What Residents Want: Perceptions of Learning During a Pediatrics Night Float Rotation

We employed a constructivist grounded theory approach¹⁸  and conducted semistructured interviews with family medicine and pediatrics first postgraduate year trainees (henceforth “interns”) at a tertiary care children’s hospital. Taking an inductive, exploratory approach to the phenomenon of interest, constructivist grounded theory locates research in situational contexts, acknowledging that researchers’ perspectives influence data collection and interpretation.¹⁸  To address inherent subjectivity,¹⁹  we constructed our research team to assure a diversity of perspectives and interpretations. BLT and MD, pediatric hospitalist attending physicians at the study site, brought local contextual knowledge. MR, a pediatric intensivist with qualitative research experience provided perspective from a different institution. WW, a pediatric chief resident who previously served as a night float intern provided a near-peer perspective. A health professions qualitative researcher and internal medicine physician (JLB) and an education scholar and biomedical researcher (DC) brought perspectives outside of pediatrics.

We modeled our interviews after similar studies exploring workplace experiences^20,21  (see Supplemental Information). We elicited participants’ stories of remembered experiences working at night on the pediatric wards, utilizing a modified critical incident technique^22,23  to prompt participants’ recall of situations they thought went well and when they felt nervous or uncertain in providing care as an intern. Additional questions explored participants’ perceptions about curricula and support for learning at night. We neither explicitly defined teaching and learning for participants nor asked participants to distinguish teaching from learning or patient care and learning. MR obtained participants’ consent, conducted all interviews, and deidentified data. A third-party service transcribed the interviews. Using constant comparison,²⁴  the research team reviewed and discussed initial interview transcripts and modified the interview guide. The institutional review board overseeing research at the hospital approved this study.

Three authors (BLT, MR, MD) introduced the study to family medicine and pediatric interns between February 2020 and August 2021. We purposively sampled interns rotating on the general pediatrics inpatient team at the study hospital to explore their experiences working at night. Participants came from 4 outlying family practice and pediatrics programs; interns from these programs work 1 to 3 weeks of nights over the course of the year. We selected interns because we expected their recent assignments to night float would facilitate the recall of detailed experiences. Participation was voluntary, and participants were assured privacy and confidentiality.

A tertiary care children’s hospital in the Northwest served as the study site. At night, the pediatric or family medicine intern works one-on-one with an in-house pediatric attending hospitalist (henceforth “attending”). Typical intern night shift activities are shown in Table 1.

The pediatrics setting is uniquely suited to studying night learning as it is a field with high resident oversight compared with internal medicine,^25,26  thus allowing our research to capture interns’ perceptions of learning in settings of close supervision.

We took an inductive approach to data analysis. We independently developed initial codes for the first 2 transcripts and discussed the findings. We applied these preliminary codes to additional transcripts, refining codes and definitions through discussion until the code book appeared stable. We used Dedoose 9.0.46 for Mac (SocioCultural Research Consultants, LLC, Manhattan Beach, CA) to code and analyze the data. Two researchers applied the code book to each transcript. The coding of the tenth transcript was consistent with previous findings. MR conducted 2 additional interviews. No new themes were identified, so we deemed the sample sufficient for the purpose of the study.²⁷  We reviewed and discussed code reports to develop our themes, comparing and contrasting related reports to enhance and challenge interpretations, and develop a conceptual interpretation of thematic relationships.²⁸  To finalize our results, we discussed our conceptual model with 2 consultants familiar with our purpose who helped to refine our thematic interpretation.

We identified 5 themes describing what interns wanted to support their work and learning at night: (1) opportunity to learn through work, (2) informal case discussions, (3) supported independence, (4) orientation, and (5) feedback. The relationship among these themes is shown in Fig 1. Opportunities to initiate patient care assessments independently are the central feature, with attending physicians providing support through informal case-based teaching and ready availability. Interns benefited from orientation and feedback that was not exclusively before or after their nightshifts, as shown by overlapping circles. We elaborate later and provide data excerpts in Table 2, using representative quotations from interviews. We identify individual participants by number and training specialty (“P” for pediatrics and “FM” for family medicine). Excerpts are edited slightly to improve readability.

Participant characteristics are summarized in Table 3. Twelve participants completed interviews, 10 between February and June of the academic year 2020 to 2021, suggesting most of our data came from experienced interns. To understand if our results were related to the experience level of interns, we separately reviewed the 2 interviews conducted with interns in the early academic year 2021 to 2022. Thematic review of these 2 transcripts was similar to the previous 10, suggesting that the stage of training did not substantially influence the results.

The core of nighttime learning occurred through conducting direct patient care (see Table 2). Evaluating undifferentiated patients, having more time to think, fewer competing demands, and monitoring patients over time characterized their experiential learning. Being the first to evaluate a patient was particularly useful “when a kid is coming from an outside hospital” because “they’re more fresh and I get to think through it from the beginning” [P6]. In addition, less interference from typical daytime activities (rounds, documentation, care coordination) facilitated learning. A participant said, “… I have a lot more time at night to come up with a good differential and actually look things up” [P4]. Another participant echoed the gratifying sentiment of making a diagnosis at night, saying “I got to the bottom of [the diagnosis] by doing the interview with the mother and it was just really cool. I felt really good about myself” [FM9].

To support their learning through work, participants generally expressed a preference for informal, one-on-one interactive learning with attending physicians while downplaying the utility of structured didactics (see Table 2). One participant stated that the exchange of formal didactics for “more one-on-one time with the attending” was “a fair trade off” [FM9]. Another participant echoed the value of “informal and…on the fly” [P6] conversations, especially when discussing new admissions where the attending was “throwing in little teaching points” [P6]. Participants repeatedly cited that “learning tidbits” [P7] were beneficial.

Participants did occasionally identify the value of formal learning at nighttime but expressed no desire for formal didactics or required modules while working at night. Teaching, they suggested, should be brief, focused, and timely and support their growing independence. Delivery should be “30 minutes to an hour before or after night rounds…” [P4]. Desired content would be “some of those bread-and-butter lectures” [FM8] or just “one general topic like fluid management” [P4]. After reviewing key topics with attending physicians, participants expressed increased confidence in their abilities to conduct nightwork responsibilities. Participants recognized the link between functioning “a little bit more autonomously” at night and the value of “know[ing] exactly what to do” [FM8].

Although participants valued being the first to evaluate patients, they also wanted access to and support from attending physicians when they felt uncertain (see Table 2). One participant described that most attending physicians stated, “‘Don’t hesitate to call me for anything’…so I didn’t feel bad about calling them if I was concerned about a patient, which is always nice” [P4]. Multiple participants appreciated having access to attending physicians, with 1 expressing the benefit of having “the expectation … in place” to call attending physicians when “questions came up that I felt uncomfortable with” [P6].

Situations with higher medical acuity highlighted participants’ desire for ready access to attending support. Such situations were sometimes “terrifying,” amplifying the need for immediate support. Another participant recounted the “nerve wracking experience” of being the first to arrive at a code, and when asked what could have made the situation go better, replied, “if I would have shown up at the same time as the attending” [P12].

Interns desired orientation to aspects of the nighttime care environment to help manage their stress and streamline their clinical learning (see Table 2). Examples included orientation to clinical care protocols, resources for responding to emergency situations, and self-care advice.

Participants frequently identified protocols for caring for patients with behavioral problems as high-yield orientation topics. Participants felt it would be helpful to have “a few targeted learning things. It could be a video or a printout on something like an eating disorder” [FM2] and it “would be really nice if there was…a central repository of information about those protocols” [P5].

Participants also wanted anticipatory guidance for more urgent situations. A participant said, “I had no idea where the PICU even was” [P4]. Another said, “…at night we don’t do a workup. We just do what we need to do to save people’s life, so, resuscitation stuff….more review of the PALS would be good” [FM11]. Finally, participants desired orientation to self-care at night such as “advice about ordering food” [FM11] or “what time do you eat during the night? …What’s the sleep schedule?” [P12].

In addition to support from attending physicians in medical decision-making, participants valued feedback (see Table 2). Participants received 3 kinds of feedback: specific actionable feedback from attending physicians, nonspecific positive feedback from attending physicians, and feedback from clinical outcomes. Examples of specific feedback were uncommon yet useful. Sometimes specific feedback was corrective, which at times felt punitive, with one participant recounting they “got chided by one of the attendings” [FM2] for a decision.

Conversely, participants provided multiple examples of nonspecific positive feedback, such as “the attending I was working with commented that I had done a good job taking care of everyone” [FM2] or “one attending said to me, ‘You’re a doctor. You’re smart. You’re a doctor. You can make decisions’” [FM1]. Participants generally had positive emotional responses to receiving such feedback. Finally, although not generally labeled as feedback, participants provided numerous examples of clinical feedback²⁹  as they monitored their decisions and the evolution of a patient, often at the bedside.

We set out to investigate concerns that night float has low educational value to inform our development of a night curriculum aligned with supporting interns’ learning needs. Our exploratory approach identified distinctions between teaching and learning. We discovered that interns do not want a formal didactic teaching curriculum at night. Rather, they want support to optimize workplace learning: opportunities to initiate patient care assessments independently, informal teaching arising from patient care, reassurance that support from supervisors is readily available, orientation to resources, and feedback on their decision-making.

Previous research addressing concerns about low educational value in the nighttime setting included several attempts to implement structured curricula.^2–7  Our results suggest that residents perceive rich learning to occur through caring for patients at night, calling into question the necessity of formalized curricula modeled after classroom learning traditions. Distinguishing informal learning from formal learning provides an alternate lens for unraveling the nature of nighttime learning.³⁰  When learning arises from informal, unstructured experiences such as those described in our participants’ stories, interns likely have greater individual agency for determining their learning agendas. Yet, informal learning may be invisible and difficult to recognize.³⁰  Educators responsible for developing and implementing nighttime curricula may need to see structured teaching conferences as visible evidence of their work, making it difficult to shift to supporting informal learning.

Results of our story-telling interview approach suggest that residents’ informal learning at night is unplanned and responsive to patient care and supported through case-based discussions with supervisors. In addition to being informal, such learning is fundamentally experiential and situated in the workplace.³¹  Formalized educational practices dominate in all settings, which may, in turn, lead educators to overlook the importance of such workplace learning.³¹  When educators fail to consider workplace learning, formal curricula may become a reflexive approach for addressing concerns about insufficient educational value.

We acknowledge the limitations of our study. We interviewed interns rotating at a single pediatric hospital with a unique intern-only nighttime rotation structure, which may have magnified their desire for aspects of orientation. The in-house attending model at the study hospital likely influenced participants’ perceptions. Studies involving other supervision models may find differences. Consistent with the field of pediatrics, most of our participants identified as female. We collected data during a global pandemic, which may have influenced the hospitalized patient population (eg, higher number of patients with behavioral concerns). Participants’ retrospective storytelling is subject to recall bias, and direct observation may have yielded additional findings. Finally, our last 2 interviews, conducted to assure data sufficiency, spanned a new academic year, although we did not observe new findings related to our research purpose.

Our findings underscore the importance of actively supporting a model of workplace learning at night. This model aligns with conceptions of learning-as-participation^31,32  more so than learning-as-acquisition typical of conference-style teaching sessions.³³  In considering learning-as-participation, what should a night float curriculum look like? We recommend shifting away from traditional views of curricula as formal, didactic teaching sessions to instead emphasize learning through supported participation in patient care activities occurring in the workplace. Such educational programs should focus on:

• Allowing interns initial independent assessment of undifferentiated patients.

• Ensuring attending availability when interns request support.

• Orienting interns to preexisting patient care protocols.

• Giving guidance for self-care (access to meals, orientation to the physical environment, where to find resources).

• Delivering specific and timely feedback.

Over the past decade, others have proposed supporting nighttime learning through structured curricula using a formal teaching approach.^1–7  Our findings suggest that informal workplace learning is already occurring at night. Attempts to implement formal curricula may be at odds with the informal and implicit mechanisms of nighttime learning. Hence, we recommend a curricular frameshift to support resident learning at night to emphasize informal teaching responsive to learning needs that arise as part of patient care delivery, integrating but not emphasizing formal didactics when necessary.

The authors wish to thank Dr Jennifer LeTourneau of Legacy Graduate Medical Education and Dr Shaban Demirel of Legacy Research Institute for their assistance in conceptualization of this study.