Communication study
Patterns in clinicians’ responses to patient emotion in cancer care

https://doi.org/10.1016/j.pec.2013.04.023Get rights and content

Abstract

Objective

To investigate how patient, clinician and relationship characteristics may predict how oncologists and nurses respond to patients’ emotional expressions.

Methods

Observational study of audiotapes of 196 consultations in cancer care. The consultations were coded according to Verona Coding Definitions of Emotional Sequences (VR-CoDES). Associations were tested in multi-level analyzes.

Results

There were 471 cues and 109 concerns with a mean number of 3.0 (SD = 3.2) cues and concerns per consultation. Nurses in admittance interviews were five times more likely to provide space for further disclosure of cues and concerns (according to VR-CoDES definitions) than oncologists in out-patient follow-up consultations. Oncologists gave more room for disclosure to the first cue or concern in the consultation, to more explicit and doctor initiated cues/concerns and when the doctor and/or patient was female. Nurses gave room for further disclosure to explicit and nurse initiated cues/concerns, but the effects were smaller than for oncologists.

Conclusion

Responses of clinicians which provide room for further disclosure do not occur at random and are systematically dependent on the source, explicitness and timing of the cue or concern.

Practice implications

Knowledge on which factors influence responses to cues and concerns may be useful in communication skills training.

Introduction

How do clinicians respond when patients express emotional concerns in medical interviews? And what factors in terms of patient, clinician and relationship characteristics predict how they respond?

The first question has been investigated in a number of different recent studies. For instance, Adams et al. recently audio-recorded 79 physician–patient admission encounters in two American hospitals [1]. They found that physicians responded by focusing away from emotions in every fourth occurrence of a negative emotional expression from the patient, by providing information, asking a question or changing the topic. In 43% of the responses physicians focused neither toward nor away from emotions. Finally, the physicians focused toward emotions in one third of the instances when a negative emotion was expressed.

Similar findings have been reported in other studies [2], [3], [4], [5], [6], [7], [8]. For instance, Mjaaland et al. found in a study of medical interviews in a general hospital across specialties that physicians provided room for further disclosure in response to about half of all emotional cues and concerns, but more often with reference to the medical than the affective content of the cue or concern [2]. Butow et al. reported that oncologists effectively identified and responded to the majority of informational cues; however, they were less effective in addressing cues for emotional support [3]. In another study from cancer care, Pollak et al. found that oncologists responded with empathy to 29% of patients’ expression of negative emotion [4]. Similar findings are reported in studies of non-physician clinicians [9], [10].

The second question raised above, regarding what factors may predict how clinicians respond, have been less studied. A number of variables could be considered as potential predictors of a clinician's response to patient worries, including patient factors, such as age, gender and emotional state, and corresponding clinician factors. Studies have found that female clinicians engage in more emotionally focused talk than male clinicians do [11]. Other studies have reported associations between patient attributes as well as situation factors and clinician empathic behavior [4], [12]. However we know little about the communicational dynamics, what relationship factors predict how clinicians respond to emotions. Potential relationship or communication variables include the source of the expression of concern; i.e. whether the cue or concern is initiated by the patient or by the clinician (e.g. by asking a question, or providing a facilitative remark); characteristics of the content of the cue/concern, such as the emotional explicitness of the expression; and the timing of the emotional expression in the consultation [13].

Source. Patient-initiated cues are reported to be more frequent than those initiated by clinicians [14], [15]. Studies have found that facilitative behaviors and even silence on the part of the clinician may promote the expression of cues and concerns [15], [16], [17]. Del Piccolo et al. found recently that the psychiatrists provided space for further disclosure of a concern more frequently when the concern had been initiated by the psychiatrist in the first place [18]. However, we need more research on how the initiation of cues or concerns by patient or clinician may influence the clinician's response.

Explicitness. The content of the cue or concern is characterized by an unambiguous or presumed emotional content, which may vary both in emotional intensity [3] and degree of explicitness. The distinctions between diffuse and ambiguous cues and explicit expressions of specific emotions have been discussed by several researchers [13], [19], [20], [21]. In the present study we have chosen the term explicitness to indicate the extent to which an emotion is unequivocally expressed or indirectly alluded to in the patient statement. We know little about the association between explicitness and clinician responses.

Timing. We have little knowledge on the impact of timing of the cue or concern on how clinicians respond to emotions, for instance if response patterns are different to the first cue or concern as compared with subsequent cues/concerns in the consultation.

One recently developed approach to study emotional communication in clinical encounters is the Verona Coding Definitions of Emotional Sequences (VR-CoDES), developed by a network of researchers who have met annually at the University of Verona to develop this coding system [22], [23]. A basic feature of the system is the division of patient expressions of emotion into cues and concerns. A cue is defined as a verbal or non verbal hint which suggests an underlying unpleasant emotion that lacks clarity, whereas a concern is defined as a clear and unambiguous expression of an unpleasant current or recent emotion where the emotion is explicitly verbalized [22].

The distinction is based on the finding from many studies indicating that emotions are often expressed as an indirect hint about underlying emotions, often referred to as cues [13], [24] or clues [14]. Similar distinctions have been suggested by other researchers, for instance by Suchman et al. [25].

In coding responses to cues and concerns the VR-CoDES apply two main dimensions. One dimension is whether the clinician explicitly refer back to the emotional expression or not in his or her response. The other dimension is whether the clinician provides room for further disclosure of the emotion. Providing room for further disclosure may occur by way of a brief pause or a facilitating remark (“Yes…”), an empathic statement (“That must be difficult for you”) or a question. A distinction is made between a remark or question regarding the content level of the concern (often a medical issue) or the affective level. See Table 1, Table 2 for definitions and examples.

Providing room for disclosure of emotions is an important aspect of empathy. A number of studies indicate that opening up for patient emotions and providing empathic responses may be associated with positive patient outcome in terms of reduced distress [26], [27], patient adherence [28], [29] and symptom resolution [30], [31], and the phenomenon of empathy is rooted in biological processes with potential health promoting qualities [32], [33], [34], [35].

The purpose of the present paper is to apply the Verona CoDES to investigate potential patterns in how clinicians provide space for further disclosure of patient emotion. Do clinician responses to emotion occur at random, or are they part of a pattern of associated preceding and subsequent variables? We were also interested to see whether there are differences in response patterns between different health professionals in different clinical settings, in this case exemplified by physicians in out-patient follow-up consultations and nurses in admittance interviews in an in-patient ward, both in cancer care.

The other dimension of responding to emotions studied with the VR-CoDES system, explicit vs. non-explicit responses, is not investigated in the present paper.

Section snippets

Design and sample

The present study is a secondary analysis of audiotapes of consultations between cancer patients and clinicians (physician or nurse), from an intervention study, testing out the effect of the Choice interactive tailored patient assessment (ITPA) tool [35], [36]. A convenience sample of adult patients 18 years or older with leukemia, lymphoma, testicular cancer, or multiple myeloma at two hospital wards and two outpatient clinics at Oslo University Hospital in Oslo, Norway was studied. The

Results

There were 471 cues and 109 concerns in the total of 196 consultations. The mean number of cues/concerns sequences per consultation was 2.4 cues (SD = 2.5) and 0.6 concerns (SD = 1.1). In multi-level analyses 8 cues and concerns are omitted due to lacking data on at least one variable.

Discussion

In this study of consultations with cancer patients clinicians were more likely to provide space for further disclosure of emotion if the patient had been more explicit in the expression of emotion and if the patient expression of emotion had been elicited by the clinician. There were, however, differences between oncologist and nurses. Nurses in admittance interviews in a bed-unit were five times more likely to provide space for further disclosure of cues and concerns than oncologists in

Funding

This study was funded by the Research Council of Norway grant no. 177500/V50.

Conflicts of interest

None.

Acknowledgements

We will thank Prof. Magne Thoresen for statistical assistance.

References (43)

  • L. Heyn et al.

    Effects of an interactive tailored patient assessment tool on eliciting and responding to cancer patients’ cues and concerns in clinical consultations with physicians and nurses

    Patient Educ Couns

    (2012)
  • A. Smith et al.

    Sharing vs. caring. The relative impact of sharing decisions versus managing on patient outcomes

    Patient Educ Couns

    (2011)
  • S. Davenport et al.

    How psychiatric disorders are missed during medical consultations

    Lancet

    (1987)
  • P.N. Butow et al.

    Oncologists’ reactions to cancer patients’ verbal cues

    Psycho-Oncology

    (2002)
  • K.I. Pollak et al.

    Do patient attributes predict oncologistempathic responses and patient perceptions of empathy?

    Support Care Cancer

    (2010)
  • R.A. Bell et al.

    Unmet expectations for care and the patient–physician relationship

    J Gen Intern Med

    (2002)
  • K.I. Pollak et al.

    Oncologist communication about emotion during visits with patients with advanced cancer

    J Clin Oncol

    (2007)
  • H. Ryan et al.

    How to recognize and manage psychological distress in cancer patients

    Eur J Cancer Care

    (2005)
  • R.M. Epstein et al.

    “Could this be something serious?” Reassurance, uncertainty, and empathy in response to patients’ expressions of worry

    J Gen Intern Med

    (2007)
  • Alexander et al.

    How do non-physician clinicians respond to advanced cancer patients’ negative emotions?

    Support Care Cancer

    (2011)
  • R. Uitterhoeve et al.

    Nurse–patient communication in cancer care: does responding to patient's cues predict patient satisfaction with communication

    Psycho-Oncology

    (2009)
  • Cited by (58)

    • How physicians respond to negative emotions in high-risk preoperative conversations

      2022, Patient Education and Counseling
      Citation Excerpt :

      The Verona Coding Definitions of Emotional Sequence (VR-CoDES) is a valid and theoretically underpinned coding tool focusing on physician-patient communication sequences which has been used to identify patients’ negative emotions (cues and concerns) and code the accompanying health providers’ responses to these emotions [14,15]. The VR-CoDES has been applied in various medical consultations including multiple sclerosis [7], cancer [16,17], psychiatry [18], primary care [19], and other settings. The Chinese version of VR-CoDES has been approved as a reliable tool and has been utilized in standardized patients [20] and ophthalmic consultations [21].

    • The cue-responding behavior of nurses to cancer patients: Implication to oncology nursing education

      2020, Nurse Education Today
      Citation Excerpt :

      No patients used level 3 cue expressions (clearly expressed an emotion). It may be that the pressure of time, which is closely tied to the heavy workloads of nurses, prevents patients from clearly disclosing their needs and concerns (Finset et al., 2013; Song et al., 2016). Studies have consistently indicated that poor nurse staffing has negative effects on clinical outcomes, such as the quality of patient care and meeting the emotional needs of patients (Hong et al., 2014; Lemonde & Payman, 2015).

    View all citing articles on Scopus
    View full text