Are clinicians’ self-reported empathic concern and perspective-taking traits associated with their response to patient emotions?

Abstract

Objective:

To understand whether clinicians’ empathic concern and perspective-taking traits are associated with their response to patient emotions.

Methods:

We audio-recorded 41 HIV clinician interactions with 342 patients at two academic medical centers. We assessed clinicians’ self-reported empathic concern and perspective-taking traits using the Interpersonal Reactivity Index and coded emotional communication using the Verona Coding Definitions of Emotional Sequences. We used random effects models to assess associations between clinician traits and clinician responses to patients’ negative emotions, accounting for clustering of emotions within encounters and patients within clinicians.

Results:

Clinicians with more self-reported empathic concern received fewer emotional expressions from their patients (β −0.06; 95% CI −0.10, −0.01) and had greater odds of responding to emotions by giving information/advice (OR 1.10; 95% CI 1.01, 1.20). There were no associations between empathic concern or perspective-taking and any other clinician responses.

Conclusion:

Clinicians with higher levels of empathic concern respond to patient emotions by giving information and advice, a response traditionally thought of as a missed empathic opportunity, not by exploring emotions or providing empathy. Whether this is helpful to patients is unknown.

Practice Implications:

Clinicians should be aware of their tendency to give information to patients with emotional distress, and consider whether this response is helpful to patients.

1. INTRODUCTION

It is broadly understood that emotional communication in medical encounters is fundamental to creating and maintaining relationships between clinicians and patients [1], and the benefits of empathy are widely cited [2]. Yet, the term ‘empathy’ is a complex concept with many definitions, which is reflected in the array of measurement approaches that consider empathy as a clinician character trait, communication behavior, or patient experience [3]. When citing the benefits of empathy in patient care, there should be greater conceptual clarity about which forms of ‘empathy’ are being referred to and how these related phenomena fit together (e.g., do empathic providers communicate empathically and make patients feel their emotions are understood and accepted).

Several studies have examined the association between clinician self-reported empathy and patients’ outcomes, with mixed results. Two studies found that patients of clinicians with higher scores on the Jefferson Scale of Empathy (JSE) had fewer metabolic complications from their diabetes [4,5]. Yet, a more recent 2019 study of 4,176 patients who received care from one of 51 primary care physicians found no association between JSE scores and diabetes outcomes [6]. In terms of the relationship between provider self-reported empathy and patient satisfaction, one study found a modest positive association between JSE scores of 41 providers and the satisfaction ratings of 1,308 patients in the Emergency Room setting [7], whereas another study that used the Interpersonal Reactivity Index (IRI) to measure clinician empathy found no association with patient ratings of clinician communication in the context of HIV care [8]. However, this later study found that clinicians who self-reported higher scores on the IRI conveyed more information to patients, with less emotional communication, and their patients had greater odds of reporting the highest medication self-efficacy. This finding suggests a possible causal pathway between clinician’s empathy and patient clinical outcomes, one that does not necessarily operate through what traditionally is considered ‘empathic’ communication.

In this study, we seek to disentangle two related concepts that are both referred to as “empathy.” We first focus on the clinician’s experience of empathy in the face of patient emotion. This experience falls into two domains: ‘empathic concern’ (the degree of emotional resonance that one has with another person’s experience, sometimes referred to as affective empathy) and ‘perspective-taking’ (the ability to understand the experience of another person, sometimes referred to as cognitive empathy). Because there is controversy over how affective and cognitive empathy influence the way clinicians communicate, we focused this study on clinicians’ behavior in response to patient emotion. We hypothesized that clinicians with greater cognitive and affective empathy respond to emotions by exploring it (due to their ability to recognize and be interested in the patient’s experience), attempting to help solve the problem underlying the emotion by giving information or advice (as suggested by previous studies) [8,9], or by acknowledging the emotion with empathic communication behavior. We also hypothesized that clinicians with greater cognitive and affective empathy would less frequently respond by blocking the emotion.

By understanding the relationship between what clinicians experience and what they subsequently say in response to patient emotion, we hope to gain conceptual clarity in interpreting previous studies examining the impact of empathy on patient outcomes. In addition, this more nuanced understanding would promote educational interventions that integrate teaching communication skills alongside emotional awareness and perspective-taking.

2. METHODS

2.1. Study Design, Subjects, and Setting.

We conducted a cross-sectional analysis within the MaRIPOHSA (Maximizing Respect and Improving Patient Outcomes in HIV and Substance Abuse) Study, conducted at two urban academic medical centers. The Institutional Review Boards at Johns Hopkins University and Oregon Health & Science University approved this observational study. Clinicians (physicians, nurse practitioners, or physician assistants) were eligible to enroll if they provided primary care to patients with HIV. Adult, English-speaking patients were eligible to enroll if they were established in HIV care for at least six months and were being seen by their regular HIV clinician for a routine ambulatory encounter.

2.2. Data Collection

All participants provided informed consent before any research procedure. Clinicians were recruited at faculty meetings and completed a baseline questionnaire that included demographic variables and the two subscales of empathy described below. Patients were recruited from the clinic’s waiting room. After obtaining informed consent, we placed two audio recorders in the examination room where the medical visit occurred between the patient and clinician. Audio-recordings were transcribed by a professional transcription company, and the transcripts were checked for accuracy by a senior research coordinator and the research assistants familiar with the clinic environment and population.

2.3. Main Independent Variables

Clinicians completed two subscales of the IRI, which is a tool that measures multidimensional facets of dispositional characteristics [10,11]. The IRI has been the most commonly used operational measure to assess clinician empathy [3]. Clinicians completed the IRI subscale of ‘empathic concern,’ which is the self-reported “tendency for the respondent to experience feelings of warmth, compassion and concern for others undergoing negative experiences” [10]. Clinicians also completed the subscale on ‘perspective-taking,’ which is the self-reported “tendency or ability of the respondent to adopt the perspective, or point of view, of other people.” The IRI uses a five-point Likert scale (0–4), with four being the most representative or descriptive of the person being assessed. Each subscale has seven items. For example, the item, “I often have tender, concerned feelings for people less fortunate than me,” represents empathic concern. An example of perspective-taking is, “I try to look at everybody’s side of a disagreement before I make a decision.” Higher scores out of the 28-point scale indicate greater amounts of the trait. In our sample, the internal reliability for the perspective-taking scale was 0.774, and the internal reliability of the empathic-concern scale was 0.585.

2.4. Dependent Variables

Our dependent variables were the number of patient emotional expressions and the types of clinician responses. To measure these, we coded the transcripts using the Verona Coding Definitions of Emotional Sequences (VR-CoDES), which has been used internationally and across specialties to explore emotional communication between the patient and clinician [12–14]. The VR-CoDES has been found to be reliable and valid in terms of representing and describing emotional talk between the patient and clinician [15,16].

2.4.1. Patient Emotional Expressions.

The VR-CoDES categorizes patient expression of negative emotions into concerns and cues according to the level of directness. A concern is defined as “a clear and unambiguous expression of an unpleasant current or recent emotion where the emotion is explicitly verbalized.” Meanwhile, a cue is “a verbal or non-verbal hint which suggests an underlying unpleasant emotion but lacks clarity.” Since we coded the emotional dialogue from transcripts, we were not able to capture all non-verbal hints, such as the tone of voice. However, the transcripts did indicate when a patient expressed themselves through crying, silence, gasps, and sighs, which were coded as a nonverbal cue.

2.4.2. Clinician Response to Patient Emotional Expression.

2.4.2.1. Primary Response Categories

Our primary clinician response outcomes were the categories established by the VR-CoDES [13]. The coding system presents 17 possible types of clinician responses to patients’ emotional expressions (Table 1a). These codes are grouped into two primary categories, whether or not the clinician refers explicitly to the content or emotion of the emotional issue (i.e., explicit vs. non-explicit) and whether or not the clinician provides the patient the opportunity to express their negative emotions further (i.e., provide vs. reduce space).

Table 1a.

Distribution of Clinician Responses to Patient Emotional Expressions (N=1,028)

Overarching Responses	Specific Response Types	Definitions (Examples)	n (%)
Non-Explicit- Reduces Space 73 (7%)	Ignore (code NRIg)	Ignoring the patient.	55 (5%)
	Information/Advice (code NRIa)	Giving information or advice to the patient without direct reference to the patient emotional expression (e.g., “Everything will be fine.”).	15 (1%)
	Shutting Down (code NRSd)	Shutting the patient down (e.g., “Oh, don’t be silly!”).	3 (<1%)
Non-Explicit- Provides Space 504 (49%)	Acknowledgement (code NPAc)	Acknowledging using moderate verbal encouragement (e.g., “Are you really?”).	88 (9%)
	Back Channeling (code NPBc)	Back-channeling with minimal verbal encouragement (e.g., “Okay” or “Mm-hmm”).	340 (33%)
	Active Invitation (code NPAi)	Actively inviting the patient to talk further (e.g., “Would you like to tell me more?”).	26 (3%)
	Implicit Empathy (code NPIm)	Providing empathy that implies that the clinician recognized the emotion but does not specifically repeat it back (e.g., “I understand.”).	43 (4%)
	Silence (code NPSi)	Providing silence.	7 (1%)
Explicit- Reduces Space 127 (12%)	Information/Advice (code ERIa)	Giving information or advice to the patient with direct reference to the patient’s emotional expression (e.g., “I don’t think it is an infection”).	113 (11%)
	Switching (code ERSw)	Explicitly directing the discussion away from emotional content (e.g., “Talk to him tomorrow. He’s t- ‘ b est expert on it, I think.”).	8 (1%)
	Postponing (code ERPp)	Postponing discussion (e.g., “Hold on one second.”).	6 (1%)
	Active Blocking (code ERAb)	Actively blocking the patient from elaborating on the motional issue (e.g., “Worrying does not do you any good.”).	0 (0%)
Explicit- Provides Space 324 (32%)	Content Acknowledgement (code EPCA)	Acknowledging the circumstance that gave rise to the emotion (e.g., “the operation?”).	103 (10%)
	Content Exploration (code EpCEX)	Asking more about the circumstance (e.g., “What operation are you going to have?”).	183 (18%)
	Affective Acknowledgement (cod’ EP AAc)	Acknowledging the emotion itself (e.g., “worried?”).	14 (1%)
	Affective Exploration (code EPAEx)	Asking more about the emotional experience (e.g., “Why are you so worried?”).	16 (2%)
	Empathy (code EPAEm)	Expression of empathy that repeats back to the patient the emotion that is heard (e.g., “I’m sorry. I can understand why that would be really worrisome.”).	8 (1%)

2.4.2.2. Secondary Conceptual Response Categories

We also assessed secondary outcomes by grouping the 17 specific responses a priori into qualitatively similar groups, which were not mutually exclusive (Table 1b) and also used in two prior studies [17,18]. We regrouped clinician responses that reduced space for the patient to talk based on whether or not the clinician gave information or advice, or actively tried to avoid the emotional expression. We also regrouped responses that provided space for the patient to talk depending on whether or not the clinician expressed empathy (a verbal expression that shows clinician understanding of patient emotion), focused explicitly on the patient’s emotion, explored the emotional issue, provided acknowledgement, or gave a passive response through silence, back-channeling, or non-explicit acknowledgement.

Table 1b.

Distribution of Clinician Responses to Patient Emotional Expressions (N=1,028)

Secondary Conceptual Categories		Definitions	n (%)
Subcategories of Reducing Space	Information/advice (codes NRIa, ERIa)	Gave information or advice.	128 (12%)
	Blocking (codes NRIg, NRSd, ERSw, ERAb)	Actively tried to avoid the emotional expression.	66 (6%)
Subcategories of Providing Space	Neutral/passive (codes NPSi, NPBc, NPAc)	Was passive by giving silence, back-channeling, or providing non-explicit acknowledgement.	435 (42%)
	Exploring (codes NPAi, EPAEx, EPCEx)	Explored the emotional issue by asking the patient for more information that referred to the emotion or circumstance.	225 (22%)
	Acknowledgement (codes NPAc, EPAAc, EPCAc)	Acknowledged the emotion or circumstance.	205 (20%)
	Any empathy (codes NPIm, EPAEm)	Expressed empathy.	51 (5%)
	Explicit focus on emotion (codes EPAAc, EPAEx, EPAEm)	Focused explicitly on the emotion by acknowledging it, asking about it, or providing explicit empathy..	38 (4%)

^*Secondary conceptual categories are not mutually exclusive and do not add to 100%.

2.5. Covariates

Clinicians self-reported their age, gender, and race/ethnicity on questionnaires.

2.6. Statistical Analysis

We used descriptive statistics to explore patient, clinician, and visit characteristics. Then, we evaluated whether clinician demographic characteristics (age, gender, and race/ethnicity) were associated with empathic concern, perspective-taking, and any of the response types. Finally, we evaluated whether clinician dispositional traits (self-reported empathic concern and perspective-taking) were associated with each other, the number of patient emotional expressions, and clinicians’ style of responses. We conducted multilevel mixed effects linear regression to assess the association of clinicians’ empathic traits (empathic concern and perspective-taking) with the number of patient emotional expressions. We then used multilevel mixed effects logistic regression to assess associations between clinicians’ empathic traits (empathic concern and perspective-taking) with types of clinician responses. All regression analyses accounted for nesting of patient emotional expressions within each visit and nesting of patient visits within clinicians.

3. RESULTS

3.1. Sample and Visit Characteristics

There were 41 HIV clinicians and 342 patients in our sample. Table 2 presents clinician and patient participant demographic characteristics, respectively. Clinicians had a mean age of 46 years of age; most were female (66%) and white (66%). Most clinicians were physicians (73%), and the remainder were nurse-practitioners or physician assistants. Patients had a mean age of 53 years; most were male (64%) and black/African American (77%). On average, clinicians interacted with eight patients (range of 2 to 10 patients per clinician).

Table 2.

Demographic Characteristics

Characteristics		Clinician (N=41)	Patient (N=342)
Age, mean (SD)		45.7 (10.3)	53.2 (10.2)
Gender, n (%)
	Male	14 (34.1%)	218 (63.7%)
	Female	27 (65.9%)	124 (36.3%)
Race, n (%)
	White/Caucasian	27 (65.9%)	69 (20.2%)
	Black/Afric;an American	6 (14.6%)	262 (76.6%)
	Hispanic/Latino	2 (4.9%)	-
	Asian	4 (9.8%)	2 (0.6%)
	Other	2 (4.9%)	9 (2.6%)

The mean (SD) clinician empathic-concern score was 22.0 (3.2), with a range 14–27 of 28. The mean (SD) perspective-taking score was 19.9 (4.0), with a range 8–27 of 28. The correlation between empathic concern and perspective-taking scales was 0.696 (p-value<0.001). There were no differences by clinician age, race/ethnicity, or training (physician vs. other clinician types) in empathic concern nor perspective-taking. However, female vs. male clinicians had a trend towards higher levels of empathic concern (mean 22.6 vs. 20.6, respectively, p-value=0.07) without differences in perspective-taking (mean 20.3 vs. 19.1, respectively, p-value=0.39).

The mean length of the medical visits was 30.4 (range of 9.5–75.0; SD 11.9) minutes, and there were 1,028 emotional expressions made by patients within the 342 visits. Most visits had at least one emotional expression (66.7%). The mean number of emotional expressions was 4.5 per visit (range of 1– 24; SD 3.6) among the 228 encounters that contained at least one emotional expression and 3.0 (range of 0–24; SD 3.7) among all 342 encounters. Among the 1,028 total emotional expressions, there were 750 cues (73%) and 278 concerns (27%).

3.2. Clinician Response to Emotional Expressions

Table 1a presents the distribution of clinician responses to patient expression of negative emotion. Clinicians most often provided space for patients to speak about their negative emotions, totaling 81% of their responses; responses that reduced space were less common (19%). Non-explicit responses were slightly more common (56%) than explicit responses (44%). The most frequent specific clinician responses were back-channeling (33%) and exploration of the content behind patients’ emotions (18%). As presented in Table 1b, statements that encompassed any empathy (5%) or an explicit focus on emotion (4%) were relatively rare. Any type of blocking response, such as shutting down or ignoring the emotional expression, were also uncommon (6%). There were no differences in clinician response types by clinician age, gender, nor race/ethnicity.

3.3. Clinician Empathic Tendency and Characteristics of Emotional Communication

There were fewer patient emotional expressions as clinician empathic concern increased (β −0.06; 95% CI −0.10, −0.01), with a similar nonsignificant trend for perspective-taking (β −0.03; 95% CI −0.07, 0.004). There were no differences in the odds of explicit versus non-explicit clinician responses nor in the odds of clinician responses that provided versus reduced emotional space, based on either empathic concern or perspective-taking (Table 3).

Table 3.

Association of Clinician Empathic Tendency with Patient Emotional Expressions and Clinician Response to Emotion

Patient and Clinician Emotional Communication	Clinician Characteristics
	Dimensions of Empathy
	Empathic Concern	Perspective-T aking
Patient Emotional Expression	β (95% CI) p-value	P (95% CI) p-value
Number of cues/concerns per visit*	−0.06(−0.10, −0.01) 0.016	−0.03(−0.07, 0.004) 0.083
Primary Clinician Response Categories^	OR (95% CI) p-value	OR (95% CI) p-value
Explicit	1.00(0.95, 1.06) 0.994	0.99(0.95, 1. 3) 0.605
Provides space	0.96(0.89, 1.04) 0.303	0.97(0.91, 1.03) 0.3 8
Secondary Clinician Response Categories^
Empathy	1.03(0.88, 1.19) 0.742	0. 96(0.86, 1.08) 0.533
Explicit focus on patient affect	1.03(0.91, 1.16) 0.677	0.98(0.89, 1.08) 0.653
Acknowledgement	0.94(0.87, 1.01) 0.113	0.97(0.92, 1.03) 0.392
Exploring	0.99(0.94, 1.05) 0.830	0.98(0.93, 1.02) 0.341
Neutral/passive	1.01(0.94, 1.08) 0.899	1.02(0.97, 1.08) 0.349
Gives information/Advice	1.10(1.01, 1.20) 0.032	1.06(0.99, 1.14) 0.076
Blocking	0.95(0.86, 1.05) 0.239	0.96 (0.88, 1.04) 0.342

^*We conducted a linear regression to assess clinician empathic traits and the frequency of patient emotional expressions accounting for clustering of patients within clinicians.

^*We used a random intercept multilevel logistic regression to assess associations between clinicians’ empathic traits with types of clinician responses, accounting for clustering of emotional expressions in each visit and clustering of patients within clinicians.

There was a statistically significant association between empathic concern and the provision of information or advice in response to patient emotions, with a similar, nonsignificant trend for perspective-taking. For each 1-point increase on the 28-point empathic concern scale, clinicians had 10% greater odds of responding with information or advice. For each 1-point increase on the 28-point perspective-taking scale, clinicians had 6% greater odds of replying with information or advice.

4. DISCUSSION AND CONCLUSION

4.1. Discussion

Our study found that clinicians who reported higher levels of empathic concern did not engage in more emotional communication with their patients. Instead, clinicians with greater empathic concern received fewer patient expression of emotions, and they responded more frequently to patient emotional expressions by giving advice and information, a behavior that has previously been considered a ‘missed opportunity’ to express empathy [19,20]. Several implications of these findings are worth discussing.

Our study challenges the relatedness of two constructs referred to as empathy: clinicians’ own internal experience with patient emotion and their communication with patients in response to it. The observation that there is little or no relationship between these constructs supports a recent critique that empathy as a term is used far too broadly to be useful in creating a coherent body of research on the concept [3]. We must, therefore, distinguish studies examining self-reported clinician empathy from studies examining empathy as a form of emotional communication, and consider each body of research as separate to make sense of them.

Why might clinicians with greater self-reported empathic concern and perspective-taking traits respond to emotional situations with information and advice, rather than with more emotional communication? Clinicians who experience distress when faced with another’s misfortune may seek to alleviate it by trying to solve the situation that has initially caused emotional distress for the patient. Clinicians who internalize patients’ distress may find it more challenging to be present with the emotion,

which is what an empathic statement requires. Future research should explore this topic by using the ‘personal distress’ subscale of the IRI. Alternatively, perhaps clinicians who rate themselves highly on the IRI subscales have different personalities, where they generally think well of themselves and feel more empowered to offer advice or solve problems. If this behavior is not helpful to patients, then perhaps patients respond by providing fewer emotional expressions in return, defeated in their attempts to be heard, which explains why clinicians with higher IRI scores receive fewer emotional expressions overall.

However, it could also be that these clinicians, who also rate themselves highly on perspective-taking, rightly perceive that information or advice would be helpful to the patient, particularly in the context of longitudinal care of long-term patients. Although studies have suggested that empathic statements can reduce patient anxiety [21,22], we do not know from our study nor from previous studies whether information and advice that clinicians give to the patient under some circumstances are similarly, or potentially even more, helpful. In an earlier study, we found that in many cases of “missed empathic opportunities,” clinicians try to problem-solve with patients, presumably to address the issue causing distress [9]. If this behavior is helpful to patients, then perhaps they are satisfied with the response and feel less need to repeat themselves, which would also explain the finding of fewer emotional expressions overall. One meta-analysis examining correlates of clinician communication behaviors found that information giving was strongly associated with patient satisfaction, but this was an effect seen for the visit overall and not specifically related to how clinicians responded to particular emotional expressions [23]. The effectiveness of information-giving likely depends on the particular circumstance.

A next logical step in this research would be to explore what types of clinician communication behaviors patients perceive to be empathic. Patient ratings of clinician empathy have been associated with increased satisfaction [24,25], reduced emotional distress [25], greater patient understanding and lower unmet information needs [26,27], and higher self-efficacy and enablement [25,28]. Patient ratings of clinician empathy have been associated with clinical outcomes such as less depression [29,30], more significant improvements in migraine symptoms [31], trauma recovery [32], more favorable biological markers in the common cold [33], and lower risk of all-cause mortality [34]. These studies demonstrate that patient perceptions of clinician empathy impart clinical benefits. However, it is important to remember that neither clinician’s empathic tendency nor the empathic statements that they make are necessarily what patients perceive when they report higher levels of clinician empathy. It is not known whether the ‘spirit’ of the clinicians’ response – the emotional resonance or understanding of the patient – is more or less important than the words that are said. If we want to understand how clinicians handle patient emotion in terms of what is helpful to patients, we must be open to possibilities about how emotional engagement is conveyed and experienced.

The results of this study should be interpreted in light of its limitations. First, due to the methodology of assessing verbal behavior through audio-recorded encounters that were transcribed, we were not able to evaluate nonverbal behaviors that could provide clues to patients’ negative emotions and contribute to empathic responses by the clinician. Non-verbal behavior, including non-verbal empathy, such as gaze, has been shown to impact patient experience with the clinician [35]. Perhaps clinicians who report more empathic concern or perspective-taking express themselves more through nonverbal channels. Therefore, the results may reflect an underestimation of patient emotional expressions and clinician response in our sample. Second, as with all studies of observed communication, there is a possibility that the presence of the recorder could have altered patient or clinician behavior. Third, our measurement of empathic concern and perspective-taking required self-awareness on the part of clinicians. Although it seems likely that clinicians are best suited to gauge their empathic tendency, clinicians may overestimate or underestimate their capacity to take the perspectives of another person. Reassuringly, Hall et al. found that higher self-reported empathy was an indicator of better interpersonal accuracy [36], and Joseph et al. found that higher self-perceived emotional intelligence (EI) was associated with better job performance [37].

Fourth, our study had a relatively small sample of clinicians, which limited our assessment of several themes. For example, the clinician sample was insufficient to allow for the disintegration of the two types of empathic tendencies: empathic concern and perspective-taking. As these scales were strongly correlated, we do not know for sure whether empathic concern or perspective-taking is independently related to clinician response type. Also, our study sample is comprised of primarily white, female clinicians and African American, male patients, which may influence clinician-patient emotional communication due to racial and gender discordance. In this study, there was a trend towards female clinicians having higher empathic concern scores than male clinicians. With mostly female clinicians in our pool, there could have been less variability in empathic concern scores, making it more challenging to appreciate differences in emotional communication behaviors.

Fifth, the study sample was fixed by the design of the parent study. The study was conducted at HIV specialty centers at two urban academic medical centers located in the United States, which may limit the generalizability of our results to other healthcare settings and populations. Finally, we did not address patients’ experience or other outcomes related to clinicians’ information-giving practices in response to patient expressions of negative emotions, which limits our conclusion on whether information-giving is an effective strategy.

4.2. Conclusion

Clinicians with higher levels of self-reported empathic concern and perspective-taking tend to respond to patient emotions by giving information and advice, which they may perceive as helpful. Whether these responses are helpful to or perceived by patients as empathic is unknown. Yet, there may be circumstances where the spirit in which these responses are given may be appreciated.

4.3. Practice Implications

When faced with patients’ emotional distress, clinicians who experience more empathic tendencies should be aware of the source of their impulse to give information and assess whether this response is helpful to the patient. Communication researchers and the interpreters of this research should be careful about combining various emotion-related topics into one broad category called ‘empathy’ and should distinguish the separate effects of clinician emotional experiences, communication behaviors, and patient experiences.

Highlights.

• Clinicians who report more empathic concern (EC) and perspective-taking (PT) traits did not focus more on patient emotion

• Clinicians who report more EC and PT tended to give information or advice in response to patient emotions

• Whether this specific clinician response is helpful to the patient is still unknown