Accomplishing 'the case' in paediatrics and child ... - Wiley Online Library

Sociology of Health & Illness Vol. 24 No. 4 2002 ISSN 0141–9889, pp. 409–435 Accomplishing Susan White ‘theLtd case’ in paediatrics and2001 child health © Sociology SHIL 24 Original Blackwell Oxford, 0141–9889 2002 0 1 4 00 Blackwell UK Article Science ofPublishers Health & Illness Ltd/Editorial Board

Accomplishing ‘the case’ in paediatrics and child health: medicine and morality in inter-professional talk Susan White Department of Applied Social Science, University of Manchester

Abstract

This paper presents data from a recently completed ESRC funded ethnography of social relations and case formulation in an integrated child health service, comprising paediatric inpatient and outpatient, child and adolescent mental health and child development services. Children present to the services with symptoms or troubles for which there are often competing biological, neurological, genetic and/or psychosocial models of causation. As a consequence, clinicians’ talk is oriented to deciding between three main potential types of case formulation – medical, psychosocial and not just medical. These three formulations are not static ideal-types. They are highly contestable and require complex practical and rhetorical work, through which facts and evidence are selectively invoked and different parties to the case are granted attributes which construct and reconstruct past events to render ambiguous symptoms or events understandable. In particular, moral judgements and complex characterizations about the child’s parents, or significant others, often form an indispensable warrant for these formulations. By analysing professional narratives about cases, this paper develops previous ethnographic work on the classification in medical work of children and adults as good or bad, appropriate or inappropriate, culpable or blameless, and renders visible a repertoire of moral formulations about childhood and child care. In particular, judgements about the adequacy of parental love are central to clinical reasoning.

Keywords: paediatrics, ethnography, professional talk, case formulation, moral judgement

© Blackwell Publishers Ltd/Editorial Board 2002. Published by Blackwell Publishers, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main Street, Malden MA 02148, USA

410

Susan White

Introduction Over the last decade, as a result of rising managerialism and consumerism in public services, an increasing need to ration services and a series of highprofile media scandals, professional judgement has received unprecedented levels of political and academic scrutiny. The dominant policy response has been the promotion of evidence-based practice (EBP), which promises rational foundations for clinical decisions based on a secure, external knowledge base uncontaminated by the contingencies and emotions of practice. EBP, however, provides a very partial representation of the processes of clinical reasoning. In particular, a preoccupation with ‘what works?’ has led to a neglect of the processes and practices by which professionals negotiate and carry out problem formulation, in what are often uncertain and ambiguous circumstances. This paper examines case formulations in a child health setting. It is concerned with how professionals order clusters of symptoms and troubles into a recognisable case. The paper argues such work can inform current debates about clinical judgement and create more sophisticated understandings of the moral complexities of child health work. The data presented in this paper are part of a corpus generated from an ESRC funded, ethnographic study of an integrated child health service situated in a district general hospital in the North of England. The service comprises paediatric inpatient and outpatient, child and adolescent mental health (CAMHS), child development (CDS) and social work services. Together, the services provide general secondary care to a socio-economically diverse community, with tertiary specialist services provided at regional centres. A single case study design was used, with the author as sole researcher. Methods included non-participant observation of clinics, ward rounds and staff/team meetings, audio-recording of interprofessional talk in meetings and other less formal settings, such as before and after clinics, the tracking of a number of individual cases through the services and a documentary analysis of medical notes. The fieldwork took place between July 1999 and October 2000. There is a rich, established ethnographic literature on medical practice in child health settings (inter alia, Strong 1979, Silverman 1987). However, as Anspach notes: Although much has been written concerning how doctors talk to patients, very little has been written about how doctors talk about patients . . . This analytic focus on the medical interview occurs even though the way in which physicians talk about patients is a potentially valuable source of information about medical culture. Rarely do doctors reveal their assumptions about patients when they are talking to them (1988: 358). Where attention has been paid to interprofessional talk, such as in Jeffrey’s (1979) study of classification of ‘good patients, bad patients and rubbish’ in an A&E department, this has rarely been grounded in the analysis of © Blackwell Publishers Ltd/Editorial Board 2002

Accomplishing ‘the case’ in paediatrics and child health

411

talk in context. With some notable exceptions (for example, Anspach 1988, Atkinson 1995), there has been a tendency for studies to present decontextualised extracts from fieldnotes, which are often used ironically to contrast what ‘really’ happens in the service with the ethics and standards upon which professional activity purports to be based. Clearly these studies have considerable utility, but they miss the opportunity to examine in detail the socially-shared discursive resources used by professionals when they talk together. In the analysis below this paper shows how the situated accounts and the stories told and retold by paediatricians, psychiatrists, psychologists, nurses and social workers in the service have certain expectable features and do particular work. They perform and reproduce aspects of occupational and service identities and they accomplish particular classifications of cases. They also do important moral work, and in this sense this study confirms and develops previous work on the classification of patients as ‘good’, ‘bad’, or indeed as unremarkable and ordinary (e.g. Jeffrey 1979, Dingwall and Murray 1983). Before considering these arguments in more detail, it is important to explain the often complex and contested nature of case formulation in paediatrics.

The problematics of case formulation in child health As in other medical settings, the ordinary work of the services in the study is oriented towards establishing relations of cause and effect. When members account for their work, they do so in terms of identifying what (if anything) is wrong with the children (and/or families) referred to the service, deciding what should be done about it, or getting on with whatever it is that needs to be done. In child health settings, however, the attribution of causation can be particularly complex. Clearly, many children present to services with symptoms which may be unproblematically categorised and treated according to the established nosologies of bio-medicine, but for many others this is not the case. For example, children might present with a physical complaint for which there may be a biological, neurological, genetic and/or psychosocial aetiology. The obvious examples are enuresis, encopresis, constipation, various forms of developmental delay, communication or behavioural disorders and psychological distress. In accomplishing diagnosis and establishing causation in such cases, the boundary between problems with a biological, and those with a psychological or psychosocial aetiology, is particularly important for members, and is underscored in professional literature (e.g. Woodward et al. 1998, Garralda 1996). The diagnostic categories themselves frequently reflect the same preoccupation (e.g. ‘failure to thrive’ is routinely subdivided into organic (intrinsic) and non-organic (psycho-social) varieties). The decision about whether a problem is part of the child’s biological make-up, or a product of their environment, clearly has a direct bearing on © Blackwell Publishers Ltd/Editorial Board 2002

412

Susan White

the management of a case. For example, the suspicion that poor weight gain is not the result of genetics, or a metabolic disorder, but is an indication that the child is not being fed or is emotionally deprived, may precipitate referral to psychologists, child psychiatrists and social workers, rather than admission to a paediatric bed. The terrain is further complicated by the classification of a range of psychological or emotional sequelae to physical illnesses (see e.g. Garralda 1994), which can create problems in the attribution of cause and effect. For example, are a child’s frequent hospital admissions the result of intrinsic diabetic instability, a consequence of the child’s emotional adjustment, or of poor parental supervision, or do all three apply? These are important empirical questions to which members are routinely oriented in their daily practices. Whilst clinicians and professionals in the services are concerned with doing this, this paper focuses on how they do it. In particular, it is concerned with how professionals render their formulations recognisable and accountable to colleagues. So, how are categories, clinical symptoms, histories and normative judgements invoked and assembled to do the descriptions upon which the work of the clinic depends?

Telling the case There are many contexts and situations in which professionals in the services talk to each other about cases. Such talk takes place in formal meetings where detailed formulations of cases are delivered, often in long narrative turns. It takes place in abridged form in regular ‘updates’ before clinics, or during ward rounds or nursing ‘handovers’. It takes place over the telephone and over coffee. Often the same case may be talked about over and over again in many different forums. Wherever it takes place, talk does particular ordering work. It turns symptoms and events into cases which are recognisable to members and can be processed using one or more of a range of potential disposals. At the beginning of the fieldwork for this study, a good deal of time was spent shadowing various professionals and clinicians as they went about their business. During this time I witnessed many chance encounters in corridors where particularly unusual, distressing, amusing or urgent anecdotes might be exchanged. With the exception of impromptu referrals to another professional – ‘could you see this child I saw this morning, I think she needs . . .’ – for the most part these encounters were used to give brief updates about shared cases. The formulation of the case was rarely explicitly stated, but nevertheless was apparent to the hearer, and usually to me. I have many years of experience as a professional and a sociological researcher in child health settings, so my ability to read the case is not particularly remarkable. Yet, it is also intriguing. How is it that, through the telling, this case becomes evident to the hearer as this or that kind of case and what is the © Blackwell Publishers Ltd/Editorial Board 2002


413

available repertoire of formulations? By this or that type of case, I do not refer to the nosological category to which the child or young person has been assigned (e.g. whether they have diabetes); this is usually explicitly stated, or a range of hypotheses are presented. Rather, I want to attend to ways in which members work up versions of causation through their talk in what is often highly contestable terrain. It has already been noted that clinicians in child health settings routinely make distinctions between medical and (psycho)social cases. In this respect, paediatrics is similar to geriatrics where Latimer (2000) has shown how patients are constituted through professional talk as ‘medical’ or ‘social’. Whereas in geriatrics the ‘bringing off’ of a social categorisation is likely to result in patients being seen as inappropriate, or as ‘bed-blockers’, in child health services, particularly multi-disciplinary ones, the classification ‘psychosocial’ still carries service entitlements, but of a different nature to ‘medical’ cases. So, how do clinicians ‘tell’ cases so that they can be recognisable as medical or social or a mixture of the two. When looked at in transcript, it is clear that very different linguistic devices are used to signal particular readings of the case. This does not mean that clinicians’ talk straightforwardly describes different kinds of cases, rather the case is, at least in part, constituted through the telling, and other possible readings are closed down. In the telling of cases, clinicians are not only using knowledge and ‘know-how’, they are very clearly making knowledge (Taylor and White 2000). That is, the same case may be told in many different ways (and often is when a new professional becomes involved and sees things differently). Neither does the propensity to see a case as either medical or psychosocial divide neatly by occupational group. For example, during my time in the service, a difference of opinion occurred between a child psychiatrist and a paediatrician over the diagnosis and management of a child referred because he was having ‘funny turns’ The child psychiatrist hypothesised that this was epilepsy and ordered an EEG (a medical reading). The paediatrician, on the other hand, whom we might assume to be less inclined to see potentially medical problems in relational terms, argued that the child was hyperventilating in response to problems in his relationship with his mother (a psychosocial reading). The EEG did not show any abnormalities, which was again variously invoked as a warrant for very different case formulations. The paediatrician used it as confirmation of the psychosocial formulation, whereas the psychiatrist invoked the fallibility of the EEG as a diagnostic tool in the identification of seizures, arguing for more tests and observations. There is thus a strong case for examining the argumentative strategies of professionals in working up versions of cases. In the analysis that follows, as part of a more eclectic approach to examining the rhetorical features of the talk, I have made some use of Sacks’ work on membership categorisation analysis (e.g. 1972, 1992, or for a more accessible summary, Silverman 1998). Sacks argues that, along with certain © Blackwell Publishers Ltd/Editorial Board 2002

414

Susan White

sequential features of an utterance, the use of social categories such as ‘mother’ or ‘child’ can operate as a means of referencing deviance or normality in talk1. That is to say, membership categories are associated with certain likely activities (category-bound activities – CBAs), and by describing behaviour which either conforms or fails to conform to these expectations we may establish positive or negative moral identities. So, if ‘mother’ is associated with nurturance and care, a description of behaviours departing from these expectations will serve to reference deviance. Membership categories are in turn associated with membership categorisation devices (MCDs), so ‘mother’ and ‘child’ are categories of the membership categorisation device ‘family’. Some categories tend to occur in pairs signalling mutual rights, duties or obligations, of which mother-child or doctor-patient are examples. These pairings are termed standardised relational pairs (SRPs). Often membership categories are explicitly stated (e.g. this is a patient), but sometimes they are referenced by association to some activity or attribute associated with a category (e.g. I have observed these symptoms), that is, categorisation can be done by invoking the predicates of a category, as Housley notes: [I]f the topic was the moral evaluation of an individual, one might state ‘I don’t like him/her, s/he is a bad person’ . . . or one might refer to the same person as being ‘lazy’, recount stories of their behaviour on previous occasions (they drank too much at a party), or suggest that their outward appearance conceals some dark motives (e.g. their eyes are too close together, etc.). In both cases these strategies can be used to do various types of work within occasioned settings. . . . (2000: 104–5) Clinicians routinely use both categories and predicates of categories in their work. However, membership categorisation analysis (MCA) has been used here as part of a more general orientation to the rhetorical features of the talk, and particularly to the kinds of warrants clinicians invoke for their versions of cases.

Telling a medical case During an observation of a paediatric clinic, towards the end of the fieldwork, one of the paediatricians, stated ‘My cases are all really boring, not like Jane and David’s [other paediatricians], mine are purely medical’. Jane and David are both associated in different ways with ‘messy’ social cases. The idea that ‘purely medical’ cases are in some way of less interest, is both untrue and in itself interesting. Telling a medical case and rendering it intelligible even to an audience with a shared interpretive repertoire is a complex business. Moreover, medical cases are often far from straightforward (although, of course, many patients are processed, quickly, routinely and unremarkably). © Blackwell Publishers Ltd/Editorial Board 2002


415

The following extract is taken from a briefing session before an outpatient clinic. The speakers are a consultant paediatrician and a registrar. Extract 1 Cons: Matthew Long Reg: [Mmm] Cons: He’s been in with asthma but that’s not why he comes to see us (.4)2. The main reason is some hydronephrosis – I think I’ve got the last ( ) seems to have a problem attending [reading] Repeat ultrasound October 99, it’s still hydronephrosis, further up (.5) urinary tract infection, yeah, for definite. Reg: That’s back in (.) April Cons: Back in April. DMSA [dimercaptosuccinic acid – test to assess scarring and relative function of kidney] clear. Mild right sided hydronephrosis with prominent renal pelvis mainly extra renal, no scarring and (.) no (.) reflux. (.7) So, I suppose I thought that the best way was to do repeat the ultrasound if the kidney was blowing up and we needed a ( ), so that’s fine. It’s difficult sometimes with these mild hydronephrosis. You never know whether it’s the beginning of – Reg: Or whether it’s borderlineCons: Or whether it’s just the [way] they’re madeReg: [Yeah] yeah In this account, the patient is identified first by his name ‘Matthew Long’, but thereafter the account is ‘depersonalized’ (Anspach 1988) with reference to the diagnostic categories, asthma and hydronephrosis. This talk does not have the pedagogic flavour of conversation between junior and senior clinicians, with the junior encouraged to practise case presentation for evaluation by the experienced colleague (Atkinson 1999). Rather, the two clinicians are telling the case together. The account has a number of markers of certainty ‘urinary tract infection, yeah, for definite’, but these are juxtaposed with markers of uncertainty, warranted principally by clinical experience ‘It’s difficult sometimes with these mild hydronephrosis’. References to the limits and fallibilities of tests’ technology are typical of this kind of telling. References to character(istics) in this account are confined to the interior of Matthew’s body. Although Matthew could have been categorised as a child, he remains in the membership category ‘patient’ which mean that other categories in the device ‘family’ are not immediately relevant. It is typical of ‘medical’ cases, that there is very little character work in the accounts and where parents are invoked it is generally through ‘reported speech’, e.g. ‘mother says she’s managing well without the oxygen’; or to signal moral worth, e.g. ‘mum’s lovely with her’. So, medical cases are rendered recognisable as such by the particular ways in which they are told. © Blackwell Publishers Ltd/Editorial Board 2002

416

Susan White

These tellings induct novices and affirm, transmit and legitimate medical knowledge, but more mundanely they get diagnosis done.

Telling a psychosocial case Here, the speaker is a clinical psychologist in the Child and Adolescent Mental Health Service (CAHMS) weekly professionals’ consultation group. Extract 2 Clin psych:

Michael Eaton is a five-year-old boy who was referred for behaviour problems and when they came, Mrs Eaton has been very depressed for five years, she’s had occasional admission to [psychiatric hospital], she’s had cognitive therapy at [psychiatric hospital] she’s had psychotherapy and counselling there as well that they’d organised. She’s been on medication. Various different types and as I’ve said she’s had inpatient admissions. Both Mr and Mrs Eaton say that Michael actually probably behaves much like any other five-year-old and that he doesn’t have any problems, school aren’t worried they don’t think he’s got any problems but Mrs Eaton can’t stand him and that’s the problem. She’s very depressed, she can’t bear him, she gets absolutely no pleasure out of his company whatsoever, ehm she you know to the point where she’s actually got a two-year-old as well. She’s completely bonded with the two-year-old and feels very warm towards him and deals with his misbehaviour perfectly appropriately but can’t-. I mean, when you see the two of them in a room when her eyes are on Michael she’s all kind of gritting her teeth and by him and yet when she looks at Bradley, the younger one, she kind of softens and smiling and indulgent and ehm and she’s broken hearted about this, but she can’t stand her own son. Ehm, we talked a lot about, I mean she’s so depressed that, you know, her sort of negativity colours everything.

This is typical of the narrative practices in psychosocial cases. The emphasis is on the child’s significant others, and there are reports of particular ‘unhappy incidents’ (Pomerantz 1978) which, together with assigning the mother to the membership of the category ‘psychiatric patient’, accomplish her responsibility for the problems. The standardised relational pair, motherson is used to signal departure from category-bound expectations and obligations (mothers love sons) and from category-bound rights (sons are loved by mothers). In psychosocial cases, this pairing often takes the place of the SRP doctor-patient which situates the patient as worthy and in need of © Blackwell Publishers Ltd/Editorial Board 2002


417

care and the doctor as obliged to offer help for medical troubles. ‘Extreme case formulations’ (Pomerantz 1986), which invoke the maximal or minimal attributes of a person or event, are powerful devices for referencing blame (or creditworthiness) in talk. These devices are common in psychosocial tellings (e.g. ‘she can’t bear him, she gets absolutely no pleasure out of his company whatsoever, irritated beyond belief, she’s broken hearted about this, but she can’t stand her own son, she’s completely bonded with the twoyear-old’). These further facilitate the characterisation of the parent both as failing and as deviant. The account is also designed to convey intractability ‘Mrs Eaton has been very depressed for five years, she’s had occasional admission to [psychiatric hospital], she’s had cognitive therapy at [psychiatric hospital], she’s had psychotherapy and counselling there as well that they’d organised. She’s been on medication’. This is important in psychosocial cases, for the amelioration of which no ready technologies exist, but about which professionals are supposed to make a difference. So, invoking the list of ‘tried and failed’ here functions as a kind of prospective self-exoneration and ‘expectation management’ device for the clinician who is speaking. The case above was already packaged (by the referrer) on referral to the service as a ‘psychosocial’ case. This is not always so. Problems referred as ‘medical’ issues may, over time, become redefined (or socialised) into a psychosocial reading. In these cases, the paediatrician has powerful definitional privilege to adjudicate on whether it is the inside, or outside, of the child’s body that is causing the trouble. The following extract, like the one above, begins with a characterisation of the mother, achieved through the invocation of the membership category ‘psychiatric patient’. The speakers are a consultant paediatrician and a registrar. Extract 3 Cons: Mark Smith – you’ve not had the pleasure, of this mother. Mother is under our psychiatrists. She is a (2.0) oh (2.0) factitious illness gives the wrong impression. She’s got a [neurotic] state really, somatisation Reg: [Right] right. Cons: [Somatisation], really severe somatisation disorder Reg: [right] yeah Cons: You, you may have met her [ . . . as soon as you meet her, she’ll go on]Reg: [I think I probably-. What’s he got?] Cons: He’s constipated, severely constipatedReg: Yes, it’s all, yes Cons: She looks ill and as soon as you meet her she looks ill and she’ll come out with all of her complaints. He has severe constipation actually required a ( ) when they first brought him in to extract the masses of faeces, but recently he’s relapsed and the problem © Blackwell Publishers Ltd/Editorial Board 2002

418

Susan White

seemed to be that mum had relapsed as well so everything went (.) down and he had to come in for an enemaDespite the presence of a physical complaint, constipation, this case is marked as psychosocial by its telling. The character work in the telling is heard by experienced clinicians as part of the diagnosis and clear attributions of cause and effect are made: ‘but recently he’s relapsed and the problem seemed to be that mum had relapsed as well so everything went (.) down’. These extracts demonstrate the very different ways in which medical and psychosocial cases are told, but the most complex rhetorical work takes place in relation to children who have an identified and named ‘medical’ problem, which is generally agreed to exist independently of any issues about parenting, but this medical problem is seen as being exacerbated by parenting practices (e.g. children with unstable diabetes whose parents are suspected of mismanaging diet). Members do not have a discrete name for these kinds of formulation. They are identified by medical diagnosis with accompanying narratives about parents/carers, or references to ‘possible child protection issues’. The formulations can however be distinguished by aspects of their telling; for the purposes of differentiation, I have called them ‘not just medical’ cases. This is as close as I can get to a members’ term for what is a recognisable members’ category of case formulation. These formulations involve particularly complex story-telling, since the presence of an ‘intrinsic’ disorder requires that any psychosocial component be worked up in the talk. Narratives about these cases have the flavour of detective stories with anomalous physical findings, such as failure to gain weight, set alongside characterisations of carers. Cases may begin as ‘medical’ and evolve gradually to a ‘not just medical’, or psychosocial formulation through formal and informal case-talk between professionals. Once they have shifted in this way, they rarely return to a purely medical reading, since the relevances for storytelling and observation are extended to the child’s relationships and social circumstances, which once exposed are almost always found wanting.

The natural and the social: ‘not just medical’ cases ‘Not just medical’ cases are common in paediatrics and may subsequently be referred to CAMHS or social work services. They are the most analytically interesting since, once clinicians have agreed that there is something medically wrong with the child, this skews the ages and stages of likely physical, emotional and social development as calibrated by developmental psychology and surveillance medicine (White 1998). Thus, the boundaries © Blackwell Publishers Ltd/Editorial Board 2002


419

between normality and abnormality become more fluid and contestable. For example, if a child has cerebral palsy and has difficulty swallowing and chewing, it is likely that her weight gain will be slow. This exists as an available explanation for poor weight gain and, in the absence of dramatic weight loss, clinicians need not necessarily question further. Further investigation must thus be triggered by something, and this is a highly contestable and often practically onerous process, often relying substantially on moral judgement and techniques of persuasion. As well as requiring artful and persuasive telling, these formulations may involve practical detective work, rigorous questioning of ‘witnesses’, cross checking of parental accounts and an almost forensic attention to detail. The propensities to look inwards to the child’s body, or outwards to their social circumstances are not evenly distributed among clinicians. Those who favour particular explanations, or who are especially likely to undertake the necessary detective work to shift a case into a ‘not just medical’ reading, are well known in the organisation and the network beyond. This can affect the type of case they are asked to see. The following more extended example shows the rhetorical and practical work involved in bringing off a ‘not just medical’ reading of a case. It is taken from a professionals’ meeting (parents absent), convened to consider the need to invoke child protection procedures in respect of a child, Sarah, who has a rare syndrome (omitted to protect anonymity), associated with multiple abnormalities. Sarah has had frequent admissions to hospital for a variety of reasons and the consultant has noticed over time that, on her return home, she loses the weight she has gained in hospital. The transcript of the meeting is some 19 pages long. These extracts have been chosen because they illustrate the major strategies of argumentation and the various warrants used for a potentially contestable case formulation. The speaker is the consultant paediatrician responsible for Sarah’s care, who begins a long turn by outlining some of the child’s medical problems, specifically directing much of the talk to Jenny, the social work team leader, reflecting the purpose of the meeting (whether or not to use formal child protection procedures). Whilst there is only one speaker for much of this story, the talk is clearly oriented to the possibility of reply, that is to the audience and to the contingencies of the situation as social and consequential. Extract 4a Cons: . . . Right, OK, if I just take you through my report, Sarah’s 20 months old now and I’ve only known her since July. She has [ ] Syndrome and if you look at page 3 of your pages, [ ] Syndrome is the name given to a child with a collection of differences. . . . Right so as part of Sarah’s [ ] Syndrome she has a heart problem that has actually required surgery but is now off © Blackwell Publishers Ltd/Editorial Board 2002

420

Susan White

medication for that. She had a problem with her stomach and needed surgery on that. She’s got a mild kidney problem and requires antibiotics for that. She’s got some deafness and she’s got a problem with one of her eyes and she’s relatively short, so there are a collection of medical issues. . . . When I took her over, she was under nine different consultants for various bits of her care. When I do a full report I’ll expand on all of those, Jenny [social work team leader], but I think it would take a long time to go into them and I think it’s better to stick to the salient features of it really. I first met her because she came in with problems with diabetes, that is not related to the [ ] Syndrome and she’s also got asthma. In this first extract, despite reference to her age, Sarah is located in the category patient and the account is depersonalised with a list of her clinical features. It reads like a medical case, but the explicit direction of the talk to Jenny, the social work team leader, alludes to what is to come. The case must at some point become constructed as the proper business of the social services department. The consultant continues by listing the care Sarah needs as a result of these complex medical problems. Extract 4b . . . She requires a certain amount of medical intervention. She can’t chew and swallow normally so her main nutrition is through a gastrostomy but she needs feed to be offered you know smooth, like Rusks, stewed fruit to be offered at meal times, alongside finger food which sometimes she takes and pump feeds via her gastrostomy for her main meals and for snacks, and she’s fed through her gastrotomy of a night. So she’s on a fairly hefty regime. She has inhalers for her asthma and night-time antibiotics to prevent infection and she requires hearing aids for periods of time in the day. . . . Here the consultant references Sarah as in need of particular sorts of care. In the next turn, he invokes the technologies of surveillance medicine, such as the centile charts on which children’s growth and weight are plotted. His account uses the language of scientific/forensic neutrality and objectivity and is linguistically coded as fact/certainty. Extract 4c Now if you just look at this overall growth chart first, which is page 4 of your charts, you can see that she was born a couple of months early with her weight being on lowest centile, which is the first centile, and that she fell away from it. Now initially she had bowel surgery so we could understand that her weight gain wouldn’t be that good, she had heart surgery half way through the first year of life but there has been a © Blackwell Publishers Ltd/Editorial Board 2002


421

marked deterioration. Well a deterioration at six months, and a marked deterioration at about a year, then effectively after the three months or even four months after the first year of life she really didn’t put on significant weight. So, the bit between the two arrows that looks a muddle on the chart. If you turn over to the next page, I’ve blown the centile up so that the line you can see going through the circles is the third centile. That’s the one across the top, the weight’s on the left axis and the dates across the bottom. The bits highlighted in pink refer to the hospital admissions. The first admission had the most dramatic weight gain, but she was ill at that time with diabetes so a considerable amount of that would have been because of that. So say one kilo of that or the first little rise would have been the fluid, but there does seem to be a pattern of rapid weight gain in hospital and a tendency to plateau or lose weight on going home. The distance from the third centile, we’d got her on the 6th August, we’d got her really approaching the third centile as near as she’d been to the third centile for a long time. If you look back she was back to where she was at three months of age and we’ve lost the distance from the third centile since. So . . . these weight gains are quite . . . I mean this is a 3lb weight gain . . . I mean she was 8 kilos and she’s gone up to 8.8 which is getting on for 2 pounds in weight and she’s been in hospital a week. You must remember that some of the difficulties with the weight will depend on when she’s being weighed in relation to her feeds because she does have bolus of feed, so if it’s after a bolus she will weigh more. Referenced here is the forensic work of the consultant, both in ensuring the child is weighed at the same time in relation to her feeds and in annotating and highlighting the various artefacts so that their salient features are recognisable to a mixed professional audience. Throughout this account, the consultant presents potential alternative readings of the periods of weight gain and weight loss, demonstrating that all things have been properly considered. This is followed by a summary of the formulation so far, invoking the legal concepts of ‘avoidable impairment’ and ‘significant harm’ imported directly from the Children Act 1989. This achieves the construction of the case as multi-agency business, and relies on the paediatrician’s category entitlement to adjudicate on ‘significant harm’ in sick children. Extract 4d Then I basically go on to say in view of these experiences with Sarah’s weight our concern that Sarah could be doing significantly better and that her growth and development are be avoidably impaired i.e. she’s suffering significant harm. A significant, prolonged improvement in weight gain would also lead to an improvement in height and head circumference which would enable Sarah to consolidate her development. © Blackwell Publishers Ltd/Editorial Board 2002

422

Susan White

This formulation is supported in the next extract by characterisations of Sarah’s mother. This is a potentially contentious issue, since caring for a child with this range of complex medical needs may reasonably be expected to provoke professional sympathy. Here, the consultant invokes his status as eye witness ‘I have observed the following . . .’ and the account is linguistically coded as fact not opinion – ‘the mother concentrates on the medicalisation of all Sarah’s care . . .’. To a lay reader this may seem a rather strange statement, since much of Sarah’s care is self-evidently medical. However, this is the first stage of a process of argumentation designed to persuade the audience that this is a case of, or has features associated with, Munchausen Syndrome by Proxy (e.g. Meadow 1980, 1985), which is the production by a parent or carer of factitious illness in a child, or exaggeration/exacerbation of an existing condition. Use of ironicisation such as ‘she went on about . . .’ and accounts of the mother’s questions about the consultant’s expertise serve to signal that this mother is a ‘troublesome patient by proxy’. Extract 4e I have also observed the following. The mother concentrates on the medicalisation of all Sarah’s care. When she first presented, she went on about all the nine consultants she was under, the number of medical problems that she had, that she was very difficult to manage that no one would be able to manage her medically and questioned us as to how many children with [ ] Syndrome I’ve managed and that they were totally unpredictable. She appears to have a very negative outlook for Sarah, she wasn’t going to grow, she wasn’t going to develop, she was going to go to a special school and her husband hadn’t really taken any of this on board. . . . Her mother has also said that if Sarah puts on weight it will put a strain on her heart and she will die of heart strain, I’ve reassured her that that’s not true and if she doesn’t put weight on she won’t do lots of other things. She’s convinced that there is more and more wrong with Sarah and to a certain extent there is something in that. I mean she has developed diabetes and we also think there is another metabolic problem to do with how she handles protein which might make her more prone to sickness episodes, but in reality that doesn’t seem to be a problem. The formulation of Munchausen Syndrome by Proxy also requires the consultant to establish that Sarah’s medical problems are not the cause of the weight problems and this involves a degree of minimisation in response to a question from the social work team leader. Extract 4f T.L.: What about her heart problem, has that been resolved? Cons: She had two holes that she’s had closed and she has a trivial leak in one valve that is probably not of any consequence but it needs © Blackwell Publishers Ltd/Editorial Board 2002


423

monitoring. Her kidney problems should be resolved by the age of five, she’s got asthma which a lot of children have, she’s got diabetes which is in remission now and may stay in remission for a number of years. . . . her main problem is the gastrotomy and the feeding and her co-ordination but then it’s difficult to know how far that could be brought on with perseverance with offering appropriate food. Here, the leak is defined ‘trivial’, the kidney problems ‘resolved’, the diabetes ‘in remission’ the asthma as something ‘a lot of children have’. The main problem is feeding – recognisable to the audience as a normal trouble of childhood with which parents may reasonably be expected to deal. Further character work completes the formulation. Extract 4g Cons: . . . I don’t personally believe that the maternal instinct, or whatever-. To have a child in the house of Sarah’s age, not be able to feed her and not have any feelings of need to feed her. Most parents would not be able to tolerate that. They would be force-feeding the child, they would be beside themselves with worry about her not eating and there’s none of that. She could go through a day and she would have 50 mls which is less than two oz of feed in a whole day and she would not be anxious about her. And if someone is at that level of dysfunction for whatever reason, I mean I don’t know if there are elements of, if you think of the Mary Eminson scale of illness perception you know from neglect to the frank Munchausen, I would see her as scoring fairly high. I’m not sure if she’s in the neglect end, which is where a lot of the failure to thrives are, I think she’s more the excess perception end, making a problem end. She’s sort of creating situations if you like. Here the ‘voice of journal science’ (Atkinson 1995: 143) (the Mary Eminson scale) and category entitlement of the consultant to adjudicate on good enough parenting combine to produce a powerful case. Noteworthy also is the use of contrast structures (Smith 1978), the first half of which set up an expectation of proper behaviour and the second a deviation from it. The quote above – ‘Most parents would not be able to tolerate that . . . and she would not be anxious about her’ – is an example of this. These devices also reference deviation from category bound obligations and expectations of parenthood. There are clearly alternative ways in which this case may have been told. For example, Sarah’s medical needs could have remained the primary formulation and the mother’s reactions constructed as ‘understandable in the circumstances’. The consultant’s particular telling relies on his strategies of argumentation and persuasion, considerable behind-the-scenes forensic © Blackwell Publishers Ltd/Editorial Board 2002

424

Susan White

detective work on the part of the consultant and nurses, an audience with a shared professional understanding of ‘significant harm’ and the category entitlement of the consultant paediatrician to authorise the bracketing of Sarah’s medical problems. The formulation depends not only on centile charts and scales, but on the working up of a convincing characterisation of the mother, and on the availability of the nosological category, Munchausen Syndrome by Proxy. In the meeting, all these warrants are treated with equal evidential weight. The paediatrician’s turns are followed by detailed accounts given by nursing staff of the mother’s behaviour on the ward and of Sarah’s presentation on various admissions. There is some discussion about whether further work is necessary to gather more evidence, or whether the case should be taken into the formal child protection system. The team leader summarises this discussion as follows: Extract 4h T.L.: To use case conferencing positively, which is the way it should be used, you would go in and say this is the purpose of the case conference to bring everyone together, to get consent and plan. If you do it that way, you can use the case conference. If you intend to plan and then say go, we’ll go to conference, it’s usually like a punishment which it’s not because it’s there to provide support . . . But, in a way, the last meeting you had before they went home was a support package, in a way you know. Cons: That’s how I feel, I offered [local support service]. We reinforced the feeding We reinforced the hearing aids and there’s not really any evidence that any of that has been taken to fruition and OK it was a low-key planning meeting, but nevertheless the elements were there. The health visitor was there, the CDC [Child Development Centre] staff were there, the hearing support teacher was there, Mandy [home care nurse] was there, [local support service] were involved. We bent over backwards to try and listen and sort things out for her. T.L.: The other thing that the case conference will serve to ensure is that social work should undertake an assessment of the family situation which is what you need as well. Cons: Well, I think that’s what we need mostly, because. . . . Sarah’s mother, in my view, has major problems and I’m not really sure what the nature of these problems is, and how treatable those problems are and how much this is personality that won’t change and how much this is a protective mechanism from the desperate situation that she thinks she’s in. Here, the team leader and consultant are arguing the case together. The team leader’s first utterance draws on her knowledge that, in order to justify taking a case to conference, it must be demonstrated that ‘family support’ © Blackwell Publishers Ltd/Editorial Board 2002


425

has failed, or that the level of risk is too great to attempt such an activity. However, it is not delivered as a challenge to the consultant’s reading, but as a ready-made rationale for taking the child protection route. The list of ‘tried and failed’ interventions given in response by the consultant reinforces the argument in favour of a child protection intervention. The allusion to the mitigatory potential of the mother’s psychological problems is delivered ironically – ‘and how much this is a protective mechanism from the desperate situation that she thinks she’s in’. The end of the meeting is devoted to planning the encounter with Sarah’s parents and how the team leader may best be involved in this. Extract 4i T.L.: I think that either myself or Kay [job-share team leader] can come in towards the end of the interview and meet the parents. I don’t mind which it is, to do the whole interview, or half of it, or come in at the end, whatever you decide. Cons: I’ll see when they can come in. I would perhaps like to start it off myself and say, ‘look I’ve been assessing the situation, I know we’ve found these bits, but this is the reality and clearly things are happening at home that aren’t doing her as well as those that happen in hospital and that has to change’. Then, if you could be around to take on the planning bit and the idea of a support package, but through a case conference. I’ll bring that up as well. This was precisely the way the case was managed after the meeting, with the concerns presented to the parents and a case conference held shortly afterwards. The potential for backstage discussions to provide opportunities for this kind of rehearsal collapses the common distinction drawn between frontstage (the doctor/patient encounter) and backstage (interprofessional talk). The backstage and frontstage are not insulated, mutually exclusive zones of activity. Backstage most definitely affects frontstage activity and vice versa (Sarangi and Roberts 1999). The backstage provides a space where professionals can shore up and contest their formulations of cases and often rehearse their next encounters with patients and their families.

Contesting ‘not just medical’ readings The analysis above shows how Sarah’s consultant produced a sophisticated and persuasive case formulation, in what is potentially contestable territory, which in turn depended on his commitment and attention to detail in weighing, measuring and recording. However, not all attempts to construct cases as ‘not just medical’ are so successful. The following extract is taken from a meeting about a family with three children aged five, four and eight months, the eldest of whom (‘Paul’ in the © Blackwell Publishers Ltd/Editorial Board 2002

426

Susan White

transcript) has severe physical problems and learning disabilities for which no diagnosis has been found. The meeting has been convened by the social work team leader in response to a number of concerns raised by the social worker and other professionals about all three children who have poor weight gain and developmental problems. Present are the team leader, the family resource worker (FRW) from social services, who has been providing practical help, the health visitor (HV) and the head teachers of the eldest child’s school for children with disabilities and from the middle child’s school. The dietician and speech therapist have sent their apologies but have provided reports for the meeting. The consultant paediatrician is invited but is late. During the meeting the professionals, in the absence of the paediatrician, have together been building an argument that the children are failing to develop appropriately due to less than adequate parenting. All professionals have given accounts of the children being smelly, and of parents failing to follow advice. The paediatrician eventually joins the meeting some 45 minutes late. At this point, for the benefit of the paediatrician, the team leader summarises the case so far. Extract 5a T.L.: . . . Now we accept that Paul has special needs ehm and his attendance at [special school] has improved slightly, I would say it’s not sufficient well it’s not as much as we would’ve liked and we still think that his sort of ehm development may be being impaired by his parents’ lack of doing the things that they should for him . . . On the 4th of January it was 12.6 kg on 1 February it was 13.2, so there was a slight increase between January and February but he was having Pediasure to supplement his diet. We don’t think the children are being fed adequately and we don’t think they’re being stimulated adequately either but the other issue the thing that occurred at the planning meeting was they actually made a statement that they didn’t if they were concerned about feeding Paul too much because it would make him too heavy to carry upstairs ehm so obviously that was sort of put back to them that you know that wasn’t a good an acceptable reason for not feeding him, but we wondered whether he does attend [school] but he doesn’t attend regularly . . . We wondered whether you would perhaps have another look at him. Cons: Well I can do yes, yes, you could say that he was failing to thrive and there is that way his weight creeps up and it was a bit higher in the first year, it has dropped off a bit through the second year in fact it’s crept up from the bottom centile, reasonably satisfactorily I would sayHV: -but it’s probably due to the Pediasure that he’s having not to hisCons: -I don’t know how Paul feeds, and whether he he’s, they obviously feed him don’t they. . . . I don’t know well or how easy he is to feed. © Blackwell Publishers Ltd/Editorial Board 2002


FRW: Cons: FRW: Cons:

T.L.:

Cons: T.L.:

427

Terrible apparently. Oh I imagine he’s very . . . I think he has everything pureed Yeah, . . . I think given the parents’ limitations and . . . They find it difficult to set time aside to feed him, I don’t know, but I think that might be one part. If he gains weight better on the Pediasure then I must say I think . . . I mean I find it very difficult to say whether he’s worse because of lack of stimulation than he would otherwise be. Whether he would be slightly better in a different family, who can know . . . Trouble is we don’t have a diagnosis for Paul, I thought I had but I haven’t ehm and he is waiting to see the geneticists to see if they have more ideas about ehm . . . I find this very difficult because here we are with a family who have a fairly handicapped child who appears to be getting by and is their parenting of that child good enough, I don’t know. Well I think that I think that’s why we’re here really because that’s what we’re not sure about. It’s not sure whether it is good enough or not whether it’s acceptable enough. and how could we make it better for them Well we’re trying to make it better. I think what we’re being faced with is the resistance of the family to accept some of the supports that we’re putting, or even if even if they don’t totally resist it, they certainly make it very difficult for professionals to provide the sort of help.

In her first turn, the team leader delivers a somewhat faltering formulation which acknowledges some improvements, but argues that these are ‘not sufficient’. In the consultant’s first turn, we can see the powerful definitional privilege that paediatricians have in marking cases as ‘not just medical’. He begins by stating that there had been some failure to thrive, but that weight has crept up ‘reasonably satisfactorily’. This is met by an immediate interjection from the health visitor who claims this as the success of Pediasure, a food supplement. Considerable character work in relation to the parents had taken place before the consultant arrived at the meeting. For example, there had been a number of florid accounts of the children’s smelly condition and the parents’ intellectual limitations and the team leader had also told the meeting about the mother’s traumatic childhood which she suggested might well be affecting her ability to parent. Yet, the consultant’s first few turns were enough to destabilise what had, hitherto, been a very robust version. There is a recognisable moral struggle towards the end of the extract with the consultant constructing the family as ‘in need of help’, which receives a powerful ‘tried but failed’ rebuttal by the team leader. The consultant’s version certainly did not return this case to a ‘just medical’ reading (indeed, some time later a child protection case conference was held). However, it problematised the positions that the other speakers had taken and rendered further © Blackwell Publishers Ltd/Editorial Board 2002

428

Susan White

practical work necessary to ‘persuade’ the consultant that the child’s parents and not his body were the problem. For example, arrangements were made for Paul to be fed at school so that the ‘difficult feeder’ hypothesis could be tested. We can see in the data above that clinicians have a wide range of potential relevancies from which to choose when telling a case. Moreover, routine work in the hospital and clinics relies on this case-telling competency, which can deliver with brevity a clear signal to other professionals about what kind of case this is. The reading of the case is accompanied by sometimes subtle and sometimes quite overt warrants. These can range from the invocation of knowledge of anatomy, physiology and pharmacology, through to judgements about parental love. It is worth examining in a little more detail the use of normative warrants in medical work with children, as it is an aspect of professional sense-making fast becoming obscured by the plethora of guidelines and policies associated with EBP.

Good patients/parents, bad patients/parents: invoking parental love This study confirms and develops some previous findings. For example, it confirms that children are a category exempt from classification as bad patients (Dingwall and Murray 1983). Whilst children or young people may sometimes be described as difficult, sensitive, challenging or damaged, this is attributed either to their embodied condition (e.g. they have autism), to their parents’ or carers’ (mis)management, or to some other aspect of their biography. This includes those children and young people whose behaviour breaches moral codes, for example those who self-harm, or engage in behaviour dangerous to others, and those whose chronological age places them very close to adulthood. However, whilst Dingwall and Murray in their study of an accident and emergency department found that moral judgement did not routinely pass to parents, this study confirms Strong’s (1979) earlier findings that, in the more holistic domain of paediatrics, normative judgements about parents are a routine feature of the work. Yet, Strong locates this ‘character work’ in the ‘backstage’ areas of the clinic and argues that, with notable exceptions, clinical encounters are conducted in a bureaucratic format, concerned with the maintenance of distance, politeness and the preservation of parents’ ‘face’. This study suggests something rather different. As the case of Sarah shows, moral judgements and character work behind the scenes frequently affect the extent to which clinicians undertake face work, or shift into an interrogative/pedagogic style in their encounters with parents. The ascent of child protection practices in the last two decades has made its mark on the ceremonial order of the clinic. Whilst for most ‘medical’ cases consultations follow the bureaucratic format identified by Strong, for cases which have shifted into the psychosocial and ‘not just medical’ groups the encounter and case management are much more confrontational (similar to Strong’s charity format). In fact, the imperative to © Blackwell Publishers Ltd/Editorial Board 2002


429

‘confront’ parents with the case formulation is frequently invoked in interprofessional talk. There are two analytically separable, but overlapping, broad moral categories to which parents may be assigned. They may be classified as good/ bad parents and/or as good/bad patients (although in reality they are patients by proxy only). This develops Dingwall and Murray’s (1983) typology and also relates to Silverman’s (1987) and Strong’s (1979) work on paediatric practices. Here, I have made use of Dingwall and Murray’s distinction (derived from McHugh 1970) between theoretic (an agent responsible for their own behaviour) and pre-theoretic (an agent who lacks the capacity to be responsible for their own behaviour) actors. Thus, a person may be deemed a ‘bad parent’ because they are believed to have wilfully neglected or deliberately abused their child, have put their own needs first, or have acted in an evasive and deceitful way. These parents are in the theoretic category and almost always simultaneously defined as both bad patients and bad parents. Extract 4, the Sarah case, is an example of this kind of formulation. As Strong (1979) also noted, however, references to parents’ intellectual limitations are common in paediatrics. Parents may be described as ‘not very bright’, hopeless or helpless. Parents so described may have mental health problems or learning disabilities. Whilst they may be seen as bad, or less than adequate parents, providing they are help-seeking and help-accepting they may avoid categorisation as bad patients. They are assigned to a pretheoretic category and are not held morally culpable for their poor parenting, even when they are dealt with in the formal child protection system. So, whilst they are bad parents, they may still be ‘good patients’, who are grateful and can be helped. However, once parents breech the category-bound expectations (of themselves as parents and as users of expert help) to accept or follow advice, or do not ‘see the need to change’, they become potentially classifiable as both bad parents and bad patients. It is precisely this kind of moral struggle that is taking place in Extract 5. There is consensus that Paul’s parents have limitations for which they cannot be held responsible. They have learning disabilities and have endured traumatic childhoods (they are pre-theoretic), the struggle relates to their willingness to accept help. So there is agreement that they are bad (or less than good) parents, but their status as bad patients is still contestable. I have argued that the ‘not just medical’ cases are particularly revealing. Here, the good parent/bad parent judgement may rely on an attention to forensic detail in the case, designed to uncover anomalies and inconsistencies in the parent’s account, but it also depends very much on judgements about the quality of parental love. Clinicians routinely invoke their ‘feelings’ about the family, often developed and reinforced by story-telling. In the following extract, a paediatrician is briefing a registrar about a family (the Kings) before an out-patient clinic. The consultant and registrar together work up a particular version of the parents, summarised by the consultant as follows: © Blackwell Publishers Ltd/Editorial Board 2002

430

Susan White

Extract 6a Cons: Every tummy pain, there has to be a sort of, every time she cries it must be her tummy. Although we showed that Infacol had no effect, didn’t make any difference on the unit at all, they’re really into the Infacol. Then, they’re into CONStipation because if she hasn’t pooed for so many hours then she’s CONStipated. That’s the problem, but having said that, just to fill you in Sue [directed to researcher], she has got a chromosome abnormality, so she isn’t a – she was very small and is abnormal and we haven’t got a clue what her outcome is going to be. It’s understandable that they’re having difficulty sorting out what is to do with her and what is to do with normal and you suggest things and in the end they do their own thing erm and it is really quite, quite hard work. So, so you’ll, you’ll have good time with them [to registrar]. By invoking their expertise and describing particular instances of the parents misunderstanding or disregarding medical advice, the doctors appear to be constructing these parents as potentially blameworthy as both bad patients and bad parents. In response to this, I make an ironic remark to the registrar ‘Lucky you’. This is rapidly followed by a repair to my erroneous reading of the situation. Extract 6b Cons: No they’re very grateful and they’re notReg: Oh yeah they’re nice enough parentsCons: They’re lovely, but they just need a lot of reassurance. . . . These qualitative judgements are particularly interesting as the parents are clearly being classified as troublesome patients, but whilst perhaps ‘naive’ they remain morally good parents. This contrasted sharply with the case of Sarah above. Clearly the reference to gratitude is important and in keeping with the findings of earlier work. However, the descriptions of parental responses to the baby’s smile are also relevant and later in conversation with me after the clinic, the consultant draws explicit contrasts with the ‘Sarah’ case: Extract 6c Cons: At least they’re [the King family] saying isn’t she lovely, have you seen her smile, she’s pulling that funny face again d’ y’ know SW: yeah mmmm Cons: They’re like, ‘do you think that’s a little bit of chromosome 9’, whereas the mother upstairs [on the ward] would be ‘well that’s [ ] Syndrome isn’t it, I mean look at her she’s got this that and the other’, not ‘isn’t she lovely she’s my daughter’. © Blackwell Publishers Ltd/Editorial Board 2002


431

In clinical practice, judgements about ‘appropriate affect’ form part of a repertoire of rationalities upon which clinicians draw in making sense of cases. Despite their qualitative nature they are indispensable warrants for certainty in case formulation.

Conclusion It was noted at the beginning of this paper that current government policy in health care sponsors a version of evidence-based practice, which assumes that clinical activity should properly rely on one specific form of rationality, which Harrison (1999) has named scientific-bureaucratic medicine: Scientific-bureaucratic medicine . . . centres on the assumption that valid and reliable knowledge is mainly to be obtained from the accumulation of research conducted by experts according to strict scientific criteria . . . It further assumes that working clinicians are likely to be both too busy and insufficiently skilled to interpret and apply such knowledge for themselves, and therefore holds that professional practice should be influenced through the systematic aggregation by academic experts of research findings on a particular topic, and the distillation of such findings into protocols and guidelines which may then be communicated to practitioners with the expectation that practice will be improved (1999: 3). So, this model is ‘scientific’ in the sense that it promises a secure knowledge base which ostensibly provides a proper foundation for clinical decisions. It is bureaucratic in the sense that knowledge is summarised, codified and manualised through the use of protocols, guidelines and computer models, adherence to which may be monitored by managers, or through internal and external audit. This rationality fosters a view of knowledge as external to the clinician, and hence as insulated from the murkier domains of subjectivity, emotion and moral judgement. This paper, however, has shown how much more complex are the forms of reasoning used by clinicians in a paediatric service when they try to establish relations of causation. Shared understandings of patients and their troubles emerge out of interaction between clinicians. By examining how clinicians tell cases, we can see how science, seasoned professional ‘know how’ and moral judgement coexist as warrants for action. Moral judgement in this context is usually ‘affective’ judgement – it is warranted in informal talk principally by invoking emotion – the clinician’s ‘feel’ for the family, or for the appropriateness and/or inappropriateness of parental ‘affect’. There are no algorithms to help clinicians decide about the quality of love, but without these categorisations, much of the work of the clinic would be © Blackwell Publishers Ltd/Editorial Board 2002

432

Susan White

difficult or impossible. They are resolutely part of professional competence and are dimensions that evade any analytic scrutiny in EBP. The paper has argued that the stories clinicians tell have clear material consequences for the various parties to the case, including the clinicians themselves. For example, asking the questions and undertaking the work to shift a case to a ‘not just medical’ formulation (or failing to do so) is not without personal risk, as it often involves a breach of one of the principles of mundane reasoning. There is a common-sense imperative to ‘call into question only so much of the situation as is required for a socially supportable solution to the immediate problem in hand’ (Garfinkel 1967: 108). In the context of a paediatric clinic, the most common and self-evident explanation is often the child’s physical problems. Clinicians who raise concerns about parenting may develop reputations for excessive zealotry or unnecessary punitiveness, but those who fail to explore some social dimension to the problem may (retrospectively) risk charges of naivety or collusion with parents. Obviously, this failure may also very directly expose children to further risk or abuse. It is for this reason that accounts of ‘not just medical’ cases often start as ‘fragile stories’, that is they are defensively designed (Silverman 1998: 93), oriented to the need to persuade and to the possibility of challenge. With repeated retellings, however, and/or through exposure to a receptive audience, they can quickly attain the quality of certainty. Indeed, once robust characterisations of parents have been accomplished, they become one of the principal warrants for certainty in case formulation. This is evidenced particularly clearly in Extracts 2, 3, 4 and 6. For the families in psychosocial or ‘not just medical’ cases there are also a number of clear consequences. The amount of professional time, the number of professionals, the range of services involved and the amount of time spent on co-ordinating activity increases as the zones of relevance expand outside the child’s body to include family and social relations. Once these zones are opened to professional scrutiny it is rare indeed for a case to revert to a purely medical formulation as more information usually breeds continued concern. In these cases, the professional encounter with the family also shifts in focus and style, becoming more confrontational and interrogative, as clinicians address moral questions about the parents’ capacity for change, or about the extent to which they have ‘taken responsibility’ for the problem. This is illustrated in Extract 4 where clinicians are explicitly rehearsing their subsequent meeting with Sarah’s parents. The consequences for families of a psychosocial or ‘not just medical’ reading may be both positive and negative. For example, Sarah’s name was subsequently placed on the register of children at risk and a formal child protection plan instigated. This sustained the high levels of service provision, but also increased surveillance and censure, and carried the implicit ultimate threat of removal of the child. This clearly may have kept Sarah safe, but it also silenced a potential alternative reading of Sarah’s mother as a distressed or depressed parent who was struggling to care for her child and needed help, © Blackwell Publishers Ltd/Editorial Board 2002


433

but was not herself morally culpable for the predicament. On the other hard, it may have secured a safer, healthier future for Sarah. It is not the purpose of this paper to adjudicate on the correctness of decision-making in relation to the cases presented. Rather, I have sought to illustrate the complexity and ambiguity of many presenting problems in child health care. The protocols and procedures of scientific-bureaucratic rationality provide a poor fit with these ambiguities. In their clinical work with such cases, professionals carve certainty from uncertainty, not so much by consulting guidelines or protocols, as by engaging in artful rhetoric and persuasion. In working up causal accounts with other professionals, clinicians do not simply draw upon an external body of knowledge, rather they are literally arguing the case. It is my contention that ethnographic studies such as this can fruitfully render visible these discursive aspects of multidisciplinary clinical work which are taken-for-granted by practitioners, and hence may open these up for analysis and debate by clinicians, academics and policy makers. Address to correspondence: Susan White, Department of Applied Social Science, University of Manchester, Williamson Building, Oxford Road, Manchester M13 9PL e-mail: [email protected]

Acknowledgements This research was supported by the Economic and Social Research Council, grant number R000 222892. I should like to thank the clinicians who allowed me access to their daily work over such an extended period. I am unable to name those clinicians whose contribution to this research was above and beyond the call of duty, but I would particularly like to thank R. and J. and P.T. who will know who they are and also fondly to remember K, whom unfortunately I am no longer able to thank. I am also grateful for the comments and suggestions given by the audiences at the ESRC seminar on clinical governance held at Leeds Health Authority in July 2001, and at the Orders of Ordinary Action conference, Manchester Metropolitan University, July 2001.

Notes 1 It should be noted that there is an ongoing debate within ethnomethodology and conversation analysis about the status of membership categorization analysis (MCA). I do not intend to go into that here, where I have used MCA pragmatically, but hopefully fruitfully. However, for the purist or enthusiast I suggest Hester and Eglin 1997, Silverman 1998 and Housley 2000. 2 Transcription Symbols [] overlapping talk () inaudible, and hence untranscribed, passage © Blackwell Publishers Ltd/Editorial Board 2002

434

Susan White

(0.8) (.) talk TALK tal =

pauses timed in tenths of second audible short pause italics indicate emphasis upper case indicates loudness in comparison to surrounding talk abrupt end to utterance noticeable slowing of tempo of talk latching of utterances

References Anspach, R. (1988) Notes on the sociology of medical discourse: the language of case presentation, Journal of Health and Social Behaviour, 29, 357–75. Atkinson, P. (1995) Medical Talk and Medical Work. London: Sage Atkinson, P. (1999) Medical discourse, evidentiality and the construction of professional responsibility. In Sarangi, S. and Roberts, C. (eds) Talk, Work and Institutional Order: Discourse in Medical, Mediation and Management Settings. Berlin: Mouton de Gruyter. Dingwall, R. and Murray, T. (1983) Categorisation in accident departments: ‘good’ patients, ‘bad’ patients and ‘children’, Sociology of Health and Illness, 5, 2, 127–48. Garfinkel, H. (1967) Studies in Ethnomethodology. Cambridge: Polity. Garralda, M.E. (1994) Chronic physical illness and emotional disorder in childhood, British Journal of Psychiatry, 164, January, 8 –10. Garralda, M.E. (1996) Somatisation in children, Journal of Child Psychology, Psychiatry and Allied Disciplines, 37, 1, 13 – 33. Harrison, S. (1999) New Labour, modernisation and health care governance, paper presented at the Political Studies Association Conference ‘New Labour, New Health’, London, September. Hester, S. and Eglin, P. (1997) Culture in Action: Studies in Membership Categorisation Analysis. Washington D.C.: Washington University Press. Housley, W. (2000) Category work and knowledgeablility within multidisciplinary team meetings, TEXT 20, 1, 83 –107. Jeffrey, R. (1979) Normal rubbish: deviant patients in casualty departments, Sociology of Health and Illness, 1, 1, 90 –107. Latimer, J. (2000) The Conduct of Care: Understanding Nursing Practice, Oxford: Blackwell Science. Meadow, R. (1980) Factitious epilepsy, The Lancet, 1, 25. Meadow, R. (1985) Management of Munchausen syndrome by proxy, Archives of Disease in Childhood, 60, 385. McHugh, P. (1970) A common-sense conception of deviance. In Douglas, J.D. (ed) Deviance and Respectability: the Social Construction of Moral Meanings. New York: Basic Books. Pomerantz, A.M. (1978) Attributions of responsibility: blamings, Sociology, 12, 115 –121. Pomerantz, A.M. (1986) Extreme case formulations: a new way of legitimating claims, Human Studies, 9, 219–30. Sacks, H. (1972) On the analyzability of stories by children. In Gumpertz, J. and Hymes, D. (eds) Directions in Sociolinguistics: the Ethnography of Communication. New York: Holt, Rinehart and Winston. © Blackwell Publishers Ltd/Editorial Board 2002


435

Sacks, H. (1992) Lectures on Conversation (ed. Jefferson, G.) Oxford: Blackwell. Sarangi, S. and Roberts, C. (eds) (1999) Talk, Work and Institutional Order: Discourse in Medical, Mediation and Management Settings. Berlin: Mouton de Gruyter. Silverman, D. (1987) Communication and Medical Practice. London: Sage. Silverman, D. (1998) Harvey Sacks: Social Science and Conversation Analysis. Cambridge: Polity. Smith, D. (1978) K is mentally ill: the anatomy of a factual account, Sociology, 12, 23 –53. Strong, P. (1979) The Ceremonial Order of the Clinic. London: Routledge and Kegan Paul. Taylor, C. and White, S. (2000) Practising Reflexivity in Health and Welfare: Making Knowledge, Buckingham: Open University Press. White, S. (1998) Time, temporality and child welfare: notes on the materiality and malleability of time(s), Time and Society, 7, 1: 55 –74. Woodward, L., Taylor, E. and Downdney, L. (1998) The parenting and family functioning of children with hyperactivity, Journal of Child Psychology and Psychiatry, 161– 9.

© Blackwell Publishers Ltd/Editorial Board 2002