, to the analysis of the results and to the writing of the manuscript.

Additional information

Funding

This work was supported by JSPS KAKENHI Grant-in-Aid for Young Scientists (B) Grant Number JP19K19377.

Notes on contributors

Junji Haruta

Junji Haruta is a family physician and researcher in primary care, medical education and interprofessional education. He was given PhD degree from Department of International Research Center for Medical Education, the University of Tokyo. In 2017, he acquired the Certificate in Measurement and Assessment in Medical Education in Iowa University. He has worked as associate professor in Department of Primary Care and Medical Education, Faculty of Medicine, University of Tsukuba since April 2019.

Yu Yamamoto

Yu Yamamoto is a family physician with research in interprofessional education. She has been a PhD student in Department of Primary Care and Medical Education, Faculty of Medicine, University of Tsukuba since April 2019.