[Skip to Navigation]
Sign In
Invited Commentary
Medical Education
July 16, 2020

Gender Differences in Resident Assessment: The Glass Ceiling and Sticky Floor for Women in Medicine Begin Early

Author Affiliations
  • 1Department of Medicine, Southern Illinois University School of Medicine, Springfield
  • 2Department of Internal Medicine, University of New Mexico School of Medicine, Albuquerque
JAMA Netw Open. 2020;3(7):e2010985. doi:10.1001/jamanetworkopen.2020.10985

Studies in health care and the business world demonstrate that diversity, equity, and inclusion are important. They help to strengthen the work environment through improved engagement, enhanced communication, increased productivity, and better workforce retention. More importantly for health care, diversity improves patient care through more effective communication, increased patient adherence, improved patient satisfaction, and an overall better patient experience. Despite these benefits, important work remains to address the barriers that deter us from achieving that goal. Understanding how, when, and where those barriers are manifested can help us better understand why they happen and how to mitigate them. In JAMA Network Open, Klein et al1 aim to understand whether and how gender is associated with assessments in graduate medical education.

This multi-institutional study examined the association of gender, in both residents and faculty, with the distribution of faculty ratings for the Accreditation Council for Graduate Medical Education’s 6 core competencies across all 3 years of internal medicine training. The findings were quite compelling. There seemed to be a clean slate at the beginning with no gender difference in evaluations during the postgraduate year 1 (PGY1). From there, the differences begin. In the PGY2, female residents scored significantly higher in 4 competencies and then plateaued, whereas male residents peaked later but ended training with significantly higher ratings. The authors found that gender-congruent pairs favored male residents. It is important to understand the reasons for this variability, because there are important implications for medical education and health care. Although not the primary focus of the study, the authors also found that, despite no baseline difference, PGY3 female residents had lower scores on the In-Training Examination than male residents. This should be further studied to see whether it is a pervasive issue, because medical educators hold up standardized tests as the criterion standard, and they may be biased.

Although unsurprising, these findings are necessary so that their underpinnings can be addressed at personal levels and—more critically—program and institutional levels. Social norms create gendered expectations from patients and peers2 with different expectations of women and men, so it is likely that these expectations also subconsciously factor into assessment and possibly teaching. Women encounter stumbling blocks to recognition in medicine, including overcoming expectations that favor stereotypically male characteristics, such as confidence, that engender entrustment by faculty and that may affect how female residents are perceived by their faculty evaluators.3,4 They also encounter a glass ceiling of unseen barriers where fewer opportunities are available for mentorship and sponsorship, possibly resulting in female residents not receiving feedback and leadership development that are necessary for successfully navigating training.5 Particularly in a hurried training environment, how faculty speak about trainees can influence and anchor their peers’ impressions; consequently, the language we use about trainees is critical, and we must sponsor them by speaking about their achievements. In addition to the glass ceiling, women also face a sticky floor where they are less likely to advance owing to inadequate investment and resource allocation at the beginning of their careers. This can undermine their professional development and reinforce imposter syndrome.6 Biased training and clinical environments hold women back,7 and women may be more likely to doubt their competence owing to training environments that favor agentic behaviors, struggle to accommodate women in foundational ways, and have commonplace microaggressions and bias.

There are many potential implications of this study. Although competency-based assessment has been part of graduate medical education since 1999, this study highlights how difficult it may be to truly assess competency. This is important during discussions on using competency-based assessment as part of time-variable graduation that will allow learners to finish training early once deemed competent. If the peak and plateau findings by Klein et al1 are replicated by others, men could finish training earlier than women, affecting women’s abilities to attain subspecialty fellowship positions. If women take longer to finish training, this could further negatively affect the gender compensation gap that persists and worsen the imposter phenomenon that affects many women. The purpose of competency-based education ostensibly is to assess competency objectively.

This study gives pause and questions whether we are measuring the right thing and have the right tools to do so.

Where do we go from here? We should confront the persistent problem of inequity with the rigor we have applied to other challenges in medicine by developing and implementing rigorous structures and processes to achieve the best outcomes. Foundational, structural steps that training programs can take to advance equity include requiring training on implicit biases, microaggressions, and upstanding for all faculty and trainees, as well as adopting robust antiharassment, antidiscrimination, and family-friendly leave policies. Programs should also explicitly commit to accountability for gender-disparate assessments and outcomes.

Processes that reduce inequity can include blinded review of clinical and academic work; providing leadership skills training for women; providing faculty development on implicit biases in assessment, feedback, and mentorship; and finding creative solutions to address the “second shift” that women often have, in which they perform most of the household duties, such as program-coordinated grocery delivery and institutional support for child care. In addition, programs should report gender-specific outcomes, such as deidentified In-Training Examination scores, clerkship evaluations, and award recipients, as well as achievements and accomplishments, such as publications, posters, and trainee awards, systematically rather than through a self-reported process. Programs and institutions should then be transparent about the implications of any gender differences found and what tangible steps will be taken to resolve them within a specified time frame. As more data become available and more effective strategies become known for reducing gender inequity, best practices should be spotlighted and accrediting bodies should consider mandating them.

Where there is gender inequity, there is likely also inequity for other historically excluded and marginalized groups in medicine that must be studied and addressed. Inequity indicates a system requiring improvement. When we understand and accept the benefits that come from inclusion and equity, it is clear we have a fiduciary duty to our trainees and patients to ensure it.

Back to top
Article Information

Published: July 16, 2020. doi:10.1001/jamanetworkopen.2020.10985

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Hingle S et al. JAMA Network Open.

Corresponding Author: Susan Hingle, MD, Office of the Dean for Human and Organizational Potential, Southern Illinois University School of Medicine, 913 N Rutledge, Mailcode 9623, Springfield, IL 62794 (shingle@siumed.edu).

Conflict of Interest Disclosures: None reported.

Klein  R, Ufere  NN, Rao  SR,  et al.  Association of gender with learner assessment in graduate medical education.   JAMA Netw Open. 2020;3(7):e2010888. doi:10.1001/jamanetworkopen.2020.10888Google Scholar
Linzer  M, Harwood  E.  Gendered expectations: do they contribute to high burnout among female physicians?   J Gen Intern Med. 2018;33(6):963-965. doi:10.1007/s11606-018-4330-0 PubMedGoogle ScholarCrossref
Pingree  EW, Huth  K, Harper  BD,  et al.  Encouraging entrustment: a qualitative study of resident behaviors that promote entrustment.   Acad Med. 2020. Published online May 5, 2020. doi:10.1097/ACM.0000000000003487PubMedGoogle Scholar
Lee  KB, Vaishnavi  SN, Lau  SK, Andriole  DA, Jeffe  DB.  “Making the grade:” noncognitive predictors of medical students’ clinical clerkship grades.   J Natl Med Assoc. 2007;99(10):1138-1150.PubMedGoogle Scholar
Shakil  S, Redberg  RF.  Gender disparities in sponsorship—how they perpetuate the glass ceiling.   JAMA Intern Med. 2017;177(4):582. doi:10.1001/jamainternmed.2016.9411 PubMedGoogle ScholarCrossref
Mullangi  S, Jagsi  R.  Imposter syndrome: treat the cause, not the symptom.   JAMA. 2019;322(5):403-404. doi:10.1001/jama.2019.9788 PubMedGoogle ScholarCrossref
Spector  ND, Overholser  B.  Examining gender disparity in medicine and setting a course forward.   JAMA Netw Open. 2019;2(6):e196484. doi:10.1001/jamanetworkopen.2019.6484 PubMedGoogle Scholar
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words