How can we teach empathy to medical students? A scoping review

Empathy is essential for healthcare professionals, enhancing patient-physician relationships, clinical outcomes, and patient adherence while reducing medical errors. It comprises emotional, cognitive, moral, and behavioral components. Empathy levels among medical students can be influenced by educational and environmental factors. Our study aims to identify and analyze educational determinants that shape empathy during medical training. Following PRISMA-ScR guidelines, a scoping review was conducted, searching seven databases (e.g., MEDLINE, APA PsycINFO, Embase, Web of Science, PubMed, CINAHL, and LiSSa) for studies published between 2002 and 2025 in English or French. Inclusion criteria: (1) assessment of empathy using validated scales (2), focus on undergraduate medical students (3), examination of educational determinants. Exclusion criteria: solely qualitative studies (mixed method studies were included, but only the quantitative data were analyzed), non-educational factors, or non-medical student populations. Study quality was assessed using the MERSQI tool before a cross-sectional and descriptive analysis. Ninety-two studies were included. The Jefferson Scale of Empathy (JSE-MS) was the most commonly used tool ( n = 56; 60.9%). Five categories emerged: (1) psychosocial skills development (2), empathy-focused curricula (3), patient encounters (4), clinical simulations, and (5) conceptual teaching on communication/empathy. Single-category studies showed significant improvements in empathy, as in n = 38 (41.3%) of studies with psychosocial skills interventions. Combined interventions were often reported to have positive effects ( n = 72, 78.3%). Results highlight a heterogeneity in educational approaches and their variable impact on empathy. The most effective interventions combined multiple determinants, such as theoretical teaching, clinical simulations, and patient interactions. However, empathy measurement tools (mostly self-report scales) have limits, including social desirability bias and low sensitivity to change. Observed variations often fall below detectable thresholds, warranting cautious interpretation. A mixed-methods approach may provide more robust assessments. Educators can combine diverse educational interventions, including early clinical experience supported by educators, in an integrative, longitudinal, and reflexive approach to foster empathy among medical students.