A Qualitative Study of Resident Experiences With Clinical Event Debriefing

We used a focus group methodology with an application of the nominal group technique to explore participants’ rich experiences.

This study took place from November to December 2020 at a tertiary-care children’s hospital with 575 inpatient beds. The pediatrics residency program consists of ∼160 residents ranging from postgraduate year (PGY) 1 through 4 in both categorical and combined programs. Inpatient rotations range between 2 and 4 weeks, spanning multiple hospital units and services. Beyond informal CED experience, each resident participates in CED as part of required simulation curricula beginning in PGY-1. Led by chief residents or other program leaders, these curricula generally target residents but occasionally involve other roles (eg, nurses). Residents may also participate in hospital-wide CED procedures after various safety events.²⁶  Managed locally by nurse and physician leaders, these procedures vary by unit and event but are generally interprofessional (ie, involving multiple professional roles), including both immediate (“hot”) and delayed (“cold”) CED. An estimated ∼ 50 CEDs take place monthly across the hospital. Residency program leaders (eg, chief residents) are consistent participants in these proceedings, but resident involvement is variable.

Residents from all levels and programs were eligible to participate in this voluntary study. Recent CED participation was not required. Recruitment strategies included the use of verbal and written study advertisements and use of “snowball” recruiting,²⁷  wherein participants were encouraged to refer peers for potential enrollment. Each focus group was organized on the basis of participant availability, consisting of residents from different training levels. All participants provided verbal consent before starting the focus groups.

Informed by our previous study,²³  we hypothesized that residents’ CED practice could be affected by their perceptions of the work environment, and, in particular, by psychological safety. We developed a focus group moderator guide using an adapted version of Edmondson’s conceptual model,^24,25  which posits that psychological safety is an antecedent for team learning behaviors and that it is affected by team characteristics and other structures (Fig 1).

Investigators (S.G., K.K.) with training in qualitative methods facilitated 60 minute focus groups via the BlueJeans video conferencing platform (BlueJeans Network, San Jose, CA). Facilitators used a moderator guide developed by the study team based on the objectives, a review of the literature on both CED and organizational behavior, and the conceptual model (Supplemental Information).

We defined CED as a facilitated discussion of a significant event that could take place in the hours to days after an event (hot CED) or in the days to weeks after an event (cold CED). Facilitators stated that CED may include emotional processing and provided examples of events that might prompt CED (eg, resuscitations).

In addition to asking residents about their CED experience, we asked about their perceptions of the hospital environment, which we hypothesized could affect CED. For example, given that many institutional CED procedures involve unit-based practice with interprofessional teams, we asked residents how (if at all) their role impacted their experience of unit culture.

The guide also included an application of the nominal group technique,²⁸  wherein we presented residents with a list of CED practice recommendations generated from our previous interview study with other clinicians (eg, nurses, advanced practice providers).²³  Among other recommendations, this list proposed that CED should be timely and use open communication. We asked residents to collectively review, refine, and prioritize this list (Supplemental Information). Facilitators determined consensus when a majority of participants verbally agreed with proposed changes and no additional input was proposed.

After each focus group, we administered an anonymous web-based survey addressing participant demographics and previous CED experience, which we reasoned could affect the interpretability of study results (Supplemental Information). The hospital’s institutional review board approved this study.

Focus groups were digitally recorded, deidentified, and transcribed by a third-party transcription service (ADA Transcription, Mt. Holly, NJ). Deidentified transcripts were entered into NVivo 12 Plus (QSR International, Burlington, MA) for coding and analysis after 4 focus groups were completed. We used a modified grounded theory approach to the analysis.²⁹  We identified an a priori set of codes derived from key concepts in the interview guide. Then 2 investigators (S.G., K.K.) performed an iterative close reading of one transcript in its entirety to identify codes represented in the data. Both sets of codes were defined and decision rules for their use were included in a codebook. Coders worked together to iteratively code 2 (50%) transcripts. Code definitions were adjusted until no new codes or themes were generated. After the codebook was refined, one coder (SG) analyzed the remaining transcripts (n = 2) in a similar fashion to ensure consistent application of codes. Other investigators (M.G., L.G.) then reviewed the coded transcripts to ensure codes were applied appropriately and no new themes were identified. Additional data collection was deferred given thematic saturation.²⁹  Of note, all data were deidentified before analysis; quotes were not attributed to individual participants or training levels.

Techniques to ensure trustworthiness included facilitation and coding by nonclinical investigators without residency program affiliation (S.G., K.K.), triangulation of resident perspectives across data collection methods via nominal group technique, and post hoc member checking with respondents.³⁰ 

We conducted 4 focus groups with 26 residents, including 7 (27%) PGY-1s, 7 (27%) PGY-2s, 11 (42%) PGY-3s, and 1 (4%) PGY-4. Groups ranged from 5 to 9 participants of mixed training levels. Twenty-five (96%) participants completed the follow-up survey (Table 1).

Our analysis identified several major themes related to resident CED experiences, including their mental model of CED and the barriers and facilitators they perceive to resident involvement (Table 2). Unless specified, we use “involvement” to describe both participation (ie, attendance) and engagement (ie, engaging in reflective discussion) in CED.

Residents generally described CED along 2 dimensions: the participant mix and the primary goals and objectives. They often described these dimensions as interrelated.

Residents primarily categorized CED on the basis of the participant mix. They described 2 scenarios, interprofessional and role-specific CED, which could be performed in a team or a one-on-one setting.

Residents generally described interprofessional CED as a valuable team-building opportunity that promotes cross-disciplinary learning. However, although they regarded it as a positive experience, residents acknowledged infrequent participation in interprofessional CED.

Instead, they described greater experience with CED among physicians, often in a one-on-one setting. They emphasized that these interactions are rarely designated as “debriefing,” yet they viewed them in this light. Residents highlighted that one-on-one CED with a senior physician (eg, a fellow or attending physician) offers an important opportunity for both clinical learning and emotional processing. Many residents also praised one-on-one CED with peers, particularly during clinical rotations that require them to function independently.

Residents also categorized CED on the basis of their perceptions of the goals and objectives, such as emotionally processing an event, appraising individual and/or team performance, and reflecting on medical management decisions to facilitate clinical learning. Across all groups, residents articulated the importance of establishing clear goals in CED because they felt these could influence how participants engage in reflective discussion.

Whereas facilitators explicitly asked about hot and cold CEDs, residents generally felt that timing was not a critical factor affecting their CED experience. Instead, residents focused on the interplay between CED timing and goals and objectives. For example, they cited the timing of hot CED as a factor affecting practice goals, as residents often have competing responsibilities in the aftermath of an event that can preclude emotional processing.

Residents identified several barriers to their involvement in CED. Although some derived from logistical factors, the most salient barriers related to unique features of the resident role.

Many residents described a feeling of ambiguity around who should initiate CED and which events should be debriefed as a common barrier to participating in CED. Many residents also reported experience with a “bystander effect”³¹  in which diffused accountability for initiating CED had culminated in a failure to debrief. Some residents attributed practice ambiguity to attending physicians, who may variably defer to other roles to initiate CED or who may initiate themselves. However, respondents speculated that learners and experienced physicians have different thresholds for considering an event “significant” enough to warrant CED. Accordingly, they felt that relying on attending physicians to initiate CED could lead to inconsistent practice.

Many residents described competing responsibilities as another major barrier to CED participation. They described that urgent clinical responsibilities after an event can frequently preclude their participation in hot CEDs. They named similar issues with participating in cold CEDs, which may be scheduled when residents are busy with new clinical rotations.

Most respondents noted that limited availability of backup resources specific to residents can be a barrier to CED participation, particularly when working overnight when fewer physicians are present to help offset the workload. During these periods, residents described feeling unable to step away from clinical duties to participate in CED. Several residents contrasted these resources against perceived supports in place for nurses and other interprofessional team members, who residents felt could more consistently lean on colleagues and supervisors for support, particularly overnight.

Overwhelmingly, residents highlighted the rotational nature of their clinical requirements as a barrier to CED participation, particularly among interprofessional teams. They consistently referred to themselves as “guests” and “outsiders” in describing their work among unit-based interprofessional teams during inpatient rotations. They lamented the challenge of having to quickly and continually integrate into new clinical environments with core prescribed teams of clinicians, such as unit nursing staff.

Consequently, many residents described feeling overlooked for interprofessional CEDs. They described instances in which they played a substantial role in an event yet were unaware of the interprofessional CED that subsequently took place. Some residents attributed this lack of awareness to logistical factors (eg, a cold CED occurred after the completion of their rotation). However, other residents speculated that unit organizers had intentionally excluded them because they were relative outsiders to the core interprofessional team.

Residents also cited the rotational nature of their role as a significant barrier to engaging in reflective discussion during interprofessional CEDs. They lamented that each clinical rotation requires them to quickly establish a shared sense of trust with a new interprofessional team, and this trust may not be established before a CED, leaving residents hesitant to speak up during interprofessional CEDs. Several respondents also worried that interprofessional team members could assume residents are less invested in unit issues because of their temporary duties.

Many residents noted that their dual role as both “the doctor” and as a trainee made it challenging to engage in interprofessional CED. They acknowledged that being a learner made it difficult to command authority as a physician, and they worried that expressing emotional vulnerability and/or self-criticism in a CED could threaten trust in their abilities.

Several residents also described a fear of being reprimanded during CED for perceived mistakes. They worried that nonphysician members of the team, who may be less familiar with the resident role, could be more apt to blame residents for adverse event outcomes.

Residents identified several facilitators of their involvement in CED. Often, these are related to individual factors such as perceived comfort working among interprofessional groups.

Uniformly, residents cited increased comfort with participating and engaging in all types of CED as they advanced in their training. They described that this comfort stemmed from an increased familiarity and confidence in working with senior staff (ie, physicians and nurses), increased comfort with interprofessional teaming, particularly among unit teams with whom they had previously worked, and felt an increase in respect from all team members because of their seniority. Respondents cited that these factors allowed them to advocate for CED initiation more effectively, participate in CED more often, and engage in reflective discussion more easily.

Many residents noted that a desire to learn through the reflective discussion of their individual and team performance often outweighs perceived barriers to participating and engaging in CED. They described many instances in which the educational value of CED had motivated their proactive involvement in CED despite logistical and situational barriers such as competing tasks and unfamiliar teams, respectively.

Most residents agreed that they were more likely to participate and engage in CED if organizers explicitly invited their involvement. Indeed, despite feeling like guests among unit-based interprofessional teams, respondents expressed that being intentionally included in CED makes residents feel valued, which can mitigate feelings of alienation.

During the nominal group exercise, residents generally agreed with previous CED recommendations but also proposed additional considerations (Table 3). Three of the 4 groups independently named clear goal setting as a key recommendation for CED. When asked to prioritize the composite list of recommendations, 2 groups prioritized participation from clinical staff caring for the patient surrounding the event and the establishment of a nonjudgmental environment (ie, through the use of a normalizing statement). Other priorities included timeliness (eg, minimizing delays for hot or cold CEDs) and the use of a private location.

With this study, we aimed to characterize how pediatric residents experience CED, with a specific focus on the barriers and facilitators of their involvement. To our knowledge, this is the first qualitative study to characterize how residents experience CED. Several of the logistical barriers we identified are consistent with previous survey studies.^18,20,32  However, we also identified novel barriers to CED that are derived from unique features of the resident role and are salient for interprofessional CED. For example, residents continually highlighted the transience of their duties in relation to unit-based interprofessional teams that work together with greater consistency and which, thus, represent a relatively stable team. In contrast, residents described themselves as outsiders, who may, at times, feel unwelcome or uncomfortable participating and engaging in CED with unit-based teams.

These findings echo the organizational behavior literature, which has revealed that newly formed teams function differently than “stable” teams.^33,34  In particular, team stability has a strong, direct relationship with team learning activities,³⁵  which require established trust and understanding among team members. Yet, team stability is rare in health care, which relies heavily on dynamic, ad hoc team formation.³⁶  Although unit-based staffing can allow for relative team stability for roles specific to the unit (eg, nurses), residents must rotate across services and locations to meet graduate medical education (GME) program requirements. This forces them to quickly integrate and establish trust among a team before moving on, leading residents to feel like guests during any given clinical rotation.

Our findings also validated the study hypothesis that psychological safety could affect resident CED practice, a finding supported by Edmondson’s conceptual model.²⁴  Participants highlighted that their involvement in interprofessional CED is sometimes limited because of their feeling unwelcome and/or fearful of judgment and other negative repercussions. Yet, these same dynamics were felt to be less salient for senior residents, perhaps because they have greater confidence in their skills and greater experience with interprofessional teamwork. Senior residents may, thus, feel more comfortable participating and engaging in CED with a relatively new interprofessional team. In this way, our study builds on literature from Edmondson and others that reveals psychological safety is necessary for team learning activities such as CED²⁴  and that it is positively associated with professional status.³⁷ 

Notably, the residents in this study clearly valued CED among peers and other physicians. Although GME programs should continue to support residents in these activities, it is critical that GME programs prepare residents for the realities of an interprofessional practice environment by supporting resident involvement in interprofessional CED, which holds enormous opportunity for cross-disciplinary learning. Hospital staff should also consider how to promote resident involvement in unit-based safety and quality activities more generally because residents (and other rotational staff) share an important perspective that may be underrepresented.²⁶ 

On the basis of our findings, we propose several recommendations for CED practice that GME program facilitators and hospital staff should consider, including the provision of interprofessional skills training for residents and staff, such as through team-based simulation training (Table 4). Among residents, in particular, interprofessional simulation may help advance skills and experience working across disciplines under stress. Future studies will assess the impact of these proposed interventions on resident CED practice, as well as on resident wellbeing.

Our study has several limitations. This is a single-center study that may have limited generalizability for other hospitals and trainee populations. There is also potential for both recall and selection bias, as recent CED participation was not required for enrollment, and we used snowball recruitment methods. As such, residents who enrolled may have felt more opinionated and/or comfortable voicing their thoughts and may not represent the breadth of resident experiences. Some resident perspectives could also have been limited by insufficient experience with interprofessional or other forms of CED. Additionally, our focus group guide did not explore how team structure and/or CED participant mix (eg, the presence of an attending physician or senior resident) affects residents’ psychological safety. We also did not perform an iterative analysis, and although 2 investigators cocoded 50% of transcripts to generate a codebook, 1 investigator coded the remaining 2 transcripts independently, which could contribute to Type I and/or II errors. However, other investigators reviewed the coded transcripts and felt the themes were captured appropriately, with no new themes identified. Regrettably, because of time constraints, we did not explore resident perceptions of how CED leads to practice change. Additionally, we used an adapted nominal group exercise to facilitate discussion, but we did not generate consensus across focus groups, the traditional outcome when employing this method. Finally, we did not account for resident demographics (eg, level of training) or previous experience with CED, which could have impacted resident perspectives. Although our follow-up survey addressed these factors, it was designed to simply describe participants and did not offer contextual data for our results.

Overall, pediatric residents in this study highlighted CED as an important opportunity to learn from and process events, but they identified multiple barriers to involvement. Although some barriers were logistical, the most salient barriers derived from the unique features of their role, such as the transience of their duties. Residents also identified several facilitators of their involvement in CED, including advancing professional experience. Our findings propose several CED practice recommendations for GME programs and hospitals to consider promoting resident involvement in CED. Future studies will assess the impact of proposed practice recommendations.

The authors wish to thank Ms Eda Akpek and Ms Rebecca Neergaard for their feedback in designing this study. The authors also thank Dr Joseph St. Geme and the Department of Pediatrics at the Children’s Hospital of Philadelphia for supporting this project.