DURHAM, N.C. - Racial and gender biases appear to be present in medical students at the earliest stages of their medical training, even before clinical experience can shape their perception about the impact of those variables on diseases, a study has found.
Researchers found that some first and second-year medical students already exhibited biases based on patient race and sex.
Specifically, the researchers found that white male medical students tended to rate a black female heart patient as having a less desirable health status (quality of life) than a white male patient, yet they considered the white man's coronary condition to be more severe. Both "patients" in the study were actors who were carefully scripted to say the same thing and reflect the same degree of illness.
In contrast, minority and female medical students rated the actors as having the same quality of health status, based on what they saw and heard based on a video presentation.
Researchers from Duke University Medical Center, the University of California at San Diego, and Georgetown University Medical Center conducted the study, the findings of which are published in the May 1 issue of The American Journal of Medicine and was funded by the U.S. Agency for Healthcare Research and Quality and the National Library of Medicine.
While other studies have found bias in health care, many have suggested that it may be due to patient preference for less aggressive treatment, to physiological differences between race and gender that influence the course of a disease, or to socioeconomic factors.
One issue not addressed by prior research is whether students come to medical school with specific subconscious biases about patient groups, or if this behavior is learned during medical school, said the senior author, Duke internist Dr. Kevin Schulman.
"In this increasingly multicultural society, the issue of whether patient characteristics influence clinical decision making is of critical importance," said Schulman, who heads the Center for Clinical and Genetic Economics at the Duke Clinical Research Institute. "Knowing that race and sex biases may be present in medical students even before the start of their clinical training suggests the need for cross-cultural education programs at the earliest stages of medical school."
The study was conducted with a multimedia survey tool that allowed first and second-year medical students to interact with a simulated patient. The students, who were not told the nature of the study, were randomly assigned to interact with a video of either a 55-year-old black female actor or a 55-year-old white male actor, who read from identical scripts describing chest pain. The professional actors were dressed in identical hospital gowns, filmed in the same rooms, and directed so that symptoms sufficient for a diagnosis of clinical angina (chest pain) was presented in the same way. The multimedia tool also provided the medical students with the same medical history and personal information, regardless of the actor.
All of the students had been taught the underlying physiological relation between coronary artery disease and chest pain, but they had not had much exposure to heart disease patients "and thus have limited clinical knowledge of the prevalence of coronary artery disease by race and sex," said Saif Rathore, first author of the paper. He helped conduct the study when he was at Georgetown University Medical Center and is now at the School of Public Health at the University of North Carolina at Chapel Hill.
Students rated the perceived quality of life, or "health value," of the angina on a 100-point scale. They also classified the actor's chest pain as definite angina, probable angina, or non-anginal pain, and rated their probability of heart disease. The students then rated how likely the "patient" was to over-report or exaggerate pain and discomfort, not show up for follow-up appointments, participate in prescribed cardiac rehabilitation, sue for malpractice and comply with medical treatment.
Thus, the study was not designed to assess whether patient race or sex, or combinations of race and sex, influenced the medical students, "but whether all medical students made the same assessment of quality of life when identical symptoms were presented by different patients," Rathore said.
Of the 164 students who enrolled in and completed the study, 57 percent were male, and 14 percent were minorities. The researchers found:
- White students reported a significantly higher mean health value for the white male patient actor compared with the black female patient actor, whereas minority students' assessments did not vary by patient actor.
- Male students assigned a significantly lower health value to the black female patient actor. In contrast, female students' assessments did not vary by patient actor.
- Although the case presentations were identical, students were less willing to provide a diagnosis of definite angina for the black female patient actor compared to the white male patient actor.
- The black patient actor was viewed as friendlier, a better communicator, and as having a more positive affect than the white patient actor. However the black female patient actor was perceived to be less likely to obtain follow-up care.
"The data suggest that white male medical students were responsible for most of the differences," reported Dr. Kevin Weinfurt of Duke University Medical Center, lead statistician on the study. "The white students reported higher values for the white male patient, while minority students did not differ by patient, though there were too few participants in the study to detect any difference other than a large one."
Journal
The American Journal of Medicine