Understanding the Influence of the Junior Attending Role on Transition to Practice: A Qualitative Study

The transition from residency to independent practice is a daunting adjustment for physicians.^1–3  New graduates may feel adequately prepared for the medical expert components of independent practice, but are often apprehensive about other domains, including leadership, collaboration, financial management, work-life balance, conflict resolution, and in academic centers, supervision of trainees.^1,3-8  Insufficient preparation is linked to burnout, which negatively affects patient care.^7,9  In the Competence by Design curriculum in Canadian postgraduate residency education,¹⁰  the Transition to Practice (TTP) stage spans the end of residency before beginning independent practice. Despite the importance of this juncture, how to optimize smooth transitions and prepare residents for all aspects of independent practice is not well-characterized. The body of literature exploring TTP is limited. The traditional cognitive apprenticeship model with graduated responsibility and tacit expertise development through clinical learning, though ubiquitous, has little objective support from the literature.¹¹  Many recommendations propose mentorship, graded responsibility, feedback, and longitudinal assessment to determine competence and promote independence.¹² 

Within many residency programs, one educational model that aims to facilitate TTP is the Junior Attending (JA) role, wherein a very senior resident assumes the roles and responsibilities of an attending physician, with supervision. In Canada, the Royal College of Physicians and Surgeons requires several blocks of JA in the training experiences of many of its specialties during their final years of training. Similar models have been employed elsewhere; for example, a neurosurgical program in the United States found the JA role to be a promising experience in the TTP.¹³ 

While the JA model is a potentially useful experience, minimal evidence exists on achievable learning objectives, appropriate expectations for supervising attendings, impacts on junior learner experiences, and whether it is effective in fulfilling its goals. Within our institution, informal feedback suggests substantial variation in implementation of the role. Therefore, we aimed to understand the potential of the JA model, to determine its usefulness in the TTP stage, and to develop a framework to guide both learners and supervising attendings on how it may best facilitate achievement of educational goals.

We performed a qualitative collective case study informed by a constructivist grounded theory (CGT) analytical approach.^14,15  The central premise of a CGT approach to analysis is that knowledge is co-constructed by participants and researchers.

We conducted our study across 6 academic hospitals affiliated with the Department of Medicine at the University of Toronto. We selected 2 cases to enhance our understanding through comparison and contrast: in one, the Most Responsible Physician (MRP) JA role was situated in experiences as a sub-specialty general internal medicine (GIM) resident leading an in-patient internal medicine team (Figure). In the other, the Consulting JA role was situated in experiences as a sub-specialty resident supervising infectious diseases (ID) or rheumatology team consultations and follow-ups to admitted inpatients. We intended each case to illuminate unique aspects of how the JA model can impact TTP.¹⁶ 

We sampled purposively across the 2 cases, including 3 groups of participants: (1) faculty who supervise residents in the JA role; (2) residents who have done the JA role in the last 5 years (to limit the effects of historical differences in work volumes and educational settings on our findings); and (3) faculty and residents who have not participated in either aspect of the JA role.

We snowball sampled by asking participants to suggest individuals who they believed may have convergent or divergent views regarding the JA role. Sampling continued until saturation of meaning was reached and new data did not inform or challenge our framework.¹⁷  Because approximately 5 to 8 “sampling units” are usually sufficient for samples that are relatively homogeneous,¹⁸  we anticipated the need to interview approximately 15 to 20 individuals representative of the 3 participant groups. A trained research assistant (T.K.) or one of the co-principal authors (R.D.Y., P.E.W.) conducted interviews using a semi-structured interview guide (provided as online supplementary data). Interview guides were iteratively adapted concurrently with data collection and analysis.¹⁹  We conducted 20 interviews between September 2017 and May 2020: 9 faculty JA supervisors (2 rheumatology and 7 GIM), 8 resident JAs (2 ID, 2 rheumatology, and 4 GIM), and 3 without any experience of the JA role.

Interviews were audio-recorded, transcribed verbatim, anonymized, then analyzed using CGT.²⁰  NVivo 12.6.0 (QSR International, Melbourne, Australia) for Mac OS was used for line-by-line coding. Three team members (R.D.Y., P.E.W., T.K.) individually read and inductively analyzed the transcripts for emergent themes during the open coding process. The remainder of the team reviewed quotes and selected transcripts. We met regularly during and after data collection to discuss themes, to iteratively adjust the interview guide, and to evolve and establish the coding structure. We stopped performing interviews when they neither challenged nor added to any component of our framework.

Our team included past and current program directors in internal medicine and its subspecialties (W.L.G., L.S., H.M.B.), as well as physicians who had supervised (W.L.G., L.S., H.M.B.) and experienced the JA role (R.D.Y., P.E.W.), and an education scientist with expertise in qualitative methodology whose research area focuses on the development of expertise (M.M.). This afforded multiple perspectives on the process and analysis and ensured multiple iterative opportunities to challenge, refine, and elaborate the developed framework. We were mindful of our own experiences, and this added a dimension of reflexivity to our analysis.

This study received approval from the University of Toronto Research Ethics Board.

We identified 3 major themes (see Table 1 for themes, subthemes, and core elements of each; see Table 2 for selected exemplar quotes for each theme, as numbered below): (1) the JA role facilitated a balance between independence and supervisory feedback; (2) the relationship between the supervising faculty and the JA was a central element of a successful JA experience; and (3) within a strong supervisory relationship, the JA is legitimized as a functional member of the health care team. Saturation was reached after the 18th interview.

Participants perceived that the purpose of the JA role is to give senior residents a nearly autonomous experience with a supervisory “backstop” in the form of the supervising attending. Consistently, it was considered important that the JA functioned as though they were the attending physician when interacting with the clinical team, patients, and their families. In the absence of other team members, touchpoints with their supervisor allowed for review of patients and engagement in teaching activities (Table 2, Quote 1).

Participants felt it was important for the JA to be recognized as the team leader by all stakeholders (trainees, other members of the health care team, patients, and families) and felt that the presence of the supervising attending during rounds could undermine the autonomy of the JA (Table 2, Quote 2).

Participants also reported that the JA experience could ease the transition from residency to independent practice by allowing the JA to make independent clinical decisions with the opportunity to reflect on their implications with the supervising faculty. Participants noted parallels between progressive independence and confidence in dealing with uncertainty, which some equated with the development of expertise (Table 2, Quote 3).

While helpful, most participants did not feel that a JA experience was necessary for all physicians transitioning to independent practice. Many noted value in those transitioning to an academic medical practice so that JAs could familiarize themselves with issues specific to teaching hospitals (clinical supervision, recognition of learners in academic difficulty, and balancing research with clinical demands).

The JA experience differed slightly when implemented within the GIM MRP setting compared to within the consultant services. In contrast to GIM, where non-medical expert skills were considered the primary focus of JA learning, the focus for ID and rheumatology JAs leaned more toward the development of medical expertise. Residents in subspecialty GIM had completed 3 prior years of internal medicine training; whereas ID and rheumatology residents, who also had 3 years of internal medicine training, had a shorter duration of experience managing the challenging clinical cases within their areas of practice and were seeking to build content knowledge (Table 2, Quotes 4 and 5).

It was clear that providing the JA with appropriate autonomy with feedback was important for development of confidence in achieving independence. This relied heavily on the relationship between the JA and their supervisor.

Participants identified several key enablers in this relationship, including clear role definition through contracting discussions, a priori understanding of the goals and objectives of the experience, mutual awareness of the individual skills brought to the relationship, trust, and an acknowledgement of potential tensions associated with the relationship.

Trust was central to a positive experience because it enabled decision-making autonomy for the JA with minimal interference from the supervising faculty. Conversely, the absence of trust between the supervising attending and the JA was associated with negative experiences that could have significant repercussions for the trainee (Table 2, Quote 6).

Although not always possible, participants found that a pre-existing relationship between the JA and supervising attending in earlier stages of residency provided both parties with knowledge of each other's skill sets.

Unearned or blind trust was also associated with negative outcomes of the experience (Table 2, Quote 7).

The supervisory style of the supervising attending was most responsible for tensions in the supervising faculty-JA relationship. In general, excessive “hands-on” supervision, or micromanagement, was associated with negative experiences (Table 2, Quote 8).

Participants indicated that there was a spectrum of supervisory styles from micromanagement to complete disengagement from the supervisor and that it required negotiation and individualization on the part of supervising faculty-JA pairing. Supervision styles on either extreme of this spectrum were less productive (Table 2, Quote 9).

Participants suggested that early-career faculty were not ideal candidates for supervising JAs because they were still developing their own clinical identities and might be less comfortable in providing the trainee with the autonomy required for a successful experience (Table 2, Quote 10).

Lack of accessibility of the supervising faculty also provoked tension in the relationship. Despite the need for autonomy, participants recognized the need for potential swift intervention from the supervising faculty (eg, for a deteriorating patient or a challenging communication issue). Therefore, having the supervisor accessible was considered important (Table 2, Quote 11).

Participants indicated the future-oriented intent of the JA experience as preparation for independent practice. Despite this, some participants acknowledged the external perception of having a JA as a “perk” to the supervising faculty through having most direct patient care assumed by the JA. There was a perception that the assignment of JAs to supervising faculty was treated as a form of currency, recognizing that this could help fill vacancies in the schedule or allow faculty to concurrently attend to nonclinical or personal duties. However, many participants acknowledged that appropriate supervision and providing a valuable learning experience to the JA requires the same, if not more, time and effort, in comparison to attending without a JA (Table 2, Quote 12).

We further identified that a strong relationship between the supervising faculty and JAs was key to ensuring a successful experience built on appropriate trust, contracting of expectations, and a supervisory style that allowed for independence coupled with support.

In addition to the supervising faculty-JA relationship, participants described important interactions between the JA and other members of the health care team, including other learners, as well as patients and their families.

It was felt that the JA should be the designated point of contact for patients and their families to legitimize the role, promote authority, and foster independence. This was also applicable to communication with other members of the health care team. It was felt that there could be circumstances when involvement of the supervising faculty should be proactive, typically in making high-risk treatment decisions or in the event of conflict between medical services (Table 2, Quotes 13 and 14).

The impact of the JA on the roles and functions of other learners on the team was less clear. The role of the JA, including team leadership, can overlap with the roles and responsibilities of the second- or third-year (senior) resident on the team. It was suggested that there should be sufficient distance in training between the JA and the senior resident whenever possible (eg, a fourth-year JA should not be paired with a third-year senior resident on the same clinical team). Participants also felt that learners could be disappointed with limited direct contact with the supervising faculty due to the presence of a JA and its potential impact on their assessment, mentoring, career counselling opportunities, or obtaining a letter of reference. Some participants described a hybrid model for teaching and evaluation, so that the supervising faculty could participate in teaching activities and independently evaluate the other learners (Table 2, Quote 15).

Another potential negative impact of the JA experience is that early-career faculty might not receive teaching evaluations that are required for continuing faculty appointment and promotion (Table 2, Quote 16).

Finally, legitimizing the JA as a team leader, often while introducing themselves and their roles in interactions with learners, other members of the interdisciplinary health care team, and patients, was associated with positive experiences. Efforts to anticipate and mitigate challenges with learner evaluations and reference letters favorably influenced the experience.

Our study found that the JA role offers promise in supporting the transition to independent practice, particularly in non-medical domains and in managing clinical uncertainty, and it is optimized when the relationship between the JA and the supervising faculty preserves the autonomy of the JA and legitimizes their status as team leader. While others have described such a dynamic as misrepresentative,²¹  this can be mitigated through role acknowledgement to patients, families, and other team members and adequate supervision.

While there was a significant focus on development of medical expertise for consultant JAs, the MRP-based JAs in contrast focused on mastery of non-medical expert domains. From a reflexivity standpoint, we found this to be surprising. All but one author who had either undergone or supervised an MRP JA perceived the experience to facilitate acquisition of at least some medical expertise. The contrast in acquisition of content expertise between MRP-based vs consultant JA pairings may be attributable to the longer prior relevant clinical experience of residents in an MRP versus Consultant role before their JA experience.

It is perhaps not surprising that the participants associated confidence and autonomy with development of medical expertise and dealing with uncertainty. Though the traditional view of development of expertise focuses more on successful diagnostic reasoning or surgical skill,²²  tacit knowledge of how to manage challenging clinical situations may be attainable only through direct practice.^23,24  Such practice may be impossible to simulate at earlier levels of training due to inadequate content knowledge or underdeveloped heuristics, making it ideally suited to JA experiences that are situated toward the end of residency training and immediately prior to beginning practice.

Westerman et al⁴  studied the transition from residency to independent practice, noting that “uncharted territory” within the attending role (supervision, final responsibility, technical challenges, practice management, and financial matters) represented a chasm between resident and attending responsibilities. Like our residents, their participants felt more prepared for clinical than nonclinical components of their new roles. A study of emergency medicine trainees revealed major gaps in their ability to teach and provide feedback, a core competency of independent practice at a teaching hospital.²⁵  Similarly, a survey of preparedness for independent orthopedic practice reinforced a preparedness for clinical work but not for the managerial aspects of the independent surgeon. And paralleling the views of the Consultant JAs in our study, many junior orthopedic surgeons also felt less prepared for complex and specialist procedures, despite feeling adequately prepared for routine clinical work.²⁶  Westerman concluded that navigating this transition would be eased by introducing trainees to some of the nonclinical tasks prior to their transition, much of which was accomplished in the JA experiences described by our participants.⁴ 

The literature regarding what the transition from residency to independent practice should look like is limited. The most similar description of a JA role was from a neurosurgical training program, but this intervention lacked published detail and evaluation.¹³  In the United States under the learner-manager-teacher paradigm, the teacher position in postgraduate year 3 most closely resembles the JA experience; it is intended to separate the initial years of residency during which knowledge acquisition is paramount from the senior year of residency where teaching and other responsibilities and competencies are the focus.²⁷  An Australasian study of new geriatricians recommended the creation of a “consultant-like” role.²⁸  de Montbrun et al identified 4 phases in the anxiety-provoking transition from surgical resident to attending surgeon, wherein final responsibility for patient care was suddenly assumed.²⁹  A study of radiation oncologists identified similar needs among trainees and junior faculty and recommended a minimally supervised longitudinal rotation for senior residents to mitigate the gap between residency and independent practice.⁵  In support of a JA-like experience to facilitate transition to practice, recent expert consensus guidelines emphasize the importance of providing an authentic environment, encouraging progressive independence, inclusion of non-medical expert domains, and psychologically preparing trainees for independence with final responsibility.³⁰ 

Based on our findings, we perceive that a JA experience may lessen the intensity and anxiety of TTP. The JA role addresses most, if not all, of these shortfalls. Our participants identified mastery of nonclinical skills as major learning opportunities, and the level of unprecedented independence as a trainee likely lessens some of the “transition shock”³¹  associated with starting independent practice. Exposure to some of the nonclinical tasks during training, such as supervision of trainees, has been associated with improved confidence as a junior faculty member.¹ 

Our study had several limitations. First, we conducted our study in GIM, ID, and rheumatology within a single university. There may be different components to the transition to independence in other specialties or at different institutions, although our literature review suggests similar themes are present in other medical disciplines. Second, our participants may not have been representative of those in non-academic or remote centers, given most worked or trained in large urban referral centers. We attempted to mitigate this by snowball sampling participants with suspected divergent views. Third, our study focused on a Canadian experience. Regional differences in certification time, residency programs, patient expectations, care delivery, and medico-legal norms may limit generalizability. Finally, we did not correlate the quality of reported experiences to objective outcome measures in education or patient care, which could be areas for further study.

Further research priorities should evaluate provision of additional guidance and the development of best practices for both supervising faculty and JAs to ensure this educational experience is successful in supporting learners through this phase of the physician lifecycle.

The JA model can help prepare residents for independent practice. Critical to the optimal implementation of this role is attention to the balance between autonomy and supervision, team relationships, and role legitimacy.