Adding patient demographic information to medical exam questions can influence diagnostic decision-making, according to a new study that raises questions about how future clinicians are trained and assessed.
Researchers surveyed 376 third- and fourth-year medical students and first-year residents using clinical vignette multiple-choice questions. Participants were randomly assigned either standard questions or identical questions that included patient descriptors such as race, gender identity, or sexual orientation. The findings, published in Academic Pathology, suggest these additional details altered answer selection, even when they were not essential to reaching the correct diagnosis.
The greatest effect was seen in scenarios where demographic information could distract from the underlying clinical presentation. Participants who received questions without demographic descriptors performed better on these "bias-obscuring" cases than those who received the demographic versions.
Differences also emerged in scenarios involving sexually transmitted infection testing in a homosexual demographic, sarcoidosis in a Black population, Takayasu arteritis in patients of Asian descent, and interpretation of hemoglobin results in transgender patients.
Although the study focused on medical education rather than laboratory medicine directly, the findings have important implications for diagnostics. The authors argue that assessments should better reflect real-world clinical reasoning by reducing reliance on demographic cues, unless they are genuinely relevant to diagnosis.
The study further recommends expanding assessment methods beyond multiple-choice examinations to include competency-based evaluations, alongside structured education on recognizing and mitigating bias. Such changes, the authors suggest, could support more equitable assessment of clinical reasoning while reinforcing evidence-based diagnostic practice.
Lead researcher Kamran Mirza from Michigan Medicine says, “When demographic descriptors consistently influence how learners answer otherwise identical clinical questions, we should see that not as a failure of individual trainees, but as an invitation to reflect on the lessons embedded throughout medical education and society. The hopeful part is that education is one of the few systems we can intentionally change. If we've inadvertently helped shape these patterns of thinking, we also have the opportunity, and the responsibility, to reshape them.”
The researchers acknowledge that recruitment fell short of the planned sample size and that participants may not fully represent the wider medical trainee population. Even so, they conclude that the findings highlight the need for medical education reforms to foster an unbiased approach to patient care.
Newsletters
Receive the latest pathologist news, personalities, education, and career development – weekly to your inbox.
