Introduction

In postgraduate medical education, there has been an increased focus on supervision [1, 2], which is highlighted as an important structure ‘to ensure patient safety and promote professional development’ [1]. While definitions of the term supervision vary with contexts and agendas [3], the most-sited definition put forward by Kilminister [4] describes supervision as:

the provision of monitoring, guidance and feedback on matters of personal, professional and educational development in the context of the doctor’s care of patients. This would include the ability to anticipate a doctor’s strengths and weaknesses in particular clinical situations in order to maximize patient safety. (p. 828)

Several different types of supervision are described in the literature. Educational supervision (ES) is defined as ‘regular supervision taking place in the context of a recognized training programme in order to determine learning needs and review progress’ [5]. Educational supervisors are expected to facilitate reflection, help identify learning needs, plan learning activities and formatively assess residents’ performance [3]. This supports trainees’ progress throughout their training by establishing a trusting environment, providing feedback, using direct observation of clinical practice and planning structured, timetabled sessions [2]. Educational supervisors should provide educational, professional and personal support in a longitudinal commitment that can help residents develop resilience [4, 6, 7].

Another common type of supervision is clinical supervision (CS), which describes the supervising day-to-day management of clinical cases and issues arising from such cases [3]. CS varies from discussions on ward rounds or clinics to more extended and reflective case-based discussions, but it is suggested to be evaluative in nature and closely linked to feedback on performance [8].

CS and ES share many of the same functions; Kilminster’s definition of supervision [4] has been used to define both CS and ES in earlier studies [2, 9]. Whereas CS is fairly well established as part of a longstanding apprenticeship model of residency training, ES was first formally introduced with postgraduate reforms in the 1980 and 1990s [10, 11]. Due to the increased focus on supervision, a guide for effective ES and CS was developed [1, 3]. However, previous studies have found that supervision practices are variable, and that supervisors and trainees can have different perceptions about how much supervision is required and how it should be approached [12].

Since then, a number of other support structures for residents’ learning, such as mentoring programmes [13] and coaching [14], have surfaced. The literature asserts that ES contains elements of mentoring, usually understood as guidance and support from a more experienced colleague, and coaching, which is commonly understood as helping people reach their potential [3]. While mentoring can be informal in being arisen naturally between colleagues or as a part of a formal scheme, coaching can be seen as a form of supervision that does not involve management or assessment [3]. Passi [5] claims that ‘educational supervision offers the unique opportunity to be an effective mentor to the student. The mentoring can be informal or more formal within the scheduled supervising meetings’ (p. 195). Patel [2] argues that although mentoring is not the same as ES, it should be a part of ES. Consequently, many of these educational interventions overlap in their definitions and operation in practice, which leads to confusion about what ES actually is.

Overall, there is limited research on ES in postgraduate medical training [2]. In some fields, such as psychiatry, there has been a longstanding therapeutic tradition, drawing on experiential and social learning theories [15, 16], whereas in somatic medicine ES has received less attention. Therefore, we conducted a scoping review [17] of the literature to explore how ES in specialist training for doctors in somatic medicine is carried out in daily hospital practice. A branch of somatic medicine was selected to narrow down the literature search, and internal medicine was specifically selected because it is a large branch of somatic medicine with distinct supervisory traditions. Our point of departure is that a shared understanding of terms is needed to better understand current educational practices and facilitate clear communication about how to help residents learn.

Methods

We conducted a scoping review based on Levac et al.’s modification of Arksey and O’Malley’s six-step framework [18, 19]. Scoping reviews are considered appropriate for informing practices, exploring how research is conducted in a given field, providing a foundation for developing additional research or related projects and addressing knowledge gaps. A systematic search in relevant databases was conducted to obtain an overview of research in this field. Studies were selected in line with the predetermined inclusion and exclusion criteria that are described in detail below.

Identifying the research question

The first step was to formulate the research question and develop a clearly articulated scope of inquiry [17]. Scoping reviews allow for more general exploration of selected literature, have a broad conceptual range and can be appropriate when the literature is vast and complex [17, 20]. Our overall research question for this scoping review is: What is known about ES practice in internal medicine training. In answering this we will focus on definitions of terms, descriptions on how ES is organized and findings on how ES supports learning and empirical evidence.

Identifying relevant studies

Guided by librarians from the University of Bergen, three of the authors (CNB, MK and CN) determined the inclusion criteria and a search strategy (see Appendix 1) using a Population, Intervention, Comparison, Outcomes and Study (PICOS) design (see Table 1). We piloted the search strategy to get an impression of which search terms yielded relevant hits. The search was conducted in April and Mai 2022 in the following databases: Medline, Embase, Web of Science and the Educational Resources Information Center (ERIC). By using internal medicine as an inclusion criterion to refine the search, we found through random sampling that it could exclude potentially relevant articles that dealt with ES but in which internal medicine was not mentioned or in which internal medicine was only one of many specialties considered. Therefore, we ran a separate search for systematic review articles published from 2015 to April 2022 (not limited to internal medicine) to compensate for this. We also conducted a handsearch to include all volumes of Medical Teacher in the past five years that were not covered in the Ovid databases. In addition, we performed handsearch in Google Scholar for ‘educational supervision’ in ‘internal medicine’ and ‘educational supervision’ in ‘core medical training’.

Table 1 Population, Intervention, Comparison, Outcomes and Study (PICOS) design
Full size table

Selecting studies to be included in the review

We followed PRISMA guidelines for systematic research to document the screening process [21]. We used EndNote version 20 and Rayyan (intelligent systematic review) to systematise our data during the screening process, as illustrated in the PRISMA flow chart (Fig. 1). All articles were screened based on title and abstracts using the specified inclusion and exclusion criteria.

Fig. 1
figure 1

PRISMA flow chart

Full size image

CNB screened all 3,284 articles, and the other three authors (KIR, CN, MK) screened a third each, thus ensuring a double-blind review. The resulting 180 articles were full-text screened by MK and CNB, blinded. Conflicts between the two authors’ assessments were resolved through discussions amongst the authors.

We included publications that described a support structure that was formalised, longitudinal and aimed to facilitate reflection on professional and personal development learning needs and provide pastoral support. These inclusion criteria are in line with the definition of ES put forward by Passi [5]. We excluded publications that described CS (understood as observations, feedback and daily support in clinical practice), formative assessment practices and studies on mentoring – which is often associated with reciprocal, voluntary and confidential relationships between a younger and a more experienced colleague – that typically focus on career enhancement, work-life balance and professional identity formation [3].

Importantly, we did not include or exclude articles solely based on terminology. We reviewed all the articles with the intention of identifying educational support structures that fell under the definition of ES. The process of screening was challenged by the overlapping definitions of ES and other support structures such as mentoring, CS and coaching, particularly in cases in which definitions of supervision/mentoring were not offered in the abstracts or in the full text. Full-text articles were excluded based on eight exclusion criteria (see Table 2).

Table 2 Reasons for exclusion during full-text screening
Full size table

Data characterisation and analysis

From the articles included in the review, we extracted information about year of publication, demographics, study population, publication type, research method and design. Three of the authors (KIR, MK and CNB) extracted information about what terms and definitions were used to describe ES and, if available, what educational theories the publications referred to. Finally, we obtained information about how educational supervision was conducted and organised during training and what evidence was presented to describe how ES supports learning.

Collating, summarising and reporting the results

The results and analysis are presented below.

Results

Of the 3284 articles screened, we included 18 studies published between 1996 and 2022, in our data analysis. Most of the articles came from the United States (n = 7); the rest were from the United Kingdom (n = 4), Australia (n = 1), Canada (n = 1), Japan (n = 1), Norway (n = 1), Singapore (n = 1) and Spain (n = 1), as well as a crossover study from the Netherlands, the United States and Canada (n = 1).

Study characteristics, demographics and population

The articles varied in study design (see Table 3): eighteen were empirical research studies, seven were qualitative studies, six were quantitative studies and four were mixed methods studies.

Table 3 Study characteristics, demographics, population, organizing and support
Full size table

Definitions of terms and theoretical anchoring

All the included studies defined the terms used or provided descriptions of the support structures. Four of the articles used the term ES (see Table 3) In one of these [22], an ‘educational supervisor’ is described as a senior doctor formally appointed for each trainee, who is expected to engage in formal appraisal exercises at regular intervals, provide feedback to trainees and identify shortfalls in training.

According to another study [23]:

The underlying idea of educational supervision is to support the professional development of trainees as they progress through the specialist training programme. Supervision aims to strengthen the institutional affiliation of each trainee, further the trainee’s reflection on theory and practice in their discipline and engage the trainee in ethical considerations. (p. 337)

This was the only study that drew on a theoretical framework to explain the purpose of ES – with reference to Schön [24] and ‘the reflective practitioner’ – framing ES as a practice that aims to facilitate reflection.

Instead of defining ES as such, a third study [25] described the responsibilities of an educational supervisor as follows:

Educational supervisors must provide evidence of suitable attitudes and behaviours (for example, from multisource feedback, evaluations of teaching, placement feedback forms, General Medical Council survey results, trainee audits and analysis of critical incidents) which is reviewed in an annual appraisal of the trainer role. (p. 2)

One study [7] did not include a description or definition of ES, but its authors used the term and added empirical knowledge on ES practices through analyses of quantitative data.

The remaining 14 studies described support practices in line with the definition of ES but used other terms (see Table 3). The most common term was ‘mentoring’, which was used in six studies [26,27,28,29,30,31], followed by ‘feedback’, which was described in three studies [32,33,34], ‘supervision’ in two studies [35, 36], ‘advising’ [37], ‘guidance’ [38] and ‘precepting’ [39] in one study each. Figure 2 illustrates the variations in the terms used to describe ES in our material.

Fig. 2
figure 2

Illustration of other terms used in 14 studies published between 1996 and 2022 to describe practices that fell under the definition of ES

Full size image

How ES is structured and conducted at work

Seventeen studies provide descriptions or findings on how ES practices were organised (see Table 3). Fourteen studies described time spent on ES practises (see Table 4).

Table 4 Studies describing time, duration, and background of supervisors
Full size table

Some of these studies reported on how the organization of ES occurred in practise. One study [7] showed that 6% of core medical training (CMT) trainees have ES meetings lasting less than 10 min, 48% reported meetings lasting 10–20 min and 46% more than 20 min. Furthermore, the length of attachment to a single educational supervisor was 12 months for 45% of trainees, 4 months for 23%, 24 months for 21% and 6 months for 7%. Another study found that most doctors considered training and educational task only a minor part of their work [22]. A study of medical training for doctors in Norway [23] found that training and career questions were discussed in 55% of the supervisory meetings, medical questions in 23% and job performance in 13% of the meetings. The supervisors ranked their most important roles as advocates (30%), promoting professional judgement (26%) and encouraging reflection on actions (17%). Another study revealed that mentors are responsible for supervising 5 or fewer residents, with 50% reporting inadequate dedication and 30% lacking formal training [26]. One study [28] reported that most residents met with their mentor every 6 to 12 months reported as sufficient by the majority of residents.

One study described implementation and evaluation of a novel mentor, advisor, and coach (MAC) program created for residents, showing that individualized relationships and meeting content were key to success of the program [38]. One study reported that the internal medicine training program included 150 residents with assigned preceptor, and while expectation for regular feedback conversation is communicated, there were no specific structure or frequency recommended [39]. Two studies presented findings on engagement in reflective feedback based on a structured model for feedback and coaching (R2C2) showing that it enabled meaningful and productive feedback conversations [33, 34].

Additional three studies discussed the various roles attached to being an educational supervisor [31, 32, 37]: One described overlapping educational roles and the importance of balancing individualised mentoring needs with consistent mentoring approaches [31]. Another study argued that since residents are constantly engaged in clinical work, they are frequently engaged in feedback conversations with supervising physicians as a part of their daily routines [32]. Finally, the differences between advising and mentoring are described, suggesting that residents usually had one advisor during training but could have multiple mentors [37].

How educational supervision supports learning and the empirical evidence of educational supervision

Seventeen studies included in our analysis presented empirical evidence addressing how ES supports residents’ learning (see Table 3). Twelve studies focused on the supervisor role and found that involvement over time, investment in learners’ growth, the value of confidential discussions, successful career development, growth of junior doctors, holistic support, and responsibilities in feedback settings, were important [23, 26,27,28, 30,31,32,33,34, 36,37,38]. Five studies identified inadequate supervision, such as a power imbalance between the supervisor and trainee, residents not receiving support or weak support such as vague action plans, a lack of continuity in the programme, a lack of investment in learners’ growth and failure to improve supervision practices [7, 22, 32, 35, 39].

Among the studies that used ES terminology, one described the need to implement ES due to inadequate supervision practices [22]. More specifically, the authors called for more structured training and supervision to help residents in their professional and personal development. With reference to the 1993 Calman proposals in which a new specialist grade was suggested, the authors found it worrisome that most doctors considered training only a minor part of their work.

One of the other studies using ES terminology [23] report changes in educational support structures such as better established residency committees, more extensive use of formal education plans and formal appointment of supervisors for trainees, following a systematic supervisory training program. The authors suggest that extensive faculty development for supervisors might have contributed to these changes.

A study conducted in the United Kingdom [7] reported that many trainees felt that ES was unsatisfactory, the authors also pointed out the lack of specific national guidelines in the United Kingdom for the optimal duration of an educational supervisor meeting.

Another study [25] showed that most of the educational supervisors participating in this study believed that a three-party agreement between educational supervisors’, local education providers and the Wales Deanery would help professionalise the ES supervisor role and increase supervisors’ accountability. The authors also suggested providing supervisors leverage to negotiate time for supporting professional activities (SPA) and continuing professional development. The results showed that 63% of respondents agreed or strongly agreed that the EdSA would ensure the quality of ES and indicated an expectation that the agreement would promote and enhance the standards of postgraduate medical education and training.

Discussion

The 18 articles included in our analyses, offered definitions and descriptions of ES with limited theoretical frameworks. Seventeen studies provided descriptions of how ES is structured and conducted at work and included description of individualized mentoring, regular feedback conversations, different roles on supervisors and varying structures or length of time spent on ES. Furthermore, seventeen studies showed how ES supports learning and the importance of supervisors’ involvement, investment in learners’ growth, confidential discussions and successful career development. However, inadequate supervision, power imbalance and challenges such as unsatisfactory ES and lack of national guidelines were also identified.

Although there is agreement in the literature that ES is an important part of residency training, we found that this was not reflected in the research on internal medicine supervisory practices. Interestingly, the term ‘educational supervision’ was only used in four publications: two from the 1990s, one from 2014 and one from 2017. Only one of these articles drew on a theoretical framework to define ES [23]. A study from 2016 [2] aligns with our impression that there are few studies of ES in internal medicine. Furthermore, approaches to ES remains highly variable and no studies included direct observations of ES sessions. However, we found several articles describing overlapping practices, such as mentoring, feedback and precepting, that fell under the definition of ES. Many of the studies provided incomplete descriptions of how ES is conducted at work. In particular, ES and mentoring seem to share many characteristics, which might imply that the choice of terminology is primarily a result of local traditions or trends.

As Launer [3] points out, ‘terminology can be confusing, and also varies between countries, but definitions are probably less important than understanding the context and purpose of any encounter’ (p. 179). In addition, with the implementation of competency-based medical education (CBME), the increased emphasis on formalised assessments, structured feedback, entrustment ratings and documentation of learners’ progress has led to the questioning of the relationship between supervision, feedback and assessment [40]. This might have contributed to the unclarities of what ES is. It is also important to consider the various supervisory traditions that exist between medical specialities. Whereas procedural-based specialties such as obstetrics and gynaecology and surgery, are more prone to focus feedback on observational tasks, internal medicine relies more on backstage oversight with support from supervisors [41]. Hatala et al. argue that ‘the ward-based system of learning includes regular periods when clinical supervisors are expected to be absent (e.g., nights and weekends). During these times, it is the system that entrusts trainees’ (p. 741). This means that in a hospital setting, CS and ES are performed by various individuals, whereas in specialties such as general practice, the same person can fill both roles [3]. Research from the field within psychiatry where ES has been particularly important has shown significant variations in definitions, roles and expectations of ES [15, 16].

There seem to be differences between the traditions in some European countries and the United Kingdom, where ES is a clearly defined element in residency training. This is also in contrast to the North American countries (US and Canada), where recent trends seem to have brought more focus on assessment into residency training, thus blurring the lines between formative feedback with the purpose of supporting residents and feedback aimed at assessing performance [40]. This can be problematic because the assessment focus, despite being labelled as formative, may supress and replace other support mechanisms that are there to promote learning. Blurring lines between learning and assessment can be problematic, as judgemental and supporting functions are hard to align. Similar tensions might arise if performance- and procedural-based supervision are not sufficiently delineated from the non-judgemental educational support that is integrated in ES. This suggests that there is a need for both CS and ES in the education of internal medicine residents. Trainees and supervisors must also understand the differences between these types of support, especially in work environments in which colleagues and faculty members fulfil many overlapping and sometimes conflicting roles [3]. Similar findings from research within psychiatry supports this impression [16].

The terms dominating the discourse are not without significance. Reflecting on why there are so many partly overlapping and poorly delineated concepts of support, we need to recognize that in comparison to the traditional apprenticeship model, medical education has become increasingly professionalized over the last decades. Formal assessment requirements have been given a key role in the pursuit of better training and thus better patient treatment. One might suggest that we have become alien to a practice that, in fact, has long roots in medical education. Although supervision – clinical and educational – has historically been implicit in the training of new doctors, it now needs to become explicit. We argue that it is time to revitalise ES to better align educational support structures with local needs.

Implication for future research

The few studies on ES in internal medicine residency indicates that the concept is not widespread or well known. The low search yield illustrates that there is a research gap on how ES practise is carried out in daily practice. There is a need for both theoretical and empirical research to improve our understanding of how internal medicine residents are supported in their training, how it is experienced from the perspective of both supervisors and trainees. We also suggest that similar studies should be carried out in other field of somatic medicine for broader exploration of existing knowledge. Furthermore, observational studies on ES practice that could provide knowledge on the content of ES would probably contribute to a broader understanding of what ES actually is.

Strengths and limitations of this study

We performed an extensive and broad search and screened publications to identify the practice of ES. Several reviewers conducted blind screening as well as additional searches in Google Scholar and Medical Teacher. We consider it a strength that our search was guided by a definition of ES, allowing us to map supervising practices, regardless of the terminology used.

A weakness of this study is that limiting our search to internal medicine made it difficult to comment on ES of other junior doctors or residency training in other specialties.

Conclusions

Our findings indicate that there is no universal definition of ES and that it has a weak theoretical foundation in internal medicine residency. There is also a lack of empirical studies exploring how ES is conducted and how the term is understood in relation to overlapping concepts by both trainees and supervisors. In such explorations, explicit definitions of terms and rich descriptions of practices are needed to avoid silos and to enhance knowledge production across terminological cultures and local supervising regimes. Considering the growing need for education of new doctors it is likely that there will be fewer opportunities for spontaneous individual follow-up of learners. We need to further explore the connection between the intention of what ES is supposed to be and knowledge of what it actually is, based on research on ES practises in internal medicine.