It is a big day for Michaela. A fourth-year medical student, she is just beginning a sub-internship in internal medicine at City Hospital, her top choice for residency. She took her first call yesterday and is about to present on rounds for the first time the patients she admitted. Knowing how important a letter of recommendation from her attending Dr. Ross will be for her residency application, she is eager to make a good impression. Although she is a bit nervous, she is well prepared and confident.
When it comes her turn to present her first patient, Michaela takes a deep breath and begins, “Ms. Adler is a 76-year-old woman, with a history—”
Before she can finish her sentence, Dr. Ross interrupts her: “76-year-old black woman, right?”
“Uh, yes,” Michaela answers, after a brief hesitation. “As I was saying, she has a history of coronary artery disease and presented to the ED after a fall and a brief loss of consciousness.”
Although she is a bit unsettled by the interruption, the remainder of the presentation goes well and she gets positive feedback from Dr. Ross on her management plan for Ms. Adler. With this confidence boost, Michaela begins presenting her next patient. “Mr. Rocha is a 26-year-old man with recently diagnosed Crohn’s disease presenting—”
To her surprise, Dr. Ross interrupts her again. “White?”
“Well, I… I didn’t ask…” Michaela stutters, as she sees Dr. Ross raise an eyebrow briefly.
“OK,” he responds with a wave of his hand, “Go on.” Although she is worried that she is not making the impression she was aiming for, Michaela gets through the rest of her presentation uneventfully.
Later that day, as Dr. Ross is about to leave, Michaela approaches him to get some feedback on her presentations. “I’m sorry I didn’t ask the patient’s ethnicity. It’s just that it wasn’t relevant to the case.”
“Well, I can’t decide what’s relevant if you don’t tell me, can I?” Dr. Ross responds. “Listen, you’re doing a great job for a beginner in managing these patients, but you have to make sure that your presentations are complete: name, age, race, chief complaint, same old script every time. I’ll give you a pass for today but just don’t forget next time, OK? See you tomorrow, bright and early!”
On my way to a Fulbright at the Hospital Civil de Guadalajara, I asked my Virtuous Mentor, Dr. William Greenough, what might be helpful during ward rounds. He replied that knowledge at the hospital would be as good as mine and clinical skills perhaps a bit better. What we have to offer, he said, is the tradition of asking questions. In many places, attending physicians proclaim and learners record the proclamations. What we’re looking for, instead, is a creative, mutual search for the best way to take care of our individual patients. He was pointing out that knowledge is incomplete and that teaching more closely resembles learning than it does downloading.
Ward teams are complex, stylized social groups, with aspects of family, classroom, guild, and municipality. Case presentations are a focal point on rounds, and several important agendas are in play. In Michaela’s case, the attending physician asserts that the presentation must include early mention of the patient’s race. This demand makes Michaela uneasy. I agree with her. It is problematic both clinically and as a matter of social justice. What should she do?
Does the requirement lead to better patient care? The number of situations in which diagnosis or management is, or should be, affected by patient race is small. To apply evidence from randomized trials, for example, the patient at hand should be as similar as possible to subjects in the relevant clinical trials. In trials that have found significant differences associated with patients’ races , race was generally assigned based on patient self-identification. If we assert that race is relevant, we should use the same criteria as in the trials, the patient’s self-identified race. This is not usually done and is rarely as simple as it sounds. A medical records study found
that many respondents had trouble identifying with the concepts of race and ethnicity as understood by health researchers, many respondents described themselves in ways that were inconsistent with the categories included in the registration database, and many respondents were assigned categorizations in the database that were inconsistent with their self-reported identities .
One respondent identified her race/ethnicity as “Beautiful.” The idea that each person can be assigned to one of a few objective racial categories by someone else is demonstrably false; race is far more complex and subtle than that. When Michaela is accosted the second time, then, her reply is brilliant: she doesn’t know the patient’s race because she “didn’t ask.” Dr. Ross is not appeased.
The disparity seen here is part of a bigger problem with case presentations. In the sentence “This is our delightful, 86-year-old, black female,” a boilerplate formulation for some housestaff, every single word is wrong and fraught with meaning except “86 year-old.” “This” is not “ours.” She is not “black.” Unless you have the karyotype, she is not “female.” And if a 20-something calls me “delightful” when I am hospitalized, we may not get along. I am guessing that President Obama’s ward team would not, as a matter of respect, use this construction. For every patient, in my opinion, this is a matter of respect.
That initial sentence serves to establish dominance and demonstrate sophistication. How much more useful to say “Ms. J. is a 53-year-old auto mechanic and flea market enthusiast who recently lost her job,” or “Mr. O. is a 53-year-old reader, athlete, and President of the United States.” Labeling Mr. Obama as black precisely illustrates the social power and scientific incoherence of specifying race.
On the merits, then, race as a one-word identifier has a paltry scientific underpinning and is usually irrelevant to providing the best medical care. It’s hard to imagine that this vague descriptor is generally central to patient care. Featuring it prominently in a presentation is at best a distraction.
In a small 1999 study where housestaff presentations at morning report and chief’s rounds were tabulated, “Race was specified more often…(and) more often specified prominently and repeatedly during presentations of black patients. Among patients to whom ‘possibly unflattering’ characteristics were attributed, race was more likely to be specified for blacks (10 of 10) than for whites (4 of 9)” . I feel sure that these housestaff would honestly deny bias, but there it is. Implicit bias is well recognized, and unfavorable characteristics are often ascribed to members of racial and ethnic minorities. No one benefits.
Within the uneven power dynamic of a ward team, what is Michaela to do? Is Dr. Ross’s requirement so wrong that she should take the risk and speak up anyway? There are certainly such times; bigotry is not extinguished by appointment to faculty. Regardless of risk, students should speak up if they believe something wrong is being done. (In speaking up we must of course always listen up.)
If the attending physician were to insist on using the “n” word, the answer would be yes, Michaela should speak; we all should. If he calls the station clerks “girls,” however, whether or not to speak up becomes a judgment call—it’s certainly evidence of prejudice, but merely condescending, rather than a hostile slur. Dr. Ross’s belief that race is clinically relevant may be evidence of a certain bias but I don’t think that is harmful or unprofessional. In my opinion, it does not rise to a level that requires Michaela to challenge her attending physician.
Harm occurs when race is used as a proxy for characteristics stereotypically ascribed to members of a group, much as the obligatory mention of age is intended to provide an indication of the patient’s place on the vitality-to-decrepitude continuum. The ability of race and age to stand in as proxies for information that medicine needs is being questioned—often by younger members of the profession like Michaela. Michaela will make the call for herself. Other means are available to raise the issue, although they are regrettably few.
If she believes that Dr. Ross is open to discussion of the idea that race is an outdated proxy for relevant information, she could ask to meet and present her views. This should be done with an inquisitive spirit and the respect due to someone with far more experience on the subject. Exceptional clinical intuition develops with scrupulous attention to detail and conscientious follow-up over time. Dr. Ross may believe that early focus on patient race during presentations leads to better patient care. He may have confirmatory anecdotes. (As an incidental matter, Dr. Ross’s manner, which appears to do violence to a great tradition of mutual respect and openness to ideas, could instead be a pedagogical device.) Michaela would be suggesting that the risks of using race in this manner—from scientific imprecision and from unconscious bias—exceed those benefits. Michaela could submit that, in the New England Journal of Medicine’s weekly “Case Records of the Massachusetts General Hospital,” a generally reputable source, race is not given in the first sentence and often is not mentioned at all.
If Michaela feels that a collegial interaction with this faculty member is unlikely, her course evaluation might be a useful tool. The same high level of care should be used in writing this as in a personal interaction. The goal is to invite reflection about how a patient’s race functions in case presentations, not to accuse and seek punishment. If even this seems too risky, a trusted adviser might have perspective and ideas. And finally, a thoughtful letter after graduation might nudge the situation in a good direction; these small acts, gently done, are sometimes unexpectedly consequential.
In summary, at our current level of knowledge, patient race is in general not clinically useful in knowing a patient, understanding a patient’s disease, or creating a treatment plan. For a case in which it is relevant, I favor reporting race during the physical exam or, as Michaela has taught me, as part of the social history. The harm from presenting it by rote in the first sentence surely exceeds the benefit. Michaela has a difficult decision in front of her. She can confront Dr. Ross, more or less gently, or she can address him, more or less directly, through other channels. Michaela’s task now is to be involved in learning and teaching. The sociology and politics of doing this are, as always, highly local.
- Exner DV, Dries DL, Domanski MJ, Cohn JN. Lesser response to angiotensin-converting–enzyme inhibitor therapy in black as compared with white patients with left ventricular dysfunction. N Engl J Med. 2001;344(18):1351-1357.
- Moscou S, Anderson MR, Kaplan JB, Valencia L. Validity of racial/ethnic classifications in medical records data: an exploratory study. Am J Public Health. 2003;93(7):1084-1086.
- Finucane TE, Carrese JA. Racial bias in presentation of cases. J Gen Intern Med. 1990;5(2):120-121.