Time to ditch the ERAS application photo

Pretty damning results about the impact of perceived attractiveness on residency application success. Suffice to say, what came up didn’t exactly make it into the NRMP Program Director’s Survey.

There’s a new paper in Academic Medicine titled “Bias in Radiology Resident Selection: Do We Discriminate Against the Obese and Unattractive?” coming out of Duke. Hint: the rhetorical question posted in an academic paper title is always answered with a yes. But while the study used their own radiology program, I have zero doubt that this is universal and probably substantially worse in other fields.

The idea was to grade mock applications and see who you’d invite:

Reviewers evaluated 5,447 applications (mean: 74 applications per reviewer). United States Medical Licensing Examination Step 1 scores were the strongest predictor of reviewer rating (B = 0.35 [standard error (SE) = 0.029]). Applicant facial attractiveness strongly predicted rating (attractive versus unattractive, B = 0.30 [SE = 0.056]; neutral versus unattractive, B = 0.13 [SE = 0.028]). Less influential but still significant predictors included race/ethnicity (B = 0.25 [SE = 0.059]), preclinical class rank (B = 0.25 [SE = 0.040]), clinical clerkship grades (B = 0.23 [SE = 0.034]), Alpha Omega Alpha membership (B = 0.21 [SE = 0.032]), and obesity (versus not obese) (B = -0.14 [SE = 0.024]).

The breakdown:

  • Facially attractive and nonobese applicants had a 24% chance of getting an interview
  • Less attractive, nonobese applicants had a 12% chance
  • Obese and unattractive had a 10% chance

At the end of the day, the top three factors for selecting candidates were Step 1 > Race > Facial attractiveness. And being both skinny and attractive literally doubled your chances.


While programs will always eventually meet their applicants and may always “like” applicants who are easy on the eyes, I don’t think any residency (except derm kidding not kidding) actively wants to screen their applicants by appearance.

Is there really any legitimate justification for having access to an ERAS photo in the first place prior to selecting interview candidates? I don’t need to know what you look like.

In the meantime, this study just further confirms the advice I’ve given before: you want your ERAS photo to be good.

Leave a Reply