Interprofessional Education Research: Disciplines, Authorship Practices, Research Design, and Dissemination Trends

Background: In 2007, the World Health Organization launched the Programme on Interprofessional Education and Collaborative Practice to emphasize the importance of interprofessional education (IPE). Since then, numerous IPE reviews have focused on what has been implemented and evaluated. The purpose of this review was to explore how IPE has been implemented and evaluated. Methods: The following aspects of IPE research were explored: 1) disciplines included; 2) authorship patterns, 3) assessment focus, and 4) dissemination trends. Abstracts were screened before a full text screening, review, data extraction, analysis and audit. Frequency and percentages were reported for categorical data while means and standard deviations were reported for continuous data. Chi-squared analyses examined differences between groups for categorical variables. Results: Three-hundred and fty-nine articles met the inclusion criteria. Authors from medical, nursing, and pharmacy schools published most frequently, with a majority using quantitative or mixed-methods techniques. Most studies involved a student discipline without an author from that discipline. Most studies also evaluated student perceptions. Studies were published in 98 journals, most of which were interdisciplinary journals, with the largest increase in publications in the most recent year of the review. Conclusion: IPE research has grown substantially over the past 10 years. Consideration should be given to expanding IPE research methodologies, strategically publishing IPE ndings, and promoting authorship representation for student disciplines involved in IPE.

*"Health Support" refers to supporting professions (e.g. social work, clinical psychology); "Not-health" includes non-health elds (e.g. educational researchers, statisticians). "Other health" included professionals like respiratory therapists. ! Percent based on all studies for that profession (i.e. studies authored) Studies may include individuals from multiple professions, therefore percentages may sum to exceed 100%; "More than 1 Profession" not included in Total *"Health Support" refers to supporting professions (e.g. social work, clinical psychology); "Not-health" includes non-health elds (e.g. educational researchers, statisticians). "Other health" included professionals like respiratory therapists.
! Percent based on all studies for that profession (i.e. studies authored) Studies may include individuals from multiple professions, therefore percentages may sum to exceed 100%; "More than 1 Profession" not included in Total *Some studies included multiple professions, therefore percentages may sum to exceed 100%; "More than 1 Profession" not included in Total ! Percent of studies out of all studies for that profession (i.e. Studies Authored in Table 1) Percent of studies out of all studies with student participants from that profession (i.e. Student Participants in Table 1) Variables of interest for this study included: author discipline, student discipline, research design, assessment focus (e.g., student perceptions, faculty perceptions, and student knowledge), and journal type (e.g., IPE or discipline-speci c journal). We analyzed author order since the rst author in health professions research is commonly considered the main contributor and the last or senior author is generally responsible for providing guidance to the research team, making author position an important factor in the clinical disciplines [16]. Authors were categorized using their credentials (PharmD, MD, etc.), their university a liation, their university biography and/or their Google Scholar or ResearchGate account. In addition, author and student disciplines were classi ed according to the WHO categorizations [1]. "Health support" was used to group professions such as social work or clinical psychology, "other health" included respiratory therapists, while "not-health professionals" was used to group non-health professions such as educational researchers and statisticians. Further, the categories included all possible degree holders within those categories, meaning that medicine, for example, would include individuals who were either Medical Doctors (MD), Doctors of Osteopathic Medicine (DO), or Doctors of Philosophy (Ph.D.) working in medical schools.
Two investigators each extracted data from half of the articles. A third investigator conducted an independent audit on 15% of randomly selected articles, resulting in more than 95% agreement across the variables of interest. We present the results as frequency (percent) for categorical data and mean (standard deviation ± SD) for continuous data. We also used chi-square analyses to examine differences between groups for categorical variables. We considered a p-value < .05 as statistically signi cant.

Results
As seen in Figure 1, 3,058 articles were imported for screening. After we removed the duplicates, 1,518 abstracts were reviewed and 680 articles were advanced to full text screening. Con icts regarding inclusion in the study were resolved by a third reviewer. There were 359 articles that met the inclusion criteria and were included in the review.
of 148 articles), and medical practitioners at 41.57% (n=69 out of 166 articles). Physician assistant authors had the lowest percentage of rst authorship in articles authored by the profession at 7.41% (n=2 out of 27 articles).
Medical practitioners were most frequently last author (n=85 out of 166 articles, 51.20%). Authors from non-health professions and nursing served as last authors 38.74% (n=43 out of 111 articles) and 36.92% (n=72 out of 195 articles) of the time, respectively. Occupational therapy had the lowest presence of last authorship in articles authored by the profession at 22.72% (n=10 out of 44 articles). Chi-square tests reveal a signi cant association between discipline and author position ( rst or last) for the 5 most frequent authorship disciplines (nursing, medicine, pharmacy, health support, not-health), as these categories had a su ciently large sample size for analysis (p=.01). Nursing professionals and pharmacists were more likely to be rst authors while medical practitioners and not-health professionals were more likely to be last authors.
When comparing student and author disciplines, a majority of articles contained a mismatch, when a student discipline was included in a study without an author from that discipline (n=190, 53.92%). Twelve disciplines had more studies involving student participants than studies with authors. Studies including medical students, for example, lacked an author from medicine 31.44% of the time (n=72 out of 229 articles), with pharmacy and nursing following at 25.13% (N=47 out of 187 articles) and 24.60% (n=62 out of 252 articles), respectively. Physiotherapy students were involved in 33.15% of studies, yet physiotherapy authors were only included in 20.89% of studies. In contrast, authors from environmental health and not-health authored more studies than they had students as participants. Not-health professionals, for example, authored 30.92% of articles (n=111 articles) while not-health students were in only 11.70% of articles (n=42 articles).

Discussion
Providing IPE is critical for preparing aspiring healthcare providers for the complexities and realities of team-based care [17]. Research suggests that IPE can enable knowledge and skills necessary for teambased care, and is frequently used to enhance practice models and healthcare services [18]. We have extended previous IPE reviews by exploring the multidisciplinary edge effect and characterizing core research elements of published IPE in an effort to better understand how, and by whom, IPE has been studied and published [9]. Speci cally, we examined the author and student disciplines engaged in IPE, along with research designs and journal characteristics.
IPE and its related research has grown signi cantly since the WHO Programme on Interprofessional Education and Collaborative Practice was launched in 2007 [1]. Historical calls for collaborative practice date back more than 30 years [17,19], yet we have shown substantial increases in IPE research over the past decade. Possible explanations include increased adoption of the term "interprofessional education," increased implementation of IPE across health professions, increased numbers of professions engaged in IPE, growth in the number of health professions schools and associated personnel, and increased numbers of educators engaged in the scholarship of teaching and learning (SoTL) [20].
Most frequently, IPE research involved authors and students from nursing, medicine, and pharmacy.
These are the oldest health professions and the most likely to interact due to proximity and job responsibilities, especially in inpatient medical settings [21]. Lawlis et al. also noted logistical barriers related to scheduling IPE activities and challenges of varying academic calendars across health professions schools [22]. These barriers are less prevalent for nursing, medicine, and pharmacy working in close proximity, which may enable IPE and IPE research between these groups. Engaging additional professions in IPE may require explicit efforts to reduce barriers to interaction between professions and their trainees.
We found the prevalence of mismatches between author and student disciplines uncovered in this review as somewhat surprising. The most common and potentially problematic mismatch occurred when students from a discipline were included in a study without an author from the same discipline. Since health professionals are trained in a tradition that aligns with the responsibilities of that profession [23,24], a study lacking authorship representation could suffer from limited interpretation and subsequent impact. Said differently, IPE research with ghost authorship at the level of a discipline may fail to su ciently evaluate, interpret, and discuss the implications of IPE outcomes for a discipline speci cally or for healthcare broadly. It leaves some question as to whether the IPE activity adequately addressed and discussed the needs of the students from the profession without the contribution of an author from the same profession.
Along the same lines, other professions such as occupational therapy or dentistry may be less involved in this body of research due to fewer number of schools and students, or scarcer requirements for scholarly work as part of academic appointments [25]. Incentives for educators to engage in SoTL and publicly disseminate IPE ndings could be an important strategy for increasing engagement in IPE research [26].
In addition, educators have responded to the WHO's call for more IPE by establishing collaborative committees and providing resources to prioritize interprofessional collaborations [22], which highlights the importance of international and national advocacy for the engagement of various health professions in IPE and IPE research.
Within IPE research collaborations, medical practitioners were more likely to be last author. This may be re ective of medicine's leading role in promoting IPE and related research. Alternatively, it could re ect the perception that the physician is responsible for the health care team and thus assigned the senior author position [27]. This is a well-documented challenge in dissemination, also called gift authorship [16,28]. If this is the case, to improve transparency and integrity in IPE, authorship should be established based on contribution to the study from beginning to end, addressing authorship issues at the outset of the study to ensure all involved are clear. Authors should also explicate their roles within articles so readers are able to discern the contributions of each individual to the IPE initiative and its related research, as currently required by some (yet not all) journals [16].
Further, a majority of studies conducted were either mixed-methods or quantitative designs. Although there are bene ts to each type of design, there is an opportunity to expand the use of qualitative designs in IPE research, as this type of study design may be better positioned to elucidate the human experience. In addition, almost all studies included measures of student perceptions with some also studying faculty perceptions. While it is important to understand how students feel and their attitudes towards IPE, there is a clear lack of focus on IPE effectiveness. This is especially important in light of the multidisciplinary edge effect, as students need to understand both their own discipline and the traditions of others to optimize the healthcare system. Without assessing the effectiveness of IPE, we may be unable to capitalize on the generative potential of working with individuals from other disciplines, which is a key tenant necessary for the multidisciplinary edge effect to occur.
Moreover, where scholars submit and publish their research can in uence its visibility and reach. Approximately half of IPE articles in this review were disseminated through interdisciplinary journals, suggesting that individuals who prioritize reading discipline-speci c journals may miss ndings from relevant IPE initiatives. This is especially important for professionals in medicine, nursing, and pharmacy, as less than 15% of IPE articles were published in each of these individual disciplines' eld-speci c journals even though these disciplines produced over 50% of the IPE research. In contrast, IPE journals build capacity for publishing IPE ndings and expand opportunities for IPE scholars to disseminate ndings. Strategies that ensure IPE research, from both discipline-speci c and IPE journals, is visible to the relevant professions could be an important step towards advancing IPE initiatives.
We found the largest growth in IPE research was in the most recent year of the study, suggesting that IPE research is continuing to expand. This review gives way to numerous questions that could help elucidate IPE further as this body of research grows and as similar reviews are conducted across more recent years. For example, how is authorship and authorship order determined in IPE research, and why might some professions be less commonly represented as rst or last authors? Since different disciplines have varying standards and expectations for authorship order [16], the interdisciplinary nature of IPE research offers unique authorship challenges -and some might suggest that who leads analysis and writing could be associated with who needs the manuscript for promotion, rather than those driving or contributing to the IPE in certain ways [29]. Second: why, and in what cases, are students from a discipline engaged in a study without an author from that discipline Fifth: to what extent are IPE studies reaching appropriate stakeholders in light of current publication practices in IPE?

Limitations
As with any systematic review, this study has several limitations. First, the databases used in this search were domain speci c. While it is possible the review missed some relevant articles, diverse databases from education and health professions were used to minimize this risk. Second, the review focused on work using the phrase "interprofessional education," which may have excluded articles with related terms. Third, publication bias may limit which IPE initiatives were published, as studies with statistically signi cant results are more likely to be published and cited within the literature [30]. While the limitations associated with publication bias could not be addressed in this study, the authors advocate for more transparency in IPE, including the dissemination of research resulting in negative or non-signi cant ndings. Finally, this study did not analyze the nature of the studies that were reported, speci cally, whether they were program evaluation studies or student self-reports on evaluations.
Despite these limitations, this review provides insight into the IPE research to-date, highlights potential opportunities for future research practices, and informs the development of an IPE research agenda for health professions education. Further research must be conducted to better understand the impact and utility of various research practices in IPE, how IPE research is incentivized and operationalized (e.g. authorship), and how these practices in uence the extent to which we understand the impact of IPE on learners. Promoting the use, evaluation, and dissemination of IPE initiatives, and improving transparency in IPE research, is imperative for advancing collaborative care models within a rapidly evolving healthcare system.

Conclusion
IPE research can elucidate strategies for promoting the skills students must develop for success within a rapidly evolving and highly collaborative healthcare system. Since the WHO declared IPE a priority in 2007, health professions have increased their engagement in IPE research. However, variation between how differing professions approach IPE research, apparent mismatches in author and student disciplines, and varied publication practices raise some questions about IPE research practice. In light of this study, health professions must pursue IPE research that is inclusive and accommodating of scholars from various professions, fosters collaboration between traditional and emerging health professions, and positions students for success within a collaborative healthcare environment. Consent for publication -Not applicable.
Competing interests -The authors declare that they have no competing interests.
Funding -There was no funding for this study.
Authors' contributions -AO collected all data, lead the analysis, and drafted the manuscript. CLS contributed to data analysis and critical review and revision of the manuscript. GBD contributed to critical review and revision of the manuscript. JEM oversaw all aspects of the study and provided critical review and revision of the manuscript. All authors read and approved the nal manuscript.