Journal club: Does undergrad competitiveness predict USMLE performance

Levy J, Kausar H. et. al. College competitiveness and medical school exam performance. BMC Medical Education volume 22, Article number: 780 (2022)

This article asked the question if students from “competitive” colleges scored the same on the USMLE (the United States Medical Licensing Examinations) as those from non-competitive colleges.

They used the definition of competitive as a college, “having greater than 10% of its student body scoring 1400 or higher on the SAT.”

The data was gathered from all current students at the Kirk Kerkorian School of Medicine at UNLV. It was a retrospective observational cohort study. They looked at each student’s undergraduate institution, MCAT score, and medical school exam scores and then divided them into students who went to a “competitive” vs. a “non-competitive” college.

Since UNLV is a newer medical school there were only two cohorts that had completed USMLE Step 2, with only a sample size of 96. Compared to a sample size of 221 for the MCAT.

The data was as follows on each of the exams:

This data is incredibly limited given that it is from a single institution and there are many confounding factors that were not controlled for, however, the data almost argues against their point. The only statistically significant data is that those students from competitive colleges had higher average MCAT scores (511.13 vs. 508.44). There is abundant data demonstrating that the MCAT score is the single best predictor of doing well on the USMLE. It is interesting to note that for the students at UNLV the mean MCAT score for noncompetitive colleges is 508.44 which is still scoring in the 79% percentile and therefore in and of itself predictive of doing well on the USMLE. Also, the n of 137 students from a non-competitive college vs. 84 from a competitive college suggests that admissions are not biased against students from a non-competitive college and are more likely to select students based on an MCAT threshold, rather than their undergraduate institution.

The journal article they cite which indicates medical schools use institutional selectivity as criteria during admissions is from 2000 and evaluated data from 1992-1995, now almost 30 years ago, before a transition emphasizing holistic admissions.

This is incredibly limited as a single institutional study. It also more likely answers the question about what MCAT score a student has the baseline ability to do well on the USMLE when educated at that school since the mean MCAT score for both cohorts was high. This article felt a bit like a reason to write a long conclusion espousing the merits of holistic admission which many medical schools already stand behind. Ultimately, when choosing an undergraduate institution it still should be about “fit” and where the student will learn and thrive best.