Beyond Evidence Based Policy in Public Health

The Interplay of Ideas

K. Smith

• Publisher: Palgrave Macmillan
• Year: 2013
• ISBN:
  9781137026583
• Topic: Ciencias sociales
• Subtopic: Estudios de desarrollo global

This book explores the complex relationship between public health research and policy, employing tobacco control and health inequalities in the UK as contrasting case studies. It argues that focusing on research-informed ideas usefully draws attention to the centrality of values, politics and advocacy for public health debates.

1

The Fluctuating Fortunes of ‘Evidence-Based Policy’

[T]here is no such thing as evidence-based policy. Evidence is the six-stone weakling of the policy world. Even its most enthusiastic advocates are inclined to prefer the phrase ‘evidence-informed policy’ as a way of conveying a more authentic impression of research’s sway. [ … ] [T]he big problem [is] the four-hundred pound brute called politics.

Pawson (2006: p. viii)

Introduction: The struggle to move beyond ‘science versus politics’

Over the past decade, there have been regular media stories in which scientists are depicted either as embattled critics of politicians or as out-of-touch individuals whom politicians might be well advised to ignore. A classic example of this, encapsulating both characterisations, was the media coverage of the UK government’s decision to sack a scientist, Professor David Nutt, from its Advisory Council on the Misuse of Drugs (ACMD). This decision followed the publication of an article by Nutt in Guardian newspaper, in which he challenged the government’s decision to reclassify cannabis to a higher (more harmful) category (for more a more detailed account of the affair, see Henderson 2012; Monaghan 2011). A media storm quickly ensued with a Guardian editorial accusing the responsible minister, Alan Johnson, of lacking ‘the strength of character to listen to people who tell him difficult truths’ (The Guardian (Editorial) 2009), while a subsequent Daily Mail commentary dismissed Nutt as ‘barmy’ and ‘dangerous’ (Glover 2010).

Even for those sympathetic to the government’s position on cannabis, the decision to sack Nutt jarred uncomfortably with New Labour’s well-publicised commitments to ‘evidence-based policy’ (EBP) (Blunkett 2000; Cabinet Office 1999b, 2000) and Blair’s insistence that ‘what matters is what works’. Key figures in the scientific community were quick to publicly condemn the government for sacking Nutt, a response which the author of The Geek Manifesto argues is evidence that ‘cavalier ministerial treatment of scientific advice’ (Henderson 2012: p. 166) is increasingly being resisted by supporters of science (‘the geeks’ in Henderson’s terms).

There does indeed seem to be an emerging movement of actors keen to promote the utility of scientific evidence and to draw attention to any perceived misuses. This includes the recent growth in organisations aiming to promote a better understanding and use of scientific evidence, from blog sites such as Ben Goldacre’s ‘Bad Science’ (http://www.badscience.net/) and Henderson’s ‘Geek Manifesto’ (http://geekmanifesto.wordpress.com/) to organisations like Sense About Science (http://www.senseaboutscience.org/) and the health-focused Behind the Headlines (http://www.nhs.uk/news/Pages/NewsIndex.aspx). The common thread connecting these various organisations/websites is the emphasis that each places on the potential value of scientific evidence for decision-making. Indeed, both Henderson and Goldacre have contributed to publications that argue science, or evidence produced through the application of scientific methodologies, can provide clear answers to difficult political questions. For example:

Science matters when it has amassed good evidence that defines policy challenge, such as climate change, and the effects that different solutions might have. In many ways, it matters still more when the evidence base is weak and questions are more open. If ministers want to know how best to teach children to read, or how best to rehabilitate drug offenders, they can use the evidence of science to find out. [ … ] Politics asks tough questions of our leaders, but science provides a great tool for answering them.

(Henderson 2012: p. 9)

Randomised controlled trials (RCTs) are the best way of determining whether a policy is working. They are now used extensively in international development, medicine, and business to identify which policy, drug or sales method is most effective. [ … ] However, RCTs are not routinely used to test the effectiveness of public policy interventions in the UK. We think that they should be.

(Haynes et al. 2012: p. 4)

From this evidence-orientated perspective, ‘politics’ is often framed as a ‘barrier’ to rational decision-making or, in Pawson’s terms, ‘the four-hundred pound brute’ that quashes the ‘six-stone weakling’ that is evidence (Pawson 2006: p. viii). Yet, as Geoff Mulgan argues, in democratically elected countries, ‘the people, and the politicians who represent them, have every right to ignore evidence’ (Mulgan 2005: p. 224). Indeed, an acceptance of the multiple, valid influences on policy have led some involved in promoting the use of evidence in policy to conclude that it is more realistic to advocate for ‘evidence-informed’ rather than ‘evidence-based’ policy (see Boaz et al. 2008).

Such compromises have provoked significant consternation among some involved in efforts to increase the use of evidence in decision-making. Pawson, for example, dismisses ‘evidence-informed policy’ as ‘a horrible expression, all thin-lipped, prissy and politically correct’ (Pawson 2006: p. viii), while Sanderson argues that ‘advocacy of a more “realistic” position of “evidence-informed policy making” conflates description with prescription, legitimises the status quo and potentially undermines our ambitions for stronger use of evidence in policy making’ (Sanderson 2009: p. 700). One of the problems with placing such a central focus on the role of evidence in policy, however, is that it makes it difficult to conceive of politics as anything other than a stumbling block for evidence. Hence, despite Sanderson’s (2006, 2009) call for pragmatic, ‘intelligent policymaking’, ‘evidence’ and ‘politics’ continue to be commonly depicted as opposing forces, locked in an endless battle (e.g. Monaghan 2011; Pawson 2006).

The existence of conflicts between evidence-based and ideological (or populist) approaches to politically contentious issues, such as illicit drugs, is widely recognised (again, see Monaghan 2011). However, for policy issues in which there seems to be rather more of a consensus about the overarching objectives, it seems less obvious how or why ‘politics’ might obstruct the use of evidence within policymaking. Public health is precisely such an issue and, indeed, the majority of civil servants and politicians in a post-1997 UK context have (rhetorically, at least) signed up to taking an evidence-based approach to improving population health and reducing health inequalities, an agenda which is strongly supported by many leading public health researchers (e.g. Macintyre 2011). The existence of such a cross-sector consensus suggests public health might be one area in which evidence-based (or at least evidence-informed) policy and practice are feasible. Yet, disappointingly, most assessments of public health policies continue to conclude that they are not evidence based (e.g. Bartley 1994; Hunter 2009; Katikireddi et al. 2011; Macintyre et al. 2001; Smith 2007).

A popular explanation for this disjuncture is that it results from communicative, institutional and cultural gaps between researchers and policymakers. In other words, the gaps between research and policy are perceived to reflect divisions between those involved in producing research evidence and those involved in constructing policies (see, for example, Innvær et al. 2002; Mitton et al. 2007). Concern about these ‘knowledge to action gaps’, as they are commonly conceived (Wehrens, Bekker and Bal 2011), has been particularly acute in countries where the evidence-based movement has been most active, such as Australia, Canada, the Netherlands and the UK. Such concerns have stimulated efforts to develop ‘bridges’ that might help connect the ‘two communities’, including the creation of ‘knowledge broker’ posts and organisations and a growing support for the ‘co-production’ of research, in which decision makers are involved from an early stage (see Smith and Joyce 2012; Wehrens, Bekker and Bal 2011).

Yet, if we turn, instead, to the available literature concerning policy change within the political sciences, sociology and policy studies, the question becomes not, ‘why is public health policy not evidence-based?’ but ‘why would we ever assume it could (or even should) be?’ This is because the literature within these disciplines tends to highlight the normative, democratic and interest-driven aspects of politics and policymaking (see Smith and Katikireddi 2013). Sociological studies of science go further, demonstrating that research itself is also a value-laden activity (e.g. Bartley 1992; Knorr-Cetina 1981; Latour and Woolgar 1986). Yet, despite periodic acknowledgement of these issues within the public health community, there appears to be a reluctance to let go of the possibility that we might someday achieve rational, ‘evidence-based’ public health policies (e.g. Lavis et al. 2004; Macintyre et al. 2001; Orton et al. 2011). In many ways, this is understandable and simply reflects the desire of those working in public health research to have a positive impact on population health outcomes. Indeed, if public health research does not contribute to more informed policy decisions (and, ultimately, to better health outcomes), then the justification for the industrial levels of research currently being funded in this area seems questionable.

This book aims to weave together normative (political) and rational (evidence-focused) approaches to thinking about the relationship between evidence and policy in public health by focusing on two contrasting case studies: health inequalities and tobacco control. The central claims of the book, informed by over a hundred interviews and an analysis of hundreds of relevant documents, are as follows. First, the book argues that it is both more accurate and more useful, from the perspective of those trying to improve the use of evidence in policy, to focus on the influence of ideas (rather than evidence). Doing so highlights the malleability of ideas as they move between actors, without necessarily discarding the potential importance of evidence (which, for example, may play a role in making particular ideas more or less persuasive in some contexts). Second, by exploring researchers’ own accounts of the work that they do, the book highlights the relatively unexplored role of research funders in the relationship between evidence and policy and draws attention to the need to investigate how policy influences research (as well as the other way round). Third, in empirically exploring the interplay of ideas as they move between actors concerned with health inequalities and tobacco control, the book draws attention to the broad networks of actors involved in the translation of ideas including, for example, journalists, charities and advocacy organisations, community activists, public sector researchers, think tanks, knowledge brokers and politicians, as well academic researchers and civil servants. Finally, the book suggests research is likely to have more impact on policy where there are strong links between research, advocacy and policy communities and where researchers work with others to (i) develop charismatic ideas (within persuasive ‘policy frames’) and (ii) monitor and actively try to counter the efforts of actors seeking to influence policies in alternative ways (e.g. corporations involved in marketing health-damaging products or think tanks promoting free-market policies and minimal state intervention).

The remaining sections of this opening chapter briefly assess key accounts of EBP within academic and policy circles over the past 50 years. The chapter begins by summarising theories and models of knowledge transfer that emerged in the 1970s and early 1980s (e.g. Blume 1977; Bulmer 1982; Pahl 1977; Weiss 1977; 1979), before moving on to consider more recent studies (e.g. Lomas 2000; Nutley and Davies 2000; Petticrew et al. 2004; Walter, Nutley and Davis 2005). It then introduces some popular theories of policy change, contrasting the normative basis of these perspectives with the rational, instrumental conceptualisations of policymaking that are evident in many recent investigations into the relationship between evidence and policy. Finally, the chapter considers how a variety of more complex theories, including science studies, complexity theory and Foucauldian analyses, are beginning to be applied to accounts of the interplay between research and policy.

Overall, the chapter challenges many popular conceptualisations of the relationship between research and policy, arguing that we need to move beyond the idea that research (or, more narrowly, science) and policy are distinct domains that need to be ‘bridged’, or that ‘politics’ is a ‘barrier’ which science must overcome. It suggests, instead, that we need to get better at exploring how research, ethics, politics, values and democratic ideals are intertwined (see also Sanderson 2006, 2009). To do this, we need to think not only about the policymaking process and the individuals and institutions involved in constructing policy outputs but also about researchers and why they choose to focus on particular issues (and not others), how and for what reasons they construct research-informed messages for policy audiences and what happens to these ideas as they move in and around policy settings.

Theories and ideas emerging from efforts to link research and policy in the 1960s and 1970s

Many years before the mantra of EBP was adopted by policymakers in the UK, Australia, Canada and elsewhere (Cabinet Office 1999b, 2000; Canadian Academy of Health Sciences 2009; Rudd 2008), policymakers in the UK and the US were experimenting with efforts to improve the utilisation of research in policy (Weiss 1977, 1979, 1982). Policy efforts to achieve research-informed policy, back in the 1960s and 1970s, stimulated a wealth of academic literature exploring the relationship between science and policy in the 1970s and early 1980s. A variety of theories and ‘models’ of the relationship were put forward (e.g. Blume 1977; Bulmer 1982; Caplan 1979; Pahl 1977; Weiss 1977, 1979, 1982):

1. A knowledge-driven model in which research findings (i.e. knowledge) provide the necessary pressure for policy to develop in line with new knowledge.

2. A problem-solving model, in which a policy problem is first recognised, prompting research with the aim of providing evidence on which to base policy solutions.

3. A political model, where research is used in a predetermined manner to support policies that have already been considered for implementation for political reasons.

4. A tactical model, where research is used as a method of delaying the decision-making process, providing policymakers with some ‘breathing space’.

5. A two-community model, which depicts policymakers and academics as contrasting communities with different (often conflicting) values, reward systems and languages. This makes it difficult for academic research to inform policy in a meaningful manner.

6. An interactive model, in which research is just one factor among many that have the ability to influence policy (other important ones include political ideology, external pressures and the personal experiences of the policymakers involved).

7. An enlightenment model, in which research influences policy through indirect, diffuse processes over long periods, often contributing to a change in the way policy problems are framed and conceptualised, rather than addressing specific problems.

Models 1 and 2 summarise the simplest way of thinking about the relationship between research and policy, depicting a direct, linear connection in which policy-relevant research findings either drive policy change (the ‘knowledge-driven model’) or provide direct solutions to problems that policymakers have already identified (Davies, Nutley and Smith 2000; Weiss 1979). In other words, for both these models of the relationship, knowledge is utilised by policymakers in an ‘instrumental’ manner (Knorr 1977; Weiss 1980). Although there have been occasional examples of research feeding into policy in this manner, such simple models have long been discredited for failing to capture the intricacies of the actual relationship between research and policy (see Nutley and Davies 2000; Nutley, Walter and Davies 2007). The other five models/theories listed above provide alternative conceptualisations of the relationship between evidence and policy. While each is distinct, they are linked by the view that policymakers rarely utilise research in the direct, linear sense depicted by models 1 and 2.

The political model, for example, highlights the dominance of political values and ideologies within policy. From this perspective, research is only likely to play a role in policy if it is consistent with dominant ideological perspectives. This is precisely the approach to research that many scientists accused the UK government of taking following the ‘David Nutt affair’ described above (e.g. Henderson 2012). Indeed, despite New Labour’s commitment to EBP, there have been multiple analyses which suggest that a political model of the relationship between research and policy remains apparent, with ‘challenging’ research still being ignored, notably in the case of illegal drugs (e.g. Black 2001; Naughton 2005; Stevens 2007). Others, however, argue that such a model is overly cynical and simplistic for many policy issues, despite acknowledging the inherently political nature of policymaking (see Davies, Nutley and Smith 2000).

The ‘tactical model’ of the relationship suggests a similarly symbolic approach to evidence in which policymakers encourage/fund research activity with the aim of delaying awkward decisions or distracting attention (Davies, Nutley and Smith 2000). The English government’s decision to commission a post-2010 review of health inequalities evidence (Marmot 2010), at a time when it was becoming increasingly obvious that the government was not on track to meet its national health inequalities targets, could be interpreted in this way. However, it could also be interpreted as an expression of policymakers’ desire to learn from policy failures. The complexities and opaqueness of decision-making mean that it can be very hard to empirically ascertain how instrumental/political particular decisions actually were (Cohen, March and Olsen 1972; Donnison 1972).

A rather more structural, less political, approach to explaining the difficulties in achieving evidence-informed policy focuses on the differences between researches and policymakers. In 1979, Caplan famously outlined what he called a ‘two-communities’ thesis, which posits that the gap between research and policy results from a ‘cultural gap’ between researchers and policymakers (Caplan 1979). While not always referring to Caplan’s work directly, contemporary assessments of the limited use of research in health policy and practice frequently mirror Caplan’s observations by highlighting perceived ‘gaps’ between researchers, policymakers and/or practitioners as a fundamental barrier to the use of research (as mentioned earlier on p. 4 and discussed further in following section). As Lindquist (1990) points out, however, this way of thinking about the relationship excludes a range of potentially important actors, such as journalists, consultants and lobbyists.

In contrast, the interactive model, developed by Donnison (1972), evokes an image of policymaking in which decisions emerge out of highly chaotic and complex processes, with policymakers seeking (and receiving) information from a variety of sources, including practitioners, journalists and interest groups, as well as academic researchers. In this model, research ideas travel back and forth between a wide variety of groups, transforming as new connections and interactions occur, potentially developing into policy initiatives over time. Similarly, Cohen and colleagues’ (Cohen, March and Olsen 1972) ‘garbage can’ model of policy decision-making suggests that a range of actors and interests feed into an interactive process in which the outcomes are difficult to predict. These models suggest that complexity and disorganisation are hallmarks of the policymaking process, meaning that chance often plays a significant role in the outcomes. Indeed, Cohen and colleagues (1972) specifically describe decision-making within policy as a process of ‘organised anarchy’. From this perspective, research-b...

Table of contents

• Cover
• Title Page
• Copyright
• Contents
• List of Figures, Tables and Boxes
• Foreword by Malcolm Chisholm
• Foreword by Hilary Graham
• Acknowledgements
• List of Abbreviations
• 1. The Fluctuating Fortunes of ‘Evidence-Based Policy’
• 2. Evidence-Informed Policy in Public Health
• 3. The Power of Ideas (Over Evidence)
• 4. Institutionalised Ideas and Policy Facilitators
• 5. Critical/Charismatic Ideas and Advocates
• 6. Chameleonic Ideas and Flexian Policy Actors
• 7. Institutional Amnesia and the Rise of Public Health Knowledge Brokers
• 8. Politics and Advocacy in Public Health – A Way Forward
• References
• Index