Impostor syndrome, associated factors and impact on well-being across medical undergraduates and postgraduate medical professionals: a scoping review

Abstract

Objectives

Impostor syndrome (IS) is a psychological state whereby individuals doubt their abilities despite evidence of competence. Though IS has been studied in specific medical groups, no review to date compares findings across groups. This study aimed to: (1) determine the range of IS rates among medical undergraduates versus postgraduates and (2) examine associated factors across both groups.

Design

This scoping review used the Joanna Briggs Institute methodology for scoping reviews, using a five-step framework.

Data sources

PubMed, Scopus and PsycINFO databases were searched from inception until September 2024.

Eligibility criteria

Studies were included if they were (1) empirical studies with a defined IS rating scale, (2) involving medical undergraduates, residents or clinicians and (3) published in English.

Data extraction and synthesis

Three independent reviewers used standardised methods to screen and review selected studies, and extract key variables.

Results

54 studies (77.8% from the West) were included. There was equal study distribution between undergraduates (46.3%, 25 studies) and postgraduates (46.3%, 25 studies), with the rest covering both groups. IS prevalence was substantial across all groups when assessed using the Clance Impostor Phenomenon Scale, ranging from 30.6% to 75.9% among undergraduates, 33.0% to 75.0% among residents and 23.5% to 50.0% among faculty and clinicians. In undergraduates, IS was associated with learning breaks, transition periods and poor academic performance. Among postgraduates, IS was correlated with younger age, junior ranking, fewer work years, inadequate faculty support or self-perceived poor clinical and teaching skills. Additionally, IS affected physical and psychological well-being (stress, anxiety, depression, burnout) and was associated with sociodemographic factors (single status, females), personality (neuroticism, perfectionistic traits, with conscientiousness, agreeableness and extraversion as protective) and interpersonal issues (conflicts, poor sense of belonging).

Conclusions

Given the high IS prevalence and associations with specific factors, practical measures are recommended to address IS and optimise learning and care for medical undergraduates and professionals.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• Employed Joanna Briggs Institute approach for scoping reviews, which has its foundations in two influential scoping review frameworks proposed by Arksey and O’Malley (2005) and Levac and colleagues (2010).

• We compared the range of impostor syndrome rates between medical undergraduates and postgraduate medical professionals.

• Only studies with a defined rating scale for impostor syndrome were included.

• Longitudinal evaluations and comparisons were not done as most studies were cross-sectional in nature.

Introduction

Impostor syndrome (IS), also termed as impostor phenomenon or impostorism, is a psychological phenomenon whereby individuals doubt their own accomplishments and have a persistent fear of being exposed as a fraud, even with external evidence of their competence.¹ IS was first described by Clance and Imes² as “an internal feeling of intellectual phoniness” among highly successful professional women. IS affects up to 70% of people at least once in their lives across a variety of professions, including medical professionals.^{1 3} Of note, it is associated with anxiety, poor self-worth, despair, feelings of inadequacy, burnout syndrome, heightened conflicts at work and at home and decreased job satisfaction.^{4 5}

Within the practice of medicine, the working environment can be demanding, stressful and inherently filled with interactions with individuals of varying seniority across different disciplines.⁴ This can potentially lead to different medical professionals experiencing IS with variable intensity, which can affect their ability to perform at work.¹ Earlier studies have focused separately on specific groups, such as medical students or clinicians across different specialties, but there is no review to date that compares the findings across groups such as medical undergraduates and postgraduate medical professionals across disciplines. Such comparisons can proffer insights into the differences in clinical correlates and suggest particular areas to address and ameliorate IS within the different subgroups.

Hence, this scoping review seeks to synthesise the data from extant empirical studies of IS within the medical profession, specifically on the prevalence of IS, associated factors, comparisons of these findings across medical subgroups and suggest practical implications going forward.

Methods

This scoping review was conducted in accordance with the methodology of the Joanna Briggs Institute for scoping reviews.⁶ We adopted the Arksey and O’Malley framework which consists of the following five steps.⁷ Step 1 was to identify the main research question. Our study aimed to answer the following two main research questions, namely, (1) what were the ranges of prevalence rates of IS among medical undergraduates versus postgraduate medical professionals? and (2) what were the relevant factors associated with IS among medical undergraduates versus postgraduate medical professionals? Step 2 was to identify relevant studies. We searched the PubMed, Scopus and PsycINFO databases for empirical studies reported until September 2024. The search strategy for our literature search comprised: ((impostor syndrome) OR (imposter syndrome) OR (imposterism) OR (impostorism) OR (impostor phenomenon) OR (imposter phenomenon)) AND ((medic*) OR (residen*) OR (physician) OR (surgeon) OR (surgery) OR (psychiatr*)) (see online supplemental table 1). Step 3 was study selection. An article was included if (1) it was an empirical study that evaluated IS with a defined rating scale; (2) included subjects who were medical undergraduates, residents in training or full-fledged clinicians; and (3) the publication was in English. Studies were excluded if they were case reports or opinion pieces. Three authors (SMC, IYKT, MET) independently screened the titles and abstracts of identified reports to evaluate whether they met the inclusion criteria, before reviewing the full reports of selected studies independently. Duplicates were removed through manual review by the three authors. The list of included studies was finalised based on discussion and agreement among all authors before inclusion in the review. Step 4 was data charting. For included studies, three authors (SMC, IYKT, MET) reviewed an equal number of studies each. They extracted relevant information including author names, year of publication, number and types of subjects, sociodemographic characteristics, IS scale used, rates of IS and salient correlates including gender, age and level of training. The remaining two authors (QHC, KS) then reviewed all extracted information to ensure its accuracy. Step 5 was the collation, summary and reporting of results. Extracted data were digitally organised into an Excel spreadsheet and summarised concisely for ease of critical assessment by the readers.

Patient and public involvement

None.

Results

Overall findings

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses flowchart for this study can be found in figure 1. A summary of salient details of the included studies is found in online supplemental table 2. Overall, 54 studies were included in this review. Most studies (52/54 studies, 96.3%) were cross-sectional in nature, with two studies^{8 9} analysing pre- and post-workshop outcomes. There were no longitudinal studies.

Figure 1. Preferred Reporting Items for Systematic Reviews and Meta-Analyses flowchart showing the inclusion of relevant studies in this review.

In terms of geographical region, there was a preponderance (42/54 studies, 77.8%) of studies from the West (Europe, Americas and Australia), with the rest (12/54 studies, 22.2%) from Asia.

Across all 54 studies, the total number of subjects was 17 426. In terms of gender, excluding four studies¹⁰,¹³ which did not provide the gender breakdown, the gender distribution of all participants was almost equal, with 55.6% (7561/13 599) of overall subjects being females.

Regarding the level of training, almost half of the studies (25/54 studies, 46.3%) focused on medical undergraduates, with another 25 studies (46.3%) focussing on postgraduate medical professionals (residents, clinicians or both). The remaining four studies (7.4%) included a mixture of both undergraduate and postgraduate subjects.

In terms of specialty, excluding undergraduate studies (25 studies) and studies which did not state the specialty (10 studies), more than half of the remaining studies (11/19 studies, 57.9%) evaluated IS in surgical specialties, four studies (21.1%) focused on medical specialties and one study (5.3%) evaluated IS within emergency medicine. The remaining three studies assessed IS within a mixture of specialities. Regarding IS scales used, the majority of studies used the Clance Impostor Phenomenon Scale (CIPS) (47/54 studies, 87.0%), followed by the Young Imposter Syndrome (YIS) Scale (4/54 studies, 7.4%) and Leary Impostor Scale (LIS) (2/54 studies, 3.7%). One study used a combination of CIPS, LIS and the Harvey Impostor Phenomenon Scale (HIPS) (1/54 studies, 1.9%).

Prevalence rates of IS

Among our studies, there is some variation in the cut-off scores used for the determination of IS. For the CIPS, a score of <40 denotes few impostor experiences, 41–60 moderate impostor experiences, 61–80 frequent impostor experiences and >80 intense impostor experiences.¹¹ In studies where an explicit cut-off score was not provided, we reported the rates of IS based on the number of respondents with frequent and intense impostor experiences. Four studies⁸¹⁴,¹⁶ that used the YIS had a cut-off score of five or greater to denote the presence of IS. Six studies did not report the prevalence of IS.

In terms of IS prevalence rate, among the 39 studies that evaluated the prevalence of IS using CIPS with cut-off scores, 64.1% of these studies (25 of 39 studies) adopted a cut-off score of 61 and above, and 35.9% of these studies (14 of 39 studies) adopted a cut-off score of 62 and above to determine the presence of IS. Specifically, within 21 studies involving medical undergraduates using CIPS and a cut-off score of 61 or 62 and above, the IS prevalence ranged between 30.6% and 75.9%. Within 10 studies involving residents in training across different disciplines using CIPS and a cut-off score of 61 or 62 and above, the IS prevalence ranged between 33.0% and 75.0%. Among nine studies involving faculty and clinicians using CIPS and a cut-off score of 61 or 62 and above, the IS prevalence ranged between 23.5% and 50.0%.

Training-related factors (age, year of training, academic performance, type of specialty)

Age and year of training

Overall, 42 out of 54 studies examined the relationship between the rate of IS and age, year of medical school, postgraduate year of residency, years of work experience and rank. Overall, 47.6% of studies (20/42) noted relevant associations between the rate of IS and factors such as age, year of medical school, postgraduate year of residency, years of work experience or rank.

Among the 25 studies involving undergraduate medical students only, 18 studied the association between the incidence of IS and age and/or year of training in medical school. Of the 18 studies, 11 studies found no significant association between the IS rate and age/year of training in medical school. However, one study found that taking more than three gap years before continuing with medical school was associated with IS.¹⁷ Three studies¹⁸,²⁰ found that being in the first year of medical school was a predictor of IS. One study¹² reported an increase in CIPS scores from pre-Year One to post-Year Three clerkships, and two studies^{15 21} found a significant association between IS and being in the fourth year of medical school (the year of transition to clinical clerkships).

Among the 25 studies involving postgraduate medical professionals, 7 studies assessed residents only, 8 studies assessed full-fledged clinicians only, while 10 studies assessed both residents and full-fledged clinicians. Of the seven studies that assessed residents, two studies^{5 22} evaluated the association between the IS rate and postgraduate year of residency, while three²³,²⁵ studied the correlation between the IS rate and both age and postgraduate year of residency. All five studies reported no significant association between the rate of IS and age or postgraduate year of residency. Of the eight studies that assessed full-fledged clinicians only, five studies^{910 26},²⁸ examined the relevant associations with IS. Of these five studies, three studies^{10 26 28} reported higher rates of IS in those of younger age, fewer years of work experience and lower rank. In one of the studies²⁹ that did not specifically assess the relationship between IS rate and years of work experience/rank, it was noted that manifestations of IS became more pronounced during periods of transition to a new job position, or when attaining positions of higher rank. Of the 10 studies that assessed both residents and physicians, nine studies^{34 30},³⁶ reported an increased frequency and severity of IS with younger age, an earlier postgraduate year of residency, fewer years of work experience and/or lower rank (particularly in residents).

Among the four studies⁸³⁷,³⁹ involving both medical undergraduates and postgraduate medical professionals, two studies^{37 38} assessed medical students and residents only, while the other two studies^{8 39} assessed medical students, residents and full-fledged clinicians together. Of these four studies, one study³⁷ reported no significant difference in the CIPS scores between residents when compared with medical students, but students who were enrolled in the 8-year programme (whereby students entered after graduating high school) had significantly higher CIPS scores than those enrolled in the 4+4 programme (whereby students entered with at least an undergraduate degree in a field other than medicine). The other three studies reported a higher incidence of IS in those in the first year of medical school, those who are in residency, with fewer years of work experience^{38 39} and those of younger age.⁸

Academic performance and support

Among medical undergraduates, despite a few exceptions,^{16 24 40} most studies^{1819 41},⁴³ found a positive association between poor academic performance and IS. For postgraduate medical professionals, other training-related factors were associated with IS, such as insufficient requisite knowledge,⁹ or when they perceived their clinical²⁶ or teaching²⁷ abilities to be below par.

For residents in training, inadequate faculty support had been associated with IS.^{9 23 26} Two studies found an inverse relationship between the amount of training received and IS,^{31 42} contrary to another study.²⁷ For both medical students and residents, poorer perception of the quality of the programme was associated with IS.⁴² Of note, greater self-motivation and satisfied psychological needs (autonomy, competence and relatedness) in medical school were associated with reduced severity of IS.⁴²

Type of specialty

Three studies found that there was no significant difference in IS between residents in medical and surgical specialties.^{25 44 45} Within non-surgical specialties, a study found that the rate of IS was higher among residents in rheumatology compared with cardiology,³⁷ whereas another study found that the IS rate was higher in residents within paediatrics and emergency medicine versus those in radiology, ophthalmology and orthopaedic surgery.¹⁰

Interaction of IS with other physical, psychological and social factors

Biologically, IS was associated with poorer physical health.^{20 46} In the psychological arena, several studies found that IS was associated with stress, anxiety, depression,113 20 21 25 33 37 39,⁴¹ increased suicidal ideation,^{10 47} loneliness,²⁰ burnout^{9 10 15 23 29 34 37 48} and poor self-esteem.^{16 21 25 33 38 49}

Other factors associated with IS included neuroticism,^{20 38} maladaptive perfectionism,⁴⁷ intolerance of uncertainty,⁵⁰ lower resilience⁵¹ and self-compassion.^{20 46}

Some protective psychological factors against IS included a lower sense of fear of failure,¹⁹ personality traits of extraversion, agreeableness and conscientiousness,³⁸ being more self-determined and being more supported in basic psychological needs.⁴²

Social factors can be divided into social background, cultural differences and transitions, as well as interpersonal issues. In terms of social background, 40 studies examined the relationship between gender and IS. While half of the studies (50.0%) noted no association between gender and IS,14 9 11 18 24 26 27 30 34 36 38,40 45 47 49 51 the other half^{38 10 14},^{17 19 21} found an association between IS and female gender. From 24 available studies with gender distribution among those with IS, there was a greater proportion of females (2623/4291 participants, 61.1%). In terms of marital status, being single was associated with IS,^{18 41 49} with the exception of a study done on family medicine residents which found no correlation.²⁵ For income, Elnaggar et al¹⁸ found an association between higher income and IS, but not in the study by Alsaleem et al.¹⁶

Regarding cultural differences and transitions, several studies found that foreign graduates^{22 41} and non-permanent residents⁴⁴ tended to experience higher levels of IS. In terms of interpersonal issues, IS was associated with conflicts with friends and family,⁴⁶ imposed expectations from family,⁸ fewer personal connections in profession, poor sense of belonging and unequal social capital distribution.⁵⁸

Discussion

There were several main findings. First, the IS prevalence rate was substantial across all groups when assessed using the CIPS, which ranged between 30.6% and 75.9% among medical undergraduates, 33.0% and 75.0% among residents in training across disciplines, and 23.5% and 50.0% among faculty and clinicians. Second, IS in medical undergraduates was associated with break in learning, transitions (adjustment to medical school or progression into clinical clerkships) and poor academic performance. IS in postgraduate medical professionals was associated with younger age, junior rank or residency status, inadequate faculty support, fewer years of work or having the perception that they had poor clinical and teaching skills. Third, IS affected physical and psychological well-being (including stress, anxiety, depression, burnout) and was associated with specific sociodemographic factors (such as single status, females), personality factors (neuroticism, perfectionistic traits but conscientiousness, agreeableness and extraversions were protective) and interpersonal issues (conflicts, poor sense of belonging).

Among medical undergraduates, a common theme is that of transition points and periods (such as entering medical school in year one, progressing into the clinical phase of training further down, having gap years)^{21 29} that could pose challenges and require individuals to adapt quickly to a new environment or arrangement. This may entail having to develop a new method of studying¹⁸ and adjust to the undergraduate university experience¹⁹ in the first year of medical school, or transitioning to the year of clinical clerkships which demands greater accountability and independence in one’s interactions with patients and preparation for clinical assessments.^{15 21} In the circumstance of taking gap periods away from medical undergraduate studies, the individual has to adjust to a different living environment and then readjust back to the busy learning environment which may have evolved in terms of peer and faculty composition, curriculum content and schedules over time.¹⁷ It was also observed that poor academic performance was associated with IS, as this may affect the confidence of the individual and self-appraised ability to manage the educational content and medical training received.^{13 18 21}

Among postgraduate medical professionals, common associated factors were younger age, fewer years at work, junior status in training or rank, self-perceived lack of knowledge, clinical and teaching skills. For residents in training, they faced different challenges and transitions as they had to tide through a competitive and rigorous selection and matching process for residency, and then take on more responsibilities in new job roles while undergoing further training.^{15 29} In the situation of inadequate faculty support (perceived or otherwise) and time pressure in a fast-paced working environment, this may lead to further self-appraised lack of knowledge and ability in clinical and teaching skills, which can form a vicious cycle^{33 36 59 60} and possibly account for the relatively higher IS rate (67.1%) within residents in this review. For full-fledged clinicians, they gained more work experience over time, and could develop greater self-confidence which in turn ameliorated IS-related thoughts and feelings.^{26 32 34}

Of note, IS can potentially affect both medical undergraduates and postgraduate medical professionals in their physical and psychological well-being, and is associated with sociodemographic, cultural and interpersonal factors. Regarding physical health, Rosenthal et al²⁰ found in their study that when asked to rate their own general health status, medical undergraduates with IS tended to rate it less favourably, consistent with findings of Buathong et al⁴⁶ who found that the presence of physical health problems was associated with higher IS rates in medical undergraduates.

In terms of psychological well-being, IS was associated with greater stress, anxiety, depression, loneliness, burnout among others, which can take a toll on the affected individuals.19 13 15 20 21 23 25 29 33 34 37 39,⁴¹ Personality factors such as neuroticism^{20 38} and perfectionistic traits,⁴⁷ known for their links with anxiety, depression and burnout, were also associated with IS. It is possible that neuroticism and perfectionistic traits mediate underlying pathways from stress to anxiety/depression/burnout and IS separately, and IS is further predisposed to the onset of anxiety, depressive symptoms and burnout. However, the presence of other personality traits such as conscientiousness, agreeableness and extraversion seemed to provide countervailing protective effects against IS.

In addition, being single and female gender were associated with IS. The reasons underlying these factors are not entirely clear, but may be related to additional social support for those who are married versus single (hence less likely to experience IS)⁴¹ and different coping styles associated with gender. For example, female postgraduates were found to manage IS by facing their doubts directly and competing harder, whereas males tended to avoid areas in which they were more vulnerable.¹⁵ On the other hand, a study conducted among Family Medicine residents found that IS did not vary with marital status²⁵ and half of the relevant studies did not find the association of IS with female gender. Also, cultural adaptations when one moves to work in another country can pose transition challenges,² apart from other reported factors such as perceptions of discrimination, implicit bias and inequity,⁹ which can influence the experience of IS. Interpersonal conflicts can weaken the level of social support and affect an existing poor sense of belonging, both of which were associated with IS.⁵⁸

There are several practical implications that can be considered to address and alleviate IS in view of the extant findings among medical undergraduates and postgraduate medical professionals. First, it is important to raise psychological awareness about the prevalence of IS, its impact on personal well-being and daily functioning, and possible ways to tackle it among these specific groups as well as within the medical schools (for medical undergraduates), residency programmes and organisations (for postgraduate medical professionals) which they are affiliated. Second, pre-emptive and early identification would be appropriate. One can start from the identified associated factors and be attentive to the profile of individuals that may be vulnerable to the experience of IS. For example, additional attention and support may be given to medical undergraduates during periods of transition (such as those just starting Year One and those moving on to clinical clerkship postings), those learners who need to take a gap period or are returning from their gap period, or underperforming learners who struggle to clear their regular formative and summative assessments. For residents and full-fledged clinicians, additional support may be given to those who have expressed concerns about their clinical skills, educational and other contributions despite being deemed clinically competent and a team player in their daily work performance. This is especially germane for individuals who are younger, more junior in their job status and with fewer years of work experience, those with certain personality traits (such as neuroticism, perfectionism) and experiencing interpersonal conflicts. Third, provision of appropriate support can be via individual, faculty, mentor and peer support. In one study,¹⁴ it has been found that discussing one’s experience related to IS with peers and mentors, reflection, sharing resources to foster a greater sense of belonging, being mutually encouraging and learning to accept positive feedback can be useful. There is room for faculty and supervisor development programmes to foster understanding of IS, provide resources to manage IS and encourage the setting up of community of practice groups to empower self-determination efforts and allow for discussion of challenges faced including IS.⁴² Fourth, it is also crucial for persevering commitment on the part of medical schools and healthcare organisations to pay heed to issues related to the well-being of the learners and employees (such as IS, burnout) and to enhance the training and work environments for optimal learning and clinical practice which will benefit patient care.

There are several limitations. First, most of the studies were from the West and studies from non-Western sites may allow for comparisons of associated factors to glean insights about cultural factors that may influence IS. Second, the majority of studies employed a cross-sectional design, thus making it difficult to determine causal relationships between the various factors and IS. Future papers may want to consider longitudinal designs to better establish how detected IS varies over time, especially if interventions for IS have been implemented. Third, the sample size varied with some studies reporting smaller sample sizes. In addition, notwithstanding that CIPS was the predominant rating scale used in most studies (87% of included studies), there is scope for larger, multi-site collaborative efforts to determine IS using similar rating scales and relevant clinical correlates. Fourth, most of the IS scales are unidimensional in nature. Thus, subscale interpretations are exploratory and may require cultural adaptation in non-Western contexts. The derived rates of IS based on these measures should be interpreted with caution.⁶¹ Fifth, the exclusion of grey literature may have omitted relevant findings, particularly unpublished or preliminary research. However, this was a deliberate methodological choice aimed at maintaining consistency in the inclusion of relevant peer-reviewed, empirical literature to identify established knowledge areas and conceptual gaps within the psychiatric literature. The search could be expanded beyond extant databases and the publication bias of studies assessed in future systematic reviews and meta-analyses which seek to answer more specific questions regarding IS and clinical correlates.

Conclusion

In conclusion, our review found substantial rates of IS across the medical profession ranging from medical undergraduates, residents in training to full-fledged clinicians. In view of the associated factors related to training, personality, sociodemographic factors and impact on physical and psychological well-being, practical measures are suggested to address and ameliorate IS so that learning and clinical practice can be enhanced to benefit patient care.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.