[Editor’s note: The following essay is adapted from Do No Harm founder Dr. Stanley Goldfarb’s opening statement at the USF Health Morsani College of Medicine’s October 2, 2026, debate, “Does ‘DEI’ Help or Hurt Medicine?”]
Patients deserve the very best care that we physicians can provide. I spent the last part of my 50-year career at Penn’s medical school as an associate dean of curriculum devoted to that end. The selection of students to become physicians is probably the single most important function of a medical school. It is particularly critical because only about one-third of the 55,000 applicants can be accepted into allopathic medical schools. It is a zero-sum game.
What should be the criteria? I believe students should have a keen intellect, a strong work ethic, a strong moral code, and a penetrating curiosity to learn medicine. They should demonstrate their potential through an exceptional performance in undergraduate courses and in the MCAT, the two most effective predictors of performance in medical school, in residency training, and, as recent data suggest, in practice. Race and ethnicity are irrelevant, and using them as characteristics for success as a physician is unfair and illegal.
Holistic admission, according to the AAMC’s model, includes a myriad of factors that appear unrelated to medicine or patient care, including “citizenship,” “distance traveled,” “relationship status,” involvement in “cultural events,” and “sexual orientation.” In addition to these attributes, the model explicitly includes “race” and “national origin,” both of which are prohibited by Title VI of the Civil Rights Act of 1964. At many schools, selecting students for admission to medical school using these criteria has been associated with substantially lower MCAT scores and grade-point averages.
This approach is supposed to produce a student body that is more attuned to patients’ needs, culturally and racially aligned with demographically similar patient groups, and therefore more likely to produce improved healthcare outcomes. What is the evidence that those characteristics are associated with improved clinical care for patients?
The question is not new, and it has been explored in over 60 studies in the peer-reviewed medical literature and has also been aggregated in six systematic reviews. All of these pose the question, “Does racial concordance — that is, does racially matching patients and physicians — lead to better communication, more trust, and better outcomes for patients?”
The answer from this robust medical literature is “no”! Everyone in this room has been or will be a patient, and as patients we expect the best care possible. The race of the caregiver does not improve that outcome.
This is not to deny the reality of healthcare disparities. But they result from poverty, cultural factors, nutrition, and even genetic factors. The argument has been forwarded that these disparities are due to bias on the part of physicians. But studies that claim bias are often fatally flawed.
For example, the most widely quoted study performed on a Florida cohort purported to show that black neonates had a survival advantage if cared for by black pediatricians. This really attributes the worst biases and prejudice to white neonatologists. Imagine being accused of bias to explain a poor patient outcome. This study was even cited by Justice Jackson in the Supreme Court in the case of Students for Fair Admissions v. Harvard as justification for DEI criteria in undergraduate admissions.
But this study was terminally flawed, as it failed to properly account for the lower birth weight of the black neonates cared for in neonatal intensive-care units. Those units are mostly staffed by white doctors, yet the study continues to be cited as the best evidence to promote racial concordance between physicians and patients. Once birth weight was properly balanced, the differences disappeared.
We need a colorblind admission process to medical school. Life experiences and leadership traits are all fine to consider, as long as academic achievement and cognitive ability are the primary criteria. For admission to undergraduate institutions, the dynamic is between the school and the student. In medical school, there is a third agent in the process, the patient, and the patient must always be the prime focus. Academic performance before medical school is the best predictor of success. It should be the key metric that determines admission.
Finally, the use of racial and ethnic characteristics as part of the admissions process diminishes the accomplishments and brilliance of black physicians such as my beloved colleague, the late doctor Marilyn Singleton. She was a leader in American medicine, a graduate of Berkeley law school and Stanford medical school, and a fierce advocate for colorblind admissions to medical school. She always felt that she needed no special accommodations to achieve all that she achieved.
A colorblind, merit-based admissions process at all medical schools will produce diversity as a consequence of the excellence of all of the students who are accepted. This is what has been achieved at the Morsani College of Medicine, and this is what we must strive for in recruiting the future American healthcare workforce.
