It has been more than 2 months since the death of George Floyd at the hands of a white police officer in Minneapolis, MN, USA. The ensuing shock and outrage led to protests across the globe, in union with the Black Lives Matter movement against institutionalised racism and systemic oppression. Discrimination permeates all facets of society, and health care is no exception.
In a media briefing on the current COVID-19 pandemic, UK politician Michael Gove said that “the virus does not discriminate”. But as evidenced by the shockingly disproportional impact the virus has had on some racial and minority ethnic groups, this statement could not be further from the truth. Data from the UK and USA have shown that Black and minority ethnic individuals are more susceptible to the virus than white people and have a greater risk of mortality. Such disparity is associated with socioeconomic inequalities and has exposed the substantial impacts these can have on the health of Black and minority ethnic individuals.
Despite their promises, digital health approaches have to some extent contributed to these disparities by not considering race or ethnicity; for example, melanoma diagnostic apps trained on images of white skin have difficulty detecting cancerous skin lesions in darker skin. Even failing to sufficiently account for variables that are correlated with race, such as health-care expenditure, can create a wedge between health needs and care received. When race clearly plays a causative role in health outcomes it should be incorporated into digital health approaches to prevent bias.
For instance, using multivariate logistic regression, Black race was shown to be a negative predictor for obtaining MRI/ultrasound fusion biopsy in cases of suspected prostate cancer—despite Black men being at greater risk of developing and dying from this disease than white men. It has been suggested that almost 85% of this excess mortality risk is due to reduced access to care and treatment. The driving force for this could be socioeconomic, but even when this barrier is removed, Black patients could be subjected to racial prejudice on the part of providers.
Racial bias also affects those doing health research. Poor diversity in the scientific workforce is well known and extends to acquisition of research funding. Black scientists are less likely to be awarded funding, partly because of their topics of interest, which ironically include health disparities and patient-focussed interventions. Those that receive funding often receive less than their white counterparts.
It is encouraging to see that steps have been taken towards addressing these problems, such as the introduction of new legislation and the development of dedicated research centres to identify and narrow disparity gaps and improve access to health care. The WHO might introduce prequalification of digital tools to ensure these tools do not deprive marginalised groups of the care they need. But the paucity of research into interventions to improve patient and provider relations and overcome bias is telling, as is the limited representation of Black and minority ethnic individuals in clinical studies. These biases are further exacerbated by the shortage of appropriate stratification of research findings according to race and ethnicity. Although surveys such as the National Health and Nutrition Examination Survey and the UK Biobank include several classifications, studies based on these classifications often group individuals into much broader categories (including the ominous “Other” category), and so inadvertently cover up important racial disparities.
The challenges facing the lives of Black people and ethnic minorities are considerable and the editors of The Lancet Digital Health cannot be complicit. In congruence with The Lancet Group's anti-racism pledge, we stand in allegiance with the Black Lives Matter movement and commit to redressing racism for global health equity. We endeavour to commission content investigating racial bias in digital health, showcasing progress in the field towards overcoming this issue, and highlighting areas of medicine and health care that are most affected. We will profile researchers who are actively working to tackle health disparities, and strive for greater racial and geographical diversity in our International Advisory Board.
But we humbly acknowledge that much more needs to be done to bring about sustainable change. We would like to take this opportunity to reach out to you, our readers, and invite your ideas on how The Lancet Digital Health can better serve you and be an agent for change: to fight against social injustice and racism.
For more on socioeconomic differences and COVID-19 see BMC Med 2020; 18: 160
For more on bias in melanoma apps see BMJ 2020; 368: m363
For more on bias in a US health-care algorithm see Science 2019; 366: 447–53
For more on reduced access to prostate cancer care and treatment see Prostate Cancer Prostatic Dis 2019; 22: 125136
For more on the disparity in grant awards see https://www.newscientist.com/article/mg24632882-500-minority-scientists-still-face-many-forms-of-institutional-racism/
For more on changes in legislation in the USA see JAMA 2020; published online June 4. https://doi:10.1001/jama.2020.10531
© 2020 Michal Urbanek/Shutterstock.com