[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address 35.171.146.16. Please contact the publisher to request reinstatement.
[Skip to Content Landing]
Views 1,914
Invited Commentary
Medical Education
July 5, 2019

Implicit Bias in Surgery—Hiding in Plain Sight

Author Affiliations
  • 1Division of General Surgery, Department of Surgery, University of Toronto, Toronto, Ontario, Canada
  • 2Department of Surgery, St Michael’s Hospital, Toronto, Ontario, Canada
JAMA Netw Open. 2019;2(7):e196535. doi:10.1001/jamanetworkopen.2019.6535

As the popularized opinion goes, our minds harmonize 2 systems of thought: a rapid, instinctive, unconscious system (system 1) and a slower, more deliberate, conscious system (system 2).1 Although system 1 has clear evolutionary benefits—for example, we can see an expression of anger on an individual’s face, quickly intuit their emotional state, and alter our behaviors to avoid that individual—the automaticity of system 1, ie, its reliance on so-called gut feelings, can lead to unconscious stereotyping and discriminatory behaviors. Implicit bias refers to biases we carry beneath our conscious awareness that alter our behaviors. Detecting and quantifying implicit bias poses obvious challenges—how can something that occurs unconsciously be measured? The Implicit Association Test (IAT), developed in 1998, is a well-known tool for measuring implicit bias. The IAT functions on the premise that it is easier to connect concepts that we have already developed mental associations between. For example, individuals with implicit racial biases are hypothesized to more quickly connect white faces with words such as good and black faces with words such as bad than vice versa.2 Participant reaction times are purported to offer a window into the unconscious mind, revealing the implicit biases they carry. Furthermore, constructs tested in the IAT can be altered to explore different types of bias, such as sexual orientation–based and gender-based biases.

Salles et al3 examined data from more than 42 000 self-identified health care professionals who participated in the Gender-Career IAT. This version tests how quickly participants associate gendered words with words related to career or family to examine whether participants hold implicit biases leading them to more commonly associate women with family and men with careers. Participants were also directly asked how strongly they associate careers with men and family with women to test for explicit bias. Unsurprisingly, the authors found health care professionals exhibited both implicit and explicit biases; both male and female participants were more likely to associate men with careers and women with families.3 To further explore biases that exist within surgery, a field where women are underrepresented and face significant barriers to career advancement, Salles et al3 developed and administered a Gender-Specialty IAT, replacing terms for career and family with words and images related to surgery and family medicine. Among the 131 surgeons who participated in the Gender-Specialty IAT, the authors found signs of both implicit and explicit bias associating men with surgery and women with family medicine.3 These findings add to our understanding of how women are perceived in surgery and how this contributes to limiting their careers. Notably, Salles et al3 demonstrate that this is not an external phenomenon—surgeons themselves hold biases that may affect the progression of women within the profession.

The results of the IATs administered by Salles et al3 appear to tell a clear and consistent story of ongoing gender bias in surgery that likely manifests in ways that affect the career success of women. But is looking into our unconscious selves and predicting the downstream effects of our unrecognized biases really such a simple feat? The validity and reliability of the IAT have been questioned—whether higher scores can identify individuals who are more likely to exhibit discriminatory behaviors has yet to be shown. Previous studies suggest a low correlation between behavior and IAT scores.4 In fact, the IAT has been shown to be right biased, whereby individuals who exhibit no discriminatory behaviors can receive scores indicating bias.5 When the association between scores on the IAT and discriminatory behaviors is unknown, what conclusions should be made about participants who receive high scores? More importantly, could a low score on the IAT falsely reassure an individual who actually exhibits discriminatory behaviors? It is known that IAT scores have poor test-retest reliability—a single individual can receive very different scores with repeated administrations of the IAT. Additionally, scores are sensitive to environmental and social contexts. A commonly used example is an individual skilled at crossword puzzles; based on their abilities at word association games, they may receive a low score on the IAT independent of true implicit biases.6 Similarly, loud environments, stress, and fatigue can all influence IAT scores.

Despite the limitations of the IAT, the findings of Salles et al3 are consistent with ample data that substantiate the ubiquity of implicit gender bias.7 Audit studies, where participants are presented with resumes for candidates applying to a hypothetical job that differ only in the gender of the applicant, have shown that when equally qualified male and female candidates apply for jobs in the fields of science, technology, engineering, and mathematics, male candidates are considered more competent and are more likely to be hired, more likely to receive a higher starting salary, and more likely to be offered greater mentoring.8 Implicit bias colors our perception of women’s competence in ways that can have harmful effects. In a study of referral practices,9 when a patient experienced a poor outcome after surgery by a female surgeon, physicians were less likely to refer subsequent patients to the same surgeon or to other female surgeons; the same pattern was not seen when patients experienced a poor outcome after treatment by a male surgeon, suggesting that gender bias in the perception of culpability disadvantages female surgeons. The bias exhibited against women in surgery is not always implicit—in a survey of surgeons and surgical residents,10 women were substantially more likely to report forms of sex discrimination, such as pay inequity, sexual harassment, and delays in promotion.

These microaggressions and macroaggressions against women can be particularly pronounced in surgery and other procedural specialties where the schema of the agentic physician is incongruent with the stereotypical notion of a collegial and compassionate woman. When there is discordance between the stereotypical gender role and reality, such as when women pursue a career in surgery, women are perceived negatively and as less competent. Such perceptions of women in surgery can lead to poor evaluations, delayed promotion, and being overlooked for leadership roles, all factors that can contribute to burnout and poor career satisfaction.

Encouraging individuals to take the IAT has merit. The IAT is a quick and accessible tool that can begin the conversation about implicit and explicit bias in medicine and surgery and can help individuals feel as if they are working toward a solution. The IAT can be particularly illuminating when individuals carry unconscious biases that are antithetical to their worldviews, the existence of which would otherwise not be revealed or even entertained. However, knowledge of these biases will not automatically translate into behavior change. A starting point for change may be acknowledging that we all carry these biases and considering how they may be affecting our perception of reality. By understanding how our interpretation of situations varies based on the biases we carry, we can attempt to predict how these biased interpretations might affect our behaviors and identify tangible strategies to mitigate the effects of implicit bias. This stepwise approach can help make progress in diversifying our workforce, appropriately supporting our colleagues, and reducing burnout and career dissatisfaction in surgery. We applaud the efforts of Salles et al3 in attempting to quantify the degree of implicit bias in medicine; however, much more research remains to be done to further understand how our biases shape our perceptions and, more critically, how to use this knowledge to alter behavior.

Back to top
Article Information

Published: July 5, 2019. doi:10.1001/jamanetworkopen.2019.6535

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2019 Dossa F et al. JAMA Network Open.

Corresponding Author: Nancy N. Baxter, MD, PhD, Department of Surgery, St Michael’s Hospital, 30 Bond St, 040-16 Cardinal Carter Wing, Toronto, ON M5B 1W8, Canada (baxtern@smh.ca).

Conflict of Interest Disclosures: None reported.

References
1.
Kahneman  D.  Thinking, Fast and Slow. New York, NY: Farrar, Straus and Giroux; 2013.
2.
Project Implicit. About the IAT. https://implicit.harvard.edu/implicit/iatdetails.html. Accessed May 20, 2019.
3.
Salles  A, Awad  M, Goldin  L,  et al.  Estimating implicit and explicit gender bias among health care professionals and surgeons.  JAMA Netw Open. 2019;2(7):e196545. doi:10.1001/jamanetworksopen.2019.6545Google Scholar
4.
Oswald  FL, Mitchell  G, Blanton  H, Jaccard  J, Tetlock  PE.  Predicting ethnic and racial discrimination: a meta-analysis of IAT criterion studies.  J Pers Soc Psychol. 2013;105(2):171-192. doi:10.1037/a0032734PubMedGoogle ScholarCrossref
5.
Blanton  H, Jaccard  J, Strauts  E, Mitchell  G, Tetlock  PE.  Toward a meaningful metric of implicit prejudice.  J Appl Psychol. 2015;100(5):1468-1481. doi:10.1037/a0038379PubMedGoogle ScholarCrossref
6.
Singal  J. Psychology’s favorite tool for measuring racism isn’t up to the job. https://www.thecut.com/2017/01/psychologys-racism-measuring-tool-isnt-up-to-the-job.html. Published January 11, 2017. Accessed May 20, 2019.
7.
Dossa  F, Baxter  NN.  Reducing gender bias in surgery.  Br J Surg. 2018;105(13):1707-1709. doi:10.1002/bjs.11042PubMedGoogle ScholarCrossref
8.
Moss-Racusin  CA, Dovidio  JF, Brescoll  VL, Graham  MJ, Handelsman  J.  Science faculty’s subtle gender biases favor male students.  Proc Natl Acad Sci U S A. 2012;109(41):16474-16479. doi:10.1073/pnas.1211286109PubMedGoogle ScholarCrossref
9.
Sarsons  H. Interpreting signals in the labor market: evidence from medical referrals. https://economics.stanford.edu/sites/default/files/sarsons_jmp.pdf. Published November 28, 2017. Accessed October 16, 2018.
10.
Cochran  A, Hauschild  T, Elder  WB, Neumayer  LA, Brasel  KJ, Crandall  ML.  Perceived gender-based barriers to careers in academic surgery.  Am J Surg. 2013;206(2):263-268. doi:10.1016/j.amjsurg.2012.07.044PubMedGoogle ScholarCrossref
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
    ×