I had to be honest—I was uncomfortable with my new patient, a woman in her late thirties who was in my office for a general medical checkup. Maria Vincent was petite in stature but massive enough in width to meet the medical criteria for morbid obesity. Her pendulous belly hung like a third appendage between her legs and impeded her gait. A lovely-featured face was entirely swallowed up in layers of neck and jowl. Her arms could not hang straight down at her sides because of her girth.
She struggled onto the exam table, and the table shuddered under her 350 pounds. When I parted her gown at the back to listen to her lungs, waves of adipose tissue spilled over in tiered layers, muffling her breath sounds. When I palpated her abdomen, my hands were engulfed and I could not even attempt to feel her liver. Nor was I able to find her thyroid or the lymph nodes of her neck.
My job is to be nonjudgmental but the reflexive discomfort I was experiencing was impossible to deny, and I was upset at my unease. Many studies have shown that doctors display a pronounced bias against obesity—they have less respect for obese patients1 and they develop less rapport with them.2 The optimist in me would have hoped that medical professionals who pride themselves on caring for the ill and the vulnerable would do better than the rest of society, but alas we don’t. I guess I shouldn’t have been surprised, but I was dismayed. Why do doctors react that way? (Interestingly, patients have the same obesity bias toward doctors. Patients, no matter what their weight, express less trust in doctors who are obese.)3
No doubt one reason is that obesity—like alcoholism and drug use—is viewed as self-induced, even by doctors who are well aware of genetics and the other confounding factors involved. From the perspective of a group singularly steeped in the discipline and deprivation that got us through medical training, it’s hard to jettison the idea—despite mountains of scientific evidence to the contrary—that these medical conditions could be effortlessly alleviated by just a slightly brisker sense of self-regulation.
Perhaps in obese patients we see the feared reflections of ourselves, should we lose our carefully honed discipline. My own adolescent battles with weight and body image, however modest compared with my patient’s, left me with an aversion to junk food and overeating. Maybe Ms. Vincent represents my worst nightmare, what I would become if I ceased being vigilant and lost control altogether.
Maybe the pure physicality of obesity is the issue. In a society that worships svelte bodies to an unhealthy extreme, even someone like Ms. Vincent—well groomed, tidy, with an attractive face—can be perceived as slovenly just because of her weight.
These reactions are entirely irrational, of course; but emotions were never billed as rational, and doctors are as susceptible as anyone else. I don’t want to be the type of doctor who prejudges her patients, and I certainly don’t want to contribute to the very tangible stigma that obese people face. Yet I couldn’t help my unease as I struggled to examine Ms. Vincent in my office that day.
The more I actually talked with Ms. Vincent, however, the more manageable things became. She spoke of the stress of raising children while tending to her own medical problems, many of which she recognized stemmed from her obesity. She admitted she had trouble controlling her eating and that stress only caused her to eat more. Being overweight made her depressed and that depression made her crave sweets. She related a family history of obesity, emotional abuse, and neglect. She talked forthrightly about how humiliating it was to go to the gym, and how it was nearly impossible to even find gym clothes in her size. Getting a job was hopeless: “No one ever calls back a fat person for a second interview.” She couldn’t reach around to her back to scratch an itch. And if one of her kids made a beeline on the street, she knew she couldn’t keep up.
The more she spoke, the more my feelings moderated. Initially I’d just seen a very obese patient. By the end I saw a nuanced, exquisitely human person who suffered terribly from the obesity that overwhelmed her life. After we finished our visit, I pondered my initial response to her. Was it any different than racism, any less repellent? Even if her condition was self-induced, even partly, how could I countenance how I reacted?
Like most doctors, I like to think I treat all my patients equally, but I know that it’s not true. Medicine carries a long and shameful history of ill treatment, particularly toward racial and ethnic minorities, though it has exhibited equal-opportunity bigotry toward women, gays, transgender patients, immigrants, drug addicts, uneducated patients, obese patients, patients with HIV—pretty much anyone who falls outside the rigid lines of the medical establishment. Disparities in health care have been extensively documented.4 African Americans and other racial and ethnic minorities consistently receive less-aggressive cancer treatment, fewer cardiac catheterizations, fewer screening tests, less mental health treatment. The list is depressingly long.
Some of the disparities can be accounted for by—though not excused by—socioeconomic differences and unequal access to care. However, these two things contribute only a portion of the inequity; there are many other components in the mix. Given how essential communication is to the parlaying of medical care (in many cases it is the medical care), I wondered if the way doctors and patients speak to one another could explain some of the disparities in medicine.
Debra Roter and her colleagues undertook a number of studies to address this question. Using the RIAS system, they analyzed the medical visits of 61 doctors and 458 patients in the Washington, DC, metropolitan area.5 Roughly half the patients were white and half African American. In all the encounters—as expected from other studies—doctors dominated the conversations, taking up more speaking time than the patients. (The doctors, incidentally, were about half white, a third African American, and the rest Asian.)
However, doctors were much more verbally dominant with African American patients than with white patients; they spoke 24 percent more than their white patients, but 43 percent more than their African American patients. The encounters with African American patients were correspondingly less patient-centered than the encounters with white patients. The overall emotional effect of the visit was also lower (less positive tone of voice and general positive emotions). You could see why African American patients often walk away from their experiences in the medical system with a worse feeling than white patients do.
Did it matter if the doctors and patients were of the same racial or ethnic group? In a separate study, the same researchers analyzed 252 doctor-patient encounters with this question in mind.6 When doctors and patients were “race-concordant” (African American patients with African American doctors, white patients with white doctors), the visits were longer by about 10 percent. Patients rated the experience and their doctors’ communication skills higher compared with patients who saw doctors of a different race. The overall emotional tone and affect were more positive.
But there was no difference in the actual communication skills. Verbal dominance was no different (doctors hogged time equally with patients of the same race or a different race). There was no difference in measures of patient-centered communication: eliciting the patients’ thoughts, involving the patients in decision-making, and so on.
But patients perceived that things went better when they saw doctors more like them, and ultimately rated the experience higher. This may be from the comfort of familiarity. Perhaps patients—and doctors—are more relaxed in such a situation, leading to a more easygoing conversation. Or maybe it has to do with more deeply ingrained attitudes, some of which are entirely unconscious.
While there are still blatant racists out there, explicit racism is increasingly unacceptable in our society and you don’t see as much of it in daily life compared with years past. However, unconscious racial bias is still an obstinate challenge and probably underlies many of the inequities we see in health care and in society at large. This is an intriguing but uncomfortable area of inquiry, as it threatens the Hippocratic ideals that medicine holds up as its paragon of professionalism.
The test that researchers use most commonly to uncover unconscious bias is called the Implicit Association Test, or the IAT.7 The test is clever, and relatively simple—at least at first. In round one, the program asks you to rapidly classify photographs of faces as either European American or African American. The classification is done by pressing one button on the left side of a keyboard or a second button on the right. The photos are unambiguous and the task is easy. Round two has you rapidly classify words as “good” or “bad” using those same two buttons. The words are similarly unambiguous and easy to classify quickly (“joy,” “love,” “peace” versus “agony,” “terrible,” “horrible”).
Now things get tough. For the next round you get a random mix of words and photos to classify—good or bad words, African American or European American photos. But there are still only two buttons. So now each button has to represent two unrelated categories together. It’s tough because first your brain has to register whether you are seeing a word or a photo, and then has to decide which category it falls into (black/white or good/bad), then decide if it goes to the left or the right button. You are supposed to do these as swiftly as possible, without stopping to think.
The rub of the test is that it mixes up the associated categories. In one round, you have to press the left button for any faces that are European American or words that are bad, and the right button for faces that are African American or words that are good. In another round, it pairs them in the opposite way: the left button is for African American faces or words that are bad and the right button is for European American faces or words that are good.
The assumption is that you will do better when you have a stronger internal association between two otherwise unrelated categories. If you classify more accurately when one button associates European American with good than when the button associates African American with good, then you are felt to have a certain amount of “white preference,” and vice versa.
Versions of the test have been created to test implicit bias regarding women, Muslims, obese people, Native Americans, Asians, disabled people, and skin tone. It’s not a perfect measure of reality, of course, but it does offer some sense of our unconscious biases.
To elucidate if and how implicit bias might affect medical communication, researchers administered the IAT to 40 doctors and then audiotaped visits of these doctors and 269 of their patients.8 The higher the doctors scored on the measures of implicit racial bias, the worse their communication was with African American patients. There was more verbal dominance, less patient-centeredness, and less positive affect. These patients rated their experiences with distinctly low marks.
For white patients, a higher level of implicit bias on the part of the doctor (against African Americans) was also associated with more verbal dominance by the doctor. But the white patients did not experience the visit as poorly. They actually rated their experience more positively. Thus, implicit bias was associated with poorer communication skills on the part of the doctors to all patients. But the perception of poor communication differed between white and African American patients.
Another study from the Midwest polled nearly 3,000 patients of 124 doctors who’d taken the IAT.9 Doctors with higher levels of implicit bias again did not do well with their African American patients. These patients gave their doctors low marks when it came to patient-centered care. And white patients, as in the previous study, gave these doctors higher ratings, despite the implicit bias. Interestingly, Hispanics gave their doctors the lowest ratings of all, and, further, their low ratings were entirely unrelated to the doctors’ level of implicit bias.
When doctors as a group are evaluated for implicit bias, white physicians do worse than African American physicians. Male physicians show more bias than female physicians. African American physicians, it turns out, do not show any bias on these tests; that is, they do not demonstrate an unconscious preference for (or bias against) any race.10
I don’t think anyone is surprised that unconscious or implicit bias is prominent in medicine, even among doctors who are paragons of egalitarianism. And I don’t think it’s a surprise that African American patients pick up on this and come away with a more negative view of medical care.
This negative response can feed on itself and impact the experience of future medical visits.11 A group of 350 patients about to see their doctors were interviewed about past experience with race and class bias. The medical visits were then audiotaped and the doctor-patient communication was analyzed. For African American patients who’d experienced more race and class bias in the past, the visit demonstrated more negative affect from both doctor and patient—compared with African American patients who’d experienced less bias in the past, and compared with white patients. In post-visit surveys, these patients felt that the doctors were not listening to them and were not treating them with respect.
The researchers hypothesized that past bad experiences prime patients for another bad experience by setting low expectations and giving the patient a more negative overall affect. Doctors can unconsciously pick up on and then mirror that affect, leading to a downward spiral of poor communication and connection. One potential solution is to make doctors aware of this unconscious dynamic. Perhaps by noticing the downward spiral as it’s starting, doctors can make an extra effort to make their affect more positive and set the interaction on a more positive course.
Positive affect may sound like cheery window dressing but it is a critical component of communication. When expressed genuinely, it evinces a confidence in the other person. Thinking back to the “conarrator” research of Janet Bavelas, the positive affect and confidence from one person will enhance the story told by the other person. So it’s not a surprise that both white and African American patients are more likely to trust doctors who speak with positive affect.12 For African American patients, this holds even for patients who’d experienced prior bias (who are generally less likely to trust doctors). This is not to say that positive affect makes for a better doctor, but a trusting relationship is essential before good medical care can have the chance to flourish.
In an interesting study designed to see if implicit bias can be overcome, nurses were shown videos of patients who were visibly experiencing pain and asked to decide how much pain medicine to administer.13 When the nurses were instructed to use their best clinical judgment, they ended up giving less pain medicine to the African American patients than to the white patients. However, when the nurses were instructed to imagine how the pain was impacting the patients’ lives, they ended up giving the same amount of pain medication to all patients, regardless of race. Taking another person’s perspective is the building block of empathy, and this may be one of the prime factors needed to help eliminate disparities in medical care.
Clearly much systemic work needs to be done in the medical system to address inequities in care—everything from where hospitals are built and how patients get insurance to who gets into medical school and how research dollars are distributed. In terms of what individual doctors and patients can do, much of it depends on one-to-one communication. For doctors, recognizing their unconscious biases is a formidable but necessary step. Eliminating unconscious bias is not easy, but making a determined effort to see the other person as a unique individual and then taking the next step of envisioning that person’s perspective can make a tangible difference. For patients—who in an ideal world shouldn’t have to do anything to get their doctors to treat them equitably—allowing doctors to get to know them as a person offers fertile ground for empathy to develop.
Jose Santiago was a patient of mine some years ago. He’d done time at Rikers Island, though he never told me the reason for his incarceration. He possessed the craggy, almost grandfatherly look of someone who had survived—and was now officially retired from—the drug-addled 1980s. His skin was corrugated from years of injecting and his tattoos had faded to a murky bluish-gray. His raspy voice came across as exhausted but steady.
Like all new admits to Rikers, Mr. Santiago had been given a standard battery of tests for the medical ailments that run rampant behind bars. He received the unwelcome news that he was HIV positive, though luckily his T-cell count was still in the normal range. He also discovered that he had hepatitis C, plus diabetes and hypertension. He required massive doses of methadone to combat his heroin habit. What made his life most miserable, though, were the stubborn leg ulcers from his old skin-popping habit. Like many addicts of the day, when he ran out of veins, he’d simply injected drugs right into the meat of his limbs. These ulcers never seemed to heal.
When Mr. Santiago was released from prison, he made his way to our hospital’s HIV clinic. He was surprisingly reliable with his complicated medical regimen and he showed up for all of his appointments. However, he didn’t get the frequent lab tests to monitor the progress of all of his diseases because his veins had been obliterated by decades of drug use.
About once a year he’d allow his methadone doctor to draw blood from a ragged vein in his neck, and that was the extent of the medical evaluation that was possible. Despite the many medical assaults on his body, his immune system remained intact. His T-cell count stayed high enough to protect him from opportunistic infections (this was before viral load was easily measured, so we checked only T-cell levels).
In those days before super-strong antiviral medications turned HIV into a manageable chronic illness, it was a rare—and celebrated—circumstance to be one of the lucky “nonprogressors.” I’d had only one other patient in my practice who was a nonprogressor, and it was as though a fairy godmother had surveyed the vast, dismal landscape of AIDS and selected these two patients to sprinkle with fairy dust. Such patients were beacons of hope in an otherwise bleak chapter in medicine.
With such infrequent blood draws, it was hard to follow the vagaries of Mr. Santiago’s health, but he continued to have robust T-cell counts—year after year. One year the nurse practitioner who had been monitoring him in the HIV clinic made a special request. When it came time for the annual blood draw from that very last standing vein in his neck, the nurse asked the methadone doctor not just for a T-cell count but also for a new HIV test. She just had a hunch.
Lo and behold, it came back negative; Mr. Santiago did not have HIV. The initial HIV test at Rikers had been a false positive, and the diagnosis had been wrong all along. Mr. Santiago was promptly discharged from the HIV clinic and sent to the general medical clinic, where I became his doctor.
When we first met, I was astounded by the story, amazed that the misdiagnosis could have persisted for so long. I wondered why we chose to explain his vigorous T-cells by classifying him as a nonprogressor, rather than considering that his initial HIV test might have been incorrect. Was it that we put too much faith in the test’s objectivity? Or was it simply that Mr. Santiago fit the HIV picture so perfectly—Rikers prisoner, drug user, hepatitis C, tattoos—that we never imagined he could be negative?
When Mr. Santiago and I began working together, his calves were filleted open from knee to ankle. These ulcers were years in the making, products of ongoing drug use in settings that were the least conducive to wound healing imaginable. At our first visit, Mr. Santiago handed me a page-long list of supplies he needed for the ulcers: sterile saline, gauze wraps, surgical tape, iodine, Silvadene cream, latex gloves, occlusive dressings—and I admit I was a little taken aback. I’d never seen an addict, even an ex-addict, so organized and responsible.
He didn’t fit the picture.
Amazingly, over the ensuing years, thanks to Mr. Santiago’s assiduous care, the ulcers painstakingly healed. In the end there were only thin snaking scars on his calves to mark their sites.
He didn’t fit the picture.
And then one day, after nearly two decades of taking methadone at a dose high enough to knock our entire front-desk staff unconscious, he abruptly tapered himself off. Though methadone doesn’t offer the high that heroin does, the body typically remains equally dependent on it and most former heroin users stay on methadone indefinitely. For many ex-addicts, methadone remains the focus of their lives ad infinitum.
But Mr. Santiago stopped the medication himself, without any fanfare. “I’ve had enough,” he told me.
He didn’t fit the picture.
When I asked him how he felt about the misdiagnosis of HIV, he simply shrugged. “It is what it is,” he told me. Perhaps, given all the other things he’d suffered in his life, eight years of carrying a false diagnosis wasn’t the worst thing.
Over the years, Mr. Santiago shattered myth after myth. But in reality, he didn’t do anything; he simply was who he was. It was the medical profession that had to shatter its myths. If we’d observed the conflicting data more carefully in the beginning and if we’d examined our own biases before being so positive in our judgment, he would have been spared this brush with HIV—the stigma, the costly and unnecessary medical care, the medications and their toxic side effects.
Ultimately, Mr. Santiago bequeathed to me a profound lesson about my profession’s penchant for stereotypes and snap judgments. I just wish he hadn’t had to suffer all those years in the process. Once his leg ulcers had healed and he’d severed his methadone ties, he now had only the comparatively tame issues of diabetes and hypertension to contend with. (His hepatitis C had already been successfully treated by then.) Both the diabetes and the hypertension were mild and easily controlled. Without any need for specialty medical care, he eventually left Bellevue and continued with a local primary doctor near his apartment in Queens.
In one respect, Mr. Santiago was a medical success—he’d started out with a slew of grave medical issues and ended up with just two manageable ones. But in another respect, Mr. Santiago represented a failure of the medical system. In addition to the stereotypes into which we pigeonholed him—with potent consequences—there was a staggering lack of communication. I realized that I was probably just as guilty as the doctors before me who’d misdiagnosed him with HIV. Admittedly, Mr. Santiago wasn’t the easiest person to communicate with. He was a laconic man of few words and seemed content to let our visits focus on his leg ulcers. I followed his lead but I’m not sure if that was a respectful mirroring of tone or a lazy acquiescence to stereotype. I recognize that I didn’t make as much effort to get to know him as I usually do with my patients. Maybe my biases got the better of me; I let the florid tattoos, the prison time, and the drug history keep me at a distance. But there was a whole person inside there, someone with remarkable fortitude. I wish I’d taken the time to learn what enabled him to cultivate those traits despite the long odds. I’m sure there was much I could learn.
When I think about how doctors respond to patients who are labeled “undesirable” for whatever reason—for being homeless or drug-addicted or mentally ill or obese or malodorous—I try to make the analogy to physical discomforts that arise in the medical setting. Over the years, I’ve tugged off socks with lives of their own. I’ve changed dressings on putrid, oozing wounds. I’ve encountered maggots, roaches, semen, and diarrhea during physical examinations. I’ve felt close to vomiting many times—I’m as squeamish as the next person. I can’t control my physical reactions, as I can’t control some of my emotional ones. But I can endeavor to control what I do with them. With enough focus, I can tame my outer behavior.
But is that enough? Even if I hide how I feel when I am uncomfortable with a patient, my feelings still may influence how I communicate in ways that could result in poorer medical care. This is a genuine fear of mine. Will my unconscious—or conscious—biases send signals of disrespect no matter how hard I try to corral my visible reactions? Will I drive these patients away from medical care, even when these patients are often the ones who need care the most?
Modifying our external behavior and how we communicate is clearly important, but I believe we in the medical profession have a duty to work to change our inner landscapes as well. It’s a tall order, I realize, but if we wish to claim the high mantle of professionalism, we need to at least be actively attempting to challenge our gut feelings. The first step is to own up: doctors and nurses need to be honest about biased feelings, however distasteful and awkward this process may be. We need to catch ourselves in the act of jumping to a conclusion, to notice that we’re doing it, and then to question ourselves about the conclusions. We need to talk with our colleagues about biases in our practice to figure out where we might have blind spots. Perfection will never be achieved but that shouldn’t be an excuse to resign ourselves to the status quo. The very act of paying attention, of attempting to notice our shortcomings, is how any change begins.
Another approach, to borrow a technique from behavioral psychology, is to “act as if.” If a doctor can act as if an obese, or a smelly, or an irritating, or an alcoholic patient doesn’t bother her, in time the uncomfortable feelings may begin to lessen. It’s a bit like smiling when you feel sad—weird at first but then you grudgingly start to feel better. Pressing yourself to go against the grain is by nature an awkward action. But in time it can influence how you feel, in this case by allowing the doctor to get to know the patient better. Again, this may seem like window dressing, and if that’s the only thing a doctor attempts, it will indeed amount to just that. But if it’s part of a genuine effort to recalibrate how you feel and act in situations in which you might be biased, it will likely chip away at those biases. And even if the changes are only external at first, a doctor’s behavior serves as a model to the students, interns, and medical staff around him. The benefits of setting the behavioral tone, even if not (yet) fully heartfelt, cannot be underestimated.
Joan Noonan was a nurse in our clinic when I started as an attending physician at Bellevue, and even she would refer to herself as an old-school nurse. She proudly wore her nursing school pin on her white coat and still treasured her nursing cap, though she joked that if she wore it to work most of the younger staff would think she was sporting a coffee filter on her head. Ms. Noonan was a nurse extraordinaire, and what stood out to me was her impeccable reverence for each and every patient. She referred to every male patient as a gentleman. She could have the most disheveled, smelly, obstreperous alcoholic ranting in her exam room and she would never utter a disparaging word. “There’s a gentleman in my room who might need a little extra medical attention,” she would say calmly to one of the doctors. “Do you think you might be able to pop in for a minute?” Her tone of voice was always exquisitely respectful, and it would be identical whether the patient was one of the guys who hung out near the homeless shelter on First Avenue or the president of the United States, for whom Bellevue is the designated hospital should something untoward happen to our head of state while visiting New York City. Her attitude was genuine through and through, and the effect of her behavior on others was remarkable. You could be the most raving misanthrope on the staff and you’d find yourself inexplicably rising to her level of civility.
Respectful behavior is contagious, so even if your inner emotions haven’t quite caught up yet, the actions you exhibit will inform those around you, especially if the attempt is genuine and not just a masquerade. Your subconscious will eventually be prodded along.
I was relieved to come across a more expansive population study of nearly seventy thousand patients that showed that the medical care given to overweight patients is no different than what non-overweight patients receive.14 Despite an ingrained societal bias against obesity—one that affects physicians as well as patients—the medical profession seems to be able to deliver comparable treatment.
I’m not sure that suffices, though. Doctors may swallow their gut feelings, hold their noses, and offer adequate treatment to patients whom they deem undesirable. But that approach—even if it achieves acceptable medical outcomes—still conflicts with the foremost tenet of medical professionalism: that we treat all patients with compassion. Compassion can’t be faked. It presumes genuine sentiment, within which lies respect.
When we train medical students, we talk a lot about empathy. In its most literal sense, empathy is the attempt to appreciate the emotions of another, to feel the world from their perspective. As I talked more with Maria Vincent during that first visit, I began to get a sense of what her life was like. I couldn’t presume to actually know how she felt, but I could begin to imagine it and how I might cope with an extra two hundred pounds and the attendant stress.
When I saw her in my office recently, I felt a difference in myself. The physical exam was still somewhat uncomfortable for me (I’ll be candid here) but I pushed myself to keep Ms. Vincent—the person, not the body—in the forefront of my mind. My gut emotions still pulled, but they felt more manageable.
Maybe that’s what doctors should strive for: to prod our negative feelings out of the shadows, no matter how ill at ease it makes us feel. Disrespect has no place in the doctor-patient relationship. To provide good medical care, doctors must first ensure that every patient feels comfortable in their presence. If we doctors don’t feel comfortable ourselves, we need to be honest about it. Only then will our biases have the chance to dissipate.