Health technology adoption and the politics of governance in the UK

ARTICLE IN PRESS

Social Science & Medicine 63 (2006) 3102–3112 www.elsevier.com/locate/socscimed

Health technology adoption and the politics of governance in the UK Timothy Milewa Centre for the Study of Health and Illness, Sociology, School of Social Sciences and Law, Brunel University, Uxbridge UB8 3PH, UK Available online 18 September 2006

Abstract The manner in which clinical and cost-effectiveness data are used to inform decisions about the funding and availability of drugs, therapies and medical devices is inherently politicised within collectively financed systems of health care. The National Institute for Health and Clinical Excellence (NICE) was established by the British government in 1999 to reach evidence-based decisions on whether selected health technologies should be made available by the National Health Service in England and Wales. But NICE is also required to involve a broad range of interested parties in the decision-making process, provide detailed rationales for its rulings and defend appeals from aggrieved parties. Debates about the emergence of ‘‘deliberative’’ forms of policy governance—based upon participation by a broad range of stakeholders rather than reliance on scientific, bureaucratic or political expertise alone—are thus particularly apposite. This article draws on a study of decision-making within NICE by focusing upon the tenor and orientation of deliberation about the adoption of health technologies. Does such deliberation take place upon a level playing field for different interests? Or do implicit parameters and understandings in the deliberative process tend to privilege some interests by structuring debate and attendant outcomes? Findings suggest that deliberative assumptions and parameters pertaining to fluid and contestable ideas of transparent reasoning and domain competence both reflect and shape relationships of influence and marginality among participants. Broader analytical implications centre on a distinction between ‘‘deliberative democracy’’ and ‘‘democratic deliberation’’. The extent to which this distinction is acknowledged and addressed in policy and practise will have marked implications for the substantive nature of attempts to broaden involvement in decision-making within public sector bodies such as NICE. r 2006 Elsevier Ltd. All rights reserved. Keywords: UK; Health policy; Health technology appraisal; Health technology assessment; NICE; Public involvement

Introduction The manner in which clinical and cost-effectiveness data are used to inform policy decisions on the public funding and availability of drugs, therapies and medical devices is of fundamental consequence to patient groups, clinicians, health technology Tel.: +44 1895 265450.

E-mail address: [email protected]. 0277-9536/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.socscimed.2006.08.009

manufacturers and—ultimately—citizens and taxpayers. The National Institute for Health and Clinical Excellence (NICE) was established by the British government in 1999 to reach transparent, evidence-based, decisions on whether selected health technologies should be made available by the National Health Service in England and Wales (Syrett 2003). This role has been accompanied by a requirement that decisions reflect consultation with stakeholder constituencies, the involvement of ‘‘lay’’

ARTICLE IN PRESS T. Milewa / Social Science & Medicine 63 (2006) 3102–3112

individuals in decision-making and reasoned, minuted, deliberation. Decision-making within NICE thus encompasses a tension. In one respect, reflecting the tenor of contemporary medical orthodoxy, NICE is charged with the production of evidencebased decisions (Elwyn & Edwards 2001; Gibbons, Smith & Antman 2003). But in a second regard NICE is politicised in that it has to afford interested parties access to the decision-making process, demonstrate cognisance of these myriad voices and reflect transparency in attendant rationales for its rulings. Contemporary debates about the emergence and substance of participatory or ‘‘deliberative’’ forms of governance—based upon reflection by a broad range of stakeholders rather than reliance on scientific, bureaucratic or political expertise alone—are thus particularly apposite (Fung & Wright, 2003; Hajer, 2003; Innes & Booher, 2003; Levine, Fung & Gastil, 2005; Newman Barnes, Sullivan, & Knops, 2004). Accordingly, this article draws upon a study of decisionmaking within NICE by focusing upon influences on the tenor and orientation of deliberation about the adoption of health technologies. Does such deliberation take place upon a level playing field for different interests? Or do implicit parameters and understandings in the deliberative process tend to privilege some interests by structuring debate and attendant outcomes? Medicine, evidence and policy The British National Health Service (NHS) has provided health care free at the point of delivery for the large majority of British citizens since 1948. But, like other public services, the NHS and its political masters have, in recent decades, had to adapt to a more critical citizenry, accommodate the rise of ‘‘expert’’ clients and engage increasingly with advocacy bodies (Appleby & Alvarez Rosete, 2003; Baggott, Allsop, & Jones, 2005; Fox, 2005). But the New Labour government, returned in 1997, has provided an additional focus for debate and conflict by stressing the importance of ‘‘evidencebased’’ health policy. This emphasis has its roots in a series of post-war studies that drew attention to disparities in the effectiveness of different medical interventions in relation to the same conditions, geographical variation in the efficacy of treatments and the need to develop robust techniques to measure and compare treatments (Timmermans & Berg, 2003, pp. 14–15). There followed a growing

3103

academic and professional veneration of statistically significant results based upon rigorous randomised controlled trials; an increasing emphasis, from the 1960s onwards, on the systematic review of scientific literatures on given conditions and—particularly in the last 15 years—a proliferation of clinical guidelines with regard to the diagnosis and treatment of particular conditions (Chalmers, 1974; Petticrew, 2001). There has remained, however, an awareness that evidence is often incomplete, contestable and— through ongoing research and publications—cumulative (Byrne,2004; Harrison, 1998). Accordingly, when the National Institute for Clinical Excellence was established in 1999 (renamed the National Institute for Health and Clinical Excellence in 2005), its remit centred upon assembling evidence on the clinical and cost-effectiveness of selected health technologies and issuing guidance on whether, how and when they should be used within the NHS in England and Wales. This remit was extended in 2004 to encompass the review of interventions to promote public health (Department of Health 1997, par. 7.11; Department of Health 2004, Annex B, par. 15; Secretary of State for Wales 1998, par. 3.10). In procedural terms, suggestions for health technologies to be ‘‘appraised’’ by NICE may be made by anyone to the Department of Health (England) or National Assembly (regional government) for Wales. All proposals beyond this stage are then assessed by the Advisory Committee on Topic Selection (ACTS) with reference to the prima facie evidence on clinical and cost-effectiveness and the severity of related medical conditions (Hill, Garattini, Loenhout, O’Brien, & Joncheere, 2003). Selected proposals are then passed to a subcommittee of ACTS, the Joint Planning Group. This group considers the topics in relation to their technical feasibility and the capacity of NICE. Once a health technology has been accepted for appraisal, NICE seeks to identify all those with a likely interest in the attendant decision—a constituency that typically embraces patient advocacy groups, manufacturers, professional medical associations (such as the Royal College of Surgeons) and any other groups or individuals who cast themselves as interested parties. This constituency is then used as a sounding board while NICE refines questions to be asked of a particular technology. These questions, in turn, inform a review of published and grey literature by at least one independent ‘‘Technical Assessment Group’’ contracted by NICE for this

ARTICLE IN PRESS 3104

T. Milewa / Social Science & Medicine 63 (2006) 3102–3112

purpose. The group’s report, ideally encompassing the best and most recent evidence is considered by one of two branches of the main decision-making body within NICE, the Appraisal Committee. The latter’s membership—typically encompassing clinicians, health economists, representatives of patients’ groups, senior NHS managers, pharmacists and research specialists from health technology manufacturers—meet to reach a provisional decision on the admissibility of the technology in question. This decision-making reflects detailed discussion among members but also dialogue with invited ‘‘witnesses’’. A provisional decision is then publicised with a view to attracting further comments (ideally concerned with matters of fact). The Appraisal Committee subsequently reconvenes to endorse, amend or defer a final decision that is then translated into operable guidelines. National Health Service organisations across England and Wales are then ‘‘required to provide funding for medicines and treatments recommended by NICE’’ (National Institute for Health and Clinical Excellence 2005, p. 9). NICE had, by the summer of 2005, reviewed over 90 health technologies. These included, for example, drug treatments for conditions such as obesity and motor neurone disease, surgical techniques for heavy menstrual bleeding, patient education models for diabetes and new hip prostheses. In terms of policy outcomes, an analysis of 33 sets of recommendations by the Appraisal Committee in relation to 28 disease areas (encompassing about a third of all appraisals at the time) indicated that four recommendations were explicitly negative and that approximately a third recommended first-line utilisation of the technologies. The largest proportion of recommendations were qualified in that they tended to reflect the reality that many health technologies are only useful to certain groups of patients under particular conditions and that many drugs only help a proportion of patients suffering from the same illness (Abacus International, 2005). Perhaps not surprisingly, given the necessity for such detailed judgements, even ‘‘final’’ decisions can be referred to an Appeal Panel if an aggrieved party maintains that one or more of three criteria have been breached. The latter encompass a claim that NICE has not acted in accordance with published procedures; a submission that the final decision reached in the appraisal process is perverse in light of the evidence submitted or that NICE has exceeded its statutory powers.

NICE and deliberative governance The work of NICE has to be set against a marked political concern with the appearance of transparency, accountability and lay influence in the field of health and health care. This contrasts with the ‘‘passive’’ citizenship and detailed centralised direction often associated with the NHS in the immediate post-war period (Offer, 1999; Webster, 1988) but also with the neo-liberal political rhetoric of the late 1980s and early 1990s that attempted to cast patients and citizens as ‘‘consumers’’ (Calnan & Gabe, 2001; National Health Service Management Executive, 1992). In the wake of the New Labour government’s election in 1997 citizens, patients and voluntary sector organisations have increasingly been portrayed as stakeholders or partners in the governance of health care at the level of service planning and provision (Baggott, 2005; Milewa, 2004) but also within related, more rarefied, spheres such as human genetics, research governance and health technology assessment (O’Donnell & Entwistle, 2004; Office of Science and Technology, 1999; Oliver et al., 2004). This emphasis upon active participation by various stakeholders in decisionmaking or advisory forums—in place of more episodic consultation or occasional exercises in involvement—appears to constitute a more inclusive, deliberative, approach to decision-making than that associated with competitive pluralist and elite paradigms of public sector governance (Haywood & Hunter, 1982; Lauman & Knoke, 1987; Walt, 1994). But broad characterisations of governance tell us little about the substantive nature of the relationship between experts and managers and the ‘‘lay’’ stakeholders now involved in deliberation about the admissibility of selected health technologies. Clearly—due to the focus upon clinical and costeffectiveness—NICE is not a simple aggregative democracy wherein the majority of ‘‘votes’’ determine decisions, regardless of available evidence. Nor is the organisation just a conduit for the expression of wider societal preferences via surveys of public opinion. The institutional concern with evidence-based decision-making does however suggest the elevation of reasoned deliberation over authority based purely on the mantle of expertise. More specifically, Cohen (1997) engages with Rawls (1999) concern with the development of decisionmaking institutions oriented to the common good, by arguing that deliberative modes of decisionmaking ideally reflect four procedural character-

ARTICLE IN PRESS T. Milewa / Social Science & Medicine 63 (2006) 3102–3112

istics. Two of these characteristics—that deliberative procedures should aim to maximise consensus and that consenting participants ideally regard themselves as bound by the outcome of deliberative procedures—appear prescriptive in that they seem to reflect a normative view on the form that deliberation should take. But two other characteristics seem more descriptive in nature and are thus of more use as analytical foci. First, it is argued that deliberative modes of decision-making reflect an emphasis upon reasoned argument that is conducted according to explicit procedures (rather than partisan conflict and simple majority-rule). Second, all participants can be regarded as equals in the deliberative sphere in the sense that ‘‘Everyoneycan put issues on the agenda, propose solutions, and offer reasons in support of or in criticism of proposalsyThe participants are substantively equal in that the existing distribution of power and resources does not shape their chances to contribute’’ (Cohen, 1997, p. 74). In other words, participants are equal due to the primacy given to the quality of reasoning rather than their social or professional status (Gutmann & Thompson, 2004, p. 7). NICE appears to reflect the two ‘‘ideal deliberative’’ criteria—reasoning according to explicit criteria and an equality of status among participants. In terms of reasoned argument, the appraisal process—as described above—is premised upon the systematic and transparent evaluation of evidence according to prescribed procedures. Indeed, as noted, appellants can challenge decisions on the grounds that they are perverse in the light of the evidence considered or that NICE has operated beyond its procedures and statutory powers. The extent to which all participants in the appraisal process can be depicted as equal in terms of deliberation is less straightforward. It does not follow that a focus on reasoned argument obviates factors that may be used to exclude or discriminate against particular actor-groups (such as the absence of recognised qualifications, perceived social status and ethnicity). There does however appear to be a conscious effort by NICE to ensure that a diverse range of stakeholders are heard and to foster a degree of parity among stakeholders. This effort is reflected in the consultative and evidentiary role of stakeholders but also in three forums associated with NICE. A quasi-independent Patient Involvement Unit was founded when NICE was established in order to provide less experienced stakeholder

3105

organisations with guidance and support in the preparation of submissions for the Appraisal Committee—a literal attempt to level the playing field for stakeholders in the appraisal process (National Institute for Clinical Excellence, 2004). In addition, a statutorily based ‘‘Partners Council’’—composed of individuals drawn from the NHS, the voluntary sector, professional bodies and commerce—convenes to consider an annual report that NICE is obliged to produce and to offer advice and comment. And, more recently, an influential Government report informed a decision by NICE to establish a ‘‘Citizens’ Council’’ in 2002. The ‘‘Kennedy Report’’, produced in the wake of a highly publicised medical scandal, took the view that ‘‘the perspectives of patients and of the public must be heard and taken into account wherever decisions affecting the provision of healthcare are made’’ (Kennedy 2001, pp. 456–457; National Institute for Clinical Excellence, 2002). The 30 members of the Citizens’ Council, selected from over 4000 respondents to a national advertising campaign, are asked to meet approximately every 6 months to deliberate upon topics suggested by NICE and to supplement technical debates on clinical and cost-effectiveness with public or social values that might, in time, be taken into account in the decision-making process (Davies, Wetherell, Barnett, & Seymour-Smith, 2005). It is not enough, however, simply to categorise observed reality against some ideal-typical characteristics of deliberative governance. The nature of any formal decision-making space—and, potentially, the relative influence of different interests— is not based solely on given procedural precepts but may also be shaped by implicit and explicit beliefs that shape deliberation. Deliberation may be imbued with potentially contentious assumptions about the capacities and motives of speakers, tacit beliefs about authority and expertise and other ‘‘givens’’ concerning different actor-groups (Parkinson, 2004; Stokes, 1998; Stokkom, 2005). Deliberation founded largely on scientific language and concepts, for example, might—when underpinned by ideas about who is and is not qualified to speak with authority and expertise—act to marginalise the views of participants not versed in relevant bodies of knowledge. And even a more accessible and inclusive form of deliberation that accommodates the expression of ideas in ‘‘lay’’ language might— through implicit understandings of who is truly ‘‘representative’’ of particular interests—facilitate

ARTICLE IN PRESS 3106

T. Milewa / Social Science & Medicine 63 (2006) 3102–3112

only a selective elite or neo-corporatist form of participation. This would contrast with a third, more broadly drawn, participatory form of deliberation that facilitated contributions from actors from a variety of backgrounds—expert and nonexpert; seasoned committee member and novice and large, well-resourced, organisations as well as those with just a few members (Skelcher, Mathur, & Smith, 2005, p. 579). But in any type of deliberative forum there is no reason to suppose that the demarcation and juxtaposition of all deliberative starting-points or givens is immutable. Indeed, the potential fluidity of boundaries and assumptions through which notions of expertise, valid knowledge and admissible evidence are maintained or challenged is a familiar concern in sociological studies of health and science-related policy (Brown et al., 2004; Gieryn, 1995; Prior, 2003). In applying these ideas to deliberation within NICE, we did not presume to second-guess or challenge sometimes very technical interpretations placed on the clinical and cost effectiveness of particular health technologies under consideration. Instead the focus here is upon deliberative boundaries and assumptions as they relate to a tenet of ideal-typical models of deliberation—the idea that participants in truly deliberative arenas are ‘‘equal’’ by virtue of an institutional commitment to reasoned (and, in this case, evidence-based) argument. Previous research suggests that patient and carer advocacy organisations are sometimes frustrated with the degree to which their concerns are taken seriously in the appraisal process (Quennell, 2003). And our own research has suggested that clinical and ‘‘professional’’ actors in the decisionmaking process within NICE appear to play a major role in debarring, reshaping or admitting more subjective or experientially based forms of evidence submitted by patients and carers (Milewa & Barry, 2005). So, with reference to the perceptions of stakeholders and the conduct of relevant meetings, do deliberative parameters and assumptions within the appraisal process overseen by NICE act to structure debates such that the formal equality of participants is brought into question? Methods A review of documents and informal discussion with a range of stakeholders was used to generate themes for a semi-structured interview schedule and to initiate a purposive snowball sample that

eventually generated 33 interviews. Core themes in the schedule included the recruitment (and exclusion) of participants in health technology appraisals orchestrated by NICE; patterns of interaction within key meetings; the framing and prioritisation of evidence and perceived deliberative boundaries and assumptions in the appraisal process. Informants included expert witnesses and consultees (called by the Appraisal Committee) from a ‘‘lay’’ background (n ¼ 7). These included representatives from a small advocacy group concerned with women’s reproductive health and the Director of a national patient advocacy body in the field of multiple sclerosis. Expert witnesses/consultees with ‘‘professional’’ backgrounds included a nurse specialist in multiple sclerosis, an individual from the College of Occupational Therapists and a research specialist in obstetrics and gynaecology (n ¼ 9). Other interviewees included employees of NICE closely involved with the appraisal process (n ¼ 6) and members of the Appraisal Committee (n ¼ 7). The latter included health economists, clinicians, a lay (non-clinical) member and a Chief Executive of a Primary Care Trust (a local body responsible for planning and providing most primary health care services and commissioning some aspects of secondary and social care). The remaining interviewees encompassed individuals from the independent groups commissioned to collate evidence as a basis for appraisals (n ¼ 2) and personnel from relevant sections within the Department of Health and NHS, such as a manager in an NHS purchasing consortium (n ¼ 2). The interviews—lasting an average of approximately one hour each—were tape-recorded and transcribed. In a second respect, discussions with staff from within NICE (including the Chair of the Appraisal Committee and the Director of Health Technology Appraisals) were used to identify broadly ‘‘typical’’ meetings within NICE that could be observed in the period available. The four meetings included an information forum convened by NICE for those consultees it had identified as appropriate for an impending appraisal of cannabinoids in the symptomatic treatment of multiple sclerosis; a meeting of the Appeal Panel concerned with objections to a decision not to recommend wider use of a growth hormone and two meetings of the Appraisal Committee. One of these meetings centred on appraisal of a drug for the treatment of aggressive non-Hodgkin’s lymphoma—the other covered a surgical procedure for endometrial ablation (surgery

ARTICLE IN PRESS T. Milewa / Social Science & Medicine 63 (2006) 3102–3112

for heavy menstrual bleeding) and drug treatment for rheumatoid arthritis. In all instances, except the meeting of the Appeal Panel, data capture centred on non-participant observation; contemporaneous annotation; post hoc formal and informal discussion; a review of written submissions by interested parties and study of meeting minutes. Access to the Appeal Panel meeting was granted on the basis that contemporaneous notes were not made—all accompanying documentation, including meeting papers and minutes were however made available and subsequently examined. The interview transcripts, notes on meetings and records of informal discussions were examined with initial reference to a coding frame based on the themes within the semi-structured interview schedule and then—on an iterative basis—in relation to more detailed themes and sub-themes that emerged through re-reading of the texts and codes. Pencilbased codings were then digitised, refined and used in topic-specific analyses. This discussion draws on data concerned with the implicit or explicit deliberative boundaries and assumptions perceived by informants in relation to considerations by the key decision-making body within NICE—the Appraisal Committee. The parameters of deliberative governance Official documentation suggested that the appraisal of health technologies by NICE was based on dispassionate but rigorous interrogation of data on clinical and cost effectiveness together with supplementary and contextual information provided by a broadly drawn range of stakeholders. But interview data and observation of selected meetings illustrated the social, interactive, nature of deliberative decision-making—a characteristic that at least qualifies the positivist ethos with which health technology appraisal has come to be associated (Leys 2003). More specifically, four deliberative assumptions/ parameters associated with the appraisal process were marked by a high degree of consensus or no demonstrable contradiction when tested against other data derived from interviews, observation of meetings and a review of associated documents. First, at a fundamental level, a handful of informants—including an Appraisal Committee member, a NICE employee and a member of the Public Involvement Unit attached to NICE—highlighted a caution towards over-reliance upon decision-analytic models. The latter encompass

3107

explicit and replicable procedures and techniques for the synthesis and evaluation of evidence relating to the costs and outcomes of health technologies (Philips et al., 2004). But collective discussion and reflection by the Appraisal Committee was also seen to encompass more porous and broad-ranging points raised by participants. The legitimacy of such reflection was perceived to rest on the expertise of committee members—a flexible, negotiated, social rationality rather than a wholly formal rationality. This approach was acknowledged but not always supported by informants: When a body like NICE or any decision-making body looks at a [decision-analytic] model they usually say ‘‘It’s too simplistic’’yThere’s a whole range of criticisms and they are valid criticismsyHowever, the implication of that is usually that somehow an alternative way of making a decision can eliminate the flaws of the modelyI don’t believe that people sitting around talking can do better than the modelyAll [NICE] can do is say ‘‘We got twenty people sitting around this room—including some health economists, some GPs, some hospital people, some people representing NGOs (whatever)—and because they are all there and we had a two-hour discussion, somehow this is a good decision’’. (Appraisal Committee member, ap23:1,2) But in a second respect—as two members of the Appraisal Committee suggested—this emphasis upon flexible deliberation by committee members did have a very specific parameter. Notions of costeffectiveness employed by the Appraisal Committee were consciously framed in terms of statutory sector expenditure. The ‘‘hidden’’ costs of informal care for patients, their carers and families tended to be overlooked: The big problem that I feel isn’t tackled is social costs. In the economics they only take into account costs to the NHSy. And sometimes what actually is expensive is the social care or the fact that this person cannot get a job because their current regimen of medication is so complicated or makes them feel so ill they can’t workyAnd if we got them better, they’d get back to work. (Appraisal Committee member, ap17–15) In a third regard, the nature of deliberation was coloured, according to two Appraisal Committee

ARTICLE IN PRESS 3108

T. Milewa / Social Science & Medicine 63 (2006) 3102–3112

members and a NICE employee, by a perception that patient groups, manufacturers and expert witnesses hardly ever argued against health technologies (one very rare example of a patient group being hostile to a technology centred on criticism by mental health service users of electroconvulsive therapy). Deliberation, in the context of health technology appraisals, was thus marked, apparently, by a gate-keeping role on the part of NICE rather than disinterested open dialogue between participants: The [patient advocacy groups] will come along and they will say ‘‘Our members with [a serious degenerative disease] badly need this drug’’. But the problem is that nobody represents the ones who aren’t thereyIt’s very difficult for a patient to suggest that there’s a trade-off between different groups—but there is. And the fact that patients there are always positive towards the technology means it’s almost a waste of time them being there. (Appraisal Committee member, ap23–11) And in a fourth, related, respect, about a quarter of informants (including four of the seven lay expert witnesses) suggested that ‘‘non-professional’’ witnesses were more likely to have their credibility or legitimacy questioned openly—a challenge to the idea of formal equality in the deliberative process: We were talking about hysterectomy because so many women are still told that ‘‘While we’re there we might as well pop your ovaries out’’yThe doctor who was doing the presentation [to the Appraisal Committee] was really quite angry. He just turned round and said ‘‘That’s not true’’, really forcefullyyThere was this kind of hushed silenceyAnd the Chair said ‘‘Oh, of course we don’t take it seriously’’yin acknowledging the GPs anger. You know, I suppose making reassuring noises. And I said ‘‘Are you saying the women are lying?’’yThey were disbelieving our callers and just, you know, completely disempowered us. (Expert witness from patient advocacy group, ep4-30,31) Some of the interaction within the meeting of the Appeal Panel that was observed illustrates this partial, human, element of dialogue and deliberation: [An expert from a professional medical association] was very correct in all his procedures—

remembering to address his comments to the Chairman, asking the Chairman politely if he could respondyThis seemed to give him enough space to say as much as he wanted, which wasn’t the case with everyoney[Another expert from a professional medical association] tried to make a point by raising his hand or just starting to talk— he was silenced on two occasionsyby [the Chair]. [The Chair] often did not allow various appellants to speak or interrupted them to move things along. [The Chair] never did this to representatives from the Appraisal Committee. (Appeal Panel, observation 4–9) That this sentiment and the other findings suggest a distinct social and partial element in the deliberative process is hardly surprising. No dialogue can be wholly removed from wider contexts of interaction, imbalances in knowledge and different degrees of acquaintance with conversational conventions. But the findings also point to a fundamental distinction within deliberative approaches to the formulation of decisions and policy—that between ‘‘deliberative democracy’’ and ‘‘democratic deliberation’’. As I argue below, the extent in which this distinction is acknowledged and addressed will have marked implications for the substantive nature of attempts to broaden involvement in decision-making within public sector bodies such as NICE. The politics of deliberative governance The study—admittedly based on a purposive sample surveyed over a relatively short period— highlighted a reliance upon expert judgement rather than formal decision-making models; a downplaying of private or social costs in calculations of the cost-effectiveness of health technologies; some cynicism among ‘‘insiders’’ regarding the objectivity of patients and carers and variation in the opportunities given to different parties to communicate their points. Taken at face value, the apparent role of these assumptions and boundaries in elevating ‘‘professional’’ judgement simply reflect the specialised training, recognised qualifications, shared methodological precepts and professional collegiality of clinicians, health economists and health service managers. But NICE does not exist in a vacuum. The organisation sits within a system of health care that, like all others, is imbued with specific (often conflicting) attitudes and aspirations around equity, social justice, personal responsibility

ARTICLE IN PRESS T. Milewa / Social Science & Medicine 63 (2006) 3102–3112

and the role of the state (Marshall, 1963; Walzer, 1983). There thus exist, echoing Bourdieu, conditions for a conscious and unconscious deliberative ‘‘struggle between agents to influence each other’s perceptions of their respective positions and, more generally, their perceptions of the reality as a whole’’ in relation to evidence, equity and need pertaining to particular health technologies (Contandriopoulos, 2004, p. 322). This was clear, for example, in the way in which evidence concerning the emotional costs and private or familial burden of certain illnesses was often downplayed in favour of a concern with public expenditure. Similarly, we saw that there was overt conflict between representatives of a patient advocacy group and clinicians in relation to anecdotal evidence concerning the attitude of doctors towards informed consent for the removal of ovaries. Both instances involved the drawing of boundaries—attempts to demarcate admissible from inadmissible evidence and reliable from unreliable testimony. Such implicit or explicit ‘‘boundary politics’’ can be used by clinicians, patient group advocates, health economists and any other form of stakeholder in attempts to privilege some forms of evidence over others and to legitimise or discredit the testimonies of particular speakers. Specific instances of such processes are relatively easy to discern. But there are two broader, more widely applicable, foci of boundary politics implicit in the findings—the idea of transparent reasoning and that of domain competence. Transparent reasoning in decision-making appears, at first, self-explanatory. As Rawls (1997) suggests, citizens/consumers in democratic polities who are coerced (through taxation) to fund public sector institutions and who depend upon attendant services can, at the very least, expect that the reasoning that underpins institutional decisions is rigorous, open to some scrutiny and that those ultimately responsible can be held to account with reference to their actions. Indeed, this imperative has become more explicit in recent years through political emphasis in countries such as the United Kingdom, Australia and the United States upon transparent demonstrations that public expenditure, trust and deference are warranted and should be continued (Goldsmith & Eggers, 2004; Kelly, Mulgan, & Muers, 2002; O’Flynn, 2005). But this does not entail a simple decision on the part of organisations such as NICE to render their deliberations transparent. Instead, transparent reasoning can be regarded as a ‘‘boundary object’’—an idea or

3109

focus that may be captured, utilised or ignored by different interests for different purposes. Stakeholders within public sector institutions such as NICE will of course vary in their proclivities and capacities with regard to this boundary object. But if, as we have seen in this case, transparent reasoning is linked to a very particular role (the formulation of recommendations based on the critical interpretation of the results of randomised controlled trials, statistical models, grey literature and patient testimonies), those who lay greatest claim to the function in question may exercise a disproportionate influence. In other words, transparent reasoning is a lexical terrain that can be colonised by the language and implicit understandings of particular groups at the expense of other interests. A second object of boundary politics in the shaping of deliberative givens is that of domain competence. The National Institute for Health and Clinical Excellence is part of a publicly funded health service that, despite at least two decades of political rhetoric about managerial devolution and community empowerment, remains largely subordinate to the whims of central government. However, the apparently straightforward control or competence exercised by the state is opaque, malleable and susceptible to challenge. The state’s impact upon deliberative arenas such as that within NICE has to be considered ‘‘in terms of its structural coupling and co-evolution with a wider set of institutions and social practices’’—institutions and practices that might not be immediately subordinate to the elected and non-elected spheres of the state (Jessop, 2004, p. 51). Examples in this context include the relationship between the state and British medical profession. The latter obtained and retains a remarkably high degree of self-regulation and thus considerable autonomy in influencing the relationship between ‘‘professional’’ and ‘‘lay’’ actors (Salter, 2004). Accordingly, even the most elaborate mechanisms and procedures designed by the state to ensure a plurality of voices in domains such as the planning of health care and health technology appraisal do not guarantee equality on the part of discussants in relation to the tenor and parameters of debate. Indeed, the reality of a-symmetries between actorgroups in relevant knowledge, resources, motivation, access to decision-makers and political leverage remains a feature of the ‘‘lay involvement’’ literature (Milewa, Harrison, & Dowswell, 2003; Skelcher, 1993). Such a-symmetries may influence

ARTICLE IN PRESS 3110

T. Milewa / Social Science & Medicine 63 (2006) 3102–3112

the deliberative assumptions, repertoires and contests evident within contrived arenas such as the Appraisal Committee. This was evident, for example, in the comments recorded above from an Appraisal Committee member who argued that one set of stakeholders (patient advocacy groups) ‘‘are always positive towards the technologyyit’s almost a waste of time them being there.’’ We thus have to look outward towards broader relationships between stakeholders—beyond specific deliberative domains—as well as inward if the parameters and assumptions that appear to underpin particular decision-making spheres are to be understood. The relationship between deliberative boundary objects, such as transparent reasoning and domain competence, and the nature of deliberation is obviously fluid in nature. But from a broader perspective we can think in terms of two extremes when gauging the nature of deliberative governance in the field of health policy and beyond. Democratic deliberation involves conscious institutional engagement with a broad variety of stakeholders, relatively easy access by these stakeholders to decisionmaking forums, explicit procedures for deliberation, detailed rationales for eventual decisions and, perhaps, scope to appeal against outcomes. NICE appears to reflect all these procedural characteristics. But as the findings suggest, democratic deliberation does not necessarily amount to deliberative democracy—a scenario premised upon the involvement of relevant stakeholders, participation by these stakeholders in the deliberative process, rationales for decisions but also the explicit negotiation, understanding and acceptance by all parties of the deliberative assumptions and parameters that anchor debate and orientate eventual decisions. The extent to which such factors are foregrounded and explicit may influence the relative capacities of different interests to challenge or negotiate the latent procedures, norms and modes of interaction that underpin deliberative arenas. Such challenge and negotiation—or its absence— may in turn be central to orienting dialogue and reflection towards democratic deliberation, deliberative democracy or essentially closed, elitecentred, decision-making. Conclusion This article drew on a study of decision-making with regard to the admissibility of health technologies in order to gauge whether deliberative para-

meters and assumptions therein privilege some parties over others by shaping the nature of debate. Relevant influences included an overt preference for ‘‘expert judgement’’ over reliance on formal decision-analytical tools; an approach to cost-effectiveness that privileged the implications for public expenditure over private or familial costs; intimations that patients and carers are rarely objective with regard to the value of health technologies and some variation in opportunities given to different actors to voice their opinions in a supposedly neutral deliberative arena. Such factors will inevitably vary in terms of their nature and implications over time and between deliberative arenas. But findings suggest that these assumptions and parameters—particularly those pertaining to the contestable ‘‘ownership’’ of the ideas of transparent reasoning and domain competence—can reflect and shape relationships of influence and marginality among participants. Accordingly, more inclusive, sustained and transparent deliberative approaches to health policy governance—epitomised by the National Institute for Health and Clinical Excellence in the United Kingdom—raise analytical challenges that go beyond consideration of issues such as access, representation and voice within decision-making forums. An empirical focus upon these issues is important but gives little indication of whether democratic deliberation (based upon an inclusive approach to participation and the expression of opinions) amounts to deliberative democracy wherein the ‘‘givens’’ that underpin deliberation are explicit to all parties and reflect a degree of consensus. The fact that such consensus is, at best, likely to be fleeting when decision-making coincides with conflicting demands and finite resources accentuates the significance of these parameters and assumptions to the relative influence exercised by different actor-groups in the governance of health policy.

Acknowledgements The co-operation of NICE, informants and their organisations is acknowledged with gratitude as are comments from anonymous reviewers and funding from Brunel University. Dr Christine Barry, formerly of Brunel University, played a major role in the collection of data. The content of the article and the interpretations expressed therein are the responsibility of the author.

ARTICLE IN PRESS T. Milewa / Social Science & Medicine 63 (2006) 3102–3112

References Abacus International. (2005). NICE guidance implementation tracking: Data sources, methodology and results, 27 January 2005 (report). London: National Institute for Clinical Excellence. Appleby, J., & Alvarez Rosete, A. (2003). The NHS: Keeping up with public expectations? In A. Park, J. Curtice, K. Thomson, L. Jarvis, & C. Bromley (Eds.), British social attitudes: The 20th report: Continuity and change over two decades. London: Sage. Baggott, R. (2005). A funny thing happened on the way to the forum? Reforming patient and public involvement in the NHS in England. Public Administration, 83(3), 533–551. Baggott, R., Allsop, J., & Jones, K. (2005). Speaking for patients and carers: Health consumer groups and the policy process. Basingstoke: Palgrave Macmillan. Brown, P., Zavestoski, S., McCormick, S., Mayer, B., MorelloFrosch, R., & Altman, R. (2004). Embodied health movements: New approaches to social movements in health. Sociology of Health and Illness, 26(1), 50–80. Byrne, D. (2004). Evidence-based: What constitutes valid evidence? In A. Gray, & S. Harrison (Eds.), Governing medicine: Theory and practice. Maidenhead: Open University Press. Calnan, M., & Gabe, J. (2001). From consumerism to partnership? Britain’s National Health Service at the turn of the century. International Journal of Health Services, 31(1), 119–131. Chalmers, T. (1974). The impact of controlled trials on the practice of medicine. Mount Sinai Journal of Medicine, 41(6), 753–759. Cohen, J. (1997). Deliberation and democratic legitimacy. In J. Bohman, & W. Rehg (Eds.), Deliberative democracy: Essays on reasoning and politics. Cambridge, MA: MIT Press. Contandriopoulos, D. (2004). A sociological perspective on public participation in health care. Social Science & Medicine, 58(2), 321–330. Davies, C., Wetherell, M., Barnett, E., & Seymour-Smith, S. (2005). Opening the box: Evaluating the citizens council of NICE (Report). Milton Keynes: School of Health and Social Welfare, Open University. Department of Health. (1997). The new NHS: Modern, dependable, Cm. 3807. London: Stationary Office. Department of Health. (2004). Choosing health: Making healthy choices easier, Cm. 6374. London: Stationary Office. Elwyn, G., & Edwards, A. (2001). Evidence-based patient choice? In A. Edwards, & G. Elwyn (Eds.), Evidence-based patient choice: Inevitable or impossible?. Oxford: Oxford University Press. Fox, J. (2005). The role of the expert patient in the management of chronic illness. British Journal of Nursing, 14(1), 25–28. Fung, A., & Wright, E. O. (2003). Countervailing power in empowered participatory governance. In A. Fung, & E. O. Wright (Eds.), Deepening democracy: Institutional innovations in empowered participatory governance (Real Utopias Project, vol. IV). London: Verso. Gibbons, R., Smith, S., & Antman, E. (2003). American College of Cardiology/American Heart Association clinical practice guidelines, part II: Evolutionary changes in a continuous quality improvement project. Circulation, 107(24), 3101–3107.

3111

Gieryn, T. (1995). Boundaries of Science. In S. Jasanoff, G. Markle, J. Petersen, & T. Pinch (Eds.), Handbook of science and technology studies. Thousand Oaks: Sage. Goldsmith, S., & Eggers, W. (2004). Governing by network: The new shape of the public sector. Washington, DC: Brookings Institution. Gutmann, A., & Thompson, D. (2004). Why deliberative democracy?. Princeton, NJ: Princeton University Press. Hajer, M. (2003). A frame in the fields: Policymaking and the reinvention of politics. In M. Hajer, & H. Wagenaar (Eds.), Deliberative policy analysis: Understanding governance in the network society. Cambridge: Cambridge University Press. Harrison, S. (1998). The politics of evidence based medicine. Policy and Politics, 26(1), 15–31. Haywood, S., & Hunter, D. (1982). Consultative processes in health policy in the United Kingdom: A view from the centre. Public Administration, 60(2), 143–162. Hill, S., Garattini, S., Loenhout, J. van, O’Brien, B., & Joncheere, K. de (2003). Technology Appraisal Programme of the National Institute for Clinical Excellence: A review by WHO (report). Copenhagen: WHO Regional Office for Europe. Innes, J., & Booher, D. (2003). Collaborative policymaking: Governance through dialogue. In M. Hajer, & H. Wagenaar (Eds.), Deliberative policy analysis: Understanding governance in the network society. Cambridge: Cambridge University Press. Jessop, B. (2004). Multi-level governance and multi-level metagovernance. In I. Bache, & M. Flinders (Eds.), Multilevel governance. Oxford: Oxford University Press. Kelly, G., Mulgan, G., & Muers, S. (2002). Creating public value: An analytical framework for public service reform (report). London: Cabinet Office Strategy Unit. Kennedy, I. (2001). Learning from Bristol: The report of the public enquiry into children’s heart surgery at the Bristol Royal Infirmary 1984– 1995 (The Kennedy report), Cm. 5207(1). London: The Stationary Office. Lauman, E., & Knoke, D. (1987). The organizational state: Social choice in national policy domains. Madison: University of Wisconsin Press. Levine, P., Fung, A., & Gastil, J. (2005). Future directions for public deliberation. In J. Gastil, & P. Levine (Eds.), The deliberative democracy handbook: Strategies for effective civic engagement in the 21st Century. San Francisco: Jossey-Bass. Leys, M. (2003). Health care policy: Qualitative evidence and health technology assessment. Health Policy, 65(3), 217–226. Marshall, T. H. (1963). Citizenship and social class. In T. H. Marshall (Ed.), Sociology at the crossroads and other essays. London: Heinemann [1950]. Milewa, T. (2004). Local participatory democracy in Britain’s health service: Innovation or fragmentation of a universal citizenship? Social Policy and Administration, 38(3), 240–252. Milewa, T., & Barry, C. (2005). Health policy and the politics of evidence. Social Policy and Administration, 39(5), 498–512. Milewa, T., Harrison, S., & Dowswell, G. (2003). Public involvement and democratic accountability in primary care organisations. In B. Dowling, & C. Glendinning (Eds.), The new primary care: Modern, dependable, successful?. Buckingham: Open University Press. National Health Service Management Executive. (1992). Local voices: The views of local people in purchasing for health (report). London: National Health Service Management Executive.

ARTICLE IN PRESS 3112

T. Milewa / Social Science & Medicine 63 (2006) 3102–3112

National Institute for Clinical Excellence. (2002). Applicants’ Introduction to the Citizens Council (document). London: National Institute for Clinical Excellence. National Institute for Clinical Excellence. (2004). A report on a Study to Evaluate Patient/Carer Membership of the first NICE Guideline Development Groups (report). London: National Institute for Clinical Excellence. National Institute for Health and Clinical Excellence. (2005). A guide to NICE (document). London: National Institute for Health and Clinical Excellence. Newman, J., Barnes, M., Sullivan, H., & Knops, A. (2004). Public participation and collaborative governance. Journal of Social Policy, 33(2), 203–223. O’Donnell, M., & Entwistle, V. (2004). Consumer involvement in decisions about what health-related research is funded. Health Policy, 70(3), 281–290. Offer, J. (1999). Idealist thought, social policy and the rediscovery of informal care. British Journal of Sociology, 50(3), 467–488. Office of Science and Technology. (1999). The advisory and regulatory framework for biotechnology: Report for the government’s review. London: Office of Science and Technology. O’Flynn, J. (2005). Adding Public Value: A new era of contractual governance? (Conference paper). Public administration and management conference, University of Nottingham, 5–7 September 2005. Oliver, S., et al. (2004). Involving consumers in research and development agenda setting for the NHS: Developing an evidence-based approach (report). Health Technology Assessment, 8(15). Parkinson, J. (2004). Why deliberate? The encounter between deliberation and new public managers. Public Administration, 82(2), 377–395. Petticrew, M. (2001). Systematic reviews from astronomy to zoology: Myths and misconceptions. British Medical Journal, 322(7278), 98–201. Philips, Z., et al. (2004). Review of guidelines for good practice in decision-analytical modelling in health technology assessment (report). Health Technology Assessment, 8(36).

Prior, L. (2003). Belief, knowledge and expertise: The emergence of the lay expert in medical sociology. Sociology of Health and Illness, 25(3), 41–57. Quennell, P. (2003). Getting a word in edgeways? Patient group participation in the appraisal process in the National Institute for Clinical Excellence. Clinical Governance, 8(1), 39–45. Rawls, J. (1999). A theory of justice (revised edition). Oxford: Oxford University Press. Rawls, J. (1997). The idea of public reason. In J. Bohman, & W. Rehg (Eds.), Deliberative democracy: Essays on reasoning and politics. Cambridge, MA: MIT Press [1993, 1996]. Salter, B. (2004). The new politics of medicine. Basingstoke: Palgrave Macmillan. Secretary of State for Wales. (1998). NHS Wales: Putting patients first, Cm 3841. London: Stationary Office. Skelcher, C. (1993). Involvement and empowerment in local public services. Public Money and Management, 18(3), 13–20. Skelcher, C., Mathur, N., & Smith, M. (2005). The public governance of collaborative spaces: Discourse, design and democracy. Public Administration, 83(3), 573–596. Stokes, S. (1998). Pathologies of deliberation. In J. Elster (Ed.), Deliberative democracy. Cambridge: Cambridge University Press. Stokkom, B. van (2005). Deliberative group dynamics: Power, status and affect in interactive policy making. Policy and Politics, 33(3), 387–409. Syrett, K. (2003). A technocratic fix to the ‘‘legitimacy problem’’? The Blair government and health care rationing in the United Kingdom. Journal of Health Politics, Policy and Law, 28(4), 715–746. Timmermans, S., & Berg, M. (2003). The gold standard: The challenge of evidence-based medicine and standardization in healthcare. Philadelphia, PA: Temple University Press. Walzer, M. (1983). Spheres of justice: A defence of pluralism and equality. Oxford: Blackwell. Walt, G. (1994). Health policy: An introduction to process and power. London: Zed Books. Webster, C. (1988). The health service since the war, volume I: Problems of health care: The National Health Service before 1957. London: HMSO.