Determinants of interprofessional collaboration in complementary medicine to develop an educational module “complementary and integrative medicine” for undergraduate students: A mixed-methods study

ABSTRACT

As use of complementary medicine (CM) is increasing, health professionals may require proper training in CM to offer evidence-based advice to their patients. The aim of this study was to explore interprofessional collaboration (IPC) in CM in order to gain a comprehensive overview of determinants and to facilitate the definition of objectives for an undergraduate interprofessional educational module. Pursuant to a sequential mixed-methods approach, focus groups and individual interviews with health professionals, who actively use CM in patient care, were conducted. All hospital employees at the University Hospital Lübeck, Germany, were asked to complete an online questionnaire that was based on the previously obtained qualitative results. The interviews with 20 participants revealed four main themes: profession-specific aspects, communication, health care system-associated factors, and the influence of CM on interprofessional dynamics. An analysis of the 157 responses in the online questionnaire showed that team meetings (n = 124, 79%) were most frequently stated as promoting IPC. In binary regression analyses, nursing (OR = 6.17 [2.02–18.84]) and medical professions (OR = 3.77 [1.27–11.18]) predicted evidence-based care as enabler for IPC. Hence, teaching professional competencies and an equal understanding of evidence-based medicine seems necessary to promote IPC within CM.

KEYWORDS:

• Interprofessional collaboration
• interprofessional education
• complementary and integrative medicine
• professionalism
• evidence-based medicine
• evidence-based nursing

Introduction

Teaching complementary medicine (CM) in an interprofessional setting could promote patient-centered care (Kligler et al., 2015; Rivera et al., 2018; Templeman et al., 2016) and team-based collaboration in the future (Homberg, Klafke, Loukanova et al., 2020; Kutt et al., 2019; Rosenthal et al., 2019). Successful approaches in interprofessional education (IPE) on CM have been described (Brett et al., 2013: Kutt et al., 2019; Rivera et al., 2018; Sierpina & Kreitzer, 2014), and competencies and frameworks in interprofessional education have been established (Academic Collaboration for Integrative Health, 2018; Thistlethwaite et al., 2014).

CM as “a non-mainstream practice is used together with conventional medicine” and includes a variety of health approaches, e.g., natural products, yoga, chiropractic or osteopathic manipulation or homeopathy (National Center for Complementary and Integrative Health, 2018). An increase in the use of CM has been documented (Johansen et al., 2016). Between 40 and 62% of the German population have used at least one CM therapy at one point in their life (Linde et al., 2014). The World Health Organization (World Health Organization [WHO], 2013) recommends adapting curricula in health professionals’ education and training to the regional needs of the population. Thus, the empirically recorded regional differences in the need, use and provision of CM therapies (Peltzer & Pengpid, 2018; Schwarz et al., 2008) justify the need to conduct studies in the region in which a CM curriculum is to be developed.

The six-step approach to developing a curriculum for medical education includes problem identification and general needs assessment as the first step (Kern et al., 1998). Since a theory-practice gap is described when students enter clinical practice settings (Brewer et al., 2017; Levett-Jones & Lathlean, 2009; Newton, 2014; Newton et al., 2009), it seems especially important to include aspects of IPC within the thematic component of the educational module. Incorporating relevant aspects of hospital-based IPC into the learning objectives may prepare students better for future experiences, since their education takes place in and around university hospitals. The role of CM in the hospital setting is scarcely addressed in published literature (Lim et al., 2017; Waddington et al., 2019); oftentimes, it is only addressed from the perspective of individual disciplines, e.g., gynecology and obstetrics (Grimm et al., 2020).

The aim of this study was to explore IPC in the area of CM in a German university hospital in order to gain a comprehensive overview of determinants of IPC within CM and, in turn, to facilitate the definition of goals and objectives for an undergraduate interprofessional education (IPE) module in CM.

Background

Though recommendations on learning objectives in CM are included in the German National Competence-Based Catalog of Learning Objectives for Undergraduate Medical Education, it is the medical schools that are responsible for the implementation of these recommendations. In order to be able to describe and explain therapeutic principles of CM therapies, critically appraise CM therapies, prescribe CM methods adequately, apply indication criteria for and discuss benefits and risks of CM methods (Medizinischer Fakultätentag, 2015), in-depth teaching of CM is necessary.

The medical school of the University of Lübeck offers various degree programs: medicine, nursing, physiotherapy, midwifery sciences, logopedics, and occupational therapy. Offering these degree programs on a single campus provides a promising foundation for IPE in CM to obtain necessary advisory competencies in CM. Based on the results of this study, and a targeted needs assessment realized with a paper-based questionnaire survey among the students of the medical school of the University of Lübeck, we implemented an IPE module. It comprises weekly seminars on various complementary therapies, shadowing of health professionals using CM therapies in patient care and writing a seminar paper on the evidence-base of a specific CM therapy used in the treatment of a disease or symptom. The learning objectives include, amongst others, describing relevant terms in the area of CM, identifying reliable sources for external evidence in CM, using professional and communication competencies to discuss CM with patients and colleagues, and recognizing organizational structures as well as time and economic aspects as determinants of IPC.

Well-functioning IPC among the various healthcare providers strengthens efficiency and coordination in the health system, and leads to better clinical outcomes (World Health Organization, 2010). The Joint Commission (2007) attributes the safety and quality of patient care to teamwork, communication and a collaborative work environment. Since regulatory or financial gatekeepers are missing for CM therapies (Fønnebø et al., 2007), a widespread use of treatments may precede research of efficacy. Thus, patient safety is probably the most important factor when it comes to the application of CM. It has been described as creating risks “through patients failing to seek effective health care, direct harm from intrinsically dangerous preparations or procedures, wastage of healthcare resources on ineffective therapies, promulgation of antiscientific perspectives in medicine and amongst the populace, and exploitation of poorly informed patients and vulnerable persons including children and non-competent adults” (Smith et al., 2016). Hence, training in evidence-based CM is important for all health professionals in order to prevent patients from physical, psychological and financial harm.

Insufficient involvement of CM in interprofessional teams has been ascribed to “limited resources and lack of education about complementary medicine” (Grace, 2015). Nevertheless, patient-centered care would require the inclusion of all health options (Grace, 2015). CM practitioners have shown a desire to offer their services as part of an interprofessional approach (Steel et al., 2020). The fields of integrative medicine – the coordinated combination of conventional medicine and CM (National Center for Complementary and Integrative Health [NCCIH], 2018) – and IPE/IPC have been described as clearly related by bringing together multiple professions to improve patient outcomes (Hollenberg & Bourgeault, 2011; Sierpina & Kreitzer, 2014). Thus, combining both in undergraduate education seems self-evident to address health professionals’ limited knowledge in CM despite its widespread use (Keene et al., 2020).

Method

The present study is based on a sequential exploratory mixed-methods approach. Qualitative focus groups and individual interviews were followed by a quantitative survey based on the results of the obtained qualitative data. While the qualitative data collection aimed at generating a variety of determinants in IPC in the context of CM, the quantitative data collection intended to assess how important these determinants are in the hospital setting. Thus, the set of qualitative data informed the quantitative data set in order to gain a deeper understanding of this complex issue in health care (Halcomb & Hickman, 2015).

Focus groups and individual interviews

The qualitative study design was used to obtain a subjective opinion on IPC and CM from employees of the University Hospital Schleswig-Holstein, Campus Lübeck, as well as teaching practices and practice partners of the University of Lübeck.

The interview guide for the semi-structured interviews was compiled by the research team (PP, JS, KF) in a consensus meeting discussing relevant content areas identified through a selective literature search on IPC, IPE, and CM (Antoni, 2010; Frisby et al., 2015; Klafke et al., 2016; Mette et al., 2016; Schärli et al., 2017; Sorbero et al., 2008). This paper reports on the answers to the following questions:

1. What do you think about collaboration with representatives of other professions in the area of CM?

2. In your opinion, what enables and what hinders IPC (in CM)?

3. What strategies have you developed for yourself to improve your collaboration with other professions?

Participants were recruited by means of purposive sampling, either face-to-face, by e-mail, or by telephone. Inclusion criteria included being a health care professional working at the University Hospital Schleswig-Holstein, Campus Lübeck or at one of the University’s teaching practices or practice partners. To ensure that participants would be able to provide insights into their experiences with IPC in CM, it was required that participants had already actively incorporated CM in their routine patient care, e.g., by performing joint mobilization on a regular basis.

Focus groups were formed with representatives from the various health professions to allow for an open discussion. As an instrument of qualitative research, focus groups facilitate exchange of experience and knowledge, the exploration of contexts, and discussions on the basis of a shared foundation (Bohnsack, 2015; Kühn & Koschel, 2017). Some participants were unable to participate in these focus groups andwere interviewed individually instead. Data source triangulation by combining focus groups and individual interviews has been described as a useful approach for a better understanding of the observed phenomenon (Carter et al., 2014).

All focus groups and individual interviews took place at the Institute of Family Medicine in Lübeck between October 2017 and March 2018. The semistructured focus groups were moderated by a researcher with a medical background (KF, medical doctor; PP, medical student and doctoral candidate). The interviews were conducted by either KF or PP. The research team fully informed all participants regarding background, aims and risks of the study, both verbally and in writing, prior to the beginning of the focus groups and interviews. All participants signed a corresponding consent form. No participant subsequently dropped out of the study. Since the study was part of a process to develop an educational module, it was inevitable that some participants were professionally acquainted with the researchers beforehand.

All focus groups and individual interviews were digitally recorded, pseudonymized, and transcribed verbatim. One researcher took field notes during the interviews to facilitate data analysis and determine theoretical saturation. After reaching saturation, the research team abstained from conducting further interviews.

Survey

A questionnaire was drawn up with the online tool Survey Monkey following completion of the qualitative data analysis. Since no summary of CM methods used or employees who would use CM methods was available, we asked quantitative survey participants about their personal use of or knowledge about CM methods used at the University Hospital Schleswig-Holstein, Campus Lübeck. Based on results from the qualitative research, other questions surveyed the level of agreement with regard to the following statements: “within CM, IPC is desirable”, “teamwork improves the quality of patient care”, “the subject of CM is neglected in vocational training”, “I feel sufficiently informed about CM therapies.” By asking the participants to indicate their level of agreement ranging from 1 “completely agree” to 4 “completely disagree” the survey aimed at assessing the overall opinion on IPC and CM education. Furthermore, participants were asked to describe their general interest in CM on a scale from 0 “no interest at all” to 10 “strongest interest” and their general attitude toward CM on a scale from -5 “very negative” to 5 “very positive.” Another question surveyed the perceived importance of CM therapies for the participant’s profession on a scale from 0 “not important at all” to 10 “absolutely important.” Additionally, participants were given a list of eleven factors derived from the qualitative data analysis and were asked to select all of those factors that would, from their own point of view, promote IPC. They were also given the possibility to add factors. Five sociodemographic questions were also included: year of birth, gender, hospital department in which the participant was currently working (in the following called “discipline”), years of work experience, and profession.

To ensure that all participants had the same understanding of CM, a definition was given in the study invitation as well as in the questionnaire itself: “Complementary medicine supports conventional therapies in the treatment of diseases or complaints, and thus enables a holistic approach in patient care. It does not serve as a substitute for conventional medicine.”

Five representatives of various health professions piloted the questionnaire in July 2017 leading to slight changes in the answer options. The final questionnaire is added as supplemental online material to this manuscript (see Supplementary file 1).

The questionnaire was sent to all employees at the University Hospital Schleswig-Holstein, Campus Lübeck via a mailing list that we compiled ourselves. Since not all e-mail addresses clearly indicated whether the recipient was working at Campus Lübeck or Campus Kiel of the University Hospital Schleswig-Holstein, a question about the place of work was added to the questionnaire.

The official invitation to participate in the survey, which included a link to the online survey and informed the recipients about the aim of exploring CM and IPC for the development of an IPE module on CM, about risks, and about data safety, was sent by e-mail to 6,113 employees on March 19, 2018.

Data analysis

Three researchers (KF, PP, and JS, family physician) analyzed the transcripts according to Mayring’s structuring qualitative content analysis. After deriving deductive categories from the interview guideline, inductive categories were developed from the transcripts (Mayring, 2015). KF and PP independently analyzed all material. In this analysis process, the two resulting coding agendas with all categories and prototypical citations were discussed and adapted by the three researchers in a consensus meeting to generate a consistent coding agenda and realize intersubjective traceability (Mayring, 2015; Schmidt, 2015). During this process JS acted as supervisor and mediator and took a neutral position. Finally, PP repeated the coding process to confirm completeness of the final coding system. Inter-rater reliability was not assessed since the primary aim was not agreement in itself but generating a variety of concepts relevant in IPC within CM (McDonald et al., 2019).

A statistical analysis of the quantitative data was performed using SPSS 25.0 (SPSS Inc., IBM Corporation, Armonk, NY, USA). Descriptive analyses were reported as frequencies for categorical data and as means and standard deviations for continuous data. Since data did not follow normal distribution, non-parametric tests were used for further analyses. Subgroup analyses between CM users and non-users were performed using the chi-squared test for nominal scaled data and the Mann-Whitney U test for ordinal and interval scaled data. Effect sizes ϕ for chi-squared tests and r for Mann-Whitney U tests were calculated. Evaluations were based on Cohen, with ϕ or |r| < 0.3 being a weak effect and ϕ or |r| > 0.5 being a strong effect (Cohen, 1988). Furthermore, binary logistic regression analyses were performed with the binary variables of IPC promoting factors (interprofessional team meetings, setting common team goals, appreciation for other health professions, knowledge about the competencies of other health professions, evidence-based care by the different professions, short, informal communication, flat hierarchies, superiors as role models, interprofessional training of students, colleagues as role models, academization of health professions [0 = no, 1 = yes]) as the outcome variable. Sociodemographic characteristics (gender [0 = male, 1 = female], age, professional experience in years, discipline, profession, CM usage [0 = non-user, 1 = user]) were handled as explanatory variables and entered into the model in a single step. The Hosmer-Lemeshow test was used to evaluate the suitability of the logistic regression model (Hosmer & Lemeshow, 2013). Models were only reported on if the omnibus test was significant and the Hosmer-Lemeshow test did not show statistical significance. Nagelkerke’s R squared explained the variance in the binary models (Field, 2009). For subgroup and regression analyses dummy variables for the professions “physicians” and “nurses” were built in comparison to other professions. Dummy variables for disciplines included “internal medicine”, “surgery”, “anesthetics and intensive care”, “psychiatry, psychotherapy and psychosomatic medicine” and “pediatrics” in comparison to other disciplines. An alpha level of p < .05 was considered statistically significant for all tests.

Ethical considerations

The study was carried out in accordance with the Declaration of Helsinki. The ethics committee of the University of Lübeck approved the execution of the study in July 2017 (Reference number: 17–200). Completion of the survey was voluntary and anonymous. Submission of the completed questionnaire was considered as consent to participate. Interview participants were fully informed – verbally and in writing – about aims and risks as well as the use of data in this study, and signed a consent form beforehand.

Results

Characteristics of qualitative and quantitative study samples

Three focus groups with four to five participants each and seven individual interviews (six telephone interviews and one face-to-face interview) were conducted. The average duration of the focus groups was 52.4 minutes (SD = 3.8; min: 47.52 minutes, max: 55.18 minutes), whereas the individual interviews lasted 12.4 minutes on average (SD = 2.3; min: 9.35 minutes, max: 16.20 minutes).

Out of the 20 interview participants, 13 were female. Their age ranged from 21 to 60 and they had 4 to 35 years of work experience.

The online questionnaire did not reach the intended recipient in 97 cases, which resulted in 6,016 potential participants. In total, 220 surveys were returned resulting in a response rate of 4%. 157 data sets (3%) were valid and could thus be analyzed. The average age of participants was 41 years and the average work experience 16 years. Seventy-six percent of the participants were female. The average rating of general interest in CM was 7.0 on a scale from 0 to 10, the general attitude toward CM was +2.5 on a scale from −5 to +5, and the perceived importance of CM therapies for the participants’ professions was on average rated as 6.2 on a scale from 0 to 10.

Table 1 presents sociodemographic characteristics of the interview and survey participants including the distribution of participating professions.

Table 1. Sociodemographic characteristics of the qualitative and quantitative study sample

Characteristic		Qualitative sample (n = 20)	Quantitative sample (n = 157)
Age in years, mean		49 (SD = 9.5) min: 21, max: 60	41 (SD = 11.0) min: 19, max: 64
Gender, % (n)^a	Female	65 (13)	76 (120)
	Male	35 (7)	23 (36)
Profession, % (n)^a	Nurse	45 (9)	36 (57)
	Physician	30 (6)	29 (46)
	Medical assistant		5 (7)
	Physiotherapist	20 (4)	4 (6)
	Occupational therapist		3 (4)
	Psychotherapist		3 (4)
	Pharmacist	5 (1)	1 (2)
	Psychologist		1 (2)
	Other profession in health care		6 (10)
	Other profession in research, administration and IT		11 (17)
Years of work experience, mean		21 (SD = 7.9) min: 4, max: 35	16 (SD = 11.4) min: 0, max: 42
Discipline, % (n)^a	Internal medicine		15 (23)
	Psychiatry, psychotherapy and psychosomatic medicine		14 (22)
	Surgery		11 (18)
	Anesthetics and intensive care		11 (18)
	Pediatrics		11 (17)
	Others		36 (56)

min. minimum, max. maximum, SD standard deviation

^an varies due to missing data

Focus group and interview results

The qualitative content analysis of the focus groups and interviews showed various barriers and enablers, so-called determinants for IPC. These determinants were grouped into four main themes: profession-specific aspects, communication, health care system-associated factors, and the influence of CM on interprofessional dynamics. Table 2 shows themes and subthemes that resulted from structuring qualitative content analysis with their respective anchor examples.

Table 2. Themes, subthemes and anchor examples identified through structuring qualitative content analysis

Theme	Subthemes	Anchor examples
Profession-specific aspects	Professionalism	“It is also up to me personally to what extent I contribute and what I want to change. If we all just sit back and watch and say, ‚the others can do it’, nothing will ever happen. […] But at the same time, you can also create something big, but only if you actually start doing something” (FP6). “I have experienced that while in training, not only of the medical students […] that there is a lot of interest and curiosity, and also the will to do something and change things. But then they do their first internships in the wards and their enthusiasm is dampened […]: ‘How can I, as an individual person, do anything about it?’” (FP5).
	Collaboration	“The conventional therapeutic professions that we work with so well on a daily basis in the hospital should be […] at one table, so that everyone contributes their part to the care of the patient. There is no such thing as superior or inferior. All are equally important if what we care about is the patient’s well-being and if no on feels the need to be the alpha animal” (IP4).
	(Continuing) education	“[…] so that they realize from the very beginning: ‘Okay, we’re in the same boat because we are working on a common project.’ So that it is already taught from the very beginning, it is something they learn in school” (FP2).
Communication	Structural requirements	“Are there any structures in the system that prompt us to practice communication as well? […] Well, if you look at it that way, I’d say there’s no such thing. That doesn’t happen, because everyone’s just doing their own thing. Barely managing that. We only have few redundancies in this system, thus few opportunities to ask each other: What are you doing exactly?” (FP11). “That means, it is fair to say: One of our special attributes are our daily discussion rounds with all those involved, but you have to practice that too, so that is not a four-hour long discussion about everything, but rather something brief and concise … everyone has to use the same vocabulary and understand it the same way. I think this would lead to a certain feeling of satisfaction. But it also has to be something planned, not something that is the result of a random constellation and then applauded or–the contrary–lamented. It has to be something the leadership wants” (FP5).
	Potential	“When you then get to know each other and really learn to appreciate each other, the timidity fades away” (IP2). “If you don’t want to work against one another, but rather together for the benefit of the patient, I think the method of choice would actually be to talk to each other about the patient” (FP1).
	Limitations	“[…] There is a certain kind of communication but everyone just writes their notes in the patient’s medical chart. And we as physiotherapists often have the problem that our notes are not even looked at. They rather ask: ‘How far along is the patient?’” (FP1).
Health care system-associated factors	Organizational structures	“What I experience here in the psychiatry department is, that there are clearly defined hierarchies, but everybody is working together to care for the patient. […] These flat hierarchies on the wards are very important to me” (IP1). “That is a general shortcoming of our health system that therapies need to be ordered by a physician who oftentimes does not know much detail about the particular therapy” (FP5).
	Economic aspects	“Lack of resources, very clearly, in the hospital. This is a huge problem […] when there is not enough nursing staff, when there are not enough experienced therapists. […] It’s not about the patient’s well-being, it’s just about maximizing profits” (IP4).
	Time	“You don’t really have the time anymore to connect with the people” (FP11). “In my opinion, the factor time plays a very important role. Unfortunately, it is true that physicians can’t be … Well, one also has to stand up for the physicians, right? […] They are also told how much time they have for each patient, how many patients they have to attend to. And it’s normal that this puts pressure on them too” (FP8).
	Demographic change	“At the same time, it has been recognized, that ‘this is one profession’, that we can’t continue like this with our multimorbid patients that get older and older. Physicians cannot manage this alone anymore, therefore, we have to work together. And I would wish that teaching this kind of understanding becomes a more integral part of the training of our young colleagues–physicians as well as physiotherapists, nurses, occupational therapists, everyone that has part in this whole thing. It’ll only work out that way” (FP2).
	Academization	“It is also that … there, this profession has been academized for much longer. We are still working on that here” (FP4). “Due to the fact that all the professions are here on the same campus, that this leads to a new understanding of their fellow human beings. But that also requires from us who work and teach here that we should much more frequently organize courses jointly” (FP6).
Influence of CM on interprofessional dynamics	Status of CM	“I am convinced that in complementary medicine in particular, it is important to appear and act just as professional as in other areas of expertise. Maybe even more so because it is oftentimes underestimated” (FP5). “And I think, naturally, many are skeptical, and increasingly so, I would say because of this bashing in the media, that we have been witnessing lately” (IP7). “In the hospital setting, according to my experiences, patients were scared a lot, they were told: ‘If you decide to go with our chemotherapy, you can’t do anything else. That would reduce the effect’” (FP7). “[…] and that is why I really like the term integrative medicine, because I think, that is … a wide variety of therapeutic approaches that you have. And the larger the better” (IP7).
	(Continuing) Education	“It is therefore worthwhile to at least give an overview of the different types of complementary medicine, what they can and cannot do, how they can be questioned through a second or third opinion, without necessarily directly teaching these procedures. More an orientation about the basics” (FP5). “It is important to ensure good training though” (FP2).
	Evidence	“In a way this was, for me personally, my way of getting into working scientifically. Because this way I use a nomenclature that is immediately understood by physicians, because that is what they are taught, right?” (FP2). “Well, in my opinion, that this is a big issue in complementary medicine, just like it is in your area, in physiotherapy, or in nursing. For us, we are only starting to work scientifically, it’s also still a project for the future” (FP2).

FP: focus group participant, IP: interview participant

Profession-specific aspects

The subtheme professionalism included a discussion about reflecting on self-perception and generating self-confidence from one’s own profession. Well-founded specialist knowledge as well as knowledge about the other professions were found to be necessary. It was said that social competence, looking up to role models, the ability “to clearly represent one’s own opinion” (focus group participant [FP] 5) and being proactive would influence IPC. Moreover, patient-centeredness and the awareness of a common goal and “setting the goal together” (FP2) were emphasized as enablers for IPC.

Under the heading of collaboration, interviewees discussed the importance of “having trust in the team” (interview participant [IP] 4). It was said that it would be crucial to ensure good handling of errors that would allow every team member to make mistakes. Participants described emotional factors such as skepticism and antipathies against colleagues or other professions in general as hindering factors for IPC. Negative experiences in practical settings could reduce previous commitment to IPC. Additionally, it was found that there are certain disciplines such as psychiatry and palliative medicine that have already practiced well-functioning IPC.

Education or continuing education were discussed and described as approaches to convey the competencies of different professions, one’s own contribution to other professions’ ability to work well, and all in all stimulate interprofessional discourse. An early IPE was discussed as being beneficial for collaboration in the future.

Communication

Participants discussed the importance of well-functioning communication. A common vocabulary, interprofessional discussion rounds, electronic health cards, therapy reports, establishing standards for communication in the hospital setting, and defining communication as a quality criterion were described as structures that would allow successful communication. In addition, participants highlighted that proper structures for communication within the health system were missing.

Communication was seen as a good approach to clarify competencies, to allow for the coordination of a collective multimodal concept with the patient, and to reduce risks and fears by getting to know each other. Participants discussed the use of communication for the delivery of information, for the distribution of tasks, and for reflecting on collaboration efforts. Eventually, communication would save time by improving processes, and would allow for an open discussion of difficulties and errors.

While nurses and physiotherapists would discuss patients more frequently, consultations with physicians were described as scarce. Further limitations were a lack of motivation, missing therapy reports, and communication that was oftentimes ineffective and concentrated too much only on facts. Measures in place for quality management and documentation requirements were said to reduce personal and verbal communication.

Health care system-associated factors

Structure-related benefits were said to start with “a corporate that also wants and supports [team work]” (FP2). The traditionally steep hierarchies in hospitals were seen as a barrier and as preventing further development. On the other hand, clearly defined hierarchies were said to also enable IPC. The fact that prescriptions and the management of medication and devices was solely in the hands of physicians was seen as barrier for other health professions to apply their skills. Participants described a chronic shortage of staff and a lack of resources because the health system focused on profit maximization. Interviewees discussed the importance of good time management. It was said that time was not sufficiently compensated and the length of hospital stays hadc been drastically reduced. In times of demographic change, working in interprofessional teams was said to be indispensable in order to care for elderly and chronically ill patients. The academization of the various health professions was discussed as an approach to regulate hierarchies and structure health professions’ education in a more balanced way.

The influence of complementary medicine on interprofessional dynamics

CM showed to have its own influence on interprofessional dynamics. Its special position in the field of medicine was said to lead to a diverse reporting in the media. Thus, it would require a particularly professional appearance for practitioners not to be seen as incompetent. Patients would find themselves in a conflict between conventional medicine and CM, leading to secrecies regarding their use of CM. Moreover, something as simple as the terminology used to describe CM was said to have impacts on IPC. The terms inclusive or integrative medicine were said to enhance collaboration while the term alternative medicine had negative connotations. Regardless of professional affiliation, there were said to be conventional hardliners, health professionals who use CM and, between the two extremes, health professionals who do not personally use CM, but do not object its use. Dogmatism toward CM was said to hinder IPC. Disciplines such as psychiatry and palliative medicine were said to be more open toward CM.

Corresponding to the fact that many patients demanded complementary therapies, participants found that a general knowledge of CM would be important for all health professionals. This was said to reduce the subjectivity of CM and promote an objective approach. Participants expressed a wish for compulsory courses on CM in undergraduate education. Such education in the field of CM could give an orientation about different therapies, enable self-awareness of CM approaches, and focus on adverse effects.

Interview participants discussed that evidence was lacking in CM; the empirical value of complementary therapies was said not to be accepted, especially by physicians. A scientific approach was said to introduce a common nomenclature approximating the health professions. Scientific studies and a stronger evidence base was said to be beneficial to CM.

Survey results

Overall, 130 participants (83%) completely agreed that teamwork improved the quality of patient care, one participant rather disagreed, no one completely disagreed. With the question of whether IPC was desirable in CM, 133 participants (85%) completely agreed or rather agreed. With regardto CM education, 46% of participants (n = 72) completely agreed that CM was neglected in vocational training, 27% (n = 43) rather agreed. In total, 71% (n = 112) of participants rather diasagreed or completely disagreed that they were sufficiently informed about CM therapies. Further details are presented in Figure 1.

Figure 1. Levels of agreement on statements regarding IPC, IPC in CM, and CM education and information

Interprofessional team meetings (n = 124, 79%), the setting of common team goals (n = 121, 77%) and appreciation for other health professions (n = 119, 76%) were most frequently stated as factors that promote IPC. Academization was, by far, the factor least frequently thought of as promoting IPC, 40 participants (25%) indicated this. More details are displayed in Figure 2.

Figure 2. Relative and absolute frequencies of factors chosen by the participants as enablers for interprofessional collaboration

Subgroup analyses revealed that CM users and non-users did not differ in the assessment of factors promoting IPC. However, CM users rated CM therapies as more important for their professions (median = 7, M_Rank = 85.01, U = 2182.50, Z = −2.506, p = .012, r = −0.200) and agreed more often on being sufficiently informed about CM therapies (median = 3, M_Rank = 68.34, U = 2282.0, Z = −2.144, p = .032, r = −0.171) than non-users (median = 5, M_Rank = 67.34 and median = 3, M_Rank = 82.47). Moreover, CM users were more often working in psychiatry, psychotherapy and psychosomatic medicine (20%, p = .041, χ² = 4.166, ϕ = 0.214) and internal medicine (23%, p = .007, χ² = 7.244, ϕ = 0.215) than their colleagues who were not using CM (9% and 8% respectively). There were no statistically significant differences between CM users and non-users regarding professions, age or work experience.

Binary regression analysis models could be calculated for 5 out of 11 enablers for IPC that were included in the questionnaire. Explained variance ranged between 0.189 and 0.274. The nursing profession (OR = 8.83 [1.62–48.03], p = .012) and female gender (OR = 3.41 [1.24–9.41], p = .018) served as predictors for having chosen setting common team goals as a promoting factor for IPC. The nursing profession (OR = 3.92 [1.23–12.52], p = .021), age (OR = 1.18 [1.04–1.35], p = .010) and less professional experience (OR = 0.87 [0.76–0.98], p = .022) were predictors for identifying knowledge about the competencies of other health professions as promoting IPC. Both the nursing (OR = 6.17 [2.02–18.84], p = .001) and the medical profession (OR = 3.77 [1.27–11.18], p = .017) predicted evidence-based care by the different professions as a factor that would enhance IPC. Working in the psychiatry department (OR = 12.10 [2.67–54.86], p = .001) and the medical profession (OR = 3.20 [1.11–9.22], p = .031) were predictors for flat hierarchies as promoting factors for IPC. No sociodemographic variable explained the IPC promoting factor of interprofessional team meetings on a significant level. The CM usage did not predict any of the 5 factors. Table 3 shows the binary regression models in detail.

Table 3. Binary regression models with sociodemographic characteristics as independent variables for IPC promoting factors (N = 139)

	Interprofessional team meetings		Setting common team goals		Knowledge about the competencies of other health professions		Evidence-based care by the different professions		Flat hierarchies
	B (p-value)	OR (95% CI)	B (p-value)	OR (95% CI)	Β (p-value)	OR (95% CI)	Β (p-value)	OR (95% CI)	Β (p-value)	OR (95% CI)
Gender	0.44 (0.402)	1.55 (0.56–4.30)	1.23 (0.018)	3.41 (1.24–9.41)	0.09 (0.860)	1.10 (0.40–2.99)	−0.67 (0.198)	0.51 (0.18–1.42)	−0.23 (0.644)	0.80 (0.30–2.09)
Age	0.08 (0.219)	1.09 (0.95–1.24)	0.02 (0.668)	1.03 (0.92–1.15)	0.17 (0.010)	1.18 (1.04–1.35)	0.10 (0.061)	1.11 (1.00–1.23)	−0.03 (0.428)	0.97 (0.89–1.05)
Work experience	−0.08 (0.247)	0.93 (0.81–1.06)	<0.01 (0.949)	1.00 (0.90–1.13)	−0.15 (0.022)	0.87 (0.76–0.98)	−0.09 (0.094)	0.91 (0.82–1.02)	0.06 (0.154)	1.07 (0.98–1.16)
Surgery	−0.42 (0.534)	0.66 (0.18–2.44)	−0.02 (0.983)	1.02 (0.24–4.28)	−0.86 (0.197)	0.43 (0.12–1.56)	0.43 (0.525)	1.54 (0.41–5.84)	0.77 (0.219)	2.17 (0.63–7.43)
Anesthetics/intensive care	0.96 (0.279)	2.62 (0.61–14.95)	1.22 (0.294)	3.39 (0.35–33.13)	−0.70 (0.303)	0.50 (0.13–1.88)	−0.41 (0.538)	0.66 (0.18–2.45)	0.66 (0.314)	1.94 (0.54–6.99)
Psychiatry	1.86 (0.097)	6.42 (0.72–57.55)	0.52 (0.486)	1.69 (0.39–7.38)	0.48 (0.496)	1.62 (0.41–6.43)	1.22 (0.087)	3.38 (0.84–13.62)	2.49 (0.001)	12.10 (2.67–54.86)
Pediatrics	0.53 (0.492)	1.71 (0.37–7.82)	−0.31 (0.692)	0.74 (0.16–3.37)	1.93 (0.086)	6.87 (0.76–62.08)	0.72 (0.328)	2.06 (0.49–8.74)	0.57 (0.374)	1.77 (0.50–6.26)
Internal medicine	0.82. (0.353)	2.27 (0.40–12.84)	−0.52 (0.502)	0.59 (0.13–2.72)	0.42 (0.538)	1.52 (0.40–5.80)	−0.03 (0.965)	0.97 (0.27–3.48)	0.48 (0.433)	1.62 (0.48–5.45)
Physician	−0.81 (0.186)	0.45 (0.13–1.48)	0.25 (0.667)	1.28 (0.42–3.91)	0.80 (0.171)	2.23 (0.71–7.03)	1.33 (0.017)	3.77 (1.27–11.18)	1.16 (0.031)	3.20 (1.11–9.22)
Nurse	1.30 (0.075)	3.66 (0.88–15.22)	2.18 (0.012)	8.83 (1.62–48.03)	1.37 (0.021)	3.92 (1.23–12.52)	1.82 (0.001)	6.17 (2.02–18.84)	−0.03 (0.953)	0.97 (0.36–2.62)
Use of CM	0,26 (0.602)	1.29 (0.50–3.36)	−0.19 (0.699)	0.83 (0.32–2.16)	−0.48 (0.278)	0.62(0.26–1.47)	0.28 (0.509)	1.32 (0.58–2.98)	−0.47(0.237)	0.62 (0.28–1.37)
Nagelkerke’s R^2	0.224		0.249		0.235		0.274		0.189

B regression coefficient, OR odds ratio, CI confidence interval

Discussion

This mixed-methods study explored determinants of IPC in CM at a German university hospital in the context of a curriculum development for an undergraduate IPE module about CM. It made a distinction between characteristic CM-related factors that influence IPC – status of CM, CM education and evidence base – and general IPC-relevant components such as profession-specific aspects, communication and health system-associated factors.

In the interviews, determinants of IPC were amongst others categorized under profession-specific aspects. Especially the subthemes collaboration and professionalism comprised various determinants. In this context, clear definitions of roles and responsibilities are particularly important to enhance successful IPC and enable efficient patient care (Salas et al., 2005). The professionalism of the individual team members should not be neglected in order to enhance successful collaboration as a team. Because of its complex and multidimensional composition and identifiable deficits (Burr et al., 2016; Cruess et al., 2015; Epstein & Hundert, 2002; Hodges et al., 2011), professionalism can be improved by knowing about expected professional norms (Burr et al., 2016). Key competencies as well as competencies that enable physicians to act professionally have been defined by the Royal College of Physicians and Surgeons of Canada within the CanMEDS framework. According to this, a physician should demonstrate commitment to patients, to society, to the profession, and to physician-health and well-being (Frank et al., 2015). Applying this information to IPE, it is important to foster not only collaboration but individual professionalism, too, or at least consider and observe the individual professional stages the students are in.

Communication as an essential determinant of cooperation (Bainbridge et al., 2010; Körner et al., 2016) turned out to be highly relevant for IPC within CM. Interview participants felt that communication was a skill that needed to be acquired. The integration of communication training in IPE modules might succeed in improving communication skills with the inclusion of validated instruments, e.g., the MAAS-Global (Böhmer et al., 2019; Van Thiel et al., 2003). This reliable and valid instrument rates physicians’ communication and clinical skills, and could be adapted to the interprofessional context.

The majority of participants in the online survey stated that interprofessional team meetings, setting common team goals, and an appreciation for other health professions would be promoting factors for IPC. All of these are aspects of competencies in collaboration, professionalism and communication (Frank et al., 2015). Hence, they can be trained in IPE. A recently published Delphi study among 49 experts from Germany or German-speaking Switzerland regarding role competencies in interprofessional education in complementary and integrative medicine supports the importance of professionalism, collaboration and communication as directly following the medical expertise (Homberg, Klafke, Glassen et al., 2020). Student courses that are able to combine the promoting factors identified in this study in an IPE format have been described: Rivera et al. (2018) reported an interprofessional standardized patient case for prelicensure learners, which comprises the interview of a standardized patient interested in integrative medicine and complementary therapies, and the joint formulation of a care plan by dentistry, physical therapy, medical, nurse practitioner students and social work, nutrition and chaplain trainees. Additionally, Kutt et al. (2019) implemented an IPE course on CM combining didactic lectures, facilitated small groups with team conferences, and case studies. Participating students reported increased comfort collaborating in quantitative evaluation and appreciated the need for collaboration and communication in qualitative evaluation.

Health system-associated factors such as organizational structures, time, and economic aspects are relevant frame conditions that students in IPE courses are not or at least not completely familiar with. Accordingly, IPC enablers included in the questionnaire such as flat hierarchies, superiors and colleagues as role models, and academization cannot be directly influenced by participants even though they may like to. Explicitly addressing these issues in IPE sessions might enhance the understanding of future working conditions and avoid disappointments when transferring from artificially created positive teamwork experiences during IPE to real-life IPC.

The topic “complementary and integrative medicine” was shown to have its own impact on interprofessional dynamics. The perception of CM as a controversial area with partly dogmatic, wildly differing opinions and the supposed lack of evidence were described as barriers to IPC. Binary regression analysis showed that nurses and physicians appraised evidence-based care as an IPC enabler. Therefore, relying on and emphasizing evidence-based care might be an approach to bring together the professions albeit all existing differences. This requires the same understanding of evidence-based care that “integrates the best external evidence with individual clinical expertise and patients’ choice” (Sacket et al., 1996). IPE sessions in CM should incorporate and enhance an equal understanding of evidence-based care. That a curriculum on CM using principles of evidence-based medicine was perceived as beneficial by medical students has already been shown (Forjuoh et al., 2003); a beneficial effect of combining CM and evidence-based care in an interprofessional curriculum can be assumed.

Consistent with the identified themes “health care system-associated factors” and the described “influences of CM on interprofessional dynamics” from the qualitative results of this study, the Academic Collaborative for Integrative Health (Academic Collaboration for Integrative Health, 2018) updated their “Competencies for Optimal Practice in Integrated Environments” – “value and ethics for interprofessional practice”, “roles and responsibilities”, “interprofessional communication”, and “teams and teamwork” – with the addition of the two competencies “evidence informed practice” and “institutional healthcare culture and practice” in 2018 acknowledging that these are especially important for the disciplines represented by the ACIH.

The psychiatry department was emphasized as a location with well-functioning IPC and advanced usage of CM therapies in the interviews. CM users participating in the questionnaire survey were more often found in the psychiatry department than non-users. Furthermore, the psychiatry department stood out in binary regression analysis in the prediction of flat hierarchies as IPC enabler. In fact, internationally, there are several examples of integrated care approaches in psychiatric care – mainly to care for mental illnesses with high lifetime prevalence (Bröcheler et al., 2009; Joseph et al., 2017; Meades, 1989). Interface problems among sectors were reduced by “using evidence-based diagnostic and treatment guidelines and implementing a higher level of evidence-based standards in diagnosis and therapy” (Bröcheler et al., 2009). This long tradition in caring for people in an integrative and integrated manner might have fostered the development of flat hierarchies and emphasis on evidence-based care.

Although sociodemographic characteristics were identified as predictors in binary regression analyses, only a very low explained variance was calculated for the models. Therefore, personal experiences of participants – which are clearly difficult to survey – may have had a greater impact on their assessment of IPC-promoting factors. Since students with a background of working in health or social care settings held relatively negative opinions about interprofessional interaction (Pollard et al., 2004), it seems even more important to integrate positive practice experiences into IPE sessions to minimize the impact of previous or possible future negative experiences.

Strengths and limitations

This study surveyed determinants of IPC in CM in a mixed methods approach, collecting qualitative and quantitative data. This enabled a comprehensive overview of IPC enablers and barriers. Since the aim of using the results for student education was clearly communicated to all participants, a more open report on experiences may be assumed. Within the focus groups and online survey, the composition of participants was interprofessional and quite balanced with regard to professions. In line with WHO recommendations, participants were experts in the sense that they could give insights into IPC in CM in this specific setting, incorporating local health needs of patients and the community (World Health Organization, 2010). Clearly, participants were rather a convenience sample of local clinical experts than experts in IPE or IPC on a national or international level.

Qualitative and quantitative data was analyzed by physicians and a medical student. This prior training and professional understanding might lead to bias in data analysis and interpretation, potentially neglecting other professional perspectives. Introducing a physiotherapist into the research team (MH) as support of all stages of this mixed-methods study enabled a multi-sided approach. Objective quantitative result description apart from interpretation, as well as presenting quotations and interpretations of qualitative data in juxtaposition with each other, shall allow for the judgment of the study’s “trustworthiness” (Graneheim et al., 2017).

When interpreting the results of the questionnaire, the relatively small return rate common for online surveys (Döring & Bortz, 2016) has to be considered. This led to very small groups of participating professions and departments, which clearly influenced regression analyses.

Participants taking part in this study were somehow attracted to the areas of interprofessionalism or CM. This selection bias regarding those topics prevents us from drawing conclusions for the whole hospital staff.

Additionally, the questionnaire survey only addressed health care professionals in the hospital setting. In hospitals, large teams that are formed in an ad-hoc fashion, a traditionally steep hierarchy, ever changing team membership due to shifts and rotations in the work schedule, and team members rarely being in the same place at the same time serve as considerable limitations to functioning collaboration (O’Leary et al., 2012). These hospital-specific unique barriers to IPC might influence views on IPC into a more negative direction.

Furthermore, the questionnaire has not undergone psychometric analysis. Thus, results need to be interpreted with caution.

Conclusion

Determinants of IPC can be assigned to profession-specific aspects, communication, health care system-associated factors, and characteristic influences on IPC by the topic or area. CM showed to have unique impacts on IPC which arise from its controversial position in the medical area. Especially the perceived lack of evidence within CM was considered a barrier for IPC. Addressing CM definitions and specifics as well as principles of evidence-based medicine should be included in IPE in CM not only to overcome known differences among professions but also to foster IPC in CM. Since interprofessional team meetings were most frequently stated as an IPC enabler, training in team meetings in the thematic area of CM is argued as a promising approach to enhance interprofessional competencies in undergraduate students.

Disclosure of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.

Data availability statement

The data that support the findings of this study are available from the corresponding author, KF, upon reasonable request.

Supplementary material

Supplemental data for this article can be accessed on the publisher’s website.

Additional information

Funding

This work was supported by the Karl und Veronica Carstens-Stiftung under Grant KVC 0/082/2/2016.

Notes on contributors

Paula Prill

Paula Prill is a recently graduated physician at the University of Lübeck. She is currently working on finishing her dissertation on “determinants of interprofessional collaboration in the field of complementary medicine” at the Institute of Family Medicine at the University Hospital Schleswig-Holstein, Campus Lübeck.

Jost Steinhäuser

Jost Steinhäuser is the head of the Institute of Family medicine and specialist in general practice. His major fields of research are optimizing postgraduate training including communication, e Health and implementation science.

Minettchen Herchenröder

Minettchen Herchenröder is a physiotherapist and research associate at the Institute of Family Medicine at the University Hospital Schleswig-Holstein, Campus Lübeck. She teaches interprofessional communication, evidence-based practice in physiotherapy and the electives Complementary & Integrative Medicine and Rural & Remote Care. She is currently researching orthopedic insoles.

Kristina Flägel

Kristina Flägel is a physician working as a researcher at the Institute of Family Medicine at the University Hospital Schleswig-Holstein, Campus Lübeck. In addition to the assessment of primary health care delivery, her fields of research involve interprofessional education and collaboration, amongst others in the area of complementary and integrative medicine.