Technology-Enabled Visualization of Team Typologies at a Multi-Institutional IPE Event

Abstract

Background: Preventable medical errors caused by ineffective teamwork are substantial contributors in the majority of patient harm events. Nonetheless, the interprofessional education (IPE) research to date has not reported on within-group interactions and discourse to determine if students in IPE teams are practicing effective teamwork at IPE events. Purpose: The overall objective of this mixed methods study was to develop IPE team typologies based on a multi-institutional IPE event in order to characterize and provide actionable knowledge for improving IPE teamwork. Methods: A total of 127 students and 18 faculty, representing eight pre-professional programs and three universities, participated in this study. The IPE teams were videotaped during their case-solving work. These recorded sessions were analyzed using a mixed methods design that included event-based scoring for cognitive level and team behaviors and development of IPE team typologies using a constant comparative analysis (open, axial, and selective coding) of 14 IPE teams during the process of developing care plans for novel patient cases. Team typologies were cross-validated with the current literature. Discussion: Four IPE team typologies emerged: Facilitated, Cohesion, Consensual Validation, and Silo Mentality (listed in rank order from most to least effective). Only the Facilitated team type demonstrated effective teamworking behaviors. Decreasing team effectiveness was met with a dose-dependent, concomitant decrease in average cognitive level and beneficial team behaviors. Conclusions: The results of this study provide the knowledge required to implement recommendations for targeted interventions to improve IPE teamwork. Effective teamwork is crucial to optimal patient care; therefore, future research should critically analyze and seek to improve IPE teamwork in order to prepare a practice-ready next generation of healthcare professionals.

1. Introduction

This paper presents the findings of research about a multi-institutional interprofessional education (IPE) event. The researchers employed constant comparative analytical methodologies with micro-level coding to elucidate IPE team dynamics and develop team typologies used by interprofessional teams of students working to solve patient case scenarios [1,2]. The authors also present event-based scoring for cognitive level and TeamSTEPPS behavioral markers during IPE teamwork. The study was conceived based on three observations about the IPE event. Namely, (1) there was an unknown baseline level of knowledge regarding effective teamwork for the student participants, who were enrolled in eight pre-professional programs at three universities, (2) teamwork training was not provided as part of the IPE event, and (3) the patient case scenarios used in the IPE event did not appear to require metacognitive team thinking in order to solve them. The researchers reasoned that the best way to determine what the teamworking dynamics actually were during these student team deliberations would be to record the sessions and complete content analyses on the video-taped sessions. Additionally, the researchers aimed to provide actionable knowledge for how to improve IPE teamwork. This research is important because teamwork failures have been identified as substantial contributors in the majority of patient harm events, making teamwork failures a major source of preventable medical errors [3]. The current research adds to IPE research that has been conducted to date by elucidating the inner workings of IPE teams during the process of solving patient cases [4,5].

Team Training for Healthcare Professions

Teamwork is an essential component of good healthcare. Patients are more satisfied when they engage with providers who understand and express their interpersonal competencies, and development of interpersonal skills is a goal for teamwork training [6]. In a 2018 study by Anandarajah and Quill, teamwork was found to be the second-most important factor for physicians in preventing burnout and sustaining resilience, with only patient contact itself being more important [7]. Reduced burnout is also associated with lower healthcare cost. The importance of teamwork in healthcare clearly spans both sides of the care equation [6].

Fortunately, several models for teamwork training are available [8]. One such model, TeamSTEPPS, has been developed specifically to meet the needs to universally improve teamwork in healthcare and healthcare education [9,10]. TeamSTEPPS addresses many characteristics and skills deemed as requisite for effective teamwork [11,12] (leadership, shared mental models, communication, team metacognition) as well as some that are distinctively relevant to healthcare such as situation monitoring, mutual support, and team structure. TeamSTEPPS was released in November 2006 as an option for effective team training in healthcare and has since been used extensively in this setting [9,10]. More recently, TeamSTEPPS is being utilized in academia-based IPE training [10,13,14,15].

2. Methods

This study was approved by the Institutional Review Boards at the University of Dayton and Wright State University in Dayton, OH, USA and Cedarville University, in Cedarville, OH, USA. All student participants provided written informed consent to participate.

2.1. Participants and the IPE Event

Study participants were recruited by their respective program directors (medicine, pharmacy, nursing, physical therapy, physician assistant, dietetics, social work, and clinical mental health counseling) at three universities. All participants (n = 127) were in good standing in their respective programs.

The care collaboration event was a multi-institutional experience that allowed students from eight different pre-professional programs to collaborate in developing a care plan for novel patient cases. The event provided programmatic evidence of interprofessional education (IPE) for accrediting agencies of pre-professional programs.

Prior to the event, participating faculty members (n = 18) developed three patient cases from adapted material [16] or from actual patient care scenarios. On the day of the event, students were stratified into disciplines and then randomly assigned to one of 14 IPE teams in order to have an even distribution of pre-professional programs across the groups. Each group was randomly assigned one of the three patient case scenarios to discuss and develop care plans for. Before the students dispersed into their assigned IPE groups, all students and faculty congregated in a large room and were introduced via a brief slide show to the Interprofessional Education Collaborative (IPEC) competencies and provided instructions for how the event would be run [17]. The IPE team sessions lasted 90 min, followed by group presentations to panels of faculty experts for critique.

2.2. Study Design and Data Analysis

In the current mixed methods study, fourteen IPE teams were populated by 127 students from eight pre-professional programs. All fourteen IPE teams were videotaped for 90 min while they worked to develop a care plan for their novel patient case. Each IPE team met in a private room, which minimized contamination by researchers and background noise or other interference in the videos. Constant comparative analysis of the fourteen IPE team videos yielded four team typologies and a single video was selected from each of the four team typologies to be transcribed and analyzed for cognitive level (Bloom’s Taxonomy) and team behaviors (TeamSTEPPS) using event-based scoring.

Data triangulation was used to provide as much information as possible regarding IPE team typologies [4,18,19]. Quantitative data for cognitive level and team behaviors were triangulated with qualitative data from the constant comparative analysis of the IPE team videos and then validated with team typologies identified in the current literature [1,2]. Findings from the statistical analyses of quantitative data were used to determine if the mean differences in cognitive level between groups and between pre-professional programs were statistically significant. The data were combined into graphs and diagrams of team typologies to portray the relationships between IPE team typologies, cognitive levels, and TeamSTEPPs behaviors. The details of the analytical processes are described below.

2.3. Qualitative Analysis Using the Constant Comparative Method

An overall intent of the present study was to transform a monumental volume of data regarding IPE teamwork into typologies of IPE teams using the constant comparative method (CCM) of data coding and analysis [2]. The researchers did not intend to identify an emerging substantive theory. Therefore, the use of CCM in this study differed from its classical use in grounded theory (GT) methodology [20,21] in that CCM was used as an analytical technique for a more pragmatic purpose. This modified use of the CCM is documented in the current literature and has been used previously [2,20].

A single analyst used the constant comparative method [2,19] to analyze each of the 14 videos to identify central categories and their related characteristics with regard to team typologies. The constant comparative method involved the incorporation, collation, and comparison of emerging categories and characteristics both within and between IPE teams. Thematic synthesis of the emerging categories was used to look for commonalities and diversions within and across the student IPE teams [21]. This process required the analyst to review all transcriptions and videos several times and to thoroughly compare them individually as well as to make cross-comparisons to fully understand the IPE teams.

The CCM method used in this analysis consisted of three distinct stages, namely, open coding, axial coding, and selective coding [19]. An overview of those three analytical stages is presented below.

(1)

Open Coding. Identifying categories of behaviors that emerged while reading transcripts and viewing videos. The analyst read through the transcripts and viewed the videos several times and then started to create tentative categories that emerged. She summarized the observations by jotting down memos to herself regarding participants’ words and actions. The observations were not based on an existing theory, but rather on the meaning that was emerging from the data. By jotting down memos to herself, the analyst was able to establish categories and characteristics of the IPE groups.

(2)

Axial Coding. Categories and their properties were integrated by reviewing memos that had been made, making more memos, and then identifying potential interaction or modifying factors between categories. The analyst searched for relationships and connections among the categories.

(3)

Selective Coding. Identifying and choosing the core categories and systematically connecting them. Basically, the analyst identified the core categories that encompassed all instances of student discourse, then reread the transcripts and selectively coded any data that related to the categories. Ultimately, every instance of student discourse was assigned to one of the core categories.

2.4. The Team Typologies That Emerged from CCM Analysis

At the conclusion of the three stages of CCM coding (open, axial, and selective), four general team typologies had emerged that typified all 14 IPE groups. The analyst gave the four typologies the descriptive names of Facilitated, Cohesion, Consensual Validation, and Silo Mentality. It is important to note that at the point in time when these team typologies emerged and the names were assigned, the analyst had not consulted the current literature and was not looking for established team typologies to emerge during the CCM analyses. Instead, the four typologies (Facilitated, Cohesion, Consensual Validation, and Silo Mentality) emerged from the CCM analysis. After the analyst had characterized and named the typologies, she conducted a literature review to determine if they were typical team types as identified in the current literature and to determine if they were effective or ineffective approaches to teamwork. After all, the overall objective of this mixed methods study was to describe the inner workings of IPE teams by characterizing the team typologies that students used while solving patient cases and to provide actionable knowledge for how to improve their IPE teamwork. Cross-validation with typologies in the published literature augmented the current researchers’ understanding about the effectiveness of the team typologies that others had previously published and their relation to the findings of the current study [22,23,24,25,26]. That information is provided in the Section 3 of this manuscript.

2.5. Quantitative Data Analyses

The quantitative analysis included cognitive level scoring and team behavior scoring, using Bloom’s taxonomy to evaluate the cognitive level demonstrated in student discourse (Figure 1) [27,28]. In preparation for scoring student discourse for cognitive level, the primary investigator and a co-investigator (the Bloom’s raters) completed three training sessions to establish interrater reliability by assigning a cognitive level to each instance of student discourse according to a list of verbs associated with each cognitive level [27].

Interrater reliability was 94% after the third session. Student discourse at the Knowledge (K), Comprehension (C), and Application (AP) levels was considered to represent lower-order thinking (LOT) and collapsed into one category labeled LOT. Discourse at the Analysis (An), Synthesis (S), and Evaluation (E) levels was considered to be representative of higher-order thinking (HOT) and collapsed into one category labeled HOT. Comments that did not deal explicitly with case content were coded as social procedural (SP) [29] and included salutations and requests for assignment instructions.

Cognitive levels were converted to numeric values (K = 1, C = 2, Ap = 3, An = 4, S = 5, E = 6, and SP = 0) and the average HOT, LOT, and SP cognitive scores were calculated for each team and for each pre-professional program. Teams and pre-professional programs were placed in rank order from highest to lowest average cognitive level.

For the assessment of TeamSTEPPS behaviors, the PI worked with three TeamSTEPPS raters for two weeks to calibrate how they would assign TeamSTEPPS behaviors to student instances of discourse using the Performance Assessment for Interprofessional Communication and Teamwork (PACT). PACT is an instrument that was developed specifically for the purpose of assessing video captures of pre-licensure health professional students’ team performances in simulated patient scenarios. The PACT scoring rubric is a grid that includes each TeamSTEPPS behavioral marker and instructs the TeamSTEPPS raters to determine if each behavior is absent, isolated, consistent, or not applicable in videos of team deliberations [30]. In the current study, we elected to be more precise and used an event-based scoring strategy whereby every time a behavior was observed it was counted. Each instance of student discourse was assigned a number (1–26) that corresponded with the TeamSTEPPS behavior it represented (

Table 1. TeamSTEPPs Behavioral Markers 1–26.

Team Structure	Mutual Support
1. Recognize LEADER	16. Acknowledge STATEMENT
2. Understand ROLE	17. ALL PARTICIPATE
3. Understand TEAM GOALS	18. Call ATTENTION to error causing actions
4. Refer to PROTOCOLS/CHECKLISTS	19. ASK FOR HELP
5. Respond to potential ERRORS	20. CUS/Two-Challenge rule/DESC Script
6. SHARE IMFORMATION	Communication
Leadership	21. VERBALIZE activities
7. Delegate TASKS	22. REPEAT BACK
8. BRIEFS/HUDDLES/DEBRIEFS	23. HAND OFF
9. AUTHORITY vs. PARTICIPATION	24. SBAR
10. COLLECTIVE INPUT	25. Ask for CLARIFICATION
11. SPEAK UP	26. CLOSED-LOOP COMMUNICATION
Situation Monitoring
12. STEP PROCESS
13. Attend to INDICATORS
14. Maintain SITUATION AWARENESS
15. PATIENT included

). Each TeamSTEPPS rater entered their scores into the respective Google Sheets that had a separate row for each student comment.

Table 1: The five main TeamSTEPPs categories and the 26 individual behavioral markers associated with each are presented above. CUS stands for concerned, uncomfortable, safety issue; SBAR stands for situation, background, assessment, recommendation; DESC stands for describe, express, suggest, consequences. A complete description of the TeamSTEPPs behaviors is available in the TeamSTEPPs pocket guide (https://www.ahrq.gov/teamstepps/instructor/essentials/pocketguide.html accessed on 1 August 2018).

After scoring was complete, data were collapsed into the five main TeamSTEPPS categories of Team Structure, Leadership, Situation Monitoring, Mutual Support, and Communication. Frequency counts of student discourse in the five main categories were converted into a percentage of the total number of responses that were scored as Team Structure, Leadership, Situation Monitoring, Mutual Support, and Communication. The scores from both TeamSTEPPS raters were combined in this analysis.

2.6. Significance Testing Using Univariate General Linear Model

A univariate general linear model (GLM; IBM SPSS Version 27) [31] test was conducted to determine if there were statistically significant differences in the average cognitive levels among the four transcribed groups (one from each of the four team typologies) or among the eight pre-professional programs (medicine, pharmacy, nursing, physical therapy, physician assistant, dietetics, social work, and clinical mental health counseling). The univariate GLM was selected because it could also determine if there was a significant interaction between group and pre-professional program with regard to cognitive level.

3. Results

3.1. Participants

A total of 127 participants were randomized into 14 IPE teams. Each team included at least one student from each of the eight pre-professional programs, with the exception of physical therapy and physician assistant because there were fewer than 14 student participants from those programs. Each of the 14 IPE teams was randomly assigned one of the three patient case scenarios to solve. One patient had hypertension; another heart failure; and another was a bariatric surgery candidate.

3.2. Cross-Validation with the Current Literature

The typologies in the current study emerged from the CCM analysis. The CCM analytical technique does not include an extensive literature review prior to conducting the analysis, so the analyst is not guided by the literature to look for specific typologies that have already been characterized. However, after characterizing and naming the typologies that emerged in the current study, the analyst conducted a literature review to determine if the typologies (Facilitated, Cohesion, Consensual Validation, and Silo Mentality) were typical team types. A brief discussion of how each of the four typologies identified in the current study is similar to team types in the published literature provides cross-validation for the typologies that emerged in the current study.

The hope was that students in the IPE teams were using effective team processes. The reality, however, was that most IPE teams were not. The passages below address each team typology that emerged in the current study, matching up each with the team typology as identified in the current literature.

3.3. Silo Mentality Typology

The term Silo Mentality was identified as a team typology in the current study and also appears as such in the published literature on teams [22]. In general, “silo mentality” is often used to describe individual or group beliefs that can cause division and whose most common outcome is the creation of barriers to communication and the development of disjointed work processes. A silo mentality is often the result of inadequate training and inadequate development of interpersonal skills, which often hinders the formation of cooperative working relationships [22]. This sealed-off type of mentality can reduce motivation to excel at team tasks and tends to generate indifference towards the needs of others [22]. Although the silo mentality in health services seems to be a significant problem, literature on the subject is scarce. In the current study, Silo Mentality teams demonstrated the least successful team dynamics and spent the least time on higher-order thinking. They did not process the patient case together as a team, rather each team member prepared their own care plan for their profession, and these “silos” were then concatenated into a final presentation.

3.4. Consensual Validation Typology

The typology named Consensual Validation in the current study aligns with the term “shared identity” in the literature; it exists when team members perceive themselves as belonging to the same social category. Social identity theory suggests that an individual may have multiple social identities, each of which is associated with different social categories, such as race, gender, occupation, organization, department, hierarchical level, and task group. In this study, we focused on task group identity based on membership in a pre-professional program. The literature suggests that the absence of a strong shared identity can diminish within-group communication, while a strong shared identity among team members tends to promote information flow among team members [23]. In the current study, members in Consensual Validation team types formed two subgroups based on pre-professional program similarities: (1) the medical group: medicine, physician assistant, pharmacy, nursing; (2) the MH/lifestyle group: mental health, social work, physical therapy, dietetics. Most of the communication, however, occurred within the two subgroups as opposed to between the two subgroups. There was also a divergence in the two subgroups with regard to cognitive level during the 90 min session; namely, the MH/lifestyle subgroup moved into higher-order thinking (HOT) much more so than did the medical subgroup.

3.5. Cohesion Typology

The typology named Cohesion in the current study aligns with elements of what is termed social cohesion in the current literature. Social cohesion is implicated in causing “groupthink”, resulting in a group that does not engage in critical deliberations because everyone agrees. One of the major causes of groupthink and the resulting failure to be critical is social cohesion itself. Groups affected by groupthink do not explore alternatives as readily as other groups, and this phenomenon is likely to occur when groups are “highly cohesive” [24]. In the current study the Cohesion team demonstrated a team dynamic which started at the lower cognitive level as a group, and then moved together through all ranges of cognition (1–6) as a group. They spent more time in cognitive levels 1–4 and did not engage in as much discourse at the HOT level compared to the Facilitated group.

3.6. Facilitated Typology

The typology named Facilitated in the current study is also noted as such in the current literature. Facilitation is the complex skill of enabling, or empowering, a group of people to complete a task. To be a facilitator is to act as a human catalyst, sparking the chemistry that turns a collection of separate individuals into a working team. Facilitation can be a formal role within a group or it can be a skill that a manager or a group member can master and contribute improvisationally to the group. In the former case, the facilitator is often someone trained in this role who is brought in to help the group but remains uninvolved in the task itself. In the latter case, facilitation becomes more of a juggling act—the individual team member must move back and forth from participant in content (i.e., what the group is working on) to observer and designer of the process (i.e., how the group is doing the work) [25]. An effective facilitator is clearly advanced in metacognitive abilities with active control over their cognitive processes and an awareness of and reflection of their own thinking [25]. They also help the team develop metacognitive skills that are crucial in order to avoid medical errors [32,33,34]. In the current study, the Facilitated team demonstrated the most successful team dynamics and spent the most time in higher-order thinking. In the middle part of the 90 min discussion, there was a wide range of cognitive levels displayed because the team did not stay together on every point. They dissected, discussed, and re-assembled their thinking in this segment of their work. Overall, they spent the greatest amount of time in HOT of all the groups.

3.7. Average Cognitive Levels for Groups and Pre-Professional Programs

The rank order of the four groups from highest to lowest average cognitive level in this study was (1) Facilitated, (2) Cohesion, (3) Consensual Validation, and (4) Silo Mentality. The rank order (from highest to lowest average cognitive level) for the pre-professional programs was as follows: dietetics, mental health counselor, physical therapy, social work, medicine, physician assistant, pharmacy, and nursing. The average cognitive level is meaningful because Bloom’s taxonomy is hierarchical, and higher average cognitive scores represent more time engaged in higher-order thinking (HOT) discourse compared to lower cognitive scores representing more time spent in LOT and SP discourse (Figure 2).

3.8. Significance Testing for Differences in Average Cognitive Level between the Four Groups and between the Eight Pre-Professional Programs

A univariate general linear model (GLM) test was conducted to determine if there were statistically significant differences in average cognitive level among the four groups or between the eight pre-professional programs. The univariate GLM was also used to determine if there was a significant interaction between group and pre-professional program with regard to cognitive level. The univariate GLM revealed that, at the main effects level, there was a statistically significant difference in average cognitive levels among the four IPE groups that were transcribed (p ≤ 0.001). Because the difference was significant at the main effects level, pairwise comparisons (post hoc tests) were consulted to determine exactly where the differences were among the four groups. The post hoc tests revealed that one team had significantly lower average cognitive scores than the other three teams (Silo Mentality; p ≤ 0.001), one team had significantly higher average cognitive scores than the other three teams (Facilitated; p ≤ 0.001), and two teams had similar mid-range average cognitive levels (Consensual Validation and Cohesion).

With regard to pre-professional programs, the univariate GLM revealed that, at the main effects level, there was a statistically significant difference in average cognitive levels among pre-professional programs (p ≤ 0.001). Because the difference was significant at the main effects level, pairwise comparisons (post hoc tests) were consulted to determine exactly where the significant differences occurred among the eight pre-professional programs. Post hoc tests revealed that mental health, dietetics, physical therapy, and social work pre-professional programs engaged in discourse that, on average, was not at a cognitive level that was significantly different from the others but that was at a significantly higher cognitive level compared to medicine, pharmacy, nursing, and physician assistant pre-professional programs. Similarly, post hoc tests revealed there was not a statistically significant difference in average cognitive levels between medicine, pharmacy, nursing, and physician assistant pre-professional programs and these four programs, on average, performed at significantly lower cognitive levels than did mental health, dietetics, physical therapy, and social work pre-professional programs.

The univariate GLM additionally revealed there was a statistically significant interaction between the four typology groups and eight pre-professional programs (p ≤ 0.001) with regard to cognitive level. This significant interaction meant that cognitive levels for the pre-professional programs varied not only among pre-professional programs but also based on which of the four team typologies they were in. In the Silo Mentality group, the average cognitive level was lowest for all but one pre-professional program. In the Facilitated group, the average cognitive level was highest for all but one pre-professional program.

Figure 2. Percent of student discussion spent in social/procedural discourse and at lower and higher cognitive levels.

Figure 2. Percent of student discussion spent in social/procedural discourse and at lower and higher cognitive levels.

3.9. Team Presentations to the Panels of Experts

A critical analysis of the 14 team presentations to the panels of experts is beyond the scope of the current study. However, it is important to note that the final group presentations were all deemed by the investigators and faculty experts to have met performance expectations regardless of the rankings derived from Bloom’s taxonomy and the TeamSTEPPs behaviors. In one way, this is good news because the final care plans produced and presented by each of the 14 groups were acceptable. However, it is not good news because successful completion of the care plans did not require effective interprofessional teamwork.

4. Discussion

The current researchers were concerned that effective teamwork and interprofessional collaboration were not being practiced by IPE teams of students while developing care plans for patient case scenarios. The students at the IPE event were enrolled in eight pre-professional programs across three universities and therefore possessed neither a standardized nor universally known baseline level of teamworking skills. Furthermore, the current event did not include any actual team training. While the patient cases required input from several healthcare fields, they may not have restricted students from concatenating single-disciplinary solutions into one care plan to present to the faculty panelists. The current study was therefore conceived and carried out in order to determine whether the researchers’ concerns were well-founded. A video camera was placed in each of fourteen rooms, where student IPE sessions were recorded without interruptions. The researchers’ concerns were indeed proven to be true and the key finding of this study is that the majority of the IPE teams did not practice effective teamwork in their small group sessions. In fact, of all fourteen teams, only three fit into the Facilitated team typology, which constitutes effective teamwork.

The current study took place just prior to the COVID-19 pandemic before online meeting became the norm. The student IPE teams and faculty were physically present for this event, and student IPE teams were isolated in private rooms for their care plan deliberations. The faculty members who served as content experts were not present in the rooms but were on-site to watch and evaluate the final presentations. The event had been conducted in this same manner several times previously, and the current researchers became curious about what the students were actually doing and saying in their private rooms. We were concerned that successful completion of the care plans may not require effective interprofessional teamwork. At this event, the final group presentations were all deemed acceptable by the faculty experts; all groups met performance expectations.

It is both interesting and puzzling that in the current study the panels of faculty experts considered the care plans developed by all 14 IPE teams to be acceptable. Even Silo Mentality teams developed acceptable care plans. Two scenarios could explain this phenomenon. Either effective teamwork is not important in IPE events or the patient case scenarios were not designed in a way that would require effective teamwork for successful completion. Given the plethora of literature documenting the importance of effective teamwork in IPE, one must assume the case scenarios themselves did not sufficiently require it.

The participants at the current event were given no training in teamwork skills at the event and may have been unequipped to practice teamwork. Although individual programs may provide team training, it would be unlikely for students from four universities and eight professional programs to have a single shared mental model for teamwork. Additionally, the event did not include any team training to equip the students with a shared mental model for how to work together. As such, in a short-term, task-centered activity such as the IPE event, the IPE teams did not exhibit effective teamwork. The current analysis concluded that only three of the fourteen teams practiced effective teamwork, yet nonetheless they all developed acceptable care plans.

One might say that the fact that all the student teams presented care plans that were evaluated as being successful by the faculty experts would negate any concern that interprofessionalism and effective teamwork had not been required. However, because interprofessional practice is known to equate to better patient outcomes in real-world healthcare settings, considered in concert with the facts that provider burnout is reduced and patient satisfaction is increased when effective teamwork with demonstrated interprofessional skills is employed, a clear argument is presented for the importance of the current study [6,7].

Data triangulation in this mixed methods study provides more knowledge about how IPE teams function than could have been provided using a less meticulous design. The researchers did not know how students were arriving at their care plans during their 90 min collaborative sessions, and the constant comparative method was used in order to find out. Data triangulation combined qualitative (CCM) and quantitative methods (event- and time-based behavior observation and micro-level coding), along with a cross-validation procedure with the current literature to provide credibility for and a deeper understanding of the four team typologies that have emerged in our study [18,35,36].

The team typologies Facilitated, Cohesion, Consensual Validation, and Silo Mentality are listed in order from most to least effective with regard to team dynamics. In fact, recall from the Section 3 that of the four typologies, only one, Facilitated, was associated with effective teamwork according to the previous published literature on teams. Also, a Silo Mentality approach to teamwork is not associated with effective team behavior at all [22]. In light of these two facts, it is not a coincidence that the Facilitated team demonstrated higher-order thinking significantly more than the other teams, and all but one of the eight pre-professional programs had their highest average cognitive level in a Facilitated group [25]. On the other end of the spectrum, the average cognitive level was lowest for the Silo Mentality teams, and all but one of the eight pre-professional programs had the lowest average cognitive level in a Silo Mentality team. Overall, there was a dose-dependent relationship between team effectiveness and cognition whereby increased team effectiveness was met with a concomitant increase in (1) average cognitive level and (2) more effective team dynamics, as measured by TeamSTEPPs behavioral markers (Figure 3).

Actionable Knowledge for Targeted Interventions

Organizations have come to expect new graduates to have an understanding of collaborative care and the knowledge and skills to engage in interprofessional practice (IPP) and effective teamwork when they enter the workforce [37]. The results of this study identify where IPE team dynamics were less than optimal at an IPE event and provide the knowledge required to implement the following recommendations for targeted interventions to improve IPE teamwork at future IPE events, as follows:

• Provide TeamSTEPPs training before the event. Because most of the IPE teams did not demonstrate mastery of the principles included in TeamSTEPPs (team structure, leadership, situation monitoring, mutual support, and communication), the primary recommendation is to provide students in pre-professional programs with substantive team training, such as TeamSTEPPs training. In the present study, this was not carried out. TeamSTEPPs, properly understood, provides students with a shared mental model and process for working in short-term, task-centered teams [38,39].
• Employ multimodal teaching and learning strategies to improve teamwork. A multimodal pedagogical strategy that combines presentation of information (e.g., lecture), visual demonstrations (e.g., videos), and practice (e.g., IPE events or simulation) may be more likely to be effective than using a single instructional strategy. This multimodal teaching strategy was also not provided in the current study. Using a variety of instructional strategies appeals to different learning styles, involves both passive learning (e.g., information) and active learning (e.g., demonstration and practice), and engages a number of the learner’s senses [40]. Lectures facilitate an accurate mental model for effective teamwork formation. Demonstrations provide students with contextualized examples of effective teamwork. Practice involves an action-based approach to learning teamwork training material and is critical to acquiring real-world problem-solving techniques that promote skill acquisition and retention [3].
• Develop patient case scenarios that require metacognition for successful completion. Becoming interprofessionally competent requires that team members engage in problem-solving activities that require metacognition in teamwork, reflection on decisions and practices, analysis of team processes, and awareness of one’s own thinking [34]. Patient cases that are designed for use in IPE events should require more than a concatenation of unidisciplinary, domain-specific content knowledge for successful plan development.
• Emphasize both technical and non-technical skills. An IPE event should emphasize the equal importance of a summative evaluation of clinical skills and a formative evaluation of the process the participants use to derive care plans. While the current study included these evaluations, the event planners had not included such training for the participants.
• Apply rigorous evaluations of both technical and non-technical skills. Methodical approaches that take dynamics into account should be used to evaluate students working in IPE teams. These evaluations can provide actionable knowledge about interaction processes of student teamwork [5].

The Agency for Healthcare Research and Quality (AHRQ) website provides all of the TeamSTEPPs curricular materials (https://www.ahrq.gov/teamstepps/curriculum-materials.html accessed on 15 August 2019) required to implement these recommended teaching strategies, including PowerPoint presentations, videos, and learning vignettes.

5. Conclusions

The results of this study are impactful because they identify where IPE team processes break down and suggest targeted interventions to improve the effectiveness of IPE team processes. Given the high stakes associated with interprofessional competence (IPC), such as patient safety and successful patient outcomes [41,42], this study is long overdue. More studies are needed that rigorously assess IPE team dynamics and evaluate the changes that occur in response to various IPE interventions. Determining measurable objectives and implementing continual re-evaluation following each intervention will be the only way to know if an intervention is working to improve IPE teamwork.