Analysis of the Interprofessional Clinical Learning Environment for Quality Improvement and Patient Safety From Perspectives of Interprofessional Teams

In 2018 the Clinical Learning Environment Review (CLER) Program, which reviews graduate medical education (GME) engagement in 6 focus areas, reported that despite many programs' efforts to advance quality improvement (QI) training, transfer of QI knowledge and a substantive QI experience to residents has not been successful, and that resident participation in interprofessional (IP) QI and patient safety (QIPS) efforts in particular were lacking.¹  The CLER report described that, while most residents report some participation in QIPS activities, many lack understanding in QI principles and have fragmented exposure to systems improvement and safety activities.¹  As a result, the CLER Pathways to Excellence 2.0 highlighted IP QIPS work as an area for GME improvement nationally and introduced a new CLER focus area related to teamwork for patient care called Teaming.² 

However, not much is known about the interprofessional clinical learning environment (IP-CLE) for QIPS work within teaching health systems. Much has been written about improvement of the CLE for residents and IP team members within the context of routine clinical care^3–5  and around factors that promote or inhibit teamwork for patient care.^6–11  Yet, there is little understanding about the CLE in relation to IP teamwork for QIPS activity and what factors affect resident integration into these teams.

The distinction between IP teams performing routine clinical care and those working on QIPS activity is important because the teams may not always be the same, and this subset of IP teams may have its own culture and milieu. Gaining knowledge about these topics could inform health systems regarding how to promote CLEs that encourage IP teamwork for QIPS and effectively incorporate residents.

We derived our understanding of the CLE through the work of Gruppen et al,⁴  which builds on the analysis of the learning environment from Genn¹²  to describe the CLE as a complex psycho-social-physical construct that is co-created by individuals, social groups, and organizations in a particular setting and shaped by contextual climate and culture. Genn details how, within medical education, the learning environment is the “most significant manifestation” of a curriculum and to understand or change a curriculum one must consider the learning environment as the 2 are inextricably linked.¹²  Genn details the many elements that comprise the CLE and Gruppen et al further group them into 4 components: personal, social, organizational, and physical and virtual spaces.^4,12 

In this study, we seek to determine the current state of the interprofessional QIPS learning environment by drawing on perspectives of interprofessional residents, faculty, and staff. We focus on factors that affect resident integration into these IP teams for QIPS and factors that promote or detract from a rich IP-CLE.

This qualitative study used thematic analysis of scripted interviews with IP teams involved in QIPS efforts in patient care units throughout the University of California, San Francisco (UCSF) Health System, a large academic tertiary health system.

We developed our interview guide through modification of an existing interview script currently being piloted by the Accreditation Council for Graduate Medical Education (ACGME) Pursuing Excellence Pathway Leaders Patient Safety Collaborative to assess the IP-CLE for QIPS at 8 GME program sites.¹³  The Pursuing Excellence Collaborative developed this script after careful review of the National Collaborative for Improving the Clinical Learning Environment 2017 report¹⁴  as well as the Institute of Medicine's Conceptual Framework for Measuring the Impact of Interprofessional Education.¹⁵  We added questions to ensure assessment of all domains of the CLE from Gruppen et al⁴  and added questions related to resident involvement in QIPS efforts (provided as online supplementary data). This interview guide and source script was tested with one interprofessional team including 3 members from the adult ICU unit for clarity of the questions as well as the logical flow to the interview protocol prior to administration with our study group.

Scripted interviews allowed us to thoroughly assess the different facets of the CLE and to standardize the questions posed to individuals of different professions for means of comparison.

To represent a diverse and representative sample, we asked the health system and GME leadership to identify key adult and pediatric inpatient and outpatient patient care units engaged in IP QIPS activities. We approached unit medical directors to request at least 3 participants from different professions for 15- to 20-minute individual interviews. We purposefully interviewed participants from different professions to account for varied perspectives and continued interviews until we reached informational redundancy. We recorded, transcribed, and anonymized all interviews.

We analyzed interview transcripts using template analysis.¹⁶  We chose this approach as it aligns with critical realist epistemology, which postulates that the social world is real, but can be perceived subjectively and is driven by various mechanisms within specific contexts that produce outcomes.¹⁷  Through analysis of these mechanisms, one can inquire about how phenomena work and can determine how to manipulate such phenomena. Three authors (M.K.C., C.B., S.C.) engaged in an iterative consensus-building approach to develop an initial coding template and, subsequently, a code book with definitions. Transcripts were then coded independently by each author and then reconciled together. Dedoose analytic software (SocioCultural Research Consultants LLC, Manhattan Beach, CA) was used to organize coded transcripts. Using template analysis and a general inductive approach,¹⁸  the authors reviewed coded excerpts to identify cohesive themes. These themes were refined through consensus, and one representative quote for each theme was chosen and compiled into tables (key quotes are highlighted within this article). The authors also compared inpatient and outpatient interviews as well as those from different fields of medicine to assess for thematic variance.

The first author (M.K.C.) practices primary care in the division of general internal medicine and participated in IP QIPS initiatives as a resident and currently coaches QIPS initiatives. Another author (R.B.) also practices primary care in general internal medicine, has participated in and coached IP QIPS initiatives, and as Associate Dean for Graduate Medical Education and a Designated Institutional Official has led the ACGME's Pursuing Excellence in Clinical Learning Environments initiative at UCSF. The other authors (S.C., C.B.) are education researchers. Senior author (C.B.) has participated in and coached IP QIPS initiatives and has been an active member of the Pursuing Excellence Collaborative and is a professor in general internal medicine. While analyzing transcripts, authors took reflexivity notes to account for individual perspectives and discussed notes with other authors to facilitate richer data analysis and interpretation. Throughout the coding process, the authors engaged in discussion to ensure that codes arose from and were supported by the data and were not imposed by the authors' existing knowledge and experiences.

The UCSF Institutional Review Board granted our project education exemption status. We obtained verbal or written consent before all interviews.

We interviewed 20 participants from November 2019 to July 2020, including 4 attending physicians (MD), 2 postgraduate year 2 pediatric residents (Res), 4 pharmacists (PharmD), 7 registered nurses (RN), 1 QI specialist (QIS), 1 administrative staff (Admin), and 1 medical assistant (MA; Table 1). Participants belonged to 3 inpatient adult units: emergency department (ED), neurological surgery (NSY), hospital medicine (HMED); 2 pediatric inpatient units: pediatric intensive care unit (PICU) and transitional care unit (TCU); and 1 outpatient adult unit: division of general internal medicine (GIM). Interviews averaged 16 minutes and 50 seconds. The first author (M.K.C.) interviewed all participants except for ED/PharmD1 (C.B. interviewed); a second author (either C.B. or S.C.) observed the first 17 interviews to supervise interviewing technique. Three broad themes emerged: definition of IP QIPS learning, resident integration into QIPS teams, and factors influencing the IP-CLE for QIPS.

Participants described what IP learning for QIPS means to them, how it occurs, and where it occurs (Table 2). For many, it meant not only learning from and with those from other professions, but also learning more about each other's roles to understand how patient care is affected by different disciplines working together. Learning occurred when QIPS collaboration was triggered by specific patient cases, and some described this as a natural byproduct of patient-centeredness. It also occurred through experiential learning, such as through QI projects. IP learning led to personal growth in one's own field, and an environment of sharing and mutual respect was important to promote this learning.

Participants delineated the benefits that residents and teams gain from resident integration into IP QIPS teams, as well as barriers and facilitators to integration (Table 3).

Residents gained new perspectives important for patient care via integration into IP QIPS teams. IP QIPS activities also allowed residents to connect to members from other professions, which augmented daily clinical practice. “[These activities] enhance our relationships...it's a good mechanism through which to learn from each other and work together...it adds to what we do on a day-to-day basis” (TCU/Res1).

QIPS teams reported benefits from having a resident champion who could disseminate information around QIPS activities to peers more effectively than non-residents.

Participants recognized many barriers to resident integration, which can be categorized as cognitive, structural, and cultural. Several participants alluded to the high cognitive load of some rotations, namely the intensive care unit rotations, that prevented resident participation in activities beyond basic expectations of clinical care because they were mentally taxed and occasionally too overwhelmed by the demanding rotations to meaningfully contribute to QIPS activities.

While rotations of lower acuity of care may be better suited for the addition to the resident workload, the general rotational structure of training programs can be a barrier as well, especially in our large academic hospital where residents frequently switch rotations and training sites. Each rotation is “such a small amount of time to be able to routinely engage with interprofessional colleagues and to see quality evolve over usually what takes years” (HMED/MD1).

Participants also reported cultural barriers to resident participation in QIPS teams. Participants alluded to a hidden curriculum suggesting that QIPS work is not considered a core element of clinical care. As one resident within the pediatric ICU explained, “There's a huddle at the whiteboard every morning that I think they go over things like that [QIPS activities] at. But the residents are told to just keep pre-rounding” (PICU/Res1). This resident was instructed to focus on the expected typical clinical care of “pre-rounding” on patients instead of being encouraged to join QIPS presentations.

Another cultural barrier was the culture of inappropriate assumptions. Residents may hold an inappropriate assumption of self-sufficiency such that they try to take on all patient-related tasks, even those better suited to other professions, because of lack of awareness of others' roles or lack of trust in others. Non-residents may hold an inappropriate assumption that residents lack interest in QIPS and would not want to be offered opportunities to join QIPS efforts.

Participants identified institutional support structures as facilitators of resident involvement. In particular, several mentioned the unit-based leadership teams (UBLT), interdisciplinary leadership teams, developed by the health system, which meet regularly to promote IP QIPS activities that purposefully incorporate residents. Another example is the yearlong UCSF Residents and Fellows Leading Interprofessional Continuous Improvement Teams (REFLECT) program, through which residents and fellows can develop and lead QIPS proposals and can receive monetary incentives for goals achieved.¹⁹  This program prioritizes IP projects.

A second facilitator was patient adverse events, which participants mentioned as motivators for residents to engage in QIPS efforts alongside others also involved in the case. One participant described a central line-associated bloodstream infection that prompted the interprofessional care team to perform a root cause analysis and develop steps to prevent future occurrences.

Another facilitator was frequent communication through structured avenues to residents about existing QIPS projects that they were expected to engage in, which participants mentioned resulted in improved dissemination among the resident cohort. Lastly, one emerging facilitator may be the promotion of QIPS principles within undergraduate medical education (UME) and GME. One participant felt that a value for QI is “ingrained in the education and training of going through medical school and residency,” which may translate into residents internalizing and gaining familiarity with this value (ED/MD1).

Participants spoke about benefits of a positive CLE as well as barriers and facilitators to achieving a positive IP-CLE (Table 4).

Many participants were extremely positive about the improved IP relationships after working on QIPS projects. As relationships grew stronger and “silos [were] broken down” (ED/MD1), team members started to “rely on each other a lot. Not just for work, I feel a lot of us are allowed to deliver moral support too” (HMED/RN1). Strengthened relationships lowered the bar to seek help for other clinic issues and to initiate further collaboration.

In terms of barriers to achieving a positive IP-CLE for QIPS, participants considered lack of quantity of time an important factor. Participants also felt that lack of protected time was a barrier as they struggled to find others to cover their clinical duties during QIPS meetings. These time-related factors can be substantial issues for IP teams as each profession brings a unique set of scheduling challenges.

Another barrier was infrequent and fragmented communication about QIPS project dissemination and updates. As one participant mentioned, “To really get everyone together to see …[QIPS projects] through when you have so many people who are on rotating shifts, to make sure the message comes across to everyone…is a little challenging” (ED/PharmD1). Without purposeful emphasis on communication, it can be difficult to disseminate QIPS efforts and maintain forward momentum of such efforts. The consequences of poor communication can be magnified on IP teams, where each profession brings different expectations and communication styles.

Participants mentioned numerous facilitators of a positive IP-CLE for QIPS. One important facilitator was a culture of respect. Several participants, particularly nurses, mentioned that “mutual respect and... a dismantling of the power hierarchy” (PICU/RN1) is a key facilitator of IP teamwork and a positive learning environment. Mutual respect encourages team members to value each other's opinions and cultivates a value for feedback.

A commitment to learning and sharing was also vital. Participants mentioned structured learning opportunities that the health system offered, such as QI boot camps and workshops, but also that individuals would display commitment through ad hoc interactions where attending physicians reviewed a patient case to discuss future areas for improvement.

Another facilitating activity was regular team meetings. Many echoed the idea that these were “a very great mechanism for us to interact with other disciplines and conduct improvement work…The idea of having all stakeholders…present is extremely powerful” (NSY/MD1). Meetings not only provided opportunities for IP QIPS teams to brainstorm and work with stakeholders, but also were an avenue for education, project dissemination, and escalation of QIPS topics.

Lastly, participants mentioned triggers that motivated them to pursue IP QIPS teamwork. Many mentioned having an intrinsic interest in QIPS. Another trigger was awareness of deficits in patient care quality through review of quality metrics.

We did not find significant differences in the responses of those from different fields of medicine. Additionally, after analysis of 3 outpatient interview transcripts, we did not notice substantial thematic variance compared to inpatient transcriptions, and thus do not believe there to be a meaningful difference between the 2 settings.

We found that there are several facilitators and barriers to successful integration of residents into IP QIPS activities. One notable finding of our study is that cultural issues can act as prominent facilitators or barriers to the IP-CLE for QIPS. A second is that patient unforeseen events can act as a valuable trigger for interprofessional interest in QIPS among residents, faculty, and staff.

The efforts of the Association of American Medical Colleges, the ACGME, and the Health Resources and Services Administration have spurred a national shift in institutional culture that places increased emphasis on the importance of QIPS within UME and GME.^20–22  Institutional initiatives with increased emphasis on QI as a core educational activity in UCSF UME include incorporating QIPS principles early and requiring medical students to engage in novel QI projects; within GME there is the aforementioned REFLECT program as well as creation of UBLTs. These efforts contribute to significant cultural shifts in institutional support for QI work and training and facilitate participants' involvement in IP QIPS teamwork. Existing studies show mixed attitudes toward and mixed impact of this increased emphasis within UME and GME, with more successful efforts requiring special attention to the mode of QIPS education and the educators selected.^23–27  However, through carefully designed curricula that emphasize QIPS in UME and GME and value the contributions of residents in these efforts, institutional initiatives can help change the perception of QI work from being a peripheral component of medical training to a more central component.

Conversely, participants mentioned a cultural practice that inhibits IP QIPS—that of making inappropriate assumptions of others or of one's own professional identity in relation to others. More broadly, several studies show that many residents and medical students can have preconceived assumptions or stereotypes of other professions, particularly in contrast to their own professional identity, which inhibits teamwork as well as the IP-CLE.^28–30  The literature supports the effectiveness of interprofessional education efforts to moderate these inappropriate assumptions^28,29  and, in particular, supports a focus on deeper understanding of each other's roles as a stimulus for interprofessional collaboration,^31,32  which are similarly important facilitators of successful IP QIPS. Within our institution there is an attempt to increase IP education through purposeful selection of interprofessional speakers at GME grand rounds, dedicated interprofessional curriculum opportunities open to all residents, and the emphasis on QIPS teams within the aforementioned REFLECT program to be interprofessional. Other institutions wishing to overcome barriers to IP QIPS may consider that an increased emphasis on IP education may be integral to QIPS training.

Patient unforeseen events can act as a strong and natural trigger for interprofessional interest in QIPS. If reviewed through a format such as morbidity and mortality (M&M) conferences, these adverse events could inspire interprofessional QIPS projects if discussed among interprofessional teams. The conferences could become impromptu IP QI meetings where members share improvement ideas from their unique perspective. While, historically, M&M conferences focused on improvement on the individual technical level as opposed to the systems level,³³  several studies show the potential for interprofessional review of patient adverse events to be an effective springboard for systems-focused improvement and QIPS initiatives.^34–37  Thus, similar health systems may consider reevaluating M&M conferences with these concerns in mind. Moreover, through embedding continuing medical education within QI initiatives, one can further optimize the triggered events as educational opportunities.³⁸  This can be achieved, for example, through connecting triggered events to clinical areas with known quality problems and highlighting evidence-based best practices or requiring post-event deliverables as part of experiential components of the embedded curriculum.³⁸ 

Our study is limited by its observation of the CLE within a single health system, potentially limiting generalizability. A second limitation was our more prominent focus on teams within inpatient units as compared to outpatient units. Lastly, the small number of resident interviews was a limitation as only 2 pediatric residents participated; our method of recruitment of participants through medical directors of key units may have led to this limited recruitment of residents. Many non-resident participants work extensively with residents and were able to relay their perception of the resident experience but not give a firsthand account.

Additional research is needed to capture a wider breadth of firsthand resident perspectives of the IP-CLE to inform areas for improvement within GME programs and to examine variance between different specialty programs.

We defined IP learning and teamwork for QIPS and identified barriers, facilitators, and benefits to a positive IP clinical learning environment for QIPS and resident integration within these environments from resident, faculty, and staff perspectives. Two main findings were the importance of cultural factors affecting the IP-CLE and the strength of patient adverse events to trigger IP QIPS activity.

The authors would like to thank ACGME's Pursuing Excellence Pathway Innovators Collaborative for their partnership in permitting us use of their interview script; the ACGME's Pursuing Excellence in Clinical Learning Environments initiative team at UCSF as well as UCSF Health leadership for their feedback and guidance; and Emily Ko for her administrative efforts for interview scheduling and transcription.