Family Practice Advance Access originally published online on May 16, 2005
Family Practice 2005 22(4):412-418; doi:10.1093/fampra/cmi037
Explanations, explanations, explanations: how do patients with limited English construct narrative accounts in multi-lingual, multi-ethnic settings, and how can GPs interpret them?
Department of Education and Professional Studies, King's College London, Franklin-Wilkins Building, Waterloo Road, London SE1 9NN, UK
Correspondence to C. Roberts; Email: celiaroberts{at}lineone.net
Received 18 August 2004; Accepted 5 January 2005.
Moss B and Roberts C. Explanations, explanations, explanations: how do patients with limited English construct narrative accounts in multi-lingual, multi-ethnic settings, and how can GPs interpret them? Family Practice 2005; 22: 412–418.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsDiscussionDeclarationReferences |
|---|
Background. The gap is widening between understanding the subtle ways patients and GPs manage their talk, and superficial discussion of the ‘language barrier’ among linguistic minority patients. All patients have to explain themselves, not just those for whom English is their first or main language. Patients' explanations reflect how they want the doctor to perceive them as a patient and as a person: they reveal patients' identities. Yet interpretations are not easy when patients' style of talking English is influenced by their first language and cultural background.
Objective. To explore in detail how patients with limited English and GPs jointly overcome misunderstandings in explanations.
Methods. Using discourse analysis and conversation analysis, we examine how GPs and their patients with limited English negotiate explanations and collaborate to manage, repair or prevent understanding problems.
Results. 31% of patients said English was not their first language. Misunderstandings arise owing to a range of linguistic and cultural factors, including stress and intonation patterns, vocabulary, the way a patient sequences their narrative, and patient and GP pursuing different agendas.
Conclusion. When talk itself is the problem, patients' explanations can lead to misunderstandings, which GPs have to repair if they cannot prevent. Careful interpretation by skilful GPs can reveal patients' knowledge, experience and perspective.
Keywords. Communicative style, cultural differences, explanations, misunderstandings.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsDiscussionDeclarationReferences |
|---|
Explanations in the GP consultation are usually discussed in two ways: doctors' explanations, and the perennial problem of making oneself understood,1,2 and how to elicit patients' explanatory frameworks.3 What is rarely tackled is how patients make themselves understood. Despite the significant literature on patients' narratives,4,5 the problems of shared understanding, of processing meaning as the interaction unfolds, have not been specifically addressed.
But what happens when language/communication differences mean GPs perceive patients' accounts as incoherent? All patients have to explain themselves, not just those for whom English is their first or main language. Yet it is often difficult for doctors to process what patients say if they have limited competence in the dominant language(s) of the country. No matter how patient-centred a GP is, or how determined to achieve a collaborative outcome, the crucial starting point is grasping the literal meaning of patients' talk. Even relatively fluent speakers of the dominant language have differences in accent and communicative style, which make it harder to process meaning accurately. Where conditions for shared understanding are in place, a fundamental level of comprehension occurs automatically. However, in South London surgeries about 30% of patients do not have English as their first/dominant language. If talk constitutes a communication problem, rather than contributing to its resolution, doctors may not access the patient's concerns adequately. In particular, it may be difficult to establish how patients conceive of and orient to their illness and other concerns. How, then, do doctors and patients with limited English work together to explore the patient's perspective?
Studies of health, ‘race’ and ethnic minorities tend to deal with language differences only in terms of interpretation and translation:6 patients can or cannot speak English, they need an interpreter, link worker or advocate or they do not. Yet the issue is not so straightforward. In an ideal situation, doctors and patients would share the same first language or there would be professional interpreters ready at all times to mediate. However, in multilingual societies, no such ideals exist. Options are limited by availability and cost, and even when available, the interpreter-mediated consultation can produce its own misunderstandings. Also, some patients prefer to communicate directly with their doctor. Language use and understanding is as dynamic as the notion of ‘culture’,7 with no obvious marker denoting a ‘language barrier’ with Patient A but not Patient B. Just as different aspects of one's cultural identity are invoked in different contexts, so, in some settings, an individual may appear relatively fluent and be perceived as having a high level of understanding, while in others the same person's ability may be more limited (or at least may be perceived so by the GP). Factors influencing variability include: topic, GP's communication strategies, and whether GP and patient have met before. It is therefore more useful to think of a continuum of ability to use English than of a fluent-nonfluent dichotomy.
The micro-analysis of consultations has shown how patients subtly and carefully design their explanations to take account of the doctors' knowledge and power. For example, patients, typically, downplay their own knowledge when offering an explanation of their illness.8,9 Why do patients take such care over their explanations? Their careful design suggests that they are concerned with how the doctor perceives them. They recognise that how they present themselves to the doctor is relevant to the outcomes of the consultation and how their concerns and interests are met.10 Where there are language/cultural differences, it is much harder for patients to manage their identities.
The ‘race’ and health literature does not deal with issues of interaction and doctor–patient social relations. It is also limited by the oversimplification of ‘health beliefs,’ which are portrayed as fixed mental models, and stereotype ethnic groups.11,12 Patients do not arrive with rigid beliefs and simply transmit them wholesale to the doctor. Rather, their concerns, habits, and attitudes are produced as part of the ongoing interaction and are designed to be appropriate and relevant for that moment. And health beliefs are only fragments of patients' stories. The accounts of themselves and their symptoms given by patients reflect how they want to be perceived by the doctor, as a patient and as a person: they reveal patients' identities. Explanations are therefore a rich source of insights for GPs: how they are designed, what is emphasised, what omitted and what perspective is conveyed are all integral to how a patient chooses to present themselves. But when there are language and cultural differences, formulating explanations can be challenging, and making sense of the insights they yield equally so. Yet skilful GPs act as ethnographers, deciphering not only what patients with limited English are trying to say, but also ‘who they are.’
|
Methods |
|---|
|
TopAbstractIntroductionMethodsDiscussionDeclarationReferences |
|---|
The data below form part of a corpus gathered for a programme of research and educational intervention on patient–family doctor interactions, the PLEDGE (Patients with Limited English and Doctors in General practice: Education issues) project, which used discourse analytic and conversation analytic techniques to explore how GPs and patients negotiate meaning and collaborate to manage, repair or prevent understanding problems. Conversation analysis13 uses authentic, naturally-occurring examples of talk to explore how speakers design the content of each of their ‘turns’ at talking, to look at how interactions are sequenced and managed. Discourse analysis examines speakers' choices of vocabulary, grammar, intonation and rhetoric and how these vary depending on speaker background and audience. These detailed features enrich our understanding of how social relations are managed in talk.
Data collection
PLEDGE's main body of data is 232 video recorded consultations in four GP practices, including one single-handed practice, in Lambeth, South London. 19 experienced GPs were filmed during 30 surgery sessions. Consecutive patients attending clinics were approached for consent, and both emergency and routine appointment clinics were filmed, to reflect the reality of practice in busy urban surgeries. Of the 19 GPs, 11 spoke more than one language and all used English as their dominant language. However, there was no simple matching of, for example, German-speaking doctor with German-speaking patient; whatever their linguistic repertoire, each doctor regularly faced patients from a range of language backgrounds and varying degrees of ability to speak English.
All 232 videos were viewed three times, independently, by two discourse analysts, using an established and widely-accepted approach.14 Recordings were supplemented by: ethnographic observation in each general practice; a brief questionnaire completed by each patient after their consultation; feedback and comment from GPs on videos selected by the researchers. 48 of the consultations were selected for detailed transcription and analysis, on the basis that they featured episodes which gave rise to brief or protracted misunderstandings, or periods of evident unease for one or both participants. These consultations were not chosen to represent extremes of communication breakdown; rather, they reveal the work patient and GP routinely engage in when language makes reaching a satisfactory resolution more difficult, or more time-consuming.
PLEDGE patients
31% of patients said English was not their first language; these patients reported speaking a total of 30 different languages. 20% of all consultations were with patients from non-English speaking backgrounds/patients with a culturally specific style of communicating and featured frequent and profound misunderstandings.
Patients ranged from those who spoke only a few words of English to local speakers for whom English was their first/only language. Many of our observations concerning the negotiation of meaning, and collaboration to manage, repair and prevent misunderstandings, focused particularly on patients in the middle of this spectrum: those speaking English as a second language, who did not use an interpreter (though they were sometimes accompanied by an advocate), but who encountered frequent or complex misunderstandings in their consultations. The patients featured here reflect this focus: some have difficulty conveying even their literal meaning, while others are more fluent, but have culturally different styles of communicating, influenced by their first languages.
Patients' explanations
The following three excerpts illustrate patients' explanations about symptoms and/or domestic/relationship concerns. In the first two, a purely ‘technical’ understanding issue means basic processing is difficult for the GP. The third example reveals how subtle yet complex differences in interaction styles make patients work hard to create accounts of themselves to which the GP can respond meaningfully. (For transcription conventions, see Box 1).
BOX 1 Transcription conventions
|
Excerpt 1: Dog bite. A Nigerian patient was bitten by a dog on holiday:
|
GP and patient share a grammatical and lexical system, but their stress and intonation patterns differ. These patterns are strongly influenced by the systems of a speaker's first language; even long periods of residence in the UK rarely lead to significant changes. Despite the patient's fluency, therefore, a misunderstanding occurs.
The GP explores whether a rabies vaccination is necessary (lines 1 & 3). Told the dog reportedly sees a vet regularly (lines 7–8), and the patient knows its owner (lines 12–13), the GP appears satisfied the evidence is authoritative—he says ‘oh fair enough so’ (line 14)—and a vaccination is not indicated. However, his acquaintance's assertions did not convince the patient: he implies a discrepancy between the acquaintance's claims and the dog's real status (‘they told me the dog go to the vet regular but that's what they said’; lines 7, 8 & 10). He suspects the dog has not seen a vet, and wants a precautionary vaccination. In British English, contrastive stress to convey this suspicion would emphasise the verbs ‘told’ and ‘said’: ‘they told me the dog goes to the vet regular, but that's what they said’ (implication: and not what they actually do). Instead, the patient's West African intonation system15 focuses on the agent (the acquaintance), and the content of the agent's utterance—the ‘what’. The GP is faced with ambiguity. Twice, the patient hints at his scepticism by using ‘but.’ Yet the difference between the participants' intonation systems means the hint is not consolidated. The GP decides not to give the patient a vaccination.
Excerpt 2a: Dust mite (1). A patient whose first language is Italian has several concerns. One is her dust allergy:
|
The patient wonders if dust mites pose a health threat, but a misunderstanding occurs because she refers to the ‘house’ rather than her ‘home,’ and asks whether dust mites are dangerous ‘for,’ rather than the more usual ‘in,’ the home. Local judgements we make about meaning, based on the utterance we have just heard, both draw on and feed into our conventional knowledge about what certain utterances mean at certain times, or where these utterances fit into our wider semantic schema. Consequently, the GP thinks the question is about the building's structure—and surmises the patient thinks dust mites are like woodworm—instead of sanitation within the building. The patient repairs the misunderstanding by explaining, with a concrete example, the kind of danger she meant. Furthermore, she avoids indicating the GP's misunderstanding explicitly: her indirectness is a face-saving strategy. This example reveals how two very minor grammatical and lexical differences can produce a misunderstanding and lead to a clarificatory sequence. Yet it also illustrates how patients with limited English use explanations resourcefully. Further, it shows how easy it is to draw on conventional knowledge when interpreting patients' meaning. The patient's concern (it emerges later in the consultation) is intensified by misconception about dust mites' size, because she has seen a magnified image on an anti-pest spray bottle, She believes the enlarged picture is the actual size of the mites (she describes later how she thinks she feels something running up her leg) and this may account for her use of the word ‘danger.’ This is just one of a whole raft of concerns, raised here and in previous encounters. The lengthy discussions which invariably ensue are part of this patient's overall narrative, and influence the GP's perception of her as anxious and in need of reassurance.
In the second part of this excerpt, the doctor is able to use her conventional knowledge to apparently prevent a misunderstanding, in contrast to the first example. Here, doctor and patient are discussing using a nose spray. The patient explains what happens when she squirts twice into each nostril:
Excerpt 2b: Dust mite (2).
|
Here the patient is explaining that if she puts in two squirts, it ‘go down’. Her non-verbal communication suggests she is referring to her throat, although this is not clear. Although she then goes on to talk about ‘smell’ at lines 5 and 9, the GP assumes, having seen her gesture towards her lower face, that the patient means taste when she uses the word smell. The doctor suspends her literal processing of language to guess at the patient's meaning from her medical knowledge of the connection between nose and throat. What appears superficially incoherent is made coherent by the GP's appropriate inferencing. However, as with all the encounters where patients speak limited English, the apparent resolution may mask an unresolved misunderstanding. Despite her ‘yeahs’ of agreement, it may be that the patient has other worries about not being able to smell and, as often happens, these minimal responses at lines 3 and 7 are not markers of understanding and agreement at all but are just keeping the channel of communication open.
Explanations can be produced interactionally—provided GPs allow patients time and space to capitalise on their resourcefulness—even where patients' English is limited and talk appears to be problematic. In this next example, an elderly Colombian patient, whose first language is Spanish, attends for cholesterol test results. Disappointed with their outcome, they discuss her depression:
Excerpt 3a: Cholesterol (1).
|
The GP probes for suicidal thoughts, but tact renders his language too indirect for the patient to infer his intention: she interprets ‘scared’ literally, believes he is asking about living alone, and explains her security precautions. Despite the misunderstanding, the GP's minimal responses (lines 8, 11 & 20) encourage her to continue her narrative. She spontaneously turns the topic to what the GP originally sought after one second pause, line 14: she feels ‘sad’ and is becoming absorbed by her memories (lines 12–22). So, like excerpt 2, the conditions are created to encourage the patient's resourcefulness, both in resolving misunderstandings and in shedding light on their perspective.
When discussing smoking cessation, the patient makes an analogy with drug users to illustrate her difficulty:
Excerpt 3b: Cholesterol (2).
|
This analogy is part of a broader narrative of herself as a ‘good’ patient, comparing her relatively small challenge with that faced by those giving up drugs. However, her intonation and grammar, influenced by Spanish, make this hard to follow. In English, individual word stress is used to contrast ideas, and most English speakers would stress ‘I’ (the cigarette smoker, lines 20–21), contrasting it with ‘poor fellows’ (line 23). The patient stresses ‘feel’ (line 21), to convey strength of emotion. There is a danger of dismissing a patient's narrative as incoherent because their system of conveying meaning is hard for other English speakers to process.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsDiscussionDeclarationReferences |
|---|
Greenhalgh and Hurwitz point out that narrative does not simply report what individuals do, but also how they feel, showing patient perspective. Narratives, they argue, invite interpretation. However, when there are language/cultural differences, information and perspectives can be lost and patients' intentions misinterpreted. The structure of patient narratives may run counter to GPs' expectations, and far from being an unproblematic route into accessing a patient's thoughts and values, these narratives can appear bewildering, and handled badly can raise more questions than they address.
Nevertheless, these patients work extremely hard to design explanations and with interpretations by skilful GPs, explanations offered by patients with limited English can reveal much about their stance and state of mind. In contrast with most of the literature, this paper has therefore focused on patient rather than doctor explanations, and on the work both participants engage in to ensure a successful outcome. The importance of examining the contributions of both interactants is underlined by Reissman:16 "One can't be a ‘self’ by oneself, identities must be accomplished in ‘shows’ that persuade." We would concur: Example 2 (‘Dust Mites’) and example 3 (‘Cholesterol’) revolve around patients revealing aspects of themselves designed to persuade the GP they are seriously worried or conscientious. When intercultural consultations are successful, patients are able to work up and reveal a particular aspect of their identity, as part of a joint accomplishment with their GP.
Allowing patients to continue talking when they seem to be straying from the intended focus may feel counter-intuitive, and challenges GPs' strict time constraints. Yet permitting patients to control the topic, as the GPs do in examples 2 and 3, has clear benefits. For example, in ‘Cholesterol,’ the GP would undermine his gentle, collaborative style if he interrupted to tell the patient she had misunderstood. Furthermore, the patient might be deterred from revealing the extent of her depression: a particularly important consideration because earlier she announced she feels ‘fantastic,’ and this anomaly requires cautious probing. Producing a narrative in her own words, without the burden of answering a stream of questions, frees the patient to use English in a way she feels competent, concentrating on her emotional state, not on the distraction of processing English. (And, once her feelings are exposed, the GP has a clear route in to explore whether they are ‘overwhelming.’) This excerpt indicates topic involvement is a sound way to ensure patient involvement.
As noted, it would be challenging for someone unfamiliar with this patient to hear her account in ‘Cholesterol (2)’ as relevant and coherent. When comprehending explanations is difficult—or their very presence is puzzling—the patient is sometimes dismissed as unintelligible, causing GPs to ‘switch off’ and stop paying attention. All of the patient's subtle endeavour as she works on her moral self and creates a picture of the kind of person she is would therefore be lost. Fortunately, this GP has known her for many years. He is attuned to her need to construct a positive self-image; consequently he makes sense of her explanation contextually (indeed, our interpretation owes much to his detailed feedback after video viewing).
Greenhalgh17 cautions: "the relentless substitution during the course of medical training of skills deemed ‘scientific’—those that are eminently measurable but unavoidably reductionist—for those that are fundamentally linguistic, empathic and interpretative should be seen as anything but a successful feature of the modern curriculum." Many consultations are not influenced by linguistic or cultural differences between patient and GP, but where these factors do play a part, it is vitally important for doctors to reflect on the patient's frame of reference, and for them to work to agree meaning meticulously with their patients. This has implications for the amount of time such consultations may take. With less skilful GPs the consultations may be suspiciously short. But just because patients have limited English does not mean that they have limited problems. Patients may say little or appear to agree because they understand little. However, the meticulous negotiation of shared understanding is likely to take more time rather than less, as the examples above begin to demonstrate. Practices often book extra time for interpreter mediated consultations but consultations with patients with limited English may take as long.
GPs are offered support in consulting with patients with limited English in a training video from the PLEDGE project.18 Clips from our data corpus are used to demonstrate some of the problems of understanding that can occur and some ways of preventing and repairing them. Even where patients struggle to explain their concerns in limited English or a very different style of communicating, GPs can develop strategies to help more explanation lead to better understanding.
|
Declaration |
|---|
|
TopAbstractIntroductionMethodsDiscussionDeclarationReferences |
|---|
Funding: the PLEDGE project was funded by the Sir Siegmund Warburg Voluntary Settlement.
Ethical approval: no further ethical approval necessary. Ethical approval was originally granted by the St Thomas' Hospital local research committee for the project: Patients with Limited English and Doctors in General Practice: Educational Issues (funded by the Sir Siegmund Warburg Voluntary Settlement).
Conflicts of interest: none.
|
Acknowledgments |
|---|
We thank all the general practitioners and patients who took part in the PLEDGE project. We also acknowledge our co-investigators Roger Jones, Srikant Sarangi and Val Wass.
|
References |
|---|
|
TopAbstractIntroductionMethodsDiscussionDeclarationReferences |
|---|
1 Kindelan K, Kent G. Patients' preferences for information. J Roy Coll Gen Pract 1986; 36: 461–463.
2 Tuckett D, Boulton M, Olson I, Williams. Meetings between experts: an approach to sharing ideas in medical consultations. London: Tavistock Publications; 1985.
3 Arborelius E, Bremberg S. What can doctors do to achieve a successful consultation? Videotaped interviews analysed by the ‘consultation map’ method. Fam Pract 1992; 9: 61–66.
4 Greenhalgh T, Hurwitz R. Why study narrative? Br Med J 1999; 318: 48–50.
5 Elwyn G, Gwyn R. Stories we hear and stories we tell: analysing talk in clinical practice Br Med J 1999; 318: 186–188.
6 Smaje C. Health, ‘race’ and ethnicity: making sense of the evidence. London: Kings Fund Institute; 1995.
7 Agar M. Language Shock. New York: William Morrow; 1994.
8 Drew P. Asymmetries of knowledge in conversational interaction. In I. Markova and K. Foppa (eds) Asymmetries in Dialogue. Hemel Hempstead, UK: Harvester Wheatsheaf, 21–48; 1991.
9 Gill VT. Doing attributions in medical interaction: patients' explanations for illness and doctors' responses. Social Psychology Quarterly 1998; 61: 342–360.[CrossRef][Web of Science]
10 Roberts C, Moss B, Sarangi S, Wass V, Jones R. Misunderstandings in multilingual consultations. Medical Education 2005; 39: 465–475.[CrossRef][Web of Science][Medline]
11 Macbeth H. Defining the ethnic group: important and impossible in H. Macbeth and P. Shetty (eds) Health and ethnicity. London: Taylor and Francis; 2001.
12 Ahmad W. Making black people sick: ‘race’ ideology and health research. In W. Ahmad (ed.) ‘Race’ and health in contemporary Britain. Buckingham: Open University Press; 1993.
13 Drew P, Chatwin J, Collins S. Conversation analysis: a method for research in health care professional-patient interaction. Health Expectations 2001; 4: 58–71.[CrossRef][Medline]
14 Gumperz J. Discourse Strategies. Cambridge: Cambridge University Press; 1992.
15 Spencer J. The English Language in West Africa. London: Longman; 1971.
16 Reissman C. Performing identities in illness narrative: masculinity and multiple sclerosis. Qualitative Research (forthcoming).
17 Greenhalgh T. Narrative and patient choice. In Edwards A, Elwyn G. Evidence-based patient choice: inevitable or impossible? Oxford: Oxford University Press; 2001.
18 Roberts C and Moss B. Doing the Lambeth Talk. Video/DVD. King's College London; 2003, The DVD is available from the London Deanery, 20, Guild ford Street, London WCIN 1DZ.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||