Bedside Manners: Can empathy be taught in medicine?

Snow fell all day while a baby was dying. I cannot remember if it was a boy or a girl, so imagine a girl. This was years ago now. The baby had pneumonia. A few weeks into a tenuous life, her lungs were so clogged that they could hardly move oxygen into the blood or carbon dioxide out. The baby’s little heart, trying desperately to pump blood through those lungs, began to fail.

The neonatology unit was on an upper floor, with tall windows and glass doors. You could see clear across the unit and then out through the high windows to the world, where a storm was coming in from over the ocean. In the baby’s room, machines accumulated, along with all the attendants who cared for the machines. There were chairs so the baby’s parents could rest a moment, and as things progressed, a couch was brought in, then more chairs, where the doctors could sit to share serious news.

I describe this scene as if from a distance because, although I was the resident doctor assigned to the baby on that day, I never really saw her. I was across the way, on a computer. In this critical case, the compelling work of medicine—diagnosis, planning, communication, care—was reserved for more experienced pediatricians. My job was to enter orders: that is, to type things into the computer so that they could happen. Various people, using my pager or my phone or occasionally face-to-face communication, would order me to order something, and then I would order it: more labs, more blood, more platelets, more medicine. Such-and-such tiny changes to the machines. I was being exquisitely micromanaged to perform a technical task. If I fucked it up, the baby could die.

This was only the beginning of my twenty-four-hour shift, during my second year working eighty-hour weeks. Perhaps this explains why, although I was this baby’s doctor for part of her brief and irreplaceable life, I cannot remember her name or anything about her family. When I think back on my residency, moments like this are explicable but still do not feel forgivable.

God knows we tried our best. But at some point the baby died, and I had many fewer orders to enter. Late that night, with my other critically ill patients tucked in and breathing on ventilators, I walked to the on-call room to lie down. By the time I made it there, the snow had turned to rain.

Rain was still lashing the windows when my phone rang sometime later—a few minutes or maybe an hour, still well before dawn. A hospital across the state had managed to resuscitate a baby born at twenty-four weeks, and they were sending the baby to us. But the storm had grounded helicopters, so the baby was coming overland. It would take a couple of hours for the ambulance to speed through the rainy darkness to a place where we could save the baby’s life, which meant that maybe I had time to sleep. I started to doze off. Then the team in the ambulance called again to say that the baby’s heart had stopped and they were doing CPR. Then they got a heartbeat back, and I fell asleep again, with the phone by my head. Then they called to say that the transport ventilator had failed, so they were ventilating manually with a bag-valve mask. I think of them now, the headlights of the ambulance beaming through the storm, a team of experts using their hands to pump tiny breaths into that baby’s lungs.

But I was so tired, and bagging a twenty-four-weeker in an ambulance, in the rain, an hour from the NICU, is a bad situation.

Maybe that baby will die, I thought. Then I can get some sleep.

Hearing this story, a colleague once commented that I had done no harm: it’s a fallacy, or perhaps a symptom of psychosis, to imagine that one’s thoughts can harm or heal. But in modern medical education, much attention is given to a doctor’s inner life, as well as her ability to understand and communicate about patients’ emotions. Studies suggest that a physician’s capacity for empathy has measurable benefits for patients: those who rate their physicians as more empathic experience less anxiety and are more likely to stick with medical treatment. Patients are also more satisfied with empathic doctors. Physicians’ empathy may even improve patients’ physical health, with lower cholesterol levels and better control of diabetes measured among patients who describe their doctors as empathic. One recent study in JAMA Network Open showed that patients with chronic pain who perceive their doctors to be highly empathic experience less pain over time.

Empathy, however, was not always understood to be part of a doctor’s essential duties. William Osler, who practiced and taught medicine across the English-speaking world around the turn of the twentieth century, positioned aequanimitas—equanimity—as a primary virtue for physicians. As the physician and philosopher Jodi Halpern has pointed out, Osler believed that a doctor should aspire to be an island of calm in stormy seas, mastering her own unruly emotions with the force of rationality. Decades after Osler’s death in 1919, scholars described “detached concern” as the ideal relational stance of physicians toward patients: a doctor should experience and act with concern for her patients but also cultivate a keen awareness that her patients’ suffering was not her own. Such detachment was meant to allow for accurate diagnoses and clearheaded decision-making.

This century has seen a turn to empathy. In 2001, Halpern published From Detached Concern to Empathy: Humanizing Medical Practice, a seminal book arguing that physicians who engage empathically with their patients might be more effective. Halpern’s work has helped galvanize a generation of researchers who explore the neurobiology of empathy and its clinical effects. Empathy still has its skeptics. In my conversations with colleagues and students, many bristle at the implication that they need remedial education in how to feel. The bioethicists Eugenia Stefanello and Keisha Ray have argued that uncritical appeals to empathy may paper over the way physicians fail to accurately imagine the inner states of those they perceive as different from themselves. Some doctors worry that empathy will cloud their judgment. Some reject the project of empathy entirely, believing that emotional engagement with patients makes the lifelong work of attending to suffering unsustainable. Still others believe that empathy is a worthy ideal but an improbable one so long as working conditions in health care remain so bad. But on the whole, empathy has overtaken equanimity as the model for physician-patient relations.

Along with empathy’s rise has come the concern that this quality is in short supply; studies have found that physicians’ empathy varies widely and declines during medical training, and that physicians often fail to act empathically even when patients clearly demonstrate an emotional need. If empathy is useful to patients but professionally scarce, the question arises: How might we cultivate more empathic physicians?

In April 2024, a group of nurses, doctors, and scholars assembled under the aegis of the Global Empathy in Healthcare Network to venture an answer in the form of the Leicester Empathy Declaration, published in the journal Patient Education and Counseling. It is “ethically imperative, and economically advantageous for all healthcare professionals to undertake empathy training,” the authors wrote.

The declaration reads as if the question of whether empathy is teachable—and if so, how—is largely settled: all sites of medical training should organize lessons in “efficient, evidence-based methods for practitioners to deliver consistently high levels of empathy,” it claims, without explaining what these methods are. But can empathy really be systematically produced? Or will attempts to do so ultimately cheapen the very thing they seek to create?

Medical empathy training is predicated on a definition of “empathy” that diverges significantly from the lay understanding of the term as a vicarious emotional experience, one’s heart “going out” to another. Researchers break down empathic care into various observable, measurable components. Affective empathy, feeling with or for a patient emotionally, is only one of them. Others include cognitive empathy (accurately understanding a patient’s situation); communication skills (asking about patients’ lives and telling them our interpretations); and altruism (mobilizing these understandings for the patient’s benefit). If you come to a medical appointment in pain, and afraid of what the pain means, an empathic doctor will listen long enough to pick up on both your verbal and physical expressions, let you know what they understand, match their tone to your feelings, and make a plan that addresses both your pain and your fear.

To produce more such doctors, the group behind the Leicester Declaration advocates for a vast expansion of the existing empathy industry—a global network of research and training centers. At least two such centers already exist in the United States: the T. Denny Sanford Institute for Empathy and Compassion, at the University of California, San Diego, and NYU Langone’s Center for Empathy in Medicine. Scholars from both are signatories to the Leicester Declaration. The Sanford Institute studies the neuroscience of clinical empathy to “establish the neurological basis for empathy in the brain,” while the Center for Empathy in Medicine is largely focused on developing and studying methods for teaching empathy to medical students, residents, researchers, and faculty. The team there, which is led by internist Jennifer Adams, offers training in what she calls “core empathy skills.”

Empathy skills also play a role in my own teaching at the University of Texas Health Science Center at San Antonio. On a Friday afternoon in June 2024, I met with a few dozen residents in an airy room on campus. They’d stepped away from bedsides and computers, handing over their phones to senior doctors. Here, we would not be interrupted.

I took a breath and began. The topic at hand was the act of breaking bad news, a special workshop for our trainees in pediatrics. I chose my words carefully. “I am here to teach you about empathic communication,” I said. “The methods I’m teaching are evidence-based. The parents you have these conversations with will remember them for the rest of their lives.”

I walked the junior doctors through a sequence of behaviors and a set of phrases to use when breaking difficult news. Relying on a script to guide critical conversations can feel uncanny, but residents seem to understand why it’s useful. The first time that a junior doctor has to tell a parent that her child is at risk of dying or has died, the doctor’s own emotions are unlikely to match the parent’s needs. The doctor may be exhausted from lack of sleep, or distraught after having performed CPR on a baby for forty-five minutes. She may be unable to access any feeling at all. While an intern in the pediatric emergency room, I met a child with the classic signs of a cancer called neuroblastoma: agitation, bruising around her eyes, a mass in her belly. My attending doctor sat behind me as I shared my concern with the girl’s father, and even with her support, I was so nervous that I nearly vomited.

“I don’t care what you are feeling in your heart when you have these conversations,” I told my residents. “You could be feeling numb, or overwhelmed, or angry. And sometimes you will. I am teaching you empathic communication skills so that, no matter what you are feeling in your heart, you can do this well.”

In teaching my residents empathic communication, I am careful not to claim that I’m teaching empathy itself—or to imply that there is a correct way for them to feel. For pediatricians, the idea of vicariously experiencing what patients and their families go through is unnerving. The infants and children we treat are often unable to breathe on their own; they may have aggressive cancers or complex fungal infections; they may have starved for months while migrating to the United States; they may have snatched a loaded gun off a parent’s bedside table and shot themselves in the head. I do not recommend that my trainees feel all of this—certainly not a dozen times a day, every day, for years.

Even so, I am struck by how quickly we can develop a resistance to joining in the emotional lives of patients. I teach a literature course to second-year medical students who, though they have not yet begun intensive training in the hospital, often find stories of callousness and detachment more relatable and “real” than stories of connection.

In her memoir On Call: A Doctor’s Days and Nights in Residency, Emily R. Transue recounts two experiences that have helped my second-year students explore this phenomenon. In one, Transue is told that a patient in her care has died. She frowns, then later finds herself laughing hysterically as she realizes that neither she nor the student on her team can remember who the patient was. My students can relate to this; they can imagine that they will soon care for people to whom they feel no human connection.

In another chapter, Transue describes spending time with a dying patient. She writes:

Sometimes, sitting at his bedside, I wonder who is really taking care of whom. I want to comfort him, to make all this easier, to keep him from being alone. Yet I sit beside him listening less out of courtesy or kindness than for my own pleasure. . . . I am hungry for the knowledge he seems to have, of how to live a life you can be proud of at its end.

In teaching these narratives together, I mean to suggest to my students that both things are possible in medicine: profound connection and profound alienation. After all, I was once so tired during residency that I became indifferent to a baby’s survival. I also cared for a teenager so sweet and funny that she made me laugh every day until she died of a horrible complication from the treatment we’d given for her cancer. Medicine contains all of this.

But some of my students just don’t believe Transue’s more uplifting story. It has the ring of a med-school application essay for them: Herein, the writer will demonstrate empathy, compassion, and moral engagement with suffering.

“And anyway,” someone will always say, “you can’t get too emotionally involved with patients.”

“Why not?” I ask.

“You’ll burn out,” the student will answer. Even before they begin work in the hospital, my students see emotional engagement with patients as a risk to their own well-being, given the demands of the job.

Eighty-hour workweeks are not uncommon for junior doctors. Once they begin practicing, a physician’s worth is measured in relative value units (RVUs), metrics that convert every medical encounter—from a colonoscopy to a child’s routine checkup to a family conference in which a dying grandmother’s wishes are debated—into a specific amount of money. To extract more RVUs from salaried physicians, the hospital corporations, academic medical centers, and private-equity-run health systems that employ us require more and more patient encounters on a given day. This means that each individual encounter is shorter and brings more tasks: not only seeing the patient (the part we generally enjoy) but also documentation, billing, arguing with insurance companies, and so on.

Empathy takes time, the very resource that doctors most lack. But NYU’s Jennifer Adams thinks that it might not have to be so time-consuming. When we spoke, she told me about a patient who arrived two hours late to an appointment. When Adams went out into the waiting room to apologize and tell her that she couldn’t see her, the patient said she didn’t need to be seen anyway. She just needed a doctor’s note—something Adams could write in three minutes. When she handed over the note, the patient gave her a hug in gratitude.

Like Adams, I believe that helping people and connecting with them makes a life in medicine not only bearable but rich. But do physicians have three minutes to spare, over and over again? In my own life, I do not find that empathy leeches away my time. Rather, when other tasks have taken too much of me, empathy becomes impossible. These moments of barrenness are what physicians call a “sign.” Like the yellowing of the eyes that signals problems with the liver, a lack of empathy for patients is a sign that one’s practice has been whittled to the quick.

When I was a third-year medical student, I won my school’s Humanism in Surgery Award. It was an engraved cup that, for years, I used to hold my toothbrush. I won it, I joked, because I was able to identify and name an emotion. The surgeons were stunned! But the joke, it turns out, was on me: because researchers are guided toward that which can be measured, results such as the ability to identify and name an emotion are prized in empathy education.

There are two surveys commonly used to measure medical empathy. In the Jefferson Scale of Empathy, physicians and medical students evaluate such statements as “It is difficult for me to view things from my patients’ perspectives” on a scale of 1 to 7. The Consultation and Relational Empathy (CARE) measure invites the supposed beneficiaries of physician empathy to be the judges: patients rate their doctors on a scale of “poor” to “excellent” on questions such as “How was your doctor at being interested in you as a whole person?”

Scores of studies have attempted to quantify how much empathy is produced by a workshop, course, or other intervention for health trainees. In the name of empathy, today’s health trainee might read Kafka’s Metamorphosis or listen to a fifteen-minute recording of simulated hallucinations commanding her to “jump, jump now! . . . Go on, do it.” She might be encouraged to consider the human cadaver that she dissects in anatomy lab as her “first patient,” and then to share a poem or a song to thank them for their gift. She might don the Age Gain Now Empathy System (AGNES), a suit designed by researchers at the MIT AgeLab to simulate the experience of living in an elderly body, or else the Cost Effective Simulator Suit (CESS), designed by a team at Touro University. Produced for less than $300, the CESS uses braces, gloves, a collar, goggles, a weighted vest, and shoes with thick foam inserts to simulate the elderly’s experiences of blurry vision, stiff joints, poor balance, and fine-motor impairment.

Studies of empathy interventions may not yield meaningful findings. Many define empathy differently or fail to define it at all; others fail to use control groups, do not report effect sizes, or contain other errors and omissions that make the studies difficult to compare. Attempts to measure empathy are further complicated by the fact that doctors’ assessments of their own empathy levels often do not correlate with patients’ experiences of their care. In all these studies, there is little evidence that any simple intervention helps physicians maintain empathy over time. This research does not, by my lights, cohere into a body of work that convincingly shows that empathy can be taught in medical school.

It might be that inculcating lasting virtues in professionals requires years of effort, rather than the short-term methods often presented in the literature. Or the problem could be more fundamental: even if empathy is desperately needed, there is no easy or efficient way to teach it. An article in the same journal that published the Leicester Declaration argues that the concept of empathy in medicine is too vague to be scientifically useful. “The empathy concept as used in the published literature,” the paper concludes, “is overused and unclear, and potentially damaging to scholarship, medical education, and ultimately healthcare.”

But there is another approach. Even if empathy cannot be measured—or transmitted via a simulation suit—it matters nonetheless. I trained for my M.D. and Ph.D. at the University of Texas Medical Branch, where philosophers, historians, artists, and scholars of literature built an early home for the medical humanities, a field of scholarship that explores the unquantifiable in medicine and health. This movement began in the Sixties, in part as a response to the widespread dehumanization of patients in hospitals and medical research—their reduction to mere bodies, the disregarding of their stories and their higher-order needs. From the beginning, the field had a reformist bent: in addition to bringing the scholarly rigor of the humanities to questions of meaning in medicine and health, it sought to reform medical education such that physicians might be equipped to help patients navigate the existential crises that illness, pregnancy, birth, and death so often beget. While some scholars in the medical humanities might deploy communication training like my workshop on breaking bad news, or even endorse the Leicester group’s aim of producing more empathic physicians, many advocate for a less goal-directed form of moral inquiry. This requires an openness to different interpretations of what it means to live a moral life in medicine, as well as a commitment to ways of knowing that don’t rely on quantification.

In a 2001 article, the foundational medical-humanities scholar Kathryn Montgomery referred to physicians as “flat-footed positivists.” Her judgment still rings true: medical school in particular is a land of multiple-choice questions, where everything a student must know (and, implicitly, the value of the student herself) is thought to be quantifiable. But when caring for people, doctors confront problems that are complex and subjective. Pain, for example, cannot be seen. Though we try to measure it on a scale of 0 to 10, it remains in the unbridgeable realm of personal experience. Suffering is like that, too, and so is care—we understand these things more deeply through narrative than through measurement. The medical humanities encourages the asking of unanswerable questions; it attracts and cultivates physicians who invite incalculable truths into the hospital room.

As a medical humanist, I reject the shallow and brittle terms on which empathy is so often taught in medicine. I am not alone: “Since there is general consensus that physicians ‘need’ empathy, the pedagogical result is often based on ways of ‘taming’ empathy,” the psychologist and medical educator Johanna Shapiro warned in 2011. “This has meant by and large separating empathy from the unruliness and unpredictability of emotion.” I worry not only that this project overlooks feeling; I fear that our efforts to slot empathy into the framework of medical research will leave us with a dry, inhuman, technocratic vision of what empathy can be. Consider this: On that rainy night during my residency, the baby survived his transport to our hospital. The father stood in the NICU room, off to one side, while we worked to stabilize his child. Someone offered the father a chair, but he refused. “You want to be closer to him,” I said. He nodded, and I asked the team to make space. If we consider empathy to be a cognitive and communicative task, this was it: I noticed the father’s emotion, described it, and acted on it. But I am a mother now, and I can imagine my own son in a moment of desperate need. I can imagine him breathless in an ambulance, hours away from help. I can imagine that the doctor awaiting him, mired in solipsism or exhaustion, is so indifferent that she wishes my son would die. I am not prepared to celebrate that doctor for her empathy if she offers me a chair.

Jodi Halpern’s work describes a vision of empathy that is affectively rich but also workable for practitioners. (Halpern, who is a signatory to the Leicester Declaration, does not claim to know a rapid, simplistic model for how empathy can be taught.) In distinguishing between “ordinary” and “clinical” empathy, she writes that empathic doctors acquire “emotional attunement”: “a subtle nonverbal sense of where a patient is emotionally.” An attuned physician experiences a resonance that directs her attention to the parts of the patient’s story that are important to them; this in turn facilitates decisions that serve the patient’s needs.

Halpern is also sensitive to the fact that some medical trainees endure a level of vicarious emotional experience that is overwhelming to the students and unhelpful to patients. Ideally, physicians would share in the distress of a patient enough to be motivated to altruism, but not to the extent that they can’t be helpful. To that end, Halpern advocates for what she calls “engaged curiosity,” a habit of asking questions so that diagnoses and treatment plans can be based on a deeper knowledge of patients’ lives and stories. Cultivating this habit early on might lead to better experiences for patients and richer lives for medical practitioners—lives where we understand our work to be meaningful to the specific human beings whom we set out to help.

Tools for empathic communication—the scripts and rehearsed responses to intense emotion—ought to be taught in medicine. But they should not be called empathy, and doctors should not accept skills training as a substitute for moral inquiry of the sort that the medical humanities offers. Such inquiry takes time and stubbornly resists quantification; its effects are complex and diffuse, embedded in the long arc of a doctor’s moral life. When I read stories with my students, I do it in the shadow of so many trials that await them: punishing workweeks, children in horrific pain, young mothers whose bodies have determined to die, phones and pagers that will interrupt every conversation, and, through it all, a health-care system that will reduce their work to aliquots of profit. The medical system that my students will work in abhors unmeasurable truths; it punishes contemplation and rewards efficiency. Attempts to teach empathy via a similarly quantifiable model belie the dehumanizing nature of that system. So, too, do these attempts reveal how desperately empathy is needed in medicine—an empathy that is not transactional but mutual.

Sooner than my students can imagine, ambulances will be speeding over rain-swept streets toward them. When exhaustion, sadness, or shock overtakes them, I want my students to be able to respond anyway, with empathic communication and other technical skills. But I also want them to experience moments when their inner lives align perfectly with the work at hand, when they see the face of suffering and find themselves equipped to respond to it with exquisite skill. The empathy that can sustain a life in medicine is not merely delivered by a doctor to a patient but instead occurs between them. It is grounded in how we all live, together and separately, in faulty bodies, and in how tenderly and sometimes desperately we offer our stories to one another. It is grounded also in the hardest truth medicine teaches: all people suffer. Special suffering is meted out for some, but we all do suffer.

At this moment, the basic requirements for human flourishing are denied to so many. American health-care systems already frayed by a brutal pandemic are being dismantled further, and Americans’ antipathy for one another is having deadly effects on children and families. At such a time, it may be tempting to beg for scraps: for the means to see a physician at all; for a physician who has to see thirty-five patients today instead of forty, who has been trained to recognize and name an emotion. But the very erosion of so much that physicians hold dear may make true empathy in medicine more urgent, more attractive and necessary, just as it makes the mutuality of suffering more plain.