Mitigating Bias in Virtual Interviews for Applicants Who are Underrepresented in Medicine

Abstract

The COVID 19 pandemic has forced residency programs to consider virtual interview formats that can disproportionately and negatively impact unrepresented in medicine applicants. Diversity is a compelling interest for residency programs. In order to recruit and retain a diverse workforce, programs must recognize the deleterious effect virtual interviews may have on underrepresented in medicine applicants.

The novel coronavirus pandemic has forced a rapid adoption of virtual platforms to be utilized, from classroom didactics, testing, and now interviews. Every fall, 40,000 medical students travel across the country for residency interviews to determine their future training destination. Students spend time meeting program leadership, interacting with current residents, and experiencing program culture to determine a fit and create a rank list. For applicants who are underrepresented in medicine (UIM), their experiences at the interview day can also provide much needed insight into how comfortable they feel at a program. Program culture is an intangible part of the interview experience – especially when an applicant can note microaggressions - and is difficult to communicate through a virtual platform. Residency programs and the broader academic medicine community must look closely and consider how virtual interviews can uniquely impact students who are UIM.

UIM students have experienced disparities and often discrimination at every step of their journey across the medical education continuum.¹ Analysis done by Lett et al.² found that Hispanic applicants and matriculants are underrepresented by 70% and American Indian and Alaskan Natives applicants and matriculants by more than 60% when compared to the relative age adjusted population in the US. While there is no question that there is a lack of diversity in the physician workforce, it further highlights the bias in the recruitment process and lack of support in retention. Traditional metrics - often based on scores, awards, and recommendations - that are used to grant interviews have also historically negatively impacted UIM students. Black medical students are less likely to be inducted into the Alpha Omega Alpha (AOA) society³ when compared to their peers. There are significant differences in the mean United States Medical Licensing Examination (USMLE) Step 1 scores by race and ethnicity with UIM students scoring lower on the exam than their White peers.⁴ The USMLE has been used for postgraduate resident selection despite studies showing a poor correlation between these scores and clinical performance. A study by Poon et al.⁵ showed that Orthopedic residency applicants who were White were more likely to get accepted despite Hispanic students having more publications and Black applicants having more volunteer hours. Bias - conscious or unconscious - was not investigated, but it is well known to be a factor that affects those UIM.

Receiving an interview in the face of all the structural barriers is an accomplishment in and of itself. The switch to a virtual platform does not challenge the underlying systems that create implicit bias or replicate systems of oppression that are detrimental to UIM applicants. It is of utmost importance for residency programs to remain cognizant of the biases that exist in interviewing.

Residency programs should be transparent about the steps they take to ensure a diverse workforce. With knowledge that there is an overemphasis on screening applicants by USMLE scores, and that UIM students score lower on the examination despite no strong correlation between scores and clinical performance, the authors recommend a more holistic review of applicants. There are many screening tools that can measure a candidate's competencies.

Two of the core competencies of residency training outlined by the American College of Graduate Medical Education (ACGME) are interpersonal and communication skills and professionalism. A comprehensive review of a candidate's online application can provide some insight into a future physician's potential in developing these competencies. However, the interview process is essential for the program's ability to evaluate these traits in an applicant. These are vital indicators of future success within a training program and in eventual clinical practice. Therefore, robust interviewing methods within the constraints of social distancing and institutional restrictions on travel must be employed through virtual forums to afford residency programs and applicants the continued ability to make these assessments.

The ACGME updated its Common Program Requirements in 2019 to now mandate residency programs engage in practices that focus on the recruitment and retention of a diverse and inclusive workforce. This shift to virtual interviews can potentially compound existing issues related to implicit bias as well as introduce new aspects of this bias in the residency application process that will impede this directive in ways the medical community has not fully encountered previously. We must seek to mitigate bias in the virtual interview world that could decrease the chances of recruiting diverse trainees. These solutions must include faculty and residency training in recognizing and avoiding implicit bias, as well as successful virtual interview techniques.

Besides this instruction for the interviewers, the logistics of the interview format need scrutiny. Interviews should be standardized for all applicants, but the move to virtual interviewing should inspire creativity and aid us in deviating from familiar but less effective methods that are considered the norm. Initially designed to assess non-cognitive skills, traditional and panel interviewing has shown variable reliability and predictive validity for academic performance. An attractive alternative is the concept of multiple mini interviews (MMI). These are structured scenarios with matching scoring rubrics that can be designed to minimize bias. MMI has been found to be a fair process as perceived by interviewees and interviewers, as well as being validated, reliable between raters and stations, less affected by gender, race, socioeconomic status, or coaching.⁶

Residency recruitment extends beyond the interview itself. Additional activities such as away rotations and pre-interview dinners will mostly be eliminated across the country. UIM applicants will now be even more disadvantaged in showcasing themselves to a desired program. Therefore, opportunities for virtual mentoring, virtual hangouts between trainees and applicants, and attendance at virtual didactics must be developed and offered to all applicants, but even more explicitly given to those UIM applicants who are more likely to be affected by these restrictions. Residency programs must also be more deliberate in signaling their priorities of diversity through program recruitment materials and demonstrate their ongoing commitment to diversity by developing and publishing their implicit bias and anti-racism policies and measures.

Medical schools also have a responsibility to ensure a level playing field for students. Access to the proper equipment for a successful virtual interview is not a given. Not all students have access to a quality web camera, stable home internet access, or a quiet environment to conduct interviews. Medical schools can conduct needs assessments of their student body, and make sure that all students have access to the tools they need to be successful. The authors recommend that medical schools should make available interview rooms with adequate telecommunication and webcam capabilities.

In many ways, this moment has demonstrated that the pandemic is not a great equalizer. Rather this pandemic can and will expose the faults in our existing systems and the process of residency applications is not exempt. The move to virtual interviews can have a disproportionately harmful effect on UIM students and residency programs, medical schools and specialty organizations should pay special attention to this population of applicants. Moves towards holistic review, implicit bias training, and support for UIM students will help marginalized students continue to succeed.

Conflict of Interest

Conflicts of Interest: None.