Background Communication failures in healthcare teams are associated with medical errors and negative health outcomes. These findings have increased emphasis on training future health professionals to work effectively within teams. The Team Strategies and Tools to Enhance Performance and Patient Safety (TeamSTEPPS) communication training model, widely employed to train healthcare teams, has been less commonly used to train student interprofessional teams. The present study reports the effectiveness of a simulation-based interprofessional TeamSTEPPS training in impacting student attitudes, knowledge and skills around interprofessional communication.
Methods Three hundred and six fourth-year medical, third-year nursing, second-year pharmacy and second-year physician assistant students took part in a 4 h training that included a 1 h TeamSTEPPS didactic session and three 1 h team simulation and feedback sessions. Students worked in groups balanced by a professional programme in a self-selected focal area (adult acute, paediatric, obstetrics). Preassessments and postassessments were used for examining attitudes, beliefs and reported opportunities to observe or participate in team communication behaviours.
Results One hundred and forty-nine students (48.7%) completed the preassessments and postassessments. Significant differences were found for attitudes toward team communication (p<0.001), motivation (p<0.001), utility of training (p<0.001) and self-efficacy (p=0.005). Significant attitudinal shifts for TeamSTEPPS skills included, team structure (p=0.002), situation monitoring (p<0.001), mutual support (p=0.003) and communication (p=0.002). Significant shifts were reported for knowledge of TeamSTEPPS (p<0.001), advocating for patients (p<0.001) and communicating in interprofessional teams (p<0.001).
Conclusions Effective team communication is important in patient safety. We demonstrate positive attitudinal and knowledge effects in a large-scale interprofessional TeamSTEPPS-based training involving four student professions.
- Team training
- Health professions education
Statistics from Altmetric.com
An increased focus on interprofessional education (IPE) has resulted from several influences. Among the most compelling is the growing recognition and evidence that improved communication and collaboration by interprofessional teams leads to better delivery and access to care. In its 2004 sentinel event data report,1 the Joint Commission listed leadership, communication, coordination and human factors as among the leading root causes of sentinel events. Failures in communication within interprofessional healthcare teams are established causes of medical error2 and negative health outcomes,1 ,3 ,4 including death.5 In addition, team communication failures have significant economic impacts that may reduce quality and safety, or access to care.1 ,6
The relationship between team communication and patient safety4 has increased the emphasis placed on training future health professionals to work within teams.7–9 However, few studies have sought to demonstrate that prepractice interprofessional team training is effective in building the foundations for later practice within healthcare teams. Increasingly, educators have sought to create interprofessional trainings that teach the key elements of effective teamwork in simulated settings that allows for the practise of skills in a stimulus-rich but controlled environment. Interprofessional team simulation provides a means to both learn and practise safe teamwork skills.
With this paper, we describe a sophisticated interprofessional team-based training, and take the important first step of demonstrating that participating students can learn critical elements of team communication, and to value team functioning. We also demonstrate the utility of new self-report instruments. Our study employs an established team communication framework,10 ,11 Team Strategies and Tools to Enhance Performance and Patient Safety (TeamSTEPPS),12 in teaching skills using manikin simulators and standardised patients. While exceptions exist, training efforts commonly neglect to provide evidence that learning has occurred, and that learning is transferable to clinical settings. The purpose of this paper is to describe and demonstrate the effectiveness of an innovative interprofessional training effort using simulation.
The curriculum and assessment tools described in this study were developed as part of a grant funded through the Josiah Macy Jr Foundation for the purpose of improving communication within learning teams, enhancing team-based care, increasing awareness of respective roles and responsibilities, and promoting an understanding of interprofessional values and ethics. Faculty from the schools of medicine, nursing, pharmacy and the MEDEX Northwest Physician Assistant (PA) Training Program worked to create novel and distributable training tools for team communication aimed at reducing errors and improving patient safety. The goal was to create new collaborations, while strengthening and leveraging existing interprofessional activities—where students from different disciplines work together—and intraprofessional activities—where students work only with students within their discipline—designed to promote incorporation of team communication into programme curricula across the health profession schools.
Interprofessional team communication is defined by skills learned and later modified and reinforced when healthcare workers work collaboratively to provide competent care. Competence to practise safely requires effective communication with patients and colleagues, active listening, assertiveness, respect and timeliness. Failures occur when vital information is not communicated between team members, or team members incorrectly interpret messages. Failures to communicate information may result from adversarial relationships, roles that are not clearly defined, or insufficiently developed communication pathways within teams. Incorrect interpretations occur when providers use different terms to convey information, accept incomplete information, or assign different weights to communications. In each case, the result may be an error.
The educational framework for the development of the training content was based on TeamSTEPPS.12 TeamSTEPPS was developed from research and development collaborations between the Department of Defense (DoD) Patient Safety Program and the Agency for Healthcare Research and Quality (AHRQ), and is rooted in crew resource management13–15 (CRM). Increasingly, there are calls for the incorporation of team training into clinical settings16 stemming from successful applications within surgery and trauma-focused settings.17–19 More recently, clinician educators have sought to integrate TeamSTEPPS tools11 ,16) into healthcare education. Simulation-based training provides an excellent vehicle for student teams to work collaboratively in a realistic yet structured environment without risks to patients.20–22
The overall goal of the interprofessional training was for students to acquire effective interprofessional team communication skills. Taking part in these exercises allowed students the opportunity to practise and observe interprofessional communication, and through facilitated debriefings learn what proved most effective. Our study goal sought to demonstrate that interprofessional students would report enhanced readiness23 through improved self-efficacy,24 ,25 motivation, positive attitudes and practice opportunities. We proposed assessing testable research questions aligned with our training goals. Following training, would interprofessional students report
Improved attitudes, motivation and self-efficacy to working within interprofessional healthcare teams?
Having observed and practised key team communication skills?
Increased understanding of interprofessional team skills?
Three adult acute cases (two adult males and one teenage male) to demonstrate communication across members of a healthcare team were developed collaboratively by a team of 9 interprofessional faculty, 19 student volunteers from various health profession programmes and 6 staff members. They were designed to provide opportunities for an interprofessional team to demonstrate team communication strategies and skills in an acute situation, while delivering care to patients and their families. The three adult acute cases were: asthma exacerbation in a teenager (simulator with standardised family member), congestive heart failure in an elderly male (standardised patient) and supraventricular tachycardia in a male postsurgery (simulator with standardised family member). Each of the three adult acute cases, and an adapted TeamSTEPPS training, were tested with 49 students in June 2010.
Following the successful demonstration of the three adult cases, we designed three paediatrics and three obstetrics cases to reflect parallel skills requirements, and provide comparable team communication skills training in areas aligned with students’ career plans. The paediatric cases were: severe asthma, acute seizure and sepsis (each using a simulator). The three obstetric cases were precipitous vaginal delivery, mild postpartum haemorrhage and mild postpartum haemorrhage complicated by error (each using a standardised patient). Each of the cases are described and available with toolkits for implementing and developing simulations on the Center for Health Sciences Interprofessional Education, Practice and Research website (http://www.collaborate.uw.edu).
Interprofessional Team Capstone
We used an existing capstone week held by the University of Washington School of Medicine during the last week of classes as an opportunity to bring graduating medical bachelors of science in nursing, PharmD and masters PA students together for interactive interprofessional training sessions. Participation in the interprofessional training was required for all students except for the PA students who were volunteers. Students participated in one 4 h training block during the capstone week.
The Interprofessional Team Capstone experience was designed to train students from four healthcare professional programmes to practise together as a team. Students had the option to participate in one of three separate (focal area) trainings: (1) adult acute care (2) paediatric, or (3) obstetric cases. This breakdown allowed students to select an area of practice most similar to their anticipated specialty. The training sessions occurred at two academic medical centre training facilities across a 4-day period. In each student focal area, the educational intervention included a didactic session and three simulated exercises. Following an icebreaker activity developed to introduce interprofessional teamwork, students had 40 min of didactic instruction on patient safety and TeamSTEPPS communication skills.
For the simulation sessions, students were divided into interprofessional teams with balanced composition across healthcare programmes. The teams then completed three simulated exercises (approximately 15 min each). Two exercises used a manikin simulator and a standardised family member, and the third used only a standardised patient. Each simulation was preceded by an introduction (eg, case materials and ground rules), and was followed immediately by a facilitated debriefing session. When not actively participating in a simulation, or when there were too many students to accommodate, students were asked to step back and observe. Students rotated through observer and participant roles throughout the three cases. All students (observers and participants) participated in the end-of-case debriefings. Student teams met again as a large group for a final wrap-up with the facilitators to review what they had learned.
Measuring the intervention's impact
To assess the impact of the training on student learning, we developed and selected instruments to assess attitudes, skills and knowledge (table 1). Instruments were developed and reviewed by the UW Macy assessment team, consisting of a physician and PA educator (LV), a visiting physician fellow (KB), a nurse practitioner (SW), a pharmacist (DH), two medical educators (DB, DS), a nursing educator (BZ) and two nursing graduate students (CC, EAR). These instruments included attitudes towards TeamSTEPPS communication skills, self-reported knowledge, motivation to implement these skills, their value or utility and student self-efficacy in being able to implement these skills in practice.
Student respondents were described by several demographic variables. These included the student educational programme, sex, age, healthcare experience prior to entering their current educational programme, and previous familiarity in working with healthcare teams (eg, respiratory tech or medic).
Two instruments were administered, both pretraining and post-training. To assess attitudes towards team communication we administered the TeamSTEPPS Teamwork Attitudes Questionnaire (TAQ).26 The TAQ is a validated instrument containing 30 Likert-type items assessing attitudes towards the five dimensions (Team Structure, Leadership, Situational Awareness, Mutual Support and Communication) underlying the TeamSTEPPS communications model. Attitudes were assessed by the Attitudes, Motivation, Utility and Self-Efficacy (AMUSE); a 21 Likert-type item instrument constructed to assess AMUSE. The attitudes, motivation and utility items were developed by the authors in consideration of existing instruments to assess similar team constructs.23 ,27–30 The self-efficacy items were guided by Bandura's recommendations for developing scales to assess efficacy31 and following Bandura's theory of agency,25 ,32 that people act on their environment, set goals and monitor progress, learning both through direct experience as well as vicariously through observing others.
Two new instruments and the AMUSE were developed specifically to assess whether students had the opportunity to practise or observe specific team behaviours, and whether these training opportunities were positively regarded, and represented skills that students believed would be of value to carry forward, and for which they had sufficient familiarity to successfully implement in practice. One instrument asked students to report the frequency with which the training cases provided the opportunity to practise or observe key communication behaviours. Examples included whether team members ‘… were consulted for their experience’ or ‘… asked for assistance’. This scale consisted of 15 items with response options ranging on a 5-point scale from ‘Never’, to ‘Frequently’. A second instrument asked respondents to rate their understanding of key concepts both before and after training (eg, ‘BEFORE participating in training I had a good understanding of the benefits and application of Situation, Background, Assessment, Recommendation (SBAR)’ and ‘AFTER completing the training I have a BETTER understanding of the benefits and application of SBAR’). This instrument consisted of 10 item-pairs on a five-point scale from ‘Strongly Disagree’ to ‘Strongly Agree’. Copies of each of the instruments are available at the following website: (http://www.collaborate.uw.edu/educators-toolkit/tools-for-evaluation.html-0). The presurveys and postsurveys were completed online and generally took between 10 and 15 min to complete. Pretraining surveys were open to students for 2 weeks prior to training until 2 days prior to the training. Post-training online surveys were completed either on the day the subject completed training, or approximately 2 weeks post-training. For logistical reasons, we were unable to have students complete multiple postassessments. Students were randomly assigned to the group that completed the survey on the day of the training, or to the group that completed the survey 2 weeks post-training. This allowed us to assess degradation of training effects over time. All procedures were approved by the University of Washington Institutional Review Board.
Statistical analyses and instruments were selected to align with the training goals. Within-group differences (pre vs post) were analysed using paired t tests. Analysis of variance (ANOVA) was used to explore differences across interprofessional student groups (eg, medical). Instrument internal consistency was assessed using Cronbach's α. All tests applied a p=0.05 level of significance. When multiple tests were performed simultaneously, the critical values were adjusted using the Bonferroni criterion to reduce risk of Type I error.
A total of 306 fourth-year medical, third-year nursing, second-year pharmacy and second-year PA students completed the training. Of the total, 255 (83.3%) students completed the preassessment and/or the postassessment, of which 149 (48.7%) students completed both the preassessment and postassessments (completers). Table 2 provides a breakdown of the student completers by professional programme. There was no significant difference, by profession, for completer classification (χ²=5.33, p=ns). Completers did not differ significantly from non-respondents or students completing only one assessment component on profession, sex, age or previous healthcare experience (each p=ns). Therefore, the analyses reported here reflect those students who completed both preassessment and postassessments, allowing for a preassessment vs postassessment comparison on study variables. Table 3 provides a breakdown of the completers by sex, age and healthcare, and healthcare team experience.
Postassessments were completed in one of two groups: day of the training, or 2 weeks post-training. Change scores for aggregate measures between preassessment and postassessment were compared using one-way ANOVA. After adjusting for the possibility of an inflated Type I error rate, no significant differences were discovered as a function of the date of the postadministration survey (each p>0.10). The staggered postmeasures were, therefore, aggregated to a single set of postmeasures. Only seven PA students completed both the preassessment and the postassessment. This number was not sufficient to treat as a separate subgroup, and the PA students were excluded from group analyses.
First training goal
Our first training goal focused on positive attitudinal shifts (including motivation and self-efficacy). The AMUSE was used to assess changes in student attitudes, motivation, utility and self-efficacy following training. Each subscale (α=0.90–0.79) and the aggregate total (α=0.90) achieved acceptable levels of internal consistency. Change scores were used to assess impact. Inspection of table 4 indicates that significant positive changes occurred for the AMUSE total score (p<0.001), and each of the four AMUSE subscales (p<0.001 to p=0.005). This provides evidence that training increased students’ positive attitudes towards working in teams, that students were more motivated to work in teams, saw greater value (utility) to this type of training and practice and felt able to implement the skills they had learned (self-efficacy). The largest effect was seen for the AMUSE utility score (mean=0.41, 95% CI 0.32 to 0.50). The smallest effect was seen for the AMUSE self-efficacy score (mean=0.12, 95% CI 0.04 to 0.21). Individual students tended to show improvement in their attitudes, motivation, beliefs about utility and self-efficacy; this effect was relatively uniform across the professional programme and focal area.
Table 4 also provides the prescores and postscores for the TeamSTEPPS Attitude Questionnaire (TAQ). One TAQ subscale (Mutual Support) exhibited marginal internal consistency (α=0.62). The other TAQ subscales (α=0.85–0.94) and the TAQ aggregate (α=0.93) achieved acceptable internal consistency. Significant positive increases were noted for TAQ Total Score (p<0.001), TAQ Situation Monitoring (p<0.001), TAQ Team Structure (p=0.002), TAQ Communication (p=0.002) and TAQ Mutual Support (p=0.003). There was no significant change in the TAQ Leadership score (p=0.062). The largest effect was seen for the TAQ Situation Monitoring (mean=0.19, 95% CI 0.10 to 0.38), and the smallest significant effect was for Communication (mean=0.13, 95% CI 0.05 to 0.21). Similar to the AMUSE results, individual students showed improvements in most of the TAQ subscales, an effect which was not differentially related to the student's professional programme or the focal area of the training.
One-way ANOVAs were conducted on the change scores of the TAQ and AMUSE aggregate total scores and subscales to explore whether differences occurred across student groups from different professions. When conducting analysis by programme of study (medical, nursing, pharmacy), robust statistical differences between AMUSE subscales emerged only for motivation (p=0.010, η2=0.06) and self-efficacy (p=0.005, η2=0.07). For motivation, this reflected lower postscores for pharmacy students (mean=3.53, SD=0.90) than for medical (mean=4.11, SD=0.46) or nursing students (mean=4.13, SD=0.59). Medical students (mean=3.89, SD=0.55) reported higher postlevels of self-efficacy than did nursing (mean=3.67, SD=0.43) or pharmacy students (mean=3.56, SD=0.73).
Second training goal
Our second training goal sought to provide students the opportunity to observe and practise team communication skills. In the postassessment, interprofessional students were asked to rate the frequency with which they saw or participated in a series of behaviours. These behaviours are provided in table 5. Since these questions could only be delivered post-training, we have reported results for all students who completed the postassessment (n=21 971.6%). Respondents were significantly more likely (adjusted for Type I error) to report having had the experience of team leaders assigning tasks and sharing information with team members, and to report examples of communication skills that served to reduce error. Observations of team members effectively asserting patient safety concerns, offering each other help, or utilising patients and/or family members as critical members of the care team were less likely to be reported.
Third training goal
Our third training goal focused on increasing student understanding of team skills. As part of the postassessment, students reported levels of agreement that training had increased understanding of key learning objectives (table 6). The largest changes occurred in beliefs around the benefits of implementing TeamSTEPPS (mean=1.48, 95% CI 1.33 to 1.63) and ability to advocate within-teams (mean=1.27, 95% CI 1.13 to 1.41). The least change occurred in student understanding of the association between interprofessional teams and patient safety (mean=0.60, 95% CI 0.49 to 0.71), and of the importance of offering assistance and seeking help (mean=0.39, 95% CI 0.29 to 0.50).
At the conclusion of training, participants (n=292) completed a brief evaluation of the experience. Overall, students within each of the three focal areas reported the trainings to be valuable (1=Not at all valuable, to 5=Highly valuable). They especially reported value in the TeamSTEPPS introduction (mean=4.57, SD=0.70), and the final debrief following the completion of the three cases (mean=4.29, SD=0.95).
Participants were also asked their level of agreement (1=Strongly Disagree, to 5=Strongly Agree) with specific aspects of the training. Participants were generally in agreement that materials were at an appropriate level (mean=4.42, SD=0.78), provided valuable team skills training (mean=4.66, SD=0.59), provided a realistic experience of the challenges faced when working in interprofessional teams (mean=4.44, SD=0.78), and provided a valuable opportunity to communicate with students from other professions (mean=4.71, SD=0.56).
Students were also asked to describe their most valuable learning experience in the training. Three consistent themes emerged: (1) value in the opportunity to work with students from different professional schools, (2) the value of learning and practising specific communication skills in a supportive environment and (3) value of practising skills within an interprofessional team.
The Interprofessional Team Capstone was successful at several levels. Initial successes are reflected in the considerable logistic challenges we addressed. These included: recruiting and scheduling students from four separate health professions; securing support from the school deans and programme directors; recruiting sufficient volunteer faculty to staff the training sessions for four full days; training the faculty using a ‘train-the-trainer’ approach; securing the physical space necessary to conduct the simulation training. Other successes are reflected in the positive self-report from students and changes in their attitudes, beliefs and confidence resulting from the training. This may result, in part, from taking a multimodal approach to case development that combined manikin simulators with a standardised patient or standardised family member. This allowed us to capitalise on the benefits of both modes to provide a rich student learning experience. The cases were considered realistic and engaging, the communication challenges were important, and the opportunity to work within interprofessional teams was described as valuable. Students enjoyed the activities and reported they had benefited professionally from participation. This benefit was reflected in improved attitudes towards interprofessional training, an increased intrinsic and extrinsic motivation to participate in future trainings, a perceived value for the utility of TeamSTEPPS communication training, and an increased sense of perceived self-efficacy in translating the skills learned in training into practice. This was consistent whether students were surveyed on the day of the training or 2 weeks following training.
Our cases, and our case development processes, parallel the recommendations of World Health Organization (WHO)33 for the creation of multiprofessional patient safety education. We sought to develop cases that were interesting, relevant, realistic and readily applicable to practice. Most importantly, students were provided the opportunity to practise skills learnt in multiple realistic simulations, and encouraged to engage and receive feedback following each activity. Students learnt by doing, not simply by observing.
We developed two new instruments for this study and reported on a use of the previously unpublished AMUSE instrument. These instruments were designed to allow students to report experiences from interprofessional team trainings. While observer-based instruments for team training exercises have been reported, similar self-report instruments have not been widely discussed. Our instruments may prove of value to other interprofessional trainings where educators seek to better understand student attitudes and perceptions.
One of the most consistent findings was reflected in students’ written stated reports of the value of working directly with students from other professions. While most of the participating students had experience in clinical settings working with practitioners from other professional programmes, the majority had minimal experience working in interprofessional activities with the people who would be their future colleagues. Fellow students provide an opportunity for learning to occur in lower stakes and a less stressful environment than working in interprofessional healthcare teams caring for real patients.
Our study has important limitations to consider. First, this was a simple pre-post design, without a defined control group. It is possible that student postresponses resulted, in part, from other aspects of their ongoing professional training. This concern is minimised by the relatively short span of time between the preadministrations and postadministrations. However, it is also possible that students were sensitised by the preassessment to be more alert and attuned to the assessed elements of the team communication training. However, few studies meet the rigorous standards necessary to draw causal relationships between the various components of training and specific outcomes,34 ,35 and the empirical base for similar trainings has been questioned.36 Future work with randomised controlled studies of students is needed with outcomes that include later professional practice.
We have demonstrated that positive outcomes are obtainable through a short introduction to the TeamSTEPPS skillset, and the opportunity to practise these skills and receive feedback from experienced facilitators. However, interpretation of the findings is confounded by unmeasured factors, which include the effects of an individual's assigned team members, and different team facilitators on the team's learning experience. In addition, we have relied on self-assessment instruments that have not been fully validated. This is partly due to the paucity of validated measurement assessment tools for use with students. This is changing, especially with work around self-efficacy,37 and we are completing additional validation efforts on the tools we have developed for the current study.
This study does not directly address student skill attainment, or the impact of newly learned skills on practice. In addition, other researchers have questioned the impact of similar interventions.38 The former can be understood to some extent through our ongoing analysis of student performance from the videos collected during the Capstone exercise. This will provide some evidence that the skills have been demonstrated, as well as the quality of the skills performed. Capturing downstream behaviour change, and the impact of this change, are more difficult. Demonstrating the effects of training will require longitudinal studies to capture the impact of attitudinal changes on behaviour within clinical settings.
Simulation training of interprofessional student teams represents a first step in establishing improved communication skills within practising clinical teams. We have shown that student teams can have significant attitudinal shifts and practice, and observe important team skills. Our work, funded through the Macy Foundation, has allowed us to build distributable resources, including interprofessional training cases and a template model for the creation of new cases. We encourage the development of multimodal simulations employing manikin simulators and standardised patients, and standardised family members as a means to leverage the benefits of both modes while optimising the student experience. Dissemination and the application of these materials broadly across healthcare training programmes will demonstrate the achievement of a principal goal underlying funding. In addition, at the University of Washington, the trainings described in this paper have been integrated into the ongoing curriculum.
Validation of the cases and interprofessional team training tools will require close inspection and quantitative assessment of the impact such training will ultimately have on the quality of healthcare delivered. We do not report unequivocal evidence for the effectiveness of our trainings; we provide a foundation for team-communication investigators to establish the next steps to create best-practice training models. Our team is currently establishing the validity of observational tools to assess team performance, as well as criteria for the assessment of videotaped team interactions. The outcomes and successful training activity reported here, when combined with the observational work in development, takes an important step towards meeting the joint commission's call that measurement represents the ‘heart of safety,’ and that improved care first requires the examination of high-quality measures of outcomes.39
The authors would like to acknowledge funding from a Josiah Macy Foundation Board Grant (B08–05), and all the members of the University of Washington Macy Team that developed and implemented the cases, simulations and trainings.
Contributors Each author contributed to the conception of the study, the study design, writing and critical review of the manuscript. Each author approved the final version of the manuscript. Data collection and analysis was conducted by DB, EAR and CRC.
Funding The authors would like to acknowledge funding from a Josiah Macy Foundation Board Grant (B08-05), and all the members of the University of Washington Macy Team that developed and implemented the cases, simulations and trainings.
Competing interests None.
Ethics approval University of Washington Internal Review Board.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.