Sociology Without Frontiers? On the Pleasures and Perils of Interdisciplinary Research

by Alison Pilnick
University of Nottingham

Sociological Research Online, 18 (3) 9
<http://www.socresonline.org.uk/18/3/9.html>
10.5153/sro.3108

Received: 11 Mar 2013     Accepted: 17 May 2013    Published: 31 Aug 2013

Abstract

The potential benefits of interdisciplinary research are commonly stated and widely acknowledged. Amongst the many claims that are made, it is suggested that an interdisciplinary approach can lead to greater innovation, promote lateral thinking, and encourage reflexivity in the research process. This paper presents a personal reflection, drawn from experience in one specific sub-field of medical sociology, on how some of these benefits might actually come to fruition. However, it also explores something which is generally given far less consideration: the potential perils of interdisciplinary research. In particular, I will focus on two areas. First, I will raise some intellectual concerns over what interdisciplinary research might mean for the health of sociology as a discipline. Secondly, I will consider some of the ethical issues that can arise when we put our professional sociological skills at the service of another profession. I will conclude by reflecting on what the implications of these concerns are for my own work in the sociology of health and illness, and what might constitute 'successful' interdisciplinary collaboration in this field.

Keywords: Interdisciplinary Research, Sociology of Health and Illness, Interdisciplinarity, Medical Sociology

Introduction

1.1 This paper presents a personal viewpoint on interdisciplinary research, borne out of almost 20 years of research experience. Given that my career history has inevitably shaped my views, I begin with a very brief summary of this history. Following a first degree in pharmaceutical science, I got a job as a pharmacist in a large teaching hospital and found myself working in paediatric oncology, dealing with children with cancer and leukaemia and their families on a regular basis. At the time that I had trained as a pharmacist, it was (as the name 'pharmaceutical science' suggests), a very laboratory-based degree, so that I could formulate tablets or suppositories from their raw ingredients, but I had not received any communication skills training. The mismatch between my skill set and the day to day requirements of my job led me to apply for funding from the Department of Health to carry out research on communication between pharmacists and their clients, which eventually resulted in a sociology PhD. Since completing my PhD I have continued to be based in sociology departments, working with health professionals including GPs, anaesthetists, midwives, obstetricians, genetic counsellors, services for intellectual disability etc. I have also worked with academic colleagues from Nursing, Law, English, Geography, the Business School and the School of Veterinary Medicine, to name just a few. One way to view my career to date is as one long interdisciplinary research project, aimed at providing research findings that are grounded in and contribute to sociology, but also speak to the needs of professionals and policy makers. What follows is an attempt to draw from this experience my view of some of both the pleasures and the (perhaps less explicit) perils of interdisciplinary work. My primary focus is not on the ontological or epistemological conflicts which interdisciplinary research often involves, or on the fractured identity that 'being an interdisciplinary researcher' can entail, though I write with a recognition of the difficulties that these can cause and will refer in passing to some of the problems that can arise in attempting to resolve these. Instead, my focus in this piece is a rather more pragmatic or prosaic one: what is the aim or purpose of interdisciplinary research, and how might we conceptualise its success (as distinct from its quality)?

Background

2.1 To discuss interdisciplinarity, it is first of all necessary to agree on a definition. The definition I have used in this paper is as follows:
'Of, relating to, or involving two or more academic disciplines that are usually considered distinct'.
This definition comes from an EPSRC/ESRC funded project on interdisciplinarity (Conole et al. 2010). However, it is worth noting that other, more wide-ranging, definitions have also been proposed, for example the one that results from Aboelela et al.'s (2007) study. This definition is the culmination of a research project to explore firstly how the term was used in the academic literature and secondly to elicit the views of academics involved in interdisciplinary research across education, business and healthcare. The end result is as follows:
'Interdisciplinary research is any study or group of studies undertaken by scholars from two or more distinct scientific disciplines. The research is based upon a conceptual model that links or integrates theoretical frameworks from those disciplines, uses study design and methodology that is not limited to one field, and requires the use of perspectives and skills of the involved disciplines throughout multiple phases of the research project' (Aboelela et al. 2007: 341).
However, the authors themselves note that few of the studies that they reviewed which claim interdisciplinarity provide evidence to satisfy all these criteria, and so this definition might better be viewed as an aim, or an ideal, for interdisciplinary researchers to aspire to. I will return to a discussion about some of the practical problems this may entail later in the paper.

The benefits of interdisciplinarity

3.1 The next issue to address is why interdisciplinary work is seen as good or necessary. It is argued that there are many benefits to adopting an interdisciplinary approach. The report cited above notes the strengths of bringing different disciplinary perspectives to bear on a research problem, thereby exposing researchers to alternative research perspectives, literatures and methodologies. It is also suggested that it opens up the potential for the development of new theoretical insights and methodological innovations, by bringing different disciplinary perspectives together to address a particular research problem. It is argued that this happens because interdisciplinary work necessarily pushes researchers intellectually; as a result it helps to broaden mindsets and encourages thinking laterally. Pragmatically speaking, it also enables researchers to do things that they couldn't do on their own, by giving them access to the skill sets of different people; this enables them to draw on what Conole et al. (2010) call a 'different armoury of tools' (Conole et al. 2010). Where those who themselves have been involved in interdisciplinary research have been studied, they talk about this in terms of the potential benefits of putting things together that are not normally put together, and also of seeing ones' own discipline differently as a result (Lamont 2009).

3.2 From this perspective, interdisciplinarity is seen as a way to overcome the supposed narrowness and boundedness of academic disciplines, by developing new ways to address problems from different directions which transcend these bounds. A frequent statement by its supporters runs along the lines of 'The real world isn't divided neatly into academic disciplines, so why should research be'? However, given its recent prominence, it might surprise some readers to find interdisciplinarity is not in any way a new concept; Andrew Abbott in his book 'Chaos of Disciplines' (Abbott 2001) cites the US Social Science Research Council as promoting it in 1934. Nonetheless, it does seem to have been afforded much greater prominence over the last 20 years or so, and an acceptance of its benefits by policy makers means that many current funding schemes are explicitly aimed at interdisciplinary work, for example the UK NIHR Research for Patient Benefit scheme, which is a scheme only open to NHS clinicians working in partnership with academic collaborators. There are also training opportunities designed to promote it right from the beginning of careers, for example the new UK ESRC Doctoral Training Centres, which have the bringing together of social scientists from all disciplines as an explicit aim.

Interdisciplinary research : an example

4.1 In order to examine how some of the claims made for interdisciplinary research might work in practice, I will draw first on one of my own projects, which I carried out in conjunction with a Consultant Obstetrician/Professor of Obstetrics and a Professor of Midwifery (Pilnick et al. 2004; Pilnick 2004). This was a project borne out of the implementation of the introduction of a new antenatal screening programme, known as nuchal translucency (NT) screening. As a starting point, those involved in delivering the screening were interested in how they could improve the understanding of women about what the tests entailed and whether to undergo them or not.

4.2 The background to this research was that since antenatal screening (screening and testing women in pregnancy for disorders that can be identified in the fetus) became widely available in the UK, over the last 25 years or so, there have been a number of sociological and more broadly social scientific studies which have suggested that women do not really feel that they have a choice about undergoing these procedures, and that they feel they have to have the tests that are available (see Pilnick 2008 for an overview of some of these studies). At the same time, research involving midwives has suggested that they feel they go to some lengths to construct the optional nature of tests and make clear that they are not compulsory. The starting point for my input was that almost all of this existing research was interview based, so that women asked after consultations said they felt they had had no real choice, and midwives asked after consultations would say that they felt they had clearly given this choice.

4.3 The first topic of discussion between me and my collaborators, then, was how to construct a research project which would shed more light on these conflicting findings. What was quickly apparent from a literature review was that there was almost no research examining the actual consultations, and without looking at those it seemed impossible to understand how both sets of participants appeared to end up with such a different understanding of what had gone on. So I suggested that we use an ethnomethodologically informed approach, recording the actual consultations and analysing the talk using conversation analysis, to see what it was that women and midwives were orienting to as significant in these encounters. What we found was that midwives almost always did, explicitly at the beginning of consultations, state the optional nature of the tests, using words like 'It's up to you', or 'you can have the test or not have the test'. However, these consultations often went on for some considerable time and what happened in the consultation subsequently tended to undermine this for 3 reasons:

  1. Placement. These 'optional' tests were discussed immediately after much more routine and unproblematic tests, such as taking a blood sample from the mother to test for anaemia, so the placement of them tended to conflate them with much more routine and straightforward tests which no-one would be expected to refuse.
  2. Midwives tended to highlight the benefits of the tests (for example telling women it would enable them to find out sooner if there were any problems) and downplay any disadvantages, so while they didn't explicitly say it would be better to have the test, many women inferred this from their discussions.
  3. In some cases the choice that was presented seemed to be more focused on which tests the woman would choose (the 'old' blood test, which was still available, or the 'new' NT scan), rather than a choice between testing and not testing.

4.4 By approaching the project in this way, we were able to see why it was that previous research had produced such conflicting results, and understand why these apparently competing versions could co-exist. We could also see that while midwives did indeed make choice explicit at the outset, by the end of the consultations this was not the aspect that women were orienting to, given what happened in the rest of the consultation. Based on the consultations, we could also make recommendations for practice. Potentially, then, this study might be seen as an example of an interdisciplinary project that worked well. Rather than taking this success at face value, however, I will now turn to explore the potential disadvantages of interdisciplinary research in a little more depth.

The dangers of interdisciplinarity

5.1 Insofar as potential disadvantages of interdisciplinary work are considered, the first question raised is usually a practical one; how do we know whether interdisciplinary research is any good? As Lamont describes, interdisciplinarity 'often brings about a broadening and multiplication of evaluation criteria, which makes both individual judgement and group agreement much more difficult' (Lamont 2009: 209). In other words, it can be difficult to establish standards of validity across subject domains. As the previously cited ESRC/EPSRC report describes, this presents researchers with a challenge because they can lack effective criteria for evaluating interdisciplinary research but also for planning it in the first place. This problem can be exacerbated because academic vocabularies and practices are often discipline-specific, so that collaborators may not be certain they are talking about the same thing (Conole et al. 2010).

5.2 In addition to the problem of understanding one another, however, there is a deeper difficulty tied to discussions over the quality of interdisciplinary research, and one which may account for some of the lack of evidence Aboelela and colleagues (2007) found in relation to the criteria of their proposed definition. Assessing quality is bound up with issues of politics, power and identity, not least in terms of who carries out such assessment. As Lingard et al. (2007: 501) describe, writing about their own experiences of conducting interdisciplinary research in a team which spans physicians, professions allied to medicine and academics, there have to be 'negotiations not only with one another as particularly positioned individuals, but also with the ideological and organizational forces that structure our scholarly worlds'. These scholarly worlds (note the plural) may hold different views on quality, so that all ontological and epistemological positions might not be seen as equal, but even this is not the end of the story. The influence of ideological and organisational forces means that assessments of quality are bound up with not only disciplinary, but also institutional norms. Interdisciplinary research is far from the only means by which such norms might be transgressed, as Tsouroufli's (2012) account of being a feminist academic in a medical school ably illustrates, but as she also shows, those whose research does not operate within accepted boundaries of authenticity and legitimacy risk being discredited. The result of this failure to fit, Brown and Taylor (2012) suggest, is that academics who find themselves on the margins have to find a 'comfortable place'; this may result in seeking acceptance by deferring to existing institutional or organisational views and values rather than seeking to develop or challenge them.

5.3 Clearly, the issue of differentiating 'good' interdisciplinary research from 'bad' is a significant issue for both commissioners and consumers of research, but as I have described above, we need to be alert to the less explicit but no less pervasive factors which are brought to bear on these judgments. However, as I suggested at the outset, what I present here is a very personal view, and in this personal discussion of the perils of interdisciplinary I want to focus more closely at a different level. In particular, I want to raise two concerns, one which might best be conceptualised as intellectual, and one as ethical. As previously, to do this I will draw on the area of sociology which I know best, which is medical sociology, and also the methodological approaches with which I am most familiar, which are ethnomethodology and conversation analysis.

The sociology of medicine vs the health of sociology?

6.1 Andrew Abbott, in his book 'Chaos of disciplines' (2001), admits 'I chose sociology because more than any social science sociology would let me do what I pleased' (preface). This is the case, he suggests, because sociology is the most general of the social sciences, or the least defined. Abbot further argues that one of sociology's defining characteristics is that the discipline is not very good at excluding things from itself. As a result it has become a discipline of many topics. This view is encapsulated beautifully, if somewhat dramatically, in the following quote, where he suggests that it is a 'discipline like a caravansary on the Silk Road filled with all sorts of types of people and beset by bandit gangs of positivists, feminists, interactionists and Marxists, and even by some larger far off states like Economics and the Humanities, all of whom are bent on reducing the place to vassalage' (2001: 6). This quote then, paints rather a depressing picture; because we are not quite sure where our boundaries are, or even who is one of us and who is not, we are perpetually engaged in all kinds of battles with all kinds of people on all kinds of fronts.

6.2 Abbot goes on to suggest that a consequence of sociology's breadth and diversity is that while single minded disciplines like Economics may have an accepted right way and a wrong way to conduct research, sociology does not, and therefore both its scope and its responsibility are greater. In particular, Abbot warns of sociology's inability to keep 'judgements about the rightness of things separate from judgements of their actual nature' (2001: 198), so that value judgements become mixed with scientific ones. This he suggests is a potential strength but also a problem, since it allows and enables sociology to be co-opted by other disciplines for their particular purposes.

6.3 In reflecting on this danger of co-option, I want to return to thinking about medical sociology, and a distinction that is sometimes made there that parallels this danger. This is the distinction between sociology of medicine and sociology in medicine, a distinction first made by Robert Straus in 1957. According to Straus (1957) sociology of medicine is concerned with studying such factors as the organisational structure, role relationships, value systems, rituals and functions of medicine as a system of behaviours. He suggests that this type of activity can best be carried out by persons operating from independent positions outside the formal medical setting. By contrast, sociology in medicine consists of collaborative research or teaching, often involving the integration of concepts, techniques and personnel from many disciplines. Straus further suggests that these two types of sociology tend to be incompatible with each other, for very practical reasons: the sociologist of medicine may lose sociological objectivity if s/he comes to identify too closely with medical teaching or clinical research; and the sociologist in medicine risks the loss of a good working relationship if s/he tries to critically study colleagues and their practices.

6.4 Though this distinction was first made in 1957, it still has currency today. Usher, in the 2007 edition of the 'Blackwell Encyclopaedia of Sociology', states 'Sociology in medicine is the label given to the collaborative work between sociologists and medical or health personnel within medical institutions or health care organizations. This distinction represents the applied work of medical sociologists in the pure versus applied dichotomy of the social sciences. In its most extreme form, sociology in medicine encompasses sociological work aimed at the provision of technical skills and problem solving for the medical community while neglecting contributions to the parent discipline' (Usher 2007, italics added).

6.5 The sentence I have italicised in the above quote is, in my personal view, one that gets right to the heart of the first problem I want to highlight with interdisciplinarity. This neglect of contributions to the parent discipline is, I think, a criticism that can quite fairly be applied to my own work on antenatal screening which I outlined earlier in this piece. This work did indeed solve a problem for the medical community, because it provided an explanation for how and why two sets of apparently competing accounts of antenatal screening consultations could co-exist, and offered practical suggestions as to how the issue of choice might be foregrounded in the future. In terms of the contributions it made to sociology, however, these are a bit less clear cut, a bit less successful. It could perhaps be argued that it promoted the use of a specific sociological approach (conversation analysis), or that it shed some light on a group of health professionals (midwives) who had not received much sociological attention, but it would be difficult to make the case for a more substantive contribution. One view to take on this would be that, since the research served a practical purpose, and had demonstrable impact in terms of service organisation and delivery, this lack of contribution to the parent discipline should not matter. Perhaps, on a case by case basis, it is true that it should not matter very much. However, there is a wider problem here. The project I have described is, I think, a fairly classic example of interdisciplinary work. As in this case, such work is often problem driven, coming about to address or explore specific issues or contingencies, and is therefore very specific. As Abbott (2001) argues, these characteristics of interdisciplinary work mean that it does not in itself create enduring, self-reproducing communities. Interdisciplinary studies are ultimately dependent on the specialised disciplines on which they draw to generate new theories and methods. The end result is that interdisciplinary work requires a strong sociology, but it may not necessarily contribute much to the strength of that sociology itself. Such a relationship might at worst be categorised as parasitical.

The ethics of interdisciplinarity

7.1 Having said that there were two specific concerns relating to interdisciplinary research that I wanted to explore, I will now turn to the second. It is worth restating here that this is a very personal take on the issue, and it may also be that this is a particular problem in my own specific field of research, given that those I most often collaborate with are medical practitioners of various sorts and given that medicine occupies a fairly unique niche in terms of power and status in UK society. Returning for a moment to issues raised by Aboelela et al.'s (2007) proposed definition of interdisciplinarity, this means that issues of power, and of who decides what constitutes legitimate and authentic research, may be particularly pertinent. Nevertheless, the specific concern that I want to address here is still an issue that has been remarked upon by sociologists working in different fields, as the quote in the following sentence illustrates. My second concern relates to the ethical problems that can arise 'When as social scientists we put our professional skills at the service of another profession, and amplify its voice and the power it can enforce over those who become the objects of its scrutiny' (Goodwin 1994: 626). These ethical problems are, I think, compounded by the fact that sociologists have a tendency to see themselves on the side of the angels, and sociological research has historically often been designed to take the position of the underdog. In fact, one of the potential perils of interdisciplinary research is that it is possible to have quite the opposite effect. Once again I will illustrate this discussion by drawing on other research with which I have been involved.

7.2 For a number of years, I have carried out research into doctor/patient interaction, as well as other health care professional/patient interaction, with a particular interest in advice giving. Increasingly, I am invited by clinicians to discuss potential topics for collaboration or ideas for future research. In these discussions, there is a common recurring theme. Different professions may refer to it in different ways, but it invariably is not long before the topic of 'non-directiveness' or 'patient autonomy' or 'patient-centredness' comes up. Frequently, practitioners ask how research might help them be less directive, or how it can help them encourage patients to make their own choices.

7.3 To set the context for these requests, I draw specifically on work carried out in collaboration with another sociologist, Robert Dingwall (Pilnick & Dingwall 2011). As we have previously noted, researchers have traditionally treated doctor/patient interaction as a site where doctors exercise power over patients, grounding this in the work of Parsons (1951). The assumption that the resulting asymmetry is problematic has underpinned the quest for 'patient-centred' medicine, and patient-centred medicine has become the stated aim of many branches of healthcare. However, what is sociologically interesting is that this a position assumed without much in the way of research evidence; it seems to rest on a moral assumption that doctors having power is bad and that rebalancing some of that power with patients is therefore desirable. Despite this assumption, a 2001 Cochrane systematic review of interventions aimed specifically at increasing patient-centredness in consultations found a mixed picture. While such interventions are generally successful in modifying styles of communication and increasing rates of patient satisfaction, it is much less clear as to whether they result in positive health outcomes (Lewin et al. 2001). Some studies, such as Stewart et al. (2000), have shown a link between patient-centred practice and health outcomes, but others, e.g. Kinmonth et al. (1998) have shown clear negative effects or no effect at all (Jaen et al. 2010; Lee & Lin 2010).

7.4 These studies show us that the clinical evidence that patients do better from patient-centred medicine is mixed. But what about the sociological evidence that underpins this power-based analysis? Sharrock (1979), expressing a view I share, attacks analysis focusing on the power held by doctors for acting as an indictment rather than an inquiry. From this critical point of view, descriptions of medical encounters as 'oppressive' not only find fault but also assign blame to doctors, and viewed through this lens the doctor–patient encounter becomes a struggle, which can only ever end in the stifling of patients. Emanuel Schegloff, an eminent researcher in the Conversation Analytic tradition (see e.g. 1991, 1997) expresses the case more generally. As Pilnick and Dingwall (2011) have noted, Schegloff contrasts two research positions. The first is the position of critical analysts of doctor–patient interaction like Howard Waitzkin, who begin by stipulating the a priori existence and relevance of power and status. The analytic focus then always falls on how this power and status is played out in relation to social classes, ethnic groups, genders etc. The second position is taken by those who instead are trying to understand how participants are actually producing the observable social order through the way in which they orient, moment-to-moment, to whatever matters are locally relevant at the time of production. According to Schegloff, the first of these:

'allows students, investigators, or external observers to deploy the terms which preoccupy them in describing, explaining, critiquing, etc. the events and texts to which they turn their attention. There is no guaranteed place for the endogenous orientations of the participants in those events; there is no principled method for establishing those orientations; there is no commitment to be constrained by those orientations. However well-intentioned and well-disposed towards the participants – indeed, often enough the whole rationale of the critical stance is the championing of what are taken to be authentic, indigenous perspectives – there is a kind of theoretical imperialism involved here, a kind of hegemony of the intellectuals… whose theoretical apparatus gets to stipulate the terms by reference to which the world is to be understood – when there has already been a set of terms by which the world was understood – by those endogenously involved in its very coming to pass.'(Schegloff 1997: 167)

7.5 Schegloff's warning of theoretical imperialism reproduced above concludes with the unforgiving comparison of this kind of 'critical' analysis to those who speak of Columbus 'discovering' America as if indigenous people were not already living there. The implication is that sociologists can see their research as 'uncovering' or 'revealing' features of everyday life (such as oppression or power struggles) as if ordinary people were not already managing everyday life quite satisfactorily without any identifiable reference to these features. It follows that the task for sociologists should not be the privileging of sociology's concerns but the discovery of members' concerns, with an acceptance that the two might or might not overlap.

7.6 Schegloff, as I have noted above, is himself an eminent researcher in the Conversation Analytic (CA) tradition, and as Pilnick and Dingwall (2011) have noted, CA research which aims to establish how the interactional dynamics of doctor/patient interaction function suggests three key findings. Firstly, patients do defer to doctors, but – and secondly – they do so actively (for example by withholding their own assessments of things). Thirdly, they do this for good reason, because they share with doctors the functional goal of diagnosis and/or treatment. Practically speaking, in choosing to put their problem in the hands of a doctor, patients recognize that some level of deferral is likely to be the most efficient strategy for achieving their goal. Put bluntly, if a patient claims greater expertise than a doctor, this is likely to undermine their claim to seeking help in the first place. Additionally, since it is the patient's condition that is under review and not the doctor's, it makes functional sense that a doctor should drive the questioning, at least in the initial stages of the encounter. Researchers from this perspective also identify asymmetry in consultations, but the difference is that rather than seeing this as being something inherently problematic that is done by doctors to patients, they show how it is co-constructed.

7.7 It is worth noting however, that while such a perspective may avoid the theoretical imperialism warned of by Schegloff, it does not necessarily make CA researchers blameless from an ethical point of view. Building on the findings described above, some conversation analytic researchers have suggested that CA is uniquely placed to assist with the development of more patient-centred medicine, as a quote that Pilnick and Dingwall (2011) identify suggests:

'Analysing co-construction is a direct research embodiment of patient-centredeness and it facilitates the biopsychosocial approach to the interview as well as a more recent emphasis on relationship-centred care.' (Maynard & Heritage 2005: 434–5 original italics).

7.8 The implication here is that CA could be used as a tool to achieve patient-centredness. However such a suggestion rests on an uncritical acceptance that patient-centred care is the way forward for medical practice, i.e. it accepts the medical agenda wholesale and co-opts CA as a tool to bring this about. As we have already seen, the evidence for the benefits of patient-centredness to patients' health is at best mixed. Examples from my own empirical research in a variety of settings also suggest that it may raise difficult interactional issues. For example, what doctors see as patient-centredness, and the way in which they promote the autonomy which is seen to go along with this, may well be oriented to by patients as an unwanted burden. The two empirical examples reproduced below serve as a brief demonstration of some of the issues that may be raised.

7.9 Example 1 comes from a project I am currently working on with a colleague, Olga Zayts, examining antenatal screening consultations in public hospitals in Hong Kong (see Pilnick & Zayts 2012; Zayts & Pilnick forthcoming 2014).

Context: This extract comes after the nurse has delivered the information about available screening and testing option, and the nurse begins to enquire about a decision.

  1. N:     Or, can you decide today? Or you want- you prefer to talk to your
  2.           husband first?
  3. W:     It's ok, which is the best for me? [I(h)] (h)don't k(h)now.
  4. N:                [Naa,] see, (.) Naa, see now whether
  5.         you want to have the, the first thing, is (.) you decide you want to have
  6.         test, or no test, ok? And the second thing is whether you want to have
  7.         direct test, or indirect test. Whether you want to have an accurate test,
  8.         or whether you want to have a safe test. If you want to have a safe test,
  9.         then you need to undergo the screening test. But if you think that I
  10.         want to have an accurate test, then you need to undergo the, the::, the::,
  11.         the other test.
  12.         (10) ((W is looking at the papers in front of the N))
  13. W:     °huh huh huh .hhh° ((chuckle))

7.10 The framing of the woman's question in line 3 of Example 1 suggests that her difficulty with making a decision relates to 'not knowing' what would be the best decision in her circumstances (rather than 'not understanding' the information she has already been given). However, the nurse responds to the request indirectly by summarizing (at some length) the information about available choices that she has already delivered, rather than giving direct advice. The nurse also explicitly states that the decision lies with the woman. But this repetition of information does not lead to the woman's decision. Instead there is a long pause (10 seconds in line 12 where the woman stares at the papers on the desk) followed by a wry laugh that may signal that she is still encountering difficulties in reaching a decision. This consultation brings to mind Bosk's vivid metaphor that 'the dark side of patient autonomy [is] patient abandonment' (1992: 158). Whilst it is clear that the woman here is given autonomy to make her own decision, it is far less clear that this autonomy is helpful or desirable to her.

7.11 Example 2 is taken from some work on genetic counselling (Pilnick 2002a, 2002b), where practitioners are particularly concerned with non-directiveness as a component of patient-centred counselling. The context for this extract is that a client with myotonic dystrophy has just reported to the clinical geneticist his mother's view that his condition might best be tackled through diet.

Example 2:

7.12 In this extract, it is notable that the geneticist avoids explicitly contradicting the view of the client's mother. In fact, when the client himself attempts to make a distinction based on the fact that the geneticist is a 'learned man' while his mother has no specialist knowledge, the geneticist himself aligns with the mother by suggesting they share a desire for explanation. In this instance, then, the desire to be non-directive over the issue of diet appears to result in a disavowal of the clinical geneticist possessing any greater expertise than the client's mother.

7.13 The two brief examples above give a snapshot of the consequences that a commitment to patient-centred or non-directive practice can have, suggesting at the very least that it may not be of unproblematic benefit to patients. As I have noted previously, my collaborators often tell me that this is what they would like joint research to achieve (this would be their idea of 'success'), and sociologists may feel they have the tools to bring this achievement about, but it is not clear that patients share this goal, or that sociology might not better be positioned as an objective observer of medical practice. If patients accept or even prefer the presence of asymmetry in healthcare consultations, then why should sociologists be co-opted into achieving these reforms? As Pilnick and Dingwall (2011) argue, if research is consistently finding embedded asymmetry, the research task for sociology might properly be conceptualized by starting with a focus on whether and what functional purpose this asymmetry might serve, rather than aligning directly with the medical reformers' goal of trying to overcome it. Any subsequent suggestions for reform would need to be grounded in an understanding of both how and why this asymmetry persists. Ultimately, this might result in less radical reform but it would be reform grounded in sociological evidence rather than co-option into a reform agenda identified by the medical profession. A further implication of this line of thought is that it is not appropriate to present any sociological method as a 'magic bullet' to bring about reform, as opposed to a means of understanding the nature of the institution being studied. This latter point has particular implications for interdisciplinary research, given the fact that it is often targeted at the resolution of a specific problem.

Conclusions

8.1 This paper has ranged across a variety of issues, and I aim to conclude with some final thoughts on interdisciplinarity, and some advice for the aspiring interdisciplinary researcher. Given the dangers of co-option identified above, and the ethical implications that can follow, I would suggest that for research to be truly interdisciplinary, it is of fundamental importance that sociologists are not just to be the people who turn up with the 'armoury of tools' when the plans have already been drawn up. Instead, there is a need to be involved in the formulation of the research questions from the outset, and to critically engage with the aims, aspirations and desires of collaborators. As I have acknowledged at various points throughout this paper, this is not necessarily an easy task. Aboelela et al.'s (2007: 341) definition requires 'the use of perspectives and skills of the involved disciplines throughout multiple phases of the research project', but I would like to place emphasis here on what happens before the project takes shape. This is necessary for two reasons. Firstly, it is necessary in terms of the strength, or success, of the output, so that the comparison with Columbus 'discovering' America can be avoided. From this perspective, it is important to avoid transmuting participants' understandings through the theoretical imperialism of sociology, but equally important to avoid 'discovering' as a sociologist what those in other disciplines may already be well aware of. Secondly, an involvement in the conceptualisation of projects helps to ensure a successful focus not just on what sociology might bring to an endeavour, but also how sociology itself might develop as a result. Put simply, it helps to guarantee that research that is carried out also contributes to the ongoing health of the discipline as well as providing the answers those in other disciplines might seek.

References

ABBOTT, A. (2001) Chaos of Disciplines. Chicago: University of Chicago Press.

ABOELELA, S.W., Larson, E., Bakken, S., Caranasquillo, O. et al. (2007) 'Defining interdisciplinary research: conclusions from a critical review of the literature', Health Services Research, 42(1) (Part 1) p. 329–346.

BOSK, C. (1992) All God's Mistakes: Genetic Counseling in a Pediatric Hospital. Chicago: University of Chicago Press.

BROWN, K.E. & Syme-Taylor, V. (2012) 'Women academics and feminism in professional military education', Equality, Diversity and Inclusion: An International Journal, 31(5/6) p. 452–466.

CONOLE, G., Scanlon, E., Mundin, P. & Farrow, R. (2010) Interdisciplinary Research- Report from the Technology Enhanced Learning Research Programme (funded by ESRC/EPSRC). London: Institute of Education.

GOODWIN, C. (1994) 'Professional vision', American Anthropologist, 96(3) p. 606–633.

JAEN, C.R., Ferrer, R.L, Miller, W.L., Palmer, R.F. et al. (2010) 'Patient outcomes at 26 months in the patient-centered medical home national demonstration project', Annals of Family Medicine, 8 p. S557–567.

KINMONTH, A.L., Woodcock, A., Griffin, S., Spiegal, N. & Campbell, M. (1998) 'Randomised controlled trial of patient-centred care of diabetes in general practice: Impact on current well-being and future disease risk', British Medical Journal, 317 p. 1202–1208.

LAMONT, M. (2009) How Professors Think: Inside the Curious World of Academic Judgment. Cambridge, MA: Harvard University Press.

LEE, Y-Y. & Lin, J.L. (2010) 'Do patient autonomy preferences matter? Linking patient-centered care to patient–physician relationships and health outcomes', Social Science and Medicine, 71(10) p. 1811–1818.

LEWIN, S.A., Skea, Z.C., Entwistle, V., Zwarenstein, M. & Dick, J. (2001) 'Interventions for providers to promote a patient-centred approach in clinical consultations', Cochrane Database of Systematic Reviews, (4) CD003267.

LINGARD, L., Schryer, C.F., Spafford, M.M. & Campbell, S.L. (2007) 'Negotiating the politics of identity in an interdisciplinary research team', Qualitative Research, 7(4) p. 501–519.

MAYNARD, D.W. & Heritage, J. (2005) 'Conversation analysis, doctor-patient interaction and medical communication', Medical Education, 39 p. 428–435.

PARSONS, T. (1951) The Social System. London: Routledge and Kegan Paul.

PILNICK, A. (2002a) '"There are no rights and wrongs in these situations": identifying interactional difficulties in genetic counselling', Sociology of Health and Illness, 24(1) p. 66–88.

PILNICK, A. (2002b) 'What "most people" do: exploring the ethical implications of genetic counselling', New Genetics and Society, 21(3) p. 339–350.

PILNICK, A. (2004) '"It's just one of the best tests that we've got at the moment": informed choice in nuchal translucency screening for fetal abnormality', Discourse and Society, p. 451–465.

PILNICK, A. (2008). '"It's something for you both to think about": choice and decision making in nuchal translucency screening for Downs Syndrome', Sociology of Health and Illness, 30(4) p. 511–530.

PILNICK, A. & Dingwall, R. (2011) 'On the remarkable persistence of asymmetry in doctor-patient interaction: a critical review', Social Science and Medicine, 72(8) p. 1374–1382.

PILNICK, A.M., Fraser, D.M. & James, D.K. (2004) 'Presenting and discussing nuchal translucency screening for fetal abnormality in the UK', Midwifery, 20(1) p. 82–93.

PILNICK, A. & Zayts, O. (2012) 'Let's have it tested first: directiveness, culture and decision-making following positive antenatal screening in Hong Kong', Sociology of Health and Illness, 34(2) p. 266–282. (Also published in Armstrong, N. & Eborall, H. (Eds.) (2012) The Sociology of Screening and Testing. Oxford: Wiley-Blackwell).

SCHEGLOFF, E. (1991) 'Reflections on talk and social structure', in Boden, D. & Zimmerman, D. (Eds.), Talk and Social Structure: Studies in Ethnomethodology and Conversation Analysis (p.44–70 ). Cambridge: Polity Press.

SCHEGLOFF, E. (1997) 'Whose text? Whose context?', Discourse and Society, 8 p. 165–188.

SHARROCK, W. (1979) 'Portraying the professional relationship', in Anderson, D.C. (Ed.), Health Education in Practice (p. 125–146). London: Croom Helm.

STEWART, M., Brown, J.B., Donner, A., McWhinney, I.R. et al. (2000) 'The impact of patient-centered care on health outcomes', Family Practice, 49(9) p. 796–804.

STRAUS, R. (1957) 'The nature and status of medical sociology'. American Sociological Review, 22(2) p. 200–204.

TSOUROUFLI, M. (2012) 'Breaking in and breaking out a medical school: feminist academic interrupted?', Equality, Diversity and Inclusion: An International Journal, 31(5/6) p. 467–483.

USHER, C. (2007) 'Sociology of Medicine', in Ritzer, G. (Ed.), Blackwell Encyclopaedia of Sociology. Oxford: Blackwell.

ZAYTS, O. & Pilnick, A. (forthcoming 2014) 'Genetic counselling in multicultural and multilingual contexts', in Hamilton, H. & Chou, S. (Eds.), Routledge Handbook of Language and Health Communication. Abingdon: Routledge.