1498_3-1_04 43.54

The deep play of medicine: Discursive and collaborative processing of tiful. Statistical and computerized models, decision-analytic techniques, and other standardizing princi- Ethnographic research was conducted in the thoracic ples have been developed over time in order to ward of a Norwegian university hospital in order to overcome the mistakes and analytical shortcomings study collaborative medical problem solving. As a gen- of individual doctors and to rationalize, standardize, eral principle, evidence-based medicine is supposed to and make medical problem solving more reliable. A lead the process of medical problem solving. However, major problem with such models has been and is medical problem solving also requires evidence of a dif- that they ignore the social complexities of cognitionin general and of medical work in particular (Ma˚seide ferent kind. This is the more concrete form of evidence, 1983; Cicourel 1987, 1990; Berg 1997). At the same such as X rays and other representations, that guides time, organizational models have been developed to medical practice and makes sure that decisions are govern medical practice within hospitals and more grounded in sound empirical facts and knowledge. In extensive healthcare systems. Clinical governance, medicine, ‘evidence’ is on the one hand an abstract cat- standardization of practice, and the demand for ac- egory; on the other hand, it is a tool that is practically countability have been critically discussed in many enacted during the problem-solving work. Medical evi- publications (Kitchener 2000; Wiener 2000; Gray dence does not ‘show itself ’. As such it has an emergent and Harrison 2004; Degeling et al. 2004); again the quality. Medical evidence has to be established and problem seems to be the neglect of the social com- made practically useful in the collaborative settings by plexities, situational dynamics, and concreteness of the participants in order to make conclusions about di- clinical practice within healthcare organizations.
agnoses and treatment. Hence, evidence is an interac- Evidence-based medicine (EBM) is an instrument developed for clinical governance and rational med- tional product; it is discursively generated and its appli- ical problem solving (Field and Lohr 1990). It em- cability requires discourse. In addition, the production phasizes the rationalization and regulation of clin- of medical evidence requires more than medical dis- ical medicine through the systematic and empirical course and professional considerations. This paper grounding of clinical practice guidelines. Timmermans looks at the production processes and use of medical and Berg (2003) have provided a sociological study of evidence and the ambiguous meaning of this term in EBM that is critical of some understandings of it. Yet they are positive with regard to the principle of stan-dardization in clinical medicine provided its multi- Keywords: medical evidence; collaborative work; med- dimensionality is recognized, and they are critical of ical problem solving; hybridity; situated practice.
many of the critics of medical standardization.
This paper deals with evidence in medicine. It does not criticize the principles of evidence-based medicine and standardization in healthcare as they Attempts to improve and rationalize medical work are understood by Timmermans and Berg. It is in- and accomplish clinical governance have been plen- stead an attempt to describe, from below, whereand when practical medical work is conducted, andwhat takes place when evidence is made useful and A‰liation(s): P. Ma˚seide (Bodø Regional University)Correspondence to: Per Ma˚seide used within a medical regime where the principles of EBM are adopted. An important focus is on Communication & Medicine 3(1) (2006), pp. 43–54 !eeerrreeeiiitttgggeeesssttteeellllllttt      vvvooonnn      |||      NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk      (((NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk))) AAAnnngggeeemmmeeellldddeeettt      |||      111777222.111666.111.222222666 HHHeeerrruuunnnttteeerrrgggeeelllaaadddeeennn      aaammm      |||      111666.000222.111222      111222:::333666 the collaborative and discursive dimensions of these ular relationship between the speaker and the world’.
Di¤erent medical ‘communities of practice’ within the Di¤erent forms of evidence influence and regulate hospital organization have di¤erent voices, and di¤er- the judgments and decisions of medical practitioners.
ent problems and situations require di¤erent voices.
The first form to be mentioned is the scientifically In many cases, ‘the voice of medicine’ includes a mix- established evidence of EBM; second, evidence from ture of professional and nonprofessional voices, some personal medical experience as mentioned by Tim- of them belonging to the institution and some to the mermans and Berg (2003: 3); third, evidence as life world. The idea behind my use of the term ‘voice’ medical representations or artifacts; and finally the is simply to suggest that when an agent speaks, his or evidence that is cooperatively and discursively con- her talk is not necessarily what I would term a ‘per- structed by practitioners in clinical meetings and sonal expression’. An agent’s verbal expression may conferences. The last form may be termed ‘practical be used to mediate information from other agents, to evidence’; it is closely integrated with forms two and present professional or collective opinions and mean- three. To borrow a statement from Mol (2002: 164), ings, and to provide ‘institutional discourse’ (Hall et the practical evidence of situated practical medical al. 1999; Roberts and Sarangi 1999). Di¤erent voices work does not appear ‘because the body itself leaves may provide di¤erent perspectives on the ongoing ac- us with no alternative’. To this it should be added tivities; they may for instance be used to switch that neither do scientifically based procedures and between moral and medical frames. As a result, col- laborative medical work does not only require collab- Practical medical evidence is generated, developed, oration between various professional agents, it also and made useful locally by medical practitioners. It requires collaboration between di¤erent ‘voices’. In responds to a complex set of demands, intertwines this sense, it might have been relevant to distinguish medical and moral frames, and has an immediate im- between medical voices. For the present purpose, pact on practical medical problem solving. It includes however, the relevant distinctions are between ‘the di¤erent kinds of descriptions and images of biolog- voice of the institution’, ‘the voice of the practitioner’, ical or pathological conditions of a patient’s body, or ‘the voice of representations or images’, ‘the voice of of other conditions made relevant to the ongoing and patients’, and sometimes what Mishler (1984) called patient-directed problem-solving activities. In this ‘voices of the life world’. ‘The voice of the institution’ sense, ‘medical evidence’ does not only refer to bio- is given a relatively wide definition. It refers to profes- logical and physical realities of a patient’s body; sional medicine’s institutionalized stock of knowledge sometimes it also refers to psychological or moral as well as to various discursive forms that regulate and often standardize medical work in hospital settings.
The empirical material is taken from fieldwork in a thoracic ward of a Norwegian university hospital.
The data were collected in various kinds of ward con-ferences; most of them in a weekly interdisciplinary A major topic of the following pages is the generation conference called the thoracic meeting. Data exist and use of practical medical evidence in collaborative partly as audio recordings and partly as field notes.
processes of problem solving in thoracic medicine. A In the thoracic meeting, surgeons, oncologists, radiol- second topic is the role of moral considerations in the ogists, and pathologists participated and had vital production and use of such evidence. When referring roles. They were supposed to conduct evidence-based to medical practice, the term ‘moral’ has a broad medicine in its various senses, and they probably did.
meaning taken from Go¤man (1956, 1971, 1981). It However, Norwegian and international studies show refers to common moral problems, to moral identi- that in spite of the existence and recognition of such ties, statuses, or characters involved in medical work, standardized clinical guidelines, there is considerable and to the ‘moral work’ conducted by these charac- variation between hospitals using them and between ters (Ma˚seide 2003). In this sense, moral kinds may countries with regard to the treatment of lung can- represent problems facing medical practitioners, but cer. A recent publication suggests for instance that they may also represent resources for institutionally more Norwegian patients with lung cancer should be adequate solutions to the medical problems at hand.
o¤ered surgical treatment than the case has been The paper has two major empirical and analytical sections. Section 4 deals with how ‘evidence’ may beunderstood in practical medicine. The longer Section5 focuses on the notion of ‘voice’. This term has certain theoretical connotations (Wertsch 1991), andseveral analysts have described the institutional voice In important settings where medical work is con- of medicine (Mishler 1984; Anspach 1988; Ma˚seide ducted, various forms of representations or ‘cognitive 2003). According to Atkinson (1995: 129), medicine artifacts’ (Hutchins 1995) replace the patient’s body.
has di¤erent voices, and ‘each voice realizes a partic- These representations become the practical and !eeerrreeeiiitttgggeeesssttteeellllllttt      vvvooonnn      |||      NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk      (((NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk))) AAAnnngggeeemmmeeellldddeeettt      |||      111777222.111666.111.222222666 HHHeeerrruuunnnttteeerrrgggeeelllaaadddeeennn      aaammm      |||      111666.000222.111222      111222:::333666 concrete objects of medical work and contribute to establishing the empirical evidence required for med- A resident physician presents the case; the pa- tient is a man in his forties, earlier diagnosed As medical know-how develops, the requirements with an inoperable lung cancer and treatment of knowledge, competence, physical capacity, and with cytostatics was initiated. Two sessions of technology have grown far beyond the capability of chemotherapy are finished; in this meeting the individual practitioners. To meet the cognitive and e¤ect of these sessions is evaluated before a deci- physical limitations of professional agents, a complex sion about a third session can be made.
working environment is generated to solve intricate The resident physician reports that the patient medical problems. It includes di¤erent medical spe- has not had much discomfort from the two pre- cialties and clinical units, representing diverse ‘com- vious sessions. After this he shows the relevant munities of practice’, technologies and tools for x-rays on a light box. An x-ray from before the standardization, systems for storing and processing sessions started and an x-ray taken after the first information, etc. Such working environments repre- session was finished are compared with the x-ray sent what at least metaphorically may be called ‘so- taken after the second session. On the first x-ray cially distributed cognitive systems’ (Hutchins 1995).
the resident shows multiple abnormal spots and In this sense, the hospital is a socially distributed shadows located in both lungs. The spots and cognitive system. The ward conferences represent shadows are clearly visible. On the x-ray taken another form of distributed system. In what philos- after the first session the spots are almost gone ophers of mind call an externalist perspective (Wil- and the shadows have become distinctly lighter.
son 2004), individual human cognition is hardly re- This visual di¤erence between the two x-rays stricted to brain processes and may also be described counts as clear evidence of improvement and it as socially distributed. In distributed systems, how- is concluded that the first session has had a sig- ever, even the ‘core’ cognitive processes (individual acts such as perception, calculation, etc.) are so- The second x-ray is then compared with the most cially distributed. Without implicating conceptions recent. On both x-rays the resident shows multi- of ‘group minds’ or ‘super organisms’, a ward con- ple small spots and thin shadows in both lungs.
ference, with a number of individual members with One of the consultants declares that in his opin- di¤erent qualifications, functions, responsibilities, ion the shadows look thinner on the new x-ray.
skills, and experiences, has knowledge and memory Some other doctors comment that they do not structures, procedures of reasoning, and practical see any di¤erence. This di¤erence of view is qualifications that are socially distributed and di¤er from the cognitive capabilities of the individual Then one of the most experienced consultants puts forth in a low voice that in his opinion,this is an unanimous indication supporting anew session.
Evidence and objects in clinical medicine The chief physician responds by saying that he isambivalent, he is not sure if the new x-ray reallycan count as indicator of improvement.
Another consultant, however, declares in a firm The most common treatment alternatives in cases of voice that in his opinion the x-ray should pass as inoperable lung cancer are radiation or chemother- apy, often in combination. Evaluation of the e¤ect There is no verbalized opposition to this view of the chosen treatment regime requires empirical and the new x-ray passes as evidence for im- evidence; in the thoracic ward, such evidence is usu- provement and thus as indicator for a new ally provided by comparing new X rays with older The chief physician takes the final and conclud- When comparing X rays in such cases, the size of ing verbal turn by declaring that this is also a the image of the tumor on the di¤erent X rays is mea- patient who wants to fight; he is a robust person.
sured to see if it has grown or is decaying, or thee¤ect of the treatment is determined by visually com- In this case the final meaning of radiological evidence paring X rays to see if the shadows representing the was established not so much on the basis of pure vi- tumor look thinner or more opaque now than before.
sual scrutiny as on talk. Professional medical vision Decisions about further treatment should depend on is not only dependent on an individual medical prac- such radiological evidence. However, the question of titioner’s biomedical coding of an X ray; instead, a radiological evidence is not necessarily a question of process of collective vision based on discourse and plain vision or individual perception. An episode moral conditions unfolds. The doctors who know (Extract [1]) reconstructed from field notes of a ward the patient are sympathetic toward him and want to give him whatever chances there may be. Prolonged !eeerrreeeiiitttgggeeesssttteeellllllttt      vvvooonnn      |||      NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk      (((NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk))) AAAnnngggeeemmmeeellldddeeettt      |||      111777222.111666.111.222222666 HHHeeerrruuunnnttteeerrrgggeeelllaaadddeeennn      aaammm      |||      111666.000222.111222      111222:::333666 treatment is the only hope the doctors can o¤er for a the surgeon’s thinking about how to solve the specific longer and better life, but they need radiological evi- dence to justify a third session. When a positive for-mulation of the visual evidence is accepted, the chief ‘Objects of knowledge’ versus ‘actual things’ physician emphasizes the patient’s moral character(the patient wants to fight, he is a robust person) to A diagnosis is not a disease; it is a form of knowledge.
support and legitimate their conclusion.
The distinction between a diagnosis as the name of a Here evidence was not used to solve a medical certain medical condition and the actual bodily condi- problem; it was discursively transformed to support tion may be similar to the distinction Knorr Cetina and legitimize the solution to a moral problem. The (2001) makes between ‘epistemic objects’ and ‘actual evidence itself does not create a solution; it is made nature’. A diagnosis refers to the disease as ‘actual na- into something that will support a preferred solution.
ture’, but the diagnostic category represents an episte- The solution is in principle evidence-based, except mic object or ‘an object of knowledge’. The distinction that the significant practical evidence is locally and between these two levels of reality is often blurred in collaboratively produced, not so much from scientifi- collaborative medical work. The same kind of distinc- cally based knowledge about e¤ective treatment of tion is important with regard to representations and lung cancer as from moral considerations.
evidence. Medical representations are not ‘actual na- However, the case continues after Extract (1).
ture’ (except in a trivial physical sense) but ‘objects of When they have reached a shared understanding of knowledge’. They are ‘social objects’ in the sense that the radiological evidence, a resident reports that a we in Mead’s words (1972 [1938]: 292) may carry on ‘suspect change’ in the patient’s left hip has just been social intercourse with them. Since they may at the found. This new information may change their pro- same time have multiple or indeterminate meanings, fessional vision of the X rays. The thin shadows still often related to diverse communities of practice, they to be seen, which just now was evidence of remission, may also be understood as ‘boundary objects’ (Star may by the remark about the ultrasound be made and Griesemer 1986). An important part of the discur- into a disease that has spread to the patient’s skele- sive practices of collaborative medical work is to bring ton; if that is the case, a new cell treatment will be di¤erent divisions of the medical expert system or the out of the question. A short discussion follows about hospital’s diverse ‘communities of practice’ together whether the ‘change’ in the hip might be metastases to generate shared definite meanings out of the am- before the chief physician closes the case and saves biguous medical representations. Medical evidence the situation by characterizing the new discovery: should thus not be confused with ‘nature’. It belongs ‘We will say that it has not progressed’. With this for- to systems of knowledge, discourse, and institutions.
mulation he makes the information about new visual Hence, much of the medical work is oriented more to- ward objects of knowledge, representations, cognitive Even if the general principles of evidence-based artifacts, and social objects than toward actual na- medicine are accepted and empirical evidence from ture. These boundary objects have di¤erent meanings X rays is used, the problems of how to follow the clin- and relate to di¤erent practices for those who attend ical guidelines of evidence-based medicine and how to to them. As indicated by Extract (1), diagnoses and establish and use evidence in practice, applied to con- representations are not singularly medical objects; crete cases, remains to be solved. These general and they are also moral and social objects belonging to abstract principles need not be the only principles in- the life world, thus bringing the life world into the forming or directing decisions about diagnosing or practice of evidence-based medicine.
treatment. Another episode from the ward illustrates The representations or objects of knowledge con- this. A female patient in her forties was diagnosed centrated on in this study belong mainly to two with lung cancer. The main question was whether forms. One form consists of X rays and similar visual to perform surgery or not. The chief physician ar- images. The other form of representations, verbal for- gued in favor of an operation, while the surgeon mulations such as descriptions, references, proposi- was hesitant. After a long and heated discussion, tions, claims, statements, definitions, identifications, the surgeon declared that from the information he presentations, etc., is even more frequently used in had about the case, he would not operate. The chief collaborative medical work. To a large extent, they physician responded immediately by asking, ‘would function as collective representations and refer to is- you say the same if she were your wife?’ The sur- sues of everyday life shared by doctors and patients.
geon answered, ‘I don’t know, I have to think itover again’. The chief physician brought a nonmed- Voices and voicing in collaborative practice ical perspective into the discussion. He spoke withwhat Mishler (1984) termed ‘the voice of the life world’. According to the principles of evidence-basedmedicine, this should be an irrelevant perspective for The patient’s participation in the problem-solving this kind of problem solving, but it obviously a¤ected processes of a ward conference was enacted through !eeerrreeeiiitttgggeeesssttteeellllllttt      vvvooonnn      |||      NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk      (((NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk))) AAAnnngggeeemmmeeellldddeeettt      |||      111777222.111666.111.222222666 HHHeeerrruuunnnttteeerrrgggeeelllaaadddeeennn      aaammm      |||      111666.000222.111222      111222:::333666 the talk of a doctor. It was not very common, but if it ‘idiosyncratic’ voice than the collective ‘institutional happened, it was because the situation required the patient’s voice. Extract (2) refers to the opening of acase presentation in a ward conference (see the Ap- pendix for transcription conventions). It provides an may alternate between using ‘the doctor’s voice’, ‘an example of how the patient’s voice is used by practi- institutional voice’, and ‘the patient’s voice’. We may tioners in collaborative work. In this case, the pa- then use the term ‘hybrid voices’ (Roberts and Sarangi tient’s voice provides part of the relevant evidence.
1999). There is an interplay between the doctor’s, the The patient’s voice may not always be trusted, and institution’s, and the patient’s voices. This can be seen when the doctor rehearses the patient’s voice, this ele- in Extract (3), which refers to a case presentation from ment of trust may be manipulated according to a combination of institutional and moral standards by the doctor’s way of rephrasing the patient.
The resident starts by presenting institutional facts about the patient’s medical history. After this (from end of line 3), subjective information from the patient is presented. Her voice is forwarded by the resident’s talk as a kind of ventriloquation (Wertsch 1991).
‘Her’ information about subjective experiences (line 6) is rephrased by the doctor as brute facts. However, a change takes place when he reaches the topic of smoking and states that the patient ‘maintains’ she has stopped smoking. Now the doctor formulates a distance or a weaker commitment to the patient’s subjective information, which was not expressed The first part of the patient’s history is presented as straight factual information. But when it comes to smoking, the resident specifies that this is what the patient said, and by that he adds a subjective quality to the patient’s information, which was not there be- fore. This provides the information about smoking habits and the patient who provided the information with a certain moral quality. Tiredness and loss of weight are not morally contaminated in thoracic med- The first part of the case presentation is without refer- icine (as it may be in other parts of medicine), and the ence to biomedical information. Instead the resident patient is obviously trustworthy on such topics, but physician, who knows the patient, provides her with not necessarily so with regard to smoking.
personal properties such as age and character. Thepatient is described as ‘elderly’, but she is ‘clear’. Thelatter is a popular cognitive category; she is ‘clear in her head’, but related to age it also becomes a moral Doctors work within a collective system and have category. She is also described as ‘funny’, which their identities and qualifications from this system; means she may be lively and entertaining, but also however, they also represent part of the agency of amusing in an unintended and less respectable way.
the system and they bring with them individual ex- Hence, a moral description becomes part of the evi- periences and opinions. ‘The doctor’s voice’ is the dence made relevant for the case. The moral presenta- voice of the specific practicing physician and relates tion lasts into line 7 before the chief physician indi- to his or her individual and direct medical practice, cates that enough has been said about personal and knowledge, experience, and judgment. It is a more moral characteristics. The resident then introduces a !eeerrreeeiiitttgggeeesssttteeellllllttt      vvvooonnn      |||      NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk      (((NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk))) AAAnnngggeeemmmeeellldddeeettt      |||      111777222.111666.111.222222666 HHHeeerrruuunnnttteeerrrgggeeelllaaadddeeennn      aaammm      |||      111666.000222.111222      111222:::333666 medical problem (‘. . . but the point with her anyway is that she has been tormented by pains in her chest for years . . .’) and keeps strictly to biomedical matters until line 28. During this talk, he switches between the institutional voice (lines 10–20) and the doctor’s voice (lines 20–28). Then he refers to the patient’s voice and says, ‘she strongly wishes to try this new wonder medicine Serevent’ (lines 28–30).
In using the patient’s voice, he does not just express Information about a tumor that is sharply limited the patient’s wish to try a specific drug. He adds against the aorta and information about an enlarged ‘strongly’ which emphasizes agency as well as person- lymphatic gland are both medically significant. The ality in the patient, and he describes the drug as ‘this first may indicate a surgical solution while the second new wonder medicine’. By adding this phrase to the speaks against it. However, a failure to identify a de- patient’s expressed wish to try Serevent, he formulates scribed visual fact, even when such evidence may be a message with a meaning di¤erent from the plain critical for the case, need not make the radiological fact of the patient’s wish. He formulates a message description invalid. By using the ‘doctor’s voice’ to that characterizes the patient; it is directed to the au- express the authority of personal experience (and dience and requires some kind of response to be com- probably also to appeal to his local identity as radio- pleted. The response comes immediately as laughter logical expert), the radiologist indirectly assures the so the resident’s phrasing has made laughter into an audience that what seems to be invisible (the tumor appropriate response. Something nonmedical, belong- as an institutional fact), is in fact visible. He simply ing neither to the doctor nor to the institution, but to declares that the lymphatic gland is described. This is the patient’s moral character, has been made relevant a direct emphasis of the institutionally confirmed exis- for the case. The medical problem is collectively made tence of the invisible facts. He makes the institutional into ‘a funny-old-lady-disease’, which is a mixture of description of a visible fact overrule the situational medical and moral matters. Such ‘hybrid diseases’ are not uncommon in clinical medicine (Sarangi and In principle, then, a radiological fact need not be Roberts 1999). Mingling of di¤erent voices and pair- seen, but it must at least be heard, and in cases like ing the phrasing of presentations with the audience’s this it was common to ask who had made the descrip- responses have produced a form of ‘moral evidence’ tion. If the radiologist who made the institutional description has a local identity in the hospital as ahighly valued and respected radiologist, then the de- scription of an invisible lymphatic gland might be ac- The meaning of visual evidence is often questioned.
cepted as positive, even if the evidence is visually Some times verbal formulations may even replace vi- negative. In such cases, we might talk about ‘virtual sual evidence; or more precisely, an invisible fact is evidence’, which is not based on direct observations made visible through verbal formulations. The radiol- but on reports and formulations from practitioners ogists who presented X rays referred more than once with professional authority or the authority of per- to discoveries described by the radiology department without being able to see and show these discoveries.
In order for X rays or other radiological images to X rays used for medical problem solving are de- count as significant radiological evidence, a local and scribed. These descriptions are practical tools in med- shared agreement had to be established about what ical work. Some X rays are described by the ward’s was actually or virtually observed. In most cases, the doctors; radiologists from the radiology department questions related to actual observations, and not in- describe other X rays, and most notably pictures frequently to the meaning of what was observed.
from computed tomography or ultrasound examina-tions. Such descriptions inform how the X ray should be seen and understood by those who use it. The radi- ologist who presents the X rays to the users is often not the same one who has written the descriptions; but descriptions from the radiology department are in- stitutionally authorized and represent the department’s voice. However, the individual radiologist’s voice may also be activated in order to generate adequate evidence. An example of this is shown in Extract (4).
!eeerrreeeiiitttgggeeesssttteeellllllttt      vvvooonnn      |||      NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk      (((NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk))) AAAnnngggeeemmmeeellldddeeettt      |||      111777222.111666.111.222222666 HHHeeerrruuunnnttteeerrrgggeeelllaaadddeeennn      aaammm      |||      111666.000222.111222      111222:::333666 constructions and presentations of evidence in ward Institutional voice and the virtual patient.
The process of collaborative problem solving in the thoracic meeting starts with the resident providing in- formation, which for the moment is granted rele- vancy, about a given case. It has to do with what Berg (1992) termed ‘the construction of medical dis- posals’. A problem that the ward can relate to has to be formulated and accepted. The institutional voicepresents this information and creates an impression Extract (5) refers to a discussion among doctors in the of factuality. It draws the attention toward isolated thoracic meeting about an ultrasound image from the parts of the patient’s body, to specific and limited bi- examination of a patient’s liver. A malignant tumor ological processes, to procedures without individual had previously been discovered in one of this patient’s agency, and to a large extent to representations and lungs. Following common routines for such cases, the images. It rarely focuses on the processes or problems abdomen of the patient was examined by ultrasound.
of generating evidence. At the same time, however, a The resident physician who presents the patient has kind of personal corporeality is discursively gener- just reported that something suspected to be a tumor ated, so that in a biomedical situation without bodies is described in the patient’s liver after the abdominal the medical problem is connected to a specific patient.
ultrasound. This may indicate metastasis from the Extract (6) shows how a case usually is presented in lung tumor, which would then be considered inopera- ble. When Extract (5) starts, the radiologist reportsthat he is skeptical of the correctness of the reported result and has talked directly to the doctor who had described the suspected tumor; he has also discussed the ultrasound discovery with another doctor, identi- fied by name, who is considered the most prestigious local authority on ultrasound examinations. The lat- ter has also seen the video recordings in question and shares the radiologist’s skepticism. After the expert’s judgment was formulated, the examiner had, accord- ing to the radiologist, declared that he might want to change his own description. The meaning of the ultra- sound image is thus altered, and so are its medical consequences. The chief physician of the ward acceptsthe radiologist’s account and dismisses the liver. A The resident starts the case construction by naming a conclusion is made. It is grounded in visual evidence, person, the patient, followed by a short pause. The but the meaning of what is seen, the practical evi- pause creates what Goodwin (2002) has termed ‘high- dence, is a product of personal involvement, experi- lighting’ of a piece of information. It makes the ence, individualized professional authority, and col- named person stand out and become significant for laborative talk. Such transformations of medical what is to follow; it also introduces a moral dimen- evidence may have to do with their moral conse- sion to the case. In a situation where formulations quences. If the initial description of a tumor in the and artifacts represent medical evidence, the initial patient’s liver had been accepted as a biological fact, naming of a person connects the abstract world of potentially life-saving surgery would have been ruled images, representations, and pathology to a specific person—sometimes even to a life world. A virtualpatient is generated by a verbal utterance; and ab-stract medical information may be attached to this patient. In this sense, the institutional and ‘disem-bodied’ medical discourse of the thoracic meeting In certain institutional situations and activities, the is also embodied. Even if a specific body is lacking, it ‘institutional voice’ of medicine has a collective and is formulated as a social fact and made relevant for depersonalized form; it is characterized by frequent the following discourse. The naming of the patient use of collective or impersonal pronouns like ‘we’ or serves as an organizational resource. Doctors will be ‘one’ and a dominant passive form (Anspach 1988).
able to identify the patient; he is someone they have It is a voice without individual agency but with met, talked to, perhaps examined, ascribed identities institutional authority. It is typically used for case and moral values to, and they have usually talked !eeerrreeeiiitttgggeeesssttteeellllllttt      vvvooonnn      |||      NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk      (((NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk))) AAAnnngggeeemmmeeellldddeeettt      |||      111777222.111666.111.222222666 HHHeeerrruuunnnttteeerrrgggeeelllaaadddeeennn      aaammm      |||      111666.000222.111222      111222:::333666 about in other settings. A web of cultural, moral, and interpersonal conditions is thus made relevant for the After the naming, the problem at hand is formu- lated as a chronologically ordered story about sig-nificant facts. The facts are presented in an institu- Extract (7) starts with the resident saying, ‘This is tional voice as connected and coherent. In this story, Hans Hansen who has a small density in his chest’.
only one instance of personal agency is identified, the When the initial word ‘this’ is uttered, the doctor named patient, who went to see a doctor and started turns to an X ray of the thoracic region attached to a process that brought a medical problem to the ward.
the light box. Visually he thus orients to the X ray After that, the health services take over the problem.
and proceeds by using a pointer to point at the X ray A general practitioner has treated the patient for ten- of Hans Hansen’s chest. The indexical expression donitis, the tendonitis (not the patient) did not im- ‘this’ and the simultaneous naming of a person refer prove, an X ray was then taken (by an institution), on the one hand to a specific patient, who is physi- and this X ray shows (seemingly without anyone’s cally absent, and on the other hand to a physical help) a ‘density’ in the patient’s lungs. This absence object, a representation, which is there to be seen.
of agency or identifiable agents is typical of institu- These communicative acts—talking, pointing, and tional processes described in medicine (Atkinson looking—work together in relation to the representa- 1995). The presented story is not responded to with tion and do at least two things at once, related to questions or other comments from the audience; it is what Go¤man (1981) called ‘footing’. The resident’s accepted and an institutional starting point for the simultaneous expressions make the patient into the further problem-solving process has been established.
representation or artifact, which constitute the episte- While the voice of medicine may represent insti- mic object to be talked about; and they make the rep- tutional procedures and express professional knowl- resentation into a person. The physical and cognitive edge, rationality, and accuracy, it relies to a large artifact attached to the light box is Hans Hansen, extent on indirect and unspoken messages. The insti- who in this setting is an X ray. The distinction be- tutional voice of medicine requires something like a tween person and representation is blurred. This is ‘conversational or conventional implicature’ (Levin- the virtual patient, who is also a boundary object.
son 1983) to make the restricted institutional dis- The virtual patient permits di¤erent approaches and course work. The institutional voice implicates a cer- discourses, from the medical to the moral ones.
tain attitude among the professional audience based The ambiguity points to an aspect of many kinds of on medical and institutional knowledge, competence, medical evidence. They have a quality of ‘openness’ and information. This also means that a set of med- with regard to meaning, which has to be ‘closed’ dur- ical and nonmedical felicity conditions are required to ing the problem-solving process. This is also a charac- communicate medical evidence with an institutional teristic of boundary objects. Is the medical ‘thing’ at hand or the object to be approached by the team ofdoctors a person or an artifact? The object of medicalreasoning is both, but not simultaneously, not all the time, not at all places, and not for all those present.
‘The voice of representations’ is used when represen- In the context of the thoracic ward, the medical ob- tations are major objects and when they are made to ject is a patient and often a person with symptoms ‘talk’ (like a pathological report) or ‘display’ a med- and signs of disease, who is there to be examined, ob- ical object (like an X ray). Statements like, ‘the X ray served, talked to, cared for, and socialized with. In shows’, ‘histology shows’, or ‘the tests are good’ make the context of the thoracic meeting, the patient is split representations act as agencies. The boundary be- into several things. They include representations and tween expressions made in the voice of representa- pieces of evidence with di¤erent meanings for those tions and other relevant voices may be fuzzy.
present. The ambiguity of the medical object is oftenintroduced in the case presentation, and it permits the multiplicity of voices that may participate in the When formulating the location of the density seen on the X ray, the patient’s condition, and results from other examinations, the resident uses a verbal technique of pausing to make each part stand out and to produce significance. This form of presenta- tion seems to complete a picture of an innocent radio- logical discovery. In lines 9 and 10, however, the resi- dent provides information that dramatically changes the picture. An ultrasound examination has found !eeerrreeeiiitttgggeeesssttteeellllllttt      vvvooonnn      |||      NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk      (((NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk))) AAAnnngggeeemmmeeellldddeeettt      |||      111777222.111666.111.222222666 HHHeeerrruuunnnttteeerrrgggeeelllaaadddeeennn      aaammm      |||      111666.000222.111222      111222:::333666 ‘an expansive process’ (this expression usually refers to malignancy) in the patient’s liver. That informa- tion about the liver makes the density in the lung sus- By careful use of the voice of representations, the resident carves out the empirical grounding for the collaborative process. The density is accessible on the X ray and it can be talked about in terms of the X ray, ultrasound, and recordings from other medical examinations. The density ‘is discovered by a routine x-ray screening’, the patient has ‘normal lung func- tions’, another product of medical technology, and ‘bronchoscopy is normal’. ‘CT thorax is also by and large ok’, but ‘ultrasound has discovered an expansive process’. This impersonal and nonsubjective voice of representations contributes to shaping facts and evi- dence. It makes medical matters and objects stand out as self-evident and not contaminated by the sub- jectivity of individual professional agents. In this re- spect they serve the principles of EBM as they are described by Timmermans and Berg (2003).
A number of di¤erent professional agents are involved in the collaborative problem solving of the ward, and these agents have diverse ‘voices’ at their disposals.
The ‘voices’ are selected and used for practical pur- poses. In the problem-solving process, then, an ex- change of di¤erent ‘voices’ takes place.
When Extract (8) starts, the resident physician has Several actors are involved in the process of generat- just presented a case in the thoracic meeting. It is a ing evidence that may qualify the patient for surgery.
male patient; his X ray has shown a lung tumor and There is the resident and the chief physician, but there bronchoscopy has been conducted. The first line re- is also the pathologist, the radiologist, the surgeon, fers to the result of this examination. The patholo- and the nurse. And finally there is the X ray on the gist’s report is the first piece of evidence presented.
light box and the CT pictures on the overhead screen.
The pictures from computed tomography shown after We find a mixture of ‘voices’. In the first lines of the that become the most significant form of evidence at extract, the resident physician is about to refer to in- formation that institutionally belong to the patholo-gist. The chief physician interrupts and declares that they should leave this to the pathologist. The resident has not respected the ownership of the pathological information and is corrected. As the institutional ex- pert with the authorized medical voice, the patholo- gist reports that they have no pathological classifica- tion for the tumor. The resident then declares the patient to be a candidate for surgery and gives a ra- diological account in institutional terms for his point of view. On the CT pictures shown on the screen a white spot is clearly visible; the surgeon asks the radi- ologist whether this is the tumor. He is told it is med- ically insignificant; instead the radiologist points to her location of the tumor. While she explains, she makes it visible to the audience by drawing the form of the tumor on the screen with a pointer. By talking and pointing to the picture on the screen, the radiolo- gist makes the tumor into the visual object to be seen.
!eeerrreeeiiitttgggeeesssttteeellllllttt      vvvooonnn      |||      NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk      (((NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk))) AAAnnngggeeemmmeeellldddeeettt      |||      111777222.111666.111.222222666 HHHeeerrruuunnnttteeerrrgggeeelllaaadddeeennn      aaammm      |||      111666.000222.111222      111222:::333666 This is done very directly when she says, ‘See, the tu- mor is located in the lower lobe’, while she points to it. She shows distinctly the demarcation of the tumor in order to demonstrate that it has not grown into nearby tissue. Also this requires talk in addition to pointing. Not only does she make the audience see the tumor; she localizes it precisely in the lower lobe of the lung. If surgery is the solution, it may be done with a lobectomy, which will have less impact than removing the lung. By pointing to and talking about the CT pictures in relation to the audience, the radiol- ogist creates visual evidence for a medical reality to be shared by those present. When the surgeon de- clares that they will operate, he explicitly accepts the The collaborative process of professional problem solving proceeds smoothly until the head nurse gives the patient a ‘voice’ and introduces what might be termed a ‘nonmedical’ aspect of the problem. It hasto do with the patient’s expressed need and wish to Extract (9) starts after the resident who presented the take care of his wife. This information brings in a patient has suggested that she may be a case for sur- new dimension; it belongs to the patient’s life world gery. The surgeon’s question about the patient’s ‘vital but is still significant for a medically adequate conclu- capacity’ refers to her measured respiratory capacity, sion of the case. By bringing the patient’s voice in, the which counts as significant evidence in a case like this.
problem has in a way developed out of the doctors’ Even if the patient’s lung function is declared to be hands. Now it cannot be properly solved without in- normal, however, such terms as ‘normal’ are medi- volving the patient. The chief physician will go into cally ambiguous; the meaning of such terms is often this moral part of the problem, while the surgeon pre- informed by contextual conditions. Medical knowl- edge about ‘normality’ is abundant; it is statisticallygenerated, standardized, and decontextualized. The Negotiating moral and corporeal evidence.
measured indicators of the patient’s lung capacity Descriptions of a patient’s moral character are often may be within the standardized limits of ‘normality’.
formulated and used when trying to avoid certain However, the concreteness and particularity of the undertakings or find adequate solutions to many problem, the patient, and the situation may change medical problems. This is illustrated in Extract (9), the meaning and applicability of such test results.
which is an excerpt from the thoracic meeting. The The resident has declared the lung capacity for nor- patient is a woman with a diagnosed lung cancer.
mal, but he pauses slightly before he continues, ‘but The question is whether she will qualify as a candi- she is eighty years old you know’. Then he pauses again. The pause and the initial ‘but’ are acts of foot-ing and indicate a change of commitment to his pre- ceding declaration. He puts the description of the pa- tient’s lung condition within a specific frame, where age becomes a prominent and significant condition and the medical meaning of her measured lung capac- ity changes and becomes less significant.
By formulating age as a vital property, another un- derstanding of the patient’s body is introduced, which may have negative consequences for the question of surgery. As the resident continues, however, he de- clares that the patient is very fit (lines 3 and 4). An- other specific bodily property with other medical pre- suppositions is formulated. It counters the previously suggested e¤ect of age. The patient is also quoted as wanting an operation. The last mentioning may be indicative of moral character and bravery, qualities that might support a decision about surgery. The way the resident finishes his description seems to express uncertainty about whether this is a surgical !eeerrreeeiiitttgggeeesssttteeellllllttt      vvvooonnn      |||      NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk      (((NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk))) AAAnnngggeeemmmeeellldddeeettt      |||      111777222.111666.111.222222666 HHHeeerrruuunnnttteeerrrgggeeelllaaadddeeennn      aaammm      |||      111666.000222.111222      111222:::333666 When the first surgeon enters the floor, he formu- was established for this case, which depended on and lates a tumor he doubts is operable. On the other supported the significance of a certain kind of general hand, he also formulates a peripherally located tumor.
and abstract medical evidence. The significance of the This latter remark undermines his first statement and patient’s corporeal particularities was taken out of the indicates that the surgical procedure may not be very case. This kind of evidence-based solution does not complicated. His formulations move the patient be- come from a pre-established scheme; it has the char- tween di¤erent medical statuses. However, the second acter of an emergent process requiring discursive and surgeon’s utterance is more deliberate. It is her age, collaborative manipulations of medical and moral and indirectly not the location of the tumor, that counts. This urges the other participants to ignorethe previously formulated facts about her physicalcondition. The second surgeon is senior to the first so his statement is privileged in this setting. The residentexpresses no direct opposition to the turn the case The principal object of biomedical problem solving may take; however, he introduces the possibility of includes patients, bodies, and diseases. But at the letting the patient’s voice count in this case (lines 14 same time, the concrete situations in which problem- solving activities take place are often characterized by In this process of problem solving, verbal formu- the absence of concrete, and therefore in a certain lations and descriptions of personal and physical sense ‘real’, patients, bodies, and diseases. Present conditions count as evidence. The process of collab- are instead medical practitioners and di¤erent kinds orative professional talk has now formulated and of artifacts. These parties interact and collaborate dis- manipulated various relevant facts and produced a cursively to produce practically relevant and useful risk case. The significant risk criterion, age, is an evidence for the problem at hand, as this problem is abstract statistical measure, as opposed to the re- developed during the problem-solving process. Dur- ports about specific and actual qualities and condi- ing this process, doctors work with colleagues and tions of this particular patient. Both forms of criteria other health professionals; together they relate will be relevant in evidence-based decisions. In this actively and practically to various kinds of represen- case, the general condition ‘age’ seems to be preferred tations and other decision-supporting tools, and to over the specific condition ‘vital capacity’ when the diverse medical and managerial systems of clinical group of problem solvers tries to formulate relevant governance. ‘Real patients’, ‘real bodies’, and ‘real and valid conditions to guide the medical judgments diseases’ are to a large extent discursive phenomena to be made. This move is di¤erent from other cases when plans are made and solutions formulated in described above, where particularities, and often hospital medicine. As they are not firmly anchored in moral particularities, of the patient are made most what for simplicity’s sake may be called ‘concrete relevant for the solution of the problem, either as for- physical or biological realities’, they should be mulated evidence or for the generation of evidence thought of as ‘objects of knowledge’ or ‘boundary ob- jects’. They are epistemic phenomena that may be A risk case is di‰cult to deal with; it is not decisive transformed from biological kinds to moral ones and and often without obvious solutions. Arguments and back, and they are often in a state of transfer. To facts will always support di¤erent alternatives. When quote Hacking (1999), terms such as ‘facts’, ‘reality’, the resident suggests that they ask the patient herself or ‘evidence’ are ‘elevator words’, words that change what she wants, he makes the case into the patient’s meaning. In Victor Turner’s terms (1969), they refer own problem. Through exchange of verbal formula- to ‘liminal’ phenomena. Elevator words do not be- tions, the patient’s problem is transferred from med- long to ‘the world’; they belong to language, but not ical to moral and moved out of the meeting’s agenda.
language alone, they also belong to discourse, to the When the resident starts talking about alternative situation, and to the institutional context. There need forms of medical treatment (line 28), he suggests a not be a necessary equivalence between a ‘fact’ and a comeback for the patient as a medical case. Quickly condition of ‘the world’ or a ‘state of nature’. While the chief physician interrupts and obstructs this at- certain aspects of nature may be unchanged, the tempt. The surgeons complete the chief physician’s meaning and practical use of ‘facts’ or ‘evidence’ move by a‰rmative responses and the chief physician change with situations and participants, with the col- laborative processes, with the ongoing transforma- In this case, the solution is guided by principles tions of frames and perspectives, and with the prac- from evidence-based medicine. The problem is that tical understanding of the problem at hand. In this di¤erent kinds of adequate and accountable evidence process, evidence is essential, but its meaning and use might be used to support di¤erent solutions. Finally, is often grounded in moral and situational considera- the participants made the general principles overrule tions rather than in scripts based on general scientific the specific and particular ones. In this sense, a pro- and evidence-based guidelines or principles of admin- cess of ‘disembodiment’ took place. A moral frame !eeerrreeeiiitttgggeeesssttteeellllllttt      vvvooonnn      |||      NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk      (((NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk))) AAAnnngggeeemmmeeellldddeeettt      |||      111777222.111666.111.222222666 HHHeeerrruuunnnttteeerrrgggeeelllaaadddeeennn      aaammm      |||      111666.000222.111222      111222:::333666 Hutchins, E. (1995). Cognition in the Wild. Cambridge, MA: Kitchener, M. (2000). The ‘bureaucratization’ of professional roles: The case of clinical directors in UK hospitals. Organi- Knorr Cetina, K. (2001). Objectual practice. In The Practice Turn in Contemporary Theory, T. R. Schatzki, K. Knorr Cetina, E. von Savigny (eds.), 175–188. London: Routledge.
Brackets with dots refer to short but noticeable pauses.
Levinson, S. C. (1983). Pragmatics. Cambridge: Cambridge The more dots the longer pause. (No specific time Mead, G. H. (1972 [1938]). The Philosophy of the Act, C. W.
Brackets without dots refer to inaudible utterances.
Morris (ed.). Chicago: University of Chicago Press.
Indicates that words or utterances are drawn together.
Mishler, E. G. (1984). The Discourse of Medicine: Dialectics of Underlining indicates emphasis or stressing of words.
Medical Interviews. Norwood, NJ: Ablex.
Colons are used to indicate that the sound/vowel is Mol, A. (2002). The Body Multiple: Ontology in Medical Prac- tice. Durham: Duke University Press.
Ma˚seide, P. (1983). Analytical aspects of clinical reasoning: A discussion of models for medical problem solving. In TheSocial Organization of Doctor–Patient Communication, S.
Fisher and A. D. Todd (eds.), 241–265. Washington, DC:Center for Applied Linguistics.
— (2003). Medical talk and moral order: Social interaction and Anspach, R. (1988). Notes on the sociology of medical dis- collaborative clinical work. Text 23 (3): 369–403.
course: The language of case presentation. Journal of Roberts, C. and Sarangi, S. (1999). Hybridity in gatekeeping Health and Social Behavior 29: 357–375.
discourse: Issues of practical relevance for the researcher.
Atkinson, P. (1995). Medical Talk and Medical Work. London: In Talk, Work and Institutional Order: Medical, Mediation and Management Settings, S. Sarangi and C. Roberts (eds.), Berg, M. (1992). The construction of medical disposals: Med- 473–503. Berlin/New York: Mouton de Gruyter.
ical sociology and medical problem solving in clinical prac- Rostad, H., Naalsund, A., Norstein, J., Jacobsen, R., and tice. Sociology of Health and Illness 14: 151–180.
Aaløkken, T. M. (2002). Er behandling av lungekreft i — (1997). Rationalizing Medical Work. Decision-Support and Norge god nok? [Is the treatment of lung cancer in Norway Medical Practices. Cambridge, MA: MIT Press.
adequate?]. Tidsskrift for den Norske Lœgeforening 122: Cicourel, A. V. (1987). The interpenetration of communicative contexts: Examples from medical encounters. Social Psy- Sarangi, S. and Roberts, C. (1999). Discursive hybridity in chology Quarterly 50 (2): 217–226.
medical work. In Talk, Work and Institutional Order, S.
— (1990). The integration of distributed knowledge in collabo- Sarangi and C. Roberts (eds.), 61–74. Berlin/New York: rative medical diagnosis. In Intellectual Teamwork. Social and Technological Foundations of Cooperative Work, J. Ga- Star, S. L. and Griesemer, J. (1986). Institutional ecology, legher, R. E. Kraut, and C. Egido (eds.), 221–242. Hill- ‘translations’ and boundary objects: Amateurs and profes- sionals in Berkeley’s Museum of Vertebrate Zoology, 1907– Degeling, P., Maxwell, S., and Iedema, R. (2004). Making clin- 1939. Social Studies of Science 19: 387–420.
ical governance work. British Medical Journal 329 (18 Sep- Timmermans, S. and Berg, M. (2003). The Gold Standard. The Challenge of Evidence-Based Medicine and Standardization Field, M. J. and Lohr, K. N. (eds.) (1990). Clinical Practice in Health Care. Philadelphia: Temple University Press.
Guidelines: Directions for a New Program. Washington, Turner, V. (1969). The Ritual Process. Structure and Anti- Go¤man, E. (1956). The nature of deference and demeanor.
Wertsch, J. V. (1991). Voices of the Mind. London: Harvester American Anthropologist 58: 473–502.
— (1971). Relations in Public. Micro Studies of the Public Order.
Wiener, C. L. (2000). The Elusive Quest: Accountability in Hos- pitals. New York: Aldine de Gruyter.
— (1981). Footing. In Forms of Talk, E. Go¤man (ed.), 124– Wilson, R. A. (2004). Boundaries of the Mind. The Individual 159. Philadelphia: University of Pennsylvania Press.
in the Fragile Sciences. Cambridge: Cambridge University Goodwin, C. (2002). Professional vision. In Qualitative Re- search Methods, D. Weinberg (ed.), 281–312. Oxford:Blackwell.
Per Ma˚seide is Professor of Sociology at the School of Social Gray, A. and Harrison, S. (eds.) (2004). Governing Medicine: Sciences, Bodø Regional University, Norway. He is also Di- Theory and Practice. Maidenhead: Open University Press.
rector of the Center of Disability Research, Bodø Regional Hacking, I. (1999). The Social Construction of What? Cam- University. He teaches sociology of health and illness, social in- bridge, MA: Harvard University Press.
teraction, social theory, and qualitative methodology. He has Hall, C., Sarangi, S., and Slembrouck, S. (1999). The legitima- extensive experience from fieldwork in diverse healthcare insti- tion of the client and the profession: Identities and roles in tutions. His research interests lie in doctor–patient interaction, social work discourse. In Talk, Work and Institutional Or- multiprofessional collaboration, the social organization of med- der: Discourse in Medical, Mediation and Management Set- ical work, distributed cognition, and pragmatics. Address for tings, S. Sarangi and C. Roberts (eds.), 293–322. Berlin/ correspondence: School of Social Sciences, Bodø Regional Uni- versity, N-8049 Bodø, Norway. E-mail: [email protected] !eeerrreeeiiitttgggeeesssttteeellllllttt      vvvooonnn      |||      NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk      (((NNNooorrrdddlllaaannndddssssssyyykkkeeehhhuuussseeettt      SSSooommmaaatttiiikkkkkk))) AAAnnngggeeemmmeeellldddeeettt      |||      111777222.111666.111.222222666 HHHeeerrruuunnnttteeerrrgggeeelllaaadddeeennn      aaammm      |||      111666.000222.111222      111222:::333666

Source: http://www.mffn.no/pdf/maseide2006ComMedTheDeepPlayofMedicine.pdf