Communicating empathy sympathy and compassion in healthcare consultations: a systematic review of applied linguistic research

Abstract

Background

Communicating with empathy is an essential skill for healthcare professionals (HCPs). However, current communication training curricula fail to show how empathy communication is done, hindering the effectiveness of this training. This systematic review synthesizes applied linguistic research that shows how empathy is communicated, and what functions empathy communication serves.

Method

We included research that analysed naturalistic recorded interactions between HCPs and patients in all healthcare settings. We systematically searched five databases for studies published in English up to August 2024. We analysed results and discussion sections of included studies using aggregative thematic synthesis.

Results

We identified 1112 studies through systematic searches and included 43 studies in our review. We found eight categories of empathy communication: 1) naming emotions or characterising experiences, 2) sharing similar emotions or experiences, 3) compassionate witnessing and minimal response, 4) claiming or displaying understanding of patients’ experiences and emotions, 5) displaying empathy multimodally, 6) response cries, 7) claiming of similar emotions or experiences, and 8) others. Regarding functions of empathy communication, we found four groups: 1) facilitating advice-giving or problem-solving, 2) assisting patients to work with emotions, 3) facilitating patient-centred care, 4) facilitating history-taking activities.

Conclusion

The reviewed studies provide evidence that empathy communication can be seamlessly incorporated into healthcare consultations. Empathy enhances patient engagement, aids emotional expression, supports problem-solving and helps in situations where patients’ and HCPs’ views diverge, making it an essential component of healthcare.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12913-026-14609-9.

Introduction

There is a strong consensus in the literature on the vital role of empathy, sympathy, and compassion in healthcare. Healthcare professionals’ (HCPs) displays of empathy are associated with higher patient satisfaction [1, 2], better adherence to medical advice [3], and improved clinical outcomes [1, 2, 4–6]. Compassion may be associated with improved patient well-being, better health outcomes [7, 8]. Likewise, sympathy is also believed to play important roles in healthcare [9–11].

However, communicating with empathy, sympathy, and compassion can be challenging for HCPs [12–16]. Current training curricula often draw on theoretical discussions of empathy, sympathy and compassion without showing real-life examples of these practices [17]. The past three decades have witnessed the emergence of a growing body of applied linguistic research1 that shows what empathy, sympathy, and compassion look like in practice; however, the impact of this research in healthcare remains limited. This is largely because these studies are often published in applied linguistic journals, using specialized terminology, thus hindering access from healthcare practitioners and researchers. To assist HCPs in interpreting the findings, developing training materials, and translating research findings into practice, we conducted a systematic review synthesising applied linguistic research, which investigates how HCPs communicate with empathy, sympathy, and compassion in naturalistic interactions with patients.

Empathy, sympathy and compassion are interrelated concepts and the distinction between them is debated in the literature. Therefore, this review spans all these terms instead of focusing on one. Exemplifying the interrelation between these concepts, empathy has been often used interchangeably with compassion in the literature [18], though originally it was conceptualized very similarly to sympathy [19, 20]. E Von Dietze and A Orb [21] attempt to distinguish between sympathy, empathy and compassion. They view compassion as a virtue that involves “deliberate altruistic participation in another person’s suffering” (p.168), sympathy as the emotion that arises in someone when they observe another’s suffering, and empathy as an active response that involves not just emotion but also a deeper understanding of the suffering witnessed [21]. This view highlights the affective and cognitive aspects of empathy, the affective characteristic of sympathy, and the ethical feature of compassion. Instead of treating these three terms as separate concepts, D Jeffrey [18] reconceptualized empathy to encompass both sympathy and compassion. According to Jeffery (2016), empathy comprises cognitive, affective, behavioral, and moral dimensions. Lains (2024), however, views compassion as a broader construct that encompasses empathy: in their framework, empathy involves emotionally resonating with another person, while compassion extends this by adding the motivation and readiness to alleviate the person’s distress [22]. In this review, we align with Jeffery’s (2016) position and treat empathy as the overarching concept, as it includes observable behavioral components, whereas compassion relies on underlying motivational states that cannot be observed in interactional data used in applied linguistic research.

This review synthezises evidence from applied linguistic research to answer:

• How is empathy communicated in healthcare?

• What functions does empathy communication serve?

Collating the functions of empathy will provide further insight into the nuances of empathy communication. Various approaches in applied linguistics, take the view that language doe s not simply represent inner states or objective truths, but functions as a tool for performing social actions and accomplishing social purposes [23]. For example, when someone responds to a picnic invitation by saying, “that sounds like a lovely idea, but I’m afraid I’m not available,” the utterance is not primarily a report of whether they truly find the idea lovely. Rather, it performs the social work of softening a refusal, preserving rapport, and mitigating the potential face-threat of declining an invitation. This perspective indicates that when a participant deploys a language practice to show empathy, they can also achieve other social actions (cf. double-barrelled turn [24] and composite social actions [25]). Understanding how empathy is enacted in real interactions, and the effects it has on the unfolding of clinical talk, offers valuable insights for strengthening empathy training. By identifying and synthesizing the specific communication practices through which empathy is accomplished, this review provides evidence that can inform the design of training curricula, particularly by highlighting the observable, teachable behaviours that support empathy communication in healthcare settings [17].

Methods for the systematic review

This study employed a systematic review design to synthesise applied linguistic research on how healthcare professionals communicate empathy in naturalistic clinical interactions. Our approach followed established guidelines for qualitative evidence synthesis and involved three stages: (1) identifying and selecting empirical studies that examined empathic communication through discourse-analytic methods; (2) extracting and organising reported interactional practices and their functions; and (3) developing descriptive themes through an aggregative thematic synthesis. This design allowed us to integrate findings across diverse studies while remaining closely aligned with the analytic frameworks used in the primary research.

We adopted the methods and procedures described by Parry and Land (2013) for systematically reviewing conversation analysis and discourse analysis research in healthcare. Generally speaking, discourse analysis and conversation analysis are similar in its focus on how language is used in real contexts to construct meaning, enact social actions, and shape relationships within interaction. The major difference is that conversation analysis is data-driven while discourse analysis draws on theories to account for communication behaviors. This guide outlines eight stages and provides specific guidance and justification for adapting traditional approaches to systematic reviews in clinical research utilising conversation analysis and discourse analysis [26]. Our review protocol was registered in Prospero (ID: CRD42023450902).

Inclusion and exclusion criteria

We included studies which met the following criteria: (a) included analyses of naturalistic audio-recording or video-recording of real life communication between patients and HCPs in all healthcare settings; (b) utilized applied linguistic approaches such conversation analysis, conversation analysis and ethnomethodology (EMCA), discourse analysis, interactional sociolinguistics; and (c) focused on empathy, sympathy and/or compassion communication practices displayed by HCPs. No date limit was applied to eligible studies.

We excluded studies reporting on interactions involving medical students, or simulated patients; studies using content analysis to investigate interaction; dissertations or conference papers, and not-peer reviewed grey literature; studies written in languages other than English. We did, however, include studies of communication in other languages, which were reported in English.

Search strategy

We searched five electronic databases: Medline (Ovid), Embase (Ovid), PsycINFO (Ovid), and ProQuest Linguistics & Language Behavior Abstracts (LLBA). Each database was searched using synonyms for a) empathy, compassion and sympathy; b) applied linguistic approaches; and c) healthcare, before combining the three concepts. In addition, we also searched the bibliographic database for research in Ethnomethodology and Conversation Analysis (EMCA Wiki), which Parry and Land (2013) found to be particularly suitable for identifying relevant papers. The initial database search was conducted in September 2023 and updated in August 2024. Records of Embase and ProQuest LLBA searches are provided in Supplementary Appendix 1.

Data extraction

Screening and data extraction were conducted using Covidence. Two reviewers independently performed each stage of the review: screening, eligibility, and inclusion. Each paper was compared to the pre-defined inclusion criteria. Conflicts were resolved by a third reviewer. Data were extracted independently by three reviewers (CA, RD, VC), and checked by BT. All reviewers have PhD qualifications in Conversation Analysis and Applied Linguistics. In addition to extracting findings regarding empathy, sympathy and/or compassion communication practices, we also extracted other relevant information: i.e., research aim, language, country, healthcare setting and whether empathy, sympathy or compassion was used as the key term in each study.

Quality assessment: data breadth and analysis depth

According to Parry and Land (2013), two broad dimensions must be considered when assessing the value and contribution of a conversation analysis or discourse analysis study: the type and amount of data, and the detail and depth of analysis. An in-depth analysis drawing on small amounts of data offers nuanced understanding of the focal communication practice, while a study drawing on large amounts of data contributes evidence that the focal communication practice is widespread in its use. Therefore, we assess both the breadth and the depth of each study. Regarding data breadth, we indexed the number of sites included in the analysis, the numbers of HCPs and patients, and the number of extracts analysed. We assessed the depth of analysis using the check list adapted from Parry and Land (2013) (see “Quality assessment: Data breadth and analysis depth”).

Methods for the thematic analysis

The data synthesis process utiliz ed an aggregative thematic synthesis method, as outlined by J Thomas and A Harden [27], which aligns with current practices for synthesizing conversation analysis and discourse analysis research in healthcare [26]. We chose an aggregative thematic synthesis approach because our aim was to integrate and summarize reported conversational practices across diverse Conversation Analysis studies, without generating new theoretical constructs. Other qualitative synthesis methods, such as interpretive or integrative methods aim to construct new interpretations, generate new theories, or generally take an interpretive focus [28]. This might be appropriate for data which has initially be analysed (in the primary sources) using thematic analysis, which is grounded in interpretivism, However, linguistic data is empirically grounded, where phenomena are identified and described rather than interpreted. Therefore, synthesis approaches grounded in interpretivism are not appropriate for the data in our review. Aggregative thematic synthesis aims to aggregate only and so is (a) an appropriate method to sue on empirically grounded linguistic data (b) an established method for synthesising data of this type in similar reviews (c) aligns with our review’s descriptive and practice-focused objectives.

This approach involved a thorough examination of the results and discussion sections of all selected studies, followed by line-by-line coding and the development of descriptive themes through inductive reasoning. Similar themes from various studies were then organized using the “one sheet of paper” (OSOP) technique [29]. This technique involves listing all identified issues on a single sheet of paper to visually map connections, cluster related ideas, and identify overarching themes across studies. This technique helped to condense and group conversational practices into high-level descriptive themes. Coding was completed by the first author (BT). A 10% sample of papers were then independently coded by the second author (RD) and discrepancies discussed. Initial themes were discussed with the entire team, and themes and subthemes were revised, through iterative discussions.

To identify how empathy is communicated in healthcare (research question 1), we began by noting all empathy communication practices reported in each study. We then identified similarities and differences in their linguistic formulations and the sequential contexts in which they occurred. Finally, we grouped practices that shared similar features and interactional contexts into higher-level themes, as presented in the One Sheet of Paper (Supplementary Appendix 1).

To identify the functions that empathy communication serves (research question 2), we began by examining the functions of empathy communication practices described in each study, noting both their local interactional contexts and broader healthcare settings. We then identified similarities and differences across studies and grouped these functions into four overarching categories.

Results

Findings of the systematic review

Search results

Forty-three studies were included for this review (Fig. 1).

Fig. 1.

PRISMA chart

Characteristics of included studies

Table 1 documents, for each included study, the research aim, the terminology used (i.e., empathy, sympathy, and compassion), the research method, the language used in the healthcare consultations, and the healthcare settings. The majority of studies used conversation analysis either exclusively (n.36) or in combination with another method (n.3) [34, 62, 71]. Discourse analysis was used in three studies [37, 47, 61] and one study did not explicitly specify its linguistic research method [58]. Two studies focused on sympathy and empathy [43, 54], and one study on compassion [32]. The remaining 43 studies focused on empathy exclusively. Regarding languages, most of the studies involved interactions in English (n.25). Other languages include Finnish (n.5) [54, 55, 66, 72, 73], German (n.4) [34–36, 39], Dutch (n.3) [56, 57, 62], Norwegian (n.2), and Swedish, Danish [38], Arabic [30] and Chinese [68] were represented in one study each. Study settings included psychotherapy (n.15), primary care (n.11) (including 7 general practice, 2 nursing, 1 immunisation clinic and 1 geriatrics clinic), counselling and helplines (n.6), oncology (n.3) [32, 49, 71], palliative care (n.3) [42–44], psychiatry (n.2) [38, 59], and audiology (n.1) [41].

Table 1.

Included study characteristics

Study ID	Study Aim	Terminology used	Research Method	Country & Language	Healthcare Setting
RAM Abu-Elrob [30]	To explore empathy sequences and its role in medical encounters	Empathy	Conversation analysis	Jordan Arabic	Not specified
RK Barnes, M Jepson, C Thomas, S Jackson, C Metcalfe, D Kessler and H Cramer [31]	To evaluate the implementation adherence to a talk-based intervention using CA methods in the context of a UK feasibility trial where General Practitioners were trained to apply the Background, Affect, Trouble, Handling, Empathy (BATHE) technique to screen for psychosocial issues during consultations patients.	Empathy	Conversation analysis	UK English	General Practice
WA Beach [32]	To explore how “woundedness” is expressed and responded to with empathy and compassionate acknowledgment.	Compassion	Conversation analysis	USA English	Oncology (comprehensive cancer centre)
T Bradshaw, A Le Couteur and M Oxlad [33]	To examine how psychologists respond to clients’ accounts of suicide attempts.	Empathy	Conversation analysis	Australia English	Psychotherapy
MB Buchholz [34]	To explore enactments of empathy in psychotherapy interactions.	Empathy	Conversation analysis mixed with cognitive linguistics	Germany German	Psychotherapy
MB Buchholz, J Bergmann, M-L Alder, MM Dittmann, F Dreyer and H Kächele [35]	To examine how psychoanalytical, depth-psychological, and behavioural therapy psychotherapists manage the challenges of displaying empathy at different stages of consultations.	Empathy	Conversation analysis	Germany German	Psychotherapy
MB Buchholz and H Kachele [36]	To examine how empathy is enabled in psychoanalytic, psychodynamic, and cognitive-behavioural therapies.	Empathy	Conversation analysis	Germany German	Psychotherapy
T Crawford and J Leask [37]	To investigate the basis of a therapeutic relationship, focusing on the expression of empathy and efforts to establish rapport during consultations between immunization specialists and vaccine-hesitant parents.	Empathy	Discourse analysis; interactional sociolinguistic	Australia English	Primary care (Immunisation Clinics)
AS Davidsen and CF Fosgerau [38]	To examine how general practitioners and psychiatrists respond to emotional disclosures during consultations with patients suffering from depression.	Empathy	Conversation analysis	Denmark Danish	General Practice & Psychiatry
A Deppermann and T Spranz-Fogasy [39]	To examine how doctors utilize various types of questions to demonstrate their understanding of the patient’s prior responses, and how different syntactic structures of questions serve to show empathy.	Empathy	Conversation analysis	Germany German	Unspecified
DW Easter and WA Beach [40]	To develop quantitative and empirical tools for measuring “missed” and “met” empathic opportunities.	Empathy	Conversation analysis; mixed with non-interactional methods	USA English	Oncology
K Ekberg, C Grenness and L Hickson [41]	To explore how audiologists address patients’ concerns about hearing aids.	Empathy	Conversation analysis	Australia English	Audiology
J Ford [42]	To examine how doctors’ expressions of empathy work to maintain the focus on treating older patients as whole persons.	Empathy	Conversation analysis	UK English	Palliative Care
J Ford, A Hepburn and R Parry [43]	To explore how empathic statements from palliative care doctors can help bridge the gap when it becomes evident that their perspective differs from the patient’s.	Empathy	Conversation analysis	UK English	Palliative care
A Hepburn and J Ford [44]	To investigate empathy as an interactional phenomenon, rather than a purely internal psychological one, across various everyday and institutional contexts. To examine the interactional factors influencing the expression of empathy within the changing knowledge and professional landscapes of each setting.	Empathy and sympathy	Conversation analysis	UK, Australia English	Multiple setting: UK child protection helpline, palliative care, and emotion focused therapy sessions
A Hepburn and J Potter [45]	To examine how call takers respond to clients’ crying on a child protection helpline.	Empathy	Conversation analysis	UK English	Child protection helpline
J Heritage and A Lindström [46]	To examine how a community nurse responds to her client’s concerns about difficulties in bonding with her new-born daughter.	Empathy	Conversation analysis	UK English	Home visits by community nurses.
J Hilder, M Stubbe, L Macdonald, P Abels and AC Dowell [47]	To identify the discourse characteristics and patterns that illustrate effective communication practices in a high-risk antenatal clinic.	Empathy	Discourse analysis; interactional sociolinguistic	New Zealand English	High-risk Antenatal outpatient clinic
I Hutchby [48]	To examine the practice of active listening in counselling psychology, which involves the methods counsellors use to demonstrate attentiveness and responsiveness to what the child is expressing.	Empathy	Conversation analysis	UK English	Counselling (British child counselling and family therapy practice)
C Iversen [13]	To analyse suicide helpline calls to explore how call-takers navigate the challenge of reframing callers’ suicidal thoughts while validating their experiences.	Empathy	Conversation analysis	Sweden Swedish	Counselling (Swedish suicide helpline)
AV Mellblom, L Korsvold, E Ruud, HC Lie, JH Loge and A Finset [49]	To investigate communication practices related to emotional concerns during follow-up consultations between adolescent cancer survivors and their paediatricians	Empathy	Conversation analysis	Norway Norwegian	Oncology (paediatric department)
P Muntigl [50]	To illustrate how clients with depression narrate their difficulties and how psychotherapists, in an emotion-focused therapeutic setting, can empathically engage with these issues.	Empathy	Conversation analysis	Canada English	Psychotherapy
P Muntigl, NK Knight and L Angus [51]	To investigate how therapists highlight a client’s verbal or nonverbal expressions of emotional stance, thereby directing the discussion towards the client’s present experience.	Empathy	Conversation analysis	Canada English	Psychotherapy
P Muntigl and AO Horvath a [52]	To examine how therapists and clients co-manage relational disaffiliation in emotion-focused therapy sessions.	Empathy	Conversation analysis	Canada English	Psychotherapy
P Muntigl and AO Horvath b [67]	To analyse storytelling episodes where clients do not reveal the emotional effects of their narrative and examine the interactional strategies therapists use to topicali ze the emotions the client might have experienced.	Empathy	Conversation analysis	Canada English	Psychotherapy
HT Nguyen [53]	To examine how the prime questions, in both their original and modified forms are addressed by pharmacists, and what interactional functions these questions may have in pharmacy consultations with patients.	Empathy	Conversation analysis	USA English	Pharmacy
J Ruusuvuori [54]	To examine “empathy” and “sympathy” as situated practices and sequential processes that are collaboratively constructed by the participants.	Empathy and sympathy	Conversation analysis	Finland Finnish	General practice & homeopathic consultations
J Ruusuvuori [55]	To investigate how practitioners affiliate with patients’ expressed concerns, how patients respond to these affiliative interactions, and how these responses are made relevant to the medical agenda	Empathy	Conversation analysis	Finland Finnish	General practice & Homeopathy
W Stommel and J Lamerichs [56]	To examine counsellors’ empathetic responses in an online counselling service. It investigates whether and how typed empathic responses to client posts align with the opportunities and limitations of the chat communication channel.	Empathy	Conversation analysis	Netherlands Dutch	Counselling
W Stommel and Ht Molder [57]	To examine GPs’ explicit verbal expressions that show empathy in response to clients’ descriptions of or statements about their problems and related emotions.	Empathy	Conversation analysis	Netherlands Dutch	Counselling (alcohol and drugs helpline)
AL Suchman, K Markakis, HB Beckman and R Frankel [58]	To develop an empirically based model of empathic communication in medical interviews by detailing the specific behaviours and interaction patterns linked to verbal expressions of emotion.	Empathy	Applied linguistics	Unspecified English	General practice
L Thompson and R McCabe [59]	To investigate how viewing language as an action can reveal the discursive practices within psychiatry and offer a new, bottom-up perspective on what constitutes effective communication in schizophrenia.	Empathy	Conversation analysis	Unspecified English	Psychiatry
CK Tietbohl [60]	To explore interactional practices for expressing empathy, called empathic validation.	Empathy	Conversation analysis	USA English	Primary care (geriatrics clinic)
PAE Velasquez and CJ Montiel [61]	To explore how the core client-centred conditions of unconditional positive regard, congruence, and empathy are communicated bodily in psychotherapy interactions.	Empathy	Discourse analysis	USA English	Psychotherapy
W Versteeg and H te Molder [62]	To examine how coaches (ADHD coaching centre) use their own bodily sensations to reflect their clients’ internal states, a technique referred to as “my side empathy formulations.	Empathy	Conversation analysis mixed with interactional sociolinguistic	Netherlands Dutch	Psychotherapy
L Voutilainen and A Peräkylä [63]	To examine the connection between empathy and challenge in a therapist’s responses to clients’ problematic emotions	Empathy	Conversation analysis	Finland Finnish	Psychotherapy
L Voutilainen, A Peräkylä and J Ruusuvuori [64]	To explore how patients’ emotional experiences are addressed in cognitive-constructivist psychotherapy.	Empathy	Conversation analysis	Finland Finnish	Psychotherapy
A Weatherall and L Keevallik [65]	To explore how expressions of pain are constructed through speech, sounds, and physical gestures. It also looks at the sequential contexts in which these expressions happen, how clinicians react to them, and the impact these reactions have on the on-going activities.	Empathy	Conversation analysis	New Zealand English	General Practice
E Weiste [66]	To examine two types of empathy formulation sequences used by clinicians to address discussions about clients’ emotional states.	Empathy	Conversation analysis	Finland Finnish	Psychotherapy (psychiatric outpatient clinic)
Y Wu [67]	To explore the specific interactional practices therapists use to demonstrate empathy toward clients.	Empathy	Conversation analysis	USA English	Psychotherapy
Y Wu [68]	To examine how is empathy collaboratively accomplished in nursing interactions.	Empathy	Conversation analysis	Unknown country Chinese	Nursing (hospital)
R Wynn [69]	To analyse conversational sequences through which empathy was achieved, with a focus on the form, function, and categorization of these sequences in general practice consultations.	Empathy	Conversation analysis	UK English	General practice
R Wynn and M Wynn [70]	To provide insights into how empathy is expressed and its role within psychotherapy interactions.	Empathy	Conversation analysis	Norway Norwegian	Psychotherapy

Quality assessment: data breadth and analysis depth

Following Parry and Land (2013), two broad dimensions must be considered when assessing the value and contribution of each study: the type and amount of data, and the detail and depth of analysis. In terms of data type and amount, we index whether the study uses audio or video recordings, how many settings and how many participants are involved, how diverse the settings are, how many occurrences of the focal communication practice in the data set, and how big the data set is. In terms of the detail and depth of analysis, studies vary in regards to whether they focus on the level of turn construction units, a turn of talk, or sequence of turns of talk. Studies also vary in terms of the extent to which analysis is grounded in previous empirical findings, and the extent to which they describe the social function of the focal practice.

We note that using smaller amounts of data and focussing on less diverse settings does not necessarily signify lower quality. Similarly, including fewer interactional features, discussing these features in less detail, or being less grounded on previous empirical findings are not necessarily indicators of poorer quality. While an in-depth analysis drawing on small amount of data offers nuanced understanding of the focal communication practice, a study drawing on large amounts of data contributes evidence that the focal communication practice is widespread in its use. Therefore, it is important to characterise and consider many features of each study before evaluating the quality of evidence it offers.

Table 2 provides information regarding the breadth of data and the depth of analysis for each study. Note that analysis depth was assessed using the following eight-question checklist, which was adapted from Parry and Land (2013).

1. Does the analysis predominantly examine all the participants involved?

2. Does the analysis examine data in fine-grained detail?

3. Does the analysis examine more than just the topical/semantic content, i.e. does it attend to aspects of grammatical, pragmatic, and/or prosodic features of talk?

4. Does the analysis include examination of aspects of the sequential environment in which the practice(s) occur(s)?

5. Does the analysis include examination of interactional effects and consequences of the features analysed?

6. Does the analysis include examination of atypical cases?

7. Does the analysis support central analytic claims by direct references to the data extracts?

8. Does the analysis use established empirical findings as analytical tools?

Table 2–

Data breadth and analysis depth

Study ID	Data Breath					Analysis Depth
	Setting number	HP number	Patient number	Data amount	Number of extracts analysed	1.	2.	3.	4.	5.	6.	7.	8.
RAM Abu-Elrob [30]	NA	NA	NA	NA	6	Yes	Yes	Yes	Yes	No	No	No	No
RK Barnes, M Jepson, C Thomas, S Jackson, C Metcalfe, D Kessler and H Cramer [31]	1	>5	10–29	10–29 hrs	31	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes
WA Beach [32]	1	1–2	<10	<10hrs	6	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes
T Bradshaw, A Le Couteur and M Oxlad [33]	1	3–5	<10	NA	5	Yes	Yes	Yes	Yes	No	No	Yes	Yes
MB Buchholz [34]	NA	NA	NA	NA	5	Yes	Yes	No	Yes	Yes	No	Yes	No
MB Buchholz, J Bergmann, M-L Alder, MM Dittmann, F Dreyer and H Kächele [35]	>2	NA	30–50	NA	4	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes
MB Buchholz and H Kachele [36]	>2	NA	10–29	NA	11	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes
T Crawford and J Leask [37]	1	>5	10–29	10–29 hrs	2	Yes	Yes	Yes	No	No	No	No	No
AS Davidsen and CF Fosgerau [38]	2	>5	>50	>50 hrs	8	Yes	Yes	No	Yes	No	No	Yes	Yes
A Deppermann and T Spranz-Fogasy [39]	NA	3–5	10–29	<10 hrs	7	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes
DW Easter and WA Beach [40]	1	NA	NA	<10 hrs	4	Yes	Yes	Yes	Yes	No	No	No	No
K Ekberg, C Grenness and L Hickson [41]	1	>5	>50	NA	9	Yes	Yes	Yes	Yes	No	No	No	Yes
J Ford [42]	>2	NA	NA	10–29 hrs	6	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes
J Ford, A Hepburn and R Parry [43]	1	3–5	30–50	NA	6	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
A Hepburn and J Ford [44]	>2	>5	>50	<10 hrs	6	Yes	Yes	Yes	Yes	Yes	No	Yes	No
A Hepburn and J Potter [45]	1	NA	10–29	NA	13	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes
J Heritage and A Lindström [46]	1	1–2	<10	NA	7	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes
J Hilder, M Stubbe, L Macdonald, P Abels and AC Dowell [47]	1	>5	10–29	10–29 hrs	8	Yes	No	Yes	Yes	Yes	No	No	No
I Hutchby [48]	1	3–5	<10	10–29 hrs	7	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
C Iversen [13]	1	>5	>50	NA	7	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes
AV Mellblom, L Korsvold, E Ruud, HC Lie, JH Loge and A Finset [49]	1	>5	>50	>50 hrs	11	Yes	Yes	Yes	Yes	Yes	No	No	No
P Muntigl [50]	1	3–5	<10	NA	3	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes
P Muntigl and AO Horvath a [51]	1	NA	<10	10–29 hrs	5	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
P Muntigl and AO Horvath b [52]	1	3–5	<10	10–29 hrs	7	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes
P Muntigl, NK Knight and L Angus [67]	1	1–2	<10	<10 hrs	8	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes
HT Nguyen [53]	1	1–2	NA	NA	13	Yes	Yes	Yes	Yes	Yes	No	No	No
J Ruusuvuori [54]	>2	>5	>50	NA	3	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
J Ruusuvuori [55]	2	>5	>50	NA	8	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
W Stommel and J Lamerichs [56]	1	NA	NA	30–50 hrs	8	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes
W Stommel and Ht Molder [57]	1	NA	NA	<10 hrs	7	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes
AL Suchman, K Markakis, HB Beckman and R Frankel [58]	>2	>5	NA	30–50 hrs	6	Yes	Yes	No	No	Yes	No		No
L Thompson and R McCabe [59]	1	NA	30–50	NA	8	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes
CK Tietbohl [60]	1	NA	>50	NA	9	No	Yes	Yes	Yes	Yes	No	Yes	Yes
PAE Velasquez and CJ Montiel [61]	1	1–2	10–29	10–29 hrs	3	Yes	No	Yes	No	No	No	No	No
W Versteeg and H te Molder [62]	1	3–5	10–29	NA	5	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes
L Voutilainen and A Peräkylä [63]	1	1–2	<10	10–29 hrs	4	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
L Voutilainen, A Peräkylä and J Ruusuvuori [64]	1	1–2	<10	<10 hrs	8	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes
A Weatherall and L Keevallik [65]	1	>5	>50	NA	6	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes
E Weiste [66]	1	3–5	<10	10–29 50 hrs	4	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Y Wu [67]	1	>5	10–29	>50 hrs	13	Yes	Yes	No	Yes	Yes	No	No	No
Y Wu [68]	2	>5	10–29	10–29 hrs	5	Yes	Yes	No	Yes	Yes	No	Yes	No
R Wynn [69]	1	3–5	10–29	10–29 hrs	5	Yes	No	Yes	No	Yes	No	Yes	Yes
R Wynn and M Wynn [70]	1	NA	10–29	NA	8	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes

Across the 43 studies, most exhibited strong adherence to core conversation analytic principles. The overwhelming majority analysed all participants (42/43), used fine-grained analysis (41/43), incorporated grammatical/pragmatic/prosodic features (40/43), and attended to sequential organisation (39/43).

Fewer studies went beyond identifying practices to examining their interactional consequences: 32 out of 43 explicitly analysed how the features under study shaped the unfolding interaction. Only seven studies reported atypical cases, indicating that while deviant-case analysis is valued, it remains relatively uncommon across the sample. Most studies supported their claims with direct references to data extracts (37/43) and drew on established empirical findings to inform their analysis (28/43).

Findings of the thematic analysis

The findings of each included study that are relevant to the two review questions – (i) How is empathy communicated in healthcare? (ii) What function does empathy communication serve? – are summarized in Table 3. We provide a thematic synthesis of findings, below.

Table 3.

Included study findings

Study ID	How is empathy communicated in healthcare?	What function does empathy communication serve?
RAM Abu-Elrob [30]	Naming emotions or characterising experiences E.g., why are you so sad? This question encourages the patient to share the experience, accounting for the emotion. Sharing similar emotions or experiences The doctor shares a personal experience which works to reassure the patient.	Facilitating advice-giving or problem-solving Through displaying empathy, doctor demonstrate an understanding of the health issues, encourage patients to share experiences, and give advice on healthcare.
RK Barnes, M Jepson, C Thomas, S Jackson, C Metcalfe, D Kessler and H Cramer [31]	Naming emotions or characterising experiences E.g., that must be really hard	Not discussed
WA Beach [32]	Displaying empathy multimodally The doctor displays empathy and compassion through a soothing voice, gently touching the patient’s knee, offering a tissue, standing by the patient and maintaining direct eye contact during their crying and sobbing.	Facilitating advice-giving or problem-solving The doctor shows compassion and empathy in the middle of the talk where two doctors enquire about a cancer patient’s support network and suggest support services.
T Bradshaw, A Le Couteur and M Oxlad [33]	Naming emotions or characterising experiences E.g.1, that must be difficult E.g.2, so that must be quite tough, irritating, frustrating? E.g.3, must be a tricky situation for you	Assisting patients to work with their emotional experiences Empathy is a therapeutic tool to work with patient emotions
MB Buchholz [34]	Naming emotions or characterising experiences E.g., and you’re thinking now going home you did that so you wouldn’t to have to be alone Claiming or displaying understanding Empathy is displayed via extensions/co-completion, which are constructed as they were a continuing part of the client’s previous turn Displaying empathy multimodally Although this study focuses on empathy as embodied practice, it doesn’t provide evidence of how empathy is displayed bodily in the therapy interaction. Instead, it talks about embodied metaphors.	Not discussed
MB Buchholz, J Bergmann, M-L Alder, MM Dittmann, F Dreyer and H Kächele [35]	Naming emotions or characterising experiences E.g., I get the impression that you have recently really really pulled yourself together here	Not discussed
MB Buchholz and H Kachele [36]	Naming emotions or characterising experiences E.g., it is really difficult to understand that now you shall be no longer together with him ain’t you that a door fell closed when he shut of you and let you stand outside in front of it In this example, the metaphor of a door fell closed was used to depict the client’s situation of breaking up with her boyfriend.	Assisting patients to work with their emotional experiences It serves to build a common ground for the therapeutic work to proceed.
T Crawford and J Leask [37]	Naming emotions or characterising experiences E.g.1, Lots of things happening at home, isn’t it? E.g.2, Yeah. You poor thing	Not discussed
AS Davidsen and CF Fosgerau [38]	Naming emotions or characterising experiences E.g.1, then you get both angry but also really sad E.g.2, and you’re fond of him	Not discussed
A Deppermann and T Spranz-Fogasy [39]	Naming emotions or characterising experiences E.g., It doesn’t work?	Facilitating history-taking activities Empathic declarative questions serve to invite patients to elaborate on the topic and shift the focus to psychological aspects of the illness, which aligns with the history-taking interview that the doctors are doing.
DW Easter and WA Beach [40]	Claiming or displaying understanding E.g., I understand	Not discussed
K Ekberg, C Grenness and L Hickson [41]	Naming emotions or characterising experiences E.g., that’s really common Compassionate witnessing and minimal response E.g., via continuers such as yep, yep, yep	Facilitating advice-giving or problem-solving Empathy is tied to the institutional task at hand, which is addressing patients’ concerns about hearing aids.
J Ford [42]	Naming emotions or characterising experiences E.g.1, Must have been really frightening E.g.2, Now I know that’s hard Response cry E.g., Oh, gosh	Facilitating patient-centred care Empathy expressions allow doctors to appreciate older patients in end-of-life care as a as individuals rather than as a mere biomedical condition.
J Ford, A Hepburn and R Parry [43]	Naming emotions or characterising experiences E.g.1, It’s difficult isn’t it because I hear what you’re sa:ying it’s helpful to have hope and treatment might work. Someti:mes: you come to a point whe:re people know treatment won’t work, In example 1, the doctor addresses the patient’s optimistic expectations which is not realistic from a medical perspective. E.g.2, I know that’s ha:rd and I know you’ve got lots of other stuff going on as well but we do know that maintaining a little bit of activity is the thing that’s going to keep you living. In example 2, the doctor displays empathy and gives advice that addresses a patient who is sceptical of the medical perspective. E.g.3, So it feels really frightening, what’s really hard when you can’t get your breath is that you can’t help but panic. In example 3, the doctor responds to a patient who has an overtly negative view of their illness.	Facilitating advice-giving or problem-solving Empathy displays serve two-fold purposes: to acknowledge the patient’s emotion and to address the discrepancies between the patient’s perspective and a medical perspective.
A Hepburn and J Ford [44]	Naming emotions or characterising experiences E.g.1, You sound as though you’re very upset E.g.2, That’s a little bit frustrating ˚isn’t iht.˚ Compassionate witnessing and minimal response E.g., Dohn’t worry, ˚th- i-˚ tahke your ti:me. Response cry E.g., oh, gosh Note that in authors’ definition, this response is considered as sympathy rather than empathy.	Facilitating advice-giving or problem-solving In helpline context, empathy is linked to the institutional agenda of advice giving Assisting patients to work with their emotional experiences In the psychotherapy context, empathy is a tool through which therapists assist clients to get in touch with their emotions.
A Hepburn and J Potter [45]	Naming emotions or characterising experiences E.g., I can hear that you’re very very upset Compassionate witnessing and minimal response E.g., Take your time Through this construction call-takers offer “a license for the late delivery of talk and a formulation of and account for” what happened” (p.109).	Facilitating advice-giving or problem-solving Empathy is tied to advice-giving, which is the institutional task of the helpline service
J Heritage and A Lindström [46]	Sharing similar emotions or experiences E.g.1, I used to wake up °and think that dratted baby is crying again E.g.2, I think the first weeks they take a lot out of you and and nothing is	Facilitating advice-giving or problem-solving Empathy is tied to advice-giving on baby care.
J Hilder, M Stubbe, L Macdonald, P Abels and AC Dowell [47]	Compassionate witnessing and minimal response Empathy is displayed through the use of response tokens such as “yeah”, “wow”, which acknowledge the emotional impacts of the patients’ recounted experiences.	Not discussed
I Hutchby [48]	Naming emotions or characterising experiences E.g.1, That doesn’t sound like a ver[y nice thing E.g.2, So she’s good at sort of shouting E.g.3, And you’re gunna be stuck in London E.g.4, So you know it needs to be different? The author considers these formulations as practices for doing active listening, which encourage children to discuss their family experiences and emotions in the child counselling and family therapy context.	Assisting patients to work with their emotional experiences Empathy is a tool for facilitating discussions about clients’ feelings and experiences regarding family relationships.
C Iversen [13]	Naming emotions or characterising experiences E.g.1, I t still sounds like it oscillates within you because you bought flowers and then you like you had hope in some way. E.g.2, You’ve gone so far so that it feels like the only way out is it. E.g.3, It feels so heavy. Compassionate witnessing and minimal response Call-takers attentively listen to callers and display empathy through minimal response tokens such as “yes”, “yeah” and “mm”.	Facilitating advice-giving or problem-solving Empathy is linked to the institutional task of suicide prevention.
AV Mellblom, L Korsvold, E Ruud, HC Lie, JH Loge and A Finset [49]	Naming emotions or characterising experiences E.g.1, But it- yes it sound like it bothers you actually, doesn’t it? E.g.2, It is not easy Claiming or displaying understanding That I understand. Compassionate witnessing and minimal response Doctors use minimal response tokens such as “yes” and “yeah”, which are termed as minimal encouragers in the study.	Assisting patients to work with their emotional experiences Empathy is a therapeutic tool to address patients’ emotional concerns.
P Muntigl [50]	Naming emotions or characterising experiences E.g.1, you felt really. (.) ashamed of it’ E.g.2, ‘it’s like.hhh you’re dampening yourself. In the author’s words, example 1 is typical of formulating emotions, while example 2 is typical of illustrating emotions through metaphorical descriptions. Displaying empathy multimodally In situations where clients struggle to find the right words for their difficult emotions, therapists use nods, body gestures, and a soft voice to connect with and empathize with the client’s expressed emotional state	Assisting patients to work with their emotional experiences Empathy is a tool through which therapists assist clients to get in touch with their emotions.
P Muntigl, NK Knight and L Angus [51]	Naming emotions or characterising experiences E.g.1, you must have been terrified E.g.2, you’re just stu:nned sorta like a slap in the face? Example 1 is an example of a mere formulation of the client’s emotion. In contrast, example 2 shows that the therapist uses vivid language or metaphors to illustrate what the client felt. Display empathy multimodally Emotion formulations or illustrations are often combined with embodied actions (E.g., multiple nods), which demonstrate the therapist’s understanding of the client’s emotions.	Assisting patients to work with their emotional experiences Empathy is a tool through which therapists assist clients to get in touch with their emotions.
P Muntigl and AO Horvath a [52]	Naming emotions or characterising experiences E.g.1, I can see some sadness in your eyes E.g.2, As I’m hearing you I sense a bit of anger in your voice The emotion formulation is prefaced with such markers as “I can see …”,“you look …”,“I’m hearing you …”	Assisting patients to work with their emotional experiences Empathy is a tool for facilitating discussions of clients’ emotions.
P Muntigl and AO Horvath b [67]	Naming emotions or characterising experiences E.g.1, so you don’t have the strength to leave- to break out. E.g.2, so somehow thes:e.hh these things seem.hh worth sacrificing your happiness for?	Assisting patients to work with their emotional experiences Empathy is a tool for facilitating discussions of clients’ emotions.
HT Nguyen [53]	Compassionate witnessing and minimal response The pharmacist attentively listens to the patient’s story about the health issue, providing response tokens such as yeah. The pharmacist provides an upshot of the story at the end.	Not discussed
J Ruusuvuori [54]	Naming emotions or characterising experiences E.g., it feels like life become so rotten Sharing similar emotions or experiences E.g., I remember #in my past I# (0.7) was maybe eight years old I came home from a choir-rehearsal cat had given been away in the meantime	Facilitating advice-giving or problem-solving Empathy is treated as expressions of understanding, which orient to problem-solving.
J Ruusuvuori [55]	Naming emotions or characterising experiences E.g.1, that’s a bit hard E.g.2, it feels like you don’t care to do anything that you become so tired Response cry E.g., oh, dear Display empathy multimodally The gesture of touching patients’ shoulders is used to display empathy.	Facilitating advice-giving or problem-solving Empathy is accomplished alongside with achieving the institutional goal of addressing the patient’s health issue. Empathy formulations work “towards closing the sequence of troubles-telling and serves to shift back to problem-solving activity, whilst in homeopathy, it may also serve as a means to problem-solving and thus help to complete the institutional task at hand” (p.597).
W Stommel and J Lamerichs [56]	Naming emotions or characterising experiences E.g.1, what a distressing message. E.g.2, oh, that sounds disturbing. E.g.3, wow! I can image that this has been very distressing for you. Claiming or displaying understanding E.g., yes, I can imagine that	Facilitating advice-giving or problem-solving Empathy displays are integrated with other actions relevant to advising activities. Empathy formulations usually do not stand alone but are accompanied with other formulations and questions which serve to explore patients’ problems.
W Stommel and Ht Molder [57]	Claiming or displaying understanding E.g.1, yeah no I understand that E.g.2, yeah no I can imagine that very well Displaying empathy multimodally Through the prosodic quality of response tokens such as yeah, no plus other empathy emotions.	Facilitating advice-giving or problem-solving Empathy serves to close off the trouble-telling episode and transition to advice. Advice is regularly launched after empathy with clients’ emotions is displayed. “The analysis reveals that these responses open up advice sequences that vary in the extent to which they treat the client’s articulated feelings as valid. Most are affiliating, treating the client’s feelings as the basis for advice, while some are less affiliative, putting the client’s feelings into perspective or implicitly questioning their legitimacy. Hence, empathically designed responses are pivots to advice-giving.” (p.523)
AL Suchman, K Markakis, HB Beckman and R Frankel [58]	Naming emotions or characterising experiences E.g.1, You seem a little upset E.g.2, you seem a little teary-eyed talking about it E.g.3, Oh just to see her suffering that long Compassionate witnessing and minimal response Clinicians used a continuer Uh-huh, yeah with empathic voice.	Not discussed
L Thompson and R McCabe [59]	Naming emotions or characterising experiences E.g., So, you’re not feeling well? Psychiatrists name the emotion in a declarative question format, which is prefaced with so.	Facilitating advice-giving or problem-solving Through empathy formulations, psychiatrists display understanding of psychiatric concerns, and close down the trouble-telling talk to transition to a new topic of discussion.
CK Tietbohl [60]	Naming emotions or characterising experiences E.g., Yeah falling is distressing, right? It’s not something we do when we’re younger, and we get worried- we worried about when we do it when we’re older.	Facilitating patient-centred care Clinicians used empathic validations when no medical solution was available in the context of primary care at a geriatric clinic. Empathy formulations serve three purposes (1) normalizing health changes of the aging patients, (2) acknowledging personal problems, and (3) respecting the patients’ decisions.
PAE Velasquez and CJ Montiel [61]	Sharing similar emotions or experiences I-statements were declarative sentences that described thoughts, feelings, etc. using the first-person perspective. The therapist oriented himself to the client’s point of view and typically involved reconstruction of client emotions.	Not discussed
W Versteeg and H te Molder [62]	Sharing similar emotions or experiences E.g., When you were telling about school and such stuff, my mind almost exploded	Facilitating advice-giving or problem-solving Through “my side” empathy formulation, counsellors can challenge clients’ perception of their problems, and set an alternative agenda for change in counselling involving patients with ADHD.
L Voutilainen and A Peräkylä [63]	Claiming or displaying understanding Empathy is displayed via extensions/co-completion, which are constructed as they were a continuing part of the client’s previous turn. These responses convey the therapist’s understanding as if it were spoken ‘from within’ the client’s experience.	Assisting patients to work with emotions Through empathy formulations, therapists assist clients to work with emotions/experiences in psychotherapy interactions. Specifically, empathy serves as a common ground for challenging clients’ emotions.
L Voutilainen, A Peräkylä and J Ruusuvuori [64]	Naming emotions or characterising experiences E.g.1, You would somehow want to.hhh be that good mother you never had E.g.2, Well quite a loss to you. E.g.3, Fight then she would be again tha:t mo:ther that you as a child were afraid of. E.g.4, You have indeed been awfully scared of her	Assisting patients to work with their emotional experiences Empathy is a tool for facilitating discussions of clients’ emotions.
A Weatherall and L Keevallik [65]	Response cry E.g., uw “The empathetic response cry is produced with a creaky voice lengthened high rounded back vowel u” (p.62).	Not discussed
E Weiste [66]	Naming emotions or characterising experiences E.g.1, so the anxiety passed E.g.2, and you calmed yourself with the sun	Assisting patients to work with emotions Empathy is a resource for working with emotional states of clients who have difficulties with emotion regulation.
Y Wu [67]	Naming emotions or characterising experiences E.g.1, Like you said just a moment ago, depressed? E.g.2, You sound frustrated on a on a number of levels with the family situation E.g.3, Wow it sounds like that’s – that’s really a hurtful kind of thing. Claiming of similar emotions or experiences E.g., Well I have the same problem as you. Offering comfort E.g., Well, that’s not bad. That’s in the right direction. Offering a candidate method for solving the reported problem E.g., So, well maybe you could explain what kind of problems you’re having to your family	Not discussed
Y Wu [68]	Naming emotions or characterising experiences E.g.1, People feel down when they are not in a good health E.g.2, You seems not to be happy Sharing similar emotions or experiences E.g.1, I also have sleeping problem E.g.2, I was hospitalized for three weeks last year I could deeply understand that it is downright inconvenient for you to be in.	Not discussed
R Wynn [69]	Naming emotions or characterising experiences E.g., Yeah, it’s very hard The GP articulates the patient’s feeling when the patient’s talk indicates a problem with expressing emotions. This practice is termed as cognitive empathy. Sharing similar emotions or experiences E.g., That’s frustrating for you and that equally frustrating for me. Here, the GP claims to share the emotion with the patient, which is termed as cognitive empathy by the author. Claiming of similar emotions or experiences E.g., I must admit I’m inclined in the same way. Here, the GP claims to have the same inclination as the patient’s, which is termed as sharing empathy by the author.	Not discussed
R Wynn and M Wynn [70]	Naming emotions or characterising experiences E.g., and you are very anxious too In the author’s words, this emotion formulation is categoriz ed as cognitive empathy. Displaying empathy multimodally Therapists can show empathy multimodally by placing a hand on the client’s shoulder and holding the client’s hands in combination with displaying understanding of the client’s emotion. In the author’s words, this way of showing empathy is categori zed as affective empathy. Sharing similar emotions or experiences Therapists can share a story that indicates the client’s emotional reaction is shared by all human beings. This type of empathy display is termed as sharing empathy in the original paper. Compassionate witnessing and minimal response Therapists can encourage clients to cry by saying “cry”, or “cry. it makes you feel better”	Assisting patients to work with their emotional experiences Empathy is a therapeutic tool to work with patients’ emotions in psychotherapy interactions.

How is empathy communicated in healthcare?

Overview

In the 43 included studies, we identified eight types of empathy communication practices: (i) naming emotions or characterising experiences (n.30); (ii) sharing similar emotions or experiences (n.12); (iii) compassionate witnessing and minimal response (n.9); (iv) claiming or displaying understanding of patients’ experiences and emotions (n.6); (v) displaying empathy multimodally (n.5); (vi) response cries (n.4); (vii) claiming similar emotions or experiences (n.3); (viii) other practices, including offering comfort (n.1), suggesting a solution to the problem reported by patients (n.1).

In Table 4, we describe each communication practice, we provide relevant terminologies and illustrative examples, and we discuss below the two most frequently researched practices.

Table 4.

Empathy communication practices

Main themes	Sub-themes	Terminology	Description	Example	Prevalence
Naming emotions or characterising experiences	• Naming the patient’s emotion	Formulation of emotional state [55], naming the client’s emotion [50]	HCPs produce an utterance that describes patients’ emotions.	You sound frustrated on a on a number of levels with the family situation [50]	30/43
	• Characterising the patient’s experience	Assessment of experiences [54]	In response to patients’ description of an experience, HCPs produce an evaluative statement of the experience.	That’s very bad [50]
		Formulation [53]	Formulations are utterances in which the current speaker suggests a meaning of what another participant has said in the prior turn or turns. [53]	I get the impression that you have recently really really pulled yourself together here. [39]
Sharing similar emotions or experiences	• Sharing a similar emotion	‘My-side’ empathy formulation [34]	HCPs describe their emotions if they were in similar situations to the patients.	When you were telling about school and such stuff, my mind almost exploded. [34]	12/43
	• Sharing a similar experience	“Empathic parallel” [45]	After patients tell a life story, HCPs tell a personal story describing experiences that parallel with patients’ experiences.	HCPs tell a personal story.
Compassionate witnessing and minimal response	• Compassionate witnessing	Compassionate witnessing [43]	HCPs stay silent and show empathy through non-verbal behaviours.	See the category – Displaying empathy multimodally.	9/43
	• Minimal response	Minimal response [59], active listening [46]	HCPs produce response tokens such as yeah, yes, yep, um hum, ah-ha, uh-hum, okay and right.	Child: So you get told t’do two different things [at]the same ti:me. Counsellor: [Yeh] [46]
	• Take-your-time formulation	Take-your-time formulation [40]	HCPs say “take your time” while patients are crying.	Take your time [40]
Claiming or displaying understanding of patients’ experiences and emotions	• Claiming to understand the patient	Knowledge claim [35]	HCPs claim that they know, understand, or can imagine how the patient would feel without naming the emotions.	Yes, I can imagine that. [57] Yeah, no I understand that very well. [35]	6/43
	• Displaying understanding	“Extension” [32]	HCPs say something as an extension of the client’s previous turn of talk. This conveys an understanding as if it were spoken from within the patient’s experience.	Patient: that very thing that one would feel more safe and that one would have erm such tenderness more Therapist: received. [32]
Displaying empathy multimodally	• Display empathy multimodally	Embodied empathy [43]	HCPs display empathy through using voice, gesture, and eye contact.	Witnessing patient’s crying, HCPs display empathy through a soothing voice, gently touching the patient’s knee, offering a tissue, standing by the patient during their crying and sobbing, maintaining direct eye contact [43] -	5/43
Response cry	• Response cry	Response cry [40]	HCPs display empathy by crying oh, gosh or oh, dear in response to patient recounts of troubles.	Oh, dear! [40]	4/43
	• Empathetic pain cry	Empathetic pain cry [69]	When patients display pain, HCPs produce a pain cry as if they experience the pain themselves.	Uw [69]
Claiming of similar emotions or experiences	• Claiming to share the same emotion without naming it	Sharing empathy [65]	HCPs claim to have similar emotions with patients.	I must admit I’m inclined in the same way [65]	3/43
	• Claiming to have a similar experience	NA-	HCPs claim to have a similar experience without telling what it is about.	Well I have the same problem as you. [64]
Others	• Offering a candidate method for solving the reported problem	Offering a candidate a method for solving the reported problem [64]	HCPs give advice on how patients can solve the reported problems.	So, well maybe you could explain what kind of problems you’re having to your family. [64]	2/43
	• Offering comfort	Offering comfort [64]	HCPs display a positive professional attitude toward clients’ problems which could ease or alleviate the clients bad feelings [64]	Well, that’s not bad. That’s in the right direction. [64]

While Tables 3 and 4 offer a summary of reported empathy communication practices, they fall short of conveying the inherent subtleties of these interactions. In the sections that follow, we synthesize studies that provide analysis of syntactical and contextual nuances that are important for making sense of empathy communication. Our focus is on the two most frequently observed practices: (1) naming emotions and characterising experiences; (2) sharing personal experiences or emotions.

Naming emotions and characterising experiences

Studies show that, depending on syntactic features and context, empathy formulations vary in their sensitivity to patients’ epistemic authority; that is, the recognized right to know based on direct, personal access to their own experiences and emotions [74]. Studies also vary in the degree of affiliation with the patient’s perspective. That is, the extent to which the HCP’s talk agrees with or validates the patient’s experience or perspective.

Sensitivity to patients’ epistemic authority

A number of studies (n.12) demonstrated that different syntactical formats are used to display HCPs’ sensitivity in their claims to be understanding patients’ experiences and emotions [32, 34–36, 39, 43–45, 51, 59, 62], 67,]. To manage the issue of patients being the only person with primary knowledge of their experiences and emotions, HCPs usually preface formulations of patients’ emotional states with epistemic markers such “I can hear/see/imagine that …” [cf.61]. For example, they may say “I can hear that you’re very very upset” [45] or “as I’m hearing you I sense a bit of anger in your voice” [67]. According to A Hepburn and J Potter [45] and J Ford, A Hepburn and R Parry [43], epistemic markers serve to indicate that the HCP has based this conclusion on observation, thereby justifying the knowledge claim. These knowledge markers are also termed as observer responses [36] or noticings [67].

To acknowledge patients’ epistemic authority over their emotional states, studies found that HCPs used question formats, which provide an upshot of the emotions while seeking confirmation from the patient [39, 44, 59]. For example, a question tag may be added after an emotion formulation, e.g. “It’s frustrating, isn’t it?” [44]. Emotion formulations can also be prefaced with “so” and produced in rising intonation, e.g. “so you’re feeling not so well?”, which is referred to as declarative question [39]. A Hepburn and J Ford [44] provided detailed analysis of the nuanced differences between question tags at the end of a turn (“It’s frustrating, isn’t it?”), and in mid-turn (“the hurt child there isn’t it with you at the moment”.) According to A Hepburn and J Ford [44], turn-final tags provide patients with an opportunity to confirm the claim, whereas mid-turn tags reduce the expectation of confirmation. Although both forms acknowledge patients’ epistemic authority, the latter does so in a more attenuated manner.

Affiliative versus disaffiliative responses

Besides variation in epistemic certainty, studies (n.10) also showed that formulations of patients’ emotional states differ in the extent to which they either validate or challenge patients’ emotional experiences [13, 43, 45, 50–53, 57, 60, 62, 75]. In other words, these formulations vary in their degree of affiliation with the patients’ perspective [67]. Tietbohl (2022), specifically, coined the term empathic validation to describe practices of empathically validating the experiences of patients. For example, when discussing a patient’s experience with falling, a doctor said: “yeah falling is distressing, right? It’s not something we do when we’re younger, and we get worried about when we do it when we’re older” [60]. Here the doctor was not only characterising the experience as distressing but also provided an account for the distress. This works to validate, or affiliate with the patient’s perspective.

Studies showed that in certain contexts, HCPs needed to both empathize and disaffiliate with patients. For example, in the context of suicide prevention helpline, C Iversen [13] showed that call-takers needed to skilfully express empathy without endorsing callers’ suicidal ideations. In the context of psychotherapy, challenging patients’ problematic emotions may be used as a therapeutic strategy. P Muntigl and AO Horvath [67] provided an example in the context of Gestalt therapy, where a therapist said: “you said you’re scared but you’re smiling”. This formulation, according to P Muntigl and AO Horvath [67], was considered highly disaffiliative with the patient’s affective stance.

Studies also demonstrated that disaffiliating can be done subtly. For example, W Stommel and Ht Molder [56] showed that the blurred hedge between affiliation and disaffiliation can also be expressed through HCPs’ changes in voice tone. According to W Stommel and Ht Molder [56], producing: “that’s very difficult” in a flat voice, in response to the client’s utterance: “what a world! Dam it!” was treated as disaffiliative. Note that in this example not only the voice tone but also the word choice “difficult” did not align with the client’s affective stance. P Muntigl [50] provided another example of subtle disaffiliation in an emotion-focused therapy context: when a client told a story about her husband, the therapist provided an upshot: “so he said I’m cold and you just feel like disappointed or?”. In her story, the client did not reveal how she felt about the experience; the therapist’s response worked to elicit the patient’s naming of her emotion. In this sense, the therapist’s formulation shifted the focus from the client’s husband to the client’s emotion, thereby challenging the client’s affective stance.

Sharing personal experiences or emotions

Studies have used different terminologies to refer to the practice of sharing personal experiences and emotions, including parallel experience or empathic parallel [46], second story [54], and ‘my-side’ empathy formulation [62]. J Heritage and A Lindström [46] analysed an example of “empathic parallel” in the context of home visits by a nurse to a new mother. In response to the mother’s telling of her trouble bonding with her new-born child, the nurse initially provided generalized predictions about the dynamics of the mother-child relationship. As the conversation progresses, she shared her own experience after the birth of her own son. J Heritage and A Lindström [46] showed that, while the mother responded minimally to the generalized predictions, she enthusiastically endorsed the empathic parallels offered by the nurse. However, in general practice and homeopathic consultations, J Ruusuvuori [54] found that HCPs rarely told second stories in response to patients’ first stories. By avoiding sharing personal stories, J Ruusuvuori [54] argued HCPs kept the conversation focused on the patient’s experience, ensuring that problem-solving remained the main objective of the consultation.

The term “my-side” empathy formulation, which was coined by W Versteeg and H te Molder [62], refers to HCPs’ description of their own emotions as they are listening to the patient’s story. For example, a HCP may say: “when you were telling about school and such stuff, my mind almost exploded” [62]. W Versteeg and H te Molder [62] argued that this practice can address the challenge of balancing distance and involvement with the patient’s emotion, while achieving high levels of empathy through claiming independent access to the patient’s mental state.

What functions does empathy communication serve?

Section “How is empathy communicated in healthcare?” illustrates what empathy communication practices look like across different contexts and highlights the subtle variations in the formulations used to convey empathy. The following section examines the interactional functions of these practices in their specific healthcare settings. Together, these sections offer a comprehensive understanding of both the forms and the functions of empathy communication practices.

Overview

Twenty-nine included studies cover exploration of the functions of empathy communication (Table 5). These functions were grouped in four categories: (i) Facilitating advice-giving or problem-solving2 (n.14), identified in a broad range of contexts including helplines, counselling, nursing home visits, psychotherapy, psychiatry, general practice, oncology, audiology, palliative care, and an unspecified context; (ii) Assisting patients to work with emotions (n.13), found predominantly in psychotherapy, with additional instances in family counselling and paediatrics; (iii) Facilitating patient-centred care (n.1), occurred in primary care [42]; (iv) Facilitating history-taking activities (n.1) found in a study where the context was unspecified [39].

Table 5.

Functions of empathy communication

Main themes	Relevant healthcare contexts
Facilitating advice-giving or problem-solving (n.14)	Helpline: 04 Family counselling: 01 Nursing home visit: 01 Psychotherapy: 01 Psychiatry: 01 General practice: 01 Oncology: 01 Audiology: 01 Palliative care: 01 Unspecified context: 01
Assisting patients to work with emotions (n.13)	Psychotherapy: 11 Family Counselling: 01 Paediatrics: 01
Facilitating patient-centred care (n.1)	Primary care: 01
Facilitating history-taking activities (n.1)	Unspecified context: 01

Facilitating advice-giving or problem-solving

Included studies explicate the nuances of how empathy communication facilitates advice-giving or problem-solving. To illustrate this, we report analyses of two extracts from two different studies, W Stommel and Ht Molder [57] and J Ford, A Hepburn and R Parry [43], showing respectively affiliative (extract 1) and disaffiliative (extract 2) practices.

W Stommel and Ht Molder [57] analysed an interaction between a client and a counsellor in the context of an Alcohol and Drugs Helpline in Dutch, in which the counsellor displayed affiliation with the patient’s perspective (extract 1). We used both patient and client because these terms were employed as such in the original studies, which spanned diverse healthcare and social care settings.

The client was engaged in trouble-telling about dealing with her alcoholic husband. She concluded by naming her emotion regarding the experience “I’m actually kind of sick of that hassle” (line 6). Through a knowledge claim about the experience: “I can imagine that” (line 7), the counsellor affiliated, responding with empathy. The delivery features of this empathy formulation, including its fast delivery (marked by > <) and the quick launch into advising the client not to accept her husband’s alcohol abuse (lines 8–10), suggest that empathy was not displayed solely for empathy’s sake, but serves an additional function. By characterising alcoholism both as an illness and as a problem one can quit, the counsellor transitioned from acknowledging the problem to advising the client not to accept her husband’s alcohol abuse.

J Ford, A Hepburn and R Parry [43] provided analysis of a palliative care consultation at a hospice, involving a patient who disagreed with medical advice (extract 2).

Prior to extract 2, the patient expressed dissatisfaction with a hospital doctor who suggested she should engage in physical activity. She perceived this advice as insensitive and offensive because it assumed that she was well enough to do so, thus implying that she had exaggerated her illness. Between lines 1 and 3 the palliative care doctor provided a justification for the recommendation, which was disaffiliative with the patient’s perspective that the advice was insensitive. After this disaffiliative statement, the doctor immediately launched an empathy formulation (lines 4 and 6), which was prefaced with “I know” and embedded with an assessment “that’s hard”. Here the doctor acknowledged the patient’s health condition and the associated challenges of doing physical activity. Finally, the doctor relaunched the advice (lines 7–9), with the preface “but we know …”. Here “but” suggests disaffiliation with what was said earlier while “we know” invokes medical authority, which worked to justify and while softening the disaffiliative advice.

Overall, W Stommel and Ht Molder [57] and J Ford, A Hepburn and R Parry [43] separately showed that displays of empathy are not standalone actions, but are integrated into the ongoing activity of giving advice, regardless of whether they occur in affiliative or disaffiliative contexts. W Stommel and Ht Molder [57] illustrated that in affiliative contexts empathy constructions served to transition from acknowledging the client’s difficult experience to giving advice on how to address the problem (extract 1). In contrast, J Ford, A Hepburn and R Parry [43] showed that empathy constructions in disaffiliative contexts work to soften the disaffiliation between the client’s perspective and the medical perspective, while providing opportunities for justifying and reinforcing the original advice (extract 2).

Assisting patients to work with their emotional experiences

Included studies identified different ways HCPs’ empathy constructions assisted patients in working with their emotional experiences. To illustrate this, we report analyses from P Muntigl, NK Knight and L Angus [51] and L Voutilainen, A Perakyla and J Ruusuvuori [73] which examined two different practices: 1) naming patient’s emotions to assist in articulating and topicalising the emotion; 2) characterising and expanding patients’ tellings about their experiences, which facilitates self-reflection on the experience from a new perspective.

P Muntigl, NK Knight and L Angus [51] analysed an interaction from an emotion-focussed therapy session, where a client was struggling to articulate the emotional impact of mistreatment by her husband (extract 3).

In this interaction, the client had been telling a story about the family opportunistically meeting a student of her husband. In line 16 she summarized the main point: the client’s husband introduced their daughter to the student but did not introduce her. In line 18, she then attempted to characterize the experience through a formulation starting with “which was”. However, she did not complete the formulation; and instead remained silent for 0.3 second (line 19), which indicated a difficulty in expressing the emotional impacts of the experience. The therapist subsequently named the emotion “you must have felt so slighted” (line 20), displaying an understanding of how the client would feel about the experience. The emotion word “slighted” was produced with high pitch, which showed emphasis and emotion. This turn by the therapist topicalizes the emotion, and the client responded saying that the experience was very bad (line 21). Following the therapists’ acknowledgement, through her elongated “yeah” (line 22), the client further discussed her emotion (line 23).

L Voutilainen, A Perakyla and J Ruusuvuori [73] investigated the practice of combining displaying empathy with challenging what the client’s description of their experience was, in a psychotherapy session in Finland (extract 4).

In lines 4–5, the therapist produced an interpretation of the experience the patient had just recounted about her mother. In addition, the therapist also suggested that the patient’s wish to be a mother was connected to her experience of not having a good mother. This interpretation was slightly different to the patient’s initial description, where the patient did not make a connection between her wish to have a baby and her childhood experience. Between lines 8 and 23, the patient responded by considering this new perspective, indicated by her confirming responses (lines 8), pauses (lines 9 and 11), and reflective comments (lines 12–23). Here she reflected on the possibility that her desire for a baby might be an attempt to compensate for her own lack of good mothering.

Overall, P Muntigl, NK Knight and L Angus [51] and L Voutilainen, A Perakyla and J Ruusuvuori [73] separately showed that empathy formulations serve as a therapeutic tool that facilitates patients’ articulation of challenging emotions and reflection on difficult experiences, thereby also helping in supporting patients when processing and managing their emotions.

Discussion and conclusion

Discussion of key findings

In summary, this review provides insights into how HCPs communicate with empathy when clients display painful and distressing emotions, and suffering. We identified eight categories of empathy communication practices: 1) naming emotions or characterising experiences, 2) sharing similar emotions or experiences, 3) compassionate witnessing and minimal response, 4) claiming or displaying understanding of patients’ experiences and emotions, 5) displaying empathy multimodally, 6) response cries, 7) claiming of similar emotions or experiences, and 8) others.

Among these eight categories, naming emotions and characterising experiences are the most common (n.30). These formulations of emotions and experiences are usually prefaced with epistemic markers (I see/know/understand, I can hear that, etc.), or produced with raising intonation (i.e., formatting it as a declarative question), or with question tags. The rising intonation serves to suggest a low level of certainty in the knowledge claim, and is contingent on the patient’s confirmation, thereby showing HCPs’ respect for the patient’s epistemic authority over knowledge that they have direct access to, which is their lived experiences [74]. The second practice, sharing similar personal experiences and emotions (n.12) allows HCPs to show understanding without making direct statements about patients’ mental states, thereby showing respect for the patient’s epistemic authority. Other practices, such as compassionate witnessing, minimal responses (e.g. “mm hm”), response cries (e.g. “oh gosh”) or using empathetic embodied conduct are reported in the literature to be useful tools to display empathy. However, as they are reported less frequently, they are examined in less detail and their functions are described with less nuance.

Another key finding is that practices to display empathy also vary in the degree of affiliation with patients’ perspective. Disaffiliative formulations of the patients’ emotional states or experiences occur when: 1) the emotion characterized by the HCP is more intense or different to that expressed by the patients; 2) the focus in the HCP’s characterisation is different from the patient’s one; 3) the experience described by the HCP is different from the experience the patient has just described [72, 76]. HCPs encounter patients whose perspectives diverge from the medical perspective [54]. In some healthcare contexts, challenging patients’ perspectives and emotions is the essential task [32, 51, 76]. In such circumstances, HCPs can still acknowledge or validate patients’ emotions without having to affiliate with patients’ perspective. A typical case of dilemma is to empathize with patients experiencing suicidal ideation without endorsing their suicidal thoughts [13].

Regarding the functions of empathy communication, the reviewed studies indicate that displays of empathy open the space for sequences of advice-giving and problem-solving in most of the healthcare settings examined. Specifically, empathy can serve to transition from acknowledging the patient’s emotion or difficult experience to providing advice on how the issue can be addressed. It can also mitigate disaffiliation and help HCPs regain patients’ affiliation in cases when the patient’s perspective is incongruent with the medical view. In psychotherapy settings, particularly, empathy communication facilitates discussion of problematic emotions, and reflection on difficult experiences, thus helping patients manage in situ their emotional experiences.

Altogether, these findings suggest that empathy communication facilitates the progression of healthcare tasks. This result challenges the assumptions that there is insufficient time for empathy communication and that such communication is at odds with achieving clinical outcomes [14, 15]. On the contrary, the reviewed studies offer evidence that empathy communication can be woven efficiently into the flow of healthcare consultations without significantly prolonging them [43]. This integration supports the notion that empathy is not an optional addition to care but a core component that can contribute to better health outcomes, adherence to treatment, and overall patient satisfaction.

Practical implications

This systematic review collates empirical evidence regarding empathy communication. The review shows that HCPs are already skilled communicators who routinely engage in empathic interactions. However, it also highlights the complexity and contextual sensitivity of empathy in clinical settings, which is helpful for novice HCPs. Empathic communication involves more than simply recognising patients’ emotions; it often includes subtle decisions about when and how to respond, depending on the clinical context and interpersonal dynamics. Variations in syntax and epistemic markers can influence the degree of certainty expressed. For example, assertive statements about a patient’s emotional state may inadvertently undermine their epistemic authority over their own experiences, potentially causing a rupture in the interpersonal relationship. Furthermore, as empathy communication practices can vary in their level of affiliation with patients’ perspectives, careful attention needs to be paid to when and how to disaffiliate with patients’ perspectives while not dismissing their emotions. Being able to disaffiliate without dismissing the patient’s emotions is a nuanced but vital skill, especially when clinical evidence and the lifeworld diverge, an area that HCPs report finding challenging [77–79].

Although concerns about time constraints are reported [14, 15], evidence from this review suggests that empathy communication can be integrated into routine clinical activities, such as giving advice or addressing concerns, rather than requiring additional time. However, this review does not support instrumentalising empathy, using it solely to advance a medical agenda, as this may undermine trust [16]. Ultimately, a deeper understanding of the nuances involved in empathic communication can support HCPs in continuing to provide care that is both emotionally attuned and clinically effective, contributing to enhanced patient experiences and outcomes.

The findings of this review also offer several concrete implications for empathy training curricula. Current programs often rely on theoretical definitions of empathy without showing how it is enacted in real clinical interactions. By contrast, the studies synthesised in this review provide detailed evidence of the linguistic and interactional practices through which HCPs display empathy in practice. Training programs could therefore incorporate annotated extracts from naturalistic interactions to illustrate how specific linguistic formulations function to acknowledge patients’ emotional states while respecting their epistemic authority, and how epistemic markers, intonations, touches can be used to display empathy. Highlighting these observable, teachable behaviors would help move training from abstract principles toward practical, replicable skills.

In addition, training should emphasize how different empathy practices work in context, including in high-stakes settings where HCPs must acknowledge distress without affiliating with problematic perspectives (e.g., suicidal ideation). Integrating these insights into role-play scenarios, simulation training, and feedback frameworks would help prepare HCPs to use empathy effectively in specific context.

Strengths and limitations

This review’s primary strength is the application of systematic review methods to interactional linguistic research, a field where systematic review methodology is rarely utilized. This methodology allowed us to deliver a comprehensive evaluation of the evidence on empathy communication in healthcare. As much of this work was published in linguistic journals, with 10 studies found on a specialist database, we make the evidence available to HCPs, educators and policymakers. Although we used a systematic search strategy, we might have missed some relevant studies because many studies were published in social science journals without Medical Subject Headings (MeSH) terms. To mitigate this, we adopted a comprehensive search strategy, consulted experts, and explored the specialist bibliographic database for conversation analytic research, which added a significant number of included studies (10 over 43 studies).

Nevertheless, the review had some limitations. One key limitation concerns the expertise of the review team. Three of the four reviewers are specialists in CA. Although we are familiar with other applied linguistic approaches, we have primarily used CA in our own work (all except VC, who has wider applied linguistic background), which introduces two potential risks. The first is the possibility of missing relevant non-CA studies. To address this, we consulted applied linguistics experts to help refine our search terms and ensure adequate coverage of a range of research methods. The second potential risk is bias in assessing the breadth and depth of non-CA studies. To minimise this, we refrained from making simplistic judgments about study quality. Instead, we focused on describing and characterising key methodological features, thereby foregrounding factors that readers may wish to take into account when assessing quality for themselves. We thus provided general guidance rather than prescriptive evaluations. A high-quality study may offer rich, in-depth analysis without drawing on a wide range of data; conversely, a study may be based on a large dataset while providing limited analytic depth. By contrast, lower-quality studies tend to exhibit neither analytic depth nor breadth of data.

The second major limitation was a lack of focus on patients’ responses to HCPs’ empathy display in the included studies, which limited our ability to evaluate empathy communication practices from the patient perspective. Most studies focused on linguistic constructions, with very few studies discussing the use of voice and embodied conducts. To demonstrate the affective aspect of empathy, future research should focus on how empathy is communicated across an array of channels other than linguistic constructions. Most of the reviewed studies used a small and medium quantity of data, and qualitative methods. Quantitative analysis drawing on larger data sets of interactional data may be necessary for uncovering regular patterns of empathy communication practices across contexts.

In addition to these limitations, the review included an unequal representation of healthcare settings with psychotherapy and general practice dominating. Future research in diverse healthcare contexts is needed, such as secondary care and tertiary care. Additionally, over two thirds of studies reported on data containing English language interaction, less than a third of studies reported interactions in other languages (mostly European languages such as Finnish, German and Swedish). Notably, there is a marked absence of studies from low-income or Global South countries, which limits the cultural and linguistic diversity of the evidence base and raises concerns about the global and cross-cultural applicability of the findings.

Conclusion

In conclusion, this review makes a valuable contribution by presenting the first systematic synthesis of the growing body of research that draws on observational data and applied linguistic methods to explore how empathy, sympathy, and compassion are communicated in real-world clinical interactions. It found that empathy communication can be seamlessly integrated into healthcare consultations. Far from distracting from clinical goals, empathy can enhance patient engagement, facilitate emotional expression, support problem-solving, and help reconcile differing patient and HCP perspectives, making it an essential element of care. This insight is important to inform the advancement of training and enhance communication practices in healthcare settings.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

B.T. reviewed studies, wrote the first draft of the manuscript, contributed to the conception and design the study, and conducted the analysis and interpretation of the data. R.D. reviewed studies, wrote parts of the findings and discussion, and contributed to the analysis and interpretation of the data. V.C. reviewed studies, reviewed the manuscript, and contributed to the analysis and interpretation of the data. C.A. reviewed studies, wrote parts of the method, reviewed the papers, and contributed to the analysis and interpretation of the data.

Funding

This research was not sponsored any research fund.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethical approval

Not applicable.

Consent for publication

All authors have reviewed and approved the final version of the manuscript for publication.

Competing interests

The authors declare no competing interests.