• physician-patient interaction;
  • chronic illness;
  • qualitative methods;
  • chronic back pain;
  • menorrhagia;
  • depression;
  • medically unexplained symptoms


  1. Top of page
  2. Abstract
  3. Introduction
  4. Study groups and method
  5. Framing the doctor-patient relationship in chronic illness
  6. Conclusion
  7. Acknowledgements
  8. References

Abstract  How family doctors conceptualise chronic illness in the consultation has important implications for both the delivery of medical care, and its experience by patients. In this paper, we present the results of a re-analysis of qualitative data collected in a series of studies of British family doctors between 1995 and 2001, to explore the ways in which the legitimacy and authority of medical knowledge and practice are organised and worked out in relation to three kinds of chronic illness (menorrhagia; depression; and chronic low back pain/medically unexplained symptoms). We present a comparative analysis of (a) the moral evaluation of the patient (and judgements about the legitimacy of symptom presentation); (b) the possibilities of disposal; and (c) doctors’ empathic responses to the patient, in each of these clinical cases. Our analysis defines some of the fundamental conditions through which general practitioners frame their relationships with patients presenting complex but sometimes diffuse combinations of ‘social’, ‘psychological’ and ‘medical’ symptoms. These are fundamental to, yet barely touched by, the increasingly voluminous literature on how doctors should interact with patients. Moving beyond the individual studies from which our data are drawn, we have outlined some of the highly complex and demanding features of what is often seen as routine and unrewarding medical work, and some of the key requirements for the local negotiation of patients’ problems and their meanings (for both patients and doctors) in everyday general practice.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Study groups and method
  5. Framing the doctor-patient relationship in chronic illness
  6. Conclusion
  7. Acknowledgements
  8. References

The clinical encounter between the doctor and the patient has both tremendous practical and ideological importance for the discipline of primary care. From the 1930s onwards, general practitioners have become increasingly attentive to the social nature of this encounter and have sought to develop a set of practices that secure good relations with the patient, first with the hope that this would lead to better patient compliance with treatment regimens (Armstrong 1982), and more latterly with the hope that the doctor-patient relationship itself would act as a therapeutic technology; hence the famous dictum that the ‘doctor is the drug’ drawn from the work of Balint (1957). Even so, until the 1970s, a strongly doctor-centred model of the clinical encounter held sway, in which the epistemological authority of medical knowledge and practice, paternalistically embodied in the doctor, was given as unproblematic, and ‘relationships’ through which the doctor knew the patient could legitimately be assumed to be longstanding.

From the late 1960s, the doctor-centred model of the clinical encounter came under sustained criticism. There were several reasons for this. Wider cultural and political shifts in the 1960s worked to undermine both the paternalism and the biomedical reductionism of medical practice. Patients began to push to be seen as more than the passive objects of clinical knowledge and procedure, and to demand to be understood and treated as ‘whole persons’, as part of a political shift which began with the critique of medicalization (Conrad and Schneider 1980) and extended to include a holistic impulse that connected ill-health with a wider pattern of social relations and contexts (May 1992). Shifts within medicine were also important: from the 1970s onwards, some elements of academic medicine had begun to co-opt both the theoretical positions (and practitioners) of the social sciences (Strong 1979a, Stacey 1991), and to attempt to lay claim to the much wider vision of ill-health espoused by proponents of a biopsychosocial model of medicine (Engel 1977)1. In the same light, academic general practice had begun to examine the form and content of the consultation much more systematically. A series of influential empirical studies engaged in microscopic examination of what general practitioners did in the consultation, and how they did it (Byrne and Long 1976, Tuckett et al. 1985, Neighbour 1987). This led in turn to a growing sense that the consultation was about the negotiation of two kinds of expertise: the authoritative general expertise of the doctor (often conceptualised as professional knowledge), and the specific experience of the patient (often conceptualised as lay beliefs).

The shift away from the doctor-centred consultation to negotiated consultations that accelerated after the mid-1970s was also encouraged (in the UK at least) by specific government policies intended to increase professional accountability and remake patients as active consumers of healthcare, able to adjudicate on its content and quality (Moran 1999). But more fundamentally, this shift occurred because within medicine itself there was a move towards enrolling the patient into the consultation in ways that reduced the growing strain on the epistemological authority of medical knowledge that seems to have become more evident as a result of wider shifts in its social production and organisation during this period (Lupton 1994). In this context, the shift to notions of patient-centredness, patient empowerment and patient satisfaction as proxies for quality of care in general practice (and also to different ways of conceptualising, assessing and quantitatively measuring these), can be seen not simply as technical moves towards improving the quality of care, but also as ideological moves that set out the frame in which negotiation between doctor and patient in the consultation can be understood (May and Mead 1999)2.

A number of factors, then, have combined to effect important changes in the relative status of doctor and patient in the consultation. The patient's view has become more important, the authority; the subjectivity of the doctor more problematic (Gothill and Armstrong 1999); and the business of negotiating within the consultation more complex. Abundant evidence of this may be found in the growing corpus of critical studies of doctor-patient relations and interactions in primary care. Studies ranging from accounts of gender relations in the consultation (Malterud 1999); the local negotiations that surround diagnosis, referral and treatment delivery (Stevenson et al. 2000); and studies of general practitioners’ responses to changes in the organisation of primary care itself (Calnan and Williams 1995), all reveal elements of complex problems of power and authority in doctors’ interactions with patients. Moreover, these problems seem to place a heavier burden on the doctor in the care and management of people with chronic health problems.

Understanding how doctors conceptualise and construct these problems is important for two reasons. First, because it is a necessary corrective to the increasingly quantitative adjudication of the technical quality of doctors’ actions and patients’ responses to them that now runs through much clinical research on the consultation. Second, because it reveals the negotiated consultation, and so enables us to see whole relationships as the sites of qualitatively complex rather than mundane work in primary care. Our purpose in this paper is to explore these issues, through the secondary analysis of qualitative data collected in a series of studies of the ways that general practitioners construct their ideas about the management of chronic illness in the consultation.

Study groups and method

  1. Top of page
  2. Abstract
  3. Introduction
  4. Study groups and method
  5. Framing the doctor-patient relationship in chronic illness
  6. Conclusion
  7. Acknowledgements
  8. References

Individual studies and data

Between 1995 and 2001 we undertook a series of qualitative studies that explored general practitioners’ constructs of chronic illnesses and their responses to them in the consultation through semi-structured interviews. Respondents in these studies were drawn primarily from practices in urban and metropolitan areas of North West England. Details of sampling frames, methods and the specific results of these studies have been reported in detail elsewhere (Chapple 1997, Chew-Graham 1999, Dixon 2000, Rogers et al. 2001, Chew-Graham et al. 2001, Wileman et al. 2002). The studies are described in Table 1.

Table 1. Sources of secondary data
StudyLocation, recruitment and subjectsFocus
Chapple (1997)North West England. Purposive sample of 73 general practitioners.Management and referral patterns in menorrhagia (Chapple et al. 2001).
Chew-Graham (1999)Greater Manchester. Purposive sample of 22 inner city general practitionersManagement of chronic low back pain in the consultation (Chew-Graham and May 1999).
Dixon (2000)North West England. Sample of 38 general practitioners already participating in a randomised controlled trial.Educational intervention to improve the diagnosis and management of depression in primary care (Dowrick et al. 2000).
Rogers et al. (2001)Manchester. Purposive sample of ten general practitioners and their patients with moderate depression.Understanding the points of contact and difference between doctors’ and patients’ accounts of depression.
Chew-Graham et al. (2001)Greater Manchester. Purposive sample of 22 general practitioners in inner city and suburban areas.Long-term management of depression in general practice.
Wileman et al. (2002)Greater Manchester. Purposive sample of 17 general practitionersProblems associated with the presentation of medically unexplained symptoms in general practice.

The studies from which these interviews are derived explored general practitioners’ views of the boundaries of the doctor-patient relationship in general practice; and specific studies of general practitioners’ accounts of (a) physiological symptoms of uncontested origin (menorrhagia); (b) psychological symptoms of uncontested origin (depression); and (c) contested symptoms that may be of psychogenic or organic origin (chronic low back pain, and medically unexplained symptoms). In each study, semi-structured interviews were conducted, audio-recordings of these made, and the interviews transcribed. These transcripts, edited to ensure the anonymity of respondents, formed the formal data for analysis, which normally followed the constant comparative techniques described by Glaser and Strauss (1967). From these projects 182 transcripts had been retained in a usable form, either electronically or on paper. This material formed the basis of the re-analysis presented in this paper.

In each of these studies, respondents reflected in detail on their notions of the authority and legitimacy of clinical knowledge of conditions, their resolution, and their problematic relation with the experiential character of these problems in the lives of their patients. More fundamentally, these studies raised questions about the ways that general practitioners conceived their power to solve such problems. Our view was that these questions, although born of studies which themselves addressed very different research agendas, required further enquiry and that the body of data collected during individual studies might be usefully combined to meet this objective in relation to the question: How does this aggregate body of data reflect problems of legitimacy, power and authority in doctors’ accounts of their relationships with patients?

Secondary analysis

In undertaking secondary analysis, two quite different kinds of technique were available to us. We could subject our studies to the kind of synthetic technique aimed at developing constructs from the analytic processes and conclusions of individual studies (Hawker et al. 2002, Campbell et al. 2003, Thorne et al. 2002). Such techniques have value in systematic review, but would rightly raise questions about distance and objectivity if applied by us to our own work. For this reason it was decided to follow a second path and re-analyse the data. Even so, in looking again at these data, we could not be blind to our own previous analyses. The principal risk here was that given our familiarity with this material, our re-analysis would be confirmatory rather than critical and exploratory. Our analytic procedure is shown in Figure 1.


Figure 1. Study design and analytic procedures

Download figure to PowerPoint

To ensure that the re-analysis was critical rather than confirmatory, we employed ‘naïve coding’ as a means of initial reinterpretation of interview transcripts. Gayle Allison (GA, with a background in organisational research) and Ruth Graham (RG, with a background in research on sexual assault) were unfamiliar with our previous work and remained so while they familiarised themselves with the transcripts and developed an initial coding framework. Because it was impracticable to work through the entire body of 182 transcripts a random sample of 15 transcripts was selected and ‘cleaned’ by GA, and loaded in Atlas Ti® files to facilitate qualitative analysis. GA and RG then coded these to develop an initial coding frame, and refine the set of analytic categories constructed within it. Carl May (CM) then refined the coding frame to ensure that the thematic categories developed in this way were distributed across a further 50 transcripts randomly sampled from groups a (n = 15), b (n = 20) and c (n = 15) described above3 (sampling ceased once category saturation had been achieved). The analytic process described above led to three main thematic categories from which subsequent theoretical interpretation could be launched.

  • 1
    evaluative judgments about the patient, including (a) the patient's motive for presenting; (b) the legitimacy of symptoms presented; and (c) the congruence between the doctor's and the patient's conceptual model of the illness.
  • 2
    the organisational structure of general practice, including (a) the constraints of the consultation; and (b) the possibilities for disposal of the patient; and (c) the available resources to manage patients within relatively short consultations.
  • 3
    the doctor's personal responses to the patient, including (a) co-operative rapport; (b) empathy and sympathy; and (c) frustration and hostility.

The method adopted permitted two approaches to the data set: (i) A cumulative comparative analysis that produced general thematic categories (described above), and which sought to establish the extent to which they were genuinely distributed within the frame of the combined data set4; and (ii) a condition comparative analysis that examined the differences and commonalities within these categories as they were manifest in relation to the three kinds of medical problem identified above. Once cumulative and condition comparisons had been undertaken, their results were circulated within the group of authors for confirmation and comment5.

Comparisons between types of condition were deployed to form simple tabular models (see Tables 2–4) of the ways that different symptom presentations engaged with problems of power within the clinical encounter. We have chosen to present these structural models to represent the data, although we do also present extracts from interviews for illustrative purposes. This is because our data are largely composed of generalised accounts of types of patients, and types of symptom presentations and their resolution. This means that the analyses that we can present are also generalised, since they are dependent on the rhetorical strategies that family doctors use to frame their work, rather than accounts of specific cases. Against this we can set the value of comparative analyses in opening up these rhetorical moves to a theoretical interpretation6. In this context, the objective of the exercise was not to interrogate specific cases, but instead to open up formal rather than substantive theoretical development.

Table 2. Evaluating the patient
ConditionMenorrhagiaDepressionChronic low back pain/medically unexplained symptoms
Motive for presentationCertain: normal physiological process perceived to have taken on a pathological character.Certain: patient overwhelmed by existential crisis.Uncertain: social character of symptomatology emphasised by doctors.
Legitimacy of symptomsUncertain: objective measure of abnormal menstrual blood loss impracticable in primary care; doctor relies on patient's personal history.Certain: observable state, variety of aetiological factors many of which were outside of the remit of medicine to solve.Uncertain: relationship between organic pathology, disability and severity of reported symptoms difficult to establish.
Congruence of modelsCertain: doctor and patient both operationalise a physiological model that recognises psychosocial consequences.Relatively certain: doctor and patient both operationalise a psychological model that recognises physiological consequences.Contested: patients tend to deploy organic model of pathology, doctors a psychosocial model.
Table 3. Possibilities for disposal
ConditionMenorrhagiaDepressionChronic low back pain/medically unexplained symptoms
ObjectiveCertain: Relieve symptoms, minimise discomfort and social embarrassment.Certain: Relieve symptoms, although it is understood that acting on social causes is not feasible, to permit effective engagement with the self and the social world.Certain: Relieve symptoms; permit effective engagement with the social world.
Local meansCertain: Pharmacological treatment in primary care; referral for surgical interventions.Certain: Pharmacological treatment or psychotherapeutic treatment in primary care; referral to community or in-patient psychiatric services.Uncertain: Referral for diagnostic tests to establish the existence of an organic pathology. Pharmacological or manipulative therapies.
OutcomesCertain: Effective simple disposal as symptoms relieved or passed on to experts.Certain: Palliation through complex disposal as opportunity to refer to other services limited; patient maintained in primary care.Uncertain: Containment as symptoms rarely relieved by medical intervention.
Table 4. Empathic orientations to the patient
ConditionMenorrhagiaDepressionChronic low back pain/medically unexplained symptoms
Status of disorderEpisodic/chronicAcute/Episodic/ ChronicChronic/Episodic
Congruence of clinical and lay modelsRelatively certainRelatively certainContested
Status of doctor-patient relationshipRoutine/episodicHolding patternContinuous/Chronic
Exit pointSymptoms relieved by medical intervention or menopauseSymptoms palliated by therapeutic intervention or relieved by existential changeNone visible
Empathic responseTechnicalMaintenanceFrustration

Framing the doctor-patient relationship in chronic illness

  1. Top of page
  2. Abstract
  3. Introduction
  4. Study groups and method
  5. Framing the doctor-patient relationship in chronic illness
  6. Conclusion
  7. Acknowledgements
  8. References

In the introduction to this paper we observed that clinical writings about the consultation in primary care have, over a period of 50 years, shifted their focus from a doctor-centred model that emphasised the authority of medical knowledge and practice, to a patient-centred model that lays claim to the consultation as a field of local negotiations. In this context, while the central focus of medical knowledge and practice is on revealed pathology, the consultation needs to take into account a far wider set of conditions and problems, at the centre of which is the patient's own subjective experience of illness. The latter has been the focus of considerable sociological research exploring the experience of chronic illness, and its impact on the sufferer (Bury 1982, Charmaz 1995, Kelly and Field 1996, Williams 2000, Thorne et al. 2002).

One way of understanding the basis of local negotiations around symptoms and their meanings in the primary care consultation is to reflect on a key epistemological problem in medicine: the way in which subject-object relationships are worked out in practice. This is a key point of contact between the view from both the social and clinical sciences. Sociological and historical accounts of medicine lay great emphasis on a longstanding trend towards objectification: in this view, the patient has gradually ceased to be considered an experiencing subject, and has been steadily reduced to an object of clinical practice and procedure (Turner 1995). In other words, the patient has become a passive thing upon which medicine works and upon which clinicians act. In this kind of account, medical knowledge and practice serve to objectify the patient, denying the experiential and subjective aspects of ill-health, and focusing down upon the patient as a specific instance of a generalised pathology. Evidence of this is presented in accounts of the steadily more microscopic ‘cosmologies’ of medicine (Jewson 1976), in which persons are reduced to organs, then to lesions, and most recently to problems at the molecular level. The formal rationality and practices of bioscience are therefore understood to be ever more reductionist in their effects and however much compensating attention is devoted to the psychological, emotional and ethical aspects of medical encounters, the patient is always in problematic relation to the lesion (Armstrong 1987). A second source of evidence for biomedical reductionism is seen to lie in the professional socialisation of doctors, particularly ethnographies of medical school training, e.g.Good (1994) in the US, and Atkinson (1981) and Sinclair (1997) in the UK, and of clinical residencies (Atkinson 1995). These have pointed to the ways in which medical students are taught to focus on the organic aspects of illness, and that medical education stresses that ‘proper’ medicine is primarily about objective disease states and organic pathology.

But if the patient's subjectivity has come to be situated outside the realm of organic hospital medicine, in general practice clinicians are deeply concerned, and frequently troubled by it. This concern arises from the practical knowledge that the realm of the social is intimately linked to ill-health, its experience, and its mediation through different forms of clinical presentation. This is especially the case with the kinds of diffuse problems connected with chronic ill-health and chronic unhappiness that are often encountered in the general practitioner's surgery. These are frequently troubling because it is not clear what, if anything, can be done about them – and because clinical uncertainty about aetiology and remedy threatens the authority upon which doctors rely to frame and enact medical knowledge and practice, and thus its outcomes (Salmon 2000).

So, while biomedicine is seen to undergo a long-term shift away from the patient (towards steadily more microscopic pathologies), in primary care revealed pathology has to be set against a much broader set of concerns, not the least of which is the social character and context of the patient, and the difficulty of establishing boundaries between the normal and the pathological in individual, idiosyncratic, cases. Our analysis focuses on the boundaries between revealed pathology and social character of the patient as it is evident in comparisons between three kinds of disorder: menorrhagia, depression, and chronic low back pain and medically unexplained symptoms.

The evaluation of the patient

That doctors form social evaluations of their patients is not a new observation. Indeed, it is a longstanding feature of sociological accounts of the relationship between them, from Parsons (Parsons 1951) onwards. In this study, we found evidence of this in the appearance of three kinds of problem associated with such evaluations: the patient's motivation to present symptoms; the legitimacy of these symptoms; and the consequent willingness to accept medical advice. These seem to follow directly from Parsons’ model of the sick role. Conventionally, both sociological and clinically-oriented research have focused on the types of patient and doctor behaviours that lead to problems in the consultation. Sociological accounts have emphasised the effects of doctors ‘labelling’ patients and problems in ways that undermine the legitimacy of their attempts to seek help (Stimson and Webb 1975, Strong 1979b, Lupton 1994).

Motivation may be problematic in all three cases: but in menorrhagia and depression doctors’ accounts evinced a much higher level of evaluative certainty about the reasons for which patients presented these symptoms, and their legitimacy. But as Table 2 shows, certainty about motivation and legitimacy is a different matter in the joint case of chronic low back pain and medically unexplained symptoms. In the latter case, the social character of symptom presentation was given as a significant problem, even though accounts of menorrhagia and depression also drew on the social character and variability of these symptoms.

We can account for this variable perspective on the social character of symptoms when we consider the models of disease process that underpinned doctors’ views. The most important feature of this is the congruence between doctors’ and patients’ models of menorrhagia (it features primarily as a physiological process, although interpretation of the boundary between the normal and the pathological by patients is sometimes given as a problem); and of depression (it is defined by doctors as a psychological problem, and their accounts suggest few cases of this model being resisted or contested by patients). Doctors’ accounts of menorrhagia acknowledged that it had important psychosocial consequences, but these were secondary to a physiological disease process, and in most cases the definition of abnormal blood loss was that made by the sufferer:

MEN/B/09: If they come in and say, ‘this is horrendous, and I am not coping with this, and it is far too much’, then that's menorrhagia in my view ( . . . ) I go by what the woman defines as unacceptable heavy bleeding, entirely. Because, if you go through all of the hassle of actually determining blood loss and you come back to the woman a week later and say, ‘I’m terribly sorry, you haven’t got heavy periods at all’, she is not going to believe you and it is not going to cure her perceived problem.

Similarly, accounts of depression recognised the physiological consequences of the illness, for example, in symptoms of fatigue or sleep disturbance. But in the joint case of chronic low back pain and medically unexplained symptoms, congruence between ‘lay’ and ‘medical’ models of the disease process at work was much more problematic. Indeed, these accounts stressed the contest between patients’ mainly biomedical model of an organic pathology in play, and doctors’ accounts of their attempts to employ a psychosocial model that was systematically resisted.

CLBP/GP10: You can’t get them to accept this [lack of organic pathology] ( . . . ) If you really upset them, they’re not going to trust you again.

Here, uncertainty about the motive for presentation and about the legitimacy of symptoms presented by the patients extended to disagreement about the pathological process at work7. Beyond this, we can also make the point that the crisis in authority in the consultation that doctors’ accounts reveal is the product of the apparent weakness in the epistemological authority of medicine being made evident to the patient, as diagnostic tests and physical examinations reveal an uncertain relationship between the symptoms experienced by the patient and the clinical signs observed by the doctor. Paradoxically, the hypothesis that we can offer here is that far from the epistemological authority of biomedicine being weakened by the social changes of the last 30 years, it is in fact strong enough to overwhelm the doctor who attempts to deploy another kind of model, whether this is a social or psychological one.

At this level of analysis, symptoms become matters of clinical importance in two ways. First, and most obviously, they reveal illness. All three kinds of clinical problem addressed in this article rely on the patient's subjective interpretation of experiences of the self as the basis for engaging with the doctor: in two of them, these interpretations are clinically useful because they define both the boundary of an illness and the points at which doctor and patient can agree about its nature (and so render it objective); in the third, they are a clinical problem because they form the point from which disagreement about the nature of illness can be launched, and an alternative explanation or management strategy set out. Second, because they become the vehicle by which the epistemological authority of medicine is tested in the consultation, and it is to this problem that we now turn.

The possibilities for disposal

If the basis for local negotiations in the consultation is the doctor's evaluation of the patient, their motivation, and of the legitimacy of symptoms, then this finds expression in the possibilities for action that stem from it. In Table 3 we set out a comparison of the range of possibilities for action raised by the three conditions with which we are concerned. In each of them, doctors’ accounts focused on the importance of relieving symptoms of physical and psychological discomfort, and effectively disposing of the patient's problem.

In menorrhagia, the means to achieve this were problematic, since they involved either pharmacological or surgical treatments of variable effectiveness, but which held some promise to relieve the discomfort and embarrassment attending abnormal blood loss. But if the patient's problem could not be effectively disposed of by treatment in primary care, it could be organisationally disposed of by onward referral to hospital-based gynaecologists. This simple disposal relied on local treatments either relieving, or not relieving, the subjective symptoms that the patient accounted for. In depression, doctors’ accounts were rather more complex, and also pessimistic.

Simple disposal was less likely for people with depression, because onward referral was inhibited by lack of resources, and a political view that depression ought to be managed in primary care (Rogers et al. 2001). This meant that only the most serious cases were likely to be acceptable referrals to psychiatrists, psychotherapists or community psychiatric nurses. Even so, doctors had a means of complex disposal available to them, and could either treat the illness pharmacologically, or prescribe a talking cure locally (by referring to a practice counsellor or some other agency). But palliation of symptoms, rather than their complete relief, ran through their accounts, in part because they recognised that much of the depression that they encountered was structural in origin (i.e. it was a response to poverty, poor housing, violent or unsuccessful relationships, exposure to crime, and so forth):

DEP/GP/T: Depression is, I think, very common in this practice because it's in a large council estate and there's a lot of unemployment and people are unsupported and so on. ( . . . ) If I make a diagnosis of depression, most people come out and say they feel sad, feel low, life's not worth living any more, and then physical symptoms of tiredness, not sleeping, losing weight so there's a combination of physical and emotional symptoms.

Even in circumstances where doctors raised questions about the motivation and legitimacy of some presentations of these subjectively-bounded symptoms; about the extent to which they crossed the boundaries of normal and pathological states; and about the effectiveness of medical treatments for social problems, disposal could, at some level at least, be accomplished. Even so, many patients had to be placed in a holding pattern, as disposal to other services was not an option and a ‘cure’ unlikely:

DEP/GP2/24: Some people just want to come back and I think that's OK. You don’t always have to give them a prescription. I get frustrated like anyone else at times, but yeah, I think if you can recognise that's [the consultation] got intrinsic value . . . it's OK.

In menorrhagia and depression the epistemological authority of medical knowledge and practice was not undermined by contested models of the illness, or by the absence of a means of acting for the patient. Indeed, the outcome of both kinds of symptom presentation was largely treated as certain, even if its clinical effectiveness was not. Simple disposal for menorrhagia, and more complex disposal (through palliation in a holding pattern) of depression both imbued these accounts with a degree of certainty about what action the doctor could take on behalf of the patient and these were discursively located in the arena of medical treatments.

In chronic low back pain, and in medically unexplained symptoms, the business of disposal was problematic. Simple disposals were not available, because there was often little by way of an observable link between the patient's account of the subjective experience of symptoms and their severity, and revealed pathology. Extensive diagnostic tests revealed little, and doctors came to construe the symptoms as evidence of revealed problems of psychogenic or social origin. The question that they faced was how to persuade the patient that the link between symptoms and pathology was complex (and involved the conversion of other troubles into an experience of pain), and if this was not possible, how then to contain the patient:

MUS/GP/10: What I do for a lot of somatising patients when it's obvious what is happening is, I explain how the practice works, explain how I work as a doctor, how I come to a diagnosis, I explain how much time they have with me, and I ask them what they think it would be reasonable to do in 10 minutes. So, for example, if they had a headache, bowel pain, in-growing toe-nail, and a problem with your piles, is it reasonable that in 10 minutes I address and examine you for all those issues? To which they always reply ‘No, it's not reasonable’, so I say . . . ‘Whatever is worrying you most we will address thoroughly, and then the other things will have to wait until next time you come’.

Containment was at issue here because of the general practitioner's continuing responsibility for the management of the patient's care. If patients’ subjective accounts of symptoms could not be effectively translated into a medical account of objective revealed pathology, the patient could not be moved along. By the same token, if the epistemological authority of medical knowledge and practice is tested for the patient by the doctor's interpretation of symptoms, when containment rather than disposal is the outcome, the patient can readily assume that medicine fails the test. The problem of incongruence of models between doctor and patient is amplified, and the patient (and problem) cannot be satisfactorily disposed of. This may be of greater significance than the absence of revealed pathology.

The doctor's empathic response

In Table 4 we outline some of the key features of general practitioners’ accounts of their personal responses to the kinds of presentations on which this paper focuses. Of central importance to these are two particular components of their accounts: the status of the relationship and the potential exit points which might signal its conclusion. General practitioners’ accounts of their responses to menorrhagia tended towards the empathic (although some female doctors were rather less so), even when they evaluated patients’ subjective presentation of symptoms as representing a non-pathological state:

MEN/P/GP5: If a woman comes in with bleeding that is heavy enough to interfere with her lifestyle, even if it is a lot less than 80 ml, then she's got a problem and I want to do something about it.

In menorrhagia, general practitioners had an immediate means of disposal, either by pharmacological or surgical interventions. An endpoint to the doctor-patient relationship was therefore in sight. Moreover, their orientations towards the relationship were framed in technical terms: the patient's problem was organic, could be disposed of by means of known clinical techniques, and the relationship drawn to a close if these modes of disposal were operationalised. Their encounters with these patients were therefore restricted to diagnostic and management consultations organised around considering and ‘trying out’ alternative treatments in primary care, or moving to referral to secondary care.

In depression, disposal was much more problematic, since doctors tended to frame the pathology as a response to social circumstances that were largely beyond medical intervention, and which could only be palliated by pharmacological means, but also because referral to secondary healthcare was available only for the most severe cases. In this case empathy was tempered by recognition of the pathological loss of motivation that accompanies the disorder, and patients seemed to be placed in a relationship that conformed to a ‘holding pattern’ with the doctor doing maintenance work:

DEP/GP/B: If they’ve got all of the biological symptoms, then I would try to encourage them to take medication. But as I said before ( . . . ) maybe it's my counselling skills, you know, talking about the length of time you need to get good results. But it never really works, I think there's very few.

Even though there was no immediate exit point from the relationship, doctors and patients appeared to be in agreement about its nature (even if they did not always agree about how it could be palliated or resolved). Doctors’ accounts of chronic low back pain and medically unexplained symptoms were formed quite differently. We have already noted that these conditions test the epistemological authority of medicine because of the poor fit between accounts of symptoms and their severity, and revealed pathology. Moreover, they test the practice of medicine because mechanisms for their effective disposal are not available. At the root of this, we have suggested, lies a more fundamental problem – patient and doctor are working different kinds of conceptual terrain in the consultation:

MUS/GP/14: They just seem to be stuck with the belief that they have something wrong, and they can’t move forward. They can’t move forward in their own lives because they’ve not accepted it, and you are unable as a GP to help them. So both ways you are stuck. It means that perhaps they will continue to consult, but at the end of the consultation you are still no further on.

The accounts collected for secondary analysis around chronic low back pain/medically unexplained symptoms focus intensively on contest and disagreement within the doctor-patient relationship. But beyond this, they reflect personal contests far more emphatically than accounts of consultations around menorrhagia or depression. In many accounts general practitioners were quite hostile, pointing to the ways that patients often resisted strategies and interventions that might relieve symptoms that were presented. One way to see this is as a reflection of doctors’ perceived failure to persuade patients to work within the same medical model: patients regard their problems as organic in origin, while doctors tended to see them as having social or psychological causes. Certainly some doctors felt this to be a failure, and regarded themselves as colluding with a set of explanations for symptoms that further buttressed illness behaviours and steadily undermined motivation to recover. Even so, they saw limited therapeutic benefits in such interactions for patients, but were profoundly aware that continuous relationships in which agreement was not reached about how to understand the patient's problem themselves risked chronicity. They framed these accounts in terms of deep frustration: they could neither solve the patient's problem, nor escape responsibility for doing so.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Study groups and method
  5. Framing the doctor-patient relationship in chronic illness
  6. Conclusion
  7. Acknowledgements
  8. References

Sociological and historical accounts of the development of medical knowledge and practice since the 19th century have emphasised, as we noted earlier, a steady move away from seeing the patient as an experiencing subject, towards biomedical reductionism and the formulation of the patient as a passive object of clinical practice, representing a specific instance of a generalised pathological entity. This view is echoed in studies of the professional socialisation of clinicians of various kinds, who are taught to seek out revealed objective pathologies as the proper focus of medicine. The epistemological authority of medical work therefore depends on the successful application not only of specific technical procedures and management practices, but on fundamental agreement about models of pathology, and the boundaries between pathology and normality, not just within medicine, but with its public. In much of the writing about the potential of the consultation to reveal biopsychosocial problems (Stott and Davis 1979), or on the patient as ‘expert’ (Tuckett et al. 1985) or indeed, in new policy interventions intended to develop the ‘expert’ and ‘future’ patient (Kendall 2001), this fundamental agreement is largely assumed. But, in general practice, biomedical reductionism is ultimately impossible in the way that more general critiques of medical knowledge and practice suggest, or that agreement about the conceptual terrain of symptomatology has ever been complete. Historians like Shorter (1991) describe longstanding contests between patients and doctors over diffuse symptoms, and more recent studies (for example, those that provided the data on which our secondary analysis is based) provide contemporary evidence for such contests. Indeed, one of the lessons of this study is that agreement about conceptual terrain is unevenly distributed.

Against this background, interrogating the ways that doctors frame their relationships with patients in primary care is useful, in part because it brings out the problem of these ‘models’ so strongly, and moves analysis towards an important field of inquiry: the wider social influences on the organisation of doctor-patient interaction (Wissow 2003). Certainly, the biomedical domain is at the centre of the issue, but it is intermingled with psychological and social explanations of patient presentation; nebulous boundaries between the normal and the pathological; and questions about how to dispose of patients’ problems. The indeterminacy of much general practice work is therefore brought readily into the foreground by respondents’ accounts in the studies from which we drew data for this secondary analysis. In this context, the patient's motive for presentation and the legitimacy attributed to their symptoms was important because it was the initial frame in which a more pressing question could be worked out, the extent to which doctors saw themselves as working on the same conceptual terrain as their patients. The congruence between medical and patients’ models is not, here, a matter of the invidious distinction often made between professional knowledge and lay health beliefs (a distinction which is often founded on the wrong assumption that there is neither variation in medical knowledge nor beliefs about illness amongst doctors), but rather about whether, in general terms, they are formulating the patient's symptom presentation using a similar conceptual vocabulary. The product of this agreement may be expressed in the form of a simple proposition:

P1When medical and lay models of symptoms are congruent, doctors’ evaluations of the patient are less problematic, even though the patient's motivation and legitimacy of symptoms may be in question.

Clearly, it is unwise to try to read off the experiences of individual consultations from generalised accounts like those given by many of our respondents. Nevertheless, what can be read off from this analysis are precisely the routine typifications that underpin the doctor's view of the clinical encounter, and more importantly, the deeply contextual origins of some of these. The important implication of this analysis, of course, is that the personal contests that some of our respondents described are not personal at all, but reflect much more fundamental general features of the engagement between medicine and society. Central to this is the problem of the epistemological authority that the doctor is able to bring to bear on a specific type of case, and the extent to which the model of pathology brought into play by both doctor and patient are congruent.

Recognising the typifications that underpin the consultation has utility, not in pressing for more individualised care (the ideological imperative underpinning much debate about relationships between doctors and their patients, from both the clinical and the social sciences), but in explaining the everyday difficulties that general practitioners’ work involves in solving individual problems as representatives of types. The endpoint, however, is successful disposal of the patient's symptoms (by explanation, relief, palliation or referral) and to achieve this in a way that satisfies both participants, it has to be undertaken on common conceptual terrain. This is not to claim, of course, that general practitioners do not individualise their patients, or that they treat them only as representatives of types. Malterud (2002), for example, has set out a strong conceptual basis for understanding the individualisation of care in clinical reasoning, and such views fit well with studies of clinical reasoning in practice, like that of Fairhurst and May (2001). Even so, typifications remain a consistent and important element of our data. This leads us to our second proposition:

P2When congruence exists between lay and medical models of the symptom in play, the patient's problem may be disposed in a way more likely to satisfy both doctor and patient, and the epistemological authority of medical knowledge and practice maintained.

Understanding the deeper context of the consultation as doctors experience this is both useful and important. It is useful because there is now a large body of work that sets out to understand the consultation as a body of specific interactional techniques that are amenable to modification through training, leading to an appropriately patient-centred encounter between the doctor and the patient. This work sets up doctor-patient interaction primarily as a technical problem of practice, and a variety of instruments and coding schemes have been developed to test doctors’ consultation behaviour against theoretical models (Boon and Stewart 1998, Platt et al. 2001). But much of this work relies on normative judgements about action, and is derived from prescriptive models dislocated from everyday experience of medical work. It also assumes a health gain from patient-centred practice, even though this has not been empirically demonstrated to be the case (Mead and Bower 2002). Better communication skills cannot of themselves achieve common conceptual terrain and congruence between models. It is important because doctors’ interactional behaviours and communication skills are exercised through their own contextual experiences of types of patients, types of problems, and types of disposal options – in other words, through a repertoire of routine judgements about the possibilities presented by the individual patient, and the routinely available means of solving these. All of these have to be operationalised, it must be remembered, in highly condensed time and space. This leads us to our final propositions:

P3The doctor's capacity for empathy with the patient's presentation of symptoms is associated with the doctor's perception of a successful exit point in the consultation.

P4Where no exit point is in sight, and the consultation is the focus of work to contain the expression of symptoms, rather than relieve or palliate their effects, the doctor is frustrated and the doctor-patient relationship itself has the potential to become chronic.

Our analysis has set out some of the fundamental conditions through which general practitioners can be seen to frame their relationships with patients presenting complex but sometimes diffuse combinations of ‘social’, ‘psychological’ and ‘medical’ symptoms. These are fundamental to, yet barely touched by, the increasingly voluminous literature on how doctors should interact with patients. Moving beyond the individual studies from which our data are drawn, we have outlined some of the highly complex and demanding features of what is often seen as routine and unrewarding medical work, and some of the key requirements for the local negotiation of patients’ problems and their meanings (for both patients and doctors) in everyday general practice.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Study groups and method
  5. Framing the doctor-patient relationship in chronic illness
  6. Conclusion
  7. Acknowledgements
  8. References

We thank Diane Oliver, Lindsay Wileman, Hannah Cole and Scott Hedley, who collected some of the data on which our analysis rests. Alison Chapple's contribution to this study was funded by the North West Regional Health Authority and Clare Dixon's by the NHSE NW R & D Directorate. Carl May gratefully acknowledges the contribution of Peter Salmon to his thinking around this topic. We also thank Christopher Dowrick, Eileen Kaner and Tim Rapley for their helpful comments on a draft of this paper.

  • 1

    This has been especially evident in general practice and primary care, public health medicine, and to a lesser extent, social psychiatry.

  • 2

    Substantial reviews of the conceptual and methodological basis of these shifts have been offered by Mead and colleagues in Manchester (Mead and Bower 2000b, Mead and Bower 2000a, Bower et al. 2001, Mead et al. 2002, Mead and Bower 2002).

  • 3

    We have treated CLBP and medically unexplained symptoms as a single category because of the ways that general practitioners tended to draw them together through notions of somatisation (May et al. 1999).

  • 4

    The original studies from which we drew data were organised around the broad precepts of Glaser and Strauss’ notion of constant comparison, and aimed at producing substantive, rather than ‘formal’ theory (Glaser and Strauss 1967). They thus operated at a relatively low level of abstraction. However, Glaser and Strauss specifically argue against their method as a means of verification or confirmation in theory building. The analysis presented here is a step on from substantive theory, but has been formed through a procedure not dissimilar to ‘frame analysis’ (Mason 1996).

  • 5

    Clearly we wanted our initial analysis to do more than confirm the results of the studies from which data were drawn. However, because not all of the aggregate data were re-analysed, once the coding frame had been set out, refined and employed, it was important to have agreement between the field researchers on each of the studies that the product of this secondary work ‘fitted’ with their intimate knowledge of the data they had collected.

  • 6

    The routine typification of the social world is, as Berger and Luckmann (1966) have noted, an anthropological necessity. Drawing on the work of Schutz (1962) they argue that a repertoire of types makes it possible for subjects to sort and understand the phenomena that they encounter, and to assign meanings to these, and that much social action is fundamentally organised around practices of typification. A parallel psychological model may be found in implicit personality theory, which Bower (1998) has applied to the very set of practices that we describe in this paper.

  • 7

    There is a serious underpinning to this complex of action: the problem is that the doctor cannot discourage the patient from returning, in case some further serious or sinister problem arises and which may pose a danger to the patient. Organising the patient's status in this way while maintaining continuity of trust is a difficult exercise.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Study groups and method
  5. Framing the doctor-patient relationship in chronic illness
  6. Conclusion
  7. Acknowledgements
  8. References
  • Armstrong, D. (1982) The doctor-patient relationship: 1930–80. In Wright, A. and Treacher, A. (eds) The Problem of Medical Knowledge. Edinburgh: Edinburgh University Press.
  • Armstrong, D. (1987) Theoretical tensions in biopsychosocial medicine, Social Science and Medicine, 25, 121318.
  • Atkinson, P. (1981) The Clinical Experience: The Construction and Reconstruction of Medical Reality. Farnborough: Gower.
  • Atkinson, P. (1995) Medical Talk and Medical Work. London: Sage.
  • Balint, M. (1957) The Doctor, his Patient, and the Illness. London: Pitman.
  • Berger, P.L. and Luckmann, T. (1966) The Social Construction of Reality. Harmondsworth: Penguin.
  • Boon, H. and Stewart, M. (1998) Patient-physician communication assessment instruments: 1986 to 1996 in review, Patient Education and Counseling, 35, 3, 161176.
  • Bower, P., Gask, L., May, C. and Mead, N. (2001) Domains of consultation research in primary care, Patient Education and Counseling, 45, 1, 311.
  • Bower, P. (1998) Understanding patients: implicit personality theory and the general practitioner, British Journal of Medical Psychology, 71, 15363.
  • Bury, M. (1982) Chronic illness as biographical disruption, Sociology of Health and Illness, 4, 16782.
  • Byrne, P. and Long, B. (1976) Doctors Talking to Patients. London: Department of Health and Social Security.
  • Calnan, M. and Williams, S. (1995) Challenges to professional autonomy in the United Kingdom: perceptions of general practitioners, International Journal of Health Services, 25, 21941.
  • Campbell, R., Pound, P., Pope, C., Britten, N., Pill, R., Morgan, M. and Donovan, J. (2003) Evaluating meta-ethnography: a synthesis of qualitative research on lay experiences of diabetes and diabetes care, Social Science and Medicine, 56, 4, 67184.
  • Chapple, A. (1997) Menorrhagia: general practitioners’ and womens’ perceptions of this condition and its treatment [Thesis]. University of Manchester. PhD.
  • Chapple, A. B., May, C. and Ling, M. (2001) Is objective testing for Menorrhagia in general practice practical? Results from a qualitative study, European Journal of General Practice, 7, 1, 1317.
  • Charmaz, K. (1995) The body, identity, and self: adapting to impairment, Sociological Quarterly, 36, 4, 65780.
  • Chew-Graham, C. (1999) Chronic low back pain: the interface between explanatory models and medical communications [Thesis]. University of Manchester. MD.
  • Chew-Graham, C. and May, C. (1999) Chronic low back pain in general practice: the challenge of the consultation, Family Practice, 16, 1, 469.
  • Chew-Graham, C.A., May, C., Cole, H. and Hedley, S. (2001) The burden of depression in primary care: a qualitative investigation of general practitioners’ constructs of depressed people in the inner city, Primary Care Psychiatry, 6, 13741.
  • Conrad, P. and Schneider, J. (1980) Looking at levels of medicalization: a comment on Strong's critique of the thesis of medical imperialism, Social Science and Medicine, 14a, 759.
  • Dixon, C. (2000) A multi-method evaluation of a training intervention for general practitioners in the assessment and management of depression [Thesis]. University of Manchester. PhD.
  • Dowrick, C., Gask, L., Perry, R., Dixon, C. and Usherwood, T. (2000) Do general practitioners’ attitudes towards depression predict their clinical behaviour? Psychological Medicine, 30, 2, 41319.
  • Engel, G.L. (1977) The need for a new medical model: a challenge for biomedicine, Science, 196, 129, 136.
  • Fairhurst, K. and May, C. (2001) Knowing patients and knowledge about patients: evidence of modes of reasoning in the consultation, Family Practice, 18, 5, 50105.
  • Glaser, B.G. and Strauss, A. (1967) The Discovery of Grounded Theory. Chicago: Aldine.
  • Good, B.J. (1994) How medicine constructs its objects. In Good, B.J. (ed.) Medicine, Rationality and Experience. Cambridge: Cambridge University Press.
  • Gothill, M. and Armstrong, D. (1999) Dr. No-body: the construction of the doctor as an embodied subject in British general practice 1955–97, Sociology of Health and Illness, 21, 1, 112.
  • Hawker, S., Payne, S., Kerr, C., Hardey, M. and Powell, J. (2002) Appraising the evidence: reviewing disparate data systematically, Qualitative Health Research, 12, 9, 128499.
  • Jewson, N. (1976) The disappearance of the sick man from medical cosmology, 1770–1870, Sociology, 10, 3, 22544.
  • Kelly, M.P. and Field, D. (1996) Medical sociology, chronic illness and the body, Sociology of Health and Illness, 18, 2, 24157.
  • Kendall, L. (2001) The Future Patient. London: Institute of Public Policy Research.
  • Lupton, D. (1994) Medicine as Culture: Illness, Disease and the Body in Western Societies. London: Sage.
  • Malterud, K. (1999) The (gendered) construction of diagnosis interpretation of medical signs in women patients, Theoretical Medicine and Bioethics, 20, 3, 27586.
  • Malterud, K. (2002) Reflexivity and metapositions: strategies for appraisal of clinical evidence, Journal of Evaluation in Clinical Practice, 8, 2, 1216.
  • Mason, J. (1996) Qualitative Researching. London: Sage.
  • May, C. (1992) Individual care? Power and subjectivity in therapeutic relationships, Sociology, 26, 4, 589602.
  • May, C., Doyle, H. and Chew-Graham, C.A. (1999) Medical knowledge and the intractable patient: the case of chronic low back pain, Social Science and Medicine, 48, 52334.
  • May, C. and Mead, N. (1999) Patient-centredness: a history. In Dowrick, C. and Frith, L. (eds) Ethical Issues in General Practice: Uncertainty and Responsibility. London: Routledge.
  • Mead, N. and Bower, P. (2000a) Measuring patient-centredness: a comparison of three observation-based instruments, Patient Education and Counseling, 39, 1, 7180.
  • Mead, N. and Bower, P. (2000b) Patient-centredness: a conceptual framework and review of the empirical literature, Social Science and Medicine, 51, 7, 10871110.
  • Mead, N. and Bower, P. (2002) Patient-centred consultations and outcomes in primary care: a review of the literature, Patient Education and Counseling, 48, 1, 5161.
  • Mead, N., Bower, P. and Hann, M. (2002) The impact of general practitioners’ patient-centredness on patients’ post-consultation satisfaction and enablement, Social Science and Medicine, 55, 2, 283299.
  • Moran, P.J. (1999) Governing the Health Care State: A Comparative Study of the United Kingdom, the United States, and Germany. Manchester: Manchester University Press.
  • Neighbour, R. (1987) The Inner Consultation. London: Kluwer.
  • Parsons, T. (1951) The Social System. London: Routledge and Kegan Paul.
  • Platt, F.W., Gaspar, D.L., Coulehan, J.L., Fox, L., Adler, A.J., Weston, W.W., Smith, R.C. and Stewart, M. (2001) ‘Tell me about yourself’: The patient-centered interview, Annals of Internal Medicine, 134, 11, 107985.
  • Rogers, A., May, C. and Oliver, D. (2001) Experiencing depression, experiencing the depressed: patients and doctors’ accounts, Journal of Mental Health, 10, 3, 31734.
  • Salmon, P. (2000) Patients who present physical symptoms in the absence of physical pathology: a challenge to existing models of doctor-patient interaction, Patient Education and Counseling, 39, 1, 105113.
  • Schutz, A. (1962) Collected Papers. The Hague: Nijhoff.
  • Shorter, E. (1991) Doctors and their Patients: A Social History. New Brunswick, NJ: Transaction.
  • Sinclair, S. (1997) Making Doctors: An Institutional Apprenticeship. Oxford: Berg.
  • Stacey, M. (1991) Medical sociology and health policy: an historical overview. In Gabe, J., Calnan, M. and Bury, M. (eds) The Socology of the Health Service. London: Routledge.
  • Stevenson, F.A., Barry, C.A., Britten, N., Barber, N. and Bradley, C.P. (2000) Doctor-patient communication about drugs: the evidence for shared decision making, Social Science and Medicine, 50, 6, 82940.
  • Stimson, G. and Webb, B. (1975) Going to See the Doctor. London: Routledge and Kegan Paul.
  • Stott, N. and Davis, R. (1979) The exceptional potential in every primary care consultation, Journal of the Royal College of General Practitioners, 29, 20105.
  • Strong, P. (1979a) Sociological imperialism and the profession of medicine, Social Science and Medicine, 13a, 2, 199215.
  • Strong, P. (1979b) The Ceremonial Order of the Clinic: Patients, Doctors and Medical Bureaucracies. London: Routledge.
  • Thorne, S., Paterson, B., Acorn, S., Canam, C., Joachim, G. and Jillings, C. (2002) Chronic illness experience: insights from a metastudy, Qualitative Health Research, 12, 4, 43752.
  • Tuckett, D., Boulton, M., Olson, C. and Williams, A. (1985) Meetings between Experts: An Approach to Sharing Ideas in Medical Consultations. London: Tavistock.
  • Turner, B.S. (1995) Medical Power and Social Knowledge. 2nd Edition. London: Sage.
  • Wileman, L., May, C. and Chew-Graham, C. (2002) Medically unexplained symptoms and the problem of power in the primary care consultation: a qualitative study, Family Practice, 19, 2, 17882.
  • Williams, S.J. (2000) Chronic illness as biographical disruption or biographical disruption as chronic illness? Reflections on a core concept, Sociology of Health and Illness, 22, 1, 4067.
  • Wissow, L.S. (2003) Untitled, Patient Education and Counseling, 49, 2, 103.