Date: October 3, 2011
Fast-paced and protocol- driven, trauma care should be the great equalizer, right? Then why, asks trauma specialist Adil Haider, do patients of different races who come in with similar injuries often leave with different outcomes?
By Sharon Tregaskis
Illustration by Otto Steininger
It’s the middle of July and a heat wave has had a stranglehold on the East Coast for days. In Baltimore, the steamy days and sultry nights have frayed tempers, and violence is up.
At 3 a.m. on a Sunday, Johns Hopkins Hospital trauma surgeon Adil Haider answered his pager. “Three people got shot, there was mayhem, and I came in as the backup attending,” he says. In the intervening 72 hours, he has done seven operations, including back-to-back gunshot repairs for two African-American teens barely old enough to drive.
Haider’s current patient—a Hispanic man—arrived in the hospital by ambulance over the weekend with an arrhythmia and was admitted to the Coronary Care Unit. Despite preliminary treatment with blood thinners, he seems to have thrown a clot, knocking out the blood supply to his intestines; his abdomen is firm and distended. Residents are preparing him for surgery, inserting a central line and infusing his veins with fresh frozen plasma to increase his blood’s ability to resume clotting for the operation. Haider suspects that his patient’s intestines have died—the man will probably need a colostomy.
“He doesn’t speak English, doesn’t have insurance,” says Haider, who co-directs the School of Medicine’s Center for Surgery Trials and Outcomes Research (CSTOR). “But he should get the same quality of care anybody else gets.”
While hospitals that receive Medicare funds have been required since 1986 to provide emergency care to stabilize patients regardless of their insurance status or ability to pay, what happens next varies considerably. Some emergency departments transfer such patients as soon as possible, leaving their care to overtaxed county hospitals. Few such patients get access to the ongoing monitoring a person with health insurance would receive from a family physician.
At Hopkins, says Haider, administrators allow doctors to make sure that patients who are unable to pay for emergency trauma care also get appropriate follow-up—such as outpatient care—to reduce the likelihood of an easily preventable return to the emergency department.
“This is one of the rare places where we’re given the latitude, if you started care, to go to an administrator and get permission to book a case with the charge waived,” he says. “We care for a tremendous number of patients who don’t have insurance, but we started their care and the sentiment is, we’ll just take care of it.”
Epidemiologists and policymakers have long known that treatment for chronic conditions like heart disease can vary in relation to insurance status, race, and a variety of other patient demographics.
Minorities in the U.S. have reduced access to health care, take less advantage of screening, receive later diagnoses, and generally have poorer outcomes after diagnosis than their white counterparts. Black women are less likely to develop breast cancer than white women, but their disease is more likely to be diagnosed at an advanced stage and they are more likely to die from the disease within five years. Non-Hispanic blacks, Mexican Americans, and Native Americans are all more likely to develop diabetes than whites and far more likely to develop complications or die from the condition. Minorities are also more likely to go without eyeglasses or hearing aids as nursing home residents; less likely to receive standard treatments for asthma during childhood; and have worse outcomes and lower quality of life following a heart attack.
Haider has focused his investigation of disparities on the realm in which he has the greatest expertise—trauma and emergency medicine. Due to the brisk pace at which care must be delivered in the aftermath of a medical crisis, emergency departments rely more heavily than most specialties on protocol and routine to deliver rapid, consistent care.
That context makes Haider’s findings all the more shocking—and controversial. In studies he’s conducted involving data from hundreds of trauma centers across the country, he’s documented that black children with traumatic brain injuries are more likely to have compromised speech, mobility, and ability to feed themselves when they leave the hospital than their white counterparts. They’re also more likely to end up in a residential care facility instead of going home.
Haider’s research has also shown that black and Hispanic pedestrians are more likely to be hit by cars—and more likely to die from their injuries—than their white counterparts. And regardless of how you ended up in the emergency department, he’s found that being black, Hispanic, or uninsured significantly increases the chances that your stay in the hospital will end in the morgue; an insured white person with the same injuries would go home.
“It’s quite politically sensitive to raise these issues,” says Peter Pronovost, the director of Hopkins’ new Armstrong Institute for Patient Safety and Quality, and a collaborator with Haider on investigations into the statistical validity of certain measures for assessing emergency department quality of care. “Adil has been quite courageous in saying we have to study this and improve it.”
Economics undeniably influence who gets medical care in this country, as well as the quality of that care, and minorities tend to be less well off than whites. Yet Haider suspects that a patient’s cash-flow situation explains only a portion of the story.
“Some people say that it has nothing to do with race, that it’s all about socioeconomic status and if you gave everybody insurance, everybody would do the same, but I don’t think that’s true,” says Haider, 38, an assistant professor of surgery and anesthesiology at the School of Medicine, and health policy and management at the Bloomberg School of Public Health. “There are many factors involved and we need to figure out what they are so we can make a difference.”
As head of Hopkins’ Trauma Outcomes Research Group, Haider has assembled an international team of trauma surgeons, injury scientists, and social scientists to focus his search for those factors. Much of his foundational work over the last four years delves into huge databases like the National Trauma Data Bank (NTDB), a compilation of statistics on 1.8 million inpatients reported by 700 trauma centers and hospitals nationwide and maintained by the American College of Surgeons.
His first foray into the field was, perhaps, the most controversial. In 2007, the Journal of Trauma published his analysis of brain injury recovery data from the National Pediatric Trauma Registry. He conducted the study with Edward Cornwell, former director of Hopkins’ Adult Trauma Service and now chairman of the Department of Surgery at Howard University College of Medicine.
In addition to their finding that black and Hispanic children had lower functional outcomes at discharge and lower chances of returning straight home, they found that the black children started out with worse health and their injuries were more likely to have been penetrating (rather than blunt) trauma and intentionally inflicted, all factors associated with poor recovery.
“Their conclusions are clearly flawed and could have unintended consequences,” one critic—himself a pediatric emergency care provider and researcher—wrote to the journal’s editors, “and are an insult to those that take care of injured children.” Haider and his racially diverse team were measured in their response. “We fully understand the discomfort a study such as this creates,” they wrote in the journal’s next issue. “Perhaps no topic in American dialogue generates as much emotion as race. … We feel personally challenged by the results of this study—but no less convinced of its validity.”
More recently, Cornwell and Haider have plumbed the NTDB for insights into sex-based disparities in trauma outcomes, an inquiry Cornwell began when a severely injured patient—whose multiple gunshot wounds included extensive damage to his testosterone-producing testicles—recuperated more quickly and fully than his surgeons had anticipated. Since then, they’ve revealed that women and adolescent girls suffer lower mortality rates after traumatic shock than men or teen boys, and pregnant women seem to do better than those who aren’t.
While that work may someday lead to new treatment protocols including temporary testosterone blockers for adult men, most of Haider’s work is designed to yield the insights policymakers will need to make trauma care more accessible and equitable. “All of us, especially after being injured, would like to get to the best possible place and get the best possible care,” he says. “By getting rid of disparities, we can ensure that no matter who we are and no matter where we go, we can all get optimal care.”
That mindset sets the surgeon apart, says Pronovost, whose own work also grapples with policy transformation. “Adil is clearly driven by this passion to make the world a better place,” he says. “When you think like that, you always think about scaling your research to have a big impact. It’s great you did something at Hopkins or published a paper, but the end game is making the world a better place and that means you have to broadly impact practice or public policy. You have to improve patient outcomes.”
Haider, who describes himself as “not black or brown but in the middle,” was born in Zanesville, Ohio, and traces most of his early childhood memories to Huntington, Indiana, where his father, a Pakistani immigrant, worked as an engineer. “I grew up in a very nice, middle class, Midwestern town,” he says. “People waited for you to get off the elevator, that kind of stuff.”
In the early 1980s, after the Soviet invasion of Afghanistan, the family emigrated with a wave of American-educated Indians and Pakistanis intent on giving back to their homeland. Haider was 9 years old and spoke only English. “It was very, very different,” he says, noting that the language barrier, subtropical climate, and episodic electrical outages only exacerbated the adjustment. “There was a huge disparity between people who couldn’t afford even the bare necessities, didn’t have adequate health care, and the small class of privileged people who could afford anything they wanted,” he recalls. More important, Pakistani attitudes toward those differences were at odds with the culture from which Haider had come. “In America,” he says, “the mentality was that everyone was equal. Over there, it wasn’t that way.”
Haider earned his MD in Pakistan, then returned to the U.S. for his residency and a master’s in public health at Hopkins. Like his father, he was drawn to service. “Having had this very nice childhood and having seen how if you give someone an opportunity that they can really make a difference, I always felt like I should do something and give back,” he says.
Initially he traveled to Africa but returned to the U.S. because, as he told an audience at Hopkins last winter, “you don’t have to go to Africa to care for the poor and disenfranchised—you can do that right here in Baltimore.”
The scientist has also made a point of publishing the mathematical and statistical techniques his team uses as they mine national databases for insights. In the case of the NTDB, for example, policymakers use complication rates to assess quality of care. Yet, such data is voluntarily supplied by hospitals, and not all hospitals complete all fields. Only a fraction of hospitals detail complication rates for each trauma, and they don’t necessarily report every complication for every case. As a result, researchers have to choose whether to include hospitals that don’t report complications when assessing quality of care.
In a 2008 study in Trauma—subtitled “Numerators in Search of Denominators,” Haider and Pronovost worked with trauma surgeon Elliot Haut, an assistant professor of surgery, anesthesiology, and critical care medicine, to describe how complication rates leapt upward by 25 to 70 percent when only hospitals that supplied details on complications were included.
“This is sensitive stuff and we can’t just run after headlines and rhetoric,” says Haider. “That’s why we’ve not only published papers on our findings but published methodology papers on the best possible way to do this kind of number crunching and outcomes research.” To cope with the inevitable gaps in patient records recorded in large data sets, his team uses a technique known as imputation, in which known variables are randomly assigned to fill in for missing data. “If on imputation our results change or the missing data really does impact our results,” he says, “we will not publish them.”
More recently, Haider has begun to design studies that delve into the mechanisms underlying the trends revealed by his large database analyses.
In a study currently under way, he and Lisa Cooper, professor of medicine here, surveyed 500 trauma care providers at Hopkins. Framed as a study of safety, the program presented hypothetical patients, then tested whether providers diagnosed, for example, vague abdominal symptoms as appendicitis or pelvic inflammatory disease, or the extent to which they accepted or discounted a family’s reports of a patient’s alcohol consumption.
The study aims to explore whether unconscious race and social class preferences have any relation to treatment response. Instead of focusing solely on surgeons, the researchers split their sample to include 250 trauma nurses. “Not too many people have studied the role of nurses,” Haider notes. “I think they have a huge role to play and they are integral in creating a solution.”
Haider is quick to clarify that the study—slated for publication this winter—isn’t intended to harpoon trauma care providers as bigots. “I don’t think people are racist,” he says. “Doctors, by and large, get into medicine because they want to do something noble and help people.” More likely, he says the mechanism by which disparities emerge has to do with our capacity for empathy and the imperative to make snap judgments, honed during medical training.
“Do we treat people differently because we’re able to identify with one person better than another?” he muses. “If someone looks like your grandmother, are you more empathic? If you can’t relate to someone, do you give them less pain medicine?”
It may sound naive, says Haider, but he’s confident such disparities can be eliminated from the practice of medicine in the United States. His latest study—conducted with Pronovost, Haut, Cooper, and Cornwell—lends credence to that goal. The scientists surveyed 200 first-year medical students at Hopkins, who were presented with eight patient vignettes—and concluded that the students’ unconscious race and social class preferences seem unrelated to their treatment responses. The study, published in September’s Journal of the American Medical Association, doesn’t explain the discepancy between the current generation of medical students and their mentors. Yet, says Haider, such findings bode well for the future of the medical profession.
“When you’ve lived in other countries, you get to see what you can do in America,” he says. “If anywhere we can achieve equality, it’s here.”