UW News

March 19, 2012

Pediatricians' pain-medication judgments affected by unconscious racial bias, says UW study

News and Information

Pediatricians who show an unconscious preference for European Americans tend to prescribe better pain-management for white patients than they do for African-American patients, new University of Washington research shows.

Pediatricians responded to case scenarios involving medical treatments for white and African American patients for four common pediatric conditions.

“We’re talking about subtle, unconscious attitudes that are pervasive in society. Because these are unconscious attitudes, doctors aren’t aware that their racial attitudes may affect their treatment decisions,” said Janice Sabin, a UW research assistant professor in the Department of Biomedical Informatics and Medical Education, a part of UW’s School of Medicine.

She is lead author of the study published online March 15 in the American Journal of Public Health.

Sabin’s previous research showed that pediatricians display less unconscious race bias than other medical doctors or the general population. Still, unconscious beliefs can affect how doctors interact with patients, and the current study reveals that those attitudes can influence doctors’ treatment decisions.

“Coupled with known racial and ethnic disparities in health care, our findings suggest that well-meaning physicians may unconsciously treat people differently in some areas of care,” said Sabin.

Among the 86 pediatricians who participated in the study, 65 percent were female, 82 percent were white and 59 percent were medical residents or fellows. They completed three Implicit Association Tests to measure unconscious attitudes and beliefs.

The test was developed in 1998 by Anthony Greenwald, a co-author and a UW psychology professor. The test measures implicit attitudes by asking participants to quickly classify several series of words or visual images as they appear on a computer screen. The patterns of speeds in response to varied classification instructions can reveal automatically operating biases.

Sabin chose four conditions commonly treated by pediatricians – asthma, attention deficit hyperactivity disorder, urinary tract infections and pain. Case scenarios were created for each condition for both an African-American and a white patient.

For the asthma, ADHD and urinary tract infection case scenarios, doctors did not show an association between unconscious attitudes about race and treatment decisions for the two patients. However, recommendations for optimal pain treatment decreased for the African American patient as doctors’ pro-white bias increased.

“Implicit biases are surprisingly pervasive, and in certain circumstances they can affect how people behave,” Sabin said. She said her findings “indicate that more research should be done to see if unconscious biases affect real-world medical care and treatment decisions, especially for pain management.”

“This is exactly the type of result that was anticipated by the Institute of Medicine’s landmark 2002 Unequal Treatment study,” Greenwald said. “That study and other studies found, among other indications of troubling health care disparities, underuse of pain medication for African American patients.”

Because physicians are likely unaware of unconscious attitudes and beliefs and the unintended disparities that may result, incorporating awareness of personal bias and methods to avoid the influence of bias on decision-making into medical education, continuing medical education and training of health professionals is necessary for health sciences education, Sabin suggested.

The U.S. Department of Health & Human Services, Agency for Healthcare Research and Quality, the National Institute of Mental Health and a University of Washington Magnuson Health Scholars Award funded the study.

###

For more information, contact Sabin at 206-616-9421 or sabinja@uw.edu.