Breaking Barriers: Overcoming Bias in Resident Recruitment in Cardiothoracic Surgery

Introduction

The United States prides itself on its rich ethnic and racial diversity, a trait that is yet to be fully mirrored within the realm of cardiothoracic (CT) surgery (1). Despite the clear benefits of workforce diversity, including enhanced patient outcomes, academic achievements, and career fulfillment, CT surgery lags in accurately representing the nation's demographic variety. As highlighted by Olive et al. in their 2022 and 2023 studies, the underrepresentation of certain races and ethnicities in CT surgery residency underscores an urgent need for deliberate interventions (2, 3). Addressing biases in the selection process is imperative for advancing diversity and inclusivity in CT surgery. Despite existing efforts to ensure equitable selection, there is an ongoing challenge to eliminate the nuanced biases that adversely affect specific groups, notably female applicants, international graduates, and underrepresented minorities.

Variability in Interview Questions, Communication Styles, and Solutions

​​The interview process for CT surgery residency positions presents a critical juncture at which unconscious biases can inadvertently favor certain groups over others. This is a concern that is well-documented within the medical community.​ 

​​The nature and structure of interview questions can vary significantly, leading to a process that inadvertently benefits candidates from specific backgrounds or with particular communication styles. This bias arises not only from the questions themselves, but also from the linguistic nuances and cultural contexts in which they are framed. Such disparities can cause a disadvantage for candidates whose primary communication modes or styles differ from the dominant cultural norms within the interviewing panel.​ 

​​Strategies include intentionally choosing interviewers who reflect the program’s diversity, predefining merit by identifying qualities deemed most valuable for potential residents, ​​and preparing standard interview questions that are behaviorally based. Such approaches help mitigate biases by focusing on the attributes directly related to the desired qualities in candidates. Moreover, introducing a blind interview process serves as a pivotal strategy in diminishing biases. This method, combined with other initiatives, holds the potential to enrich diversity within residency programs and guarantees that applicants from diverse backgrounds receive fair consideration throughout the selection process (4). 

​​​Disparities in Scoring Mechanisms​​ 

​​Including gender-specific language in the scoring criteria may result in implicit bias. If characteristics are described using terms traditionally associated with a particular gender, such as "assertive" or "cooperative," this may inadvertently help individuals who fit those stereotypes. In such cases when an attribute affects expectations of a narrative presentation in communication, a cultural bias can emerge because different cultures perceive narratives different. For instance, stereotypes of different attributes are developed when descriptors, such as "aggressive" or "emotional" are used without a clear and objective criteria. In other words, subjective and incorrect assessments are originated from stereotypes. ​​​​​A systematic review considering gender bias in reference letters for residency and academic medicine, performed by Khan et al. (5), demonstrated that there was a significant difference in gendered adjectives between men and women. A total of 86 percent of the articles included in the review noted that female applicants were more likely to be described using communal adjectives, such as “delightful” or “compassionate.” Male applicants, however, were more likely to be described using agentic adjectives, such as “leader” or “exceptional.” Several studies noted that reference letters for female applicants frequently used doubt raisers and mentions of the applicants personal life and/or physical appearance ​​(6-10).​​ 

​​​These examples illustrate how biased gender language, cultural prejudices, negative perspectives, and limited assessment criteria can all contribute to inappropriate policy selection.​​​​ 

​​​​​Downstream Impact on Rank Meetings and Lists​​ 

​​The acknowledgment of biases being present in the interview process precedes the opinion that biases extend their influence into rank meetings and, consequently, in the subsequent compilation of residency program lists. For instance, biases may inadvertently manifest when interviewers unconsciously favor candidates who share similar linguistic or cultural backgrounds, which can potentially overlook the rich diversity that international medical graduates (IMGs) can offer to the medical scenery of the program. If the interview scores a heavy preponderance in favor of candidates from certain linguistic or cultural demographics, it can perpetuate unintentional but systemic imbalances. In this sense, imagine a scenario where the preferences for candidates with specific language proficiencies or cultural affinities continue to reinforce this phenomenon through an involuntary influence on decisions during rank meetings. This perspective underscores the need for a more introspective approach to the interview evaluation system that acknowledges the potential for biases related to different languages, cultures, or nationalities that can influence critical decisions during these meetings. This expressed opinion consists of the idea that these biases can become an inherent feature in the CT surgery residency programs if they are left unattended and can ultimately, perpetuate disparities in the field.​ 

​​This viewpoint calls for a proactive reevaluation of the interview process to ensure that the downstream impact on rank meetings and lists is characterized by guaranteeing fairness and impartiality rather than unintentionally reinforcing systemic inequities.​ 

References

1. Population Estimates. Demographic turning points for the United States:Population projections for 2020 to 2060. United States Census Bureau. March 18. http://census.gov/content/dam/Census/library/publications/2020/demo/p25-...
2. Olive JK, Yost CC, Robinson JA, et al. Demographics of Current and Aspiring Integrated Six-Year Cardiothoracic Surgery Trainees. Ann Thorac Surg. 2023;115(3):771-777.
3. Olive JK, Mansoor S, Simpson K, et al. Demographic Landscape of Cardiothoracic Surgeons and Residents at United States Training Programs. Ann Thorac Surg. 2022;114(1):108-114.
4. Balhara KS, Weygandt PL, Ehmann MR, Regan L. Navigating Bias on Interview Day: Strategies for Charting an Inclusive and Equitable Course. J Grad Med Educ. 2021;13(4):466-470.
5. Khan S, Kirubarajan A, Shamsheri T, Clayton A, Mehta G. Gender bias in reference letters for residency and academic medicine: a systematic review. Postgrad Med J. 2023 May 22;99(1170):272-278. doi: 10.1136/postgradmedj-2021-140045. PMID: 37222712.
6. Yong V, Rostmeyer K, Deng M, Chin K, Graves EKM, Ma GX, Erkmen CP. Gender differences in cardiothoracic surgery letters of recommendation. J Thorac Cardiovasc Surg. 2023 Nov;166(5):1361-1370. doi: 10.1016/j.jtcvs.2023.03.027. Epub 2023 May 6. PMID: 37156362; PMCID: PMC10592592.
7. Reddy RM. Commentary: The importance of equity in letters of recommendation in residency and fellowship applications. J Thorac Cardiovasc Surg. 2023 Nov;166(5):1373-1374. doi: 10.1016/j.jtcvs.2023.05.006. Epub 2023 May 19. PMID: 37210071.
8. Yong V, Reddy R, Erkmen CP. Discussion to: Gender differences in cardiothoracic surgery letters of recommendation. J Thorac Cardiovasc Surg. 2023 Nov;166(5):1371-1372. doi: 10.1016/j.jtcvs.2023.07.001. Epub 2023 Jul 31. PMID: 37530692.
9. Feldman H, Blackmon S, Lawton JS, Antonoff MB. Dear sirs, your bias is showing: Implicit bias in letters of recommendation. J Thorac Cardiovasc Surg. 2023 Jan;165(1):398-400. doi: 10.1016/j.jtcvs.2022.04.017. Epub 2022 Apr 23. PMID: 35599208.
10. Feldman HA, Papageorge MV, Antonoff MB. Words matter: Gender bias in letters of recommendation. J Thorac Cardiovasc Surg. 2023 Nov;166(5):e179. doi: 10.1016/j.jtcvs.2023.07.006. Epub 2023 Aug 4. PMID: 37542482.

Disclaimer

The information and views presented on CTSNet.org represent the views of the authors and contributors of the material and not of CTSNet. Please review our full disclaimer page here.