Medical Schools' Botched Pass-Fail Experiment

Published in City Journal on March 05, 2025.

Medical schools and institutions are now among the zealous champions of progressive ideology. Within days of George Floyd’s death in May 2020, the Association of American Medical Colleges demanded that the nation’s medical schools “employ anti-racist and unconscious bias training.” The following year, the American Medical Association called on physicians to “dismantle white supremacy, racism, and other forms of exclusion and structured oppression.” But efforts to enhance diversity among the medical student body—too often by compromising standards of excellence—have long been in place at America’s medical schools, from affirmative action policies to pass-fail grading of courses and clinical rotations.

In a recent Journal of the American Medical Association commentary, however, four Stanford University-affiliated scholars pushed back on these changes—a ripple that suggests a potential academic shift. In their essay, Drs. James Agolia, David Spain, and Jeff Choi, and medical student Allen Green, denounce the “diminishing objectivity” of the residency-admissions process. “We believe that some objective standards are necessary,” they write, “for programs to identify candidates who best fit their program in a fair, consistent, transparent, and efficient fashion.”

Specifically, the authors lament that the United States Medical Licensing Exam made its initial test pass-fail. The USMLE, which all would-be doctors take, is administered in three parts. Step 1 is taken after the second year in medical school to test pre-clinical medical knowledge; Step 2 is taken after the fourth or final year to assess clinical knowledge; and Step 3 is taken after the first year of residency to evaluate clinical decision-making.

The change was several years in the making. The exam’s co-sponsors, the National Board of Medical Examiners and the Federation of State Medical Boards, first recommended making Step 1 pass-fail in 2019. Other groups, including the AMA and AAMC, collaborated in developing the proposal, which was eventually adopted in 2022.

The Stanford scholars argue that changing Step 1 to pass-fail was a mistake for four reasons. First, since most competitive residency programs use Step 1 scores to select their residents, pass-fail grading gives these programs less information about applicants’ abilities. Second, it removes an opportunity for underprivileged students to stand out. “[A] high Step 1 score was once a venue through which students from less prestigious medical schools could distinguish themselves with more competitive programs,” they note. Third, in a misguided effort to lower the stakes of Step 1, the exam’s sponsors put even more pressure on students when taking Step 2 in their fourth year. Finally, the scholars observe that pass-fail grading led students to undertake extra activities during medical school—mostly unpaid research—to stand out. Students and faculty point to the often poor quality of this research, and believe that students’ time would be better spent developing clinical skills.

These are compelling reasons to oppose turning Step 1 into a pass-fail test. But they do not address an important motivation for the change: DEI. One the “[g]uiding principles for change” listed in the USMLE’s 2019 summary report was to “[m]inimize racial demographic differences that exist in USMLE performance.” In practice, this meant lowering standards.

That 2019 report also listed “diversity in medicine” as a “guiding principle.” The important subtext here, not spelled out in the report, is the idea that minority patients fare better when treated by a doctor of the same race. Yet, a recent review found that the purported benefits of a same-race doctor were unevenly-to-negatively supported by data. More certain is the value of the full USMLE: a 2024 study in Academic Medicine found that “better provider USMLE performance was associated with lower in-hospital mortality.”

The early results of making Step 1 pass-fail are in, and they are worrisome. Scores fell to a ten-year low in 2023. “The USMLE Step 1 examination was once a high-stakes standardized examination for which students studied extensively,” the Stanford authors observed. “When it became pass-fail in January 2022, the incentive to study for high scores vanished.”

While it’s true that test scores don’t give the full picture, medical school is too far along in the educational pipeline to discount academic merit in the name of social justice. It is time for the United States Medical Licensing Exam to abandon its misguided pass-fail experiment.