In the summer of 2006, an African-American high schooler was critically injured in an East Baltimore shootout after being grabbed by a drug dealer running from the police. She was rushed to Johns Hopkins Hospital where trauma surgeon Adil Haider was the attending on call. She underwent an immediate operation to save her life, and then spent nearly two weeks in the critical care unit, requiring multiple additional surgeries. When she came to, she suddenly refused all medical care.
“Here you have this young girl who had this miraculous save initially, but now she is not willing to interact with anyone on the medical team,” Haider, now a trauma surgeon at Brigham and Women's Hospital at Harvard Medical School, says. “She is on the surgical floor, and she’s refusing all care. And when you asked folks about that, most people thought, ‘This girl is ungrateful for all that has been done for her, for all of our care. It’s OK to be angry at her, it’s OK to be mad.’ But we totally missed the boat.”
For Haider, his and others’ response to this case revealed a disturbing truth about trauma surgeons: they could be just as susceptible to racial bias in treating patients as physicians in every other field of healthcare. His patient wasn’t actually ungrateful or difficult—she was in the beginning stages of post-traumatic stress disorder (PTSD) from being held hostage and getting shot.
“We didn’t give her the benefit of the doubt because she came from a community that has endemic violence and we were quick to judge,” he says. “Meanwhile that had nothing to do with it. She just was a little girl who was scared.”
Months earlier, another teenager had come into the hospital, Haider says. She was the daughter of physicians who flipped her new convertible while driving to school and had severe, disfiguring road rash on her face. “Terrible injuries to the face, but no life-threatening injuries,” he says. “And yet, there was a line of psychiatrists and others out the door from the beginning, to help her cope with her potential disfigurement. And rightly so! But nobody was trying to take care of that other young girl like that. That’s what led to our first study thinking about unconscious bias.”
It’s been well documented that racial disparities exist in almost every area of health. African Americans have higher death rates for almost all the leading causes of death, like heart disease, cancer, stroke, diabetes, kidney disease, hypertension, and other chronic illnesses.
It's also been shown there are racial disparities in pain management, for how much and what type of pain medication patients receive. One study found that these biases could stem from false beliefs about biological differences between blacks and whites, like that African Americans' "skin is thicker." In action, this means that black patients have been shown to receive fewer opioids than non-Hispanic whites for certain kinds of pain. (Some researchers think this could explain why the opioid crisis primarily affects white people, with 82 percent of overdose deaths occurring among caucasians.)
Spanish-speaking Latinas don’t receive as many epidurals during childbirth as non-Hispanic caucasian women, and black and Hispanic women are less likely to be offered breast reconstructions after mastectomies than white women. Studies show that African Americans have more amputations than white patients, and that black mothers are three to four times more likely to die in childbirth than white mothers.
Trauma, Haider says, was thought to be different. Emergency rooms were supposed to be the “great equalizer.” “When you come to the [emergency department], we don’t check anybody’s insurance, we take you right in," he says. "We supposedly don’t care what you look like or how you present, and we give everyone the same exact care. At least, that’s what our idea is.”
But that’s not what he found when he completed his first study to see if there were racial disparities in trauma care outcomes. In 2008, Haider and a group of trauma surgeons found that black children had worse outcomes after traumatic brain injuries; they had more difficulties with eating, talking, and walking after having brain injury compared to white children. The study used data from over 40,000 children from the National Pediatric Trauma Registry.
When he presented this paper at an American Association for the Surgery of Trauma (AAST) conference, it was met with hostility. One surgeon wrote a letter to the editor to the Journal of Trauma, calling Haider’s work “extremely troublesome,” and that his “conclusions are clearly flawed and could have unintended consequences, and are an insult to those that take care of injured children.”
This was another wake-up call for Haider. He needed to keep pursuing this work, but telling surgeons who save lives that there are racial disparities in their patients' outcomes was a provocative stance. Any further research he pursued had to be even more thorough.
In Haider and his colleagues’ next study, they controlled for more factors than the first—like insurance status. They found that the interplay between race, insurance status, and health outcome also revealed some big disparities, even within the same race. An uninsured white patient had a 50 percent higher chance of dying from trauma injuries compared to a white patient with insurance. When they compared insured white and black patients, black patients still had a 20 percent increase in odds of death. “Being black does increase your mortality, even if you’re insured. But then the crazy thing was when we compared the insured white patient versus the uninsured minority patient," he says. "There we found a nearly 80 percent increase in odds of death.”
Edward Cornwell, the former chief of trauma surgery at Johns Hopkins and the current chair of the department of surgery at Howard University, was Haider’s colleague, mentor, and collaborator on those first papers. He says if you'd asked him before if race or socioeconomics would lead to trauma outcome differences, he wouldn’t have believed you.
“Why should insurance status make a difference in the one disease where critically injured, uninsured patients gains access to the system every bit as quickly as his critically injured, insured counterpart?” he says. “Why should there be differences? And still, there are.”
Marie Crandall, a trauma surgeon at the University of Florida College of Medicine Jacksonville, says that trauma already discriminates. Minorities are more likely to be overrepresented in terms of car crashes, violent crime or violent injuries, and most other injury mechanisms. But outside of that, just being black, Hispanic, or uninsured means you're more likely to die from those blunt injuries, gunshot wounds, motorcycle crashes, and head injuries, she says.
“That opens up a whole other Pandora’s box because we know this but we don’t know why,” she says. “If you control for injury severity, you wouldn’t think that race and insurance status would have that much of an impact, but it does.”
There isn’t a single answer, but understanding the different factors can help surgeons improve care. “There are a couple of possibilities,” Crandall says. “One is access to care. One is baseline co-morbid illness. Another might be the chronic stress hormones that are related to being of a lower socioeconomic status, strata, or institutionalized racism or structural violence associated with poverty.”
Watch more from VICE:
Crandall has spent many years studying one of those potential factors: access to care. She’s researched the association between trauma outcomes and transport times, finding that if you were shot more than 5 miles away from a trauma center in Chicago, you were 23 percent more likely to die, even when controlling for race, insurance status, age, injury mechanism, and injury severity. Her work has influenced the city of Chicago to open a Level 1 trauma center in the south and southeast sides of the city, which will open in May of this year, and hopefully reduce those disparities.
But even when a person of color reaches a hospital, it’s likely that they’re not at the best possible trauma center. Haider has compared hospital rankings to the proportion of minority patients they serve, and found that 85 percent of hospitals that treat mostly minority patients are trauma centers with higher mortality rates. “So there’s a handful of trauma centers that aren’t so good, and we found that those centers are predominantly caring for minority patients,” he says.
But it’s not about putting blame on those trauma centers, Haider says. It all goes back to the interplay with race and insurance. These hospitals treat a much higher portion of uninsured patients—and people of color are more likely to be uninsured—so the hospitals don’t end up with the same incomes as other hospitals for staffing, equipment, medical testing, and more. “It’s a vicious cycle,” Haider says. “They’re not getting the patients who have the money, they’re already under-resourced, so now they’re doing worse and worse, and taking care of these very complex patients. We need to support these hospitals rather than penalizing them.”
Cornwell thinks insurance status may be a marker for other important factors too: that a patient has a primary care doctor, knows their blood pressure, or their lipid profile. Patients may have different health behaviors if they’re insured, like monitoring their chronic illnesses, or simply maintaining overall wellness because they check in with a doctor more frequently. “These are the things that you bring to the table the night you’re stabbed or shot or in a motor vehicle crash,” Cornwell says.
Haider thinks these more linear reasons for disparity, like access and insurance, still don’t explain all of the differences in trauma outcomes. In his studies where all those aspects were controlled for, being black and getting a traumatic injury still meant you were more likely to die than an injured white person. And so, he returned to the question of implicit bias, and how it could be subtly impacting outcomes.
Implicit biases don’t mean extreme racism—just subtle preferences for people who look like you, or who you can relate to economically and socially, Haider says. Doctors might be more prone to give those patients extra attention or care, or be more instinctively empathetic. Like with the teenage girl in Baltimore: he says some doctors may more immediately relate to or empathize with a white car crash victim than a black gunshot victim.
Haider and colleagues conducted multiple studies on unconscious preferences, testing more than a thousand medical students, nurses, and surgeons. They used the Implicit Association Test (IAT), an online screening developed at Harvard University that measures subconscious preferences by detecting automatic associations with certain prompts. They found that two-thirds of doctors, medical students, nurses, and board certified surgeons had a preference toward white people. (For now, Hadir's bias research has only focused on the race of the patient, not the medical professional.)
Haider says this reflects IAT results of the general population as well. The caretakers' unconscious biases, in almost all cases, contradicted their explicit biases—meaning, when surveyed, those caretakers all thought that they regarded black and white people the same way.
Importantly, Haider says that despite these findings, there isn’t good data showing that unconscious preferences impact how doctors treat patients. In simulated patients, on paper, they didn’t find any relationship between unconscious preferences and how people might assess or treat patients. “But that doesn’t necessarily mean that it doesn’t exist,” Haider says. He knows from his own mistakes that it sometime does, and further research needs to be done to see how biases might play out in real patient interactions.
“Doctors are human beings,” Cornwell says. “We bring with us our experiences, our biases, our principles.”
Rather than condemning all surgeons, Haider wants them to be aware that racial disparities, even in areas like trauma, do exist. A paper he co-authored in 2016 found that less than half of surgeons reported institutional efforts to address disparities, and less than a quarter had taken efforts to investigate disparities in their personal practice.
He thinks that teaching medical professionals how to interact with patients from other cultures could go a long way to resolving at least one part of the problem. He promotes an idea called cultural dexterity, over the more traditional concept of cultural competence. “How do you become competent in another person’s culture without living in that person’s culture?" he says. "We take care of a very diverse set of patients. If you learn just a little about every culture, you’re at the risk of stereotyping.”
Instead, he’s developed a program built around curiosity, respect, and empathy that he thinks will allow doctors to care for people who are different than them—culturally and racially—without letting their own preferences or biases get in the way. Haider and several collaborators just completed a pilot study of this curriculum at the Harvard Teaching Hospitals, called The Provider Awareness and Cultural Dexterity Toolkit for Surgeons (PACTS) project. The goal for now is to introduce the fundamental ideas, and gather input from patients and doctors on how cultural dexterity influences their practice. Future classes will teach students how to work with patients with limited English-language skills, obtain a truly informed consent for surgery, and manage pain in diverse patients.
“No one goes to medical or nursing school to treat patients differently," Haider says. "Doctors and nurses want to do better. I think this training is something that hopefully can counteract these biases that we all have, and help us really provide patient-centered care—no matter what a patient looks like."
Read This Next: Here's What People Do When When There's No 9-1-1