Bias Reduction in Academic Recruitment (BRIAR)

Community Medical Center, Toms River, NJ

Status: Completed

Conditions: Education, Medical; Internship and Residency; Interview

Treatments: Other: Blinding to academic performance of candidate; Other: Usual Interview methodology

Study type: Interventional

Funder types: Other

Identifiers

NCT05616065

22-029

Details and patient eligibility

About

The aim of this study is to determine if, with respect to medical students applying for residency in emergency medicine, prior knowledge of an applicant's academic record affects their ranking during their interviews.

Attendings interviewing candidates will either be blinded to their prior academic records or be allowed to review them prior to the interview. The interviewer's final score of the applicant will be measured, comparing blinded versus unblinded assessors.

Full description

In residency recruitment and ranking, the most honest admission educators can offer is that there truly is no exact way to identify candidates who will thrive in a given discipline or residency program. Along with applications and letters of reference, the traditional residency interview process has long been thought to be one of the best available tools to evaluate a given candidate as well as assess their 'fit' for the culture of the program they wish to enter. Despite this decades-used tool, little uniformity exists in practice, with various disciplines and programs using various techniques to evaluate potential residents. To wit, prospective resident themselves, in one study, have listed their top three consideration when selecting a residency to be the friendliness of the program and faculty, the overall environment, and the interview day itself. Spending time in the department and hospital and having contact with the faculty and staff has long been seen as the best way to 'get to know' the program. This is among the main reasons why medical students will often schedule 'audition rotations' at hospitals and programs that draw their interest.

With respect to the interviews themselves, there has been some evolution of thought, despite the variability. Recently, structured interviews and other techniques have been discussed to reduce bias and increase efficiency in residency recruitment.

While this has been discussed in academic circles for some time, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has brought these issues to the fore. Forcing interviews to be done remotely has effectively removed the intimacy of an in-person visit or interview, making an assessment of 'fit' for both the applicant and the program more challenging. Many program leaders and resident educators have described 'fit' as among the most important predictors of residency success, sometimes more so than pure academic prowess.

Of the many biases in residency interviewing, among the easiest and most seductive is hinging a decision on academic achievement alone. It is well understood by many residency educators that academic success in medical school does not always translate to the same level of performance in residency. With the virtual method of interviews continuing for at least this year (and likely going forward), our study ponders the question if the removal of knowledge of an applicant's academic record changes their interviewer's perception of their potential as a resident in an emergency medicine.

This study will examine if blinding interviewers to the academic performance of residency applicants impact the assessor's recommendation for the candidate's match day rank list placement. Candidates will be assigned to have one of three of their interviews performed during the standard interview process, to have a single assessor who is instructed not to look at either medical school transcripts or standardized test scores. Assessors will be permitted to look at letters of recommendation, personal statements, research and work experience, and hobbies.

The absolute value of the difference of between blinded and unblinded assessors will be compared and tested for statistical significance by matched student T testing. The study is powered for an absolute difference of 1 point on a ten-point scale, which translates linearly to rank list placement. Additionally, this 1-point difference is considered significant from a standpoint of residency operations. Also, blinded versus unblinded assessors will be compared to the placement by the residency program director, who will remain unblinded and also makes independent recommendations for every candidate. As a subanalysis, a qualitative assessment of candidates who get ranked either substantially higher or lower when blinded compared to unblinded will also be conducted.

Enrollment: 212 patients

Sex: All

Volunteers: Accepts Healthy Volunteers

Inclusion criteria

• All candidates who have or are anticipating completing their medical school education, and have applied for a position in our residency during the 2022-2023 match cycle.

Exclusion criteria

• Candidates our program declined to grant an interview spot or have canceled their interview prior to completing it.

Trial design

Primary purpose: Health Services Research

Allocation: Non-Randomized

Interventional model: Parallel Assignment

Masking: None (Open label)

---

212 participants in 2 patient groups

Blinded

Experimental group

Description:

Interviewers will be assigned a candidate to interview. They will be asked not to review the medical school transcripts or standardized test scores (USMLE, COMLEX). They will be permitted to review personal statements, letters or recommendation, and other personal information in the application.

Treatment:

Other: Blinding to academic performance of candidate

Unblinded

Active Comparator group

Description:

Interviewers will be assigned a candidate to interview. They will be given full permission to review the full application of their candidate as per their normal interview protocol.

Treatment:

Other: Usual Interview methodology