According to a new study published in the American Journal of Medicine, many students may enter medical school with existing patient biases based on race and gender.
After conducting a study involving 164 first- and second-year med students, a team of researchers from Duke, the University of California at San Diego and Georgetown University found that such biases may not be conscious, but can have an effect on clinical decisions.
"We found that medical students' perceptions of health states differ for a white male patient actor compared with a black female patient actor," the team wrote in the May issue of AJM.
"These differences suggest that patient characteristics influence perceptions among medical students even before the start of their clinical rotations and suggest the need for cross-cultural educational programs at the earliest stages of medical training."
Specifically, the study suggests that white students are more likely to perceive a white male patient as healthier than a black female patient with similar symptoms.
In addition, a black female patient would be perceived as less likely to obtain follow-up care.
After completing a questionnaire, students were randomly assigned one of two videos which featured an actor portraying a patient experiencing chest pains.
On the first tape, a 55-year old black female actress was used; in the other tape, a 55-year old white male actor performed.
Both actors were portrayed as professionals, dressed in the same type of hospital gown and filmed in the same room. They also used the same script.
Students were asked to classify the patient's chest pain and rate the patient's health-state, the likelihood of the patient complying with treatment and the probability of the patient having coronary disease.
Researchers noted that the lack of student diversity in the study's sample makes it hard to draw any definite conclusions: Fifty-seven percent of students were male and only 14 percent were minorities.
"There were two effects of such a small enrollment of minorities," said Kevin Weinfurt, one of the study's authors and a Medical Center instructor of psychiatry and behavioral science. "Any existing bias would have to be large to be detected and it was not possible to examine results by both race and gender."
Weinfurt explained the study could not determine if a certain bias exists in white male students-it could only detect the bias in male students or white students, but not in the cross-section of the two.
Associate Professor of General Medicine Dr. Kevin Schulman, one of the study's authors, stressed that the point of the study was not to find out which bias predominates in medical students, but only to discover if there is one to begin with.
"If we believe that labeling mechanisms carry through to the patient treatment, we have to begin to learn to deal with that," he said.
Edward Holmes, Dean of the School of Medicine, wrote in an e-mail that Duke has already successfully instituted a program for medical students designed to deal with issues relating to race and bias.
"We are also developing a similar program in our new medical school curriculum that will include all clinical departments," Holmes added.
Get The Chronicle straight to your inbox
Signup for our weekly newsletter. Cancel at any time.