Unconscious Bias may Impede Diversity in Radiology
Thursday, Nov. 30, 2017
Developing an awareness of unconscious bias against women and underrepresented minorities and taking steps to mitigate its pernicious influence is vital to ensuring that radiology reach its full potential, according to presenters at a Wednesday session.
Radiology has been plagued with what session moderator Margaret Szabunio, MD, from the University of Kentucky in Lexington, KY, referred to as a "leaky pipeline," or a disparity between the percentage of women in medical school and the proportion who go into radiology. Women made up only 25.8 percent of radiology residents in 2015, down from 27.4 percent in 2005 and a far cry from the almost 50 percent of female medical students overall. Diagnostic radiology also ranks low in diversity — 19th of 21 medical specialties, according to a 2014 Journal of American College of Radiology study.
Addressing this disparity means understanding and taking on unconscious bias — a pervasive tendency, deeply rooted in unconscious feelings, to make assumptions about people based on race, gender and other characteristics, according to M. Elizabeth Oates, MD, chair of the Department of Radiology at the University of Kentucky College of Medicine in Lexington. Superficial information such as a person's first name or their photograph on an application can provoke negative assumptions, often to the detriment of women and minorities. Dr. Oates shared results from a study in which changing the name from "John" to "Jane" on the exact same application for a laboratory manager position resulted in a lower rating and a lower salary offer.
"Unconscious bias exists in all of us," she said. "We need to recognize and accept it and apply de-biasing strategies intentionally and systematically in the recruitment process."
Strategies can be implemented every step of the way to attract a more diverse pool of radiology residents and fellows, said Madelene C. Lewis, MD, from the Medical University of South Carolina in Charleston.
The pictures and descriptions an institution chooses for its web site should suggest an inclusive atmosphere, Dr. Lewis advised. She also recommended that staff members create a shared vision of the ideal candidate, conduct the interviews in a structured way and avoid making negative inferences about family obligations or gaps in training. The selection committee should be composed of people from different ages, ranks and genders, she said. Artificial intelligence may have applications too, as it can be programmed to ignore demographic information about candidates and make screening decisions based on data points.
"Awareness and willingness to engage in reflection are the first steps and how change begins," Dr. Lewis said.
One way to fix radiology's leaky pipeline is to reach out to women and minorities early in their medical education, said Katarzyna J. Macura, MD, PhD, a professor in the Johns Hopkins Medicine Department of Radiology in Baltimore, MD. Dr. Macura helped develop the Pipeline Initiative for Enrichment of Radiology (PIER), a mentoring program designed to attract women and minorities to the profession. Five students were recently selected as the inaugural class of PIER scholars.
"We want to cast a wider net," Dr. Macura said. "We don't want to overlook any wonderful candidates because we have these biases."