Patient Education and Counseling 85 (2011) 60–67

Communication Study

Examining the validity of the unitary theory of clinical relationships: Comparisonof observed and experienced parent–doctor interaction

Bridget Young a,*, Jo Ward b, Mary Forsey a, Katja Gravenhorst a, Peter Salmon a

a University of Liverpool, UKb Wrexham County Borough Council, UK

A R T I C L E I N F O

Article history:

Received 3 February 2010

Received in revised form 19 August 2010

Accepted 29 August 2010

Keywords:

Clinical relationships

Methods

A B S T R A C T

Objective: We explored parent–doctor relationships in the care of children with leukaemia from three

perspectives simultaneously: parents’, doctors’ and observers’. Our aim was to investigate convergence

and divergence between these perspectives and thereby examine the validity of unitary theory of

emotionality and authority in clinical relationships.

Methods: 33 audiorecorded parent–doctor consultations and separate interviews with parents and

doctors, which we analysed qualitatively and from which we selected three prototype cases.

Results: Across the whole sample doctors’ sense of relationship generally converged with our

observations of consultation, but parents’ sense of relationship diverged strongly from each. Contrary

to current assumptions, parents’ sense of emotional connection with doctors did not depend on doctors’

emotional behaviour, and parents did not feel disempowered by doctors’ authority. Moreover, authority

and emotionality were not conceptually distinct for parents, who gained emotional support from

doctors’ exercise of authority.

Conclusions: The relationships looked very different from the three perspectives. These divergences

indicate weaknesses in current ideas of emotionality and authority in clinical relationships and the

necessity of multisource datasets to develop these ideas in a way that characterises clinical relationships

from all perspectives.

Practice implications: Methodological development will be needed to address the challenges posed by

multisource datasets.

� 2010 Elsevier Ireland Ltd. All rights reserved.

Contents lists available at ScienceDirect

Patient Education and Counseling

jo ur n al h o mep ag e: w ww .e lsev ier . co m / loc ate /p ated u co u

1. Introduction

Consensus statements and policies emphasise the importanceof patient–doctor relationships in healthcare [1,2]. Good clinicalrelationships are associated with better outcomes for patients[3,4], and both patients and doctors value their mutual relation-ships [5,6]. The distress for patients and doctors that canaccompany the faltering or breakdown of their relationships [7–10] is further evidence of their importance to both parties.

Though clinical relationships are important they are inherentlydifficult to define. Metaphors such as bond, attachment or alliance[11–13] are common, their materiality suggesting entities which‘hover’ between people and are observable by third parties.However, relationships are not material entities which exist

* Corresponding author at: Division of Clinical Psychology, School of Population

community and Behavioural Sciences, University of Liverpool, Whelan Building,

Brownlow Hill, Liverpool, L69 3GB, UK. Tel.: +44 151 794 5525;

fax: +44 151 794 5537.

E-mail address: byoung@liv.ac.uk (B. Young).

0738-3991/$ – see front matter � 2010 Elsevier Ireland Ltd. All rights reserved.

doi:10.1016/j.pec.2010.08.027

independently of the people experiencing them [14,15]. Observa-tions of how two people interact may reveal little of how they eachexperience their relationship or the functions it serves [16–18],and the implications extend beyond theoretical and methodologi-cal questions. Interventions to improve clinical relationships haveoften produced ‘benefits’ that communication experts haveobserved in dialogue between patients and doctors but thatpatients have not experienced [19–23]. Dissatisfaction withcurrent concepts and methods for investigating relationshipshas led to recent calls for improvements [20,24–26]. Given thatrelationships exist both in the subjective experiences of eachparticipant and in the interaction between them, researchers havespecifically advocated the development of multidimensionalapproaches [16,24,27,28] combining data from different sources– participant accounts and observations of dialogue – which havehitherto usually been analysed separately.

We report a preliminary investigation of the feasibility and valueof such a multisource, triangulated analysis and its implications fortheory. The report is drawn from a qualitative study in which weaudiorecorded consultations between paediatric haematologists/

B. Young et al. / Patient Education and Counseling 85 (2011) 60–67 61

oncologists and parents of children being treated for cancer, andthen interviewed the doctors and parents about their consultations.We examined whether and how combining these data couldilluminate two concepts – emotional engagement and authority.Based largely on single perspective studies, these have becomefundamental in understanding clinical relationships, and are widelyrecognised as clinically important [29,30].

Studies have criticised doctors for focussing on biomedicalconcerns at the expense of emotional ones and have encouragedthem to attend more to the personal aspects of patients’ lives. Thispractice has become regarded as fundamental to building anemotional connection [31–36] and to patients’ sense of being caredfor as unique individuals. Clinical relationships have also beenexamined extensively through the lens of authority and power.Though some of this literature has recognised the positive aspectsof medical authority and expertise [24,37], most has criticiseddoctors for constraining, dominating or undermining patients indoctor-centred consultations and has encouraged them to adoptmore egalitarian and more patient-centred practice [38,39].

However, different views of the same relationship (i.e. fromdoctor, patient and observer of their dialogue) do not alwaysconverge in the way unitary theory would predict [18,40] so it isunclear how robust these concepts would be when examinedsimultaneously from multiple perspectives. With the ultimate aimof stimulating methodological and theoretical development forstudying relationships, we first explored whether we couldrecognise authority and emotional engagement in consultationdialogue. Then, we linked and compared the different data sourceswithin cases by comparing each consultation with the doctor’sinterview and the corresponding parent interview. Finally, weexamined the extent to which our findings were consistent orinconsistent with current formulations of authority and emotionalengagement.

2. Method

2.1. Participants and recruitment

The data are drawn from the Rapport study, a longitudinalqualitative study which examined parent–practitioner relation-ships in the care of children (aged 1–12 years, median 4 years) withacute lymphoblastic leukaemia receiving treatment in one of sixUK specialist paediatric oncology centres. The reported analysesare drawn from the study’s first data collection point whichcomprised: (a) 33 parent–doctor consultations audiorecorded 2–13 weeks (median 6) after the child’s first admission to hospital;(b) interviews with 50 parents (30 mothers, 20 fathers) and (c) 11doctors (8 males, 3 females) who were party to the consultations.Parents were interviewed 0–11 weeks (median 2) following theaudiorecorded consultation; doctors were interviewed on oneoccasion to coincide with a recorded consultation for at least one ofthe families in their care. The study was approved by a UK ResearchEthics Committee.

2.2. Procedure

The researchers prompted doctors to audiorecord their con-sultations and later conducted semi-structured interviews withparents and doctors. Interviews were conversational and informedby a topic guide with prompts about what each party usuallysought from, and how they reacted to, the other. Besides this broadfocus, the researchers also reviewed consultation recordingsbefore each interview to identify key exchanges and to developspecific prompts about these index consultations. Most doctorscared for several families in the study. As each doctor wasinterviewed on just one occasion the portion of their interviews

about consultations focussed on the one or two families withwhom they had most recently recorded consultations.

2.3. Analysis

Analysis was interpretative [41]. Working with three datasources is challenging and there is little precedent so we could notrely on any single qualitative paradigm. We examined consulta-tions first, focussing on what could be inferred from whatparticipants talked about and what they did not discuss. Thenwe examined the corresponding interviews. To explore authority,we examined the extent of patient versus doctor-centeredness,focussing on the degree of participation and influence exerted byeach party, including the number of words that each spoke. Foreach consultation we then examined the participants’ interviewaccounts of their sense of their own and the other party’s authorityand influence or, conversely, disempowerment and passivity. Foremotionality, we examined the relative prominence of biomedicalor psychosocial concerns in the consultation dialogue. Wescrutinised interview accounts for attitudes towards discussionof biomedical and psychosocial content within consultations, andfor the role of this content in participants’ accounts of relationship.The analysis was iterative; while consultations initially provided acontext for analysis of the interviews, we cycled between all threedata sources.

Procedurally, our analysis followed the general principles ofconstant comparison [42]. One member of the team led a process of‘cycling’ between the developing analysis and new data, with othermembers developing the analysis by periodic review and discus-sion. Transcripts were read several times to develop codes atmultiple levels, beginning with descriptions of turn-by-turninteractions, progressing to written narratives of each case. Thesenarratives incorporated analytical reflections on participants’global stances in relation to authority and emotionality, and onparticipants’ roles within the consultation. We compared withinand across the different data sources and cases to link the data andto attend to convergences and divergences.

3. Results

Doctors varied considerably in their communication styles andtherefore the degree to which consultations were doctor or family-centred and biomedical or emotional in focus. As we mostly foundconvergence between doctors’ consultation dialogue and theirinterview accounts, we draw on both in describing their styles.

In general, doctors’ styles varied on a spectrum. We describethis spectrum by distinguishing three styles, which encompass themain areas of variation. ‘Style A’ doctors were authority figureswith biomedically focussed, doctor-centred consultations and acontrolling and unemotional stance in interviews. Style B’ doctors– the most frequent in our sample – were managerial andunemotional in consultations but used more subtle forms ofcontrol (e.g. making indirect references to time constraints) thanStyle A doctors. In interviews, Style B doctors presented authority,but subtly, describing strategies to manage the consultation andsupport parents emotionally. Nevertheless, their engagement withfamilies was managerial rather than personal and emotional. ‘StyleC’ doctors – the least frequent – aimed for egalitarian and family-centred consultations. In interviews they described themselves asconfidant-like figures who emphasised emotional concerns andthe importance of personal engagement in clinical relationships.

While doctors’ consultation dialogue and interview accountsconverged, parents’ data indicated a more complex picture. Parents’interview accounts of the doctor–parent relationship usuallydiverged markedly from the relationships as evident in the doctors’data. Most commonly, doctors’ interviews and consultation data

Box 2. Case A

Consultation

Authority

Dr A’s use of closed questions, such as, ‘‘do you remember those

conversations?’’ and avoidance of open questions conveyed a controlling

stance. His frequent references to time restrictions constrained the

consultation and controlled its pace and sequencing:

‘‘I really want to just, over, in just a few minutes is to go back through the

conversation that we had at diagnosis, and during, while your admission

erm, and to talk about erm what’s going to happen in the future’’.

Emotionality

Only three of Dr A’s many utterances could be classified as emotional and

these were brief (e.g. ‘‘it’s nice to see him back on his feet’’) or initiated by the

nurse rather than by Dr A himself. Whereas Dr A avoided personal or emotional

language, he explicitly demonstrated his commitment by describing instru-

mental features of his care. For example, he referred to how he had worked

behind the scenes to obtain results of a test performed several weeks previously:

‘‘Ever since <the holidays> I’ve been emailing <the trial co-ordinating

centre> and trying to find out where they’re up to’’.

The doctor’s interview

Authority

Dr A explained his exercise of control and responsibility in the consultation:

‘‘It’s a complicated field of medicine, I don’twanna beasking parents what to

do, I’dratherbeknowingwhattodoandtellingthemwhatI thinkweshoulddo’’.

He labelled his approach to parents as ‘‘fairly scripted’’ and acknowledged

that the ‘‘comprehensive discussion’’ he had did not leave ‘‘much

opportunity for a family ((. . .)) to ask about something else’’. Indeed, he

pre-empted parents’ questions by giving a ‘‘complete story’’ and ‘‘reinforcing

information’’ explaining that:

‘‘If you try giving half the story or a little bit of the story then the questions

that you get will be what you haven’t given’’.

Emotionality

Dr A contrasted his enjoyment of the instrumental aspects of his work with

the emotional challenges that relationships with families could present:

‘‘One of the attractions of dealing with paediatric haematology, it’s what

we’re in there to do and erm, no I don’t find it difficult it anymore really, unless

the families make it difficult ((. . .)) by being angry, by being whatever’’.

Indeed, he described emotionality in his relationships with parents at diagn-

osis as something he worked to overcome by explaining the biomedical and ins-

trumental aspects of his work, namely that effective treatments were available:

‘‘By the time they see us they’ve had some of the upset and actually what

we’redoing is being able togo beyond that and say look ((. . .)) we are expecting

tobeable togetyourchild better ((. . .)). What we’reabletodo((. . .)) is tosay,not

only this is what you’ve got, but this is what we’re gonna do about it’’.

However, elements of emotionality were present in Dr A’s account of his

relationships with the children in his care:

‘‘I think the relationship with the child is important ((. . .)) what I want is to

give the child confidence that we know what we’re doing, and we’re going to

make them better’’.

The father’s interview

Authority and emotionality

Mr A’s account did not separate emotional aspects of the consultation from

his experience of Dr A’s instrumental and controlling approach. He clearly did

not feel disempowered by Dr A’s authority and spoke of how he could ‘‘ask Dr

A about anything ((. . .)) anything whatsoever’’. In repeatedly describing how

the doctor helped him feel ‘‘in safe hands’’, ‘‘at ease’’ and ‘‘that there was light

at the end of the tunnel’’, Mr A presented the relationship as intensely

emotional. Furthermore, his need to feel comforted and cared for was met by

the Dr A’s instrumental, expert and controlling stance:

‘‘The first time I ever met him ((. . .)) he came across, he knew what he was

B. Young et al. / Patient Education and Counseling 85 (2011) 60–6762

implied an unemotional parent–doctor relationship in whichdoctors were overtly (Style A: authority figures) or subtly controlling(Style B: managerial doctors). Despite this, parents mostly experi-enced these doctors as emotionally supportive and approachable,though a few parents felt these doctors to be unsupportive anddistant. Conversely, although some parents experienced theconfidant-style doctors (Style C) in the way they presented, thatis, as personal and emotionally supportive, a few parents describedrelationships with these doctors as unemotional and remote. Wealso found that some parents’ interview accounts even divergedfrom how they, themselves, presented in consultation, for exampledescribing themselves as passive in the face of doctors’ authoritywhile nevertheless being active participants in the consultation.Finally, while we were able to distinguish data relating to authorityand emotionality in the doctors’ data, authority and emotionalitywere closely intertwined in parents’ interviews.

To illustrate these patterns of divergence and convergence, wepresent detailed evidence from three ‘prototype’ [43] cases; that is,cases that focused the main relationship patterns that we hadidentified. First, to illustrate how these patterns arose across thespectrum of variation in doctors’ style, we identified three doctorswho represented the key elements of each style (A, B and C). Thesedoctors saw two-four families each (eight in total); from these weidentified cases that best illustrated the patterns within the widerdataset. Because divergence between parents’ experience ofrelationship and the relationships evident in the doctors’ datawas the most frequent pattern in our dataset, we chose cases thatparticularly exemplified this pattern in selecting the presentedprototype cases. For each case, we outline to what extent authorityand emotionality were evident in the consultation, doctorinterview and parent interview. While the doctor styles weretherefore our starting point for selecting cases, we emphasise thatthese are not the focus of our report. Rather, the doctor stylesprovide entry points into the data to illustrate the nature of therelationship processes across the wider dataset and the spectrumof doctor styles.

We present details and illustrative quotations from each case inBoxes 2–4, and explain the format of our data presentation in Box 1.

3.1. Case A

Overview: Dr A’s control over consultation and avoidance ofemotional dialogue (matched by Mr A’s conspicuous silence in theconsultation) was reflected in his emphasis on his instrumentalrole and authority when interviewed (Box 2). Nevertheless,although these data sources suggested a clinical relationshipwhich lacked emotional and personal engagement and excludedparent participation, Mr A described Dr A as highly approachableand as personal and emotionally supportive.

3.1.1. The consultation

This consultation, recorded 9 weeks after Mr and Mrs A’s 4-year-old son was first admitted to hospital, lasted 15 min. Present

Box 1. Data presentation

We have used pseudonym initials to identify doctors and parents and used

masculine pronouns to refer to doctors, as fewer were female.

The data concern a rare condition: to preserve anonymity we have altered

minor background details of children and consultations (though none relevant

to our argument have been changed).

For data extracts, ((. . .)) indicates omitted speech, square brackets [] enclose

speech that was overlapping or unclear, (x s) indicates hesitation of given

duration and < > enclose text entered to replace names or aid interpretation.

talking about ((. . .)), very confident ((. . .)) and he gave us the confidence that he

had ((. . .)). Really I can’t praise him enough, what, so good I feel, still feel the

same now ((. . .)) about everything. Every time I go and meet him, like last week

err I just felt better ((. . .)) through seeing him’’.

Indeed, he experienced aspects of Dr A’s controlling style, such as restricting

information, as caring and supportive:

‘‘That was something actually he told us at the beginning ((. . .)) that there’s

so much to take in, he didn’t tell us everything ((. . .)) because there was so

much, you wouldn’t take everything in ((. . .)) he didn’t bombard us with err

too much info ((. . .)) which is good ((. . .)) ‘cause I couldn’t have took anything

in at the time’’.

Mr A interpreted Dr A’s behaviour ‘‘<he> always comes over to us, asking

how we’re doing’’, as signs of genuine commitment to the family. Therefore,

even seemingly small gestures on the part of Dr A had considerable emotional

significance for Mr A.

Box 3. Case B

The consultation

Authority

Dr B clearly controlled the interaction but his use of authority was more

subtle than Dr A’s. In asking the parents if they had any ‘‘burning’’ questions

at the start he implied that his time was limited and that he expected

important questions, not trivial ones. He also repeatedly instructed Mr and

Mrs B about the need for vigilance when they returned home:

‘‘And that’s why monitoring temperature is important ((. . .)) and why if

there is temperatures that you liaise quickly ((. . .)) with us’’.

His direct reference to his own responsibility for the child’s care and his

request to be kept informed about all problems cemented his authority:

‘‘Again you know I’m responsible for K7’s care and if there’s something

you’re not happy with, even if it’s you know something which you may think

is not necessarily a medical issue ((. . .)) but maybe a nursing issue, I’d still

like to know about it’’.

However, elsewhere Dr B used anecdotes, examples and the language of

‘suggestion’ and ‘norms’ rather than explicit ‘instruction’ to guide Mr and Mrs

B on their roles and responsibilities - ‘‘we’d probably say, you know, not

venturing off <on holiday> at this moment in time’’ – which allowed him to

be clear about what he expected, whilst softening his use of authority.

Mrs B’s presentation in the consultation was complex. She referred to

herself throughout as passive (e.g. ‘‘It’s all been done for us really, so we’ve

not got to think about anything too much’’) and obedient (e.g. ‘‘I wouldn’t

even give an <indigestion remedy> without asking’’). However, this

contrasted sharply with her active participation in the consultation.

Emotionality

While Mrs B made many attempts to bring topics that were personally or

emotionally significant into the consultation, in general, Dr B avoided

personal and emotional talk. The following extract is one of 10 references Mrs

B made to her son in which she emphasised his personal characteristics:

Mrs B because he has got a very high pain tolerance hasn’t he?

Mr B Yeah

Mrs B He doesn’t- he doesn’t complain

Dr B Yeah

Mrs B So you know he did- he did unbelievably really

Dr B OK so does that start to make sense?

In this example, Dr B disregarded Mrs B’s references; at other times he

responded minimally or provided medical guidance. While appearing relaxed

and informal he remained impersonal and business-like; for example he

apologised ‘‘that I may not always have as much time for you ((. . .)) as I

would like to’’ and offered his secretary as a means of arranging a longer

consultation. Similarly, whereas Mrs B was emotionally expressive through-

out, referring to herself as ‘‘worrying’’ and ‘‘desperate’’, Dr B rarely used

emotional language, even when talking about highly emotive topics, such as

risks and relapse:

Dr B there is a small i.e. about a five per cent chance

Mrs B Yes

Mr B Mm

Dr B of leukaemia coming back whilst we are on the treatment. If it

comes back whilst we are on treatment that can be very hard to get rid of

((. . .)) so that is very unusual but if it does happen it’s

Mr B Mm

Dr B it ain’t great

Mrs B Right

Dr B And the other risk while we’re on treatment is always this risk of

infection. As I was saying we probably unfortunately lose one child a year

through overwhelming infection despite you know all the best efforts

[from] everybody involved

The doctor’s interview

Authority

Dr B was explicit about using his authority in consultations. He described

being ‘‘proactive in giving information’’ and emphasised how doctors needed

to ‘‘train and educate’’ parents. Nevertheless, his account of how he managed

consultations combined authority and control with helping parents feel

involved and supported. While he described firm rules about the length of

clinic consultations - ‘‘not the time to sit down and discuss’’ - he also

described how he offered appointments outside clinic, thereby demonstrat-

ing his accessibility. Like Dr A, Dr B referred to pre-empting parents’

questions:

‘‘I tend to ask the questions which I think they need to ask and try and give

them answers’’.

However, in contrast to Dr A (who emphasised how few questions parents

asked), Dr B spoke of the importance of listening to parents:

B. Young et al. / Patient Education and Counseling 85 (2011) 60–67 63

were Dr A, Mr and Mrs A and child, as well as a nurse and juniordoctor. For the first 7 min Dr A spoke in a monologue of 1189words, punctuated only by 11 affirmations (e.g. ‘‘yeah’’) and 15sub-vocal responses (e.g. ‘‘mmm’’) from the parents. Compared tothe 2016 words spoken in total by Dr A, Mr and Mrs A uttered only520 words. Furthermore, only 186 were in dialogue with thedoctor; most were in dialogue with the nurse or child. Mr Aparticipated minimally, uttering only 60 words and asking thedoctor just one question, compared to his wife’s nine questions.Therefore, the cumulative effect was a highly doctor-centredinteraction. Throughout the consultation Dr A’s focus wasinstrumental – that is, on biomedical care and the child’s progresssince diagnosis rather than emotional considerations. Similarly, Mrand Mrs A used no emotional language, and made no attempt toengage the doctor in emotional talk about their own individualneeds or those of their child.

3.1.2. The doctor’s interview

Dr A described how he expected to control the course of theconsultation and to take responsibility for decision-making. Hemade no reference to the importance of allowing parents to voicetheir own questions; rather he described how the experiencegained in his years of practice allowed him to anticipate whatparents wanted to know and to give appropriate explanations, andso avoided the need for parents to voice questions. Dr A seemed tovalue certain elements of emotionality in his relationships withchildren but, in his relationships with parents, emotionalityappeared as a challenge to be borne rather than as an aspect ofwork that he valued or enjoyed. On the whole, his descriptions ofthe emotional aspects of his work stood in contrast to hisenjoyment of the instrumental aspects.

3.1.3. The father’s interview

The emotional and instrumental aspects of the consultationwere largely indistinguishable in Mr A’s description of therelationship. Indeed, Dr A’s instrumental and controlling approachappeared in Mr A’s account as a source of intense emotionalcomfort to him.

3.2. Case B

Overview: Dr B’s subtly controlling dialogue converged with themanagerial stance that he presented in interview (Box 3). It alsoconverged with Mrs B’s characterization of him as an expertauthority figure, albeit a highly approachable one. However, thiscase illustrates the divergence that could occur between parents’observed behaviour in consultation and how they describedthemselves in interview. Mrs B’s presentation of her role in herson’s care in interview (and also in consultation) as limited, passiveand obedient contrasted markedly with her active participation inthe consultation.

3.2.1. The consultation

This 70-min consultation was with the parents of a boy agednine admitted to hospital 3 weeks earlier. Both parents (Mr andMrs B) were present, but not the child. This consultation waspermeated by social and personal conversation, initiated largely bythe mother, and its tone was relaxed. Dr B’s exercise of authoritywas softer than Dr A’s but he nonetheless clearly controlled theconsultation. His strategies ranged from subtle references to timeconstraints to repeated and overt instruction on how Mr and Mrs Bshould care for their son at home. Mrs B’s presentation wascomplex. While she took a particularly active role, contributing4701 words compared to Dr B’s 9497 and asking 25 questions, shemade explicit comments throughout in which she presentedherself as passive and obedient. Mrs B also repeatedly brought

‘‘I think the most important thing ((. . .)) is to listen to what parents are

saying. And to give them opportunity to ((. . .)) ask the questions and

appropriate time to sit down and answer and time to recap’’.

Emotionality

Dr B described ‘‘recognizing the doubts in parents’ minds’’, ‘‘reassuring’’

and ‘‘giving hope’’, and explained how parents ‘‘need a lot of intensive

support’’ indicating that he regarded emotion as an important aspect of his

work with parents. He particularly emphasised emotion when describing his

priorities for breaking the bad news of diagnosis:

‘‘Saying in the first breath that it’s leukaemia ((. . .)) absolutely confirming

that and also very high in the priority is giving the parents hope and

reassuring them that childhood leukaemia in a vast majority of kids is a very

treatable condition’’.

He also described his enjoyment of the scientific aspects of his work, ‘‘a

subject which is sort of the forefront of science and moving forward so I

don’t think you’d ever have an opportunity to get bored in it’’, and his long-

standing, emotional commitment to paediatrics:

‘‘I went to medical school wanting to do paediatrics ((. . .)) and I’ve always

loved working with children’’.

The parent’s interview

Authority and emotionality

Mrs B focused overwhelmingly on Dr B’s and his team’s expertise in the

clinical care of her son and cited many forms of evidence of this expertise:

‘‘Nothing will be misread because ((. . .)) they go by, you know, their

experience of the drugs they deal with, the samples that they use and so

he’ll get exactly the treatment really that he needs’’.

Consistent with this, she presented her role as passive and described how

she effectively entrusted responsibility for her son’s care to the hospital staff:

‘‘They’re just holding the reins ((. . .)) I just have to attend and make sure

he’s there and make sure he takes his tablets you know, there’s not a great

onus on me with that, the care they’re actually giving’’.

Mrs B noted Dr B’s adeptness in how he ‘‘picked up on’’ her son’s needs

‘‘straight away’’ and described him as highly approachable, responsive and

patient:

‘‘He’s very, you know, very receptive to anything we ask. He answered

everything we asked ((. . .)) I think he’s fabulous actually ((. . .)) He’s very

approachable ((. . .)) you don’t feel like you’re talking to a top consultant

((. . .)) He talks quite slowly ((. . .)) and he gives it time to sink in and he sort of

says it again in a different way’’.

However, beyond emphasising Dr B’s expertise and approachability, Mrs B

said little that suggested emotional and personal engagement in her

relationship with Dr B.

Box 4. Case C

The consultation

Authority

Dr C adopted an egalitarian approach. With questions like ‘‘do you think

she’s better now since she’s been off steroids?’’ Dr C invited Mrs C to bring

her personal and intimate knowledge of her child into the consultation and

acknowledged Mrs C’s expertise as a mother. He also mirrored her language;

for example he twice used her own colloquial term ‘‘mardy’’ to refer to the

child’s moodiness. Nevertheless, Dr C was clearly controlling the consulta-

tion. His favourable comments on Mrs C’s vigilance when her daughter

showed signs of infection signalled his authority in judging parental

competence. By following these comments with a question about the child’s

appetite he also controlled the pace and sequencing of the consultation.

Dr C I think you did exactly the right thing ((. . ..)) but erm at the moment it

looks like things are settling down so [err]

Mrs C [yeah]

Dr C I’m not too worried about that (2 s) erm (3 s) and how is she eating

at the moment?

Emotionality

By commenting ‘‘she doesn’t like being in hospital does she?’’ Dr C

demonstrated his interest in Mrs C’s daughter as an individual. Moreover,

his emotional language – ‘‘I’m very pleased with how things are going

with her’’ and ‘‘we’re very happy about that’’ – when commenting on the

child’s treatment explicitly suggested his emotional involvement.

Similarly, Dr C repeatedly anticipated the difficulties Mrs C might have

in caring for her daughter and showed personal interest in Mrs C

experiences, but in her responses Mrs C largely avoided focussing on her

own emotional needs:

Dr C They tend to be a little bit more irritable ((. . .)) but it’s it’s nothing to

compare [with what] you’ve been through already

Mrs C Yeah

Similarly, when Dr C followed his question about how Mrs C was coping by

referring to her daughter and asking ‘‘does she sleep during the day at all

or?’’ Mrs C rejected the implied invitation to describe the impact of her

daughter’s behaviour on herself. Instead, she acknowledged that being in

hospital had been disruptive for her daughter, ‘‘coming in here messed her

up a little you know’’ but emphasised that this would be temporary ‘‘I’m sure

she’ll get back round [to her like] ((. . .)) usual self again’’.

The doctor’s interview

Authority

Dr C explicitly described the parent–doctor relationship as a partnership,

‘‘we share the responsibility between all of us’’ and referred to how he liked

parents to be ‘‘very vocal’’ and to challenge doctors. Similarly, he tried to

‘‘empower’’ parents to explain the illness to their child. Nevertheless, he said

that doctors need to ‘‘manage’’ parents’ expectations about their child’s well

being, suggesting that he felt responsible for shaping how parents

understand their child’s illness and treatment:

‘‘I have to occasionally remind them. They don’t often want to be

reminded that we are fallible but they sometimes need to be reminded’’.

Emotionality

Establishing emotional relationships with parents – and children – was

central to Dr C’s stance. He described how relationships were ‘‘what makes

me tick’’ and explained how crucial this aspect of role had been in attracting

him to paediatric oncology:

‘‘In paediatrics what attracted me to oncology was that you get to know

families for a long period of time, you get to know the children and you get

the opportunity to follow them over years. That long-term relationship I

found much more valuable and satisfying than the 15 minutes you spend

with an acute <respiratory> patient’’.

He spoke of time with children and parents at diagnosis as investment

which ‘‘actually pays off later on’’ and which helped him care for children

and their families as individuals: ‘‘you sometimes can recognize this is just

how the person deals with things’’. Emotion was also explicit when he spoke

of caring for individual families: ‘‘I think I’m going to enjoy looking after

<that family>’’.

Parent’s interview

Authority and emotionality

Mrs C’s account of her relationship with Dr C was unemotional and

impersonal. She described doctors as ‘‘all the same’’ and, when asked what

she thought of Dr C when she first met him, she replied:

‘‘Now apart from please get my daughter better ((. . .)) that was about it

((. . .)) And nothing apart from that, just that I wanted them to do the best

they could to get, obviously get <my daughter> better; that was it. That

was it, yeah ((. . .)) just please just try and get my daughter better’’.

B. Young et al. / Patient Education and Counseling 85 (2011) 60–6764

personal or emotional subjects into the consultation which, ingeneral, Dr B disregarded or responded minimally, maintaining hisbiomedical focus on treatment and the child’s care at home. Incontrast to Mrs B’s emotionally expressive language, Dr B rarelyused emotional language.

3.2.2. The doctor’s interview

Dr B was explicit in describing his use of authority inconsultations but this was balanced by his account of efforts tohelp parents feel involved and supported. He also described thevalue of giving parents opportunities to ask questions and of helpingthem feel listened to. Dr B talked of emotion as an importantcomponent of his work. However, consistent with his business-likeconsultation, he neither emphasised the importance of parents’personal characteristics in his work, nor described his own feelingsabout the relational aspects of his encounters with them.

3.2.3. The parent’s interview

Mrs B focused overwhelmingly on Dr B’s expertise in the clinicalcare of her son. Though she also emphasised Dr B’s approachabilityand adeptness, there was little evidence of emotional and personalengagement in her account of the relationship.

3.3. Case C

Overview: Dr C’s consultation dialogue matched his descriptionof himself as empowering parents and attending to their emotionalneeds (Box 4). Conversely, Mrs C’s unemotional account andpresentation of her relationship with the doctor, as not extending

Mrs C praised Dr C for his ability to deliver information, ‘‘he answered

what I wanted to know ((. . .)) I understood it straight away as soon as he

said it, and he did it nice and clear’’ and commented on how the staff helped

‘‘you understand what you’ve got to understand and tell you what you need

to know’’. However, she commented on no other aspect of Dr C, and provided

no indication of recognising or valuing his personal manner and emotional

engagement. Similarly, her description of the hospital as ‘‘like a big family

[everyone is] is really nice everyone talks to each other’’ indicated, not the

emotional intimacy of a family, but the sense of a diffuse network in which

personal and emotional engagement had no particular significance.

B. Young et al. / Patient Education and Counseling 85 (2011) 60–67 65

beyond the instrumental care of her daughter, was consistent withher unemotional and passive role in consultation. In this way thedoctor’s and mother’s accounts converged substantially with theirbehaviour in the consultation. However, the resulting divergencebetween the two parties’ accounts of the relationship that this caseillustrates was striking.

3.3.1. The consultation

This 20-min consultation included a 3-year-old girl, her mother,the doctor and a nurse and was recorded 5 weeks after the child’sfirst admission to hospital. Though Dr clearly controlled pace andsequencing, his egalitarian approach was also visible throughout.Nevertheless, Mrs C remained passive, speaking only 538 words toDr C’s 2943 and asking only four questions, which centred on herdaughter’s medical care rather than her personal characteristicsand needs. As well as focusing on biomedical issues, such as testresults, treatment and the child’s care at home, Dr C demonstratedpersonal interest and emotional care of Mrs C and her daughter.However, Mrs C rejected his apparent invitations to comment onhow she was coping, responding instead with comments about theimpact of the illness and treatment on her daughter.

3.3.2. The doctor’s interview

Dr C emphasised the importance of empowering parents andlikened the parent–doctor relationship to a partnership. He wasexplicit about the value of establishing emotional relationshipswith parents, and emotion was central when he described enjoyingthe relational aspects of his work generally and with regard toparticular families.

3.3.3. The parent’s interview

Whereas Mrs C valued Dr C’s professional commitment andexpertise, her account of their relationship was unemotional andimpersonal; she remarked on his expertise in explaining herdaughter’s illness and treatment but provided no indication ofrecognising or valuing Dr C’s personal manner and emotionalengagement.

4. Discussion and conclusion

4.1. Discussion

The novel and critical feature of our study is that we examinedclinical relationships simultaneously in the consultation and fromeach participant’s perspective. The divergence between thesedifferent data sources showed that single data sources areinadequate for describing such relationships and, in turn, thelong-established principles for understanding clinical relation-ships that have generally emerged from studies of single datasources are problematic. When triangulated between data sources,both authority and personal and emotional engagement provedmore complex than usually portrayed.

The relationships that we observed often diverged from thosethat parents described, so that unitary accounts of emotionalengagement or balance of power were impossible. For example,

Case A illustrates a common pattern where the consultationimplied an unemotional parent–doctor relationship, but the parentdescribed a relationship that was intensely emotional, comfortingand reassuring. A few doctors had a more overtly emotional style,reflecting their stated concern with parents’ personal andemotional experience and their valuing of their own emotionalengagement with parents. Although some parents showed a senseof emotional connection with such doctors (as some parents didwith the less emotionally oriented doctors), divergence couldoccur in this type of consultation also. For example, emotion wasstrikingly absent from Mrs C’s account of her relationship with Dr Cdespite the priority he gave to emotional engagement inconsultation and interview.

The balance of authority also defied unitary description.Whatever stance doctors took, parents’ accounts bore no consistentrelationship to these. For example, Dr C’s egalitarian approach inconsultation and interview could not be detected in Mrs C’s accountof the authority figure on whom she depended. Similarly, Dr A’sstrongly controlling behaviour was absent from Mr A’s descriptionof a reassuring and approachable expert. Even individual parents’accounts were not always unitary, as Mrs B illustrates. Despite heractive participation in the consultation, she described subordinatingher role to the authority of the doctor and hospital.

The interview stances of doctors converged strongly with theirconsultation behaviour, a pattern which others using differentmethods have observed [44,45]. This contrasts sharply with thedivergence between parents’ sense of relationship and ourobservations of consultations. That is, whereas doctor’s sense ofrelationship was largely visible in consultations, parents’ sense ofrelationship was not. This might be explained by disparity betweenparents and doctors in the specificity of their accounts, which inturn reflects the asymmetry of the relationship. However,regardless of the explanation, comparing the different perspectivesshows the error of assuming that certain behavioural transactionsin consultation are synonymous with parents’ sense of relation-ship. It also indicates that inductive analyses of interview accountsare crucial for describing the relationship that parents or patientsexperience.

Paradoxically, such analyses have had limited influence inresearch underpinning current formulations of clinical relation-ships [28,46], meaning that these formulations might reflect theviews of clinical communication researchers and educators morethan those of patients. In particular, while current formulationsidentify emotionality and authority as conceptually discretefundamental dimensions of relationship, the parents in this studydid not. Authority and emotionality were not conceptually distinctin parents’ experience: parents felt emotionally supported bydoctors’ deployment of authority and even seemingly controllingdoctor behaviour did not prevent parents experiencing anapproachable clinician. Parents’ accounts deviate in two otherimportant ways from current assumptions in clinical communica-tion theory. Firstly, although overt personal and emotionalbehaviour is often regarded as fundamental to building relation-ship [31–36], parents’ interviews frequently showed that it wasnot necessary, consistent with other reports [47]. Secondly, whilestudies of analogue patients [48,49] or those with less seriousillnesses [49,50] have suggested that doctors’ authority and controlleaves patients dissatisfied or disempowered, the parents westudied often experienced doctors’ authority as a source ofconfidence and reassurance. This echoes reports that thoseexperiencing or fearing life-threatening illness prefer to delegateauthority to doctors [18,51–53]. It seems that in constructing therelationships they need, these parents, like patients in previousstudies, disregard behaviour which communication researchershave deemed important [54] and experience relationships in waysthat fall outside current theoretical formulations [55–57].

B. Young et al. / Patient Education and Counseling 85 (2011) 60–6766

4.2. Conclusion

Researchers have previously advocated the use of multidimen-sional approaches to study clinical relationships and have arguedthat triangulating patients’, observers’ and clinicians’ perspectivesmay be particularly important for conceptualising relationships[20,24,28]. Our findings vindicate these suggestions. The relation-ships that we described were not unitary or objective. Instead, theyusually looked very different from the perspectives of the third-party observer, parent or clinician. Thus, a complete account ofclinical relationships requires that all three perspectives beconsidered simultaneously. Our findings indicate the need forconceptual development of current formulations of emotionalityand authority that have arisen from previous, more restricted,examinations of relationships. Our presentation focussed onprocesses that were most frequent in our dataset and we neededto include data from each of the three perspectives to illustrate theseprocesses. The necessity for ‘depth’ therefore limited ‘breadth, andwe could present only three cases. A further limitation is that ourstudy was based in a specialised clinical context, and focussed onparents rather than patients. Future research in other contexts willprovide insights on how multi-perspective formulations of clinicalrelationships are shaped by context.

4.3. Practice implications

Further research is needed that triangulates between theperspectives of doctors and patients (or parents) and observationsof the communication between them. However, dialogue andinterview data have rarely been linked and compared in the waywe have done, so methodological development beyond thatattained here will also be needed. This will have to incorporatemore integrated analyses. To examine relationships from threeperspectives simultaneously will, however, be difficult. Apart fromthe practical problems in working with a large set of linked data,there are also epistemological questions [58,59] about the status ofthe knowledge derived from each data source and the significancethat each should be given. Our analyses, though iterative,simplified the task by starting with the doctors’ consultationand interview data and using these as the point of reference forexamining divergence and convergence with other data sets inrelation to two specific concepts. This leads to an importantlimitation – our linking of data identified divergences that exposedproblems with these concepts but it did not enable new conceptualdevelopment. This will require analyses that do not prioritise onespecific data source. Instead, future analyses will need to integrateand synthesise the three sources to produce theory that transcendsthe different perspectives. Therefore, methodological developmentbeyond that achieved here will need to enable, not justtriangulation but integration of the perspectives of doctors andpatients and observations of the communication between them.

I confirm that all personal identifiers have been removed ordisguised so the persons described are not identified and cannot beidentified through details of their story.

Acknowledgments

The study was part supported by Cancer Research UK (grantnumber C19412/A6913). We are grateful to all parents andpractitioners for supporting or participating in the study.

References

[1] Brunett PH, Campbell TL, Cole-Kelly K, Danoff D, Frymier R, Goldstein MG,et al. Essential elements of communication in medical encounters: theKalamazoo consensus statement. Acad Med 2001;76:390–3.

[2] von Fragstein M, Silverman J, Cushing A, Quilligan S, Salisbury H, Wiskin C. UKconsensus statement on the content of communication curricula in under-graduate medical education. Med Educ 2008;42:1100–7.

[3] Stewart M, Brown J, Weston W, McWhinney I, McWilliam C, Freeman T, et al.Transforming the clinical method. Thousand Oaks, CA: Sage; 1995.

[4] Di Blasi Z, Harkness E, Ernst E, Georgiou A, Kleijnen J. Influence of contexteffects on health outcomes: a systematic review. Lancet 2001;357:757–62.

[5] Mainous 3rd AG, Goodwin MA, Stange KC. Patient–physician shared experi-ences and value patients place on continuity of care. Ann Fam Med2004;2:452–4.

[6] Stokes T, Tarrant C, Mainous 3rd AG, Schers H, Freeman G, Baker R. Continuityof care: is the personal doctor still important? A survey of general practitionersand family physicians in England and Wales, the United States, and TheNetherlands. Ann Fam Med 2005;3:353–9.

[7] Stokes T, Dixon-Woods M, Windridge KC, McKinley RK. Patients’ accounts ofbeing removed from their general practitioner’s list: qualitative study. BritMed J 2003;326:1316.

[8] Stokes T, Dixon-Woods M, McKinley RK. Breaking up is never easy: GPs’accounts of removing patients from their lists. Fam Pract 2003;20:628–34.

[9] Hareli S, Karnieli-Miller O, Hermoni D, Eidelman S. Factors in the doctor–patient relationship that accentuate physicians’ hurt feelings when patientsterminate the relationship with them. Patient Educ Couns 2007;67:169–75.

[10] Smith AK, Buss MK, Giansiracusa DF, Block SD. On being fired: experiences ofpatient-initiated termination of the patient–physician relationship in pallia-tive medicine. J Palliative Med 2007;10:938–47.

[11] Fuertes JN, Mislowack A, Bennett J, Paul L, Gilbert TC, Fontan G, et al. Thephysician–patient working alliance. Patient Educ Couns 2007;66:29–36.

[12] Van der Feltz-Cornelis CM, Van Oppen P, Van Marwijk HW, De Beurs E, VanDyck R. A patient–doctor relationship questionnaire (PDRQ-9) in primary care:development and psychometric evaluation. Gen Hosp Psychiat 2004;26:115–20.

[13] Salmon P, Young B. Core assumptions and research opportunities in clinicalcommunication. Patient Educ Couns 2005;58:225–34.

[14] Salander P. Bad news from the patient’s perspective: an analysis of the writtennarratives of newly diagnosed cancer patients. Soc Sci Med 2002;55:721–32.

[15] Salmon P, Young B. Dependence and caring in clinical communication: therelevance of attachment and other theories. Patient Educ Couns 2009;74:331–8.

[16] Mead N, Bower P, Hann M. The impact of general practitioners’ patient-centredness on patients’ post-consultation satisfaction and enablement. SocSci Med 2002;55:283–99.

[17] Epstein RM, Franks P, Fiscella K, Shields CG, Meldrum SC, Kravitz RL, et al.Measuring patient-centered communication in patient-physician consulta-tions: theoretical and practical issues. Soc Sci Med 2005;61:1516–28.

[18] Mendick N, Young B, Holcombe C, Salmon P. Distinguishing responsibilityfrom ownership in decision-making about treatment for breast cancer: trian-gulation of consultation with patient and surgeon perspectives. Soc Sci Med2010;70:1904–11.

[19] Brown JB, Boles M, Mullooly JP, Levinson W. Effect of clinician communicationskills training on patient satisfaction. A randomized, controlled trial. AnnIntern Med 1999;131:822–9.

[20] Stewart M. Continuity, care, and commitment: the course of patient–clinicianrelationships. Ann Fam Med 2004;2:388–90.

[21] Griffin SJ, Kinmonth AL, Veltman MW, Gillard S, Grant J, Stewart M. Effect onhealth-related outcomes of interventions to alter the interaction betweenpatients and practitioners: a systematic review of trials. Ann Fam Med2004;2:595–608.

[22] Fallowfield L, Jenkins V, Farewell V, Saul J, Duffy A, Eves R. Efficacy of a CancerResearch UK communication skills training model for oncologists: a random-ised controlled trial. Lancet 2002;359:650–6.

[23] Shilling V, Jenkins V, Fallowfield L. Factors affecting patient and cliniciansatisfaction with the clinical consultation: can communication skills trainingfor clinicians improve satisfaction? Psycho-Oncol 2003;12:599–611.

[24] Mead N, Bower P. Patient-centredness: a conceptual framework and review ofthe empirical literature. Soc Sci Med 2000;51:1087–110.

[25] Katz AM, Alegria M. The clinical encounter as local moral world: shifts ofassumptions and transformation in relational context. Soc Sci Med2009;68:1238–46.

[26] Salmon P, Young B. From communication skills to skilled communication:clinical communication as creative art. Med Educ, in press.

[27] Zandbelt LC, Smets EM, Oort FJ, Godfried MH, de Haes HC. Satisfaction with theoutpatient encounter: a comparison of patients’ and physicians’ views. J GenIntern Med 2004;19:1088–95.

[28] Zoppi K, Epstein RM. Is communication a skill? Communication behaviors andbeing in relation. Fam Med 2002;34:319–24.

[29] Ong LM, de Haes JC, Hoos AM, Lammes FB. Doctor–patient communication: areview of the literature. Soc Sci Med 1995;40:903–18.

[30] Kiesler DJ, Auerbach SM. Integrating measurement of control and affiliation instudies of physician–patient interaction: the interpersonal circumplex. Soc SciMed 2003;57:1707–22.

[31] Duffy FD, Gordon GH, Whelan G, Cole-Kelly K, Frankel R, Buffone N, et al.Assessing competence in communication and interpersonal skills: the Kala-mazoo II report. Acad Med 2004;79:495–507.

[32] Delbanco TL. Enriching the doctor–patient relationship by inviting thepatient’s perspective. Ann Intern Med 1992;116:414–8.

B. Young et al. / Patient Education and Counseling 85 (2011) 60–67 67

[33] Platt FW, Gaspar DL, Coulehan JL, Fox L, Adler AJ, Weston WW, et al. ‘‘Tell meabout yourself’’: the patient-centered interview. Ann Intern Med 2001;134:1079–85.

[34] Levinson W, Gorawara-Bhat R, Lamb J. A study of patient clues and physicianresponses in primary care and surgical settings. J Am Med Assoc 2000;284:1021–7.

[35] Roter D. The enduring and evolving nature of the patient–physician relation-ship. Patient Educ Couns 2000;76:5–15.

[36] Barrier PA, James TC, Jensen NM. Two words to improve physician–patientcommunication: what else? Mayo Clin Proc 2003;78:211–4.

[37] Maseide A. Possibly abusive, often benign, and always necessary. On powerand domination in medical practice. Soc Health Ill 1991;13:545–61.

[38] Coulter A. Paternalism or partnership? Patients have grown up-and there’s nogoing back. Brit Med J 1999;319:719–20.

[39] Little P, Everitt H, Williamson I, Warner G, Moore M, Gould C, et al. Preferencesof patients for patient centred approach to consultation in primary care:observational study. Brit Med J 2001;322:468–72.

[40] Saba GW, Wong ST, Schillinger D, Fernandez A, Somkin CP, Wilson CC, et al.Shared decision making and the experience of partnership in primary care.Ann Fam Med 2006;4:54–62.

[41] Giacomini MK, Cook DJ. Users’ guides to the medical literature: XXIII. Quali-tative research in health care A. Are the results of the study valid? Evidence-based medicine working group. J Am Med Assoc 2000;284:357–62.

[42] Glaser BG, Strauss AL. The discovery of grounded theory. Chicago: AldinePublishing Company; 1967.

[43] Griffin DW, Bartholomew K. The metaphysics of measurement: the case ofadult attachment. Adv Personal Relat 1994;5:17–52.

[44] Street Jr RL, Gordon H, Haidet P. Physicians’ communication and perceptions ofpatients: is it how they look, how they talk, or is it just the doctor? Soc Sci Med2007;65:586–98.

[45] Levinson W, Roter D. Physicians’ psychosocial beliefs correlate with theirpatient communication skills. J Gen Intern Med 1995;10:375–9.

[46] Pedersen R. Empirical research on empathy – a critical review. Patient EducCouns 2009;76:307–22.

[47] Butow PN, Brown RF, Cogar S, Tattersall MH, Dunn SM. Oncologists’ reactionsto cancer patients’ verbal cues. Psycho-Oncol 2002;11:47–58.

[48] Schmid Mast M, Hall JA, Roter DL. Caring and dominance affect participants’perceptions and behaviors during a virtual medical visit. J Gen Intern Med2008;23:523–7.

[49] Aruguete MS, Roberts CA. Gender affiliation and control in physician–patientencounters. Sex Roles 2000;42:107–18.

[50] Anderson LA, Zimmerman M. Patient and physician perceptions of theirrelationship and patient satisfaction: a study of chronic disease management.Patient Educ Couns 1993;20:27–36.

[51] McKneally MF, Martin DK. An entrustment model of consent for surgicaltreatment of life-threatening illness: perspective of patients requiring eso-phagectomy. J Thorac Cardiov Sur 2000;120:264–9.

[52] Butow PN, Maclean M, Dunn SM, Tattersall MH, Boyer MJ. The dynamics ofchange: cancer patients’ preferences for information, involvement and sup-port. Ann Oncol 1997;8:857–63.

[53] Beaver K, Luker KA, Owens RG, Leinster SJ, Degner LF, Sloan JA. Treatmentdecision making in women newly diagnosed with breast cancer. Cancer Nurs1996;19:8–19.

[54] Beckett MK, Elliott MN, Richardson A, Mangione-Smith R. Outpatient satisfac-tion: the role of nominal versus perceived communication. Health Serv Res2009;44:1735–49.

[55] Fiscella K, Franks P, Srinivasan M, Kravitz RL, Epstein R. Ratings of physiciancommunication by real and standardized patients. Ann Fam Med 2007;5:151–8.

[56] Wright EB, Holcombe C, Salmon P. Doctors’ communication of trust, care,and respect in breast cancer: qualitative study. Brit Med J 2004;328:864.

[57] Orfali K, Gordon EJ. Autonomy gone awry: a cross-cultural study of parents’experiences in neonatal intensive care units. Theor Med Bioeth 2004;25:329–65.

[58] Mason J. Mixing methods in a qualitatively driven way. Qual Res 2006;6:9–25.[59] Barbour RS. Mixing qualitative methods: quality assurance or qualitative

quagmire. Qual Health Res 1998;8:352–61.