Can Empathy Be Taught? The Science and Practice of Clinical Empathy Training

Can empathy actually be taught?

It depends on what we mean by “teach.”

A medical school cannot simply give someone a lecture and install compassion.

Nor does performing a particular communication technique prove that a clinician genuinely understands another person's experience.

But clinical empathy is not only an internal feeling.

It also involves attention, interpretation, communication and response.

Those elements can be practised.

That distinction may resolve much of the argument.

Rather than asking whether educators can manufacture empathy as a personality trait, a more useful question is:

Can clinicians become better at noticing patients' perspectives and emotions, understanding what they mean, and communicating that understanding effectively?

The evidence increasingly suggests that they can.

Training the expression of empathy is different from teaching people to care

One of the useful ideas in early clinical-empathy education was that the objective should not necessarily be to “teach empathy” itself.

Instead, educators can teach the tools through which empathy functions in clinical practice.

These include listening, asking useful questions, recognising verbal and non-verbal emotional cues, perspective-taking, responding to emotion, reflecting on assumptions and regulating one's own reactions during difficult encounters.

That distinction matters.

A clinician may care deeply and still interrupt, become defensive or fail to communicate understanding.

Another clinician may know the correct empathic phrase yet deliver it mechanically.

Clinical empathy requires more than intention and more than technique.

The aim is to bring the two together.

What kinds of training have been tried?

Medical educators have experimented with an unusually wide range of approaches.

Communication-skills training asks learners to practise listening, questioning and responding to emotion.

Standardized-patient encounters allow clinicians to rehearse difficult conversations and receive feedback.

Role-play asks learners to experience an encounter from another position.

Narrative medicine, literature and reflective writing invite closer attention to another person's story and to the clinician's own reactions.

Theatre and improvisation develop attention, presence, flexibility and responsiveness when conversations do not follow a script.

Mindfulness and self-regulation approaches concentrate partly on what happens inside the clinician when an encounter becomes stressful.

The diversity of these methods reflects something important:

Empathy is not one behaviour.

Practice appears to matter more than being told to be empathic

Recent evidence increasingly favours active learning.

A 2023 review of empathy education found that interventions extending over time were more likely to report improvements than isolated one-off sessions, and that experiential and affective learning approaches generally performed better than purely didactic teaching.

A larger 2025 systematic review and meta-analysis of randomized trials reached a broadly similar conclusion.

Across 111 studies involving more than 11,000 participants, empathy interventions produced a moderate overall improvement compared with controls.

The studies were highly heterogeneous and publication bias was present, so the number should not be interpreted as a universal effect.

But the analysis provides considerably stronger support for training than simply pointing to a handful of successful courses.

Notably, combinations involving instruction plus rehearsal were among the more effective approaches.

That makes intuitive sense.

Reading about listening is not the same as listening while another person is distressed.

Patient feedback changes what we mean by success

There is another problem with empathy training:

Who decides whether it worked?

Historically, much empathy research has asked clinicians or students to rate themselves.

But being convinced that we are empathic is not necessarily the same as being experienced that way by another person.

In one observational study involving 51 physicians and 945 patients, physicians' overall self-rated empathy did not significantly correlate with their patients' assessments.

That does not make self-reflection useless.

It means that the patient's perspective should be part of the outcome.

A training programme has achieved something particularly meaningful when patients themselves detect a difference.

A small randomized trial illustrates the point

In a randomized controlled trial involving 99 residents and fellows, clinicians received three one-hour modules of empathy training or continued with standard postgraduate education.

The primary outcome was not simply whether participants felt more empathic.

Patients rated their clinicians.

The trained group showed a statistically greater improvement in patient-rated relational empathy than the control group.

Interestingly, the clinicians' own Jefferson Scale empathy scores did not show a statistically significant difference between groups.

The study was relatively small and follow-up was short, so it should not be treated as definitive proof.

But it demonstrates why measurement matters.

A clinician's self-concept may remain largely unchanged while the way patients experience the clinical interaction improves.

Role-play is useful when it is rehearsal rather than theatre

Role-play sometimes attracts understandable skepticism.

No patient wants to feel that a clinician is performing compassion from a script.

But that is not the most useful way to understand rehearsal.

Clinicians rehearse resuscitation, surgical techniques and difficult procedures because performance under pressure improves through practice.

Communication also takes place under pressure.

Practising how to acknowledge emotion, tolerate silence, respond to anger or explore disagreement gives clinicians a larger repertoire when those situations occur for real.

The goal is not to memorize empathic sentences.

It is to become sufficiently comfortable with difficult interpersonal situations that technique no longer occupies all of one's attention.

Improvisation offers another interesting model

Clinical conversations rarely proceed exactly as expected.

A patient gives an unexpected answer.

A relative becomes angry.

A carefully planned explanation produces silence.

A clinician realises that the problem they thought they were solving is not the problem the patient most wants addressed.

Medical improvisation training approaches this uncertainty differently from scripted communication teaching.

It develops close listening, presence, responsiveness and the ability to build on what another person has actually said rather than mentally rehearsing the next response.

The evidence base for medical improvisation is much smaller than the broader empathy-training literature, so it should not be presented as a proven superior method.

But conceptually it addresses something important:

Clinical empathy has to operate in conversations that cannot be scripted.

The environment can reinforce—or undo—the training

No empathy curriculum exists in isolation.

A student may spend an afternoon learning reflective listening and then enter a clinical environment where respected physicians interrupt patients, ridicule concerns or treat relational care as wasted time.

The informal lesson can overpower the formal one.

The reverse is also possible.

Learners who observe senior clinicians combine efficiency, curiosity, boundaries and respect gain something no communication slide deck can provide:

a model of what empathic clinical practice actually looks like.

That is why role modelling, supervision and institutional culture belong in any serious discussion of empathy education.

Training also needs to include self-regulation

Empathy should not require clinicians to absorb unlimited distress.

Healthcare professionals repeatedly encounter fear, pain, uncertainty, grief and anger while also managing responsibility, fatigue and cognitive load.

Effective clinical empathy therefore includes the capacity to recognise another person's experience without becoming incapacitated by it.

Training in self-awareness, reflection and emotional regulation is not opposed to empathy.

It may help make empathy sustainable.

So: can empathy be taught?

The most defensible answer is:

Important components of clinical empathy can be developed through training.

The evidence is strongest for approaches that involve active practice rather than instruction alone, and there is growing support for combining methods rather than relying on a single workshop.

But several cautions remain.

Studies use different definitions and measurement tools.

Follow-up is often short.

Self-reported empathy and patient-perceived empathy are not interchangeable.

And an educational intervention cannot completely compensate for a clinical culture that repeatedly undermines what it teaches.

So the goal should not be to produce clinicians who score highly on an empathy questionnaire.

It should be to develop clinicians who can understand patients more accurately, communicate that understanding, remain open when encounters become difficult and sustain those capacities throughout a demanding career.

That is a much more useful definition of success.

Selected evidence

Riess H, Kelley JM, Bailey RW, Dunn EJ, Phillips M. Empathy Training for Resident Physicians: A Randomized Controlled Trial of a Neuroscience-Informed Curriculum. Journal of General Internal Medicine. 2012.

Samarasekera DD, Lee SS, Yeo JHT, Yeo SP, Ponnamperuma G. Empathy in health professions education: What works, gaps and areas for improvement. Medical Education. 2023.

Ngo H, Sokolovic N, Hu J, Jenkins J. A systematic review identifying effective teaching methods and their combinations for increasing empathy in physicians: pairwise and network meta-analysis. BMC Medical Education. 2025.

Bernardo MO, Cecílio-Fernandes D, Costa P, et al. Physicians' self-assessed empathy levels do not correlate with patients' assessments. PLOS ONE. 2018.