Pergamon Soc. Sci. Med. Vol. 46, Nos 4-5, pp. 559-565, 1998 © 1997 Elsevier Science Ltd. All rights reserved

PII: S0277-9536(97)00200-1 Printed in Great Britain 0277-9536/97 $19.00 + 0.00

PATIENTS WITH M E D I C A L L Y U N E X P L A I N E D SYMPTOMS: SOURCES OF PATIENTS' AUTHORITY A N D

IMPLICATIONS FOR D E M A N D S ON MEDICAL CARE

SARAH PETERS) '2 IAN STANLEY, 2 MICHAEL ROSE 3'4 and PETER SALMON ~*

~Department of Clinical Psychology, University of Liverpool, Liverpool, U.K., 2Department of Primary Care, University of Liverpool, Liverpool, U.K., 3Department of Physiotherapy, University of

Liverpool, Liverpool, U.K. and 4Back Pain Programme, Wirral Hospital, Liverpool, U.K.

Abstract--Lay and medical beliefs are not separate systems. The beliefs of somatizing patients, in par- ticular, incorporate medical understanding and it has been argued that this increases the power that such patients exert in seeking treatment from doctors. To understand the nature and use of this power requires investigation of (i) how patients use medical ideas and language to explain their symptoms and (ii) how this process influences patients' expectations and evaluations of their doctors. We interviewed 68 patients, in whom no physical cause had been found for persistent physical symptoms. Their accounts of symptoms and of their experience of doctors were subjected to qualitative thematic analy- sis. As expected, patients used medical terms to explain their symptoms. However, these depicted expla- natory themes which have long been familiar in traditional lay models: disease as a malign entity and imbalance between bodily forces. Patients' sense of authority over doctors derived, not from facility with medical language and ideas but from contrasting their own sensory, and therefore infallible, ex- perience of symptoms with doctors' indirect and fallible knowledge. By providing explanations that questioned the reality of symptoms, doctors were perceived as incompetent and inexpert. Patients used their authority, not to seek treatment, but to secure naming of, and collaboration against, the disorder. Although these patients saw the doctors' role as limited and inexpert by comparison with their own, our analysis suggests ways in which doctors might more effectively engage with persistent somatizing patients. © 1997 Elsevier Science Ltd. All rights reserved

Key words--somatization, beliefs, doctor-patient communication

INTRODUCTION

Patients' beliefs about their symptoms are import- ant because they influence the nature of the symp- toms that are experienced (Pennebaker, 1982; Waddeil et al., 1980; Waiters, 1961), the ways in which patients manage their symptoms, and patients' interactions with health care professionals (Borkan et al., 1995; Marchant-Haycox and Salmon, 1997; Salmon and May, 1995). The term "lay beliefs" has been used to denote ideas that are culturally or personally based rather than directly attributable to medical understanding. Anthropological studies have identified such beliefs, in particular notions of bodily imbalance and pen- etration by external forces (Heiman, 1994).

It has become clear that lay explanations are not only rational and sophisticated, but also no less "scientific" than those held by doctors and in prin- ciple may be no different (Blaxter, 1989). Moreover, it is no longer helpful to view lay and medical beliefs as distinct systems. The growing importance in popular culture of biomedical models of the body and disease has been noted by Kleinman (1978) and Helman (1994). More recently, Shorter's

*Author for correspondence.

(1992) historical analysis suggests that medical accounts of common physical symptoms continually diffuse into lay discourse about biology and illness.

According to Shorter, this process is particularly important in patients who present physical symp- toms which are inexplicable in terms of organic pathology but which may in some cases reflect psychological distress (Fink, 1996). These patients are characterized by frequent contact with health- care professionals (Speckens et al., 1996). It has been suggested that their assimilation of medical beliefs both shapes the way in which symptoms are articulated (Fink, 1996; Shorter, 1992) and increases their power in interaction with doctors. That is, such patients are thought to present and explain bodily experiences in ways that engage but also constrain doctors' responses so as to access physical treatment (Marchant-Haycox and Salmon, 1997; Salmon and May, 1995). This process therefore incurs iatrogenic risk.

Previous qualitative studies in patients who pre- sent unexplained physical symptoms have revealed beliefs, organized in complex ways, that relate bod- ily experiences to social and emotional phenomena (Borkan et al., 1995; Helman, 1985). In contrast, the focus of the present study was to relate patients' explanations of symptoms to their medical encoun-

559

560 Sarah Peters et al.

ters. We investigated a group of patients with at least a 12-month history of unexplained physical symptoms. The value of this extreme group for the present purpose arises from their high level of con- tact with doctors and consequent intense exposure to medical ideas. We first explored patients' use of medical ideas and language in their understanding of symptoms. Secondly, we sought to understand the part played by medical ideas in their require- ments and evaluation of interactions with doctors. This allowed us to reexamine the view that patients' use of medical ideas shapes their expectations o f - and increases their power over - - doctors.

METHOD

Subjects

The study arose from an investigation of the effects of physical exercise on patients with persist- ent unexplained physical symptoms. All general practitioners (GPs) in Liverpool and an adjacent health district were invited to refer patients with physical symptoms which (i) had persisted for at least one year and (ii) which remained unexplained despite specialist referral and investigation. Patients were excluded if they had a history of ischaemic heart disease or psychotic disorder or were being treated for hypertension.

In all, 323 patients were referred and 228 were recruited to the exercise study; 15 referred patients were unsuitable for the study (i.e. did not meet the inclusion criteria above or were being treated for hypertension) and the remainder failed to attend for interview or declined to take part. All patients were interviewed before joining one of two exercise pro- grammes and relevant parts of each interview were transcribed from notes immediately afterwards. The first 40 transcripts were used to refine the interview. Thereafter, 188 patients were interviewed and the analysis reported here is based on 68 transcripts, chosen at random from these.

Patients' ages ranged from 19-72 years; 36 (53%) were female. Current symptoms were mainly pain (localized or extensive) and fatigue but included also bowel problems, respiratory problems, dizzi- ness, sweating and nausea. Analysis of GP records for the 6 months before referral confirmed a gener- ally high level of contact with medical care (Office of Population Censuses and Surveys, 1995): attend- ances at the GP ranged from 0-21 (median: 12) and documented contacts with secondary care specialties (e.g. outpatient appointments) ranged from 0-12 (median: 2).

Procedure

Interviews. Patients were interviewed individually by one author (SP). The pace and sequencing of each interview depended on the patient: the inter- viewer avoided closed questions and encouraged

patients to talk in their own ways about their symp- toms and their experiences of treatment. If necess- ary, patients were prompted to describe: (i) current symptoms; (ii) history of illness, medical investi- gation and treatment; (iii) explanations of symp- toms; and (iv) perceptions of doctors' role. The interviewer recorded, verbatim, any statement which was relevant to the patients' perception of symptoms and health care.

Analysis. The analysis of patients' statements was inductive. That is, no categories were specified in advance. Instead, the procedure followed estab- lished conventions to ensure that our analysis was "grounded" in the data rather than reflecting preex- isting ideas (Dey, 1993; Henwood and Pigeon, 1992; Patton, 1990). First, a preliminary categorization and thematic analysis by the interviewer and one other author was based on careful reading and re- reading of 30 records. The analysis was first modi- fied on the basis of an additional 20 records and then by the remaining authors from their reading of all records. Finally, lg further records were read and the analysis was finalized when no further modifications emerged from this process and when all the relevant text could be accommodated by the analysis. The use of different authors in different roles permitted a "triangulation" of the analysis (Stiles, 1993).

FINDINGS

Initial analysis identified four interconnected themes. These concerned: (i) the assumptions appar- ent in patients' own explanations of their symp- toms; (ii) the patients' emphasis on the accuracy of their own vs. their doctors' knowledge of symptoms; (iii) the patients' evaluation of the doctors as lack- ing authority and denying their symptoms; (iv) patients' requirements of their doctors. Subsequently, these themes were elaborated into the categories presented below. In our final analysis, each category was defined by commonality of meaning across the accounts of a number of patients: comments relevant to isolated patients were excluded. The range and commonality of con- tent of each category are illustrated by statements in the text, numbered to distinguish the patient by whom they were made. Table 1 summarizes age, sex and symptom pattern of the patients who are cited.

Patients" explanations of symptoms

Few patients had arrived at explanations of their symptoms that convinced and satisfied them. For most, explanations were incomplete or unsatisfac- tory and, in consequence, fluid. Indeed, permeating the majority of patients' accounts was evidence of continuing, intricate scrutiny of alternative expla- nations. Medical terms and ideas did not dominate patients' accounts and most explanatory ideas were described without reference to a specific doctor or

Patients with medically unexplained symptoms

Table 1. Details of patients cited in the text: symptoms are those reported by the patient at interview

561

ID Sex Age Principal symptom pattern

1 F 67 2 F 41 3 M 33 4 F 33 6 F 50 8 F 55 9 F 54 l0 F 25 11 M 26 12 M 42 13 M 47 14 M 38 15 F 66 17 F 47 18 F 32 21 M 52 22 M 33 23 M 45 24 F 51 26 F 41 27 F 40 28 F 34 31 M 46 38 M 46 42 M 39 44 M 41 52 F 69 59 M 58 66 M 35 68 F 36

neck pain, facial pain back pain, facial pain, wrist pain, non-specific pain

headaches, neck pain, back pain lower limb and hip pain, fatigue, breathlessness, anxiety

abdominal pain, anxiety symptoms, non-specific pain, nausea, fatigue back pain, headaches, lower limb pain, non-specific pain, fatigue, nausea, anxiety, sinusitis

nausea, sleep disturbance, neck pain, fatigue anxiety

respiratory infections, back pain, headaches fatigue, non-specific pain, dizziness, anxiety

depression, irritable bowel symptoms non-specific pain, parasthaesia, anxiety symptoms, sweating

irritable bowel, fatigue, respiratory infections, hand pain back pain, menstrual disorder, breathlessness

fatigue, non-specific pain, headaches, menstrual disorder, sore throats, constipation back pain, fatigue, parasthaesia, gout

fatigue, back pain, chest pain fatigue, sleep disturbance

non-specific pain, abdominal pain, back pain, hip pain, neck pain, constipation non-specific pain, fatigue

cough, anxiety, fatigue, non-specific pain, chest pain headache, non-specific pain, fatigue

back pain, abdominal pain, groin pain back pain, neck pain, nausea, chest pain

breathlesness, chest pain, dizziness, abdominal pain non specific pain, fatigue, poor concentration

sore throat, dizziness, chest pain, tinnitus irritable bowel symptoms, back pain, chest pain, headaches

fatigue, depression, non specific pain, headaches respiratory infections, fatigue, nausea, non specific pain

medica l inves t iga t ion as the source. Pa t i en t s ' cu r ren t e x p l a n a t i o n s cou ld be m o r e efficiently ca tegor ized acco rd ing to fou r n o n - m e d i c a l themes:

Disease entity. The pervas ive theme across pa t i en t s ' a c c o u n t s was o f a mal ign, a u t o n o m o u s

ent i ty . M o s t s t r iking was its capac i ty to m o v e t h r o u g h the b o d y in ways tha t the pa t ien ts could t rack in detail . A c c o u n t s ident i f ied s ta r t ing poin ts , o r trigger spots t a n d des t ina t ions : Now it just seems to be settling in my hip 2. Often , the comple t e j o u r -

ney was descr ibed: The migraine starts here and moves to here 4. Biomedica l t e rms were w o v e n in to

these t o p o g r a p h i c a l models : It goes up here into the head around this side. Could be the nerve from the neck that's infected going up to the back o f the head here I .

F u r t h e r evidence o f the a u t o n o m y o f the disease ent i ty was the way it cou ld change form: That [ME]

started from glandular fever and it just turned into the M E 15. It cou ld also be a t t acked , lurk a n d re turn: I'd beaten it and it came back 28. It could ,

therefore , evade capture : I think it wavers in and out. I've had it tested a few times. It's never got caught when you've gone for tests 68. The ent i ty had an exis tence b e y o n d the pa t i en t ' s o w n body: It just come on me and never left me24; a n d cou ld the re fore be exper ienced by others : M y brother gets the same back pain 2. Occas ional ly , the invad ing ent i ty was descr ibed in b iomedica l terms: The glandular fever left a scar on the lung. I assume that's where the virus got in. The virus in my body that makes me really tired 28. All this a d d e d up to a f r igh ten ing threat : I'm not a neurologist, I don't know what's on

the left, but it must be something sinister i f it always starts there 3.

Social influences. R a t h e r t han a d isease ent i ty or p a t h o g e n , s o m e a t t r ibu t ions to fac tors ex te rna l to the b o d y ci ted social sources o f dis t ress o r difficulty, par t i cu la r ly e m p l o y m e n t a n d close re la t ionships . T h e t e rm "stress" was c o m m o n l y used in this way: I

suppose the job is part to blame - - the stress o f the job 2t. To have ident i f ied an ex te rna l cause o f this

k ind cou ld m a k e acceptable a label o f depress ion : O f course I do get depressed, but it's reactive de- pression you know. It's because o f C-'S [daughter] ill- ness 8.

Internal imbalance. In this ca t ego ry exp l ana t i ons

desc r ibed excessive or i nadequa t e s t r eng th in any o f a var ie ty o f bodi ly systems, c o m m o n l y ident i f ied by b iomedica l t e rms inc luding chemical imbalance 3 or making too much adrenaline. 6 M o s t references to the

i m m u n e sys tem were o f this na ture : I remember my dad saying, you know when you were..we were..ill, like you had all these armies in your body who were fighting. And you felt ill 'cause it was the armies fighting. It's daft but that's how I feel, like it's my immune system. I feel ill because my body's fighting infection all the time. That was why I was so tired...- the immune system is very vulnerable 6. In this accoun t , the s y m p t o m s are a t t r i bu t ed less to the ex te rna l agen t ( infect ion) t h a n to the c o n s e q u e n t ove r l oad o f the in ternal m e c h a n i s m ( i m m u n e sys- tem).

Re la ted to ideas o f imba lance in bodi ly sys tems was imba lance be tween the b o d y ' s capac i ty a n d the pa t i en t ' s d e m a n d s u p o n it: I put it [relapse] down to

562 Sarah Peters et al.

doing too much 6. In this way symptoms could be explicitly presented as warnings: ! used to have panic attacks when driving, and it was my body tell- ing me I shouldn't be driving. I really wasn't well enough 6.

Nervous and psychological mechanisms. These ex- planations were distinguished from "stress" by the absence of an external cause and the consequent emphasis on internal factors. Occasionally, the references were to "nerves": I want to know it's 'cause o f my nerves, because the stupid thing is I always think I've got cancer or something l°. Despite the physical metaphor of "nerves", psychological processes are implicit in this comment. More com- monly, emotional or mental processes were made explicit. That is, the symptoms were thought to result from moods: It could be the way I feel all the time. That could be it. Before ! was on the tablets, I used to stare at the ceiling. Sounds a bit boring doesn't it. Maybe that's why I'm always getting these coldsll; and to be exacerbated by thoughts: I think ! am depressed... 1 think I think too much. It's two plus two makes five with me. I'm always reading about symptoms, then ! work it round to me and 1 start thinking I've got it 9.

Patients" unique knowledge o f symptoms

Most patients' emphasized their inevitably su- perior knowledge of their problems. Only the patient could feel a symptom, i.e. use a faculty which was both infallible and unavailable to the doctors: I've had ridiculous things said about me. It's as i f you don't know your own body 6. This knowl- edge could, in turn, derive from rigorous self-exam- ination: I've fel t all over the area looking for where it starts. I've found here, here and this area there, they're like trigger spots 1. Patients' infallible sensory knowledge of their symptoms was the basis for the universal assertion that the symptoms were real and must therefore have a cause: These migraines must be from something 3. In contrast with patients' intrin- sic and hence privileged sensory knowledge of their body beneath the skin, doctors were portrayed as relying on what they could see and demonstrate with imperfect techniques: Anything intramuscular is hard to see. I can feel exactly where it is. I f I dig my fingers in it really hurts 24. Even medical technology was limited: The X-rays showed nothing. But they don't show nerves do they? 59 For other patients, their superior knowledge reflected their greater fam- iliarity with the symptoms: The neurologist said it's not affected by the jaw thing, but I know it triggers it off 3. For other patients, the imbalance of knowledge reflected information that they had withheld from the doctor: My doctor doesn't know [about a recent episode of low back pain]. I don't go to him unless I have to. He only knows about the first one 2.

Because of the underlying assumption of the rea- lity of symptoms, doctors' negative investigations were universally interpreted as failure to reveal the

problem: I don't understand why nothing 's come up. I can't be imagining it 3.

Patients' evaluation o f doctors

Attributing authority and questioning professional competence. Given the subjective reality of symp- toms and patients' privileged information, it is not surprising that many patients' accounts of medical consultations clearly established their own authority over that of their doctors. Doctors' statements were carefully scrutinized and either adopted or rejected: Now they think it's a spasm. Well, spasms come and go don't they. Unless it stays in a sort o f spasm? I don't think it is. l Patients were also in a position to compare doctors against each other: I've been to all the different doctors in the surgery in case one o f them could come up with somethingl; and against their own expectations: She never asked to get the [hormone] levels checked. I have to remind her all the timerS; and to inform doctors: I've told the doctor things she only knows from me 6s. Alternative sources of "medical" knowledge were also sought: I went to B-, my friend who's got it [ME]. She gave me a book on it and it was just like they said. I had everything they described just like it saidlS; although generally these had to be sanctioned by the doctor: I saw it on the telly about it. I made myself go to the doctor and I said have I got agoraphobia and he said yes. I thought well why couldn't you have told me 27.

The competence of doctors and other pro- fessionals in applying the limited knowledge that they did possess was frequently criticized. Apart from simply "missing" the cause they had, in many patients' view, neglected certain procedures: Never tried the obvious: ear, nose and throat. You'd think they'd try that wouldn't you?52; or had used them incorrectly: X-ray came back as normal but my im- pression o f that is that I take lousy photos too 3. Doctors' incompetence extended to damage that they had, in the views of a few patients, caused: This came on me when ! went for investigation. They put a tube inside. Since then, when ! go to the toilet, it squirts everywhere. That's not normal 24

Perceived denial o f symptoms by doctors. GPs and hospital doctors were commonly seen as having denied the reality or importance of the symptoms: They don't understand it so they just dismiss it 3s and having doubted patients' veracity: They [A and E staff] laughed at me. They didn't believe me. 66 Attri- butions by doctors to anxiety or depression were generally presented to the interviewer in this way: It's not bloody psychological I'm not off me trolly. She [GP] thinks it's all in the mind 22. Rejection was, however, often more reasoned: I know what de- pression is like. This is completely different... I f it is depression it's a different type, but I really don't think it is. This is physical isn't it? ~s. In some reports it was explicit that denial of symptoms was experi- enced as rejection of the patient: I do feel let down by the GP in particular. She says it's anxiety 42.

Patients with medically unexplained symptoms

Although, as above, these accounts were often as- sociated with hostility towards the doctor, hostility from the doctor was also inferred: Pain-in-the-neck patients you're looking for aren't you? That's what the doctor said anyway 13. Correspondingly, embar- rassment and withdrawal were described: I don't tell her now. I think she'll just laugh t8.

Patients' requirements o f doctors

Having the problem named. Many patients had evidently been frightened of serious pathology: I was even convinced I'd got a brain tumour or cancer or something 26. Therefore naming of the problem was typically a source of relief: Not having a label, I think that's the real problem so I can put it in a box and go yeah it's that, l've got that, now I can get on with me life. I can't deal with the unknown 26. Diag- noses which our patients recounted clearly reflected the doctor's access to special procedures or to special names: They confirmed it was the M E [myal- gic encephalomyelitis, chronic fatigue] virus 12. Acceptance or rejection of a diagnosis was not a simple function of whether it was physical rather than psychological since even physical labels could be rejected: IBS [irritable bowel syndrome] is what they call it when they don't know what it is 6. More- over, although the label of depression had usually been rejected, patients had accepted it when it was suitably qualified: I have clinical depression. It's not normal depression. It is in fact the clinical type. The doctor explained it to me quite well actually. It's between the neurons, in these synapses. Something goes awry and that happens in clinical depression. Everything that hurts - - I know that it's because o f the brain cells not quite working 23.

At a second level of naming the problem, many patients recounted a number of mechanisms whereby doctors were felt to have adequately explained their symptoms. Some, although simple, relied on professional authority: The doctor tom me that 'cause I was tall I'd always suffer from my back 3. Others were explicitly professionally based: The blood cells weren't reproducing. Is it the mito- chondria or something 12. Again, these were carefully examined and considered before being adopted.

Provision of explanation which involves the patient. The explanations above were ones that legitimated the symptoms and exculpated the patient for them, but did not give the patient any responsibility for managing them. Rarely, the mechanism cited was one for which the patient could take responsibility: He [GP] explained about tensing myself up so the neck kept hurting 24.

Engagement with the problem. It was striking that doctors' value did not derive solely from the pro- ducts of investigation but also, for a few patients, from the process, regardless of outcome: I appreci- ate all the doctor's done. Not helped mind, but tries everything, blood levels and tests. He's a good doctor 3~. The doctor could be seen as a collaborator

563

with the patient against the problem: I'm not blam- ing him [GP]. We're in the same boat. Neither o f us know what it's about 44. Consistent with this, pro- cedures which had been carried out by doctors were sometimes recounted as evidence of the intensity of symptoms: I had to have an extreme D and C 17.

DISCUSSION

The purpose of this study was (i) to find out how biomedical ideas are used by patients to understand and describe symptoms for which doctors have failed to provide medical explanations and (ii) on this basis to clarify understanding of such patients' demands on and responses to their doctors. The persistent presentation of physical symptoms in the absence of confirmed physical pathology has been termed "somatization" and accounts for a significant proportion of demand for primary and secondary health care (Fink, 1996). Our sample contained within it many patients with frequent contact with GPs and hospital clinics. The insights provided by this study are, therefore, not simply of theoretical interest; by suggesting alternative ways in which symptoms might be handled they have the potential to reduce the high level of unnecessary, expensive and often damaging clinical investigation to which such patients are exposed.

Biomedical terms and ideas did not provide a useful way of categorizing patients' accounts. Instead, biomedical terms were mostly used to elab- orate well recognised, cross-cultural explanatory themes. In particular, notions of imbalance of bod- ily processes and humours have been a consistent feature of explanations of symptoms across cultures and, within western society, across centuries (Helman, 1994). Many of the contemporary medical ideas that our patients cited were interpreted in these terms, including imbalance of hormones and overloading of the immune system.

The most pervasive explanatory theme was the alien disease entity; this was menacing, played tricks (by retreating but then returning), and moved around autonomously within the body. Into this theme patients wove contemporary medical terms, such as "virus" and ME. The entity might be "bea- ten" but there was no implication that it could be destroyed. This theme has been described previously (Blaxter, 1989) - - although not in relation specifi- cally to unexplained symptoms - - and it, too, has a long history and wide generality across cultures (Cassell, 1976; Helman, 1994). This attribution of suffering to a discrete entity separate from the indi- vidual has been interpreted as a defence mechanism separating out negative components in order to pre- serve a view of self as basically healthy (Helman, 1985). References to stress fit a similar pattern: patients used this term to ascribe symptoms to social factors in ways that occur in other cultures (Helman, 1994; Unschuld, 1986).

564 Sarah Peters et al.

At first sight these accounts, by invoking medical ideas, are consistent with the Shorter (1992, 1995) view that medical models become incorporated into lay discourse and thereby "owned" and operated by patients as much as by professionals. Overall, how- ever, our findings are explained more efficiently by the alternative view that medical ideas are adopted to express pre-existing cultural beliefs. In the same way, attributions of symptoms to moods or thoughts resembled previous explanations of symp- toms in a community sample (Blaxter, 1989) and may represent the updating of a traditional concept of "nerves" (Helman, 1994).

It appears, therefore, that this group's require- ments of doctors are not directed by adherence to a specific biomedical model which constrains doctors' responses (Salmon and May, 1995; Shorter, 1992). Instead, from their own perspective, patients' auth- ority in consulting doctors is derived from two self- validating internal processes. The first is their direct sensory experience of symptoms. The second is the active and ingenious way in which they construct and scrutinize explanations of their symptoms. Evidence, both from their own observations and doctors' investigations, is carefully sought and ex- planations from different doctors and from nonme- dical sources compared, a process that is consistent with previous demonstrations of the fluidity of patients' explanations of symptoms and of patients' readiness to reject doctors' advice (Hunt et al., 1989; Stimson, 1974). In contrast to patients' view of their own evidence and process of enquiry as informed and expert, doctors' perceived fallibility stems from their inevitable reliance on the indirect information provided by medical investigations; they are relegated to an inexpert and essentially technical role. This perceived imbalance in expertise is most evident where doctors were described as denying the reality of symptoms; often, this arose when symptoms were labelled as "psychological" or located "in the mind".

Our findings contrast with previous literature, also, in the absence of evidence that these patients seek to constrain doctors to deliver tangible inter- ventions (Marchant-Haycox and Salmon, 1997; Shorter, 1992). It is notable that, despite unresolved symptoms, no patient directly criticised a doctor for failure to treat an underlying disorder or to palliate symptoms. Patients' accounts revealed two tasks for which they did seek out doctors. First, engagement with the problem was valued for itself, even where it failed to deliver satisfying explanation or symp- tom control. Secondly, patients recounted diagnoses or explanations in ways that demonstrated the im- portance of having the disorder named (Blaxter, 1989). Even where the patient had arrived at their own diagnosis, such as "ME" or agoraphobia, the doctor was used for validation. Some explanations attributed to doctors concerned mechanisms rather than diagnoses. These did not necessarily include

medical terms but could be based on less "expert" attributions; e.g. explaining back problems by being "too tall".

Comparison of explanations that were accepted to ones that had been rejected reveals patients' requirements for explanation. These are to recog- nize the reality of suffering and exculpate the patient by confirming the origin of symptoms in factors outwith the patient's responsibility (Pill and Stott, 1982; Telles and Pollack, 1981). Thus, although many explanations had been assimilated, others that implied a denial of symptoms' reality or that implied culpability (e.g. being unfit) had been roundly dismissed. Occasionally a physical mechan- ism was recounted, such as muscle tensing under stress, over which the patient might exercise some control. In this instance, attribution to an external cause, stress, removed culpability.

In summary, patients' explanations of persistent unexplained symptoms do not differ as markedly as recent literature would suggest from the expla- nations of symptoms that have been described in unselected groups. Our findings are therefore not consistent with the view that somatizing patients, in understanding and describing their symptoms, adopt medical models which constrain doctors to deliver tangible interventions. The contrast with previous work may reflect our recruitment of patients in primary care rather than hospital clinics, or our emphasis on use of patient interviews with a non-medical observer rather than records of their interactions with doctors. Two components of our analysis offer an alternative formulation of what these patients seek from doctors: first, their percep- tion of disease as a malign entity (and their in- terpretation of medical ideas in this culturally preestablished way); and, second, their appreciation of the doctor as an ally against this threat. These themes suggest the importance, in somatization, of a doctor-patient alliance against the disease entity, a view which extends existing models of doctor- patient interaction beyond meetings of experts (Tuckett et al., 1985) or of collaborators in problem solving. In current practice, alliances of doctors with patients presenting unexplained symptoms tend to focus on clinical investigations and sympto- matic treatments. Other ways may be more effective and less damaging: for example, the reattribution techniques of cognitive therapy (Sharpe et al., 1996; van Dulmen et al., 1996) or collaboration with patients to find ways to constrain or "outwit" the disease entity by pacing and expanding daily activi- ties.

Acknowledgements--This research was supported by the UK MRC.

Patients with medically unexplained symptoms 565

R E F E R E N C E S

Blaxter, M. (1989) The causes of disease: women talking. Social Science and Medicine 17, 59-69.

Borkan, J., Reis, S., Hermoni, D. and Biderman, A. (1995) Talking about the pain: a patient-centred study of low back pain in primary care. Social Science and Medicine 40, 977-988.

Cassell, E. J. (1976) Disease as an "It": concepts of disease revealed by patients' presentation of symptoms. Social Science and Medicine 10, 143-146.

Dey, I. (1993) Qualitative Data Analysis: a User-Friendly Guide for Social Scientists. Routledge, London.

Fink, P. (1996) Somatization - - beyond symptom count. Journal of Psychosomatic Research 40, 7-10.

Helman, C. G. (1985) Psyche, soma, and society: the social construction of psychosomatic disorders. Culture and Medicine in Psychiatry 9, 1-26.

Helman, C. G. (1994) Culture, Health and Illness 3rd edn. Butterworth-Heinemann, Oxford.

Kleinman, A. (1978) Concepts and a model for the com- parison of medical systems as cultural systems. Social Science and Medicine 12, 85-93.

Henwood, K. L. and Pigeon, N. F. (1992) Qualitative research and psychological theorizing. British Journal o f Psychology 83, 97-111.

Hunt, L. M., Jordan, B. and Irwin, S. (1989) Views of what's wrong: Diagnosis and patients' concepts of ill- ness. Social Science and Medicine 28, 945-956.

Marchant-Haycox, S. and Salmon, P. (1997) Patients' and doctors' strategies in consultations with unexplained symptoms: interactions of gynecologists with women presenting menstrual problems. Psychosomatics, in press.

Office of Population Censuses and Surveys. (1995). Morbidity Statistics from General Practice: 4th National Study 1991-1992. HMSO, London.

Patton, M. Q. (1990) Qualitative Evaluation and Research Methods 2nd edn. Sage, Newbury Park.

Pennebaker, J. W. (1982) The Psychology of Physical Symptoms. Springer-Verlag, New York.

Pill, R. and Stott, N. C. H. (1982) Concepts of illness cau- sation and responsibility: some preliminary data from a sample of working class mothers. Social Science and Medicine 16, 43-52.

Sharpe, M., Hawton, K., Simkin, S., Surawy, C., Hackmann, A., Klimes, I., Peto, T., Warrell, D. and Seagroatt, V. (1996) Cognitive behaviour therapy for the chronic fatigue syndrome: a randomized controlled trial. British Medical Journal 312, 22-26.

Salmon, P. and May, C. (1995) Patients' influence on doctors' behaviour: a case study of patient strategies in somatization. International Journal o f Psychiatry in Medicine 25, 309-319.

Shorter, E. (1992) From Paralysis to Fatigue: A History of Psychosomatic Illness. Free Press, New York.

Shorter, E. (1995) Sucker-punched again! Physicians meet the disease-of-the-month syndrome. Journal of Psychosomatic Research 39, 115-118.

Speckens, A. E. M., Van Hemert, A. M., Bolk, J. H., Rooijmans, H. G. M. and Hengeveld, M. W. (1996) Unexplained physical symptoms: outcome, utilization of medical care and associated factors. Psychological Medicine 26, 745-752.

Stiles, W. B. (1993) Quality control in qualitative research. Clinical Psychology Review 13, 593-618.

Stimson, G. V. (1974) Obeying doctor's orders: a view from the other side. Social Science and Medicine g, 97- 104.

Telles, J. L. and Pollack, M. H. (1981) Feeling sick: the experience and legitimation of illness. Social Science and Medicine 15A, 243-251.

Tuckett, D., Boulton, M., Olson, C. and Williams, A. (1985) Meetings Between Experts: An Approach to Sharing Ideas in Medical Consultations. Tavistock, London.

Unschuld, P. (1986) The conceptual determination (Uberformunt) of individual and collective experiences of illness. In Concepts of Health, Illness and Disease, eds. C. Currer and M. Stacey, pp. 51-70. Berg, Leamington Spa.

van Dulmen, A. M., Fennis, J. F. M. and Bleijenberg, G. (1996) Cognitive-behavioral group therapy for irrita- ble bowel syndrome: effects and long-term follow-up. Psychosomatic Medicine 58, 508-514.

Waddell, G., McCulloch, J. A., Kummel, E. and Venner, R. M. (1980) Nonorganic physical signs in low-back pain. Spine 5, 117-125.

Waiters, A. (1961) Psychogenic regional pain alias hysteri- cal pain. Brain 84, 1 18.