Empathy: Are Future Physicians Learning the Crucial Piece of Clinical Practice?

A first-year medical student walks into her first patient interview carrying more compassion than she may ever have again in her career. What sounds like a cynical joke is actually very close to reality, based on a growing body of research. Decades of studies using validated instruments like the Jefferson Scale of Empathy (JSE) have tracked a familiar arc: high empathy at matriculation, followed by a measurable dip once students hit the wards.¹^,² Nearly everyone agrees empathy in healthcare matters. The harder question for medical educators, hospital administrators and practicing physicians alike is whether medical training is structured in a way that keeps empathy intact.

What the Data Show

The empathy-decline narrative traces back to Mohammadreza Hojat and colleagues at Thomas Jefferson University, who developed the JSE because no tool existed to measure empathy in a clinical, rather than general, context. Their prospective study of 125 medical students found a statistically measurable decline in empathy scores between the beginning and end of the third year, the point at which students move from classroom learning into direct patient care.¹ A 2011 systematic review of studies involving medical students and residents concluded this erosion undermines the pursuit of professionalism and may threaten the quality of healthcare. It also flagged the third and fourth years, when clinical exposure intensifies, as the period of steepest decline.

More recent data complicate that picture. A 2024 cross-sectional study of nearly 900 students across all five years at Leicester Medical School in the United Kingdom found a small but statistically significant drop in empathy scores as students moved from preclinical to clinical training, with male students scoring consistently lower than female students.³

A University of Chicago research team pushed the question further by separating empathy into its cognitive and affective components. Self-reported cognitive empathy scores did decline in line with prior studies, but the picture grew more nuanced once affective empathy, the capacity to actually feel what a patient feels, was measured using different methods.⁴ Medical students, in other words, may not be losing their capacity to care so much as learning, sometimes for self-protective reasons, to regulate how much of that caring they let show.

Taken together, the research supports a more precise conclusion than the blanket claim that medical school kills empathy. Empathy does appear to soften somewhat as students enter clinical training. The drop is generally small in magnitude, and the timing, right when students begin sitting across from real, suffering patients, is the detail that should worry educators most, regardless of how one interprets the exact numbers.

Empathy and Patient Outcomes

Ask researchers why empathy dips during training and the same list of culprits recurs: chronic sleep deprivation, unrelenting time pressure, a curriculum still organized around biomedical facts rather than human narratives, weak or absent role modeling and the informal, often cynical lessons students absorb by watching how attending physicians actually behave, as opposed to what they are told in lectures.³ Add the emotional toll of encountering death, suffering and system dysfunction for the first time, often without adequate support, and a decline in outward empathic expression starts to look less like moral failure and more like an adaptive, if costly, coping response.

Physician and author Abraham Verghese, MD, MACP, who built the Stanford 25 program to keep bedside physical examination alive in an age of scans and algorithms, has argued the problem is not that medical schools fail to teach empathy, but that training actively erodes what students already bring with them. “What we need in medical schools is not to teach empathy, as much as to preserve it,” Dr. Verghese said. “The process of learning huge volumes of information about disease, of learning a specialized language, can ironically make one lose sight of the patient one came to serve; empathy can be replaced by cynicism.”⁵ Dr. Verghese also described how modern training takes lovely people and converts them into bottom-line, somewhat cynical, disease-oriented people, in part because students are taught a clinical vocabulary that strips the humanity out of a patient’s story.

The implications of empathy-free patient interactions extend beyond good bedside manner. A widely cited 2011 study of 891 diabetic patients treated by 29 family physicians found that patients of high-empathy physicians, as measured by the JSE, were significantly more likely to have good control of hemoglobin A1c (56 percent versus 40 percent) and LDL cholesterol (59 percent versus 44 percent) than patients of low-empathy physicians.⁶ In addition, a review of empathy’s effectiveness in general practice similarly concluded that empathy correlates with better patient satisfaction and adherence, lower patient anxiety and distress, and improved diagnostic and clinical outcomes.⁷

Physician empathy has also been linked to fewer malpractice claims, greater patient safety and lower rates of physician burnout, a two-way relationship in which burned-out clinicians express less empathy and low-empathy environments contribute to burnout in turn.⁸ Helen Riess, MD, a psychiatrist at Massachusetts General Hospital and Harvard Medical School who has spent much of her career studying the neuroscience of empathy, put it bluntly in a recent interview: Physicians who train in relational skills also become “more resilient” over time, because connecting authentically with patients feeds back into a clinician’s own sense of purpose rather than draining it.⁹

Can Empathy Be Taught, or Just Protected?

The most encouraging development in this field is found in a growing body of evidence that empathy is trainable. Dr. Riess developed a neuroscience-grounded curriculum built around the acronym E.M.P.A.T.H.Y. (eye contact, muscles of facial expression, posture, affect, tone of voice, hearing the whole person and your response), and tested it in a randomized controlled trial of resident physicians. Patients rated physicians in the trained group as significantly more empathic than those in the control group, and the trained residents also scored better on tests of reading subtle facial emotion.¹⁰

“Most people believed that you had a fixed amount of empathy, and you either had it or you didn’t,” Dr. Riess said of the assumption her research set out to challenge.¹¹ Her broader argument is that empathy is closer to a skill than a personality trait, one that can atrophy under stress and can also be deliberately rebuilt.

A second, increasingly common intervention is narrative medicine, a field founded by Rita Charon, MD, PhD, at Columbia University, which trains clinicians in close reading, reflective writing and attentive listening to patients’ stories. “The effective practice of medicine requires narrative competence,” Dr. Charon said. “That is, the ability to acknowledge, absorb, interpret and act on the stories and plights of others.”¹² Residency programs have begun building narrative medicine sessions directly into required didactic time, on the theory that giving trainees structured space to process difficult clinical experiences, rather than simply pushing through them, protects both empathy and professional identity during a uniquely formative and stressful period.¹³

Programs combining these approaches, communication skills training, reflective writing, humanities coursework and protected time for processing difficult encounters, are proliferating across U.S. medical schools and residencies, though implementation remains uneven and is often squeezed out by the demands of an already saturated curriculum.

Embedding Empathy into the Curriculum

The response to the empathy problem is no longer confined to a single elective or a communication skills seminar bolted onto the first year. At Leicester Medical School, the Stoneygate Centre for Empathic Healthcare, launched in 2023, has taken the unusual step of building its curriculum with patients themselves at the table. Working alongside educators and students in a series of co-production workshops, the center designed a five-year program that threads empathy training through every stage of learning rather than isolating it to a single course.¹⁴ Students conduct structured home visits with patients early in training, receive workplace-based empathy assessments built on real patient feedback and revisit communication skills repeatedly as they progress through clinical rotations. The center now trains faculty from other institutions through a three-day course, betting the model can travel beyond Leicester.

That instinct, toward sustained exposure rather than a single intervention, shows up elsewhere, too. At the University of Navarra in Spain, an obligatory medical identity journey program introduced in 2013 runs through the fourth, fifth and sixth years of the medical curriculum, pairing humanities-based workshops and case discussion with personal written reflections tied to each clinical rotation. A longitudinal study tracking one class across all six years found a significant increase in self-reported empathy between the end of year three and the end of year six, the exact window the program covers, a reversal of the pattern researchers usually find, suggesting that spaced, repeated reflection may do more than any single course.¹⁵ It’s worth noting the same research group’s earlier work found standardized patients rated trained students as more empathic, even when the students’ own self-reported scores didn’t move, a reminder that empathy measures don’t always agree with each other.

Teaching methods themselves are also shifting. An examination of empathy training across medicine, nursing and allied health increasingly favors structured role play with real-time feedback over lecture-based instruction, reinforcing something Dr. Riess’ work has long argued: Empathy behaves less like a fact to memorize and more like a clinical skill to drill.¹²  

What This Means for the Profession

The full arc of medical training, from the first year of medical school through residency and into independent practice, is where empathy is formed, tested and, the evidence increasingly suggests, at real risk of quiet erosion. That decline is modest, structural rather than a failure of character, and consequential enough for patient satisfaction and very possibly for clinical outcomes, that it deserves more than passing concern.

Dr. Verghese’s framing is the more useful one for educators to adopt: The goal isn’t to instill empathy students lack, but to protect what most already bring on day one. Curriculum reform, protected reflective time, communication skills training and narrative medicine programs all point to the same conclusion: Empathy is a clinical competency like any other, one that must be taught, assessed and defended throughout training rather than assumed to survive on its own. Whether medical education is doing that well enough remains an open question, and one the profession can no longer treat as optional.