Introduction

Medical diagnosis explains, legitimises and normalises. Following Balint (1964), it provides a sense of order in the presence of disarray. Diagnosis explains certain kinds of deviance in terms of disease rather than of moral failing (Conrad, 1975), and it situates dysfunction under the jurisdiction of medicine, rather than under that of another authority (Zola, 1972). Diagnosis provides a trajectory of treatment, prognosis and, in some cases, prevention. It provides social identity, removing the patient from his or her isolation (Chiong, 2001). However, diagnosis also controls, compelling one to become obedient to a new set of normative obligations including incapacity and therapeutic compliance (Parsons, 1958). It controls the allocation of resources (De Swaan, 1989; Dumit, 2006; Jutel, 2006b), stymies expression (Munro, 2002) and stigmatises (Sokratis et al, 2004; Raz and Vizner, 2009). And, diagnosis affects outcomes. As Hamilton and colleagues point out, different disease labels when a range of options is available may result in a different prognosis and different social outcomes (Hamilton et al, 1996). Haynes et al (1978) reported that labelling patients hypertensive, for example, increased absenteeism from work.

When diagnosis has such power to frame social reality, the issue of patient complaints that cannot be diagnosed challenges both medicine and lay people. In this article, I will explore medicine's discursive approach to illness that cannot be diagnosed. Specifically, I focus on what the medical literature refers to widely as ‘medically unexplained symptoms’. This phrase, which on the surface appears benign, is at the root of important lay-professional contests. The absence of diagnosis denies the patient an explanatory framework, a treatment, access to the sick role and legitimisation of the complaint. A patient who truly feels poorly, but for whom no medical explanation can be given is marginalised by his or her inability to, in Balint's words, ‘propose an illness’ that the doctor can recognise as a disease (Balint, 1964).

This article demonstrates that the discursive approach of both medical literature and practice is paradoxical. Multiple unexplained causes for consultation become a unitary condition under an informal diagnostic category linked to psychiatric dysfunction. Medicine fails to note the limitations of its episteme, creating a catch-all psychogenic diagnosis in the absence of a suitable existent label. In the following pages, I make a critical appraisal of this approach and show how discursive reliance on diagnosis in discussion of the medically unexplained symptom shifts responsibility for illness to the patient and interferes with the interpretative role of medicine.

The Symptom

Foucault (1963) described how the symptom took on a new role in nineteenth century clinical medicine. Previously the locus of medical interest on its own, it became instead a key, a clue or a pointer to deeper and more significant conditions. Foucault explained that the arrangement and configuration of symptoms provide clarity and insight to otherwise hidden disease. ‘Cough, chest pain and shortness of breath are not pleurisy … they are its essential symptoms’ (1963, p. 89, my translation). No longer did clinicians consider symptoms as individual phenomena; they became the external expression of disease. This interpretive change transformed the doctor's role into one of detective: unscrambling the messages of the symptom to discover the link between signifier and pathology. Diagnosis is, according to Eliot Freidson (1972), the foundation of medicine's claim to authority. ‘Where illness is the ubiquitous label for deviance in an age, the profession that is custodian of the label is ascendant’ he wrote (p. 244).

Drew Leder (1990) describes this stealth work as an interpretive process. The patient brings a situation that they are not fully able to interpret but have assigned to the realm of the medical to the medical consultation. In their desire to understand the discomfort, the cough, the rash, the swelling or the cramp, they initiate an encounter with the doctor. This is what the doctor will refer to as the symptom: a ‘subjective’ experience or change in function. In turn, the doctor channels the narrative of illness or dysfunction towards an interpretive outcome, asking questions about personal and family history, querying both the presentation and absence of symptoms. The doctor then undertakes a physical assessment, exploring the body-object, now silent while palpated, auscultated and measured, to create a narrative that explains the patient's complaint. And finally, both patient and doctor acquiesce to the numerical findings contained in instrumental texts such as x-ray, scans, blood count or manometer.

The illness-disease dichotomy is pivotal in the diagnostic process. On the one hand, illness is the undesirable change in social or personal function that the individual brings to the medical consultation in search of remedy. Disease, on the other hand, is medicine's categorisation of illness: a biological or psychophysiological dysfunction or mal-adaptation (Kleinman et al, 1978). Illness originates from a cultural context, and in turn influences the response to, or decision to access, medical services. That illness and disease are not identical concepts, and that disease labels enable access to services and status that are not otherwise available is salient in this discussion.

When illness is presented to the doctor, the diagnostic trajectory is guided by a presumptive paradigm, and measured by clinical efficacy. This means that pre-diagnosis conceptions frame the manner by which the doctor explores the complaint, and prevent an aimless ‘fishing trip’ of non-systematic explorations and questions. The collection of symptoms along with the patient's story point the doctor in a particular direction, suggest a collection of potential diagnoses or explanations, and reduce the complexity of decision making (Croskerry, 2002). It also means that the encounter cannot be considered successful if medical interpretation emerging from the interaction does not result in clinical improvement for, or cooperative comprehension by, the patient.

Yet, studies propose that anywhere from 10 to 35 per cent of primary consultations fail to alleviate the presenting complaint (Peveler et al, 1997; Feder et al, 2001; Sharpe and Carson, 2001; Reid et al, 2002; Carson et al, 2003). Frustrated by symptoms for which they cannot find a recognised medical diagnosis, doctors devote much energy to explaining and identifying a presumed root cause. This entails a significant financial and structural impact on health care (Reid et al, 2002; Woivalin et al, 2004), generates inappropriate referrals and treatment (Salmon et al, 2006) (Mayou and Sharpe, 1997), engenders physician frustration (Reid et al, 2001; Salmon et al, 2005; Ringsberg and Krantz, 2006), and of course, results in distress for the patient who generally consults with the hope of identifying an explanation and concordant cure (Malterud, 2000; Katon et al, 2001; Nettleton, 2006). One of Nettleton's interviewees explained: ‘There is a lurking fear – in the dark out there that I may be faking it. It's an awful moment, a really nasty moment a total put down when the tests don’t show anything. Despair. Quite a put down’ (p. 1173).

Methods

I undertook this study in the spirit of a traditional literature review, rather than of a systematic review, using a Foulcauldian driven discourse analysis.Footnote 1 My choice to perform a traditional literature view was driven by my interest in the discursive content of the medical articles – analysis of which cannot be captured in a system-based process. It relies instead upon what MacLure (2005) refers to as intertextual connectivity, critique, interest, expertise, independence, tacit knowledge, chance encounters with new ideas, and dialogic interactions between researcher, “literature” and “data” (p. 394). Although this process requires greater levels of subjective judgement and intuition on the part of the researcher than the sciences might normally tolerate, it leads to highly informed judgements based on extensive experience. Importantly, these judgements are subjected to levels of scrutiny and interrogation at least equal to, if not more demanding than, those in the methodologically driven systematic review.

I obtained the material through a straightforward search of the pubmed database for all articles containing the terms ‘medically unexplained symptom’ or ‘medically unexplained symptoms’ (MUSFootnote 2) in the title. My search terms were intentionally explicit. I did not wish to include terms that others might consider related, such as ‘somatoform’ disorder, ‘somatising’ or ‘functional somatic syndromes’ as the point was to consider the linguistic configuration and the use made of the term by its authors rather than to associate it with alternate and perhaps value-laden terms. After initially locating 77 references, I excluded two that proved to be unobtainable.

I classified the 75 articles according to whether they were research reports or perspective/opinion pieces, and then read them to get a sense of common themes or assumptions. I then re-read the articles and classified them according to the nine themes I had identified. The themes included:

  • referring to MUS as an entity to be diagnosed, treated and prevented;

  • using the term MUS synonymously with somatoform disorder, somatising, functional somatic syndrome or other related psychogenic term;

  • referring to MUS as a burden for the doctor;

  • referring to MUS as a burden for the patient;

  • referring to MUS as an economic burden;

  • describing MUS as resulting in (not from) psychological stress;

  • attributing it to poor diagnostic skills;

  • presenting a contrasting view critical of the tendency to psychologise, or of diagnosis by exclusion; or

  • clearly enunciating a non-psychogenic cause to MUS.

In addition to these themes – which themselves could be classified as psychogenic assessment of MUS and non-psychogenic assessment of MUS – I assessed additional assumptions contained in each individual article through careful reading and textual interpretation.

These classifications and associated note-taking assisted me to synthesise and build discussion around the prominent debates, assumptions and contradictions that this body of literature presented.

Findings

A critical review of these articles demonstrates a growing concern with the burden that medically unexplained symptoms present for patients, doctors and the health-care system, but also reveals a curious approach to the problem. Paradoxically, many of the articles resort to creating a catch-all diagnostic category in which it can place the unexplained.Footnote 3

Half of the articles use the phrase ‘medically unexplained symptoms’ interchangeably with psychiatric diagnostic terms such as somatoform disorder, somatising, functional somatic syndrome or other related terms implying psychogenic or sociogenic origin. Although medically unexplained symptoms are an important diagnostic component of these conditions, in these articles, they are subsumed within the category of psychiatric disorder and used synonymously. Less than a quarter of these articles proposed views critical of the tendency to provide a psychogenic explanation, to diagnose by exclusion or to recognise the limitations of medical knowledge vis-à-vis the physical complaints of patients. I discuss this in greater length below.

But whether the publications see medically unexplained symptoms as psychogenic or not, there is a pervasive reference to them as a unified condition that could be considered under one light, a kind of diagnosis of the undiagnosable. This is not completely surprising. Diagnoses are not only the classification tools of medicine, but also the means by which medicine does its work. As an act of classification, diagnosis is organisational, informational and embedded in practice (Bowker and Star, 1999). As with all classifications, it is potent: valorizing some points of view, and silencing others.

The literature links medically unexplained symptoms semantically with phrases such as ‘the problem of medically unexplained symptoms’; ‘diagnosing medically unexplained symptoms’; ‘treating medically unexplained symptoms’. This semantic formulation reduces chronic fatigue, lower back pain, irritable bowel syndrome and so on to the unitary label of medically unexplained symptoms. The use of the acronym ‘MUS’ reifies the notion that all physical complaints without explanation can be viewed in the same way. However, it is not just a matter of nomenclature; references to medically unexplained symptoms give them diagnostic status by framing them as a problem to be approached epidemiologically (Brown, 2004; Rosendal et al, 2005; Smith et al, 2005; Spence et al, 2005; Verhaak et al, 2006). Other articles communicate the unified nature of these symptoms by referring to them as a treatable or preventable condition (Rosendal et al, 2005; Morriss et al, 2006; Smith et al, 2006; Van der Feltz-Cornelis et al, 2006). And another group refers to sufferers as if they were a homogenous group (Jackson et al, 2004; Salmon et al, 2005; Bakal et al, 2006; Epstein et al, 2006; Morriss et al, 2006; Ringsberg and Krantz, 2006; Smith et al, 2006, 2005). This word use confirms an approach that reasons through diagnosis rather than interpretation and that contains ontological and psychiatric assumptions about that which is not readily available for medical classification.

Psychiatry to Explain the Unexplainable

Rosenberg (2006) writes that ‘Psychiatry is the residuary legatee of … developments that have always been contested at the ever-shifting boundary between disease and deviance, feeling and symptom, the random and the determined, the stigmatised and the value-free’ (p. 237). He seeks here to account for how disease categories explain behaviour and emotional pain. This is relevant to medically unexplained symptoms as the seemingly unjustified symptoms are ‘behaviours’ – repeated visits to doctors’ surgeries, complaints that do not translate into biophysical evidence, discomfort that does not respond to conventional medical advice – in a model that uses psychiatry to explain the unexplained. This psychiatric domiciliation of the unexplained is problematic because it frequently relies upon diagnosis by exclusion, that is, it is settled upon by ruling out other explanations, failing to acknowledge the limitations of knowledge, and by shifting responsibility for cause and cure in a way that ignores socio-historical realities as I show below.

First, to consider unexplained symptoms a psychiatric disorder is to make a diagnosis by exclusion; that is to say, it is a diagnosis that is made not on the basis of what it is, but on the basis of what it is not. The absence of explanation, rather than the presence of a well-defined feature, is pathognomic of the condition. The label becomes a wastebasket diagnosis – a classification in which may be placed many conditions that do not easily fit elsewhere. As Robins and Helzer (1986) point out, psychiatric diagnosis is already challenged by the absence of definitive tests to mark the presence of a biomedical disorder; there is no equivalent to the x-ray in orthopaedics or haemoculture in infectious disease. It is therefore difficult, according to Lipowski (1967), to establish a positive diagnosis, yet, facile to attach a positive psychiatric diagnosis to a medical patient. Forty years ago, he called for a theory of somatic symptoms and practical guides to differential diagnosis (Lipowski, 1967). De Gucht and Fischler (2002) confirm that it is still difficult to confirm the somatic nature of symptom; somatisation remains a complex concept that presumes a causal relation between the psychiatric condition and the unexplained symptom.

Second, this assumption of psychiatric disorder in the absence of other diagnoses presumes the infallibility of the doctor and of the field. Yet, one post-mortem study revealed discrepancies in major diagnosis of 39 per cent and of minor diagnoses in 50 per cent of cases, respectively (Aalton et al, 2006). Graber and colleagues discuss the number of diagnostic errors that we can retrospectively assign to the incomplete knowledge of the field. For example, Lyme's disease, before its designation as a specific diagnosable entity, was often cast as atypical rheumatoid arthritis (Graber et al, 2002). Similarly, Skegg et al (1988) revealed that multiple sclerosis was frequently diagnosed as a psychiatric disorder, possibly owing to the subjective nature of many early symptoms. Symptoms may also be atypical, presenting challenges for clinician and patient alike (Malterud and Taksdal, 2007).

Advances in genetics provide links and explanations between pathologies that were previously seen as completely distinct before the emergence of genetic explanations for illness (Miller, 2006). Kirmayer et al explain that ‘the lack of explanation reflects the limits of medical knowledge, available technology and the epistemological difficulties of assigning a clear cause to subjective complaints like pain and fatigue, which may have no objectively measurable correlates and may change rapidly over time in quality and intensity’ (Kirmayer et al, 2004, p. 663).

Third, this approach to the unexplained symptom shifts responsibility for the inability to explain the symptom from the doctor to the patient, positing the cause in patient mental health. Hadler reflects upon both the causes and the limitations of this catch-all psychogenic labelling, calling it a ‘contest of diagnosis’, and describing medicine as ‘not [being] likely to accept blame for subjecting the patient to months of an exercise that turn[s] out to be flawed in design and iatrogenic in execution’ (Hadler, 1996). Nettleton (2006) similarly reports the important effort of trying to compensate for and palliate symptoms in the absence of a diagnosis as being as onerous as the symptoms themselves. ‘Society does not readily give people permission to be ill’, she writes, ‘in the absence of an “accepted” abnormal pathology or physiology’ (p. 1167).

Pilowsky (1994) confirms this contest as weighted towards the physician, defining abnormal illness behaviour as an ‘inappropriate or maladaptive mode of experiencing … one's own state of health which persists despite the fact that an appropriate social agent … has an accurate and reasonably lucid explanation of the nature of the person's health status’ (p. 567). He points out that the doctor's potential to err is an important liability of this definition and that in some cases, there is a hasty progression from lack of explanation to considering a patient's behaviour ‘abnormal’. Salient in Pilowsky's definition is the recognition of the important, yet equivocal role of the ‘social agent’ or doctor.

Locating the symptom in psychology ascribes responsibility and moral blame in such a way that it encourages patients to resist (Kirmayer et al, 2004). While a physical explanation for a symptom confers, in most cases, a lack of personal responsibility for its onset, a psychiatric one implies that the patient might have the ability to both control and reverse the physical symptoms, an interpretation that may seem impossible to the sufferer and that brings with it stigma and shame. It is important to note at this juncture, the significant adverse role of distress in the effectiveness of medical treatment (Querido, 1959; Levine et al, 1996; Kamolz et al, 2001).

Finally, this psychogenic explanation relies for diagnostic criteria upon historically and socially developed patterns of behaviour that emerge from the expectations of modern medicine. An important criterion in the ‘diagnosis’ of MUS (acronym use intentional) is frequency of visits to General Practitioners (GPs). Repeated consultation for one or many complaints for which no ready explanation of disease is found may be the result of compliance to social expectations about the role of the doctor and medicine in the quotidian as I have introduced above. For example, a mother is rewarded for the medicalisation of her children's health. Many countries provide financial or other incentives for parents to engage in ‘well-child’ check-ups. Direct-to-consumer advertising admonishes the public to ‘Check with your doctor and find out if X is right for you!’ using enticing and effective models to elicit consumption of medical care generally, and of specific pharmaceutical products (Jutel and Buetow, 2007). General medical practice may include such non-medical services as family planning and ear-piercing. Women are far greater consumers of these services than men, and are more often subject to and compliant with calls to consult the doctor, resulting in a greater prevalence of ‘diagnosed’ MUS in women. This compliance, historically learned from the medicalisation of women's health, unequally slots them into the category of psychiatric illness by exclusion. Malterud's work submits that clinical approach is based on gender, and traps women into diagnostic categories and treatment protocols to their detriment (Malterud, 1999, 2000, 2001).

Health promotion organisations, on the other hand, lament the non-compliance of men, who are less likely to see their doctor regularly than women. A growing body of literature points out that men tend to delay help-seeking behaviour, are less likely to consult a GP until well past the propitious time (Galdas et al, 2005) and are more reliant upon social factors than on the gravity of symptoms to determine when they will consult (Wolters et al, 2002). Tardy help-seeking behaviour is proposed as one of the reasons for poorer health outcomes in men relative to women (Galdas et al, 2005). Addressing this deficiency is an important focus of men's health initiatives (McKinlay, 2005). As men are often reluctant to consult a GP, alternatives have been developed, including using sports clubs and pubs as venues for health promotion activities.

How Then to Explain the Unexplained?

I have shown the limitations of ‘diagnosing’ medically unexplained symptoms as an entity. It is a diagnosis of exclusion, which presumes the infallibility of the physician and the omniscience of medicine. It transfers responsibility for the disorder to the patient in a way that hinders cure, and it ignores socio-historical practices that influence patient behaviour and that place women unequally under scrutiny for psychogenic disorder.

There is, however, a range of ways of explaining complaints that are not linked with disease. Returning to Leder's hermeneutic contextualisation of the patient–doctor encounter, we can easily see that one explanation is that the patient has incorrectly attributed the origin of her complaint to the realm of medicine. Although the patient's account may not end up having a medical explanation, medicine's role as arbitrator may lead the patient to presume that it does. And, as industry marketing uses the stamp of medical approval to monger its wares, this enhances the chances of the patient making assumptions about the medical nature of the complaints (Conrad and Leiter, 2004). ‘Disease mongering’, a term coined by Lynn Payer in 1992, and more recently integrated in the academy in discussions of medicalisation, describes the degree to which a range of interested parties promote medical solutions to the public (Moynihan et al, 2002; Wolinsky, 2005; Tiefer, 2006; Jutel, 2006a). Heart burn, back pain, overweight (exercise not a condition) and mood swings, to name only a few, are examples of potentially non-medical conditions whose management is framed with the directive: ‘ask your doctor’.

Many serious symptoms and dysfunctions affecting digestion, mobility, comfort and kinaesthesia may be quite clearly ‘real’ without being medical. The tortuous course of lower back pain in the absence of structural pathology may be explained by a muscle-balance assessment, and ‘cured’ by development of core strength and exercise prescription, rather than surgical or medical or pharmacological solutions (Abenhaim et al, 2000). Similarly bloating, flatulence, diarrhoea and other digestive system distress that cannot be explained by coeliac disease or other defined clinical entity may still be related to dietary intolerances and obscure food allergies that a nutritionist could better address (see, for example, Meador's (2005) unusual case study of a specific brand of toothpaste as a causative factor in intractable diarrhoea). A biomechanical explanation may be at the heart of any one of a number of functional disorders such as lower back pain, and occupational overuse syndrome that an occupational therapist, clinical biomechanist or podiatrist would be better suited than either physician or surgeon to diagnose and correct.

In the same way, unknown and unexpected environmental factors may explain any number of respiratory, gastro-enteric, cutaneous and functional symptoms. Many conditions that are medically unexplained overlap strongly with explained conditions, such as asthma. Environmental factors are an under-researched area of potential explanation and are the basis of contested diagnoses such as in the case of unacknowledged environmental contamination (Brown, 1995; Kipen and Fiedler, 2002).

A condition may be medical, but not recognised as such. I have provided explanations above, but add here Aronowitz's (2001) assertion that socio-cultural and political, as much as biological and clinical factors influence how symptom clusters can be classified as diagnoses. He describes expert consensus panels and their use both of medical evidence and of practitioner interests in the production of disease labels. Notably, inclusion in the international classification of diseases is by application. An example of one such application is proposed as an exemplar on the Australian National Centre for Classification in Health. At odds with recognised scholarly practices, it references no publications. This particular application argues for the recognition of a new variant of disease on the basis that ‘this change would be acceptable to the whole of the Western world as it better reflects the current clinical understanding of … [disease name]’ writes the author with no justification. The exemplar cites no evidence beyond consultation with expert clinicians, and presumes an unanimity which is similarly unsupported (National Centre for Classification in Health, 2007). Some conditions are functional and treatable, but are not linked to a particular pathology. Radiotherapy damage, whiplash sequelae and even the common cough (Goldstein, 2007) can be discrete entities of their own that we could perhaps understand as physical dysfunctions, rather than as diseases.

Finally, we need to consider that the explanation may indeed be psychiatric. Physical symptoms are recognised as part of psychiatric syndromes. For example, they commonly coexist with psychological signs and symptoms in major depressive disorder (Arnow et al, 2006); are part of the diagnostic criteria for generalised anxiety (American Psychiatric Association, 2000), are often linked with psychosis and so on. Psychiatry is a legitimate form of medicine and has a diagnostic framework for disease. A psychiatrically explained symptom, is hence, not medically unexplained. However, psychiatric diagnosis should not be by exclusion; rather it should be based on positive findings in the context of a comprehensive biopsychosocial assessment.

Conclusion

Despite constructing this critical review of the way medical literature deals with unexplained, somatic complaints, I am aware that medicine is attentive to at least some of the contradictions in its approach. Pilowsky's 1994 work, as discussed above, makes important points about the risks of jumping to conclusions about illness behaviour. Almost half of the articles in this review that made reference to medically unexplained symptoms also expressed doubt about the psychogenic assumptions underpinning the management of medically unexplained symptoms. Interestingly, however, so too do the more recently published accounts of illness written by doctors-as-patient. In the articles evaluated above, the authors of a number of editorials and opinion pieces shared their own experiences of idiosyncratic illness (Sowby, 2003; Yodaiken, 2003), some specifically questioning whether medical limitation was at the root of the inability to assign a diagnostic explanation for many symptoms (Morrison, 2003). ‘Perhaps funds directed to MUS might be directed to better training of doctors in some common clinical problems that are misdiagnosed rather than unexplained’ (Goadsby, 2003). Others lamented the blame and shame associated with psychogenic explanations for physical symptoms (Butchart, 2005; Watkins, 2002). An increasing focus on patient accounts relevant to the practice of medicine and ethics is captured in the narrative work of Arthur Frank, Arthur Kleinman, Kathryn Montgomery, Oliver Sacks and others and also in new sections in such publications as Annals of Internal Medicine (On Being a Patient) and the British Medical Journal (Patient's Journey), which focus on patient perspective.

That these two publication types are at odds with the corpus is not completely surprising. Much medical research seeks to develop a mechanistic understanding of diseases and populations in which they occur, reducing the human condition into physiochemical terms. By contrast, clinical practice focuses on the idiosyncratic individual. Science explains disease, while physicians treat patients. Cassell (1991) writes: ‘the scientific basis of medicine does not recognise nor provide a methodology to deal with such individual variations on the level of patient-doctor interactions’ (p. 20). Doctors’ own accounts of illness including their experiences as patients concur with this differentiation: their narratives punctuate the uniqueness of experience that medicine, in general terms, ignores in its classificatory project.

Undeland and Malterud (2002) maintain that we err by seeing diagnosis as the be-all and the end-all of the medical consultation. The encounter between patient and doctor can be recast as a help-seeking interaction rather than pursuit of diagnosis. This way, serious matters can be diagnosed, and symptoms managed; but also, matters that end up being non-medical in nature can be so identified, and perhaps referred on. This is a particularly important point, if, as Stone et al (2002) suggest, ‘Some diagnostic labels, particularly those that sound “psychological”, can be perceived by patients as offensive by implying that the patients are “putting on” or “imagining” their symptoms or that they are “mad” ’ (p. 1449).

Stone and his colleagues (2002), who explore the response of patients to particular diagnoses explain that diagnostic labels have the ability to offend the patients so labelled, particularly those that imply a psychological rather than a physical explanation. They write that ‘Although “medically unexplained” is scientifically neutral, it had surprisingly negative connotations for patients’. I do not find this surprising, because, as I have demonstrated above, although the label is neutral, its use is anything but so. Stone calls for the restoration of the term ‘functional’ that, while value laden in the medical psyche, appears to be acceptable in that of the patient. The functional disorder provides ‘a rationale for pharmacological, behavioural, and psychological treatments aimed at restoring normal functioning … a useful and acceptable diagnosis for physical symptoms unexplained by disease’ (Stone et al, 2002).

I propose that clinical interpretation rather than clinical diagnosis is a way to reconcile medicine and the individual, the unexplained symptom and the patient distress. For this we return to Leder's discussion of clinical hermeneutics as the basis for the patient–doctor encounter. Leder's interpretive approach to medicine is not considered successful on the basis of accurate diagnosis; the hermeneutic telos, or objective of interpretation, is not to diagnose, but to explain. Medicine then leaves its role as purified scientific endeavour with its classificatory frameworks and generalisations to recognise the texts within which both patient and doctor are working. According to Leder (1990), medicine has undertaken a flight from interpretation, instead, seeking what it sees as clarity and objectivity, refuting the subjectivity, ambiguity and opacity inherent to its interpretive endeavour.

Returning to its interpretive ideal, medicine would authorise patient accounts of illness even in the absence of disease, and would permit alternative frameworks for explaining the patient complaint. Medicine could then interpret more responsibly, using diagnosis as one of many potential outcomes.