Background

The challenge of graduate medical education is to prepare physicians for unsupervised practice while ensuring patient safety. Current approaches may inadequately prepare physicians due to limited opportunities for autonomy. Recent work on how trainees gain autonomy shows that they actively influence their supervisors' entrustment decisions. If program directors more clearly understand how trainees experience increasing independence, they may better sensitize trainees to the deliberations they will face during patient care.

Objective

The authors sought to explore how trainees experience lessening supervision as their clinical training advances.

Methods

Using constructivist grounded theory, the authors recruited trainees from various specialties and training levels via email and conducted 17 semi-structured interviews from 2019 to 2020 to solicit clinical experiences during which their perceived autonomy changed. Through constant comparison and iterative analysis, key themes and conceptual relationships were identified.

Results

Seventeen trainees from 4 specialties described novel clinical situations that required “overextending,” or going beyond their perceived edge of evolving expertise. This move represented a spectrum based on perceived locus of control, from deliberate overextending driven by trainees, to forced overextending driven by external factors. Trainee judgments about whether or not to overextend were distilled into key questions: (1) Can I do it? (2) Must I do it? (3) Do I want to do it? and (4) Is it safe to do it? More advanced trainees posed a fifth question: (5) Am I missing something?

Conclusions

Decisions to move into the realm of uncertainty about capabilities carried weight for trainees. In making deliberative judgments about overextending, they attempted to balance training needs, capability, urgency, and patient safety.

Objectives

To further characterize how trainees develop clinical practice at the edge of evolving expertise.

Findings

When confronted with novel clinical situations, trainees go beyond the edge of evolving expertise, defined as overextending. This occurs on a spectrum from deliberate to forced. During deliberate overextending the locus of control lies with the trainee; during forced extending the locus of control is in external factors.

Limitations

Our study was completed at one large academic center and the findings may be unique to the culture of our institution; trainees may be more likely to report on procedural instances of overextending than others.

Bottom Line

Clinician educators can sensitize trainees to the notion of overextending and the deliberations they will face through explicit discussions about the relationship between entrustment, autonomy, and their learning.

The major challenge of graduate medical education (GME) is to prepare physicians for unsupervised practice while guarding patient safety. Progressions from wholly supervised to unsupervised work was previously assumed to occur via an apprenticeship model1  and led supervising physicians to infer they could leave trainees unattended during overnight hours and surgical procedures. Now, an increased emphasis on patient safety has resulted in work hour restrictions and enhanced faculty supervision,2,3  measures which may limit progressions in autonomy.

Learning theories help explain the developmental trajectory to manage increasingly complex situations with decreasing supervision. For example, Vygotsky conceptualized the “zone of proximal development” as a space in which learners achieve maximum gains when they function with support just beyond a skill that has been fully mastered.4,5  This perspective suggests that trainees realize their full potential only when they perform skills at the dynamic edge of their expertise, aka their “learning edge.”

Dynamic processes characterize entrustment at the point of care. Factors that influence entrustment decisions relate to: (1) trainees, (2) supervisors, (3) situations, (4) tasks, and (5) the relationships between trainees and supervisors.6,7  While many aspects of entrustment lie outside of trainees' locus of control, recent research has shown that physicians-in-training actively shape perceptions of their trustworthiness through specific behaviors that may influence their supervisors' decision-making.8  Kennedy and colleagues described 4 foundational dimensions of trustworthiness as understood by supervisors: (1) knowledge and skill, (2) conscientiousness, (3) truthfulness, and (4) discernment of limitations.9  Further, Robinson et al10  described diminishing opportunities for decision-making influenced by multiple factors while also highlighting the role of a sense of ownership.

What remains less clear is how trainees experience and shape their work environment4  to learn at the edge of their evolving expertise, as well as how they experience their own agency to do so while providing patient care. Novel clinical situations provide valuable learning opportunities, yet when and how trainees capitalize on which opportunities11  requires more exploration.

Our objective was to better understand how trainees' experience learning beyond the boundaries of their current abilities, which could benefit GME in 2 ways: (1) to guide future trainees in gaining progressive autonomy efficiently, and (2) to support supervising physicians in making entrustment decisions.

The study took place at a large, university-affiliated, quaternary care medical center comprising 3 hospitals and 840 resident trainees at an accredited US medical school in Chicago, Illinois. Residency program directors and the Vice Dean for Education supported our recruitment of participants from their programs by allowing investigator access to their trainees via email. One investigator (A.K.) sent 2 recruitment emails to trainees in pediatrics, emergency medicine, internal medicine, surgery, and obstetrics and gynecology residency programs asking them to participate. Ensuring a mix of training year and specialty, we consented and interviewed 17 participants in total. Our aim was not to achieve a representative sample but an informative sample for this qualitative study. Each participant provided informed consent and received a $10 gift card after participation.

For this exploratory study, we used a constructivist grounded theory (CGT) approach, which explores complex social processes that can benefit from further theorization.12  We chose CGT because this approach is well suited to help explore trainees' subjective clinical experiences. Further, CGT encourages researchers to reflect on their own backgrounds and perspectives that could impact the research process.13  In examining our reflexivity, all authors are physicians who experienced their own journeys to independent clinical practice. One author (A.K.) completed advanced specialist training during most data collection and analysis. All authors are versed in medical education research, and 2 authors (P.W.T., W.J.E.) have extensive backgrounds in qualitative methodologies.

One investigator (A.K.) conducted all individual semi-structured interviews14  between June 2019 and November 2020 to solicit residents' accounts of influential patient care experiences. Open-ended questions invited participants to describe their experiences while practicing at the edge of their capabilities. For instance, trainees were asked what tasks they regularly completed unsupervised. See online supplementary data for the interview guide. Initial sampling was purposive to include diversity in gender, training level, and medical specialty. Later sampling was guided by theoretical considerations in line with CGT.12  Individual interviews were conducted either in person or via videoconferencing. Following CGT methodology, we collected and analyzed data iteratively. Interviews were audio-recorded, transcribed, reviewed for accuracy, and deidentified.

The analysis proceeded first with line-by-line coding and using constant comparative analysis of initial interviews to create focused codes. Two authors (A.K., W.J.E.) coded a subset of interviews to ensure consistent coding; disagreements were resolved through discussion. Analytic meetings served to elevate and combine key concepts and to identify major themes. The entire team met to examine the relationships among major themes to theorize how trainees practiced at their learning edge. In later stages of iterative data collection and analysis, we recruited additional participants whose specialty and training level allowed us to refine aspects of our conceptual model. Data collection ended when our analysis achieved theoretical sufficiency.15  We used MAXQDA 2018 (VERBI Software, Berlin, Germany) to facilitate data management and coding.

The Office of Research Integrity and Compliance at Ann & Robert H. Lurie Children's Hospital of Chicago reviewed this study and determined that it was exempt from institutional review board review.

We interviewed 17 physicians-in-training (7 men, 10 women) from clinical postgraduate year (PGY) 1 through 5 in pediatrics, emergency medicine, internal medicine, and surgery residency programs. Fifteen participants were from one academic center; 2 were clinically active during research fellowships at this center but were primarily enrolled in clinical training programs at other academic centers. See Table 1 for an overview of participant characteristics. Anonymous participant codes identify representative quotations and year of training (eg, P08, PGY-1). All trainees reported regularly encountering novel clinical situations that required judgments about making decisions just beyond their own perceived level of current ability. These situations required trainees to either go beyond (P01, PGY-3) their current known ability, or, conversely, to remain at the level of their current known ability. We defined practice at or just past the learning edge as “overextending” and used this term to describe situations in which trainees experienced uncertainty and continued to care for patients despite being outside their comfort zones.

Table 1

Participant Characteristics

Participant Characteristics
Participant Characteristics

We identified a spectrum of overextending with 2 extremes: (1) deliberate overextending, driven by trainees, and (2) forced overextending, driven by factors beyond the trainees' direct control, such as supervisor, patient, or environmental factors. Although all instances of overextending were influenced by situational aspects, we differentiated these extremes of overextending in relation to trainee perspectives of agency: either they deliberately managed the situation or the situation forced their hand. Both agency and situational factors influenced trainees' deliberations and affected when and how overextension occurred. We now discuss each of these main findings in greater detail. See Table 2 for representative quotes that highlight the overarching concepts and themes described in this section.

Table 2

Major Themes and Representative Quotes

Major Themes and Representative Quotes
Major Themes and Representative Quotes

Deliberate Overextending

Trainees reported feeling agency during deliberate overextension, characterized by perceived confidence, presumed trust from their supervisors, and familiarity with the task at hand. Importantly, deliberate overextension typically required some back-and-forth with supervisors. These negotiations could become quite nuanced and often included multiple strategies, including trainees demonstrating their readiness through preparation and explicitly communicating their preferred level of supervision. When asked why these discussions were important, trainees cited anticipated future unsupervised practice as a main motivator.

Demonstrating Readiness for Less Supervision

Rather than simply stating readiness, trainees recognized the need to demonstrate readiness to take on tasks with less supervision in key ways, such as (1) performing an initial and complete-as-possible patient evaluation independently, (2) formulating and committing to a plan, and (3) preparing for procedures by gathering supplies, obtaining informed consent, etc. Trainees communicated their preferred level of supervision both explicitly and implicitly. This sometimes took the form of completing a task with implicit permission as supervisors watched, for example, in the operating room.

Anticipating Future Unsupervised Practice

Trainees were often mindful of performing future tasks without supervision. Especially toward the end of training, they began putting themselves in their supervisors' shoes. This anticipation of unsupervised practice transformed how they approached overextension. Their future selves loomed large in their minds and motivated them to seek out situations at their learning edge to then extend it.

Experiencing Supervision During Deliberate Overextending

In instances of deliberate overextension, close supervision was perceived as “hovering,” which we defined as supervision perceived as unhelpful, unwanted, or overbearing. Deliberate overextending sometimes appeared to be an attempt to escape hovering that evoked feelings “like someone is expecting you to fail, so you just feel the spotlight” (P05, PGY-2). To illustrate this worry, trainees reported situations in which supervisors took charge of procedures while supervising closely, prompted by lack of trainee confidence. Trainees predicted that hovering supervisors would transition from supervision into direct clinical care. These experiences undermined trainees' sense of responsibility, agency, and perceptions of supervisors' trust. Participants endorsed feeling trusted by their supervisors and the clinical team when they were allowed to assume some unsupervised tasks, even if they were unsuccessful.

Forced Overextending

Trainees engaged in forced overextending during clinical events that pushed them beyond their perceived learning edge, situations characterized by unique tensions. On the one hand, trainees realized that their patients had clear and immediate care needs; on the other hand, they faced uncertainty and discomfort in their ability to meet those needs without the supervision and support they perceived necessary but which may not have been immediately available. For example, emergency medicine residents frequently rotated away from the academic hospital to gain exposure to other practice settings and encountered different expectations at different sites.

Trainees understood that their supervisors expected them to complete tasks without direct supervision. These expectations appeared to stem from: (1) supervisors' direct knowledge of and experience with the trainee, or (2) supervisors' expectations based on training level. Trainees also recognized acute or complex medical situations that required urgent action. They described forced overextension as “being thrown into the fire” (P05, PGY-2), whether they were ready or not. These uncomfortable experiences, however, contributed to their learning, especially early in training.

During forced overextension, trainees recognized obvious tensions between their desire for supervision to ensure patient safety and pressure to fulfill urgent patient care needs. Supervisors provided a safety net; trainees appreciated having someone to check their clinical work and “bounce ideas off of” (P13, PGY-3). Their supervisors' presence helped ensure that committing to the wrong plan did not result in adverse patient outcomes, allowing them to fail safely. Although forced overextension represented one end of the spectrum of overextension, most clinical situations trainees described included both elements of negotiation and external factors. We will discuss these external factors in turn.

Factors Influencing Overextension

Supervisor Expectations

Trainees often learned about their supervisors' expectations at the beginning of a new academic year. These expectations were explicitly stated by supervisors, implied, or constructed through discussion with other trainees. Urgent clinical needs demanded overextension, and trainees' perceived lack of readiness collided with a sense that they should be ready to manage certain aspects of patient care without supervision at their current level of training.

Patient Expectations

Trainees' close contact with patients and families engendered trust, feelings of responsibility, and patient ownership. Further, family members often expected trainees to manage primary tasks. Their close patient contact placed the onus to communicate difficult news squarely on trainees, whether they felt ready or not.

Expectations Within a Medical Specialty

A trainee's specialty seemed to influence overextension behaviors, specifically: (1) which tasks to try unsupervised, and (2) how to talk about comfort and discomfort with particular tasks. For example, both surgical and emergency medicine trainees identified expectations to state comfort in novel clinical situations, regardless of their actual comfort or discomfort, to minimize risk of losing learning opportunities.

Specialty also shaped perceptions about which tasks comprised overextending at given training levels. This could represent unsupervised conversations regarding difficult diagnoses in pediatrics or unsupervised central line insertions in surgery. Awareness of these variations across specialties helped trainees construct how they practiced at their learning edge as they moved through different areas of the hospital. This awareness influenced:

  • what trainees perceived as acceptable to say, ask for, and do on their own,

  • how trainees interacted with supervisors, and

  • how trainees determined acceptability of deliberate overextension.

To Overextend or Not to Overextend?

Trainees were often explicitly aware of opportunities to overextend, some of which resulted from deliberative judgments and others that arose from acute patient care needs in particular patient care contexts. They used an in-the-moment risk-benefit analysis of contextual requirements and their current perceived clinical abilities. We distilled these deliberations to 4 fundamental questions that captured the main factors in residents' considerations related to capability, urgency, accountability, clinical development goals, and patient safety:

  • Can I do it?

  • Must I do it?

  • Do I want to do it?

  • Is it safe to do it?

Trainees leaned into opportunities for overextending when patient safety was not in jeopardy, they were sufficiently confident in their abilities, and they had demonstrated competence in prior supervised situations. When thinking about clinical situations that demanded more urgent responses, trainees' internal deliberations about “Can I do it?” were sometimes “I think so.” Importantly, questions around “Must I do it?” for the sake of patient care led to unsupervised actions. Therefore, these judgments appeared interrelated, and often depended on external factors such as patient care setting or supervisor preference. Of course, getting to “yes” in these deliberations did not guarantee success. Overextension demanded vulnerability and acceptance of possible failure.

Unsuccessful instances of overextension prompted reflections on how to improve patient care, since failures revealed previously unrecognized blind spots and familiarized them with the boundaries of their capabilities.

A Fifth Question

Trainees further along their developmental trajectory toward unsupervised practice described asking themselves a fifth question before overextending, distilled as: “What could I be missing?” Some reported that they earned their supervisors' trust if they considered potential complications or contingency plans prior to being prompted, demonstrating more clinical maturity.

Based on our analysis, we identified overextending as a specific mechanism by which trainees expanded skills and knowledge at their learning edge. Overextension occurred on a spectrum ranging from deliberate to forced, with multiple factors influencing trainees' decisions on whether or not to overextend. These factors included supervisor expectations, patient and family expectations, and local culture. Future unsupervised practice motivated physicians-in-training to seek and experience opportunities to practice at their learning edge, and they engaged in explicit deliberations about when and how to do so. We see potential in sensitizing trainees to the notion of overextending and the deliberations they will face through explicit discussions about the relationship between entrustment, autonomy, and their learning. Clinical educators can use these 5 key questions to help promote trainees' ability to engage in reflection-in-action16  about opportunities to overextend and balance patient safety, clinical needs, and trainee ability. These questions may also guide supervisor entrustment decisions. Forced overextension in particular has several implications, specifically around the necessity, inevitability, or desirability for trainees' perceived need to act on their own.

Overextending describes how trainees move tasks from the zone of proximal development4  into a zone of unsupervised problem-solving. In our conceptualization, overextension describes how trainees take on more tasks with less oversight, representing individual instantiations of progressive autonomy. As described previously, agency and responsibility17  influence concrete behaviors in clinical practice that contribute to the development of autonomy. Cantillon and Macdermott18  described how responsibility influenced learning: the trainees “valued opportunities to take individual responsibility for patient care.” In some situations, trainees sought agency even when they had to resist pressure to conform to expectations.19  Perceived agency was related to addressing the deliberative questions we identified above; trainees recognized their ability to shape expectations of supervisors, patients, and families within their learning culture.

Our work also extends findings by Pingree et al8  who recognized many similar trainee behaviors, specifically highlighting the importance of confidence and self-advocacy. In line with their findings, trainees in our study also intuitively understood that “thinking aloud”20  could demonstrate entrustability. Further, trainees may feel pressure to act independently in situations characterized by acuity, urgency, or complexity. Kennedy et al21  reported similar findings, defining some of these expectations as cultural—either endemic to the practice of medicine as a whole or to specific specialties. Our participants identified cultural norms and expectations that shaped their deliberations to earn the right to practice with lessening supervision throughout training. While they reported frequent help-seeking behaviors, they also described barriers influenced by local culture. Trainees were aware of the potential for failure during instances of overextending. Prior work has shown that clinical supervisors occasionally allow trainee failure for educational benefit and attempt to balance patient safety and trainee learning.22  Trainees acknowledged the importance of supervision that prevented patient harm and enabled them to fail safely.22 

Our study has important limitations. Our findings may reflect the unique culture of our academic medical center or individual specialties and may reflect trainee tendency to report on procedural tasks more often than others. Members of the research team (A.K., W.J.E.) who coded primary data were medical educators and emergency physicians making their own entrustment decisions, which may have affected data analysis. However, one author (P.W.T.) was from a separate institution, worked in a different specialty, and had a different perspective on entrustment decision-making, which balanced analytic discussions.

Future research should further delineate and enhance skills required for safe and successful overextension, as well as further characterize the nuances of overextending in different specialties. Another important area of inquiry is how sensitizing trainees to these processes may influence entrustment decisions. Finally, faculty development in clinical training programs may aid them in promoting safe overextension behaviors and progressive autonomy.

Overextending represents a concrete mechanism by which trainees actively shape how they push the boundaries of their learning edge to develop progressive autonomy. It exists on a spectrum from (1) deliberate, when control lies with trainee, to (2) forced, when external factors compel trainees to act. Perceived ability, patient need and acuity, patient safety, and educational value shaped decisions to overextend along the journey toward unsupervised practice.

The authors would like to thank the trainees who participated in this study and Cheryl Westercamp for transcribing.

1. 
Cox
M,
Irby
DM,
Cooke
M,
Irby
DM,
Sullivan
W,
Ludmerer
KM.
American medical education 100 years after the Flexner Report
.
N Engl J Med
.
2006
;
355
(13)
:
1339
-
1344
.
2. 
Sandhu
G,
Thompson-Burdine
J,
Matusko
N,
et al
Bridging the gap: the intersection of entrustability and perceived autonomy for surgical residents in the OR
.
Am J Surg
.
2019
;
217
(2)
:
276
-
280
.
3. 
Sandhu
G,
Magas
CP,
Robinson
AB,
Scally
CP,
Minter
RM.
Progressive entrustment to achieve resident autonomy in the operating room: a national qualitative study with general surgery faculty and residents
.
Ann Surg
.
2017
;
265
(6)
:
1134
-
1140
.
4. 
Vygotsky
L.
Mind in Society: The Development of Higher Psychological Processes.
Cole
M,
John-Steiner
V,
Scribner
S,
Souberman
E,
eds.
Cambridge, MA
:
Harvard University Press;
1980
.
5. 
Kennedy
TJT,
Lingard
L,
Baker
GR,
Kitchen
L,
Regehr
G.
Clinical oversight: conceptualizing the relationship between supervision and safety
.
J Gen Intern Med
.
2007
;
22
(8)
:
1080
-
1085
.
6. 
Sterkenburg
A,
Barach
P,
Kalkman
C,
Gielen
M,
ten Cate
O.
When do supervising physicians decide to entrust residents with unsupervised tasks?
Acad Med
.
2010
;
85
(9)
:
1408
-
1417
.
7. 
Santen
SA,
Wolff
MS,
Saxon
K,
Juneja
N,
Bassin
B.
Factors affecting entrustment and autonomy in emergency medicine: “how much rope do I give them?”
West J Emerg Med
.
2019
;
20
(1)
:
58
-
63
.
8. 
Pingree
EW,
Huth
K,
Harper
BD,
et al
Encouraging entrustment: a qualitative study of resident behaviors that promote entrustment
.
Acad Med
.
2020
;
95
(11)
:
1718
-
1725
.
9. 
Kennedy
TJT,
Regehr
G,
Baker
GR,
Lingard
L.
Point-of-care assessment of medical trainee competence for independent clinical work
.
Acad Med
.
2008
;
83
(suppl 10)
:
89
-
92
.
10. 
Robinson
M,
Bowen
J,
Aylor
M,
van Schaik
SM.
Finding (and keeping) a voice: pediatric residents' perceptions of autonomy and patient care ownership
.
Acad Med
.
2021
;
96
(suppl 11)
:
213
-
214
.
11. 
Guadagnoli
M,
Morin
M,
Dubrowski
A.
The application of the challenge point framework in medical education
.
Med Educ
.
2012
;
46
(5)
:
447
-
453
.
12. 
Charmaz
K.
The power and potential of grounded theory. Med Soc Online.
2012
;
6(3). Accessed March 10, 2022.
13. 
Ramani
S,
Könings
KD,
Mann
K,
van der Vleuten
CPM.
A guide to reflexivity for qualitative researchers in education
.
Acad Med
.
2018
;
93
(8)
:
1257
.
14. 
DiCicco-Bloom
B,
Crabtree
BF.
The qualitative research interview
.
Med Educ
.
2006
;
40
(4)
:
314
-
321
.
15. 
Varpio
L,
Ajjawi
R,
Monrouxe
LV,
O'Brien
BC,
Rees
CE.
Shedding the cobra effect: problematising thematic emergence, triangulation, saturation and member checking
.
Med Educ
.
2017
;
51
(1)
:
40
-
50
.
16. 
Schön
DA.
The Reflective Practitioner
.
Farnham, UK
:
Ashgate Publishing;
1991
.
17. 
Sawatsky
AP,
Santivasi
WL,
Nordhues
HC,
et al
Autonomy and professional identity formation in residency training: a qualitative study
.
Med Educ
.
2020
;
54
(7)
:
616
-
627
.
18. 
Cantillon
P,
Macdermott
M.
Does responsibility drive learning? Lessons from intern rotations in general practice
.
Med Teach
.
2009
;
30
(3)
:
254
-
259
.
19. 
Watling
C,
Ginsburg
S,
LaDonna
K,
Lingard
L,
Field
E.
Going against the grain: an exploration of agency in medical learning
.
Med Educ
.
2021
;
55
(8)
:
942
-
950
.
20. 
Pinnock
R,
Young
L,
Spence
F,
Henning
M,
Hazell
W.
Can think aloud be used to teach and assess clinical reasoning in graduate medical education?
J Grad Med Educ
.
2015
;
7
(3)
:
334
-
337
.
21. 
Kennedy
TJT,
Regehr
G,
Baker
GR,
Lingard
LA.
“It's a cultural expectation…” The pressure on medical trainees to work independently in clinical practice
.
Med Educ
.
2009
;
43
(7)
:
645
-
653
.
22. 
Klasen
JM,
Driessen
E,
Teunissen
PW,
Lingard
LA.
“Whatever you cut, I can fix it”: clinical supervisors' interview accounts of allowing trainee failure while guarding patient safety
.
BMJ Qual Saf
.
2020
;
29
(9)
:
727
-
734
.

Author notes

Editor's Note: The online version of this article contains the interview guide used in the study.

Funding: This work was funded by the Grainger Research Initiative Award.

Competing Interests

Conflict of interest: Dr Eppich has received salary support from the Harvard Center for Medical Simulation to teach on simulation educator courses; he receives per diem honoraria to teach on simulation educator courses with PAEDSIM e.V; and he is faculty with the Debriefing Academy.

This work was previously presented at the virtual Association for Medical Educators in Europe (AMEE) conference, September 7-9, 2020.

Supplementary data