A Picture of Modern Medicine: Race and Visual Representation in Medical Literature

Abstract

Background

There has been a recent focus on the impact of race on health equity, which has revealed unsettling results. Multiple studies have shown that the underrepresentation of minorities in medical education such as course slides, pre-clinical lecture material, case studies, and textbooks impedes racial equity in the practice of medicine.

Aim of study

In this study, we aimed to survey the landscape of published imaging in modern medicine to understand the degree of racial diversity represented in current biomedical literature.

Methods

We performed a photogrammetric analysis of medical images from the New England Journal of Medicine representing various medical fields and geographic regions to examine implicit biases with regards to human skin color.

Results

Overall, 18% of images depicted non-white skin tone but there was considerable heterogeneity in the percentage of non-white medical images published from different geographic regions and specialties (ranging from 0% to 67%).

Conclusions

Unfortunately, these results suggest that there is an underlying implicit racial bias in published images from medical literature with an underrepresentation of minorities compared to the general population, which could also contribute to inequities in health care. It is critical that health care providers, educators, and trainees promote cultural competency and work to understand the multifaceted influence of race and culture on the daily experience of patients in the modern healthcare system. We hope this study will encourage authors to critically evaluate their medical images for implicit bias so that documented photography in scientific literature may better reflect the populations we serve.

Introduction

Race and ethnicity have been used as justification for discrimination, prejudice, and marginalization of minority groups in many aspects of society.^1,2 Racism in medicine, whether explicit or implicit, is rooted in over 2500 years of history² and has pervasive effects throughout all aspects of the modern healthcare system. Although it is rarely acknowledged and often ignored, racial bias can distort clinical decision-making and directly impact the daily experience of patients as well as the quality of the care they receive.² While there are inherent racial and ethnic differences in the causes, expression, and prevalence of human disease, the interplay between these factors and bias, culture, socioeconomic status, and other related factors on the development and treatment of disease remains nebulous.¹

Race in biomedical research typically follows the classification scheme employed in the 2000 U. S. Census and includes five major classifications based on the geographic origin of populations: black or African American, white, Asian, native Hawaiian or other Pacific Islander, and American Indian or Alaska native.³ Ethnic background is a less clearly defined concept that includes influences such as cultural traditions, common histories, and religion.¹ There has been debate about whether race and ethnic identity are truly biologic constructs or if they are socially driven. Studies have shown that genetic variations are found across all populations regardless of race, and that there is greater genetic variability within the same racial group than between different ones.^3,4

Both race and ethnicity have arisen through the interplay of geographic, social, and cultural influences on the inherent biology of populations. This suggests that race and ethnicity may be fluid rather than static,¹ supporting the concept that race is a social, not scientific, construct. Regardless of historical changes in these definitions, race and ethnicity are currently used as primary identifiers of research participants. Many funding agencies, regulatory agencies, and biomedical journals including the U.S. National Institutes of Health (NIH),⁵ Office of Management and Budget (OMB),⁶ U.S. Food and Drug Administration (FDA),⁷ and the International Committee of Medical Journal Editors (ICJME)⁸ have guidelines that mandate or recommend the use of racial and ethnic identifiers in research. With recent breakthroughs in human genome sequencing and modern genetics, questions remain about the utility of racial classification in research, as there is no true genetic basis for race.¹ However, in epidemiology and clinical research, these distinctions are useful in understanding the complex interplay between societal and environmental influences with genetic and biological factors, which are key determinants of medical outcomes.

Health equity refers to the absence of disparities in health between groups with different degrees of social advantage or disadvantage.⁹ In medicine, there has been a recent focus on the impact of race on health equity, which has revealed unsettling results. Multiple studies have evaluated the representation and effect of race and skin tone in medical education, patient assessment, and clinical care. Studies have shown that there is an underrepresentation of racial minorities in course slides, preclinical lecture material, case studies, and clinical research used in medical education throughout the United States.10, 11, 12, 13, 14, 15 A similar trend is also present in medical textbook imagery, which lack skin tone diversity at the textbook, chapter, and topic level.¹³ These implicit biases present in medical education have tangible repercussions that directly impact patient care and foster inequities in clinical practice. In one study, medical students and residents were found to believe that black patients have biologic differences that cause them to experience less pain and were biased in their treatment recommendations.¹⁶ Another study revealed that black and Hispanic patients were significantly more likely to be diagnosed with late stage melanoma than white patients, likely resulting from suboptimal prevention efforts in minority populations.¹⁷ Other studies have shown that racial diversity is not preserved in clinical trials,18, 19, 20, 21 access to care and treatment options,^22,23 or in the career advancement of trainees.^24,25

Rather than ignoring differences in race and ethnicity, scientists and physicians should incorporate these issues into all aspects clinical research to better understand the societal impact of race in health care outcomes for all patients. As discussed previously, medical research is often held accountable for reporting patient demographics, including race or ethnicity, but there is no standard for medical imagery. Especially in the era of #BlackLivesMatter, it is critical for there to be racial equity and inclusivity in all aspects of medicine, including clinical research, as implicit bias can significantly affect patient care, experiences, and outcomes.¹⁵ While studies have evaluated skin-tone and racial diversity in medical education and textbooks, there is a paucity of knowledge on the racial and skin-tone diversity of images in biomedical literature. In this study, we aimed to survey the landscape of published imaging in modern medicine to understand the degree of racial diversity represented in current biomedical literature.