‘Naloxone works’: The politics of knowledge in ‘evidence-based’ drug policy

Abstract

For over 20 years, drug policy experts have been calling for the wider availability of naloxone, to enable lay overdose witnesses to respond to opioid overdose events. However, the ‘evidence base’ for peer-administered naloxone has become a key point of contention. This contention opens up critical questions about how knowledge (‘evidence’) is constituted and validated in drug policy processes, which voices may be heard, and how knowledge producers secure privileged positions of influence. Taking the debate surrounding peer-administered naloxone as a case study, and drawing on qualitative interviews with individuals (n = 19) involved in the development of naloxone policy in Australia, we examine how particular kinds of knowledge are rendered ‘useful’ in drug policy debates. Applying Bacchi’s poststructuralist approach to policy analysis, we argue that taken-for-granted ‘truths’ implicit within evidence-based policy discourse privilege particular kinds of ‘objective’ and ‘rational’ knowledge and, in so doing, legitimate the voices of researchers and clinicians to the exclusion of others. What appears to be a simple requirement for methodological rigour in the evidence-based policy paradigm actually rests on deeper assumptions which place limits around not only what can be said (in terms of what kind of knowledge is relevant for policy debate) but also who may legitimately speak. However, the accounts offered by participants reveal the ways in which a larger number of ways of knowing are already co-habiting within drug policy. Despite these opportunities for re-problematisation and resistance, the continued mobilisation of ‘evidence-based’ discourse obscures these contesting positions and continues to privilege particular speakers.

Keywords

Carol Bacchi drug policy evidence-based policy knowledge naloxone problematisation

Introduction

The idea that naloxone could be distributed for administration by drug-using peers was first put forward in the medical literature in 1992 (Strang and Farrell, 1992). Naloxone is an antagonist which reverses the effects of opioids, including heroin, and has been used by medical practitioners for over 40 years. Given that naloxone has no abuse potential, clinical researchers expressed confidence that the possible risks associated with making naloxone available to peers would be ‘minimal’ and that ‘considerable benefit may accrue if drug users could give emergency doses of antagonist to fellow injectors who inadvertently overdose’ (Strang and Farrell, 1992: 1128). Since then, there have been repeated calls for the wider availability of naloxone (Bigg, 2002; Darke and Hall, 1997; Dietze and Lenton, 2010; Lenton et al., 2009a, 2009b; Strang et al., 1996, 2006). However, the question of the ‘evidence base’ for peer-administered naloxone became a key point of contention for the drug policy field. Was there sufficient ‘evidence’ to demonstrate the effectiveness of naloxone in the hands of lay overdose witnesses and thus provide an adequate rationale for implementing this intuitively appealing intervention? In the absence of a randomised controlled trial (RCT), did knowledge claims about the effectiveness of peer-administered naloxone satisfy the rigours of ‘evidence-based drug policy’? How could we know that naloxone ‘works’?

It is the assumptions underlying these questions that we seek to interrogate herein. We suggest that the contentious discussion surrounding the ‘evidence base’ for peer-administered naloxone opens up critical questions about how knowledge (‘evidence’) is constituted and validated in drug policy processes, which voices may be heard, and how knowledge producers secure privileged positions of influence (Bacchi, 2009; Lancaster, 2014). Taking the debate surrounding peer-administered naloxone as a case study for analysis, and drawing on qualitative interviews with individuals (n = 19) closely involved in the development of naloxone policy in Australia, this article critically examines the ways in which ‘particular kinds of knowledge’ (Bacchi, 2009: 240) are rendered ‘useful’ in drug policy debates. Following key themes explored in Bacchi’s (2009) poststructuralist approach to policy analysis, we seek to destabilise taken-for-granted truths implicit within the ‘evidence-based drug policy’ paradigm, consider the productive power of ‘evidence-based policy’ discourse, and in doing so, also take up Bacchi’s entreaty to consider how discourses may be regarded as resources for re-problematisation and resistance.

Background

Given its long history of use in emergency medicine, naloxone was unequivocally regarded as a ‘tried and tested product’ (Strang et al., 1996), but questions were raised about the possible effectiveness of this medicine in the hands of lay overdose witnesses and drug-using peers including the following: concerns about the shelf-life of the drug outside of medical settings; whether overdose would reoccur in the absence of medical follow-up due to naloxone’s short half-life; whether peers would be able to administer naloxone while intoxicated; whether the availability of naloxone would discourage ambulance call-outs; and, more controversially, whether the removal of the ‘deterrent effect of overdose’ would encourage opioid use (Darke and Hall, 1997; Lenton and Hargreaves, 2000; Strang et al., 1996: 1435). Throughout the 1990s, researchers and clinicians in the United Kingdom and Australia called for ‘trial and evaluation’ (Strang et al., 1996: 1435), and began ‘pre-launch’ studies to establish both the acceptability of naloxone distribution among people who inject drugs and the likely impact of such an intervention (Strang et al., 1999). While it was suggested that the best way to respond to remaining ‘uncertainties’ would be to conduct a controlled evaluation, it was acknowledged that ‘it may not be easy to conduct a randomized controlled trial of sufficient size to detect any effect of naloxone, should it occur’ (Darke and Hall, 1997: 1198) due to the low base population rate of overdose and difficulty ensuring that the naloxone was not used in the comparison group. Despite the methodological difficulties, researchers nonetheless asserted that ‘[i]t is only by a trial of naloxone distribution that empirical answers to the questions raised […] could be obtained’ (Darke and Hall, 1997: 1199).

By the early 2000s, some sectors of the drug policy field were questioning the ‘stalemate’ which had emerged in the naloxone debate, suggesting that the ‘fears, anxiety and ethical implications being voiced’ against peer-administered naloxone were ‘reminiscent of those expressed in the past’ regarding needle and syringe programmes (Oldham and Wright, 2003: 113). Describing preliminary positive outcomes observed through the Chicago Recovery Alliance’s programme, Bigg (2002) railed against the prevailing stasis: ‘Being opposed to a potentially lifesaving practice in the absence of data proving it wrong is a dangerous proposition’ (p. 678). In the absence of RCTs or systematic evaluations, researchers and clinicians began gathering together a body of observational evidence from a series of pilots and small-scale programmes. By February 2006, more than 900 opioid overdoses had reportedly been reversed using peer-administered naloxone, but the lack of formal evaluation precipitated still more calls for ‘structured, scientifically sound evaluations’ (Sporer and Kral, 2007: 175).

In 2009, Australian researchers stopped calling for trials and began advocating for swift implementation:

In our view, the international evidence clearly indicates that increased naloxone availability will prevent many cases of fatal overdose, that conducting a trial in Australia is now unnecessary, and that naloxone should be made available without delay. (Lenton et al., 2009a: 469)

However, not everyone agreed with these conclusions. As one clinician opined, ‘The only evidence for naloxone distribution so far comes from observational studies, which are considered among the weakest form of research design. The abundance of such studies does not compensate for their inherent lack of rigour’ (Wodak, 2013).

While momentum slowly built towards implementing programmes in Australia (Lancaster and Ritter, 2014b; Lenton et al., 2015), and Scotland adopted peer-administered naloxone as a funded public health policy in 2011 (Bird et al., 2015; McAuley et al., 2012), attempts to gather ‘conclusive’ RCT evidence continued in the United Kingdom through the establishment of the N-ALIVE trial (Strang et al., 2013). Despite the policy movement in Scotland, modelling showing the hypothetical cost-effectiveness of naloxone distribution (Coffin and Sullivan, 2013), and new reports from the United States of 10,171 overdose reversals through peer-administered naloxone programmes (Wheeler et al., 2012), those conducting the trial nonetheless argued that the ‘clarity of conclusions from a suitably well-designed and objectively undertaken randomized trial would be a powerful influence on key decision-makers’ (Strang et al., 2013: 989). But as researchers noted, the need to test ‘the extent to which this intervention actually reduces deaths’ necessarily ‘poses challenges for study design’ (Strang et al., 2014: emphasis added). In December 2014, 6 years after the Medical Research Council had awarded funding for the trial, it was announced that the N-ALIVE trial had stopped randomising participants and could not go ahead as planned. Investigators found that ‘twice as many people who are given naloxone-on-release use it to save another person’s life as to save their own’, meaning that the trial would have needed to be much larger than was practicable to render meaningful results (N-ALIVE, 2014).

While the quest for ‘enough’ ‘scientifically rigorous’ ‘gold star’ evidence continued, a grass-roots counter-movement emerged. In 2011, the Eurasian Harm Reduction Network and the Harm Reduction Coalition founded a social media campaign disseminated through Facebook, YouTube and campaign websites (Eurasian Harm Reduction Network, 2014, 2015; Harm Reduction Coalition, 2015; Open Society Foundations, 2013). Their message was a simple declaration in response to the ongoing quagmire of scientific debate: ‘I’m the Evidence! Naloxone Works!’ (Harm Reduction Coalition, 2015). The video on the landing page provided a campaign overview; a compilation of personal ‘success stories’ of overdose reversals, told directly to camera by people who use drugs. The compilation of short videos was introduced by a woman stating,

These are stories of people that have witnessed overdose, and most of them have been trained in the use of naloxone and able to administer it, saving the life of their friend, family member or a stranger. The scientific evidence is developing, that naloxone works. It is important that you also hear these stories in order to better see its impact. (Harm Reduction Coalition, 2015, emphasis added)

Approach

Our analysis takes as its starting point the poststructuralist work of Australian policy theorist, Carol Bacchi. A growing body of drug policy research has applied Bacchi’s (2009) ‘What’s the Problem Represented to be?’ (WPR) approach, illuminating the ways in which the problem of drugs is not fixed and stable but rather constituted in policy and practice (Bacchi, 2015; Fraser and Moore, 2011; Lancaster and Ritter, 2014a; Lancaster et al., 2015a, 2015b; Pienaar and Savic, 2015; Roumeliotis, 2014; Seear and Fraser, 2014). In doing so, this research has also begun to raise critical questions about the concept of ‘knowledge’ in drug policy, demonstrating how knowledge is both ‘central to the constitution of political problems’ and active in ‘the production and transformation of relations of power’ (Roumeliotis, 2014: 337).

The WPR approach takes a ‘sceptical stance toward “knowledge”’ (Bacchi, 2009: 249). The relationship between ‘knowledge’ and power, and the place of science in governing, are key themes explored in Bacchi’s work and the focus of our analysis here. Bacchi scrutinises the concept of ‘knowledge production’ in contemporary modes of governance and challenges the ‘problem-solving’ premise of the dominant ‘evidence-based policy’ paradigm. Following a Foucauldian governmentality perspective (Dean, 1999; Foucault, 1991; Rose et al., 2006), Bacchi contends that academic knowledge (what we might call ‘evidence’) is not merely an objective resource for governments to use in policy decision-making, but rather central to the way we are constituted and governed as political subjects. From this perspective, ‘knowledge’ is a contested concept with a range of political effects. The question of ‘who is best placed to produce “knowledge” that will count as “truth,” and how they secure their position/s of influence’ (Bacchi, 2009: 235) is of central concern. Both ‘the power of discourses’ (to limit what can be thought and said) and ‘the power to make and/or to deploy discourses’ (to determine who may speak authoritatively) must be considered (Bacchi, 2009: 236, emphasis original). We apply this perspective here to ‘evidence-based policy’ discourse.

‘Evidence-based policy’ developed out of the evidence-based medicine (EBM) movement which sought a rational approach to clinical decision-making through the application of ‘gold star’ research (Davies et al., 2000; Lin and Gibson, 2003). This movement is regarded as one of the most significant developments in medicine in the last two decades, and has dramatically affected policy and practice far beyond the scope of health (Bell, 2012). But as Harrison and Checkland (2009) observe, EBM and its subsequent expression in the broader notion of evidence-based policy and practice are political phenomena. The dominant model of ‘evidence’ espoused in contemporary medical discourse is one based on the primacy of particular research methods (systematic comparisons, randomisation of research subjects to control and intervention groups, and aggregation of research findings in the form of reviews and meta-analyses) as the source of valid and unbiased knowledge about the effectiveness of healthcare interventions, over and above other ways of ‘knowing’ (Harrison and Checkland, 2009). Such methods have come to form what is known as the ‘hierarchy of evidence’, with systematic reviews of RCTs occupying the peak of the pyramid. Throughout the 1990s, this approach became de rigueur in medicine as well as academic and policy circles (for further discussion, see Harrison and Checkland, 2009; Marston and Watts, 2003). The seemingly incontestable ‘self-evident value’ of using evidence as an ‘objective’ and ‘accountable’ basis for decision-making allowed ‘evidence-based’ discourse to ‘creep’ out of the scope of clinical medicine and into diverse policy domains (Bell, 2012: 314).

While ‘evidence-based’ discourse has been taken from the domain of medicine and applied more broadly to the pursuit of ‘evidence-based policy’ in Western democracies, the disciplinary and methodological roots of EBM have a number of implications (Marston and Watts, 2003). Two in particular are pertinent to this study. First, where there is a preference for a particular kind of ‘evidence’, and what counts as valid knowledge is narrowly conceptualised, other voices and knowledge(s) are necessarily marginalised. As Marston and Watts (2003) note, ‘[i]f knowledge operates hierarchically, we begin to see that far from being a neutral concept, evidence-based policy is a powerful metaphor in shaping what forms of knowledge are considered closest to the ‘truth’ in decision-making processes’ (p. 145). Second, given that public health and drug policy are closely related to medicine and the role of clinicians is prominent, evidence-based policy discourse holds significant purchase in these domains (Smith, 2013; valentine, 2009). However, whether or not the principles espoused within EBM and the ‘hierarchy of evidence’ are necessarily appropriate for responding to complex policy problems (like illicit drug use) remains contested given the strong normative and ethical dimensions of these issues (Smith, 2013). Nonetheless, given its prevailing dominance within drug policy, ‘evidence-based’ discourse is likely to significantly shape the limits of what can be thought and said and also who may speak authoritatively (Bacchi, 2009) and therefore deserves close analysis.

Method

In this article, we take up Bacchi’s challenge to use her approach not only for the interrogation of policy proposals, but also as a way to scrutinise theoretical stances more generally so as to highlight the ‘political implications of theoretical “knowledges”’ (Bacchi, 2009: 233, emphasis original). We focus on the sixth question in Bacchi’s (2009) approach, examining how assumptions implicit to ‘evidence-based policy’ discourse have been ‘produced, disseminated and defended’, and how they could be ‘questioned, disrupted and replaced’ (p. 48). The purpose of this question is to destabilise taken-for-granted ‘truths’ by directing attention to practices and processes which allow these ‘truths’ to dominate and achieve legitimacy, thereby opening up opportunities for challenge and resistance (see Bacchi, 2009: 19, 45).

We critically analysed a corpus of discourse drawing on 19 semi-structured in-depth interviews with policy makers, advocates, clinicians and researchers closely involved in discussions surrounding the establishment of peer-administered naloxone programmes in Sydney and Canberra, Australia.¹ Interviews were on average 1 hour in length (ranging from 30 minutes to over 2 hours). The interview guide focussed on eliciting participants’ perceptions of policy activity, the processes through which policy activity occurs, and the roles of multiple knowledge(s) and ‘voices’. Participants were also asked what information or knowledge they found compelling in the naloxone discussions. The term ‘evidence-based’ was not used by the interviewer. When this term did arise in an interview, it was spontaneous and introduced by the participant. The interviews were audio-recorded and transcribed verbatim. Participants were sent a copy of their transcript to review for the purposes of verifying accuracy, correcting errors or inaccuracies, and providing clarifications.

Transcripts were de-identified and then initially analysed thematically with the assistance of NVivo 10 software. This initial coding identified and categorised a range of knowledge(s) and voices at work within the naloxone discussions, as well as modes and locations of knowledge production including academic publication, evaluation structures, international naloxone programmes, and grass-roots advocacy. Each of these categories was then explored in more detail using Bacchi’s (2009: 37–38) Foucauldian-influenced approach, which turned our attention to the functioning and institutionalisation of discourses, as well as to the operations of power. Initial themes were refined accordingly, thus tracing the struggles over the production and deployment of knowledge(s) in naloxone discussions and interrogating the productive power of evidence-based policy discourse through attention to practices and processes. Underpinning this analysis was a critical assessment of the conceptual logics of evidence-based policy discourse. If, as Bacchi (2009) argues, ‘what we propose to do about something indicates what we think needs to change’ (p. xi) then implicit within the evidence-based policy paradigm are a number of assumptions. The call for drug policy to be ‘evidence-based’ is underpinned by the implicit assumption that a particular kind of knowledge (‘evidence’) should be taken up and used by policy makers to make sound policy decisions. Within this paradigm, policy makers are constituted as rational, authoritative ‘problem-solvers’; knowledge is constituted as instrumental and objective; and the ‘problems’ to be addressed are constituted as fixed and known, that is, ‘the “problem” against which “evidence” is to be “applied” is presumed to exist separately from deliberations about how to “solve” it’ (Bacchi, 2009: 107). In this way, ‘evidence-based policy’ discourse is productive and constitutive. It produces ‘truths’ both about the policy process and the kind of knowledge and voices which may be legitimately invoked to inform policy.

In the following sections we examine two related themes: (1) legitimating speakers and (2) resources for resistance.

Legitimating speakers

Evidence-based policy discourse constitutes instrumental and ‘objective’ knowledge as relevant for policy, thereby legitimating speakers with access to this particular kind of knowledge in policy processes. In this case study, we identified how evidence-based policy discourse legitimated the voices of researchers and clinicians, and gave force to knowledge claims made from these domains. From a poststructuralist perspective, power is analysed as ‘something which circulates […] It is never localised here or there, never in anybody’s hands’ (Foucault, 1980: 98). Thus, we are not suggesting that researchers and clinicians always occupy privileged positions (i.e. that they ‘possess’ power by virtue of their roles) but rather that the productive power of evidence-based policy discourse makes and produces researchers and clinicians as particular kinds of political subjects (see Bacchi, 2009: 37–38). That is, the legitimacy of their voices may be understood as one power-effect of evidence-based policy discourse. This effect is by no means fixed or consistent, however, as discourses are not homogeneous (Bacchi, 2009). By understanding power as productive, and not as possessed, our attention turns in this case study to how it operates, and the effects of legitimating particular speakers.

The need to publically present an ‘evidence-based’ argument generated a perception that particular kinds of people should be the ‘face’ and ‘voice’ of drug policy discussions:

You’ve got to keep it calm, have it evidence-based; you need reliable – someone like [names a public health professor] is perfect, this man of science with a beard. […] He said, ‘You know, if you want to get something politically across, and change the laws on an issue’ he said, ‘You look at when the hemp spokespeople get on television, they always have the dreadlocks, or unkempt beards and that’. […] He said, ‘What you want is someone out there with a certain attire, clean cut, saying, “Well, this is a health issue”’. And that’s what [this professor] was. […] He’s always logical, he’s unflappable, he’s not an advocate. […] You’ve got to do it softly softly, bit by bit, and evidence-based. (Researcher)

Here, taken-for-granted ‘truths’ about the kind of knowledge which is relevant for policy legitimates a particular kind of ‘rational’ speaker, and excludes other ‘less legitimate’ voices. In this researcher’s account, demeanour, appearance and institutional position are intertwined markers of legitimacy for policy debates, trumping any argument or knowledge the advocate may have put forward (indeed, in this account the voice of the advocate is silenced; we are offered an image of an advocate on television, but not what she or he may say). This particular professor is constituted as a ‘scientific’, ‘rational’ and ‘objective’ voice, and therefore a legitimate speaker in drug policy debates. One effect of this legitimating process is that other voices, by contrast, are constituted as spurious or less reliable:

I wasn’t sure that [the other clinician] was necessarily the sort of person I would have chosen. [The other clinician is] certainly a very passionate advocate, but as a sort of very sane, sage and sober doctor-type like myself, [the other clinician] kind of reeks of all of the things that make me nervous about drug policy, where it’s a lot of emotion and not a lot of hard evidence behind it. (Clinician)

Here, the participant describes herself as a ‘sane, sage and sober doctor-type’ and contrasts her own subject position with that of the ‘passionate advocate’. In doing so, this participant highlights the privileging of ‘objectivity’ and ‘rationality’ in evidence-based policy discourse, indicating that it is not enough to simply hold a medical position to be legitimate; one must also be ‘sane, sage and sober’ (i.e. ‘rational’ and objective’). This highlights the complexity of the subjectivities produced through evidence-based policy discourse and the way it legitimates particular speakers and excludes others, through appeals to rationality and objectivity.

As participants described the development and establishment of the peer-administered naloxone programmes in both Sydney and Canberra, the privileging of ‘objective’ and ‘rational’ perspectives over and above the knowledge brought to the process by drug user organisations² was a common theme. As noted by one participant, ‘consumers aren’t treated very well, nor are their ideas, and stuff tends to be shut down’. Although most participants (clinicians, researchers, policy makers and advocates alike) spoke of the importance of including the voices of people who use drugs (and their families and communities) in policy processes, there was a sense in which the involvement of researchers and clinicians was nonetheless required to legitimise processes and reassure decision makers. As one advocate noted, reflecting on a committee process which brought a group of stakeholders together,

I think that gave the government confidence that there was – it wasn’t just a drug user group coming up with a harebrained scheme. It was across the board of researchers, doctors, nurses, ambulance, drug user organisations, users themselves. (Advocate)

Without the legitimacy afforded by researchers, clinicians and other professional medical personnel, the drug user organisation’s proposal may not have been seen as authoritative in the eyes of government. Legitimating some voices to the exclusion of others had effects for participants engaged in these processes. As another participant recalled,

we were at a meeting and [a drug user organisation representative] would say something and somebody like [one of the researchers] would say, ‘Actually they’re correct’ and then the government would be reassured. (Advocate)

The de-legitimisation of the drug user organisation representative mentioned in this exchange was acutely felt by the individual involved, who in reflecting on this same process said,

They just speak differently to me. Like we’re sitting around in a meeting and people have differing views and I’m saying something and I’m being sort of being spoken to in this really horrible way and then being disagreed with and then someone else will speak up and then [the policy maker] will change their mind and go, ‘oh okay maybe I’ll consider that’.

Here, we see the ways in which evidence-based policy discourse places boundaries not only around what is said (in terms of what kind of knowledge is relevant to policy discussions), but who delivers the message. While it may be intuitive to assume that researchers and clinicians are better placed to produce and deploy the kind of ‘objective’ and instrumental knowledge that will count as ‘truth’ (due to their institutional access to research processes, ethics boards, qualifications, and modes of dissemination in academic fora and medical journals), from these interviews we can also see the ways in which the productive power of evidence-based policy discourse continually positions and sustains researchers and clinicians as ‘objective’ and ‘rational’ legitimate speakers, while excluding ‘advocates’ as irrelevant or illegitimate.

Resources for resistance

Closer analysis revealed resources for re-problematisation and resistance. Across the interviews, it was the ‘legitimate speakers’ who most frequently articulated their frustrations with the problematic barriers that the requirement of ‘high level evidence’ (Clinician) had created for the implementation of peer-administered naloxone programmes. While policy makers, drug user organisations and advocates discussed the importance of gathering together a convincing ‘evidence base’ (perhaps attempting to legitimise their own voices through appeals to objective knowledge), many of the researchers and clinicians interviewed offered critiques of the notion of a hierarchy of evidence or unbiased knowledge, and expressed frustration at calls for ‘more evidence’ or RCTs:

I found the debate around evidence-building and RCTs interesting and a bit infuriating actually. […] I think to hold RCTs as the gold standard is very problematic. (Researcher)

It could be argued that it is from the position of legitimacy afforded to researchers and clinicians from within evidence-based policy discourse that this critique is made possible.

In critiquing the pre-eminence of RCTs and articulating their frustrations, researchers and clinicians appealed to other ways of knowing. As another researcher argued,

The sort of people that say RCTs are the only way to go, I just think are breathtakingly stupid, because there are other ways to knowledge; and there are some questions, that by their very definition, will never, ever, be amenable to an RCT. And I will go back to that wonderful BMJ satirical piece. […] We’ve never done an RCT of parachutes, says the BMJ, and under the guise of absolutely cutting edge satire, they made a very important point; that, one, how would you do it; and, two, why would you do it? We know it works. (Researcher, emphasis added)

This researcher was one of several participants (clinicians and researchers) who mentioned the British Medical Journal article, which satirically notes that the effectiveness of parachutes has not been established through RCTs, and that the extant observational evidence is insufficient to satisfy proponents of evidence-based policy and practice (see Smith and Pell, 2003). In invoking this satirical piece, participants argued that there were other ways of ‘rationally’ ‘knowing’ that ‘naloxone works’, thus re-deploying and re-problematising the ‘instrumental rationality’ concept at the heart of evidence-based policy discourse. So while ‘instrumental rationality’ is based on the notion of a particular kind of ‘objective’ and ‘instrumental’ knowledge being used by ‘rational’ decision makers to ‘solve problems’ (as discussed above), appeals to other ways that we might ‘rationally’ ‘know’ ‘what works’ here destabilise this particular representation, highlighting that it is not fixed, and thereby altering relations of knowledge and power. The concept of ‘rationality’ itself, and how we might otherwise ‘rationally’ ‘know’ ‘what works’, becomes a ‘discursive resource’ (Bacchi, 2009: 45) which can be raised and re-deployed in response to what is seen as a problematic practice (one which, in this case study, was perceived to be creating barriers to the field supporting the implementation of peer-administered naloxone programmes). This is not about the strategic framing of arguments but rather opening up ‘one position’ that evidence-based discourse ‘makes available’, thus allowing a contesting stance to be developed (Bacchi, 2009: 46).

Invoking this contesting stance, participants made appeals to a number of other ways in which we might ‘rationally’ ‘know’ ‘what works’. ‘Common sense’ and ‘logic’ were prominent themes:

In some areas common sense can – must prevail. I don’t see what the benefit of a high order, high level random controlled trial would achieve. Naloxone has got significant evidence that it reverses overdose. If it’s in a pharmacist’s cupboard, it won’t. But if it’s in a drug user’s cupboard, and taken out at the event of an overdose, it might. That’s the painfully obvious evidence from my point of view. It’s the laws of physics that a mini-jet of naloxone, won’t, of its accord, jump up and run across the street and find somebody who’s just overdosed. (Clinician, emphasis added)

This clinician re-deploys assumptions about rationality implicit within the evidence-based policy paradigm to create a new space from which to consider the effectiveness of naloxone. Here, ‘evidence’ of ‘what works’ is remade as a ‘common sense logic’ argument formed not from the scientific methods reified within the evidence-based policy paradigm which ‘prove’ effectiveness, but rather from a particular ontological view of the world. The clinician presents this account of ‘what works’ not as an anecdotal observation, but rather as a ‘rational’ and ‘logical’ response to the world as he sees it. These appeals to logic, the observed world and ‘laws of physics’ remake the notion of rationality into a different way of knowing.

The ‘obvious common sense’ rationality and objectivity of observation was invoked by another clinician:

sometimes I think we make it hard for ourselves at the altar of wanting to have evidence-based policy, letting it get in the way of doing things that just make obvious common sense. You see it with the naloxone thing as well, that yes people are overdosing, they’re using naloxone, they’re reversing the overdose, people aren’t dying. Surely that’s enough. That’s what I liked about the ‘I’m the Evidence’ campaign because that really struck at the heart of that nonsense of ‘We need experimental evidence’. It’s like look at these people, they had overdoses, they’re still alive. (Clinician, emphasis added)

Here, the empiricism at the heart of methods occupying the peak of the ‘hierarchy of evidence’ is simultaneously challenged and remade. The clinician likes the way the ‘I am the Evidence’ campaign (described earlier) challenges the call for experimental evidence, but in doing so points to a different way of observing and ‘seeing’ effectiveness. Again, ‘common sense’ observation is remade as sufficient and relevant knowledge for policy because it is ‘logically’ ‘obvious’ and ‘rational’.

Nonetheless, despite the critique and other ways of knowing put forward, participants continued to mobilise and defend the evidence-based drug policy paradigm through their every day practices. While the articulation of their vociferous frustrations highlighted the ways in which evidence-based policy discourse does make available a contesting position and create room for re-problematisation and resistance, the clinicians and researchers interviewed did not completely resist the paradigm or fully destabilise taken-for-granted ‘truths’ about the value of particular kinds of knowledge for policy. Rather, they continued through their day-to-day practices to preserve the implicit notions of rationality and objectivity, and thus maintain their legitimacy. Participants still continued their pursuit of ‘objective’ ‘evidence’ through the evaluation structures they built around the new naloxone programmes and saw their contributions to academic literature as important. As one clinician said (after earlier saying ‘we make it hard for ourselves at the altar of wanting to have evidence-based policy’),

we still thought that it would be good to do that [evaluate the program], also by way of disseminating our results and adding to the evidence-base down the track. And we also felt that then we would have to put it through ethics and that that would sort of give it a bit of extra imprimatur. So as part of the ethics process, scientific merit is assessed and so that would also give us something to fall back on should we find ourselves in a bit of a storm. (Clinician)

Despite their critique, participants perceived a need to mobilise evidence-based policy discourse, and continued to appeal to notions of ‘scientific merit’ and processes of formal evaluation in order to justify their actions and retain legitimacy.

Conclusion

Taken-for-granted ‘truths’ implicit within evidence-based policy discourse privilege particular kinds of ‘objective’ and ‘rational’ knowledge and, in so doing, legitimate the voices of researchers and clinicians to the exclusion of others. We suggest that what appears to be a simple requirement for methodological rigour in the evidence-based policy paradigm actually rests on deeper assumptions which place limits around not only what can be said (in terms of what kind of knowledge is relevant for policy debate) but also who may legitimately speak. As we noted earlier, we are not suggesting that researchers and clinicians always occupy privileged positions, but that the productive power of evidence-based policy makes them so. As calls for the voices of people who use drugs to be included in policy deliberation continue to gain traction in the drug policy field (Jürgens, 2008; Lancaster et al., 2013), these findings raise a critical question: to what extent is the reifying of ‘evidence-based policy’ limiting opportunities for participation, through the way its practice legitimates particular voices to the exclusion of others? This silencing effect applies not only to people who use drugs themselves, but also their families, communities and those who advocate from within academia and the medical profession. This is a key question not only for drug policy, but also for the consumer participation agenda in health.

Given the dominance of ‘evidence-based’ policy discourse in drug policy and more broadly as a contemporary mode of rule, it is difficult to problematise or bring into question what is meant by the seemingly incontrovertible concept of ‘evidence’. However, we have identified points of rupture which make it possible to appeal to other ways of knowing. The way in which evidence-based policy discourse has become a barrier to the implementation of peer-administered naloxone has left this discourse open to challenge. In turn, this has opened up the opportunity to redeploy the notion of ‘rationality’, thus making it possible to argue that there are other ‘common sense’ or ‘logical ways’ of ‘seeing’ and ‘knowing’ that naloxone ‘works’.³ So while the central tenet of evidence-based policy (that good policy decisions should be based on rational assessments of effectiveness) may seem incontrovertible, there are tensions present which may be regarded as resources for re-problematisation and resistance.

It is important to note here that redeploying the notion of ‘rationality’ does not necessarily devalue ‘gold star’ research evidence but rather ‘revalues upwards’ (Bacchi, 2009: 73) other ways of knowing (other ‘evidences’). As Law (2004) notes, it is not the case that standard research methods are ‘straightforwardly wrong’ (p. 4). Indeed, they have been the basis of significant developments in drug policy and public health more broadly. The critical question is whether reifying particular methods or ways of knowing places the drug field in ‘a set of constraining normative blinkers’ (Law, 2004: 4) by claiming authority.

This analysis illuminates how other ways of knowing can be relevant for policy, and suggests that the notion of the thing we call ‘evidence’ needs to be further opened up, problematised and questioned (Lancaster, 2016). The challenge posed to scientific method and the limits of the concepts of ‘objectivity’ and ‘instrumental rationality’ at the heart of the evidence-based policy paradigm are beginning to be explored in different ways in the drugs field but require further critical consideration. For example, drawing on Bruno Latour’s work, Fraser (2015) has problematised taken-for-granted distinctions between scientific fact and self-observation. Along with Fraser (2015), our analysis provides a ‘corrective’ to the catch-cry of ‘evidence-based drug policy’ by revealing contradictions and points of resistance (p. 18). As the idea of ‘objective evidence’ faces increasing public scrutiny, Latour (2013) has suggested a new way of thinking about scientific knowledge which remakes and redefines science in such a way that it may also ‘allow us to give more space to other values that are very commonly encountered but that did not necessarily find a comfortable slot for themselves within the framework offered by modernity’ (p. 11). Latour’s (2013) goal is to create an arrangement that he describes as a ‘diplomatic’ mission, making room for other knowledges and bringing ‘a larger number of values into cohabitation within a somewhat richer ecosystem’ (pp. 7, 11). The accounts offered by our participants reveal the ways in which a larger number of ways of knowing are already co-habiting within drug policy. Like the notion of addiction examined by Fraser (2015), naloxone provision is another issue within the drug policy field which ‘refuses resolution by recourse to scientific evidence’ (p. 17). Time and time again, participants bemoaned the limitations of the ‘hierarchy of evidence’ and the barriers it produced, turning instead to other ways of knowing. By applying Bacchi’s approach in this analysis, we have identified resources for re-problematisation (positions that evidence-based policy discourse makes available) which provide another way of thinking about how to make Latour’s ‘diplomatic’ mission possible within the drug policy field and produce avenues for different kinds of action.

Despite these opportunities for rethinking the evidence-based drug policy paradigm, we have also identified how the continued mobilisation of ‘evidence-based’ discourse within the field obscures these contesting positions and continues to privilege particular speakers and modes of engagement. As Fraser et al. (2014) contend, perhaps as a field we are so dependent on the ‘evidence-based policy’ paradigm that we are ‘obliged’ to present our views ‘in the taken-for-granted modes of evidencing, that is, in the format of scientific knowledge’ despite its obvious limitations (p. 236). As Bacchi (2009) notes, it has become virtually impossible for those in the research community to speak outside of the confines of ‘evidence-based’ discourse.⁴ So while researchers and clinicians bemoan the demands of the hierarchy of evidence and highlight the limits of evidence-based policy, they continue to mobilise it, thus becoming ‘enlisted’ in the task of governing through the knowledges they produce and deploy (Bacchi, 2009: 157).

While we have analysed this case study through Bacchi’s WPR lens, other critical approaches would yield further insights. For example, valentine (2009) suggests that ‘a critical values-based approach directs us to questions of political power, ideology and marginalisation of drug users’ while a science and technology studies approach invokes additional questions such as ‘how is it we know what we do about drugs?’ (p. 460), thus disallowing established divisions between science and the social. Another approach might be to explore the ‘making’ of naloxone itself as an intervention. As has recently been proposed by Rhodes et al. (2016), ‘what constitutes an intervention, and knowledge about it, is not given but made’, recognising that evidence-based sciences are only one part of its constitution. We suggest the naloxone debates provide fertile ground for such examination.

Given the seeming incontrovertibility of ‘evidence-based policy’ discourse, we recognise that suggesting a shift away from such a dominant paradigm may appear unlikely or impractical. But as Fraser (2016) notes, it is important to recall that ‘major shifts in thinking do occur’ (p. 4). Our analysis herein has sought to continue to explore the basis on which such a shift may be considered, by highlighting the already present possibilities for re-problematisation and resistance.

Footnotes

Acknowledgements

Thank you to all the participants who generously gave their time to this research. The Drug Policy Modelling Program is funded by the Colonial Foundation Trust, and is located at the National Drug and Alcohol Research Centre, a research centre funded by the Australian Government. The Centre for Social Research in Health is supported by a grant from the Australian Government Department of Health.

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Kari Lancaster is a recipient of an Australian Postgraduate Award. Alison Ritter is a recipient of a National Health and Medical Research Council Fellowship (APP1021988).

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

Notes

Author biographies

Kari Lancaster is a research associate at the Drug Policy Modelling Program (DPMP) at the National Drug and Alcohol Research Centre (NDARC) at the University of New South Wales. Since joining the DPMP in 2009, Kari has undertaken research investigating policy processes, with a particular interest in examining ‘problematisation’ and how drug policy problems and policy knowledge(s) are constituted.

Carla Treloar is director of the Centre for Social Research in Health, UNSW Australia. Carla has been a member of the NSW and Australian Ministerial Advisory Committees and numerous advisory committees for government, health agencies and non-government organisations.

Alison Ritter is the director of the Drug Policy Modelling Program (DPMP) at the National Drug and Alcohol Research Centre (NDARC) at the University of New South Wales. She is an NHMRC senior research fellow leading a multi-disciplinary programme of research on drug policy. The goal of the work is to advance drug policy through improving the evidence-base, translating research and studying policy processes.

References

Bacchi

(2009) Analysing Policy: What’s the Problem Represented to Be? Sydney, NSW, Australia: Pearson Education.

Bacchi

(2015) Problematizations in alcohol policy: WHO’s ‘alcohol problems’. Contemporary Drug Problems 42(2): 130–147.

Bell

(2012) Cochrane reviews and the behavioural turn in evidence-based medicine. Health Sociology Review 21(3): 313–321.

Bigg

(2002) Data on take home naloxone are unclear but not condemnatory. British Medical Journal 324(7338): 678.

Bird

Parmar

MKB

Strang

(2015) Take-home naloxone to prevent fatalities from opiate-overdose: Protocol for Scotland’s public health policy evaluation, and a new measure to assess impact. Drugs: Education, Prevention, and Policy 22(1): 66–76.

Coffin

Sullivan

(2013) Cost-effectiveness of distributing naloxone to heroin users for lay overdose reversal. Annals of Internal Medicine 158(1): 1–9.

Darke

Hall

(1997) The distribution of naloxone to heroin users. Addiction 92(9): 1195–1200.

Davies

HTO

Nutley

Smith

(2000) What Works? Evidence-Based Policy and Practice in Public Services. Bristol: The Policy Press.

Dean

(1999) Governmentality: Power and Rule in Modern Society. London: SAGE.

10.

Dietze

Lenton

(2010) The case for the wider distribution of naloxone in Australia. Available at: http://www.atoda.org.au/wp-content/uploads/The_heroin_reversal_drug_naloxone_FIN2.pdf (accessed 20 January 2017).

11.

Eurasian Harm Reduction Network (2014) I’m the evidence: Naloxone works! (6 videos). Available at: http://www.harm-reduction.org/library/im-evidence-naloxone-works-6-videos-ehrn-2014 (accessed 20 October 2015).

12.

Eurasian Harm Reduction Network (2015) I’m the evidence. Naloxone works (Facebook page). Available at: https://www.facebook.com/NaloxoneWorks (accessed 20 October 2015).

13.

Foucault

(1980) Power/Knowledge: Selected Interviews and Other Writings 1972–1977. New York: Pantheon.

14.

Foucault

(1991) Governmentality. In: Burchell

Gordon

Miller

(eds) The Foucault Effect: Studies in Governmentality. Chicago, IL: The University of Chicago Press, pp. 87–104.

15.

Fraser

(2015) A thousand contradictory ways: Addiction, neuroscience, and expert autobiography. Contemporary Drug Problems 42(1): 38–59.

16.

Fraser

(2016) Articulating addiction in alcohol and other drug policy: A multiverse of habits. International Journal of Drug Policy 31: 6–14.

17.

Fraser

Moore

(2011) Governing through problems: The formulation of policy on amphetamine-type stimulants (ATS) in Australia. International Journal of Drug Policy 22(6): 498–506.

18.

Fraser

Moore

Keane

(2014) Habits: Remaking addiction. London: Palgrave Macmillan.

19.

Harm Reduction Coalition (2015) I’m the evidence video campaign. Available at: http://harmreduction.org/issue-area/overdose-prevention-issue-area/i%E2%80%99m-the-evidence-video/ (accessed 20 October 2015).

20.

Harrison

Checkland

(2009) Evidence-based practice in UK health policy. In: Gabe

Calnan

(eds) The New Sociology of the Health Service. Oxford: Routledge, pp. 121–142.

21.

Jürgens

(2008) Nothing about Us without Us – Greater, Meaningful Involvement of People Who Use Illegal Drugs: A Public Health, Ethical, and Human Rights Imperative, International Edition. Toronto, ON, Canada: Canadian HIV/AIDS Legal Network, International HIV/AIDS Alliance, Open Society Institute.

22.

Lancaster

(2014) Social construction and the evidence-based drug policy endeavour. International Journal of Drug Policy 25(5): 948–951.

23.

Lancaster

(2016) Performing the evidence-based drug policy paradigm. Contemporary Drug Problems 43(2): 142–153.

24.

Lancaster

(2017) Confidentiality, anonymity and power relations in elite interviewing: Conducting qualitative policy research in a politicised domain. International Journal of Social Research Methodology 20(1): 93–103.

25.

Lancaster

Ritter

(2014a) Examining the construction and representation of drugs as a policy problem in Australia’s National Drug Strategy documents 1985 to 2010. International Journal of Drug Policy 25(1): 81–87.

26.

Lancaster

Ritter

(2014b) Making change happen: A case study of the successful establishment of a peer-administered naloxone program in one Australian jurisdiction. International Journal of Drug Policy 25(5): 985–991.

27.

Lancaster

Duke

Ritter

(2015a) Producing the ‘problem of drugs’: A cross national-comparison of ‘recovery’ discourse in two Australian and British reports. International Journal of Drug Policy 26(7): 617–625.

28.

Lancaster

Ritter

Stafford

(2013) Public opinion and drug policy in Australia: Engaging the ‘affected community’. Drug and Alcohol Review 32(1): 60–66.

29.

Lancaster

Seear

Treloar

(2015b) Laws prohibiting peer distribution of injecting equipment in Australia: A critical analysis of their effects. International Journal of Drug Policy 26(12): 1198–1206.

30.

Latour

(2013) An Inquiry into Modes of Existence. Cambridge, MA: Harvard University Press.

31.

Law

(2004) After Method: Mess in Social Science Research. Oxford: Routledge.

32.

Lenton

Hargreaves

(2000) Should we conduct a trial of distributing naloxone to heroin users for peer administration to prevent fatal overdose? Medical Journal of Australia 173(5): 260–263.

33.

Lenton

Dietze

Degenhardt

. (2009a) Naloxone for administration by peers in cases of heroin overdose. Medical Journal of Australia 191(8): 469.

34.

Lenton

Dietze

Degenhardt

. (2009b) Now is the time to take steps to allow peer access to naloxone for heroin overdose in Australia. Drug and Alcohol Review 28(6): 583–585.

35.

Lenton

Dietze

Olsen

. (2015) Working together: Expanding the availability of naloxone for peer administration to prevent opioid overdose deaths in the Australian Capital Territory and beyond. Drug and Alcohol Review 34(4): 404–411.

36.

Lin

Gibson

(eds) (2003) Evidence-Based Health Policy: Problems and Possibilities. Oxford: Oxford University Press.

37.

McAuley

Best

Taylor

. (2012) From evidence to policy: The Scottish national naloxone programme. Drugs: Education, Prevention, and Policy 19(4): 309–319.

38.

Marston

Watts

(2003) Tampering with the evidence: A critical appraisal of evidence-based policy making. The Drawing Board: An Australian Review of Public Affairs 3(3): 143–163.

39.

N-ALIVE (2014) Update for people who have taken part in the N-ALIVE pilot trial. Available at: http://www.ctu.mrc.ac.uk/13391/13399/n-alive_update_12.14 (accessed 16 November 2015).

40.

Oldham

Wright

NMJ

(2003) A UK policy on ‘take home naloxone’ for opiate users – Strategy or stalemate? Drugs: Education, Prevention, and Policy 10(2): 105–119.

41.

Open Society Foundations (2013) Naloxoneinfo.org. Available at: http://www.naloxoneinfo.org/ (accessed 18 June 2013).

42.

Pienaar

Savic

(2015) Producing alcohol and other drugs as a policy ‘problem’: A critical analysis of South Africa’s ‘National Drug Master Plan’ (2013–2017). International Journal of Drug Policy 30: 35–42.

43.

Rhodes

Closson

Paprini

. (2016) Towards ‘evidence-making intervention’ approaches in the social science of implementation science: The making of methadone in East Africa. International Journal of Drug Policy 30: 17–26.

44.

Rose

O’Malley

Valverde

(2006) Governmentality. Annual Review of Law and Social Science 2: 83–104.

45.

Roumeliotis

(2014) Drug prevention, politics and knowledge: Ideology in the making. Addiction Research & Theory 22(4): 336–347.

46.

Seear

Fraser

(2014) The addict as victim: Producing the ‘problem’ of addiction in Australian victims of crime compensation laws. International Journal of Drug Policy 25(5): 826–835.

47.

Smith

GCS

Pell

(2003) Parachute use to prevent death and major trauma related to gravitational challenge: Systematic review of randomised controlled trials. British Medical Journal 327(7429): 1459–1461.

48.

Smith

(2013) Beyond Evidence-Based Policy in Public Health: The Interplay of Ideas. London: Palgrave Macmillan.

49.

Sporer

Kral

(2007) Prescription naloxone: A novel approach to heroin overdose prevention. Annals of Emergency Medicine 49(2): 172–177.

50.

Strang

Farrell

(1992) Harm minimisation for drug misusers: When second best may be best first. British Medical Journal 304(6835): 1127–1128.

51.

Strang

Bird

Parmar

(2013) Take-home emergency naloxone to prevent heroin overdose deaths after prison release: Rationale and practicalities for the N-ALIVE randomized trial. Journal of Urban Health 90(5): 983–996.

52.

Strang

Bird

Dietze

. (2014) Take-home emergency naloxone to prevent deaths from heroin overdose. British Medical Journal 349: g6580

53.

Strang

Darke

Hall

. (1996) Heroin overdose: The case for take-home naloxone. British Medical Journal 312(7044): 1435–1436.

54.

Strang

Kelleher

Best

. (2006) Emergency naloxone for heroin overdose: Should it be available over the counter? British Medical Journal 333(7569): 614.

55.

Strang

Powis

Best

. (1999) Preventing opiate overdose fatalities with take-home naloxone: Pre-launch study of possible impact and acceptability. Addiction 94(2): 199–204.

56.

valentine

(2009) Evidence, values and drug treatment policy. Critical Social Policy 29(3): 443–464.

57.

Wheeler

Davidson

Jones

. (2012) Community-based opioid overdose prevention programs providing naloxone – United States, 2010. Morbidity and Mortality Weekly Report 61(6): 101–105.

58.

Wodak

(2013) Should naloxone be used to reduce opioid overdoses? The Conversation, 15 November Available at: https://theconversation.com/should-naloxone-be-used-to-reduce-opioid-overdoses-19896