Loading ...

Júlio César de Oliveira, MD, recalls the first time he encountered a classic medical presentation: Horner syndrome. The signs and symptoms were all there — ptosis, miosis, and anhidrosis — but the patient, a Black man with dark skin, had to go through several healthcare facilities before receiving his diagnosis.

You look for images of classic signs and almost exclusively find them in White people, said de Oliveira, who is a co-founding member of Race.ID, the Research Group on the Health of the Black Population, Division of General Practice and Propaedeutics, Hospital das Clínicas, University of São Paulo, Brazil. It is easier to identify signs that have been taught on White bodies. This affects the time to diagnosis and, often, the outcome.

The case illustrates a less visible dimension of health disparities: biases embedded in clinical practice itself that can compromise the safety of Black patients.

Racial health inequalities are largely explained by social determinants. In Brazil, Black children up to 5 years old are 39% more likely to die than White children, and Black adults have a higher burden of multimorbidity. The Black population is overrepresented in the lowest socioeconomic brackets, has less access to services, and is less integrated into the formal labor market.

“We tend to attribute these worse outcomes exclusively to access and social conditions, but that is only part of the story,” said de Oliveira. “Even when the patient enters the system, care can be different.”

One example is pain management. A Brazilian study shows that Black women receive less local anesthesia during procedures such as episiotomy, even under clinical conditions similar to those of White women. The difference reflects a bias documented in the literature: the false belief that Black people have a higher pain tolerance.

The problem is that this does not appear to be an error to those making the decision, de Oliveira said. It’s a decision that seems clinical but is influenced by bias.

This is another problem that arises in medical training, as illustrated by the case of Horner syndrome. Much of medical education has been built around White bodies. This influences what we recognize as normal, as pathologic, and as urgent.

Classic clinical signs, such as anemia, jaundice, cyanosis, or inflammatory processes, may present differently across different skin tones, but this variability is rarely incorporated systematically into medical education. Physicians learn to recognize patterns in a limited spectrum of patients and must often adapt this knowledge intuitively when faced with greater diversity.

The result is delayed diagnosis, underestimation of signs, clinical uncertainty, and less precise management — effects that are difficult to measure and remain invisible despite their cumulative impact.

The persistence of these biases is directly related to the predominance of White physicians in the healthcare system. Although more than half of the Brazilian population self-identifies as Black or Brown, only about 3% of the country’s physicians fall into these categories. 

The academic funnel reinforces this inequality. Data from the USP Statistical Yearbook show that, at the institution, there is a predominance of undergraduate students who self-identify as White (about 64%), while Black students represent approximately 5% and Brown students about 17%. Among the faculty, more than 90% of professors self-identify as White.

“Diversity influences what is perceived as a problem. If we don’t have people who experience certain situations and thus recognize where the problems lie, they don’t make it onto the agenda,” de Oliveira said.

The lack of diversity also impacts the production of knowledge. Clinical studies often include small proportions of Black participants, which limits the generalizability of results and may compromise the suitability of protocols and guidelines for the real-world population.

Certain patterns of inequity may remain normalized, without being questioned or systematically investigated. Problems that are not recognized are not measured and, therefore, are not prioritized. Diagnostic delays related to clinical biases are rarely audited as system failures or included in healthcare quality indicators.

“If certain groups are more exposed to diagnostic delays, reduced access to treatment, or inappropriate decisions, this must be treated as a care risk,” said de Oliveira.

This approach involves incorporating race and color into existing indicators. Time to diagnosis, therapeutic adequacy, hospital mortality, and other outcomes could thus be analyzed in a stratified manner. There’s no point in having guidelines if they don’t become indicators. And if they don’t become indicators, they don’t become a priority. One of the strategies adopted by de Oliveira has been to work directly with medical students and residents through Race.ID. In sessionswith residents and students, he proposes clinical recognition exercises using images and patient cases with different skin tones — something still rarely addressed in traditional training.

In addition to clinical recognition, the workshops promoted by Race.ID include structured discussions on decision-making, doctor-patient communication, and implicit biases, using concrete situations — such as pain management or ordering tests — to highlight how small variations can lead to significant differences in care.

Advancing this debate requires recognizing that racial inequality in health is not only the result of factors external to medicine but also stems from the healthcare system itself. “Seeking a biological explanation for racial inequality is a mistake,” said de Oliveira. “And, often, it is an institutional mistake.”

Change requires accepting that incorporating a racial lens into the analysis of outcomes is not just a matter of equity but also of quality of care. Patients like the one described, with Horner syndrome, could receive faster diagnoses if medicine recognized that its own systems — from training to clinical practice — need to be rethought for a diverse population.

“Often, the doctor doesn’t realize they’re making different decisions. When they see this in a structured way, they begin to question their own practice,” concluded de Oliveira.

This story was translated from Medscape’s Portuguese edition.