Implicit gender-career bias in postgraduate medical training still exists, mainly in residents and in females

Stress and career aspirations: a longitudinal study with medical students

Background

The psychological well-being of medical students has often been studied in relation to various factors and typically exhibits poorer scores than comparable student groups. This diminished well-being can have far-reaching implications, like impact on career or academic engagement. While there are some studies on specialty choice or academic performance few studies have examined the effects of stress at the beginning of medical school on career ambitions and the associated interest in research. Given the anticipated future shortage of healthcare professionals and the goal of gender equality in higher positions, it is crucial to examine this relationship as early as medical education. The question arises whether medical education and the demands of studying influence the career ambitions of different student groups.

Method

As part of a longitudinal survey, students from a medical cohort are surveyed over three consecutive years. A total of 297 students participated in the surveys, of which 66.4% were women. Stress experiences during medical school and career aspirations are assessed and analyzed using Generalized Estimating Equation (GEE) modeling.

Results

The results indicate that early academic stress is a negative predictor of the career ambitions of medical students (B = 1.637; p = 0.009). Also, female students report lower career goals than male students (B = 1.168; p = 0.018). In addition, gender moderates the impact of stress on career ambitions (p = 0.046). Finally, stress levels are highest when starting the medical program (p = 0.022) and decrease over time.

Discussion

Stress during the initial medical training impedes students to pursue higher career ambitions. Future research as well as medical programs need to elaborate more on and consider structural stress prevention interventions in order to promote well-being and strengthen also scientific career goals at an early stage, specifically for female students.

Still Higher Risk for Burnout and Low Work Engagement Among Female Residents After 10 Years of Demographic Feminisation

Objectives

We explored whether gender differences in burnout and work engagement characteristics among residents changed after the representation of female physicians has surpassed the 30% threshold of critical mass between 2005 and 2015, as well as if these gender differences are influenced by working in a surgical versus a non-surgical specialty.

Methods

This study used data of two questionnaire surveys on the well-being of Dutch residents, collected in 2005 (N = 2115) and 2015 (N = 1231). Burnout was measured with the validated Dutch translation of the Maslach Burnout Inventory, covering the characteristics emotional exhaustion, depersonalisation and personal accomplishment. Work engagement was measured with the Utrecht Work Engagement Scale, covering the characteristics vigour, dedication and absorption. Gender differences in residents' engagement and burnout characteristics in 2005 and 2015 were analysed using hierarchical regression analyses. Factorial analyses of variance were used to compare gender differences in residents' burnout and engagement characteristics in general surgery with those in internal medicine.

Results

In both years, female residents reported higher emotional exhaustion, lower depersonalisation, personal accomplishment, and vigour than males. These gender differences were similar in general surgery and internal medicine.

Conclusions

This study demonstrated unchanged gender differences in burnout and work engagement characteristics among residents after 10 years of demographic feminisation (increasing female representation), indicating higher risk for burnout and lower work engagement among females, both in surgical and non-surgical specialties. In view of the ever-increasing number of female residents, educators and hospitals need to create supporting work environments that safeguard residents' well-being.

Gender Representation of Health Care Professionals in Large Language Model-Generated Stories

Importance

With the growing use of large language models (LLMs) in education and health care settings, it is important to ensure that the information they generate is diverse and equitable, to avoid reinforcing or creating stereotypes that may influence the aspirations of upcoming generations.

Objective

To evaluate the gender representation of LLM-generated stories involving medical doctors, surgeons, and nurses and to investigate the association of varying personality and professional seniority descriptors with the gender proportions for these professions.

Design, Setting, and Participants

This is a cross-sectional simulation study of publicly accessible LLMs, accessed from December 2023 to January 2024. GPT-3.5-turbo and GPT-4 (OpenAI), Gemini-pro (Google), and Llama-2-70B-chat (Meta) were prompted to generate 500 stories featuring medical doctors, surgeons, and nurses for a total 6000 stories. A further 43 200 prompts were submitted to the LLMs containing varying descriptors of personality (agreeableness, neuroticism, extraversion, conscientiousness, and openness) and professional seniority.

Main Outcomes and Measures

The primary outcome was the gender proportion (she/her vs he/him) within stories generated by LLMs about medical doctors, surgeons, and nurses, through analyzing the pronouns contained within the stories using χ2 analyses. The pronoun proportions for each health care profession were compared with US Census data by descriptive statistics and χ2 tests.

Results

In the initial 6000 prompts submitted to the LLMs, 98% of nurses were referred to by she/her pronouns. The representation of she/her for medical doctors ranged from 50% to 84%, and that for surgeons ranged from 36% to 80%. In the 43 200 additional prompts containing personality and seniority descriptors, stories of medical doctors and surgeons with higher agreeableness, openness, and conscientiousness, as well as lower neuroticism, resulted in higher she/her (reduced he/him) representation. For several LLMs, stories focusing on senior medical doctors and surgeons were less likely to be she/her than stories focusing on junior medical doctors and surgeons.

Conclusions and Relevance

This cross-sectional study highlights the need for LLM developers to update their tools for equitable and diverse gender representation in essential health care roles, including medical doctors, surgeons, and nurses. As LLMs become increasingly adopted throughout health care and education, continuous monitoring of these tools is needed to ensure that they reflect a diverse workforce, capable of serving society's needs effectively.

A Shadow of Doubt: Is There Implicit Bias Among Orthopaedic Surgery Faculty and Residents Regarding Race and Gender?

BACKGROUND

Orthopaedic surgery continues to be one of the least diverse medical specialties. Recently, increasing emphasis has been placed on improving diversity in the medical field, which includes the need to better understand existing biases. Despite this, only about 6% of orthopaedic surgeons are women and 0.3% are Black. Addressing diversity, in part, requires a better understanding of existing biases. Most universities and residency programs have statements and policies against discrimination that seek to eliminate explicit biases. However, unconscious biases might negatively impact the selection, training, and career advancement of women and minorities who are underrepresented in orthopaedic surgery. Although this is difficult to measure, the Implicit Association Test (IAT) by Project Implicit might be useful to identify and measure levels of unconscious bias among orthopaedic surgeons, providing opportunities for additional interventions to improve diversity in this field.

QUESTIONS/PURPOSES

(1) Do orthopaedic surgeons demonstrate implicit biases related to race and gender roles? (2) Are certain demographic characteristics (age, gender, race or ethnicity, or geographic location) or program characteristics (geographic location or size of program) associated with the presence of implicit biases? (3) Do the implicit biases of orthopaedic surgeons differ from those of other healthcare providers or the general population?

METHODS

A cross-sectional study of implicit bias among orthopaedic surgeons was performed using the IAT from Project Implicit. The IAT is a computerized test that measures the time required to associate words or pictures with attributes, with faster or slower response times suggesting the ease or difficulty of associating the items. Although concerns have been raised recently about the validity and utility of the IAT, we believed it was the right study instrument to help identify the slight hesitation that can imply differences between inclusion and exclusion of a person. We used two IATs, one for Black and White race and one for gender, career, and family roles. We invited a consortium of researchers from United States and Canadian orthopaedic residency programs. Researchers at 34 programs agreed to distribute the invitation via email to their faculty, residents, and fellows for a total of 1484 invitees. Twenty-eight percent (419) of orthopaedic surgeons and trainees completed the survey. The respondents were 45% (186) residents, 55% (228) faculty, and one fellow. To evaluate response biases, the respondent population was compared with that of the American Academy of Orthopaedic Surgeons census. Responses were reported as D-scores based on response times for associations. D-scores were categorized as showing strong (≥ 0.65), moderate (≥ 0.35 to < 0.65), or slight (≥ 0.15 to < 0.35) associations. For a frame of reference, orthopaedic surgeons' mean IAT scores were compared with historical scores of other self-identified healthcare providers and that of the general population. Mean D-scores were analyzed with the Kruskal-Wallis test to determine whether demographic characteristics were associated with differences in D-scores. Bonferroni correction was applied, and p values less than 0.0056 were considered statistically significant.

RESULTS

Overall, the mean IAT D-scores of orthopaedic surgeons indicated a slight preference for White people (0.29 ± 0.4) and a slight association of men with career (0.24 ± 0.3), with a normal distribution. Hence, most respondents' scores indicated slight preferences, but strong preferences for White race were noted in 27% (112 of 419) of respondents. There was a strong association of women with family and home and an association of men with work or career in 14% (60 of 419). These preferences generally did not correlate with the demographic, geographic, and program variables that were analyzed, except for a stronger association of women with family and home among women respondents. There were no differences in race IAT D-scores between orthopaedic surgeons and other healthcare providers and the general population. Gender-career IAT D-scores associating women with family and home were slightly lower among orthopaedic surgeons (0.24 ± 0.3) than among the general population (0.32 ± 0.4; p < 0.001) and other healthcare professionals (0.34 ± 0.4; p < 0.001). All of these values are in the slight preference range.

CONCLUSION

Orthopaedic surgeons demonstrated slight preferences for White people, and there was a tendency to associate women with career and family on IATs, regardless of demographic and program characteristics, similar to others in healthcare and the general population. Given the similarity of scores with those in other, more diverse areas of medicine, unconscious biases alone do not explain the relative lack of diversity in orthopaedic surgery.

CLINICAL RELEVANCE

Implicit biases only explain a small portion of the lack of progress in improving diversity, equity, inclusion, and belonging in our workforce and resolving healthcare disparities. Other causes including explicit biases, an unwelcoming culture, and perceptions of our specialty should be examined. Remedies including engagement of students and mentorship throughout training and early career should be sought.

How long will it take to reach the gender diversity goal for orthopaedics in Japan?

BACKGROUND

In Japan, orthopaedics is one of the medical fields with the lowest proportion of women. This study analyses the change in gender diversity over the past decade and estimates the time required to achieve the 30% gender diversity goal, according to the critical mass in Japan in 2020.

METHODS

We investigated the demographic composition of orthopaedic surgeons in 2020 by age group, the gender ratio of the main clinical fields from 2010 to 2020, and estimated the time required for the bottom 10 (i.e., least diverse) medical departments in Japan to reach the proportion of 30% women. We used simple linear regression analyses to clarify the number of years.

RESULTS

In 2020, the population pyramid of orthopaedic surgeons showed that those in their 50s were the largest component with 24.1%, followed by those in their 40s and 30s with 22.3% and 19.4%, respectively. The percentage of women orthopaedic surgeons increased slightly from 4.1% in 2010 to 5.7% in 2020. This means that to achieve the proportion of 30% women at the current annual increase rate, orthopaedics would require up to 160 years, cardiovascular 149 years, and neurosurgery 135 years.

CONCLUSION

Contrary to the recent increase in the number of women physicians, there has been only a slight increase in the number of women orthopaedic surgeons over the past decade. Moreover, the number of young male orthopaedic surgeons has decreased. As current orthopaedic surgeons age and retire, Japan will soon face an overall shortage of orthopaedic surgeons. Issues that must still be addressed in Japanese orthopaedics include educating men and women about gender diversity and bias, changing stereotypes about surgical lifestyles, improving work-life balance, and diligent and collaborative efforts at both the individual and community levels.

Recent Trends in Successful Neurosurgery Resident Matriculation: A Retrospective and Bibliometric Analysis

BACKGROUND

Prior literature has demonstrated barriers to successful residency matching, including sex, medical school background, and international medical graduate status. Our aim is to characterize the recent trends in successfully-matched residents, with particular attention to geography and academic productivity.

METHODS

Resident information, including demographics and educational background, was gathered from program websites. Bibliometric analysis focused on PubMed publications from the top neurosurgery journals. A top 20 medical school was defined using the US News Rankings for research in 2022. Regression analyses were performed to explore the associations between total and first-author publications and other relevant factors, correcting for graduate studies.

RESULTS

A total of 114 institutions and 946 residents were included in the final analysis. Of the 845 with medical school information, 62 (7.3%) completed medical school internationally and 181 of 783 (23.1%) came from a top 20 medical school. Male residents had a higher proportion of residents with international undergraduate and international medical school degrees when compared to female residents [32 (7.5%) vs. 4 (2.4%), P = 0.021; 52 (8.6%) vs. 10 (4.2%), P = 0.026; respectively]. The multivariate regression analysis demonstrated a significant increase in publications for international medical school graduates (B = 8.3, P < 0.001), top tier medical school graduate (B = 1.3, P = 0.022), and male sex (B = 1.20, P = 0.019) for total number of publications.

CONCLUSIONS

Geographical factors, reported sex, and graduation status have influenced how resident candidates are perceived. Understanding these trends is vital for future resident matching. Addressing gender and educational diversity is essential to foster inclusivity and research-driven environments in neurosurgery residency programs.

What Goes In, Must Come Out: Generative Artificial Intelligence Does Not Present Algorithmic Bias Across Race and Gender in Medical Residency Specialties

Objective Artificial Intelligence (AI) has made significant inroads into various domains, including medicine, raising concerns about algorithmic bias. This study investigates the presence of biases in generative AI programs, with a specific focus on gender and racial representations across 19 medical residency specialties. Methodology This comparative study utilized DALL-E2 to generate faces representing 19 distinct residency training specialties, as identified by the Association of American Medical Colleges (AAMC), which were then compared to the AAMC's residency specialty breakdown with respect to race and gender. Results Our findings reveal an alignment between OpenAI's DALL-E2's predictions and the current demographic landscape of medical residents, suggesting an absence of algorithmic bias in this AI model. Conclusion This revelation gives rise to important ethical considerations. While AI excels at pattern recognition, it inherits and mirrors the biases present in its training data. To combat AI bias, addressing real-world disparities is imperative. Initiatives to promote inclusivity and diversity within medicine are commendable and contribute to reshaping medical education. This study underscores the need for ongoing efforts to dismantle barriers and foster inclusivity in historically male-dominated medical fields, particularly for underrepresented populations. Ultimately, our findings underscore the crucial role of real-world data quality in mitigating AI bias. As AI continues to shape healthcare and education, the pursuit of equitable, unbiased AI applications should remain at the forefront of these transformative endeavors.

Influence of a sexism and sexual harassment prevention course on medical students' perceptions of and responses to problematic situations

BACKGROUND

Issues of sexism and sexual harassment within the medical field are widespread. The aim of his study was to describe the influence of a training course on the prevention of sexism and sexual harassment on medical students' perceptions and responses to problematic situations in the workplace.

METHODS

We performed a secondary analysis of the transcripts of 16 medical students' interviews. The data were initially gathered as part of an external evaluation of the course. We decided to further explore these with a focus on potential changes about sexism and sexual harassment awareness induced by the course. Data were analyzed using qualitative thematic analysis.

RESULTS

Most medical students were aware of the existence of sexism and sexual harassment in the medical environment before the course but were not always able to delineate what could be considered as such or not. The course allowed them to broaden their ability to identify problematic situations, that were somewhat "trivialized" before, and to set a lower cut-off when judging what constitutes sexism or sexual harassment. It also provided them with tools on how to react when confronted to these situations, as well as resources to seek help. However, students stressed the importance of such courses also being offered to medical staff higher up in the hierarchy, as they are the ones in a position of power.

CONCLUSIONS

The course helped students to better identify and react to sexism and sexual harassment. However, given students' lack of power within the hierarchy, efforts to prevent sexism and sexual harassment must be undertaken at an institutional level to allow for change in the system as a whole.

The Impact of UK Medical Students' Demographics and Socioeconomic Factors on Their Self-Reported Familiarity With the Postgraduate Training Pathways and Application Process: Cross-Sectional Study

BACKGROUND

UK medical graduates can apply for specialty training after completing a 2-year internship (foundation training). Postfoundation training application requirements vary depending on specialty but fundamentally require key skills such as teaching, research, and leadership.

OBJECTIVE

This study investigated whether medical student demographics impact their self-reported familiarity with the Post-Foundation Training Pathways (PFTPs) and Post-Foundation Application Process (PFAP).

METHODS

This was a cross-sectional study using a Bristol Online Survey. We invited all UK medical students to answer a range of questions about their demographics. Students were then asked to rank their familiarity with PFTPs and PFAP on a scale of 1 to 5 (1=least familiar and 5=most familiar). The responses were collected between March 2022 and April 2022 and exported for further analysis. Statistical analysis was conducted in Stata (version 17.1; StataCorp) using chi-square tests.

RESULTS

A total of 850 students from 31 UK medical schools took part. There was a significant difference between gender and self-reported familiarity with PFTPs (P<.001) and PFAP (P<.001), with male students expressing higher familiarity. Similarly, there was a difference between ethnicity and self-reported familiarity with PFTPs (P=.02) and PFAP (P<.001), with White students more likely to express higher familiarity than their Black, Asian, or Mixed Ethnic counterparts. Lastly, there was an overall difference between medical background and age and self-reported familiarity with PFTPs and PFAP (all P<.001), with students from medical backgrounds and older students being more likely to express higher familiarity.

CONCLUSIONS

The impact of gender, ethnicity, age, and medical background on students' self-reported familiarity with PFTPs and PFAP is significant. Further studies are required to evaluate the impact of these factors on tested knowledge of PFTPs and PFAP and whether this impacts the success rate of postfoundation applications.

The Ideal Orthopaedic Surgeon: Comparing Patient Preferences of Surgeon Attributes to Notions Held by Orthopaedic Postgraduates

Introduction

Stereotypes have been a barrier to providing patients a diverse orthopaedic workforce. Our goal was to identify stereotypes and disparities among doctors and their patients regarding the attributes that should determine a competent orthopaedic surgeon.

Materials and Methods

A cross-sectional descriptive multicenter study was conducted in India. Tailored questionnaires were administered to patients and orthopaedic postgraduates to determine the attributes they believe patients prefer in their orthopaedic surgeon. Likert data and data on preferred sex of the surgeon were analyzed as categorical data sets using frequency statistics. Participants were asked to rank surgeon attributes and analysis was based on frequency of an item among top 5 surgeon attributes.

Results

304 patients and 91 orthopaedic postgraduates participated in the study. 70.4% and 73% of patients and 27.5% and 29.6% of postgraduates preferred an orthopaedic surgeon with greater physical strength as an outpatient consultant or operating surgeon respectively. 81% of patients had no preference of the sex of their doctor. 56% of postgraduates felt patients would prefer a male operating surgeon, none felt their patient would prefer female orthopaedic surgeon. 92.3% of the female postgraduates felt patients would prefer a male orthopaedic surgeon. Patients most often ranked years of experience, surgical outcomes, time spent with patients, reputation, and physical strength in their top 5 surgeon attributes and sex, religion, and community were given least importance.

Conclusion

Diversity among the orthopaedic workforce is necessary to optimize patient care. It is our collective responsibility to educate our patients and trainees and redress the misconceptions and stereotypes that plague our profession.

Supplementary Information

The online version contains supplementary material available at 10.1007/s43465-023-00988-2.

Perceptions of Gender Stereotypes among Women Residents in Surgical and Nonsurgical Specialties

OBJECTIVE

The objective of this study was to determine whether and to what degree residents experience stereotype perception by gender and specialty type (surgical vs nonsurgical).

METHODS

A cross-sectional survey was sent to resident physicians across all specialties at a single academic institution in February 2021. The survey items asked whether participants believe residents, faculty, and the public expect men or women to be better physicians on a numerical scale from 1 to 7. A χ² test compared the calculated mean and standard error for each survey item. This study took place at Washington University School of Medicine in St. Louis, Missouri, a large academic tertiary care center.

RESULTS

A total of 411 (46% of total) residents participated; 13 were excluded because of nonbinary gender or missing demographic information, for a final sample of 398. Participants perceived all three groups to expect men to be better physicians than women. Regression analysis showed a significant effect of gender on stereotype perception, with women reporting stronger stereotype perceptions than men. There were no significant differences in stereotype perceptions by specialty type.

CONCLUSIONS

Women resident physicians in both surgical and nonsurgical fields reported higher levels of gender stereotype perception compared with men, making it imperative that graduate medical education leadership support changes to the current learning environment.

Gender differences in emergency medicine resident assessment: A scoping review

Background

Growing literature within postgraduate medical education demonstrates that female resident physicians experience gender bias throughout their training and future careers. This scoping review aims to describe the current body of literature on gender differences in emergency medicine (EM) resident assessment.

Methods

We conducted a scoping review which adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews guidelines. We included research involving resident physicians or fellows in EM (population and context), which focused on the impact of gender on assessments (concept). We searched seven databases from the databases' inception to April 4, 2022. Two reviewers independently screened citations, completed full-text review, and abstracted data. A third reviewer resolved any discrepancies.

Results

A total of 667 unique citations were identified; 10 studies were included, and all were conducted within the United States. Four studies reported differences in EM resident assessments attributable to gender within workplace-based assessments (qualitative comments and quantitative scores) by both attending physicians and nonphysicians. Six studies investigating clinical competency committee scores, procedural scores, and simulation-based assessments did not report any significant differences attributable to gender.

Conclusions

This scoping review found that gender bias exists within EM resident assessment most notably at the level of narrative comments typically received via workplace-based assessments. As female EM residents receive higher rates of negative or critical comments and discordant feedback documented on assessment, these findings raise concern about added barriers female EM residents may face while progressing through residency and the impact on their clinical and professional development.