"No physician is racist, so how can there be structural racism in health care?" wrote the Journal of the American Medical Association (JAMA) in a tweet promoting its podcast in early March 2021. In the podcast, Dr. Ed. Livingston, a white male surgeon and JAMA Deputy Editor, dismissed the concept of structural racism, protesting, "many of us (physicians) are offended by the concept that we are racist." Backlash to the since-deleted tweet and the podcast was so swift that Livingston subsequently resigned from his post.
Livingston's sentiments naively suggest that a medical degree inoculates physicians from racism. It clings to the mistaken notion that racism only harms Black patients when actively schemed. This view flies in the face of countless studies, many appearing in JAMA itself, demonstrating widespread barriers to health care for Black patients from basic preventive services to cutting-edge treatments. Worst of all, it obscures how stubborn racist legacies reinvent themselves.
Indeed, even today, modern-day segregation within hospitals kills Black patients, as statistics have shown.
Take surgery for instance, commonly performed on older adults. Earlier analyses reveal that the mortality rate for common surgical procedures can be 35% or more higher among Black patients as opposed to white patients. These disparities are commonly linked to a history of redlining that has left Black communities disproportionately served by poorer hospitals. Even when elite hospitals are in Black communities, their patients tend to be disproportionally white.
But what if those inequalities are removed by examining Black and white patients treated within the same hospital? To investigate this, we analyzed national Medicare claims from older beneficiaries who underwent heart bypass surgery, a complex and technical procedure. Our worrying results, published in Circulation: Cardiovascular Quality and Outcomes, identified two issues.
First, we found physician teams treating Black patients did not often overlap with those of white patients undergoing heart bypass surgery — even within the same hospital. Hospital segregation was vastly higher than observed in care settings such as nursing homes or neonatal care units. Indeed, care by cardiac surgery teams may be more racially segregated than a high school lunchroom. In some hospitals, some providers cared exclusively for Black patients, while others treated none.
Second, after controlling for other health variables, we found Black patients had significantly higher mortality rates when treated at hospitals with higher levels of segregation. And, encouragingly, when segregation was lower and provider teams treated both Black and white patients, the mortality difference went away.
It is hard to know why such segregation continues. Hospital segregation may seem a specter from a long-past Jim Crow era. When Title VI of the Civil Rights Act prohibited the Federal government from making Medicare payments to facilities practicing racial discrimination, de jure segregation ended nearly overnight. But de facto segregation has proven harder to banish.
Of course, Black patients may prefer racially familiar doctors; yet sadly, there are too few Black surgeons to explain the mortality difference. Further, racially concordant interactions generally lead to enhanced satisfaction and better clinical outcomes. Taken together, the contemptuous view that underqualified Black doctors are the problem is unlikely.
The assignment of patients to provider teams is an opaque and complex process, largely insulated from public scrutiny and political control. It is possible that providers make profit-minded decisions, choosing patients through payment criteria like insurers, which might skew racial differences. Racial divisions in referrals among providers may also exist through informal relationships.
Even with the best of intentions, a provider team that becomes "the one for Blacks" may begin to view Black patients with a jaundiced eye, expecting poorer clinical outcomes or missing fresh insights from working across teams. One important aspect of this problem is evidence showing that physicians treat patients of different races differently; for example, falsely believing Black patients can withstand more pain than white patients and thus undertreating their pain. Most troubling, Black patients may simply be assigned to weaker provider teams.
In short, hospitals are not immune to the segregation and structural racism that infects so much of American life, from education to housing to employment. But the good news is that our research also shows the immense benefits of meeting the challenge head on.
Hospitals could start to measure how they distribute patients across provider teams and actively institute measures to desegregate patient care where needed. They could also develop activities around implicit bias and cultural humility, while advancing strategies to achieve greater workforce diversity.
Turning a blind eye because doctors do not consider themselves racist lessens our dedication to rooting out institutional harms. Surgeons do not have to set out to hurt people in order for the color of a patient's skin to determine their fate. Our findings make clear that hardly seen structural racism in health care can be a matter of life and death.
Ekow N. Yankah is a professor of law at the Benjamin N. Cardozo Law School of Yeshiva University. Dr. Brahmajee K. Nallamothu is Professor in the Division of Cardiovascular Diseases and the Department of Internal Medicine at the University of Michigan. Dr. John M. Hollingsworth is a Professor of Urology at the Center for Healthcare Outcomes & Policy at the University of Michigan.