Does bias affect physician decision making?
Study shows need for cultural sensitivity
Do physicians treat patients differently based on race, gender, or other cultural and social differences? A recent study suggesting they do has lead to soul-searching and a renewed push for "cultural competence" in medicine.
Researchers at Georgetown University Medical Center in Washington, DC, surveyed 720 primary care physicians using an interactive multimedia tool and found race and gender disparity in referrals for cardiac catheterization.1
The physicians viewed a multimedia tool in which one of eight patient actors — two black men, two black women, two white men, and two white women — recited one of three standard scripts describing their chest pain. The patients’ medical history and personal information also were standardized.
While a majority of all patients were referred for cardiac catheterization, the odds that blacks and women with chest pain would be referred were only 60% of those of white men.
"This is the first concrete evidence that people are making different treatment recommendations based on patient characteristics," says Kevin Schulman, MD, an internist and associate professor of medicine at Georgetown.
The results point to a subtle, perhaps unconscious bias, he says. "It suggests that physicians are just like everyone else in society. Automatic perceptions incorporate society’s stereotypes. In brief encounters with patients, physicians don’t have time to override those automatic stereotypes."
The Georgetown study generated national media interest, including a segment on Nightline, the ABC newsmagazine. It troubled physicians, particularly those who have devoted their careers to addressing cultural competency in medicine.
A multitude of factors may influence physicians when they see patients from different backgrounds, says Robert Like, MD, MS. Like is a family physician and director of the Center for Healthy Families and Cultural Diversity at the University of Medicine and Dentistry of New Jersey-Robert Wood Johnson Medical School in New Brunswick, NJ. For example, clinical trials have historically underrepresented blacks and women, leading to less information on differences in symptom presentation and effective therapies. Systems also can fail patients and physicians alike, creating pressure to see patients in a short time frame.
But Like adds that the study "should be a wake-up call for us to look at ourselves as well."
Greater cultural sensitivity can begin with the opening dialogue with a patient, says Like, who is also an associate professor in the Department of Family Medicine. He has developed a curriculum for culturally competent health care.
"We have to change how we listen to patients," he says. "Not all people are going to present with the classical symptoms of angina." Physicians should ask patients what they believe is causing their symptoms and what they think might be an appropriate treatment, adds Like.
Sometimes the answers may be surprising, as in the case of a Mexican man who began stuttering as a young boy when he saw a man killed. He had reached to touch the man’s shoe. After his mother warned him that the man’s spirit would hurt him, he began to stutter and became very nervous. While still a child, he received herbal therapy and visited a spiritualist healer, but his symptoms persisted. Like tried to help the man receive both speech therapy and mental health counseling. Those efforts failed because his health plan lacked Spanish-speaking health professionals in these areas.
In other encounters, cultural differences may be more subtle and may relate to educational, social class, and racial differences as well as ethnic ones, notes Like.
He and his colleagues developed a mnemonic device, ETHNIC, to help physicians and other providers focus on the aspects of culturally sensitive communication. (See box, p. 66.) Yet beyond communication issues, the Georgetown study points to a bias that affects decision making. In patient profiles, all subjects had the same insurance status and the same job. Yet when asked in the survey, physicians perceived that black patients had a lower socioeconomic status than white patients, says Schulman.
Still, that different perception didn’t explain the different referral rates, he says. "The question that we can’t answer is what caused that bias," he says. "Our suggestion is that it’s most likely a subconscious effect."
That subconscious element is difficult to change precisely because physicians aren’t aware of it, says Julia Puebla Fortier, director of Resources for Cross-Cultural Health Care in Silver Spring, MD. "A lot of physicians would be surprised if they were faced with this information and told, This is you,’" she says.
"How do you get at attitudes and practices that you may not even recognize? How does one develop a training program for something like that?" she asks. "I think that gets at the most difficult issue that exists in health care. We primarily talk about patients, how do we get patients to change their behavior. But in this case, we’re talking about physicians. How do we get physicians to change their behavior?"
In fact, the study raised issues of bias in health care just as elements of President Clinton’s initiative on race and health got under way. The initiative seeks to eliminate disparities in six areas of health that impact racial and ethnic minority populations. It involves an outreach effort by U.S. Surgeon General David Satcher and $150 million in grants for up to 30 communities to research innovative strategies to improve minority health status. (For more information on the initiative’s goals, see box, above left.)
Clearly, patients also play a key role in determining their health care. Differences in health care in some cases may relate to different expectations. "Patients’ desires may differ by race and gender," says Schulman. "We should explore and investigate that to see whether that is true or not."
Patients also should become empowered to question the treatment recommendations and request a second opinion, he says.
In the Georgetown study, patients were passive participants and didn’t ask questions of doctors, notes Like. In real life encounters, decisions occur within a dialogue. "Patients are partners in this," he says. "To what degree do they become active in making choices?"
3. A recent study by Kevin Schulman, MD, an internist and associate professor of medicine at Georgetown University Medical Center in Washington, DC, was the first to show:
A. different treatment
recommendations based on
patient characteristics
B. different outcomes based on race
and gender
C. physician variation in diagnosis
and treatment
D. different patient preferences based
on race and gender
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content