Abstract
Objective
To examine using audio-recorded encounters the extent and process of companion participation when discussing treatment choices and prognosis in the context of a life-limiting cancer diagnosis.
Methods
Qualitative analysis of transcribed outpatient visits between 17 oncologists, 49 patients with advanced cancer, and 34 companions.
Results
46 qualifying companion statements were collected from a total of 28 conversations about treatment choices or prognosis. We identified a range of companion positions, from “pseudo-surrogacy” (companion speaking as if the patient were not able to speak for himself), “hearsay”, “conflation of thoughts”, “co-experiencing”, “observation as an outsider”, and “facilitation”. Statements made by companions were infrequently directly validated by the patient.
Conclusions
Companions often spoke on behalf of patients during discussions of prognosis and treatment choices, even when the patient was present and capable of speaking on his or her own behalf.
Practice Implications
The conversational role of companions as well as whether the physician checks with the patient can determine whether a companion facilitates or inhibits patient autonomy and involvement. Physicians can reduce ambiguity and encourage patient participation by being aware of the when and how companions may speak on behalf of patients and by corroborating the companion's statement with the patient.
Keywords: patient-physician communication, companion, family, communication, qualitative research, surrogacy, ethics
1. Introduction
While the “physician-patient relationship” is given primacy in both research literature and public discourse, this dyadic view represents an incomplete understanding of the modern medical encounter. Routine adult medical encounters in the USA are accompanied between 20% and 66% of the time, with patients who are older, sicker, female, and less educated being more likely to be accompanied.(1–4)
Third-parties during medical encounters have been perceived in a range of ways, from disruptive interlocutors, to beneficial social, linguistic, and cognitive extensions of the patient. (2, 4–7) Previous research has suggested that companions can play an influential or even dominant role in medical conversations.(8, 9) Certain patients in certain clinical contexts may find extensive companion participation beneficial. In ideal circumstances, an intimate companion can create a “shared mind” with the patient. In such circumstances, open communication helps the patient and companion to bring a wider range of experience and cognitive resources, enhances the patient’s ability to successfully navigate an arduous medical journey, and promotes relational autonomy.(10–12)
Some studies, however, have implicated companion presence as an inhibitor to patient-centered care. In one study, a majority of physicians surveyed reported that companions were sometimes a barrier to shared decision-making.(13) Analysis of recordings and transcripts of actual encounters supports this physician perception. Green(14) found that patients who were accompanied raised fewer conversational topics and participated less often in decision-making. Tsai(15) found that when Taiwanese patients were accompanied, they volunteered less information to physicians. Wolff(2) concluded that multiple studies have indicated accompanied patients are less verbally active and discussion is shifted toward biomedical information-giving and away from psychosocial exploration.
This mixed literature on physician-patient-companion visits is concerning because in order to preserve autonomy, the patient must maintain enough conversational control to broadly understand the biopsychosocial situation he or she faces and to make decisions without coercion.(16)
Companion influence on the medical encounter and shared decision-making is particularly salient in the context of serious and life-limiting illnesses. Accordingly, we undertook a secondary analysis of a set of audio-recordings between patients with advanced cancer, their companions who were present in the consultation and their oncologists. In particular, we focused on high-stakes discussions about prognosis and treatment choices; these tend to be emotionally charged, difficult conversations that often influence subsequent quality of life. Building on work by Coupland(8), we chose to look at both syntactical (e.g. use of “we”, “him/her” and/or “you” when referring to the patient) and interactional (e.g. to whom questions and information was addressed) behaviors to understand the conversational role taken by companions in triadic clinical encounters.
2. Methods
2.1 Study Design
This was a qualitative study utilizing a dataset of transcribed medical encounters between patients with advanced cancer, their companions, and their oncologists. Transcriptions were chosen to preserve anonymity of the physicians, whose voices might otherwise be recognized by study personnel. Non-verbal participation and other actions that could be not transcribed from audio recording were not analyzed in this paper. The data used in this study are from the observational phase of a larger randomized trial (R01CA140419) of an intervention to improve clinical communication.(17) The study was approved by the five relevant institutional review boards.
2.2 Physician Participants
Oncologists were recruited to provide three audio-recorded clinical encounters with patients identified as having advanced cancer. Practicing oncologists in the greater Rochester, NY and Buffalo areas who were currently caring for patients with solid (non-hematologic) malignancies were eligible for the study. Recorded encounters used in this analysis took place between November 2011 and March 2012.
2.3 Patient Participants
Potentially eligible patients were identified from the office visit schedules at participating physicians’ practices by a research assistant in collaboration with the physician or a practice nurse. With the physician’s permission and patient’s assent, potentially eligible patients were approached by a research assistant who described the study and obtained informed consent. Patients were eligible if they were age 18 or older, had advanced non-hematologic cancer (stage III or IV), were able to understand spoken English and complete surveys, and for whom the oncologist “would not be surprised”(18) if the patient died within twelve months.
2.4 Coding and Analysis
We used a grounded theory approach to analysis(19), applying a coding/editing template method to the transcribed audio-recordings as described by Crabtree(20). Each companion utterance in the transcriptions was initially reviewed by one author (BM). If a companion utterance was deemed to be part of a discussion about either prognosis or treatment choices, the entire discussion was read and coded. From these conversational samples, all statements by companions that in some respect appeared to “speak on behalf of” the patient were identified. Patient-physician conversations about topics other than prognosis and treatment, and conversations that took place without companion participation were not specifically coded or analyzed.
The multi-disciplinary analytic team collectively reviewed the transcripts. Collectively, the team developed a taxonomy of companions’ utterances categorized as speaking on behalf of a patient and a consensus was reached about the correctness of each categorization. Then the team developed a mutually-exclusive coding scheme to further characterize the companion’s utterances. The process of developing the codes was informed by prior research on triadic communication, in particular, Coupland(8), who utilize the terms ‘speaking for’, ‘speaking with’, ‘speaking as’, and ‘co-experiencing’ to describe companion communication for elderly patients. In addition to adapting these terms, our analysis yielded several other companion behaviors that could be coded reliably, each referring to a different conversational role taken by the companion. Thus, our coding system included a mixture of emergent and a priori categories. Once consensus was reached about the coding scheme, all of these segments were extracted and coded by two members of the team. Examples and details of the coding process are available from the authors.
Once the actual utterances were coded, three additional contextual elements were coded:
Was the companion’s utterance spontaneous, or in response to a prompt from the physician or patient?
Did the patient express explicit agreement or disagreement with the companion's statement? Explicit agreement included utterance such as “I agree”, “That’s true”, etc.
Did the physician respond to the companion’s statement? Did the response express explicit validation or rejection of the companion statement? To whom was the statement addressed (patient or companion)? We also coded whether the companion’s statement prompted the physician to change addressee from patient to companion.
3. Results
3.1 Physician characteristics
Of the 33 oncologists who were eligible and invited to participate, 23 enrolled and 10 declined. From the thirty-four accompanied visits in our sample, there were 17 unique physicians. The participating physicians were predominantly male (13/17) and identified as either White (11/17) or Asian (6/17). A slight majority of the physicians (9/17) had participated in communication skills training since medical school.
3.2 Patient and companion characteristics
A total of 49 patients had been recruited and audio-recorded (from 17 physicians, 1–4 recordings per physician) at the time of this study. One recording was performed for each participating patient. Recordings were transcribed and de-identified by a transcriptionist not involved in data collection or analysis. Thirty-four out of forty-nine visits (69.4%) included a companion who spoke at least once. Patient and companion characteristics in our sample of 34 accompanied visits (34 unique patients and companions, 17 unique physicians) are presented in Table 1, and are similar to the total study population.
Table 1. Patient and companion demographics.
Demographic and educational characteristics of the patients and companions who were present in the medical encounters analyzed. These characteristics did not differ significantly from the patients in the larger study. Three companions who were present during encounters were not enrolled in the study but agreed to be audio-recorded; detailed information is unavailable. Characteristics of those companions that could be reliably identified from the audio recording are included (*).
| Characteristic | Patient % (n=34) | Companion % (n=31) |
|---|---|---|
| Sex | ||
| Female | 58.82% (20) | 70.59% (24) |
| Race/Ethnicity | ||
| Hispanic | 2.94% (1) | 3.23% (1) |
| Non-Hispanic White | 97.06% (33) | 93.55% (29) |
| Non-Hispanic Black | 0% (0) | 3.23% (1) |
| Education | ||
| Grades 9–11 | 2.94% (1) | 6.45% (2) |
| Grade 12 or GED | 47.06% (16) | 25.81% (8) |
| College 1–3 years (incl. Associate's Degree) | 32.35% (11) | 22.58% (7) |
| College 4 years (Bachelor's) | 5.88% (2) | 25.81% (8) |
| Graduate Degree | 11.76% (4) | 19.35% (6) |
| Household Income* | ||
| $20,000 or less | 6.25% (2) | |
| $20,001 to $50,000 | 31.25% (10) | |
| $50,001 to $100,000 | 50.00% (16) | |
| Over $100,000 | 12.50% (4) | |
| *3 non-responders | ||
| Mean (std) | Mean (std) | |
| Age (years) | 62.9 (10.5) | 58.5 (10.7) |
| Companion’s Relationship to Patient* | ||
| Spouse | 66.67% (22) | |
| Daughter | 12.12% (4) | |
| Son | 3.03% (1) | |
| Sibling | 9.09% (3) | |
| Partner/significant other | 3.03% (1) | |
| Friend | 6.06% (2) |
n=34 (includes 1 male and 2 female companions who were not enrolled in the larger study but agreed to be audio-recorded)
3.3 Conversation characteristics
From 34 accompanied visits a total of 1444 companion utterances were examined. A total of 28 conversations (12 prognosis conversations and 16 conversations about treatment choices) were identified from these utterances and formed our final analytic sample. From these conversations, a total of 46 examples of companions’ speaking on behalf of patients were coded. These 46 statements were spoken by 19 unique companions.
3.4 “Speaking on behalf of”: a spectrum
Our emergent and a priori codes collectively expanded on and modified the terminology of previous research on triadic communication, and provided a rich set of examples.
Companions’ statements represented a spectrum. We introduced the notion of “pseudo-surrogacy” to describe one end of that spectrum: instances when the companion spoke as if the patient were not in the room or unable to speak for him/herself. Our study enrollment criteria required that all patients have capacity to express their wishes and make medical decisions. Yet, crucial information about their experiences and preferences were sometimes expressed by the companion without further confirmation by the patient. When a patient lacks capacity to make medical decisions (e.g., coma, dementia), a companion's “speaking as” the patient is considered “surrogacy”.(21) Because the situations in this study all involved patients who had full capacity, we adopted the term “pseudo-surrogacy.”
Pseudo-surrogacy represents an extreme in two regards: (1) the extent to which the companion made claims about the patient's views and (2) the extent to which the companion could have had first-hand experience or knowledge about the topic. Pseudo-surrogacy represents situations in which the companion claims to represent the patient's perspective but when the topic is sufficiently internal or subjective that the companion's statement could not by itself be a complete picture of the patient's own perspective. At the other end of the spectrum were companions who used “facilitating” statements to prompt or encourage patient participation but that still presupposed certain views were held by the patient. With pseudo-surrogacy at one extreme, and facilitative communication on the other, we developed a preliminary taxonomy and continuum for “speaking on behalf of.”
3.4.1 Pseudo-Surrogacy - “Speaking-as”
The important features of pseudo-surrogacy that we observed were (a) answering a question addressed to the patient, (b) describing the patient's values, beliefs, or experiences without indicating that the companion is directly quoting a previous patient statement, and (c) not inviting the patient to collaborate.
Physician [addressing the patient]: May I ask you what are your thoughts about hospice? What do you think hospice services mean?
Companion: Well, I think [Patient Name] thinks it means the end.
Physician: I see. (Patient 544)
We return to this example later to illustrate the complex set of physician and companion responses.
3.4.2 Hearsay - "Speaking-for"
Some companions made statements that were ostensibly quotes or paraphrases of something the patient said before the visit. Hearsay was sometimes prompted by the conversational context and other times was spoken without elicitation, often in order to change the topic of conversation. Patients rarely contradicted companions when prompted to elaborate or confirm their statements.
Physician: Yeah, why not, why not. You need to get stronger, I think that’s good.
Companion: She feels good enough to go out to lunch.
Physician: Um hm.
Companion: So [patient’s name] says why… why can’t I go to therapy?
Patient: Yeah, I feel really myself, you know. (Patient 558)
In the example above, the companion spoke on behalf of the patient syntactically, by using the “[Patient’s name] says” phrase to indicate hearsay. In contrast to pseudo-surrogacy (“speaking-as”), here the companion reports something that the patient has actually said, which seems credible in this example because the patient supports the companion’s hearsay in her next utterance.
3.4.3 Conflation of thoughts - "Speaking-with"
Conflation of thoughts occurred when we observed a companion expressing his or her own beliefs or preferences while suggesting that the patient shares this belief/preference. This is often done syntactically by using the first-person plural (“we”). Companions sometimes indicated that a specific conversation had previously taken place to support this agreement.
Physician: I remember [we discussed DNR forms] but it’s always good to repeat because it’s been a while since we last chatted.
Companion: Right.
Patient: Yeah.
Companion: No, we have all those [DNR] papers. No, we don't want any artificial stuff.
Physician: And I believe hospice was brought up and discussed. May I know as to what your [patient’s] thoughts are about hospice?
Companion: Well, may I answer that just for a moment because when he was in the hospital we did discuss it…(Patient 558)
The companion conflated his thoughts conversationally with the patient’s by answering a question addressed to the patient while also using the pronoun “we” (syntactical method). When the physician repeats the question to the patient later in the conversation, the companion answers again using conflation of thoughts, and the patient never verbally confirmed agreement with the plan of care.
3.4.4 Observation as an outsider and co-experiencing - "Speaking-about"
Companions in the conversations we analyzed sometimes described an event or experience at which both companion and patient were present. Companions may describe what they believe are the patient’s own views and experiences while simultaneously expressing their own experience. We noticed two distinct types of speaking on behalf of patients that fell under this description.
“Co-experiencing” statements occur when a companion discusses a topic that both the patient and companion have experienced directly.(8) Co-experiencing statements in our data appeared most often during discussions of physical symptoms and shared experiences of events. In the quote below a companion recalled a time when the patient experienced a symptom and consulted the companion about it.
Companion: The day that she was quite bothered by her belly she actually had asked me if I would feel her belly in one spot…and what it felt like was truly a knot, like when you have a knot in a muscle or something.
Physician: And that may exactly be what it was, a knot in the bowel wall squeezing.
Patient: Yeah, that’s what [another person] said. (Patient 525)
Companions also provided an “observation as an outsider.” We defined this term to mean that the companion made a statement about the patient's cognitive processes that they inferred from external observation of the patient. Feelings, values, and beliefs are internal cognitive processes that can only be inferred unless the patient explicitly describes them. However, there might be some observable signs, such as crying, which the companion might interpret as sadness (as opposed to frustration, anger or disappointment). In the example below, the spouse of a patient observed the patient’s physical response to the physician’s recommendation and drew a conclusion about the patient’s opinion of that recommendation.
Physician: It’s rare of me to tell somebody point blank you’ve got to stop [working]. However, I will say you have my permission to set limits.
Patient: Okay. [chuckles]
Companion: He can't stand the thought of [that suggestion]. I can tell by his laugh.
Physician: I know he can’t stand the thought of it. (Patient 520)
The companion’s interpretation drew mainly on the patient’s one-word response and its accompanying paralinguistic cues, yet this interpretation was taken as authoritative by the physician.
3.4.5 Facilitation - "Speaking-to"
There were companion statements in our data that encouraged a patient to express his or her own views while simultaneously speaking on behalf of the patient in some respect. These statements are more collaborative than other examples of speaking on behalf of patients. However, there remains some ambiguity even with this facilitative language. In the example below the companion encouraged the patient to talk about her symptoms but in doing so pre-supposed that the patient’s leg pain should be prioritized over other potential topics for discussion.
Companion: I’m jumping in here because I’m thinking ahead.
Physician: Yeah, go ahead.
Companion: You [patient] have been commenting frequently about the pain in your legs. Talk to the doctor about that in terms of what you’ve been feeling, what’s been going on.
Patient: Yeah. The pain in my legs is new. (Patient 521)
Additional examples of companions’ speaking on behalf of patients are presented in Table 2.
Table 2. A taxonomy for companions’ speaking on behalf of patients.
Additional quotes from our sample that illustrate the proposed categories of companions’ speaking on behalf of patients
| Form of speaking on behalf of patients |
Illustrative quotes |
|---|---|
| Pseudo-surrogacy (“speaking-as”) | Patient: I didn’t think I would ever go back to work. Companion: It was really scary for him. (pt 518) Physician: Independent living with facilities to call if you need help? Companion: Right. And they can get that if they need it or want it. And I think they will cover any aids or additional help. I just don’t think he’s at that point. He’s too aware. Physician: Yeah. Companion: But I know it’s really hard on mom. There are places for her to go, a facility, but just in the apartment I think it’s… (pt 519) |
| Hearsay (“speaking-for”) | Companion: Her big concern is the spots on her liver. She wants those out. Physician: Yes, so do we. So do we. And there were a number of people in the clinical trial that put this combination on the map who did have spots in the liver. Patient: Okay. (pt 531) Patient: We’re driving [to our vacation]. Companion: She said she didn’t want to have a reaction. Physician: I know. Patient: I don't want anything to go wrong just – Physician: Yeah, because you are driving, right? You’re not flying. (pt 530) |
| Conflation of thoughts (“speaking-with”) | Physician: I know. Do you want to stop altogether? With the chemo? I feel – Patient: No, no. We’re going fine. Physician: Okay. You want to keep going, okay. Companion: No, we want to keep going. (pt 500) Physician: So, tell me more about your thinking. What were you asking about in terms of the schedule? Companion: Well, we just thought that if the next CAT scan if the tumor was way littler that he could go every six weeks and then every 12 weeks and then stop. Is that eventually what happens or we’re just, we want it to happen faster than…? Physician: The hope is that we’ve killed the living parts of the tumor and that, just like a weed plant it would, you know, you kill the plant but the plant doesn’t vanish. The trunk is still there and the roots, you can still see them. And it may slowly wither. Patient: Okay. (pt 518) |
| Observation as an outsider (“speaking-about”) | Companion: From my point of view, though, I want to say since he started the chemo I’ve begun to see… it was rough at first but his overall attitude, whatever, has just improved. Physician: Okay. Well, it sounds like - Companion: A lot. Patient: And the idea that all this, whatever I go through is having a positive effect is…it’s a lot easier to do, it’s an incentive. (pt 511) |
| Co-experiencing (“speaking-about”) | Physician: Is it vomiting… is it only after you eat? Patient: Pretty much. Physician: Okay. And is it like one episode and you’re done or is it repeated? Companion: Well, he feels so good after he vomits. He’s back to his normal self and he feels well. Patient: Almost like I get energized. After I’ve eaten it sits. I know it’s sitting there. And the next thing I know I’m running to the bathroom. (pt 505) |
| Facilitation (“speaking-to”) | Physician: And taking out the intubation, people can feel like they’re suffocating even with an intubation procedure. So. Patient: Just thinking about it makes me feel like I’m suffocating. Companion: So, we don’t want to attempt resuscitation? Patient: Yeah. I think we… yes, we can check that, do not resuscitate. Companion: Not gonna … attempt cardiopulmonary resuscitation or not at all? Patient: No. (pt 520) |
3.5 Initiation of and responses to speaking on behalf of patients
Nearly all of the time, when companions spoke on behalf of the patient, those statements were volunteered by the companion, rather than being solicited by physician or patient. When physicians responded to such companion statements, they predominantly continued to address the patient rather than the companion. More salient to us, however, was how infrequently patients explicitly accepted or confirmed companion statements. Patients only explicitly accepted companions’ speaking on behalf of them 17.39% (8/46) of the time. The remainder of the time patients responded ambiguously or did not respond at all. In contrast, physicians explicitly accepted companion statements as true 34.78% (16/46) of the time. This suggests physicians are often willing to believe companion statements during these important discussions even in the absence of explicit patient confirmation. Data from this contextual description of companion statements are presented in Table 3.
Table 3. Conversational context of companion utterances.
Descriptive statistics characterizing physician and patient responses to companion statements that spoke on behalf of patients. Determination of these characteristics was based on an analysis of the two conversational turns preceding and following companion statements. Addressee was inferred from verbal cues since transcripts could not account for non-verbal actions such as changing eye contact.
| Utterance characteristic | % (n=46) |
|---|---|
| Companion initiated | 97.83% (45) |
| Patient response | |
| Explicit acceptance | 17.39% (8) |
| Ambiguous response | 45.65% (21) |
| No response | 36.96% (17) |
| Physician response | |
| Explicit acceptance | 34.78% (16) |
| Ambiguous response | 56.52% (26) |
| No response | 8.70% (4) |
| Change of addressee from patient to companion | 17.39% (8) |
| Physician explicit request for confirmation from patient | 6.52% (3) |
As a whole, statements spoken on behalf of patients were nearly always accepted into the conversational common ground without explicit corroboration or clarification by either patient or physician. Physicians requested explicit confirmation from the patient for only three companion statements. We looked particularly closely for explicit disagreement, upset or conflict that arose from companions’ speaking on behalf of patients, and found no examples.
3.6 An emerging pattern of how companions speak on behalf of patients
With these findings in mind, we can return to the original example of pseudo-surrogacy quoted above (patient 544). By examining further conversational turns, we hope to exemplify what we observed to be a common pattern: (a) the physician implicitly accepting the companion statement as true and (b) the patient providing either an aligning or ambiguous follow-up. The patient statements rarely contradicted the companion statement, leaving considerable ambiguity about whether this was passive acquiescence or true agreement.
Physician: May I ask you [to patient] what are your thoughts about hospice? What do you think hospice services mean?
Companion: Well, I think [Patient Name] thinks it means the end.
Physician: I see.
Companion: Yeah.
Patient: Well, it’s a preparation for the end.
Companion: But, see, so he’s got that embedded in his mind. I don't feel that way. I feel it’s just something to give him comfort and when the time is necessary we’ll get to it but I don't think we still need it yet. (Patient 544)
While the patient has a conversational turn to express his opinion, it is ultimately expressed using the phraseology of the companion (“the end”). The companion and physician then continue to converse about the topic for twelve turns before the patient speaks again. This unaddressed pseudo-surrogacy by the companion coupled with an ambiguous statement by the patient that is not explored demonstrates one extreme in which the patient’s voice was not heard, in spite of her having had a conversational turn. This conversation then proceeded as though the patient’s statement had not occurred.
4. Discussion and conclusion
4.1 Discussion
Companions often spoke on behalf of patients during discussions of prognosis and treatment choices, even when the patient was present and capable of speaking on his or her own behalf. Companion statements appeared to fit along a spectrum ranging from speaking as if the patient lacked capacity (“speaking as”) to inviting full patient involvement in the conversation. Specifically, companions spoke on behalf of patients by: answering questions directed to the patient, conflating the companion's own values or observations with those of the patient, asking facilitating questions to the patient that presuppose certain values, or by describing events that both patient and companion experienced together and assuming a shared interpretation. Companions employed a combination of syntactical and interactional choices. For example, a particular syntactical choice, such as use of the first person plural (“we”), was often paired with an assertion of conversational control, such as answering a question addressed to the patient. In our data these ways of speaking on behalf of patients appeared to fit along a spectrum.
In the conversations we coded, physician and patient responses to companions’ statements were usually ambiguous. That is, while patients sometimes responded with additional information, they usually did not explicitly clarify whether the they agreed or disagreed with the companion statement, leaving ambiguity about whether the rest of the conversation represented passive acquiescence or true agreement. In those circumstances when patients did correct a companion, it was done quickly and without drawing attention to the disagreement. We consider this finding important in the context of serious illness; patients who are seriously ill generally wish to have their values and preferences enacted, yet they often assume a more passive role in patient-physician conversations compared to those who are less seriously ill.(22)
Pseudo-surrogacy and other forms of speaking on behalf of patients represent an ethical challenge for physicians.(23) While a companion’s perspective might sometimes allow him or her to communicate a patient's own values and concerns more articulately than the patient can, at other times it can be unclear whether it is the patient's or the companion's perspective that is being expressed, and upon what information the companion is basing the claim. One study suggests that physicians may interpret a companion’s speaking on behalf of a patient as an indication that the patient is incapable of speaking for him/herself.(3) Our data suggest that patients sometimes allow their companions to speak unilaterally for them through passivity – by only providing infrequent and terse responses. Even when pseudo-surrogacy is an infrequent event, it may be silencing patients who are uncomfortable interrupting or contradicting companions or returning to a topic already perceived to have already been discussed. Physicians did not frequently challenge companions after they spoke on behalf of patients. Physicians did, however, appear to account for this phenomenon by continuing to address the patient even after a companion would answer a question that was directed to the patient or to both parties. This may be a deliberate strategy to promote patient participation without explicitly challenging a companion. However, the effect of such an indirect approach is uncertain.
In our view, it is during discussion of the “internal life” of the patient that physicians’ should guide companions into a supportive rather than dominant role in order to respect the patient's autonomy. The patient’s cognitive processes are both dynamic and inherently unknowable by a third party unless expressed explicitly. Furthermore, medical information can be complex, preferences unstable, communication challenging, and cognition impaired in the face of serious illnesses; thus perspectives of patients may change and are not always well-aligned with those of their family members, friends and clinicians.(24–28) Furthermore, patients may be afraid to speak up for fear of losing their physician’s good will.(29)
Our study has a number of limitations. Our taxonomy applies to patients with advanced cancer and cannot be considered representative of physician-patient-companion interactions in general. Furthermore, while our study indicates that speaking on behalf of patients is common enough to warrant attention, it does not attempt to quantify how often companions speak on behalf of patients, nor how those conversations compare with those conversations on similar topics in which the companion did not participate. Future research should consider under what conditions speaking on behalf of patients is most likely (e.g., types of relationship, patient personality, or clinical scenario).
We also could not explore the outcomes of such utterances beyond the audio-recorded conversation in this study. In future studies, participant self-reports may be able to clarify the degree to which companions’ statements on behalf of the patient were truly facilitative, or whether they interfered with the patient expressing his or her values, preferences and concerns. Further studies can also confirm and expand upon our taxonomy by using a larger, more diverse sample of patients.
4.2 Conclusion
This study is a preliminary exploration of the ways companions may express values and experiences on behalf of the patient. We have elaborated a spectrum of ways in which family members and friends can interact with patients and physicians in ways that ultimately lead to (or interfere with) patient-centered care. The results of this study can guide efforts to help family members, friends and other companions be better advocates for their loved ones during clinical encounters by avoiding “pseudo-surrogacy” and engaging in more facilitative behaviors. It also can guide physicians to recognize companions’ communication behaviors that might silence the patient’s voice; it can point the way to respectful clarifications to ensure that patients’ perspectives are elicited and their preferences enacted(25), especially in situations in which these discussions have important implications for patient autonomy and quality of life.
4.3 Practice Implications
Physicians can reduce ambiguity and encourage patient participation by being aware of when and how companions may speak on behalf of patients and by corroborating the companion's statement with the patient. This can be done by interrupting the visit momentarily to seek the patient’s perspective on a companion’s statement, perhaps by asking the patient, “Have you had a chance to talk together about this?,” “Tell me what you’ve been thinking,” or saying, “It’s only natural for people to have different perspectives about these kinds of things, especially as things change.” Open communication will hopefully result in shared attentional focus, voicing of multiple perspectives, greater patient participation in deliberations and decisions, determination of the degree to which there is agreement among the parties, and avoidance of implicit assumptions about the patient’s values and preferences.(10, 12, 30) Executing this simple action, like any new communication behavior, requires awareness and training: physicians must recognize in-the-moment when a companion is speaking on behalf of a patient; decide whether this particular instance will influence decision-making; and ultimately create a “multipartial alliance” between all parties, while still prioritizing the interests of the patient.
ACKNOWLEDGMENTS
Funders: This research was funded by the National Cancer Institute (Grant R01CA140419). The principle investigator is Ronald M. Epstein, MD.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Contributors: Paul Winters aided in the construction of the tables presented in this paper.
The authors report no conflicts of interest.
I confirm all patient/personal identifiers have been removed or disguised so the patient/person(s) described are not identifiable and cannot be identified through the details of the story.
REFERENCES
- 1.Labrecque MS, Blanchard CG, Ruckdeschel JC, Blanchard EB. The impact of family presence on the physician-cancer patient interaction. Soc Sci Med. 1991;33(11):1253–1261. doi: 10.1016/0277-9536(91)90073-l. [DOI] [PubMed] [Google Scholar]
- 2.Wolff JL, Roter DL. Family presence in routine medical visits: a meta-analytical review. Soc Sci Med. 2011;72(6):823–831. doi: 10.1016/j.socscimed.2011.01.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Laidsaar-Powell RC, Butow PN, Bu S, Charles C, Gafni A, Lam WW, et al. Physician-patient-companion communication and decision-making: a systematic review of triadic medical consultations. Patient EducCouns. 2013;91(1):3–13. doi: 10.1016/j.pec.2012.11.007. [DOI] [PubMed] [Google Scholar]
- 4.Shields CG, Epstein RM, Fiscella K, Franks P, McCann R, McCormick K, et al. Influence of accompanied encounters on patient-centeredness with older patients. The Journal of the American Board of Family Practice / American Board of Family Practice. 2005;18(5):344–354. doi: 10.3122/jabfm.18.5.344. [DOI] [PubMed] [Google Scholar]
- 5.Levine C, Zuckerman C. The trouble with families: toward an ethic of accommodation. Ann Intern Med. 1999;130(2):148–152. doi: 10.7326/0003-4819-130-2-199901190-00010. [DOI] [PubMed] [Google Scholar]
- 6.McDaniel S, Campbell T, Hepworth J, Lorenz A. Family-Oriented Primary Care. 2 ed. New York, NY: Springer; 2005. Chapter 16. Looking death in the eye: facilitating end-of-life care and the grieving process; p. 261. [Google Scholar]
- 7.Street RL, Gordon HS. Companion participation in cancer consultations. Psychooncology. 2008;17(3):244–251. doi: 10.1002/pon.1225. [DOI] [PubMed] [Google Scholar]
- 8.Coupland N, Coupland J. Relational frames and pronominal address/reference: the discourse of geriatric medical triads. In: Sarangi S, Coulthard M, editors. Discourse and social life. 1 ed. Routledge; 2001. pp. 207–229. [Google Scholar]
- 9.Karnieli-Miller O, Werner P, Neufeld-Kroszynski G, Eidelman S. Are you talking to me?! An exploration of the triadic physician-patient-companion communication within memory clinics encounters. Patient EducCouns. 2012;88(3):381–390. doi: 10.1016/j.pec.2012.06.014. [DOI] [PubMed] [Google Scholar]
- 10.Epstein RM, Gramling RE. What is shared in shared decision making? Complex decisions when the evidence is unclear. Med Care Res Rev. 2013;70(1 Suppl):94S–112S. doi: 10.1177/1077558712459216. [DOI] [PubMed] [Google Scholar]
- 11.Epstein RM, Entwistle VA. Capacity and shared decision-making in serious illness. In: Quill TE, Miller FG, editors. Palliative Care and Ethics: Common Ground and Cutting Edges. Oxford University Press; 2013. [Google Scholar]
- 12.Epstein RM. Whole mind and shared mind in clinical decision-making. Patient EducCouns. 2013;90(2):200–206. doi: 10.1016/j.pec.2012.06.035. [DOI] [PubMed] [Google Scholar]
- 13.Shepherd HL, Tattersall MH, Butow PN. Physician-identified factors affecting patient participation in reaching treatment decisions. J Clin Oncol. 2008;26(10):1724–1731. doi: 10.1200/JCO.2007.13.5566. [DOI] [PubMed] [Google Scholar]
- 14.Greene MG, Majerovitz SD, Adelman RD, Rizzo C. The effects of the presence of a third person on the physician-older patient medical interview. J Am Geriatr Soc. 1994;42(4):413–419. doi: 10.1111/j.1532-5415.1994.tb07490.x. [DOI] [PubMed] [Google Scholar]
- 15.Tsai MH. Who gets to talk? An alternative framework evaluating companion effects in geriatric triads. Commun Med. 2007;4(1):37–49. doi: 10.1515/CAM.2007.005. [DOI] [PubMed] [Google Scholar]
- 16.Entwistle VA, Carter SM, Cribb A, McCaffery K. Supporting patient autonomy: the importance of clinician-patient relationships. J Gen Intern Med. 2010;25(7):741–745. doi: 10.1007/s11606-010-1292-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hoerger M, Epstein RM, Winters PC, Fiscella K, Duberstein PR, Gramling R, et al. Values and options in cancer care (VOICE): study design and rationale for a patient-centered communication and decision-making intervention for physicians, patients with advanced cancer, and their caregivers. BMC cancer. 2013;13:188. doi: 10.1186/1471-2407-13-188. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Moss AH, Lunney JR, Culp S, Auber M, Kurian S, Rogers J, et al. Prognostic significance of the "surprise" question in cancer patients. J Palliat Med. 2010;13(7):837–840. doi: 10.1089/jpm.2010.0018. [DOI] [PubMed] [Google Scholar]
- 19.Creswell JW, Creswell JW. Qualitative inquiry and research design : choosing among five approaches. 3rd ed. Los Angeles: SAGE Publications; 2013. [Google Scholar]
- 20.Crabtree BF, Miller WL. Doing qualitative research. 2nd ed. Thousand Oaks, Calif.: Sage Publications; 1999. [Google Scholar]
- 21.Beauchamp T, Childress J. Principles of Biomedical Ethics. 6 ed. New York, NY: Oxford University Press; 2008. A Framework of Standards for Surrogate Decision Making; p. 135. [Google Scholar]
- 22.Say R, Murtagh M, Thomson R. Patients' preference for involvement in medical decision making: a narrative review. Patient EducCouns. 2006;60(2):102–114. doi: 10.1016/j.pec.2005.02.003. [DOI] [PubMed] [Google Scholar]
- 23.Epstein RM. The ambiguity of personhood. The American journal of bioethics : AJOB. 2013;13(8):42–44. doi: 10.1080/15265161.2013.804746. [DOI] [PubMed] [Google Scholar]
- 24.Cassell EJ, Leon AC, Kaufman SG. Preliminary evidence of impaired thinking in sick patients. Ann Intern Med. 2001;134(12):1120–1123. doi: 10.7326/0003-4819-134-12-200106190-00012. [DOI] [PubMed] [Google Scholar]
- 25.Epstein RM, Peters E. Beyond information: exploring patients' preferences. Jama. 2009;302(2):195–197. doi: 10.1001/jama.2009.984. [DOI] [PubMed] [Google Scholar]
- 26.Fried TR, O'Leary J, Van Ness P, Fraenkel L. Inconsistency over time in the preferences of older persons with advanced illness for life-sustaining treatment. J Am Geriatr Soc. 2007;55(7):1007–1014. doi: 10.1111/j.1532-5415.2007.01232.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Lichtenstein S, Slovic P. The Construction of Preference. New York, NY: Cambridge University Press; 2006. [Google Scholar]
- 28.Slovic P. The construction of preference. American Psychologist. 1995;50(5):364. [Google Scholar]
- 29.Frosch DL, May SG, Rendle KA, Tietbohl C, Elwyn G. Authoritarian physicians and patients' fear of being labeled 'difficult' among key obstacles to shared decision making. Health Aff. 2012;31(5):1030–1038. doi: 10.1377/hlthaff.2011.0576. [DOI] [PubMed] [Google Scholar]
- 30.Epstein RM, Hadee T, Carroll J, Meldrum SC, Lardner J, Shields CG. "Could this be something serious?" Reassurance, uncertainty, and empathy in response to patients' expressions of worry. J Gen Intern Med. 2007;22(12):1731–1739. doi: 10.1007/s11606-007-0416-9. [DOI] [PMC free article] [PubMed] [Google Scholar]