Romantic partner involvement during oncology consultations: A narrative review of qualitative and quantitative studies Patient Education and Counseling

Objective: To review the currently available research into romantic partner involvement during oncology consultations. Methods: Studies were identi ﬁ ed via database searches plus hand-searching. A narrative review was performed using the principles of Thematic, and Framework syntheses. The search strategy was performed according to the principles of PRISMA. Results: From 631 results, 18 studies were included. The ﬁ ndings indicate that romantic partners are most valued by patients when they provide emotional, practical, and informational support. It is also indicated that psychosocial and sexual concerns are rarely discussed. Couples ’ self-reported satisfaction with consultations appear related to the extent of romantic partner involvement, the roles that they enacted, and the extent to which psychosocial and sexual concerns were addressed. Conclusion: This review indicates that romantic partner involvement during clinical consultations enhances the couple ’ s experience. However, there are methodological limitations to this body of research, which are discussed in this review. Practice implications: Research to date has yet to offer an exploration of the social practices and conversational actions relating to romantic partner involvement during triadic oncology consultations. Future studies that draw upon recordings of these consultations, using methods capable of analysing situated social practices can address this gap.


Introduction
Communication in clinical consultations has mainly focused upon the patient-healthcare professional dyad. Some studies have sought to typify the encounter based on the extent to which patients are involved in treatment decisions [1][2][3], while others have examined the conversational dynamics of these consultations [4][5][6]. With research emphasising dyad-based consultations, the influence of companions has received relatively little attention. This is noteworthy as research indicates that companions attend clinical encounters regularly, for example, in 16-25 % of primary care encounters, and 36-57 % for patients aged over 60 [7]. A review of research into companion involvement during clinical encounters found that studies offered descriptive evidence, noting that companions were typically helpful in supporting patients [7]. It noted that there were likely to be differences in relational dynamics depending on the companion. A more recent review of romantic partner involvement during treatment decision-making across multiple clinical contexts, noted that partners reported regular involvement, while noting a lack of direct observations as a key limiting factor of the research to date [8]. A lack of research into companion involvement for oncology consultations was also reported [7]. This is noteworthy as research into companion attendance in oncology has indicated that companions attend more frequently than other clinical settings, with companion attendance reported to be between 64 % and 86 % [7,9,10]. However, the lack of large-scale, quantitative studies, makes this distinction uncertain.
It is well understood that the experience of cancer can affect the patient and anyone who cares for them [11]. This can be particularly pertinent for romantic partners, as cancer can have significant psychological, and sexual impact on couples, with some cancers referred to as couples' or relational cancers [12]. In this respect, romantic partners are a distinct sub-group of companions, with the experience of cancer impacting upon couples in such unique ways. This has given rise to the concept of the 'cancer couple' [13]. Cancer can have profound effects on sexual relationships [14], as well as have implications for ongoing relationship roles, such as threats to masculine and feminine identities [12,15,16].
Of the research into companion involvement during oncology consultations, a qualitative meta-synthesis was conducted, which was limited in scope by constraints relating to the definition of companions to that of unspecified female companions, a patient sample of only African American men in the USA, and within the context of prostate cancer [17]. While this review offers valuable insight into companion involvement, such constraints limit its generalisability.
Given the relative lack of research into romantic partner involvement during oncology consultations, there is value in carrying out a review of the currently available research. This is the primary aim of this review. For the purposes of this review, the term 'partner' is used. Partners are defined as spouses or romantic partners, distinct from relatives, friends, and carers.

Search strategy
After initial scoping, a search strategy was developed (Table 1). Searches were performed across six databases: Medline; PsychAR-TICLES; PsychINFO; CINAHL; EMBASE; and SCOPUS. The combined results were screened according to the principles of PRISMA [18]. Papers were selected in relation to eligibility criteria that were developed using an adapted version of the SPIDER framework [19] (see Table 2). Primaryexclusion criteriawere: Papers without focus on partner involvement; a medical area outside of cancer; and communication outside of clinical or community healthcare settings. The references in the included studies were also hand-searched for additional papers. At the stages of title, abstract, and full-paper stages of screening, a random 20 % selection of papers were double-screened by all authors. This led to the final selection of papers.

Data extraction
The data extraction, and narrative review combined inductive methods related to Thematic Syntheses, and deductive methods related to Framework Syntheses [20]. This approach was selected due to the diversity of the identified studies. The inductive approach related to the principles of thematic analysis in which themes are produced largely from the data through familiarisation, coding, and generation of themes, subject to ongoing review [21]. The deductive aspect related to the production of a provisional coding frame, constructed from key concepts found in most empirical research. For this step, each abstract was read to identify candidate codes such as methods, measures, and key findings. After this, an initial exercise of 'open coding' was performed through repeat readings of the full papers. These initial codes were grouped into higher-order narratives and incorporated into the provisional coding frame. This frame developed iteratively, leading to a final coding frame ( Table 5). The coding exercise was carried out using the qualitative data analysis software, NVivo 12 (QSR International). Codes were organised into overarching themes, with each study re-assessed against the developed coding frame. The final coding frame and themes were reviewed by all authors, with both co-authors cross-checking and validating the data extraction throughout the process.
The study assessed the quality of the included papers two ways. All studies were subjected to a standardised quality assessment using the Qualsyst tool, which allows for evaluations of qualitative and quantitative methodologies. Qualsyst scoring was crosschecked and validated by the co-authors using a random 50 % selection of quality ratings, with inter-rater reliability calculated at 85 % and 81 % for qualitative and quantitative studies respectively. Differences were discussed and resolved among all authors. Further detailed methodological appraisals were made throughout the review, which are outlined in the discussion section.

Findings
The review produced four themes: 1) Extent of partner involvement; 2) partner roles; 3) psychosocial and sexual support; and 4) satisfaction with the encounter. The extent to which partners were involved was examined through studies considering types and frequencies of partner contributions. Partner roles were examined in studies seeking to describe these roles, and how they related to the patient's experience. Psychosocial and sexual support was examined through studies that considered the amount of talk related to these topics, and perspectives relating to psychosocial and sexual support. Satisfaction with the encounter was examined by considering the ways that partner involvement related to retrospective reports of participants' satisfaction.

Extent of partner involvement
Eight studies described the extent of partner involvement. This included quantifying the amount and type of partner contributions, and qualitative descriptions of partner contributions. Four studies used qualitative methods, analysing interviews or observations [24,26,28,36], while two used quantitative methods such as surveys, or content analyses [23,30]. One study employed mixed Table 2 Inclusion and exclusion criteria.

Inclusion Criteria Exclusion Criteria
Study Types Qualitative studies, quantitative studies, that can include primary or secondary analyses of data from interviews, focus groups, surveys, or consultation observations. Systematic Reviews and meta-analyses, non-empirical reviews, commentaries, and editorials. Papers that only report on prevalence of partner presence. Papers that do not distinguish partners from other companions in reporting of findings.
Papers written in English.
Observations of, or retrospective reports about consultations. Participants Patients, their partners, and healthcare professionals, with all participants being over 18 years old.
Children as patients, or adults who lack capacity.
Companions who are not identified as partners. Partners include spouse, husband, wife, and significant others (romantic partners). Studies where all or most companions were identifiable as partners, and where analyses were explicitly related to partners (e.g. sex/ intimacy) Settings A consultation involving patient, partner, and healthcare professional.
Settings outside of health, medical, and community settings. Any health, medical, or community setting in which an oncology or oncology-related consultation is taking place.
Medical area outside of oncology.
In studies examining the extent of partner involvement, two studies reported low levels of partner involvement during clinical encounters, measured by the percentage of talk they contributed [27,28]. One study coded the consultations by participant utterance, noting that partners contributed 7% of all utterances.  This was considered low, compared with measures of patient utterances at 21.5 %, and healthcare professional utterances at 71.5 % respectively. While the other study did not perform systematic measurements, it noted that partner involvement was minimal outside of exceptional cases where the patient had marked cognitive or physical impairments. Likewise, another study that observed triadic prostate cancer encounters noted exceptional instances where partners could become so dominant as to become the primary addressee [33]. While this indicates that the extent of partner involvement can vary, these studies were unable to explain the differences. For studies examining preference for involvement, findings were mixed. When asked separately, partners and patients expressed preferences for partner involvement [23]. This finding is limited due to the use of five fixed survey responses that were transformed and reduced into three 'types' of preference; active, collaborative, and passive, which lacked any further definition.
For studies examining perceptions of partner involvement, findings were also mixed. One study into patient preferences for prostate cancer treatment indicated that partners reported limited influence upon treatment decision-making [30]. When asked to assess their influence on a scale of 0 (none) to 10 (major), the mean score of 4.8 belied the fact that most responses were at the extreme ends of this scale. The study could not explain this variation. In an interview-based study into the experiences of patients and partners during cancer consultations, partners reported that, although they were treated well by the healthcare professional, they often felt excluded [26]. One study examined the perceived extent of partner involvement by interviewing eighteen patients and seventeen partners using opportunity sampling from a single location [36]. Using an interview plus six-point survey question it was reported that partners perceived themselves as more involved than patients perceived them to be. While this study offered potential insight into these differing perspectives, it did not use matched patient-partner pairs, limiting the validity of the findings, as they were not comparing patients' accounts to their partners'. By contrast, one study into the attitudes and experiences of partners during cancer consultations claimed that successful partner participation in consultations was the product of active negotiation between the patient and partner prior to the encounter [24]. This study noted that practices such as role-clarification and discussion of expectations pre-consultation were largely appreciated and led to a better experience for couples. When considering these findings, it indicates that partner involvement tends to be low, but this can be influenced by negotiation to clarify the partner's role.
There was variation in the ways that roles were defined. Upon examination of the role descriptions across these studies, it became apparent that the diversity of labels belied the homogeneity of the roles described. Twenty-six role categories across these studies were categorised into four distinct roles: Patient advocacy; emotional support; informational support; and practical support ( Table 6).

Patient advocacy
Advocating for the patient was described in four studies, with descriptions including social communicator; middleman; and pseudo-surrogate [22,24,32,36]. These labels related to representation, support, and presenting information on the patient's behalf. Two studies coded patient advocacy directly from recorded consultations [22,32], and in the other two, the role of advocate was elicited through interviews [24,36]. In retrospective studies, data collection was significantly removed from the experience, taking place up to two years after the consultation, introducing a risk of recall bias.
In observational studies, advocacy was described as arising from ways that partners supplied information to support and protect the patient's position, such as important aspects of the patient's life, and conveying information on the patient's behalf.
One observational study reconceptualised advocacy as 'pseudosurrogacy' [32], reframing it as a spectrum, showing that different aspects of advocacy within the consultation might serve to facilitate or inhibit patient contributions as partners spoke for, as, to, or with the patient. In interview-based studies, participants described ways in which they supported the patient during the consultation, describing advocacy as a key reason for their attendance, either to act as the patient's voice, or to bring a supporting perspective [24,36]. Although the findings in this area were largely descriptive, the combined findings indicate that advocacy was largely perceived by participants in positive terms, although one study suggested it may negatively impact upon patient autonomy [32]. This claim was not explicitly present in the data, and instead was offered as a speculative comment.

Emotional support
Emotional support was described in three studies, labelled as moral supporter; counsellor; and confidant [17,24,35]. All three studies adopted a qualitative approach, drawing on retrospective accounts elicited from topic guides. Crucially, in the two studies where these guides were made available, it was evident that questions were leading, with preferences for certain responses. For example, one study guide elicited the "benefits" of partner involvement [24]. These studies reported that patients placed value upon their partner's emotional support, noting that it encouraged them to express themselves [17]. Emotional support was also described as not necessarily enacted through talk, and that 'being there' for the patient was often sufficient [24]. In one study where patients and partners recalled their experience, partners described how they downplayed the impact of cancer, offering positive appraisals to protect the patient from the uncertainties of their diagnosis [35]. For example, partners actively normalised their life with the patient, talking of 'setting aside' their own emotions for the patient's sake. Combined, these findings suggest that patients value their partners' emotional support, while also showing that it is of value to the partner. However, these findings should be interpreted cautiously due to the data collection methods.

Informational support
Informational support was described in nine studies using several terms including reporter; secretary; and information manager [22,[24][25][26]29,30,[34][35][36]. These descriptions were present in five qualitative studies [22,[24][25][26]35], and four quantitative studies [29,30,34,36]. All but one of these studies used retrospective accounts elicited from interviews or surveys. In these studies, participants placed value on partners' support in sourcing information before the consultation, providing medical history during the encounter, or helping the patient to understand medical terms [22,[24][25][26]29,30,36]. In one study, when partners were asked about their roles during the consultation, they described activities including keeping a history of the patient's health, and recording anything they saw as important to the patient's condition [35]. Communication in medical encounters regularly involves taking a medical history [40]. In supporting the patient, partners can become integral in this process as couples can coproduce the narrative. One study into the role of partners, demonstrated that their presence led to a statistically significant increase in recall when compared to the recall of patients who attended alone [34]. In this, informational support is not only valued by patients, but can benefit patients, partners, and healthcare professionals.

Practical support
Practical support was described in three studies, including descriptors such as coordinator; carer; and financial assistant; [22,35,36]. Two studies adopted a qualitative approach, with a third taking a quantitative approach [36]. All but one of these studies made use of retrospective interviews that were undertaken up to two years after the consultation, introducing a risk of recall bias. Practical support related to how partners described their involvement in the daily aspects of the patient's life, such as scheduling appointments, encouraging positive behaviours, providing transportation, and managing medication. Practical support was considered by partners as important to patients' wellbeing [22], with many partners becoming carers in an official capacity, as defined by the NHS [41].

Psychosocial, and sexual concerns
Psychosocial, and sexual concerns were identified in seven studies [16,23,24,27,28,37,38] with all but one of these studies treating these concerns as interrelated [24]. One study adopted a quantitative approach, using fixed surveys and questionnaires [23]. Four studies adopted a qualitative approach [16,24,28,37], while two adopted a mixed methods approach [27,38]. Of the mixed methods studies, one study applied qualitative coding to observations followed by a quantitative analysis of coded units [27], and the other collected qualitative data using focus groups, and performed a statistical analysis of fixed response questionnaires [38]. Critically, the focus groups took place up to three years after the consultation, making responses susceptible to recall bias. Moreover, the fixed-response surveys were informed by the focus groups, constraining the scope of the survey.
Studies approached this topic two ways; the extent to which these topics should be discussed [16,23,24,38], and the extent to which these topics are discussed [16,27,28,37,38]. It is important to note that most of these studies took place in the context of prostate cancer, involving only male patients, and female partners [27,28,37,42], with only one study reversing this configuration [38]. This bias should be acknowledged as limiting the scope of this research.

The need for psychosocial and sexual support
Four studies described the importance of discussing the psychosocial and sexual impact of cancer [16,23,24,38]. They described the challenges that couples faced throughout the cancer experience, with couples expressing a desire for psychosocial and sexual support. These studies elicited responses from patients, partners, and healthcare professionals using interviews, focus groups, and questionnaires designed to examine the extent to which participants expected such support to be offered. The need for psychosocial and sexual support was further underlined by studies that described how the sexual impact of cancer could have implications for both masculine and feminine identities [16,28,37], in which concerns relating to body image impacted directly upon patients' sexuality [16,37]. In a study of couples with hereditary breast and ovarian cancer, it was reported that 61 % of participants considered that psychological support should be offered as standard practice with a further 17 % of participants expressing that it should be an obligatory provision [38]. This sentiment was echoed in a separate study that examined the experiences of partners during oncology consultations. The participants reported that they would have benefited from psychological support, adding that they felt unable to discuss sexuality during the encounter [24].
In a study involving only healthcare professionals, participants were asked to talk about their attitudes towards discussing sex during consultations. They considered that sex needs to be discussed routinely during cancer consultations [16]. In this study, participants noted how physical changes to the patient had implications for the sexual wellbeing of couples. Healthcare professionals largely constructed their talk relating to sexual concerns in biomedical terms, relating sexuality to a constrained concept of physical functions.

Psychosocial and sexual support in consultations
Five studies examined the extent to which psychosocial and sexual concerns were discussed during consultations [16,27,28,37,38]. The studies indicated that psychosocial and sexual concerns were not discussed routinely. This contrasts with studies indicating consensus among patients, partners, and healthcare professionals relating to the need for such discussions [16,23,24,38].Despite recognising thisneed, these studies indicated that this topic was rarely raised by healthcare professionals, and was reported as an unmet need by couples [27,28,37,38].Onestudyintoprostatecancerconsultationsunderlined thisbydemonstratingthattalkrelatingtopsychosocialconcernsmade up only 5.9 % of all consultation talk [27]. Most noteworthy were the findings from the study involving only healthcare professionals [16]. Despite taking the position that discussions around sex should be routine, participants in this study also took a contradictory position, stating that they rarely discussed sex during the consultation. They justified this by claiming they lacked the necessary training, and that the topic was inapposite, arguing that it may embarrass couples. Additionally, they drew upon institutional boundaries, stating that it was not their job, noting that it should be discussed elsewhere, such as with a clinical psychologist. When considering these findings, it becomes apparent that there is a need for psychosocial and sexual support to be provided prior to treatment, and that patients and partners believe that they would benefit from this. Correspondingly, while healthcare professionals also believe this, they do not consider themselves equipped to fulfil this need. Consequently, it is a topic that receives insufficient attention.

Satisfaction with the encounter
Four studies attempted to measure satisfaction with the consultation [24,27,31,37]. The extent of partner involvement, the roles they assume, and the extent to which psychosocial and sexual concerns are discussed in consultation were considered in relation to patient and partner reports of satisfaction with the encounter.
Two studies adopted a qualitative approach [24,37], one employed mixed methods [27], and one used a quantitative approach that relied upon post-consultation survey responses [31]. While these were typically collected shortly after the experience (up to six months after), the fixed responses of the surveys limited the findings. A single study used recordings from direct observations [27], while three relied upon retrospective responses [24,31,37]. In measuring satisfaction, two studies measured patient satisfaction [24,27], one study measured partner satisfaction [31], and one measured patient and partner satisfaction [37].

Partner involvement and satisfaction
Two studies considered how the extent of partner involvement in the consultation related to reports of satisfaction [27,31]. Studies measured satisfaction using post-consultation surveys and interviews. No reliable association was reported between partner involvement and post-consultation reports of satisfaction. One study reported an inverse association between postconsultation reports of satisfaction and partner assertiveness as measured by a patient survey, completed within minutes of the consultation. However, this effect did not persist over time, with no association between the two measures after twelve weeks [27]. This stands in contrast to qualitative research reporting that partners' participation in the consultation led to positive evaluations from couples afterwards, as their involvement gave partners a sense of purpose during times when they typically 'felt helpless' [24].

Partner roles and satisfaction
One study discussed partners' roles in relation to patient and partner satisfaction [24]. It reported that informational, and emotional support roles taken on by partners were valued by patients and partners. Patients reported that their partners' emotional support gave them strength and comfort, and that this emotional support could be expressed both verbally, and through non-verbal actions such as hand holding or just 'being there'. The role of advocate was also evaluated as having a positive impact upon patients' experiences of the encounter, as patients endorsed this role while reflecting upon their experience [24].

Psychosocial, sexual concerns and satisfaction
While several studies noted the extent to which these topics were covered during consultations [16,24,27,28,37,38], only one study examined the association between the extent of psychosocial and sexual discussion, and reports of post-consultation satisfaction [37]. In this study, that happened to be the only randomised control trial, a 'treatment as usual' group was compared against an intervention-based group where couples attended a seminar in which the psychosocial and sexual impact of cancer was discussed. The half-day seminar covered topics such as managing emotions, psychological impact, and sexual dysfunction. Outcomes were measured by comparing interview responses. The intervention group reported high levels of satisfaction with the intervention and reported significantly fewer unmet needs post-consultation. It is important to note that this was reported through interview questions directed only towards the intervention group, with questions explicitly requesting evaluations of the intervention. Participants were also aware that this was something 'over and above' treatment as usual. This is of methodological concern, as equipoise was not maintained, which is essential in delivering a randomised control trial.

Discussion
This paper reviews the available research into partner involvement in oncology clinical consultations. The identified studies described the extent of partner involvement, the reported roles of partners, the prevalence of psychosocial and sexual discussion, and how these three themes might relate to reports of participant satisfaction.
The studies in this review indicate that partner involvement during oncology consultations is valued by patients. Although findings suggest partner involvement was typically low, there is a shared preference reported by patients and partners for partner involvement. Research in this area has offered insight by quantifying partner involvement during consultations, and eliciting patient and partner attitudes in relation to partner involvement.
Research has yet to examine more qualitatively, what partners accomplish during consultations, when they contribute, and how their contributions impact upon the interactional trajectory. Such research may allow for the development of a participatory framework in which patient and partner expectations relating to partner involvement can be elicited and managed by healthcare professionals.
Likewise, the ways that partners might support patients during the consultation has been studied extensively. However, the support roles of advocacy, practical, informational, and emotional support can be argued to also relate to the category-bound support activities that romantic couples are expected to provide across all contexts [43]. This underlines how romantic partners differ from non-romantic partners. Romantic couples share a degree of physical, psychological, and emotional intimacy that is typically not present in nonromantic relationships. They are likely to cohabit and therefore, spend more time together, share more of their lives, and are likely to co-experience disease in unique ways. It is therefore of concern that this review has highlighted a disconnect between the need for psychosocial and sexual support, and the prevalence of such support during consultations. With the prevailing assumption that sexuality and genitourinary concerns remain social taboos [16,33], it is apparent that research needs to develop communicative frameworks that seek to normalise these topics, challenging the normative assumptions of healthcare professionals in relation to sex, intimacy, and age [16]. In the research by Laidsaar-Powell and colleagues, one spouse noted, "We've been married for 50 years and we feel that we are a 'twosome', We don't really do things individually. Where there is suffering involved, we'd rather be in it together" [24].
The research identified by this review offered relatively little insight into the experiences of the cancer couple. Accordingly, an in-depth exploration of the unique ways that cancer can impact romantic couples is beyond the scope of this review. However, this should be considered as topic for future research.
As an overarching concern to this review, the findings in many of these studies have methodological limitations that necessitate discussion. These limitations could not be highlighted by the Qualsyst calculation, underlining the limitations of this tool.

Sampling strategies and statistics
In statistics, there are practices to ensure that analyses offer a good explanation of observations, such as a power calculation for sample sizes [44], evaluating the distribution of results to ensure an appropriate analysis [45], and evaluating statistical models to assess how well they explain findings [46]. These procedures were notably absent in the reporting of all quantitative studies in this review, apart from one study that evaluated a model of associations between sociodemographic factors, and post-consultation satisfaction. The low model evaluation scores, combined with findings that were short of statistical significance underlined that the model offered little explanatory value [27]. Statistical tests intended to explore associations between partner involvement and sociodemographic variables such as marital status and education, were performed on samples as low as 18 [36], introducing potential for statistical errors. Additionally, effect sizes were absent in all statistical reporting. Several studies also reported statistics that did not reach statistical significance as 'trends' [27,34,38].
In studies using retrospective accounts, participants were asked to reflect upon experiences that occurred up to three years earlier [24,36,38], and in one study, up to ten years earlier [29]. No study offered justification for this. It is inevitable that participants would not recall events accurately. Instead, interview questions would have made salient certain concepts, leading to reconstructions of events affected by recall bias [50]. Related to this is the social desirability bias; a desire for participants to present a favourable impression as they are asked to give an account of their experiences, and by extension, themselves [48]. In this, participants manage their selfpresentation to create a favourable impression [49,51,52]. Asking couples about their joint activities implicitly asks them to evaluate their relationship [49]. Notably, there were few reports of any adverse effects of partner involvement, indicating that responses were influenced by this bias. Related to this is the understanding that carrying out research interviews or surveys proceeds as a joint activity, in which questions can affect responses [53]. The way questions are framed, elaborated, or re-phrased is important [48]. This was highlighted when examining available interview schedules. In one study, separate interview schedules were prepared for different groups, and then responses compared [37]. In another, the generated themes could be located by examining the interview and research questions [24], underlining the production of meaning as a collaborative activity. When considered together, the limitations of recall bias, social desirability bias, and the influence of interview questions limit the findings produced from retrospective accounts.

Conclusion
Of the studies relating to partner involvement during oncology consultations, there is a tendency towards describing aspects of involvement rather than explaining them. This review suggests that patients value their partners being involved in oncology consultations, indicating that partners that provide advocacy, and practical, informational, and emotional support are valued. However, the emphasis and reliance upon participant recall also means the factors that influence partner involvement, such as how they come to be involved, when this involvement occurs, and the impact it has upon the consultation have yet to be examined.

Practice implications
It is evident that partners want to contribute during oncology consultations. Likewise, it is evident that these consultations will necessitate addressing the patient and partner as a couple. Research to date has examined the extent of partner involvement, the role of partners, and the prevalence of psychosocial and sexual support discussions in oncology consultations. However, it has over-emphasised participants' reflections. The current research does not offer an exploration of the social practices and conversational actions enacted during these encounters. Observational, micro-analytical approaches can identify what partners accomplish during consultations, when they contribute, and how their contributions impact upon the consultation. Such approaches may allow for the development of a participatory framework in which patient and partner expectations relating to partner involvement can be elicited and managed.

Funding
The lead author was funded and supported by the Economic and Social Research Council South Coast Doctoral Training Partnership, Ref: ES/J500161/1

Declaration of Competing Interest
None.