Jefferson Scale of Physician Empathy (JSPE)

Abstract

The Jefferson Scale of Physician Empathy (JSPE), developed by Dr. Mohammadreza Hojat and colleagues at the Center for Research in Medical Education and Health Care at Sidney Kimmel Medical College (formerly Jefferson Medical College) of Thomas Jefferson University, is the premier psychometric instrument specifically designed to operationalize and measure empathy within the context of medical education and patient care. While general psychological measures of empathy such as the Interpersonal Reactivity Index (IRI) conceptualize empathy as a multidimensional construct encompassing both emotional contagion and cognitive perspective-taking, the JSPE explicitly defines clinical empathy as predominantly a cognitive attribute. This cognitive orientation involves the ability to understand a patient’s inner experiences and perspective, combined with the capacity to communicate this understanding and an intention to help, while maintaining professional boundaries to prevent personal distress and clinical burnout.

The JSPE comprises 20 self-report items scored on a 7-point Likert scale ranging from 1 (Strongly Disagree) to 7 (Strongly Agree), yielding a theoretical total score between 20 and 140, where higher scores indicate higher levels of clinical empathy. Ten items are positively phrased, and ten items are reverse-scored (R) to mitigate acquiescence response bias. Extensive factor analytic studies across diverse cultural and linguistic cohorts consistently reveal a stable three-factor structural model: Perspective Taking (the dominant factor capturing cognitive appraisal and patient perspective understanding), Compassionate Care (reflecting the emotional engagement, warmth, and the affective dimension of the therapeutic relationship), and Standing in the Patient’s Shoes (measuring the effortful cognitive simulation of the patient’s existential position). Across international investigations, the JSPE has demonstrated robust psychometric integrity, with internal consistency coefficients (Cronbach’s alpha) typically spanning between .78 and .89, alongside established convergent, discriminant, criterion-related, and predictive validity.

Keywords

Jefferson Scale of Physician Empathy, JSPE, clinical empathy, physician-patient communication, medical education, perspective taking, compassionate care, psychometrics, patient outcomes, emotional validation.

Authors

The Jefferson Scale of Physician Empathy was developed and validated by a distinguished multidisciplinary research team based at Thomas Jefferson University in Philadelphia, Pennsylvania, USA:

• Mohammadreza Hojat, Ph.D.: Research Professor of Psychiatry and Human Behavior, Director of the Jefferson Longitudinal Study of Medical Education, Center for Research in Medical Education and Health Care, Sidney Kimmel Medical College at Thomas Jefferson University. Dr. Hojat is the principal architect of the scale and has published foundational empirical monographs on clinical empathy.
• Salvatore Mangione, M.D.: Associate Professor of Medicine, Division of Pulmonary and Critical Care Medicine, Sidney Kimmel Medical College, Thomas Jefferson University.
• Thomas J. Nasca, M.D., MACP: Professor of Medicine and former Dean of Jefferson Medical College; Chief Executive Officer of the Accreditation Council for Graduate Medical Education (ACGME).
• Michael J. Cohen, M.D.: Department of Psychiatry and Human Behavior, Sidney Kimmel Medical College, Thomas Jefferson University.
• Joseph S. Gonnella, M.D.: Distinguished Professor of Medicine, Dean Emeritus, and Founder of the Center for Research in Medical Education and Health Care, Sidney Kimmel Medical College at Thomas Jefferson University.
• James B. Erdmann, Ph.D.: Center for Research in Medical Education and Health Care, Sidney Kimmel Medical College, Thomas Jefferson University.
• M. Magee, M.D.: Collaborating clinical investigator, Thomas Jefferson University.

Institutional Contact & Administration: Center for Research in Medical Education and Health Care (CRMEHC), Sidney Kimmel Medical College, Thomas Jefferson University, 1025 Walnut Street, Suite 119, Philadelphia, PA 19107, USA. Web repository: Jefferson Scale of Empathy Official Portal.

Purpose

The primary purpose of the Jefferson Scale of Physician Empathy (JSPE) is to provide an empirically grounded, psychometrically sound, and domain-specific assessment instrument capable of quantifying empathy as manifested specifically in healthcare delivery, doctor-patient interactions, and professional medical training environments. Prior to the operationalization of the JSPE in 2001, researchers evaluating empathy in healthcare professionals were forced to rely on generalized psychological scales, such as the Interpersonal Reactivity Index, Hogan’s Empathy Scale, or Mehrabian’s Balanced Emotional Empathy Scale. While informative within social psychological laboratories, these general measures yielded substantial ecological validity limitations when transferred into clinical environments, as they conflated cognitive clinical empathy with visceral emotional sympathy, personal distress, and affective over-involvement.

The JSPE addresses critical applications across multiple distinct domains:

• Medical Education Assessment: Tracking the development, preservation, or erosion of clinical empathy across longitudinal milestones in undergraduate medical training (preclinical to clinical transition), residency training, and continuous professional development. Multiple longitudinal studies have identified an alarming decline in empathy scores during the third year of medical school—coinciding with the initiation of intensive clinical clerkships—making the JSPE an essential diagnostic tool for educational interventions.
• Curricular Evaluation: Measuring the pedagogical efficacy of empathy-enhancing educational interventions, such as narrative medicine workshops, reflective writing courses, medical humanities integration, communication skills training, and standardized patient clinical simulations.
• Clinical and Health Services Research: Evaluating the empirical links between clinician empathy and objective patient outcomes. Research utilizing the JSPE has demonstrated direct correlations between higher physician empathy scores and superior patient therapeutic adherence, tighter glycemic control in diabetic patients (lower HbA1c levels), reduced low-density lipoprotein (LDL) cholesterol, fewer acute metabolic complications, lower rates of clinical malpractice claims, and enhanced overall patient satisfaction.
• Professional Wellbeing and Burnout Monitoring: Examining the relationship between empathy levels, professional fulfillment, depersonalization, and emotional exhaustion among practicing physicians and residents, thereby identifying clinicians at risk of systemic burnout.

Psychological Construct

The construct assessed by the Jefferson Scale of Physician Empathy is operationalized as a multidimensional, predominantly cognitive capability that enables healthcare providers to perceive, decode, understand, and communicate therapeutic appreciation of a patient’s internal psychological experiences, perspective, and suffering, coupled with an intrinsic drive to alleviate that distress. Hojat and colleagues rigorously differentiate clinical empathy from sympathy. While sympathy is characterized by affective immersion, emotional contagion, shared suffering, and potential loss of clinical objectivity, clinical empathy demands an intellectual appraisal of the patient’s state while preserving a clear cognitive boundary between the self and the other.

The scale captures three primary psychometric dimensions:

1. Perspective Taking

Perspective Taking constitutes the central core of the scale, representing the active, cognitive capacity of the clinician to mentally step outside their own subjective frame of reference and reconstruct the patient’s worldview, psychological dilemmas, and emotional landscape. In items such as “I try to understand what is going on in my patients’ minds by paying attention to their nonverbal cues and body language” and “I try to imagine myself in my patients’ shoes when providing care to them,” this dimension measures the clinician’s explicit effort to decode explicit verbal statements alongside implicit physiological and behavioral signals. It also indexes the rejection of purely biomedical reductionism, recognizing that the patient’s illness narrative and life context are intrinsic determinants of medical diagnosis and therapeutic efficacy.

2. Compassionate Care

Compassionate Care captures the clinician’s affective warmth, humanistic responsiveness, and acknowledgment that interpersonal connection is an essential therapeutic ingredient of medical healing. Rather than viewing the clinical encounter as an impersonal technical transaction, this dimension measures the degree to which the clinician values the therapeutic alliance. It is anchored in items such as “I believe that empathy is an important therapeutic factor in medical treatment” and “My understanding of my patients’ feelings gives them a sense of validation that is therapeutic in its own right.” Furthermore, this dimension assesses the clinician’s willingness to recognize emotional suffering without dismissing it as clinical interference, repudiating the belief that medical treatment is strictly confined to pharmacological or surgical interventions.

3. Standing in the Patient’s Shoes

Standing in the Patient’s Shoes reflects the complex cognitive capacity to achieve experiential simulation—the clinician’s subjective appraisal of how difficult or accessible it is to imagine themselves situated within the specific existential, social, and physical limitations confronting the patient. This dimension is measured primarily through reverse-coded items, such as “It is difficult for me to view things from my patients’ perspectives” and “Because people are different, it is almost impossible for me to see things from my patients’ perspectives.” High scores on this dimension signify that the clinician does not view individual human differences or demographic diversity as insurmountable barriers to clinical rapport and empathic understanding.

Theoretical Framework

The theoretical architecture underpinning the Jefferson Scale of Physician Empathy draws extensively from humanistic psychology, social cognitive neuroscience, psychotherapeutic theory, and biomedical ethics. Foremost among these influences is the client-centered psychotherapy paradigm formulated by Carl Rogers. Rogers conceptualized empathy as entering the private perceptual world of the other and becoming thoroughly at home in it, sensing the other’s feelings accurately without ever losing the “as if” quality. If this “as if” condition is compromised, the state ceases to be empathy and degenerates into emotional identification or vicarious personal distress, which Rogers and subsequent clinical researchers recognized as detrimental to professional efficacy.

From a contemporary cognitive and neurobiological perspective, the JSPE aligns with dual-process models of empathy, which bifurcate empathic phenomena into: (1) a bottom-up, phylogenetically older emotional contagion system mediated by the mirror neuron system and anterior insula, and (2) a top-down, neocortically mediated cognitive mentalizing network involving the medial prefrontal cortex, temporoparietal junction, and superior temporal sulcus (Theory of Mind). The JSPE purposely emphasizes the latter mentalizing circuitry. This theoretical stance is clinically vital: while unfiltered emotional contagion triggers sympathetic over-arousal and emotional exhaustion, top-down cognitive perspective-taking equips the physician with diagnostic insight and emotional stability, enabling sustained therapeutic presence across chronically traumatic environments.

Furthermore, the scale operationalizes the Biopsychosocial Model proposed by George L. Engel. Engel argued that conventional biomedical frameworks fail to account for the complex interplay of biological, psychological, and social determinants in illness presentation and treatment compliance. The JSPE translates Engel’s theoretical critique into direct psychometric indicators, systematically testing whether a clinician views psychological and social exploration as vital clinical evidence or as irrelevant distractions from somatic pathology.

Validity

The JSPE has undergone rigorous international psychometric validation across medical students, residents, practicing physicians, and diverse allied health professionals, generating substantial evidence for construct, convergent, discriminant, criterion, and predictive validity.

Construct and Factorial Validity

Structural construct validity was originally established by Hojat et al. (2001) in an initial sample of medical students and internal medicine residents. Exploratory factor analyses revealed clear three-factor solutions that accounted for substantial proportions of common variance. Subsequent confirmatory factor analyses (CFA) conducted across independent, multi-institutional international samples (e.g., in the USA, Italy, Japan, Iran, Mexico, and Australia) consistently verified the superiority of the hypothesized three-factor oblique model over unidimensional alternatives, yielding root mean square error of approximation (RMSEA) values consistently below .06 and comparative fit index (CFI) values exceeding .90.

Convergent and Discriminant Validity

Convergent validity is documented through robust statistically significant correlations with established psychological instruments measuring related constructs. In a landmark study by Hojat et al. (2005) comparing the JSPE with Davis’s Interpersonal Reactivity Index (IRI), JSPE total scores correlated positively and significantly with the IRI Perspective Taking subscale ($r = .40, p < .001$) and the Empathic Concern subscale ($r = .41, p < .001$). Crucially, demonstrating discriminant validity, the JSPE demonstrated non-significant or negligible correlations with the IRI Personal Distress subscale ($r = -.05$), confirming that clinical empathy as operationalized by the JSPE is independent of neurotic emotional contagion or personal anxiety in clinical situations.

Further convergent validity is demonstrated through moderate positive correlations with measures of emotional intelligence, such as the Mayer-Salovey-Caruso Emotional Intelligence Test (MSCEIT), social desirability independence, and positive correlations with the warmth and dutifulness facets of the NEO-PI-R personality inventory.

Criterion-Related and Predictive Validity

Criterion-related validity has been demonstrated by significant differences in JSPE scores between medical specialties. Clinicians practicing “people-oriented” specialties (e.g., family medicine, general internal medicine, pediatrics, psychiatry, obstetrics and gynecology) consistently score significantly higher on the JSPE than clinicians practicing “procedure- or technology-oriented” specialties (e.g., general surgery, orthopedics, radiology, pathology, anesthesiology). Furthermore, female physicians and female medical students consistently score statistically significantly higher than their male counterparts across virtually all published cultural samples ($d \approx 0.30 – 0.45$).

Most compellingly, predictive validity has been substantiated through objective clinical outcome markers. Del Canale et al. (2012), in a landmark study involving 20,961 diabetic patients and 242 primary care physicians in Parma, Italy, showed that physicians with high JSPE scores had significantly fewer patients with acute metabolic complications requiring hospitalization compared to physicians with low empathy scores (adjusted odds ratio = 0.59, $p < .05$), establishing that psychometric empathy directly predicts clinical therapeutic outcomes.

Reliability

The Jefferson Scale of Physician Empathy displays high reliability across varied healthcare populations, demographic sub-cohorts, and cross-cultural adaptations.

• Internal Consistency: Across extensive empirical investigations, Cronbach’s coefficient alpha ($lpha$) for the overall scale consistently ranges between .78 and .89. In the original development study by Hojat et al. (2001), the internal consistency was reported at $lpha = .89$ for medical students and $lpha = .87$ for medical residents. Similar estimates have been verified globally: Italian translation ($lpha = .85$), Japanese translation ($lpha = .80$), Persian translation ($lpha = .83$), Spanish translation ($lpha = .82$), and Korean translation ($lpha = .84$). Split-half reliability coefficients typically exceed .82.
• Test-Retest Reliability: Temporal stability assessments demonstrate robust test-retest reliability across multiple time horizons. Over a 3- to 4-month interval without intervening educational training, test-retest reliability coefficients have been documented at $r = .65$ to $r = .75$ ($p < .001$). Over shorter intervals of two to three weeks, test-retest stability coefficients reach $r = .85$, indicating that the scale measures a stable psychological attribute while maintaining responsiveness to targeted educational interventions.
• Standard Error of Measurement: The standard error of measurement (SEM) for the JSPE is relatively low, typically estimated around 5.0 to 6.2 score points across a 120-point operational range, indicating precise individual-level score estimations.

Factor Analysis

The internal structural validity of the JSPE has been extensively scrutinized using exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) across numerous sample populations.

In the original EFA reported by Hojat et al. (2001) utilizing principal axis factoring followed by oblique rotation (promax), three primary factors emerged with eigenvalues greater than 1.0, explaining a major portion of the common variance:

• Factor 1: Perspective Taking — Accounting for the largest portion of explained variance (approximately 25% to 35% across samples). Items loading heavily on this factor include Items 2, 4, 7, 8, 9, 10, 11(R), 12(R), 13(R), and 14(R), with standardized factor loadings predominantly ranging from .45 to .78. This factor reflects the active mentalization of patient experience and the explicit rejection of pure biomedical reductionism.
• Factor 2: Compassionate Care — Accounting for 5% to 9% of the variance. Items loading robustly onto this factor include Items 1, 3, 5, 6, 15(R), 16(R), and 18(R), with standardized loadings between .40 and .72. This factor captures the emotional warmth, emotional validation, and therapeutic alliance.
• Factor 3: Standing in the Patient’s Shoes — Accounting for 4% to 6% of the variance. This factor is primarily defined by the substantial negative loadings of Items 19(R) and 20(R) (standardized loadings often exceeding .65), assessing the clinician’s perceived difficulty in adopting the subjective psychological perspective of diverse patients.

Subsequent CFA studies conducted across various international samples have systematically tested multiple alternative nested models: a single-factor unidimensional model, a two-factor model combining compassionate care and perspective taking, and the established three-factor oblique model. The three-factor model consistently produces the best fit across global studies: Comparative Fit Index ($ ext{CFI} ge .92$), Tucker-Lewis Index ($ ext{TLI} ge .90$), Root Mean Square Error of Approximation ($ ext{RMSEA} le .054$, 90% CI [.048, .061]), and Standardized Root Mean Square Residual ($ ext{SRMR} le .052$), supporting the invariant multidimensional structure of the scale across healthcare specialties and languages.

Instrument / Measurement Tool

The JSPE is a self-administered, structured psychometric inventory. Key administrative and scoring parameters include:

• Target Population: Practicing physicians, medical residents, and clinical fellows. Parallel validated versions exist for medical students (S-Version), non-physician healthcare professionals (HP-Version), and health professions students (HPS-Version).
• Administration Time: Approximately 5 to 10 minutes.
• Number of Items: 20 items.
• Item Phrasing Distribution: 10 positively worded items (Items 1, 2, 3, 4, 5, 6, 7, 8, 9, 10) and 10 negatively worded / reverse-scored items (Items 11, 12, 13, 14, 15, 16, 17, 18, 19, 20).
• Response Format: 7-point Likert response scale:
• 1 = Strongly Disagree
• 2 = Moderately Disagree
• 3 = Slightly Disagree
• 4 = Neither Agree nor Disagree (Neutral)
• 5 = Slightly Agree
• 6 = Moderately Agree
• 7 = Strongly Agree
• Scoring Protocol:
  • Direct Items: For positively phrased items (1 through 10), assign direct numerical point values corresponding to the selected rating (1 = 1, 2 = 2, 3 = 3, 4 = 4, 5 = 5, 6 = 6, 7 = 7).
  • Reverse Items: For reverse-coded items indicated by (R) (Items 11 through 20), invert the scoring as follows: $7 = 1$, $6 = 2$, $5 = 3$, $4 = 4$, $3 = 5$, $2 = 6$, $1 = 7$. (Computational formula: $ ext{Reversed Score} = 8 – ext{Raw Score}$).
  • Total Score Calculation: Sum all 20 item scores following reverse transformation. Total scores range from a minimum of 20 to a maximum of 140.
  • Missing Data Rule: If more than 4 items (20% of the instrument) are omitted, the questionnaire is considered invalid. If 4 or fewer items are missing, impute the missing responses using the mean of the completed items for that individual.
  • Score Interpretation: Higher total scores reflect greater self-reported orientation toward cognitive clinical empathy in therapeutic encounters.

Permissions & Fee and Test Year

The Jefferson Scale of Physician Empathy was officially introduced and published in 2001 by Dr. Mohammadreza Hojat and the research faculty at Thomas Jefferson University.

Copyright and Licensing: The JSPE and its associated variants (S-Version, HP-Version, HPS-Version, and the patient-reported JSPE-VP) are legally protected intellectual property copyrighted by Thomas Jefferson University (All Rights Reserved). The scale is not in the public domain. While the complete item statements have appeared in academic empirical validation literature, formal clinical or institutional utilization, large-scale educational administration, and academic research protocols require explicit licensing and registration through Thomas Jefferson University.

Licensing and Fee Structure: Thomas Jefferson University requires academic investigators and educational institutions to complete a user agreement before using the instrument. A nominal per-administration or site licensing fee is typically assessed to support ongoing longitudinal research, psychometric benchmarking, and automated scoring services provided by the Center for Research in Medical Education and Health Care (CRMEHC). Reduced fee structures are accessible for student-led dissertation research upon institutional review board (IRB) approval. Detailed information regarding permissions and authorized translation protocols is accessible via the Jefferson Center for Research in Medical Education and Health Care.

References

• Del Canale, S., Louis, D. Z., Maio, V., Wang, X., Rossi, G., Hojat, M., & Gonnella, J. S. (2012). The relationship between physician empathy and disease complications: An empirical study of primary care physicians and their diabetic patients in Parma, Italy. Academic Medicine, 87(9), 1243–1249. https://doi.org/10.1097/ACM.0b013e3182628fbf
• Hojat, M., Mangione, S., Nasca, T. J., Cohen, M. J., Gonnella, J. S., Erdmann, J. B., Veloski, J., & Magee, M. (2001). The Jefferson Scale of Physician Empathy: Development and preliminary psychometric data. Educational and Psychological Measurement, 61(2), 349–365. https://doi.org/10.1177/00131640121971158
• Hojat, M., Gonnella, J. S., Mangione, S., Nasca, T. J., Veloski, J. J., Erdmann, J. B., Callahan, C. A., & Magee, M. (2002). Empathy in medical students as related to academic performance, clinical competence and gender. Medical Education, 36(6), 522–527. https://doi.org/10.1046/j.1365-2923.2002.01234.x
• Hojat, M., Mangione, S., Kane, G. C., & Gonnella, J. S. (2005). Relationships between scores of the Jefferson Scale of Physician Empathy (JSPE) and the Interpersonal Reactivity Index (IRI). Medical Teacher, 27(7), 625–628. https://doi.org/10.1080/01421590500069744
• Hojat, M. (2016). Empathy in health professions education and patient care. Springer International Publishing. https://doi.org/10.1007/978-3-319-27611-3
• Kane, G. C., Gotto, J. L., Mangione, S., West, S., & Hojat, M. (2007). Jefferson Scale of Patient’s Perceptions of Physician Empathy: Preliminary psychometric data. Croatian Medical Journal, 48(1), 81–86. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2080498/
• Shariat, S. V., Eshtad, E., & Ansari, S. (2010). Empathy and its correlates in Iranian physicians: A preliminary psychometric study of the Jefferson Scale of Physician Empathy. Medical Teacher, 32(10), e417–e421. https://doi.org/10.3109/0142159X.2010.498488
• Williams, B., Brown, T., Boyle, M., & Dousek, S. (2013). Psychometric testing of the Jefferson Scale of Empathy Health Profession Students’ version with Australian paramedic students. Nursing & Health Sciences, 15(1), 45–50. https://doi.org/10.1111/nhs.12003

Items of the Scale