KIRSTI MALTERUD

THE (GENDERED) CONSTRUCTION OF DIAGNOSISINTERPRETATION OF MEDICAL SIGNS IN WOMEN PATIENTS

ABSTRACT. Medicine maintains a distinction between the medical symptom – thepatient’s “subjective” experience and expression, and the privileged medical sign – the“objective” findings observable by the doctor. Although the distinction is not consistentlyapplied, it becomes clearly visible in the “undefined,” medically unexplained disordersof women patients. Potential impacts of genderized interaction on the interpretation ofmedical signs are addressed by re-reading the diagnostic process as a matter of socialconstruction, where diagnosis results from human interpretation within a sociopoliticalcontext. The discussion is illustrated by a case story and empirical evidence of thegendering in the doctor-patient relationship. The theoretical analysis is supported by semi-otic perspectives of bodily signs, feminist theory on experience, and Foucault’s ideas aboutmedical perception and gaze, and concludes that a medical diagnosis is seldom a biologicalfact, but the outcome of a process where biological, cultural and social elements are inter-woven. Further deconstruction of the chain of signs from a feminist perspective, assigningvalidity to the voice of the woman patient, might broaden the understanding of women’shealth, illness and disease.

KEY WORDS: diagnosis, gaze, gender, medical reasoning, signs, symptoms

INTRODUCTION

Mainstream biomedicine maintains the ideal of a dichotomous distinctionbetween the medical symptom and the medical sign.1−4 While the formeris a “subjective” matter, experienced and communicated by the patient(such as “pain in the neck”), the latter are assumed to be “objective” find-ings, observable by the doctor (for example “decreased mobility in thehip joint”). When the health care system operates in what Chinen callsthe representational mode of medical understanding,5 this dichotomy isemphasized, and symptoms are considered as secondary subjective reflec-tions of an underlying objective reality. According to Foucault, medicalsigns are considered as objective facts, observable by the authority of themedical gaze.1 The diagnosis – the name of the disease, which usually willexplain the origin of the symptoms – is supposed to emerge as a fact whichis discovered by the doctor when the puzzle of symptoms and signs hasbeen sorted out and deciphered.5,6

Theoretical Medicine and Bioethics20: 275–286, 1999.© 1999Kluwer Academic Publishers. Printed in the Netherlands.

276 KIRSTI MALTERUD

In this article I suggest that these “facts” emerge from interaction andinterpretation of medical signs, and that this is a gendered process. I willdemonstrate how medicine is inherently ambiguous on the symptom/signdichotomy: on one hand, the epistemology of the representational mode iscommonly referred to in medical theory and practice, implying that thereare distinguished sources for clinical knowledge, with objective findingsdifferent from and ranking higher than subjective symptoms. On the otherhand, subjective symptoms are nevertheless consistently valued as signi-ficant diagnostic information in everyday clinical judgement. I intend toexplore this contradiction by focusing on situations when the patient’ssubjective symptoms are not accepted as valid medical signs – morespecificly: the medically unexplained disorders of women patients.7 Fromthis point, I will reflect on gendered interaction in clinical medicine.

My approach is to replace the representational mode with a re-reading of the diagnostic process as a matter of social construction, wherediagnosis results from human interpretation within a social and politicalcontext. From this position, it is possible to adress potential impacts ofgenderized interaction8 on the construction of medical knowledge towardsdiagnosis by interpretation of medical symptoms and signs. Referring tosemiotic perspectives on bodily signs,9 I shall here deliberately use theword “sign” to denote anything which determines something about some-thing to somebody, including the interpretant of the sign as a visible andsituated actor. From this perspective, both patient and doctor are legit-imate interpretants of medical signs. The discussion will be illustrated bya case story and supported by empirical evidence on the gendering of thedoctor-patient relationship.

PAIN IN THE CHEST AND SHOULDERS –THE CASE OF “JUDITH SMITH”

Judith Smith’s story below is fiction, but draws on the author’s similarexperiences with authentic patients and doctors. The case story emphasizesexperiences and consequences of pain, and how they are communicated,perceived and interpreted towards a diagnostic conclusion.a

Judith Smith (age 55) has pain in her chest and shoulders. She believes that this sign comesfrom her muscles, not from her heart. However, her mother died from a heart attack, and sheis concerned. She has experienced that strain at work and at home increases her musculartension, but she is not able to change her situation and relieve her symptoms.

One Wednesday morning Judith Smith is not able to go to work because of her pain. Sheneeds a sick leave certificate, and she wants a medical explanation so that she can checkher own understanding of the signs of pain and disability. The doctor listens to Judith

THE (GENDERED) CONSTRUCTION OF DIAGNOSIS 277

Smith’s symptoms and story, which he recognizes from previous encounters with middle-aged women. The signs communicated by the patient are perceived by the doctor as notvery serious. The doctor knows from experience that conditions like this tend to becomechronic, and he has no effective cure. The doctor’s interpretation is processed througha cultural grid of experience, emotions, attitudes, and preconception about this kind ofpatient and this kind of knowledges.

The doctor assumes that Judith Smith believes she has heart disease, that she does notattribute her symptoms to her life situation, and that she should be able to modify herdaily life burdens. The clinical examination does not reveal any findings or sign consideredsignificant by the doctor, and the medical signs represented by Judith Smith’s experience,perception and story are not in this case considered relevant for the diagnosis of disease.The doctor’s inference is that no signs of disease have been identified.

This conclusion is given to the patient (no diagnosis), and to the community (no sick-ness leave certificate). Judith Smith feels depressed and goes to work. Her symptoms anddisability continue to increase.

GENDER IN DOCTOR-PATIENT INTERACTION

Judith Smith, a woman patient, has in this case consulted a male doctor.b

Although many women patients prefer women doctors,10−13 the genderdistribution of doctors in the Western world is such that women patientsmost often meet a male doctor. This pattern is reinforced by the fact thatwomen consult more often than men do. Referring to the typical parti-cipants of the consultation – the male doctor and the woman patient – themedical interaction is gendered from the initial moment, and even beforethat. The templates through which the doctor filters his perceptions of anysign, are embedded with his beliefs and sociocultural positioning.14 Thedoctor’s ability to perceive the signs of the patient is read through hissociocultural images of 55 year old women patients, as is the patient’simage of herself. The overview below will illustrate the potential impactof gender in the doctor-patient relationship of the case story above.

Gendered assumptions about patients influence doctors’ interpretationof medical symptoms and their management, as demonstrated in a studywhere 120 general practitioners assessed constructed case histories.15 Thecases were presented as identical, except for gender. Doctors who attrib-uted the symptoms to women patients, gave significantly higher score foremotional explanations of the symptoms than doctors who thought thesame symptoms belonged to men. Another study that compared examina-tions and tests for 181 patients with back ache, headache, dizziness, breastpain or fatigue, showed that men patients were significantly more compre-hensively investigated than women patients with matching symptoms.16

Among 2,231 patients in a postinfarction intervention trial, men were

278 KIRSTI MALTERUD

twice as likely to undergo an invasive cardiac procedure as women, despitestronger cardiac disability in these women patients.17 In 82,782 patientshospitalized for coronary heart disease, consistent male-to-female oddsratios confirmed that women underwent fewer major diagnostic and thera-peutic procedures than men with the same condition.18 Women aged 46–60years in the United States in long-term dialysis had less than half thechance of receiving a kidney transplant when compared with men of thesame age and race.19 Studies like these do not clarify whether such genderimbalances might be adequate, or whether they refer to social and genderbias. However, physician gender bias was identified in a recent longitudinalobservation study of 1,546 patients of 349 physicians, where the odds ofprescribed activity restrictions from the doctor were 3.6 times higher forwomen patients than for mens with equivalent characteristics.20

The doctor’s gender also has an impact on the medical encounter.Women enrolled in a large health plan were more likely to undergoscreening with Pap smears and mammograms when they saw womenrather than men physicians.21 A Dutch study including 47,254 consulta-tions showed that women doctors spent more time on their patients andhad a stronger tendency to provide continuity of care.22 In a Canadiansurvey where 3,000 doctors were asked about prenatal diagnosis, womendoctors reported more liberal attitudes than their men colleagues.23 Acase-control study from an urban outpatient practice in Boston demon-strated that women were more likely to be prescribed estrogen replacementtherapy if they were cared for by women doctors than men doctors.24 Asurvey among 1,000 primary care phycisians in Michigan showed thatmen doctors more often endorsed pharmacological treatment for depres-sion in the elderly, while women doctors more frequently used conselingand exercise techniques.25

Communication between doctor and patient is affected by genderon both sides. Analysis of material from 336 audiotaped consultationsrevealed that women patients ask more questions to the doctor thanmen patients do, but although they proportionately have more responsesfrom the doctor, the responses are less adequate.26 In 21 videotapedconsultations, men doctors were responsible for 67% of the interruptionsduring the medical conversation, while women doctors were responsiblefor 32%. A study of 100 videotaped consultations showed distinctivegendered patterns of doctor-patient communicative behaviour, especiallyin nonverbal communication.27 Gender operates on diverse levels inmedical interaction, and should also be adressed when it comes to theinterpretative part of the diagnostic process.

THE (GENDERED) CONSTRUCTION OF DIAGNOSIS 279

WOMEN’S “UNDEFINED” MEDICALLYUNEXPLAINED DISORDERS

The case of Judith Smith portrays the “undefined”, medically unexplaineddisorders – the commonly occurring conditions where the lack of observedfindings is seen as significant, while the doctor’s subsequent diagnosticinterpretations discard the meaning of the patient’s symptom experience.The logic of the representational mode of medicine implies that doctors candiscriminate “real” disease from “unreal” disease by means of objectivefindings.2 Symptoms without corresponding findings lack the structuralcorrelation which signifies and legitimizes real disease, and will accord-ingly be refused the status of “real disease”. The lack of signs in a strictlymedical meaning justifies that these conditions are regarded as a vagueor unspecific anomaly, classified as “other”, “diverse” or “undefined”,although the patients suffer from painful and disabling symptoms.7 Themedically unexplained disorders comprise a diversity of expressions, suchas chronic pain syndromes, chronic fatigue, tension headache, irritablebowel, urinary tract inflammation, or whiplash trauma sequelae. Thedoctor does not understand the origin of the illness, and he does not knowhow to bring relief or cure.

Male patients may certainly experience the consequences of observablefindings as privileged medical signs as compared to symptoms. However,there is a remarkable and consistent majority of women compared to menwho suffer from the commonly occurring conditions mentioned above.7 Inthese cases, where medical signs are interpreted according to the represent-ational mode of medical understanding, subjective symptoms are overruledby objective findings, or rather the lack of such.

Women patients with chronic muscular pain report experiences ofhumiliation and distrust from the health care system when they presenttheir symptoms.28,29 The culturally impregnated judgement of the medicalexpert overrules and may even pervade the woman patient’s bodily exper-ience and interpretation.30 Because an approved medical diagnosis islacking, the patients risk being refused access to welfare benefits as sick-ness leave certificates or disability pension. They are concerned with thecontested “reality” of pain, resist being blamed for this, and return theblame to the doctor who has not yet identified the cause of the pain.31

Eccleston and co-workers comment that the place of power is an importantissue in this web of blame, responsibility and identity.

Given the evidence on gender-biased interaction presented above,my hypothesis is that gender-biased interaction contributes to strategiesof diagnostic reasoning where privilege is systematically given to the

280 KIRSTI MALTERUD

doctor’s perception of medical signs, especially when the patient is awoman.

THE MEDICAL GAZE – PRIVILEGED SIGNS UNDERCERTAIN CIRCUMSTANCES

Lilleaas’s informants report the experiences of being labelled as whim-pering, complaining or irrational if they tell their story and share thesigns as perceived by them.28 Due to the alleged scientific “objectivity” ofmedicine, doctors alone have the competency to make proper judgmentsabout health and options.32 Women’s perceptions are communicated tothe doctor, whose medical gaze1 will interprete the signs and determinewhether they are valid or not. Foucault comments that although the medicalgaze is a collective mode of perception, shaped by issues related to politicsand power as codes of knowledge, it is supposed to be open and receptive,not bound by the narrow grid of predefined structure.c

The “undefined” disorders emerge as an conceptual anomaly definedfrom within the representational mode of understanding. The indisputableexistence of such disorders, portrayed by the large number of women withmuscular pain syndromes, testifies that the medical culture of knowledgeholds a hierachy of signs, where certain kinds of signs are more privilegedthan others. Symptoms are subordinate to findings, which are believedto be objective. Indirect evidence, as mediated by medical technology,seems to constitute to medical signs a stronger validity than direct evidencefrom the voice of the patient. However, this hegemony seems to deny thatany indirect representation of disease is dependent on the contemporaryscope, perspective and technological level. Illness which produces symp-toms without leaving footprints available for the medical eye, may yet bevisible tomorrow. The objectivity of indirect findings might certainly becontested.

Although gaze literally refers to visual perception, Foucault consideredthe medical gaze and its corresponding discourse to be a comprehensivemode of perception.1 Auditiory perception may also be included inthe gaze under certain circumstances. When auditive cues are mediatedthrough medical instruments and analyzed by the doctor (listening to theheartbeats through the stethoscope), the problem of subjectivity seems tobe no problem. However, when it comes to the voice of the patient with themedically unexplained disorder, this auditive cue can be discarded due toits alleged subjectivity. The medical gaze favours indirect and objectivizedlistening, preferably through technological devices.

THE (GENDERED) CONSTRUCTION OF DIAGNOSIS 281

Indirect visual signs, as x-ray pictures, are commonly considered asthe absolute standard for objective diagnosis. The x-rays of women with“undefined” disorders are typically negative, with no pathological findings.I shall not deny the existence of x-ray pictures where any observer wouldagree upon the conclusion, and the progress of the case would confirm thecorrespondence between the interpreted signs and some significant clinicalpattern. This might be the case when a broken leg is investigated and oper-ated on. Ontologically, however, signs like these are just interpreted andindirect representations of selected elements of the body, perceived fromspecific positions and perspectives, and interpreted according to consensusvalues within radiology and clinical medicine. Empirical studies demon-strate considerable interobserver variation for clinical data assumed tobe objective facts,33 for example the problems of accuracy and variab-iliy in mammographic interpretation.34 This is not surprising, for evenin experimental laboratory research, findings are the objects of humanmanipulation and interpretation.35 Even within the most “objective” fieldsof medicine, human perception and interpretation can construct clinicalknowledge presented as facts. The problem is not that this is happening, butthat medical culture on certain frequently occurring occasions insists onthe ideal of a diagnostic model in which signs and symptoms are separatephenomena.

While certain symptoms seem to require an objective correlate, othersdo not. Although patients with medically unexplained disorders experi-ence that their symptoms are discarded, clinical decision-making is notconsequent in sticking to the representational mode of understanding inany medical situation. The patient’s symptom experience and story is actu-ally very often attributed a strong impact and recognized as a medicalfact in the construction of medical knowledge – not only in contextsclassified as “undefined.” In 77% of the cases at a British outpatientclinic, the diagnostic conclusion was mainly drawn from the story of thepatient. Only in the remaining proportion, the clinical examination andtechnical tests contributed significantly to the diagnostic conclusion.36

A gap between symptoms and findings is common even in supposedly“well-defined” conditions as pneumonia and dyspepsia.37,38 Under suchconditions, however, the discrepancy seems not to be held against thepatients, as happens with the women who reported that they felt blamedfor the disconcordance between symptoms and findings.28

282 KIRSTI MALTERUD

MEDICAL INTERPRETATION IN A SOCIALAND CULTURAL CONTEXT

In the representational mode of medical understanding, a fact is a fact,unbiased by the observer as well as the context and the situation. Whatis not seen by the doctor, is not there.1,5 The interpretative componentsof this activity and their contextual dependency are seldomly questioned.Perception and presentation of bodily symptoms are not just the simpleexchange of facts, but a complicated hermeneutic process of readingand understanding the text of medical signs.39 Von der Fehr used semi-otic perspectives to elucidate the role of consciousness on the patient’sperception and interpretation of bodily signs.40 Rudebeck says that from aphenomenological perspective, a symptom is an act of perception, whichcontains both the meaning and its physical correlate, whereas the symptompresentation is a personal communication of a very private experience.41

However, perception and expression of bodily experience are not solely“authentic” features, unbiased by the cultural context of the woman,but also constituted by historical context. Scott reminds us that whatcounts as experience is neither self-evident nor straightforward, – it isalways contested and therefore political.42 Even the bodily experience of awoman is influenced by the life she lives and the meanings of illness andwomanhood in her culture.

The doctor is supposed to make sense of the patient’s narrative. Rude-beck introduces the concept “bodily empathy” as a preconception for theclinical understanding of medical symptoms. We might suggest that thedoctor’s perception and response do not accommodate the bodily empathyneeded to acknowledge the suffering of the women with medically unex-plained disorders. The task of clinicians and clinical researchers is toextend and perfect the maps of illness, although each patient, and eachinstance of illness are yet uncharted territory.43 Anomalies not fittingneatly into the puzzle of the doctor’s mind, run the risk of being discarded,like the problems of Judith Smith. The doctor might forget that “the clin-ical entity” is only a preliminary typology43 and blame the patient for notpresenting a clearcut clinical pattern.

THE CONSTRUCTION OF MEDICAL “FACTS”

From a constructionist point of view, the validation of medical signsis not a mechanical procedure exposing undeniable facts, but a matterof perception, interpretation, narration, and negotiation.33 The doctor’s

THE (GENDERED) CONSTRUCTION OF DIAGNOSIS 283

inner images, constructed by previous medical science and history, are thetemplate towards which the clinical signs are read by the medical gaze.

In fiction stories like the one presented in this article, the author hasthe freedom to exaggerate. In the story above, this was done to illustrateand emphasize the potential clinical impact of some of the gender issueswhich have been documented by empirical evidence. The reader shouldbe warned not to interpret this story in an essentialist way, implying thatall women patients are comparable to Judith Smith. Medical reality is notthat simple, although stories from this reality sometimes may be read assimplified realities – even by the doctor.

Medical theory and practice have; for a long time, been constructedby men, with the clinical gaze of men, the cultural templates of men, andthe perception and language of men. It is therefore not unexpected thatthe clinical signs presented by women patients are not always adequatelyembraced within the scope of contemporary medical epistemology. Theor-etical and empirical evidence remind us that the interpretation of clinicalsigns depends on the position of the reader – even the gendered ones. Fromsuch a perspective, it is not surprising that medical signs in women havebeen denied medical validity.

A medical diagnosis is seldom a biological fact, but commonly theoutcome of a process in which biological, cultural and social elementsare interwoven through interaction and language. A semiotic perspectiveallows us to dispute the narrow conception of a medical sign as an observ-able, objective finding. The demonstration of the potentially gendereddynamics and effects of the interpretative and interactive elements of thediagnostic process calls for feminist perspectives on medical epistemologyand clinical practice. Further deconstruction of the chain of signs from afeminist perspective, assigning validity to the voice of women patients,might broaden the understanding of women’s health, illness and disease,and of the cultural construction of medical knowledge.

ACKNOWLEDGEMENTS

This article origins from a crossdisciplinary feminist collaboration aboutwomen’s chronic muscular pain at the Center for Feminist Research,University of Oslo. I thank Ulla-Britt Lilleaas, sociologist, and Drude vonder Fehr, literary theoretician, for their important contributions and criticalcomments.

284 KIRSTI MALTERUD

NOTES

a I admit that I might recognize myself in the role of the doctor, although I prefer toconsider myself as more empathetic than the doctor portrayed here.b I would not deny the possibility that what happened in this consultation, might havehappened even with a woman doctor.c This contradiction in logic attends to the relationship between ideals and practicementioned initially.

REFERENCES

1. Foucault M.The Birth of the Clinic. An Archaeology of Medical Perception. NewYork: Vintage Books, 1994/73.

2. King LS.Medical Thinking. A Historical Preface. Princeton, New Jersey: PrincetonUniversity Press, 1982: 73–89.

3. Armstrong D. The patient’s view.Soc Sci Med1984; 18: 737–744.4. Honkasalo, ML. Medical symptoms: A challenge for semiotic research.Semiotica

1991; 87: 251–268.5. Chinen AB. Modes of understanding and mindfulness in clinical medicine.Theoret-

ical Medicine1988; 9: 45–71.6. Albert DA, Munson R, Resnik MD.Reasoning in Medicine. An Introduction to

Clinical Inference. Baltimore: The Johns Hopkins University Press, 1980.7. Malterud K. Women’s undefined disorders – A challenge for clinical communication.

Family Practice1994; 9: 299–303.8. Code L.What Can She Know? Feminist Theory and the Construction of Knowledge.

Ithaca & London: Cornell University Press, 1991.9. Hoopes J, ed.Peirce on Signs. Writings on Semiotic by Charles Sanders Peirce.

Chapel Hill/London: The University of North Carolina Press, 1991.10. Graffy J. Patient choice in a practice with men and women general practitioners.

British Journal of General Practice1990; 40: 13–15.11. Philliber SG, Jones J. Staffing a contraceptive service for adolescents: the importance

of sex, race and age.Public Health Report1992; 97: 165–169.12. Brink Muinen A van der, Bakker DH de, Bensing JM. Consultations for women’s

health problems: factors influencing women’s choice of sex of the general practi-tioner.British Journal of General Practice1994; 44: 205–210.

13. Orzano AJ, Cody RP. Gender concordance between family practice residents anddiagnoses in an ambulatory setting.Family Medicine1995; 27: 440–443.

14. Pendleton D. Doctor – patient communication – a review. In: Pendleton D, Hasler J,eds.Doctor – Patient Communication. London: Academic Press Inc, 1983: 5–53.

15. Colameco S, Becker LA, Simpson M. Sex bias in the assessment of patientcomplaints.Journal of Family Practice1983; 16: 1117–1121.

16. Armitage KJ, Schneiderman LJ, Bass RA. Response of physicians to medicalcomplaints in men and women.JAMA1979; 241: 2186–2187.

17. Peterson MC, Holbrook JH, von Hales D, Smith NL, Staker LV. Contributions ofthe history, physical examination, and laboratory investigation in making medicaldiagnosis.West J Med1992; 156: 163–165.

THE (GENDERED) CONSTRUCTION OF DIAGNOSIS 285

18. Ayanian JZ, Epstein AM. Differences in the use of procedures between women andmen hospitalized for coronary heart disease.N Engl J Med1991; 325: 221–225.

19. Kjellstrand CM. Age, sex, and race inequality in renal transplantation.Arch InternMed1988; 148: 1305–1309.

20. Safran DG, Rogers WH, Tarlov AR, McHorney CA, Ware JE JR. Gender differencesin medical treatment: The case of physician-prescribed activity restrictions.Soc SciMed1997; 45: 711–722.

21. Lurie N, Slater J, McGovern P, Ekstrum J, Quam L, Margolis K. Preventive care forwomen. Does the sex of the physician matter?N Engl J Med1993; 329: 478–482.

22. Bensing JM, van den Brink Muinen A, deBakker DH. Gender differences in practicestyle: a Dutch study of general practitioners.Med Care1993; 31: 219–229.

23. Bouchard L, Renaud M. Female and male physicians’ attitudes toward prenataldiagnosis: a Pan-Canadian survey.Soc Sci Med1997; 44: 381–392.

24. Seto TB, Taira DA, Davis RB, Safran C, Phillips RS. Effect of physician genderon the prescription of estrogen replacement therapy.J Gen Intern Med1996; 11:197–203.

25. Banazak DA. Late-life depression in primary care. How well are we doing?J GenIntern Med1996; 11: 163–167.

26. Wallen J, Waitzkin H, Stoeckle JD. Physician stereotypes about female health andillness: A study of patients sex and the informative process during medical interviews.Women & Health1979; 4: 135–146.

27. Hall JA, Irish JT, Roter DL, Ehrlich CM, Miller LH. Gender in medical encounters:an analysis of physician and patient communication in a primary care setting.HealthPsychol1994; 13: 384–392.

28. Lilleaas UB.Når forskjellen blir synlig – kvinner med kroniske muskelsmerteri et kjønnsrolleperspektiv. [Making the difference visible – women with chronicmuscular pain in a gender role perspective] Arbeidsnotat 8/95 – Senter for kvin-neforskning. Oslo: Universitetet i Oslo, 1995 (in Norwegian).

29. Johansson EE, Hamberg K, Lindgren G, Westman G. “I’ve been crying my way” –qualitative analysis of allmenn group of female patients’ consultation experiences.Family Practice1996; 13: 498–503.

30. Lupton D.Medicine as Culture. Illness, Disease and the Body in Western Societies.London/Thousand Oaks/New Dehli: Sage Publications, 1994: 131–160.

31. Eccleston C, deWilliams C, Rogers WS. Patients’ and professionals’ understandingsof the causes of chronic pain: Blame, responsitility and identity protection.Soc SciMed1997; 45: 699–709.

32. Sherwin S.No Longer Patient. Feminist Ethics and Health Care. Philadelphia:Temple University Press, 1992: 145–153.

33. Koran LM. The reliability of clinical methods, data and judgments.N Engl J Med1975; 293: 642–701.

34. Elmore JG, Wells CK, Lee CH, Howard DH, Feinstein AF. Variability in radiologists’interpretation of mammograms.N Engl J Med1994; 331: 1493–1499.

35. Latour B, Woolgar S.Laboratory Life. The Construction of Scientific Facts. Prin-ceton/New Jersey: Princeton University Press, 1986.

36. Hampton JR, Harrison MJG, Mitchell JRA, Prichard JS, Seymour C. Relativecontributions of history-taking, physical examination, and laboratory investigationto diagnosis and management of medical outpatients.British Medical Journal1975;2: 486–489.

286 KIRSTI MALTERUD

37. Melbye H. Diagnosis of pneumonia in adults in general practice.ScandinavianJournal of Primary Health Care1992; 10: 226–233.

38. Johnsen R, Bernersen B, Straume B, Førde OH, Bostad L, Burhol PG. Prevalences ofendoscopic and histological findings in subjects with and without dyspepsia.BritishMedical Journal1991; 302: 749–752.

39. Daniel, SL. The patient as text: A model of clinical hermeneutics.TheoreticalMedicine1986; 7: 195–210.

40. Fehr D. Kropp og symptom i en tegnteoretisk kontekst. [Body and symptom in asemiotic theoretical context].Sosiologi i dag1996; (4): 65–77 (in Norwegian).

41. Rudebeck CE. General practice and the dialogue of clinical practice. On symptoms,symptom presentations and bodily empathy.Scandinavian Journal of Primary HealthCare1991; Suppl 1: 72–81.

42. Scott JW. “Experience”. In: Butler J, Scott JW, eds.Feminists Theorize the Political.New York/London: Routledge, 1992: 22–40.

43. Hunter KM. Doctors’ Stories. The Narrative Structure of Medical Knowledge.Princeton/New Jersey: Princeton University Press, 1991: 3–26.

Centre for Feminist ResearchUniversity of OsloP.O. Box 1040 BlindernN-0315 OsloNorway

Address for correspondence:University of BergenUlriksdal 8CN-5009 BergenNorway