Women and minorities who are traditionally underrepresented in medicine (URM) -- a group that includes blacks, Hispanics, American Indians, and Alaskan and Hawaiian native peoples -- comprise just under one-fourth of practicing physicians and slightly more than one-fourth of radiology residents, the study team reported.
According to results of the eight-year survey, women and minorities remain underrepresented at the resident trainee level, compared with their numbers as medical school graduates. The researchers also found no significant increase in female or minority resident representation in the past eight years.
"Our main finding, in terms of groups that are underrepresented in medicine, is that these groups are underrepresented in radiology as well, compared to the U.S. population," co-author Dr. Curtiland Deville told AuntMinnie.com.
In their analysis of ethnicity and gender in the U.S radiologist workforce, the authors found that underrepresented groups and females "are significantly underrepresented as residents, academic faculty, and practicing physicians compared with the overall U.S. population and medical school graduates."
"Anecdotally, when you go to the conferences or you're in the environment, it seems like there's underrepresentation, but there were essentially no studies documenting what is the representation, how does it relate to other specialties, and the population numbers," Deville said.
Hypotheses offering explanations for the disparities are many and wide-ranging, encompassing everything from unspoken bias to different comfort levels with regard to radiation exposure. The ideas for correcting the disparities, from student interest group programming to creating opportunities for mentorship, have been just as broad, yet the trajectory of female and minority underrepresentation in radiology has remained largely unchanged, wrote Deville, Dr. Christina Chapman, and colleagues from the University of Pennsylvania and Oregon Health and Science University (Radiology, July 30, 2013).
Data were obtained from the American Medical Association, the Journal of the American Medical Association supplements, the American Association of Medical Colleges, and U.S. census registries for 2010 for practicing physicians, academic faculty, residents, subspecialty trainees, residency applicants, and medical school graduates.
The researchers looked for significant differences in diagnostic radiology resident representation for academic years 2003-2004 to 2010-2011 as well as for 2010, with data compared among the 20 largest residency training programs.
They found that women and URM -- encompassing blacks, Hispanics, American Indians, Alaskan Natives, Native Hawaiians, and Pacific Islanders -- are underrepresented as practicing physicians (women, 23.5%; URM, 6.5%), academic faculty (women, 26.1%; URM, 5.9%), and diagnostic radiology residents (women, 27.8%; URM, 8.3%), compared with the U.S. population (women, 50.8%; URM, 30%) (p < 0.001 for all).
Women and underrepresented minorities are present in higher numbers as residents than as practicing physicians; however, both groups remain underrepresented at the resident trainee level, compared with their proportions as medical school graduates (women, 48.3%, URM, 15.3%), the study team wrote.
In addition, during the past eight academic years, there has been no significant increase in women or underrepresented minority residents (both p > 0.01), suggesting that the future holds no dramatic change in representation of these groups.
"It's been pretty level over the past eight years," Deville said.
Radiology fares worse
Radiology fared worse than medicine as a whole, the researchers found. Among the 20 largest medical specialties, diagnostic radiology ranks 17th in female and 20th in minority participation, even though radiology is the ninth largest medical specialty, according to the authors.
"Radiology ranks at the bottom or near the bottom for specific groups, whether you're looking at women, blacks, Hispanics, or all underrepresented minorities combined; it ranks 6th to 20th for all those groups," Deville said. "So for whatever reason, even though there is some representation in medical schools, it doesn't seem to be translating into diagnostic radiology."
Efforts have been made to change the numbers on a variety of fronts -- for example, trying to speed up the pipeline of minority medical student graduations -- but the data have shown that these minority groups decline in representation when transitioning from medical school to diagnostic radiology residency. Current data showing rough parity between white and nonwhite students among college-age individuals means there is a lot of work to do before some type of parity can be achieved in the physician pool as well, Deville said.
Previous studies have tackled everything from work attitudes and hours to job satisfaction, length of training, patient contact, exposure to radiology mentors, and basic science and clerkships in medical schools, the authors wrote.
"In a survey of women physicians, radiologists, despite having higher incomes, reported less job satisfaction and work control, they reported working more hours, they were more likely to believe they were overworked, and they noted sexual harassment," the group wrote. "Although interest was unchanged after a seven-week introductory diagnostic radiology course to first-year medical students, the exposure revealed variable sex trends, with women rating more patient contact as a positive influence on residency selection and technological work as a negative influence."
Whatever the reasons for underrepresentation and whatever the solutions to the disparities may be, they will be multifactorial, Deville said.
"There will be different levels of interest across different groups, but then there are other factors such as exposure, mentorship, unconscious bias -- which is essentially a nice way of saying racism or sexism," Deville said. "Is it that they were exposed, but then they got harassed or saw certain teachers that didn't appeal to them, and they were turned off by it?"
Looking for clues about interest in the field of radiology, a survey of blacks and Hispanics found that these groups rated patient contact and the opportunity to make a difference as the two most important factors influencing their choice of medicine as a career. Exercising social responsibility came next, followed by educating patients about their health.
In a previous survey of radiation oncology trainees, the researchers found that women and minorities submit fewer applications, and the same phenomenon seems to apply in radiology. "They are interested, but maybe they don't have the appropriate guidance and mentorship showing them what they need to do to secure a position in radiology or radiation oncology," Deville said.
Addressing the problem
Emphasizing opportunities for patient contact and education in diagnostic radiology may attract more women and minorities, the authors wrote. Unconscious bias may be worth investigating as well, as one study showed that female laboratory manager candidates were rated as less competent and less hireable than men with equal qualifications -- and they were offered less career mentoring.
Addressing gender- and race-based disparities could also include the following:
- Initiating programs to attract preclinical students to diagnostic radiology and subspecialties
- Advocating for third-year medical student radiology rotations
- Creating opportunities for mentorship early in the pipeline of science, technology, engineering, and mathematics
- Evaluating program effectiveness by race, ethnicity, and gender to determine insights and interventions
Given the persistent underrepresentation in radiology among women and minorities in an increasingly diverse society, future training efforts should address increasing diversity in the field, the authors concluded.
In Deville's view, the best way to get at the core reasons for the disparities is also the hardest -- i.e., trying to survey the trainees themselves, such as talking to those who did or did not get a residency match. This is difficult because the National Resident Matching Program (NRMP) doesn't break down the data by race and gender, though there are other ways to approach the question, he said.
"What we're currently doing is looking more broadly at specialty selection and trying to look at whether there are themes, because there won't be a one-size-fits-all solution to underrepresentation for women and minorities," Deville said. "There are some interesting trends when you look across different specialties as to who has been increasing where, so we're looking more broadly at specialty selection in general."