https://orcid.org/0000-0001-5981-9393
Figures
Abstract
Background
Therapeutic empathy improves patient and practitioner outcomes, yet existing measures are often lengthy, conceptually inconsistent, and difficult to compare across respondent groups. Brief, universal measures (usable by patients, practitioners, students, and observers) are lacking. We developed a universal single-item scale – the Rapid Assessment of Therapeutic Empathy (RATE) scale – and conducted psychometric testing of the patient-reported version.
Methods
We used a three-phase approach: (1) item development; (2) pre-testing the scale by obtaining expert panel feedback (n = 9) and conducting cognitive interviews with stakeholders (n = 35); and (3) scale validation in an international patient sample (n = 521) assessing convergent, discriminant, and known-groups validity. Validation involved assessing correlations with the Consultation and Relational Empathy (CARE) measure and clinical neutrality measure, and by assessing differences in scores by patient ethnicity.
Results
We developed two versions (pictorial and text-based) of each scale. Expert feedback and cognitive interviews confirmed content and face validity. Pictorial and text-based versions showed high convergent validity with the CARE measure (r = 0.761 and r = 0.838, both p < 0.001), and discriminant validity with a clinical neutrality measure (r = 0.131 and r = 0.139, p = 0.003 and p = 0.001, respectively). Correlations with the CARE measure remained high (r > 0.70) and statistically significant (p < 0.001) across patient gender, ethnicity, and practitioner type. Ethnic minority patients rated practitioner empathy lower than White patients (pictorial p = 0.057; text-based p = 0.033), demonstrating known-groups validity. Patients rated doctors’ empathy higher than other healthcare practitioners’ (p = 0.001 for both pictorial and text-based); there were no significant differences in empathy scores by patient gender.
Conclusions
We developed the first universal single-item therapeutic empathy measure and demonstrated validity. The RATE scale is brief, accessible, and applicable to clinical practice, education, and research. Further research should validate practitioner-, student-, and observer-reported versions, and assess predictive and cross-cultural validity. This robust tool can support patient-reported routine measurement of therapeutic empathy and contribute to improving patient and practitioner outcomes.
Citation: Bennett-Weston A, Maltby J, Khunti K, Leung C, Narwal D, Otoo P, et al. (2026) Rapid Assessment of Therapeutic Empathy (RATE): Development of a universal single-item empathy scale and validation of the patient-reported version. PLoS One 21(9): e0346994. https://doi.org/10.1371/journal.pone.0346994
Editor: Ipek Gonullu, Ankara University Faculty of Medicine: Ankara Universitesi Tip Fakultesi, TÜRKIYE
Received: March 25, 2026; Accepted: July 25, 2026; Published: September 9, 2026
Copyright: © 2026 Bennett-Weston et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Our dataset comprises a combination of quantitative and qualitative data from three participant groups: expert panellists (n=9), key stakeholders from each of our respondent groups including patients, practitioners, healthcare educators, carers, and healthcare students (n=35), and patients (n=521). Data from expert panellists includes age, gender, ethnicity, professional role and area of expertise, geographic location, and both Likert-scale scores and free-text comments to assess the relevance of our newly developed Rapid Assessment of Therapeutic Empathy (RATE) scale. Data from key stakeholders from each respondent group includes age, gender, ethnicity, professional role and area of expertise, geographic location, and where relevant health condition, along with interview transcripts describing their thought processes as they complete the newly developed RATE scale, which involves recalling recent healthcare appointments that are likely to be identifying. Data from patients (n=521) includes age, gender, ethnicity, geographic location, details about their last healthcare consultation including the practitioner they saw and where it took place, details about their health condition and/or disability, scores on the RATE scales, the Consultation and Relational Empathy (CARE) measure, and a measure of clinical neutrality. The data are not owned by a third-party organisation. There are ethical restrictions on sharing our de-identified data set, imposed by the NHS Research Ethics Committee that approved the study (the West of Scotland Research Ethics Committee), the UK Health Research Authority, and our institutional (University of Leicester) General Research Ethics Committee and Research Governance Office (IRAS project ID: 350113; Research Ethics Committee reference: 25/WS/s0067). This is because the data (both quantitative and qualitative) contain potentially identifying and sensitive participant information, including details of healthcare consultations and health conditions, as well as participant ethnicity, gender, and age, and job role, which in combination can be identifying. We are not permitted by these bodies to share the de-identified data set publicly. Requests for data access may be sent to our departmental email (empathy@leicester.ac.uk) or our institutional ethics email (Ethics@le.ac.uk).
Funding: Jeremy Howick, Amber Bennett-Weston, Benita Iyadi-Wilson, and Patience Otoo were funded by the Stoneygate Trust (no grant reference available). Kamlesh Khunti is funded by the National Institute for Health Research (no grant reference available). Daya Narwal was funded by the National Institute for Health Research (grant reference: NIHR305951). John Maltby and Char Leung completed this work during the usual course of their employment at the University of Leicester and did not receive additional funding for this study. The funders had no role in developing the protocol. There was no additional external funding received for this study.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Therapeutic empathy (sometimes called “clinical empathy”) is a cornerstone of healthcare practice, research, and education [1–3]. For patients, therapeutic empathy can reduce pain, depression, and anxiety while improving safety and satisfaction with care [4–6]. For practitioners, therapeutic empathy is associated with reduced burnout and greater job satisfaction [7,8].
Challenges in defining therapeutic empathy have, until recently, inhibited research in this area. However, a recent systematic review and thematic analysis of definitions identified six common components across definitions: exploring (asking questions of, and actively listening to, the patient), understanding (the cognitive processes involved in trying to take the patient’s perspective), shared understanding (communicating understanding back to the patient to check its accuracy), feeling (emotionally engaging with the patient), therapeutic action (taking actions to help the patient), and maintaining boundaries (respecting the patient’s views and choices) [9–11]. These components highlight that therapeutic empathy has both cognitive and affective dimensions that are inextricably linked [11].
Despite conceptual advances, approaches to measuring therapeutic empathy remain highly variable [12]. Systematic reviews [13–15] have identified multiple practitioner self-reported, patient-reported, and observer-reported measures [16–23], although observer measures are rarely used because of reliability concerns [13]. Aside from concerns about the reliability and validity of existing measures, one of their main limitations is that none were designed for use across different respondent groups [13–15]. For example, the Jefferson Scale of Empathy (JSE) was initially developed and validated for use by physicians [24], while the Consultation and Relational Empathy (CARE) measure was designed specifically for patients to assess their practitioner’s empathy [25].
Different therapeutic empathy scales also appear to measure a variety of constructs [12–15]. For example, although the JSE defines empathy as involving cognitive understanding and therapeutic action [22], the measure focuses largely on cognitive elements [15]. Meanwhile, the CARE measure captures the relational aspects of empathic patient-practitioner interactions (for example, how well the practitioner listened), with less consideration of cognitive components of therapeutic empathy. Relatedly, observer-reported measures such as the Therapist Empathy Scale (TES) focus on the emotional aspects of empathy [13].
This variation in measures matters. If we wish to test whether an empathy intervention is effective for improving therapeutic empathy (and subsequently, patient and practitioner outcomes), we need to know which measure is optimal. Relatedly, if we wish to measure whether our training interventions [26] are effective at teaching therapeutic empathy to medical students, then it is important for observer-rated empathy measures to be suitably validated [13]. Moreover, the lack of conceptual coherence between measures makes it difficult to compare therapeutic empathy scores across perspectives or determine whether different scales measure the same construct [27], potentially contributing to discrepancies between practitioner self-ratings and patient ratings [28–30].
In an attempt to address this problem, some existing scales have undergone post-hoc adaptation for other groups. For example, the JSE has been adapted for use by patients [31], and the CARE measure has been adapted for use by practitioners [32]. However, such post-hoc adaptations repurpose instruments not specifically designed for these groups, risking misalignment between item content and the construct as experienced by different respondents. This, in turn, can compromise content validity and conceptual equivalence [33,34].
Another problem with existing scales is their complexity. Many have over 15 items and include negatively worded questions [16–24]. This can increase survey fatigue and risks incomplete responses [35,36]. Even the widely used 10-item Consultation and Relational Empathy (CARE) measure [25] may be burdensome in some time-pressured clinical settings [37].
Single-item scales reduce time burden while offering effective measurement, and higher response and completion rates [38–44]. They are particularly suited for time-constrained clinical settings where routine measurement is required [45,46]. Single-item scales also reduce the risk of criterion contamination [40], improve face validity by removing ambiguous items [40,47,48], and simplify translation across diverse populations [49,50]. While concerns have been raised about the reliability and sensitivity of single-item scales, they have not been substantiated [39,51]. Although several single-item therapeutic empathy scales have been proposed [52–55], none have been validated.
It is therefore important to have a single-item, validated therapeutic empathy scale that measures the same construct and that is designed from the outset for use by patients, practitioners, students, and observers [13–15]. We describe this as a “universal” scale. Developing such a scale is especially important given recent research showing the importance of empathy for enhancing patient satisfaction [4], safety [5], and practitioner wellbeing [7].
Objectives
This study has two objectives:
- To develop the first “universal” single-item measure that integrates all six components of therapeutic empathy [11] and is suitable for use by patients, practitioners, students, and observers.
- To test the construct validity of the patient-reported version.
Materials and methods
Study design
We conducted a mixed-methods study to develop a simple, user-friendly, and universal single-item scale of therapeutic empathy, and to psychometrically test the patient-reported version. We called this new measurement tool the Rapid Assessment of Therapeutic Empathy (RATE) scale. Following established best-practice for developing and testing scales in health, social, and behavioural research [33,34,56], we followed three phases: 1) item development, 2) scale development, and 3) scale evaluation (Fig 1) [33]. We reported the study following recommendations for reporting scale development and testing [56].
Sampling and recruitment
We recruited three groups of participants:
- Stakeholders with expertise on therapeutic empathy in healthcare research, practice, or education to assess content validity of the RATE scale.
- Stakeholders from each target population (patients, practitioners, observers) to assess face validity and pre-test the RATE scale.
- Patients to complete the final patient-reported RATE scale to enable tests of validity
Sample sizes for each participant group were chosen to meet established standards [33,34,56]. All participating patients were offered remuneration in line with national guidance [57].
Stakeholders with expertise on therapeutic empathy to assess content validity.
We aimed to recruit a minimum of five expert panellists (including patient representatives, therapeutic empathy researchers and educators, and healthcare practitioners) to assess content validity [33,34]. We purposefully sampled [58] participants with relevant expertise in across medicine, nursing, patient advocacy, medical ethics, and scale development, of different ages, genders, and ethnicities, and from different continents [34,59], and invited them to participate via email between 2nd June 2025 and 2nd July 2025. All participants provided written informed consent to participate. We defined expertise as having made a substantial contribution to the field of therapeutic empathy either through research, teaching, or practice [60].
Stakeholders from each target population to assess face validity and pre-test the RATE scale.
We aimed to recruit at least 20 patients, practitioners, healthcare students, and observers (healthcare educators, carers) to participate in cognitive interviews to pre-test all versions of the RATE scale. The final sample size was determined by reaching meaning saturation: the point at which we had a rich understanding of participants’ views and experiences of using the RATE scale [61]. We purposefully selected patients of varied ages, ethnicities, genders, and healthcare experiences across the UK via local patient groups and the NIHR People in Research platform [62]. We also recruited a diverse sample of practitioners, students, and educators from multiple professions in the East Midlands, via a Medical School and a Healthcare School, and their associated NHS Trusts. Recruitment took place between 15th July 2025 and 30th July 2025. All participants provided written informed consent to participate.
Patients to complete the final patient-reported RATE scale to test validity.
We aimed to recruit at least 200 patients aged ≥18 years, of any gender or ethnicity, who had seen a healthcare practitioner in any setting within the past year to complete the final patient-reported version of the RATE scale. We recruited an international convenience sample of patients using online survey recruitment platform Prolific [63]. We supplemented this by recruiting in-person from General Practices, Maternity Services, Infectious Disease, and Geriatrics clinics across an NHS Trust and Primary Care Network (PCN) in the East Midlands of the UK to capture patients experiencing digital exclusion or limited digital literacy. Clinics were selected for their diverse populations [64,65]: the General Practices served the most deprived ward within their PCN [66]; the Maternity Services served women and families in the UK’s first plural city; the Geriatric Clinic served older people; and the Infectious Disease Clinic served many homeless people. A screening question based on previous research [67] assessed digital exclusion and illiteracy (S1 Appendix), and at least ten percent of the sample was recruited in-person to reflect the estimated proportion of digitally excluded or illiterate patients in the UK [68,69]. Recruitment took place between 1st September 2025 and 15th November 2025. All participants provided written informed consent to participate.
Procedures
Identification of domain and item generation.
Therapeutic empathy was the construct of interest, and recent systematic reviews indicated that existing measures did not meet the aims of the scale we intended to develop; none were designed for universal use by patients, practitioners, students, and observers, and many lacked user-friendliness [13–15]. We therefore started scale development, using previous qualitative research on patients’ and practitioners’ lived experiences of therapeutic empathy [70–72] and a recent systematic review which identified six common components of the concept (S2 Appendix) as our foundation for item generation [10,11]. In line with best-practice guidance [33], we began by developing the RATE scale in one language: English.
To support the accessibility of the RATE scale across different respondent groups with varying literacy levels [72], we adopted several strategies: first, we used Plain English language and aimed for a reading age of below 13 years [73]. To achieve this, we assessed the initial scales using readability tests: the Flesch-Kincaid Grade Level and Flesch Reading Ease Test [74]. Second, following previous studies [25,75], we aimed to develop two versions of the RATE scale for each target population: a pictorial version and a text-based version. Third, we opted for using unipolar five-point measures to reduce conceptual ambiguity [33,76].
Content validity.
To assess content validity [33], our expert panel reviewed the initial versions of the RATE scale, along with the study definition of therapeutic empathy [11], and provided anonymised feedback on clarity and relevance via an online survey (S3 Appendix). This feedback informed scale refinement.
Pre-testing the scale.
We assessed face validity and pre-tested the RATE scale through cognitive interviews [33] to explore how potential respondents interpreted and formulated answers to the scale, steered by a semi-structured topic guide (S4 Appendix) [77]. We offered participants the choice of conducting their interview either online (via Microsoft Teams version 25306.804 [78]) or in-person (at the host institution). All interviews were conducted by at least one researcher (from either ABW or DN), and were audio-recorded and transcribed. Using a light-touch approach to thematic analysis [79], we summarised the interviews and used this to further refine each scale. We conducted the cognitive interviews in rounds of at least five participants (including a mix of participants from each respondent group) to enable us to modify and test each version of the RATE scale iteratively [33,77,80], and continued until we reached meaning saturation [61]. By this point we had made exhaustive modifications, leaving us with the final versions of the scale.
Survey administration.
We administered the final versions of the patient-reported RATE scale to a larger sample of patients online via Prolific [63] and in-person at our NHS sites. We first tested the patient-reported version as this has been identified as a priority [72]. To measure convergent and divergent validity (see below), we administered the CARE measure [25] and a question about clinical neutrality (S5 Appendix), which we adapted from previous research [24].
Tests of validity.
Typically, establishing internal reliability and examining latent structure (for example via internal consistency indices and exploratory factor analysis) would be the next step in scale evaluation [33]. However, these approaches are not applicable to a single-item measure [33,41,51,81]. Moreover, test-retest reliability is primarily used to assess the stability of an underlying trait. This approach is unlikely to be meaningful because therapeutic empathy is inherently interaction- and context-dependent, rather than a stable characteristic of a single consultation [82,83].
Therefore, in the final step we assessed construct validity, including convergent validity, discriminant validity, and known-groups validity [84] using R Version 4.5. We assessed convergent validity by measuring correlations between the two versions of the RATE scale and one other well-established patient-reported measure of therapeutic empathy: the CARE measure [22,25]. A high correlation would represent high convergent validity. To further explore the robustness of any correlations between our scales and the CARE measure [25], we repeated the correlational analyses across key demographic subgroups, including patient gender, ethnicity, and type of practitioner they were treated by [33].
We assessed discriminant validity by measuring correlations between our scale and a measure of a theoretically unrelated construct: clinical neutrality. This concept has been used in previous therapeutic empathy scale validation studies to assess discriminant validity [24]. A low correlation would indicate strong discriminant validity, by demonstrating that our measure does not correlate too highly with an unrelated construct [33].
We used Spearman’s Rank correlation coefficient to assess convergent and discriminant validity given the ordinal nature of the data, and interpreted correlation strength as follows: (-) 0.90 to 1.00 indicates a very high correlation, (-) 0.70 to 0.90 indicates a high correlation, (-) 0.50 to 0.70 indicates a moderate correlation, (-) 0.30 to 0.50 indicates a low correlation, and below (-) 0.30 indicates a negligible correlation [85]. In the context of convergent validity specifically, although a correlation coefficient of (-) 0.50 is broadly considered “sufficient”, it is recommended that a correlation coefficient of (-) 0.70 should be used as a benchmark to indicate that two scales genuinely measure the same construct [86].
Known-groups validity (the extent to which a scale can discriminate between two or more groups theoretically expected to differ on the variable of interest) was assessed by comparing RATE scale scores across groups expected to differ in their therapeutic empathy ratings using the Kruskal-Wallis test [33]. Research shows that perceptions and experiences of empathy in healthcare are influenced by ethnicity [59,87] and that specifically, patients from ethnic minority groups rate practitioner empathy lower than those from non-minority ethnic groups [88]. Accordingly, we conducted tests for known-groups validity between patients of different ethnicities.
To supplement our tests of validity, and following previous research [25], we conducted exploratory demographic comparisons between empathy ratings by patient gender and the type of practitioner they were treated by.
Results
Identification of domain and item generation
Having identified our domain – therapeutic empathy – we generated the single-item RATE scale, taking inspiration from the findings of previous research describing the nature of the construct [11,70–72]. For each respondent group, we generated two versions of the RATE scale: a pictorial measure using smiley-faces (ranging from not at all happy to very happy) as the response options, similar to the visual analogue scales used to measure pain [89,90], and a text-based measure ranging from “not at all empathic” to “extremely empathic” (see Fig 2 for initial patient-reported versions and S6 Appendix for other versions).
Flesch-Kincaid Grade Level scores (range 2.8–7.5) and Flesch Reading Ease Test scores (range 49.0–88.8) confirmed that the readability of the initial versions of the scale were suitable for the intended reading age of below for 13 years for each respondent group (S7 Appendix).
Content validity
In July 2025, nine participants joined our expert panel to assess the content validity of the initial versions of the RATE scale (67% female, 78% white, n = 2 patient representatives, n = 9 healthcare practitioners and educators, see S8 Appendix). Their feedback was positive; they described the scales as being easy to understand by each respondent group. Many expert panel members also believed that the description of therapeutic empathy provided was relevant to everyday clinical practice. They made suggestions for improving the scale, including refining the wording to better align with our definition of empathy and removing references to any single type of healthcare practitioner (for example, doctors or nurses), which we used to make modifications (see S9 Appendix for a detailed description of participant feedback and modifications made at each stage of development). The modified versions were pre-tested in cognitive interviews.
Pre-testing the RATE scale
In July and August 2025, 35 participants (comprised of patients, healthcare practitioners, healthcare students, and observers such as healthcare educators and carers), 49% female, 69% white, see S10 Appendix) participated in cognitive interviews lasting between 30 and 40 minutes. Three rounds of interviews were conducted; 15 participants took part in round one, 10 participants in round two, and 10 participants in round three. Each participant took part in one interview; 11 interviews were conducted in-person, 25 took place online. Several participants had dual roles (for example, as patients and carers, or as healthcare practitioners and educators), and were able to pre-test more than one version of the RATE scale.
In all three rounds of cognitive interviews, most participants endorsed the single-item nature of the RATE scale, explaining that the scale was easy to understand and quick to complete, often taking less than two minutes. Most participants (including healthcare practitioners and students) preferred the pictorial version of the scale, finding it more engaging and even faster to complete than the text-based version. Minor modifications to wording were suggested (S9 Appendix). After the third round of interviews, we had reached meaning saturation [61] and stopped interviewing. We modified the scales once more, leaving us with the final versions ready to be tested for validity (see Fig 3 for final patient-reported versions and S11 Appendix for other versions).
Scale administration and tests of validity
Between September and November 2025, 521 patients completed our newly developed RATE scales, along with the CARE measure [25] and a measure of clinical neutrality [24]. Table 1 summarises participants’ characteristics.
High convergent validity with the CARE measure was achieved for the pictorial (r = 0.761, p < 0.001) and text-based RATE scales (r = 0.838, p < 0.001). Strong convergent validity was also observed across key demographic subgroups (Table 2).
Similarly, both the pictorial (r = 0.131, p = 0.003) and text-based (r = 0.139, p = 0.001) versions demonstrated strong discriminant validity through negligible correlations with a measure of clinical neutrality [24]. We also found evidence of known-groups validity based on patient ethnicity [33]; we observed differences in median therapeutic empathy ratings between patients from non-minority ethnic groups and patients from ethnic minority groups. Patients of White ethnicity assigned higher therapeutic empathy ratings to their practitioners than patients from other ethnic groups. Although statistical significance differed slightly between the pictorial and text-based versions (p = 0.057 and p = 0.033, respectively), both showed the same directional pattern. Fig 4 summarises the results of our statistical tests of validity.
Our exploratory demographic comparisons showed statistically significant differences between patients’ ratings of doctors’ empathy versus other healthcare professionals’ empathy for the pictorial (p = 0.001) and text-based (p = 0.001) versions of the RATE scale, with ratings of doctors’ therapeutic empathy higher in both cases. We found no statistically significant difference in patients’ ratings of therapeutic empathy by gender on the pictorial (p = 0.625) and text-based (p = 0.698) RATE scales.
Discussion
Summary of findings
We developed and validated the first universal single-item measure of therapeutic empathy, which we have called the RATE scale. Both pictorial and text-based patient-reported versions demonstrated strong convergent validity with the CARE measure, negligible correlations with an unrelated construct, and sensitivity to known group differences. These findings suggest that a carefully designed single-item measure can provide a brief yet robust assessment of therapeutic empathy suitable for routine use in healthcare settings.
Comparison with other evidence
Patients from minority ethnic groups rated practitioner therapeutic empathy lower than White patients, consistent with the findings of previous research [88] and broader evidence of inequities in empathic healthcare [87]. Although statistical significance differed between the pictorial and text-based versions, the consistent direction of effects across both versions of the scale supports known-groups validity [33].
Also aligning with previous research [88], we did not find any differences in therapeutic empathy ratings between patients of different genders. This finding supports the gender invariance of the measure and indicates that it is unlikely to be biased by gender-related expectations. While previous research [91] shows that patients assign lower therapeutic empathy scores to doctors than other healthcare practitioners, in our study doctors were rated more highly than other practitioners for therapeutic empathy. It is possible that this finding reflects a sampling imbalance, as doctors were over-represented relative to other healthcare professionals. Equally, it is possible that this demographic difference has modified over recent years, given the growing recognition of the value of therapeutic empathy in healthcare [4,5,7].
Many existing therapeutic empathy measures overemphasise particular components of the concept or diverge from their underpinning definitions, like the JSE [22] which emphasises cognitive aspects of therapeutic empathy at the expense of affective or action-based aspects [15,22]. In contrast, the RATE scale captures the full breadth of therapeutic empathy through its grounding in the concepts core components [10,11]. Combining these components into a single descriptive item may offer several psychometric advantages over multi-item scales, including reduced criterion contamination, higher face validity, superior completion rates, lower response burden, while demonstrating convergent validity with multi-item measures [40–44,47]. Indeed, the high convergent validity of both pictorial and text-based scales with the CARE measure [25] in our study reinforces the growing consensus that one well-crafted item can be both psychometrically sound and conceptually coherent [41–44]. These arguments are strengthened by research showing that patients’ overall perceptions of empathic care better predict positive outcomes than the relative weighting of its individual components [92,93]. Finally, our finding that a simple scale could demonstrate validity and convergence with more complex measures also dovetails with Gigerenzer’s findings that “fast and frugal” methods can be as effective (or more effective) than complex instruments [48].
Strengths and limitations
Key strengths of this study include adherence to best-practice methodological guidance and sustained stakeholder involvement to support content and face validity [33]. Moreover, psychometric testing in a large international and demographically diverse patient sample offers broader generalisability than many previous empathy scale validation studies which are often limited to patients from single healthcare organisations [25,94]. Indeed, established guidance suggests that such results sufficiently meet, and even exceed, the criteria for validating a new measurement tool, with evidence of content, face, and various forms of construct validity [33,95].
Our study also has limitations. It was not possible to assess the test-retest reliability of our scale due to the dynamic and context-dependent nature of therapeutic empathy [33,41,51,81]. Relatedly, validation was limited to the patient-reported version; further testing is required to confirm the “universal” nature of the scale in practitioners, students, and observers. Finally, despite our international patient sample, Europe and Africa were over-represented relative to other continents, potentially limiting our ability to capture the cultural nuances influencing empathy ratings. This is further limited by the fact that we administered the scale in English to an international sample without collecting data about participants’ first languages.
Implications for research and practice
Given its brevity and validity, the RATE scale has strong potential for routine use in clinical, educational, and research settings. In practice, the scale offers a rapid, low-burden method for routine monitoring of patient experience and subsequently informing service-level quality improvement. In education, including continuing professional development, the RATE scale could support formative assessment of, and reflection on, communication skills. This could be achieved by averaging several ratings on our scale, similar to methods deployed when using the CARE measure for appraisal or revalidation [25].
In addition to validating practitioner-, student-, and observer-reported versions of the RATE scale in diverse healthcare contexts, further research is needed to continue testing the patient-reported version in different samples (for example, patients with learning disabilities), languages, and countries. Additionally, future research can use this scale for predictive validity to check whether scores measured on this scale are associated with, or can predict, patient and staff outcomes. Finally, this study conceptualises therapeutic empathy as an attribute of the practitioner, as reflected in patient-reported evaluations. However, it is possible that patients also influence how therapeutic empathy is expressed and perceived in clinical encounters. Further research is therefore warranted to examine the role of the patient in shaping therapeutic empathy expression, and in turn, measurement.
Conclusion
We developed a universal single-item measure of therapeutic empathy – the RATE scale – both in pictorial and text-based forms. Patient-reported versions demonstrated strong validity, sensitivity to known group differences, and high acceptability. In increasingly time-pressured healthcare environments, this brief, conceptually grounded, and accessible measure offers a pragmatic approach to therapeutic empathy assessment. With further validation, it can support routine therapeutic empathy measurement, strengthen feedback loops in clinical practice and education, and contribute to international efforts to enhance empathic healthcare and improve patient and practitioner outcomes.
Supporting information
S1 Appendix. Questions used to screen digital exclusion and illiteracy based on previous research.
https://doi.org/10.1371/journal.pone.0346994.s001
(PDF)
S2 Appendix. Table describing the components of therapeutic empathy.
https://doi.org/10.1371/journal.pone.0346994.s002
(PDF)
S3 Appendix. Free-text questions used to elicit expert panel feedback for the assessment of content validity.
https://doi.org/10.1371/journal.pone.0346994.s003
(PDF)
S4 Appendix. Topic guide for semi-structured cognitive interviews.
https://doi.org/10.1371/journal.pone.0346994.s004
(PDF)
S5 Appendix. Question to assess the construct of clinical neutrality.
https://doi.org/10.1371/journal.pone.0346994.s005
(PDF)
S6 Appendix. Initial version of the universal single-item RATE scale to measure therapeutic empathy.
https://doi.org/10.1371/journal.pone.0346994.s006
(PDF)
S7 Appendix. Results of readability tests for initial versions of the RATE scale.
https://doi.org/10.1371/journal.pone.0346994.s007
(PDF)
S8 Appendix. Participant characteristics: Assessment of content validity.
https://doi.org/10.1371/journal.pone.0346994.s008
(PDF)
S9 Appendix. Participant feedback and modifications made to the RATE scales at each stage of development.
https://doi.org/10.1371/journal.pone.0346994.s009
(PDF)
S10 Appendix. Participant characteristics: Pre-testing the RATE scale in cognitive interviews.
https://doi.org/10.1371/journal.pone.0346994.s010
(PDF)
S11 Appendix. Final versions of the universal single-item RATE scales.
https://doi.org/10.1371/journal.pone.0346994.s011
(PDF)
Acknowledgments
The authors would like to acknowledge the participants, without whom the study would not have been possible.
References
- 1. Howick J, Bizzari V, Dambha-Miller H, Oxford Empathy Programme. Therapeutic empathy: what it is and what it isn’t. J R Soc Med. 2018;111(7):233–6. pmid:29978750
- 2. Winter R, Issa E, Roberts N, Norman RI, Howick J. Assessing the effect of empathy-enhancing interventions in health education and training: a systematic review of randomised controlled trials. BMJ Open. 2020;10(9):e036471. pmid:32978187
- 3.
General Medical Council. Promoting excellence: Standards for medical education and training. London: General Medical Council; 2015.
- 4. Keshtkar L, Madigan CD, Ward A, Ahmed S, Tanna V, Rahman I, et al. The effect of practitioner empathy on patient satisfaction : a systematic review of randomized trials. Ann Intern Med. 2024;177(2):196–209. pmid:38285985
- 5. Keshtkar L, Bennett-Weston A, Khan AS, Mohan S, Jones M, Nockels K, et al. Impacts of communication type and quality on patient safety incidents : a systematic review. Ann Intern Med. 2025;178(5):687–700. pmid:40228297
- 6. Howick J, Moscrop A, Mebius A, Fanshawe TR, Lewith G, Bishop FL, et al. Effects of empathic and positive communication in healthcare consultations: a systematic review and meta-analysis. J R Soc Med. 2018;111(7):240–52. pmid:29672201
- 7. Wilkinson H, Whittington R, Perry L, Eames C. Examining the relationship between burnout and empathy in healthcare professionals: a systematic review. Burn Res. 2017;6:18–29. pmid:28868237
- 8. Lamiani G, Dordoni P, Vegni E, Barajon I. Caring for critically ill patients: clinicians’ empathy promotes job satisfaction and does not predict moral distress. Front Psychol. 2020;10:2902. pmid:31969851
- 9. Fernandez AV, Zahavi D. Basic empathy: developing the concept of empathy from the ground up. Int J Nurs Stud. 2020;110:103695. pmid:32736251
- 10. Howick J, Bennett-Weston A. An empathy definition at last: exposing the narcissism of small differences. Br J Gen Pract. 2025;75(751):74. pmid:39890110
- 11. Howick J, Bennett-Weston A, Dudko M, Eva K. Uncovering the components of therapeutic empathy through thematic analysis of existing definitions. Patient Educ Couns. 2025;131:108596. pmid:39657390
- 12. Cuff BM, Brown SJ, Taylor L, Howat DJ. Empathy: a review of the concept. Emot Rev. 2016;8(2):144–53.
- 13. Hong H, Han A. A systematic review on empathy measurement tools for care professionals. Educ Gerontol. 2020;46(2):72–83.
- 14. Vieten C, Rubanovich CK, Khatib L, Sprengel M, Tanega C, Polizzi C, et al. Measures of empathy and compassion: a scoping review. PLoS One. 2024;19(1):e0297099. pmid:38241358
- 15. Sulzer SH, Feinstein NW, Wendland CL. Assessing empathy development in medical education: a systematic review. Med Educ. 2016;50(3):300–10. pmid:26896015
- 16. Barrett-Lennard GT. Dimensions of therapist response as causal factors in therapeutic change. Psychol Monogr. 1962;76(43):1.
- 17. La Monica EL. Construct validity of an empathy instrument. Res Nurs Health. 1981;4(4):389–400. pmid:6916383
- 18. Baron-Cohen S, Wheelwright S. The empathy quotient: an investigation of adults with Asperger syndrome or high functioning autism, and normal sex differences. J Autism Dev Disord. 2004;34(2):163–75. pmid:15162935
- 19. Hogan R. Development of an empathy scale. J Consult Clin Psychol. 1969;33(3):307–16. pmid:4389335
- 20. Christison GW, Haviland MG, Riggs ML. The medical condition regard scale: measuring reactions to diagnoses. Acad Med. 2002;77(3):257–62. pmid:11891166
- 21. Spreng RN, McKinnon MC, Mar RA, Levine B. The Toronto Empathy Questionnaire: Scale development and initial validation of a factor-analytic solution to multiple empathy measures. J Pers Assess. 2009;91(1):62–71.
- 22. Hojat M, Gonnella JS, Nasca TJ, Mangione S, Vergare M, Magee M. Physician empathy: definition, components, measurement, and relationship to gender and specialty. Am J Psychiatry. 2002;159(9):1563–9. pmid:12202278
- 23.
Davis MH. A multidimensional approach to individual differences in empathy. Texas: University of Texas; 1980.
- 24. Hojat M, Mangione S, Nasca TJ, Cohen MJ, Gonnella JS, Erdmann JB, et al. The Jefferson Scale of physician empathy: development and preliminary psychometric data. Educ Psychol Meas. 2001;61(2):349–65.
- 25. Mercer SW, Maxwell M, Heaney D, Watt GC. The consultation and relational empathy (CARE) measure: development and preliminary validation and reliability of an empathy-based consultation process measure. Fam Pract. 2004;21(6):699–705. pmid:15528286
- 26.
Howick J, Bennett-Weston A. Empathy-Based Healthcare Education: How to Teach Empathy. University of Leicester; 2025.
- 27. Stosic MD, Fultz AA, Brown JA, Bernieri FJ. What is your empathy scale not measuring? The convergent, discriminant, and predictive validity of five empathy scales. J Soc Psychol. 2022;162(1):7–25. pmid:34726126
- 28. Hermans L, Olde Hartman T, Dielissen PW. Differences between GP perception of delivered empathy and patient-perceived empathy: a cross-sectional study in primary care. Br J Gen Pract. 2018;68(674):e621–6. pmid:30012809
- 29. Bernardo MO, Cecílio-Fernandes D, Costa P, Quince TA, Costa MJ, Carvalho-Filho MA. Physicians’ self-assessed empathy levels do not correlate with patients’ assessments. PLoS One. 2018;13(5):e0198488. pmid:29852021
- 30. Berg K, Blatt B, Lopreiato J, Jung J, Schaeffer A, Heil D, et al. Standardized patient assessment of medical student empathy: ethnicity and gender effects in a multi-institutional study. Acad Med. 2015;90(1):105–11. pmid:25558813
- 31. Hojat M, Louis DZ, Maxwell K, Markham F, Wender R, Gonnella JS. Patient perceptions of physician empathy, satisfaction with physician, interpersonal trust, and compliance. Int J Med Educ. 2010;1:83–7.
- 32. Rosa F, Ávila R, Alves-Nogueira AC, Pereira M, Melo C, Canavarro MC. Do Physicians CARE? Psychometric Validation of the Portuguese Self-Report Version of the CARE Measure. Eval Health Professions. 2025.
- 33. Boateng GO, Neilands TB, Frongillo EA, Melgar-Quiñonez HR, Young SL. Best practices for developing and validating scales for health, social, and behavioral research: a primer. Front Public Health. 2018;6:149. pmid:29942800
- 34.
Streiner DL, Norman GR, Cairney J. Health measurement scales: a practical guide to their development and use. Oxford: Oxford University Press; 2024.
- 35. Sullivan GM, Artino Jr AR. How to create a bad survey instrument. J Graduate Med Educ. 2017;9(4):411–5.
- 36.
Lavrakas PJ. Encyclopedia of survey research methods. London: Sage Publications; 2008.
- 37. Aiyegbusi OL, Cruz Rivera S, Roydhouse J, Kamudoni P, Alder Y, Anderson N, et al. Recommendations to address respondent burden associated with patient-reported outcome assessment. Nat Med. 2024;30(3):650–9. pmid:38424214
- 38.
Ziegler M, Kemper CJ, Kruyen P. Short scales–Five misunderstandings and ways to overcome them. Oxford: Hogrefe Publishing; 2014.
- 39. Konrath S, Meier BP, Bushman BJ. Development and validation of the Single Item Narcissism Scale (SINS). PLoS One. 2014;9(8):e103469. pmid:25093508
- 40. Fisher GG, Matthews RA, Gibbons AM. Developing and investigating the use of single-item measures in organizational research. J Occup Health Psychol. 2016;21(1):3–23. pmid:25894198
- 41. Postmes T, Haslam SA, Jans L. A single-item measure of social identification: reliability, validity, and utility. Br J Soc Psychol. 2013;52(4):597–617. pmid:23121468
- 42. Kost RG, da Rosa JC. Impact of survey length and compensation on validity, reliability, and sample characteristics for ultrashort-, short-, and long-research participant perception surveys. JCTS. 2018;2(1):31–7.
- 43. Liu M, Wronski L. Examining completion rates in web surveys via over 25,000 real-world surveys. Soc Sci Comput Rev. 2017;36(1):116–24.
- 44. Sahlqvist S, Song Y, Bull F, Adams E, Preston J, Ogilvie D, et al. Effect of questionnaire length, personalisation and reminder type on response rate to a complex postal survey: randomised controlled trial. BMC Med Res Methodol. 2011;11:62. pmid:21548947
- 45. Zimmerman M, Ruggero CJ, Chelminski I, Young D, Posternak MA, Friedman M, et al. Developing brief scales for use in clinical practice: the reliability and validity of single-item self-report measures of depression symptom severity, psychosocial impairment due to depression, and quality of life. J Clin Psychiatry. 2006;67(10):1536–41. pmid:17107244
- 46. Alrashdi MH, Quinn B, Clarke SA. Barriers and facilitators in implementing clinical practice guidelines among nurses in emergency departments and critical care units: a systematic review. J Clin Nurs. 2026;35(3):1060–78. pmid:40851207
- 47. Wanous JP, Hudy MJ. Single-Item Reliability: A Replication and Extension. Organ Res Methods. 2001;4(4):361–75.
- 48.
Gigerenzer G, Todd PM, ABC Research Group. Simple heuristics that make us smart. Oxford University Press; 2000.
- 49.
Baker C. NHS staff from overseas: statistics. London: House of Commons Library; 2023.
- 50.
Office for National Statistics. Population estimates by ethnic group and religion, England and Wales: 2019. United Kingdom: Office for National Statistics; 2021.
- 51.
Allen MS, Iliescu D, Greiff S. Single item measures in psychological science. Oxford: Hogrefe Publishing; 2022.
- 52. Ayers JW, Poliak A, Dredze M, Leas EC, Zhu Z, Kelley JB, et al. Comparing physician and artificial intelligence chatbot responses to patient questions posted to a public social media forum. JAMA Intern Med. 2023;183(6):589–96. pmid:37115527
- 53. Armbruster J, Bussmann F, Rothhaas C, Titze N, Grützner PA, Freischmidt H. “Doctor ChatGPT, Can You Help Me?” The Patient’s Perspective: Cross-Sectional Study. J Med Internet Res. 2024;26:e58831. pmid:39352738
- 54. Wan P, Huang Z, Tang W, Nie Y, Pei D, Deng S, et al. Outpatient reception via collaboration between nurses and a large language model: a randomized controlled trial. Nat Med. 2024;30(10):2878–85. pmid:39009780
- 55. Chen D, Parsa R, Hope A, Hannon B, Mak E, Eng L, et al. Physician and artificial intelligence Chatbot responses to cancer questions from social media. JAMA Oncol. 2024;10(7):956–60. pmid:38753317
- 56. Streiner DL, Kottner J. Recommendations for reporting the results of studies of instrument and scale development and testing. J Adv Nurs. 2014;70(9):1970–9. pmid:24684713
- 57.
National Institute for Health Research. Payment guidance for researchers and professionals involving people in research. London: NIHR; 2025.
- 58. Palinkas LA, Horwitz SM, Green CA, Wisdom JP, Duan N, Hoagwood K. Purposeful sampling for qualitative data collection and analysis in mixed method implementation research. Adm Policy Ment Health. 2015;42(5):533–44. pmid:24193818
- 59. Lorié Á, Reinero DA, Phillips M, Zhang L, Riess H. Culture and nonverbal expressions of empathy in clinical settings: a systematic review. Patient Educ Couns. 2017;100(3):411–24. pmid:27693082
- 60. Caley MJ, O’Leary RA, Fisher R, Low-Choy S, Johnson S, Mengersen K. What is an expert? A systems perspective on expertise. Ecol Evol. 2014;4(3):231–42. pmid:24558579
- 61. Hennink MM, Kaiser BN, Marconi VC. Code saturation versus meaning saturation: how many interviews are enough? Qual Health Res. 2017;27(4):591–608. pmid:27670770
- 62. National Institute for Health Research. People in Research. Available from: https://www.peopleinresearch.org/
- 63. Prolific. Available from: https://www.prolific.com/
- 64. Lu X, Yao Y, Jin Y. Digital exclusion and functional dependence in older people: Findings from five longitudinal cohort studies. EClinicalMedicine. 2022;54:101708. pmid:36353265
- 65. Rhoades H, Wenzel S, Rice E, Winetrobe H, Henwood B. No digital divide? Technology use among homeless adults. J Soc Distress Homeless. 2017;26(1):73–7. pmid:31097900
- 66. National Health Service. Hockley Farm Medical Practice: Practice Information. 2024. Available from: https://www.hockleyfarmmedicalpractice.co.uk/patient-info/practice-information/
- 67. Nelson LA, Pennings JS, Sommer EC, Popescu F, Barkin SL. A 3-item measure of digital health care literacy: development and validation study. JMIR Form Res. 2022;6(4):e36043. pmid:35486413
- 68.
Tudor S. Digital exclusion in the UK: communications and digital committee report. London: UK Parliament; 2024.
- 69.
Office for National Statistics. Exploring the UK’s digital divide. London: Office for National Statistics; 2019.
- 70. Sanders JJ, Dubey M, Hall JA, Catzen HZ, Blanch-Hartigan D, Schwartz R. What is empathy? Oncology patient perspectives on empathic clinician behaviors. Cancer. 2021;127(22):4258–65. pmid:34351620
- 71. Hall JA, Schwartz R, Duong F, Niu Y, Dubey M, DeSteno D, et al. What is clinical empathy? Perspectives of community members, university students, cancer patients, and physicians. Patient Educ Couns. 2021;104(5):1237–45. pmid:33234440
- 72. Bennett-Weston A, Howick J. Patient and practitioner perspectives on the definition and measurement of therapeutic empathy: qualitative study. J Particip Med. 2025;17:e71610. pmid:40744521
- 73.
Organisation for Economic Co-operation and Development. Survey of Adult Skills 2023 Technical Report. Paris: OECD; 2023.
- 74. Kouame JB. Using readability tests to improve the accuracy of evaluation documents intended for low-literate participants. J Miltidiscip Eval. 2010;6(14):132–9.
- 75. Murphy J, Mercer SW, Duncan EA. A pilot study to explore the feasibility, validity and reliability of a visual version of the CARE Measure. Int J Ther Rehabil. 2013;20(9):460–5.
- 76. Höhne JK, Krebs D, Kühnel S-M. Measurement properties of completely and end labeled unipolar and bipolar scales in Likert-type questions on income (in)equality. Soc Sci Res. 2021;97:102544. pmid:34045005
- 77. Beatty PC, Willis GB. Research synthesis: the practice of cognitive interviewing. Public Opin Q. 2007;71(2):287–311.
- 78. Microsoft. Microsoft Teams. 2026. https://teams.microsoft.com
- 79. Braun V, Clarke V. Using thematic analysis in psychology. Qual Res Psychol. 2006;3(2):77–101.
- 80. Collins D. Pretesting survey instruments: an overview of cognitive methods. Qual Life Res. 2003;12(3):229–38. pmid:12769135
- 81. Burns RA, Crisp DA. The development of single-item measures of flourishing: the brief scales of flourishing (BS-F). J Posit Psychol. 2025:1–16.
- 82. White CL, Khunti K, Gillies C, Meissner K, Palipana D, Nockels K, et al. How do patient and practitioner characteristics influence empathy in healthcare? Protocol for a systematic review and meta-analysis. BMJ Open. 2025;15(2):e096269. pmid:40000084
- 83. Howick J, de Zulueta P, Gray M. Beyond empathy training for practitioners: cultivating empathic healthcare systems and leadership. J Eval Clin Pract. 2024;30(4):548–58. pmid:38436621
- 84.
Raykov T, Marcoulides GA. Introduction to psychometric theory. London: Routledge; 2011.
- 85. Mukaka MM. Statistics corner: A guide to appropriate use of correlation coefficient in medical research. Malawi Med J. 2012;24(3):69–71. pmid:23638278
- 86. Abma IL, Rovers M, van der Wees PJ. Appraising convergent validity of patient-reported outcome measures in systematic reviews: constructing hypotheses and interpreting outcomes. BMC Res Notes. 2016;9:226. pmid:27094345
- 87.
The All-Party Parliamentary Group on Birth Trauma. Listen to Mums: Ending the Postcode Lottery on Perinatal Care. London. 2024.
- 88. Roberts BW, Puri NK, Trzeciak CJ, Mazzarelli AJ, Trzeciak S. Socioeconomic, racial and ethnic differences in patient experience of clinician empathy: Results of a systematic review and meta-analysis. PLoS One. 2021;16(3):e0247259. pmid:33657153
- 89. Langley GB, Sheppeard H. The visual analogue scale: its use in pain measurement. Rheumatol Int. 1985;5(4):145–8. pmid:4048757
- 90. Li L, Wu S, Wang J, Wang C, Zuo W, Yu L, et al. Development of the emoji faces pain scale and its validation on mobile devices in adult surgery patients: longitudinal observational study. J Med Internet Res. 2023;25:e41189. pmid:37067854
- 91. Howick J, Steinkopf L, Ulyte A, Roberts N, Meissner K. How empathic is your healthcare practitioner? A systematic review and meta-analysis of patient surveys. BMC Med Educ. 2017;17(1):136. pmid:28823250
- 92. Mercer SW, Reynolds WJ. Empathy and quality of care. Br J Gen Pract. 2002;52 Suppl(Suppl):S9–12. pmid:12389763
- 93.
Reynolds W. The measurement and development of empathy in nursing. London: Routledge; 2017.
- 94. Kane GC, Gotto JL, Mangione S, West S, Hojat M. Jefferson Scale of Patient’s Perceptions of Physician Empathy: preliminary psychometric data. Croat Med J. 2007;48(1):81–6. pmid:17309143
- 95.
Frongillo EA, Nanama S, Wolfe WS. Technical guide to developing a direct, experience-based measurement tool for household food insecurity. Washington, DC: Food and Nutrition Technical Assistance, Academy for Educational Development; 2004.