How did policy actors use mass media to influence the Scottish alcohol minimum unit pricing debate? Comparative analysis of newspapers, evidence submissions and interviews

Abstract

Aims: To explore how policy actors attempted to deliberately frame public debate around alcohol minimum unit pricing (MUP) in the UK by comparing and contrasting their constructions of the policy in public (newspapers), semi-public (evidence submissions) and private (interviews). Methods: Content analysis was conducted on articles published in ten national newspapers between 1 January 2005 and 30 June 2012. Newsprint data were contrasted with alcohol policy documents, evidence submissions to the Scottish Parliament’s Health and Sport Committee and 36 confidential interviews with policy stakeholders (academics, advocates, industry representatives, politicians and civil servants). Findings: A range of policy actors exerted influence both directly (through Parliamentary institutions and political representatives) and indirectly through the mass media. Policy actors were acutely aware of mass media’s importance in shaping public opinion and used it tactically to influence policy. They often framed messages in subtly different ways, depending on target audiences. In general, newspapers presented the policy debate in a “balanced” way, but this arguably over-represented hostile perspective and suggested greater disagreement around the evidence base than is the case. Conclusions: The roles of policy actors vary between public and policy spheres, and how messages are communicated in policy debates depends on perceived strategic advantage.

• Alcohol
• evidence base
• health
• policy
• public health
• qualitative research

Background

Alcohol use is responsible for over 4.8 million deaths per year worldwide (Lim et al., 2012) and price-based interventions are a key mechanism to address this (World Health Organization, 2010). The UK, and Scotland in particular, experiences high levels of alcohol-related harms compared to the rest of western Europe (Leon & McCambridge, 2006). Historically, many countries used alcohol duty (taxation) measures but there is growing interest in interventions that selectively affect the cheapest alcohol products. For example, reference pricing has been used in a number of Canadian provinces and has resulted in several health benefits (Stockwell, Auld, Zhao, & Martin, 2012a; Stockwell et al., 2012b, 2013). Minimum unit pricing (MUP) of alcohol aims to link the lowest price of an alcoholic product to its alcohol content (calculated on the basis of units). It is a key component of the Scottish Government’s policy to address alcohol-related harms (Scottish Government, 2008), and has been considered elsewhere in the UK (HM Government, 2012). Legislation to introduce the policy has been passed in Scotland (Scottish Parliament, 2012) but its implementation has been delayed following industry-instigated legal challenges (BBC News, 2012; Katikireddi & McLean, 2012).

Alcohol policy is markedly affected by the mass media (such as newspapers, television and radio; Baillie, 1996; Yanovitzky & Stryker, 2001), but in a complex and contingent manner (Walgrave & Van Aelst, 2006). The mass media can influence people’s perceptions of a policy debate, including those of the actors involved in the debate, by helping define what issues are of greatest and least importance (McCombs & Shaw, 1972). In other words, a distinction can be made between “agenda-setting”, where the importance of an issue is highlighted, and “frame-setting”, which involves the salience of specific issue attributes being foregrounded (and others necessarily downplayed; Scheufele, 1999, 2000). The framing of messages within the mass media therefore fulfils four functions: it defines the way problems are viewed; diagnoses causes of a problem; makes moral judgements about attribution; and suggests remedies (Entman, 1993). Within the policy analysis literature, this process is seen as an essential aspect of policymaking (Stone, 1997). Furthermore, the mass media do not just act as a conduit to policy actors, but also they exert an indirect influence by shaping public attitudes and hence helping define the limits of what is politically acceptable (Weiss, 1986).

Importantly, the mass media may often be used strategically by policy actors to shape and influence policy development (Stewart & Casswell, 1993; Stone, 1997). The interactive nature of deliberate framing is worth noting – while policy actors deliberately portray an issue in an attempt to influence audiences’ perceptions, audiences are not passive consumers and may actively reject or reinterpret these presentations (Kitzinger, 1999). Although there has been considerable research examining the content of media messages (Hilton, Hunt, Langan, Bedford, & Petticrew, 2010; Hilton, Patterson, & Teyhan, 2012; Wakefield, Loken, & Hornik, 2010) and on media marketing both online (Winpenny, Marteau, & Nolte, 2014) and offline (Hastings, Anderson, Cooke, & Gordon, 2005), less attention has been paid to the ways in which policy actors communicate messages through the mass media (Hansen & Gunter, 2007; Randolph & Viswanath, 2004). Public health professionals have argued that using the mass media strategically is an important component of public health advocacy (Freudenberg, 2005; Stewart & Casswell, 1993; Wallack & Dorfman, 1996). Despite this, empirical examples of how the mass media contribute to the development of specific alcohol policies, particularly those which are high-profile and challenge existing power relationships, remain unusual.

Evidence has been influential in the MUP debate. Quantitative research, especially drawn from the field of epidemiology, has documented the magnitude of alcohol harms and indicated that high-risk drinking is more likely amongst those who purchase the cheapest alcohol (Black, Gill, & Chick, 2011; ISD, 2011; Leon & McCambridge, 2006). Public perspectives of MUP have been investigated to a lesser extent, with Lonsdale, Hardcastle and Hagger (2012) finding limited public support for the intervention in a focus group study based on the belief that it would likely be ineffective, could “penalise” moderate drinkers and may result in adverse social consequences. Research based on an economics perspective has been widely discussed. The “Sheffield model” (Purshouse, Meier, Brennan, Taylor, & Rafia, 2010) which estimates the likely impacts (in health, social and economic terms) of different price-based alcohol policy interventions including MUP, has been particularly influential (Katikireddi, Bond, & Hilton, 2014b). Two crucial findings from this modelling were that MUP was expected to have a larger impact than changes in alcohol duty and that the intervention was “targeted”, in the sense that drinkers at greatest risk were predicted to change their consumption the most. The Centre for Economics and Business Research (CEBR) were funded by the brewer SABMiller to conduct a critique of the Sheffield model (Ata, Ohanissian, Pragnell, Read, & Snook, 2009). This report argued that heavy drinkers were less price-responsive and raised concerns that MUP would result in increased consumer expenditure which would benefit large retailers while resulting in relatively small health and social benefits.

Research on the policy process identified several factors as of particular importance in MUP’s development, including the high burden of harms, the institutional environment (with a devolved Scotland’s more limited legislative powers fostering policy innovation) and a re-framing of the policy debate (Hawkins & Holden, 2013; Katikireddi, Bond, & Hilton, 2014a; Katikireddi, Hilton, Bonell, & Bond, 2014c). Some studies have also focussed on the role of corporate actors in the policy process (Hawkins & Holden, 2014; Hawkins, Holden, & McCambridge, 2012; Holden & Hawkins, 2013), including studies documenting a selective use of evidence (Holden, Hawkins, & McCambridge, 2012; McCambridge, Hawkins, & Holden, 2013, 2014). Holden et al. (2012) argue that rather than considering the “alcohol industry” as a single entity, it is important to consider heterogeneity between and within sectors. In relation to MUP, a diverse range of sectors are likely to be affected, including producers, the licensed trade (e.g. pubs, nightclubs) and a broad variety of off-licenses, such as national supermarkets and local corner shops. Within sectors, supportiveness for interventions like MUP is likely to relate to perceived impact on profit-making which may in turn depend on a company’s product portfolio and customer market.

In this article, we aim to understand how policy actors attempted to deliberately frame public debate around alcohol MUP in the UK. To do so, we compare constructions of the policy issue in three different arenas: the public arena of messages presented by policy actors in the mass media; the semi-public arena of Scottish Parliamentary evidence submitted by policy actors (that is publicly available on-line but typically not achieving a similar reach as the mass media); and “private” accounts in confidential one-to-one interviews.

Methods

To study the ways that policy actors may make deliberate use of the mass media, data that illuminate discourses within the mass media and the perspectives of policy actors are ideal. We therefore draw upon two large-scale studies of MUP: the first, investigating newsprint media coverage, and the second the policy process by which the intervention developed. We summarise the primary data collection procedures below and go on to describe the comparative approach adopted. More details of the primary data have been published elsewhere (Hilton, Wood, Patterson, & Katikireddi, 2014; Katikireddi, 2013; Katikireddi et al., 2014a; Wood, Patterson, Katikireddi, & Hilton, 2014).

Content analysis of UK newspaper reporting

Ten publications with high circulation (National Readership Survey, 2012) were selected: three Scottish national newspapers and seven UK national newspapers. Tabloid, middle-market and serious genres were included to ensure diverse readerships were represented (Hilton et al., 2010; Williams, Seale, Boden, Lowe, & Steinberg, 2008). Researchers searched the Nexis UK and Newsbank databases for articles containing variants of the terms “alcohol” and “pricing” published between 1 January 2005 and 30 June 2012. The time period began prior to Scottish Health Action on Alcohol Problems’ (SHAAP) first announcement of support for MUP, and ended just after parliamentary passage of the Alcohol (Minimum pricing) (Scotland) Bill. The search returned 2076 articles. Each article was read to ensure they met our inclusion criteria that the content relates to alcohol and MUP (defined as more 50% of the text). All duplicate and Eire editions were excluded, leaving 901 relevant articles for detailed coding. A coding frame was developed by scoping the literature and reading a random selection of 100 articles to identify thematic codes. Further articles were read in batches of 20 until no new codes emerged. Codes were grouped into thematic categories. The coding frame also recorded descriptive information such as publication, page number and word count. Initially the quantifiable manifest content of each article was coded (Patterson, Katikireddi, Wood, & Hilton, 2014). From this, a subset of 262 articles were selected which contained quotes or any comment from key stakeholders constructions of the alcohol problem and claims about MUP (Hilton et al., 2014).

Examining the policy process

Three different data sources were used to investigate the process by which MUP developed. This study focused on the Scottish and UK governments and did not examine in detail attempts to introduce MUP at the local level. First, a review of existing relevant alcohol policy, with a focus on the UK over the last two decades, was conducted (Katikireddi, 2013). Second, advantage was taken of the availability of evidence submissions from interested stakeholders, as part of the Scottish Parliament’s legislative process. The Health and Sport Committee was responsible for providing oversight during the introduction of primary legislation for MUP and these documents were submitted by interested actors. Documents were retrieved, systematically coded to describe the type of actor and their supportiveness of MUP. This was followed by more detailed qualitative analysis of the framing (Katikireddi et al., 2014a). Finally, 36 confidential in-depth interviews were carried out with academics, advocates, industry representatives, politicians and civil servants involved in the policy process. In most cases, interviews were transcribed (otherwise detailed notes taken), followed by thematic analysis (Katikireddi et al., 2014c).

Comparative analysis

Primary data from the above two studies were retrieved and re-analysed to allow comparison to be made about the similarities and differences between presentations targeted at the mass media and policy audiences. Frameworks were created to organise the data from newspaper articles (using Microsoft Excel 2007), based on a combination of predetermined and inductive codes. Interview data and evidence submission documents were coded inductively, with the assistance of Nvivo 9. Themes were identified using a constant-comparative approach to identify patterns and explanations for differences across the data, with attention paid to contradictory data (Glaser & Strauss, 2009; Mason, 2002).

Results

Key findings from the newsprint reporting are first summarised to provide the necessary context to allow comparisons with the primary data presented. The main themes identified in the interview data are then presented, with areas of similarity and difference between representations in the mass media contrasted to those articulated in interview data and evidence submission documents.

Key themes in newsprint reporting

Quantitative analysis of 901 newspaper articles that were published between 2005 and 2012 confirmed that MUP attracted a high profile within newspapers (Patterson et al., 2014). Alcohol was acknowledged in most newspaper articles to be an important issue, often associated with overconsumption among young people but driven by broader factors, including its affordability. MUP was typically presented in a neutral manner, with positive portrayals becoming slightly more common over time. Much of the debate focused upon its potential effectiveness, with emotive or business perspectives less common.

In a more detailed qualitative analysis of a subset of articles, different advocates’ and critics’ constructions of the alcohol problem and claims about MUP were examined (Hilton et al., 2014). A broad range of actors were publicly supportive of MUP, including health-related organisations, third sector actors, the police, academics and many within the licensed trade, as well as key politicians. Political advocates who were cited regularly in newspapers included Nicola Sturgeon MSP and various spokespersons for the SNP Scottish Government. David Cameron, the UK Prime Minister, and later the home secretary Theresa May also were cited in support.

The main political critics of MUP were UK MP Andrew Lansley, and in Scotland, Labour MSPs Dr Richard Simpson and Jackie Baillie. The Scottish Labour party were opposed to the policy throughout the period studied; however, newspapers highlighted the support of Labour MSP Malcolm Chisholm, a high-profile back-bencher, for MUP. Alcohol producers were the other key critics cited. Organisations representing retailers, such as the British Retail Consortium and Scottish Grocers’ Federation, were consistently presented as hostile. CEBR were the most commonly mentioned academic critic, following their report about the Sheffield model (although this work was commissioned by an alcohol producer). Table 1 presents a summary of consistent differences in the framing between those in favour and those against.

Table 1. Summary of the divergence between the framing of the minimum unit pricing debate by advocates and critics in UK newspapers.

	Advocates	Critics
Overall framing of the debate	Emphasised the existing burden of alcohol-related harms to provide a reason for targeting health and social disorder issues	Emphasised that minimum unit pricing will not work, sometimes without providing supportive reasons and tended to avoid describing drivers of the alcohol problem
Drivers of the alcohol problem	Cheap alcohol, especially loss leading Irresponsible alcohol marketing	Problem people: youth binge drinkers and dependent drinkers
		Problem attitudes to alcohol
Arguments about minimum unit pricing	Targets cheap drinks and irresponsible retailing Targets alcohol misuse Will reduce social and health harms	No evidence it will be effective Will punish responsible drinkers Will punish the poor Will harm businesses Will lead to illicit alcohol trading Will be illegal

Table summarises material reported in (Hilton, Wood, Patterson, & Katikireddi, 2014).

Framing the policy debate through evidence submissions

A diverse range of actors responded to the Scottish Parliament’s Health and Sport Committee call for evidence in 2009, with their positions summarised in the Appendix. It is worth noting that because explicit pre-defined criteria were used to categorise positions of policy actors, several are “misclassified” in the sense that their position has been made more explicitly in other sources. This analysis raises several issues. First, it is clear that there were far more supportive than hostile submissions. However, it is also evident that one reason for this is local public sector organisations responded individually, as well as through their representative organisations. For example, 14 local alcohol and drugs partnerships responded calling for MUP, as well as their umbrella organisation. Second, there also appears to be considerable cohesion amongst non-industry actors, with nearly all being supportive or not stating an explicit position (although many of these organisations voiced support more explicitly elsewhere). Of those who were hostile, most had financial conflicts of interest, either directly or indirectly (e.g. The International Coalition Against Prohibition). Third, the Appendix illustrates the heterogeneity of the alcohol industry. In general, the on-trade was more supportive (with the regulated environment of licensed venues presented as an important advantage), producers were more divided and the off-trade tended towards being hostile. Diversity was found within sectors too.

More detailed qualitative analysis of these documents coupled with the interview data found that the framing of the policy issue differed markedly between advocates and critics, with some key themes summarised in Table 2 (Katikireddi et al., 2014a). Advocates tended to present the drivers of the alcohol problem in a structural manner, similar to those identified by the content analysis of newspaper reporting. However, they differed in their framing of the solution – MUP was presented as an approach that sought to reduce overall population consumption which aimed to address chronic health harms, as well as acute health harms and social disorder. Furthermore, MUP was presented as an essential aspect of a multi-component strategy to address the broad problem of alcohol-related harm. In contrast, critics argued that problem drinking (as assessed by surveys) was falling over the last few years and therefore suggested a need for more targeted approaches. In particular, they argued that education-based measures were needed to change drinking culture and that the impact of recent licensing law reforms should be awaited before taking further action. Industry actors who were in favour often agreed that education and individual responsibility remained key, but suggested MUP targeted those misusing alcohol.

Table 2. Summary of dominant framing of minimum unit pricing in evidence submissions to the Scottish Parliament (triangulated with interviewee data).

	Non-industry related advocates	Industry-related advocates	Critics
Overall framing of the debate	Need to reduce overall population consumption to address alcohol-related harms Alcohol-related harms have increased hugely over last three decades	Need to target those who are behaving irresponsibly Alcohol misuse is a serious issue	Need to target those who are behaving irresponsibly Problem drinking is reducing in last few years
Drivers of the alcohol problem	Increasing affordability of alcohol, particularly in the off-sales environment	Increasing affordability of alcohol but must remember need for individual responsibility	Culture of irresponsible consumption Problem people: young binge drinkers and dependent drinkers
Arguments about minimum  unit pricing	Evidence demonstrates is effective Essential part of multi-component strategy Targets those at greatest risk of harm Reduces health inequalities, as poor at greatest risk of harms Will improve drinking patterns, with lower strength drinks and move to licensed premises Will lead to economic gains due to less absenteeism	Targets those who misuse alcohol Complements education Experience in Canada shows it does not harm industry	Will not change behaviour of the most problematic drinkers Alternative solutions: Education and licensing law (changes had recently occurred) Will harm businesses Will punish responsible drinkers Will punish the poor Will lead to illicit alcohol trading Questions around legality

Table summarises material originally presented in (Katikireddi, Bond, & Hilton, 2014).

Interviewees’ perceptions of mass media coverage

Many interviewees acknowledged the mass media’s importance, with those in favour often seeing coverage as helpful but balanced. A recurrently identified contribution was to help keep alcohol as an active policy issue:

Advocate: I actually think they’ve [the mass media have] been fairly balanced, you know, they’ve always sort of tried to get a balanced argument, so it usually ends up being the industry against the, whoever, whichever health professional or police constable or whatever is in support. So I don’t think they’ve kind of damaged things either way, I think they’ve simply kept things up there a bit in the public eye, which has been a helpful thing, really, just to sort of raise public consciousness of the debate.

There was a feeling amongst many in favour that the mass media had become more supportive over time, echoing our quantitative analysis of newsprint coverage (Patterson et al., 2014). An interviewee related this to being fortunate that “we tend to have a health media who understand” the issues but noted this applied to health journalists in particular, rather than non-specialist reporters. Many interviewees felt alcohol policy was often portrayed in a sensational manner which might not convey the issue’s nuances. This was sometimes identified as favouring those advocating for policy change, particularly in relation to the commonly used “binge drinking” discourse:

Civil Servant: Well, I think (small laugh) the portrayal of and the creation of this thing called binge drinking, you know, in some sense it's a folk devil, it frightens middle England, it sells newspapers. It's shocking, you know, the students urinating on war memorials, the young women falling off of their high heels, all this kind of thing. You know, at one level, one knows it's media sensationalism, but it's a story that has helped to bring the problem to public attention in a way that, in public health terms, I can only see as helpful.

Some interviewees perceived a move from focussing on binge drinking to a more population-based public health framing over time. For example:

Industry: I think in the past media reports have led politicians to make policies on antisocial behaviour because that’s what’s on the front pages. We often see front pages with young people lying on benches or fighting in the streets. We never see a lonely pensioner sitting at home drinking too much because they’re lonely and harming their own health. You know, that doesn’t make the front pages of the Daily Mail. So I think you know we’ve maybe spent too much time in the past tackling antisocial behaviour which is going in the right direction and now there is a focus on health and how to get the information out on health, and I think that’s right.

Strategic use of the mass media

Interviews revealed that both advocates and critics felt influencing the mass media was an important part of trying to influence the policy process.

Civil servant: So I think there is a bit of a game, and we both play it[…] both those that are for a policy and against a policy are kind of using the media. And some of it can be helpful, but some of it, you know, isn’t necessarily helpful. But I think we live in an age where the media does have a huge influence on people.

The data did not suggest interviewees expected a simple relationship to occur between public opinion and the mass media. Rather, several spontaneously argued that a co-constitutive relationship existed with it being unclear “whether the media reflects public opinion or shapes public opinion”. Nevertheless, the mass media was clearly seen as a crucial part of the battle over public opinion. Table 3 provides two examples of how policy actors sought to deliberately frame the policy debate through the mass media. There was evidence of careful regular monitoring of public opinion and the mass media, coupled with careful coordination of media messaging by both advocates and critics.

Table 3. Quotations illustrating the ways policy actors used the mass media to further their policy interests.

Examples of strategic use of the mass media
Civil Servant: We were really, really active, so any time any new research was published we were putting press lines out. We had a guy in the press office because it was just like, let’s turn the volume up on this. And, we were pretty much every week or every couple of weeks, I had like a list of what reports are coming out, planning to get a story out off the back of those things or if the minister was doing a visit that was relevant or, you know? So, it was quite planned in that sense and working quite closely with key partners to ensure that there was a sort of degree of consistency about the messages that we were putting out.
Advocate: In terms of public opinion, if you look at the tracking of public opinion in terms of support for minimum pricing in Scotland – one of the agencies that was working, ad agencies that was working for the [Scottish] Government did some tracking – then basically there was a point in Scotland where the alcohol industry had three public affairs consultancies working for them, and the brief that the consultancies had was to get media stories in the press in order to remove minimum pricing from the Alcohol (etc) Act 2010. So, there was a period in the summer [where] there was about three anti-minimum unit pricing stories in the media every week, which clearly, they [alcohol-related industries] had three public affairs agencies working for them, so they were doing what they were paid to do. And, if you looked at public opinion, it almost tracked exactly that media coverage. So when you had lots of anti-minimum pricing stuff, public opinion went down, when you had, when, you know, Health Scotland, ourselves or others [public health organisations] were having a profile in the press for stuff, public opinion went up. So, it’s a factor, and you have to, as public health advocates, you have to maximise your use of it [the mass media] to the benefit of the public interest.

Although as noted above, mass media coverage was largely viewed positively by public health interests, alcohol policy was considered an intrinsically challenging issue to communicate via the mass media. This includes communication of research findings, particularly in relation to the Sheffield model (Katikireddi et al., 2014b). More fundamentally, the nature of the mass media was felt to constrain the extent to which a population-based public health perspective could be adopted:

Academic: In terms of the media, I guess there are different aspects to this, because clearly, one aspect of it is that the industry advertises through media, so they actually have an interest in aligning with the industry view from a commercial perspective. But also, I would say they will obviously try to reflect the sorts of views that they think their readership will be interested in – if it’s the print media or whatever. You actually see quite contradictory headlines in the media. So one week, they’re criticising sort of nanny state policies by whatever government it might happen to be, and then, the next week, they’re decrying the sort of city centre chaos or whatever drink-fuelled violence has been going on. I think there’s less willingness in the media to accept that there is, that the problem isn’t so much the binge drinking, or not only the binge drinking that is very visible, but is actually the quiet, steady drinking at home or in other settings. It comes back to the ambivalence. It’s not unknown for journalists to take a drink. It’s not unknown for politicians to take a drink so, you know, there is clearly an ambivalence in people recognising that they actually are part of the problem as well as seeing it as a sort of public policy issue.

Financial considerations coupled with the cultural norms of journalists and the public may therefore have mitigated against the whole-scale adoption of a population-based framing within the mass media.

Heterogeneity in the framing of the minimum unit pricing debate

Overall, there were striking similarities in which actors were perceived to be influential in the policy debate when assessed by interviews, evidence submission documents and newspaper articles. For example, health organisations, the voluntary sector, police and industry interests were all frequently quoted within newspaper articles. However, amongst critics, the prominence to which some actors were represented in the public debate (as reflected by newspaper articles) appeared greater or lesser than might be expected from consideration of the evidence submissions and confidential interviews. Supermarkets were not only actively engaged in the policy process but also relatively under-represented within newspaper articles. Similarly, the Scotch Whisky Association were very frequently referred to by the mass media, in comparison to other hostile industry actors. Newspaper articles also maintained a “balanced” perspective by articulating a plurality of perspectives. However, this perhaps resulted in a greater profile for those against MUP, in comparison to within evidence submission documents and amongst policy actors.

Many aspects of the framing of policy debates were consistent between public and more private audiences. Industry actors often used similar arguments in policy circles and amongst the public, with MUP consistently presented as a poorly targeted instrument. Unsurprisingly arguments about the direct impacts on industry (e.g. adverse economic effects) were less prominent in quotations from industry-related spokespeople, with potential adverse impacts on the public privileged. More notably, there appeared to be less effort expended in disputing the framing of alcohol as a “crisis” within newspaper articles when compared to evidence submissions and interviews. Instead, MUP’s effectiveness was denied in a far more clear-cut manner than in policy submissions, often refusing to engage with research evidence in any way. While evidence submissions tended to argue MUP would be ineffective for dependent drinkers, the argument presented within public fora often denied any relationship between price and harm at all. Public health interviewees sometimes expressed surprise at critics’ response to evidence:

Academic: The only thing that occurred to me is just in terms of how the industry engage with the evidence. I found it surprising how willing they are to be wrong in the sense that they are willing to stand there after I’ve given a presentation saying, ‘there is a link between price and harm and here is a systematic review of fifty studies showing clearly that link exists’, for someone to stand up from the industry five minutes later and say ‘there is no evidence of a link between price and harm’. Not to engage with what I’d said but just to deny I even said it. And I found that a very strange way of them actually entering the debate.

Disputes about evidence were also emphasised by politicians hostile to MUP in newspaper articles. In contrast, their opposition was often expressed more in terms of a conflict over values within confidential interviews. For example, one politician critical of MUP explained “my personal reason is social justice […] a policy which is supposed to be a population-based policy that is regressive is something I’m not interested in”. However, this perspective was given less weight when speaking in public (including by that same individual).

Finally, it is worth noting the distinction often made by interviewees between the UK-based and Scottish-based media. While quantitative comparisons showed little difference between newspapers (Patterson et al., 2014), interviewees referred to some aspects of the English media, particularly the Daily Mail, as “quite fierce” and therefore “shapes things a bit down south”.

Conclusions

A broad range of policy actors were involved in the UK policy debate, exerting influence both directly through Parliamentary institutions and political representatives and indirectly through the mass media. Those seeking to influence the policy debate were acutely aware of the mass media’s importance in shaping public opinion, and therefore the viability of policy change, but appreciated that no simple relationship exists between the two. They used the mass media in a strategic manner to influence policy development and often framed messages in subtly different ways, depending on their intended target audience. Mass media messages often omitted the nuance of arguments used to persuade policymakers and the actors chosen to present the message varied depending on the target audience. In general, newspapers appeared to present the policy debate in a “balanced” way, but this arguably had the effect of over-representing hostile perspectives and suggesting greater disagreement around the evidence base than exists.

The findings presented complement a number of existing studies. In UK newspapers, there appears to have been a gradual transition from alcohol having been portrayed as a relatively unproblematic commodity within the UK (Hansen, 2003), to public health concerns having been articulated but policy actions such as MUP actively resisted (Nicholls, 2011), to coverage presenting MUP in a reasonably balanced manner (Hilton et al., 2014). These shifts echo trends in newspaper coverage observed elsewhere (Azar et al., 2013). However, our findings support the suggestion that achieving a public health framing within the mass media may be challenging (Baillie, 1996) for financial as well as cultural reasons. Nevertheless, cultural perspectives around alcohol have been found to change over time (Berridge, 2005), with evolving media discourses appearing to play an important role (Törrönen, 2003). This study provides an additional perspective to previous research about the role of the alcohol industries in the policy debate. Holden and Hawkins (2013) suggested that the Scotch Whisky Association played a lead role in coordinating the response to MUP within Scotland because of the product’s local political significance. This article suggests that this has been the case within the mass media too. Industry critical of MUP has been found to make selective use of evidence (McCambridge et al., 2013, 2014). We find that the manner in which research evidence is engaged with by critical industry actors varied between public and policy circles, with “denialism” seemingly pursued more readily in public fora (McKee & Diethelm, 2010).

Our findings contrast with a similar stream of Australian research that has been developing on public and media discourses around alcohol policy. Unlike the relatively unified position observed amongst non-industry stakeholders in our data, a lack of agreement about the preferred price-based alcohol policy intervention was found amongst public health experts in Australia (Fogarty & Chapman, 2013). However, some similarities were noted amongst mass media coverage between the UK and Australia, with the relative neglect of chronic alcohol-related harms a consistent theme (Fogarty & Chapman, 2012).

Our article has a number of strengths. We draw upon several data sources to provide a more complete picture of the development of a specific policy. Confidential interviews were particularly helpful in providing evidence related to the origin of mass media messages. However, a number of important limitations should be noted. First, these data provide only snapshots of a complex and evolving policy process. It is therefore difficult to be certain that differences observed do not reflect slightly differing timescales for which the data were available. Second, we have only considered newsprint media and not other forms of mass media, including television and social media. Further analyses that examine their influence would be valuable (Nicholls, 2012). Third, while we were successful in obtaining a diverse range of interviewees, there were inevitably some individuals who could not be recruited into the study. The picture presented here is therefore inevitably somewhat incomplete. Finally, our data related to both Scottish and UK mass media and policy processes. Separating out discourses of particular relevance to Scotland and England is problematic, with newspapers covering the interactive development of the policy in both countries, often simultaneously. While differences have been explored, conclusions about the role of the mass media in explaining policy divergence are necessarily tentative, with many other factors clearly important (Katikireddi & Smith, 2014).

This article adds to the emerging body of evidence that suggests the media plays an important role in shaping discussions about public policy, with media discourses appearing to be fiercely contested amongst policy actors (Callaghan & Schnell, 2001). The respective roles of different policy actors are likely to vary between public and policy spheres, and how evidence is used in these debates alters depending on perceived strategic advantage. Understanding such nuances may well help public health advocacy.

Table A1. Stakeholder positions based on manifest content analysis of evidence submissions to the Scottish Parliament’s Health and Sport Committee.

	Position
Sector	Supportive	Not explicit	Neutral	Against
Academic	– Peter Anderson – Royal Society of Edinburgh – Centre for Addictions Research of British Columbia – University of Aberdeen – University of Stirling – University of the West of England	– Anne Ludbrook – Prevention Research Centre – Queen Margaret University – Glasgow Centre for Study of Violence – University of the West of Scotland – Violence Reduction Unit Scotland	– ScHARR – Law Society of Scotland – Medical Research Council	– CEBR
Health actor	– SHAAP – Faculty of Public Health – Scottish Association for Mental Health – Academy of Royal Medical Colleges – Alcohol Health Alliance – Dr. EM Armstrong (Former CMO) – NHS Ayrshire Clinical Forum – NHS Borders & South Borders Council – NHS Borders Health Improvement Team – NHS Health Scotland – NHS Orkney Chair – Royal College of General Practitioners Scotland – Royal College of Nurses – Royal College of Physicians – Royal College of Physicians of Edinburgh – Royal College of Psychiatrists Scotland – Royal College of Surgeons of Edinburgh – Royal Infirmary of Edinburgh – Scottish Patients’ Association – Scottish Ambulance Service – Scottish Association of Alcohol and Drug Teams +14 Local Alcohol and Drugs Partnerships +11 NHS Health Boards	– Dr. Forrester Cockburn – Dr. Macleod – Midlothian Alcohol and Drugs Partnership – Chris Record (member of UK Health Alliance) – Renfrewshire Licensing Forum – Royal College of Physicians and Surgeons of Glasgow – Nick Sheron (hepatologist)	– Canadian Centre for Substance Abuse – Health Protection Scotland	– Addiction Recovery Training Services
Individual	7 individuals		1 individual	3 individuals
Off-trade	– Broadway Convenience Store	– Mitchell’s & Butlers		– Scottish Grocers’ Federation – Scottish Beer and Pub Association – Scottish Retail Consortium
On-trade	– Scottish Licensed Trade Association – NOCTIS – Castle Leisure Group	– British Institute of Innkeeping
Producer	– Tennent’s Caledonian Breweries Ltd. – Molson Coors UK – Chivas Brothers Ltd.		– BAC Canada Brewers	– Whyte & Mackay – Scotch Whisky Association – NACM – Portman Group – SIBA – Diageo – Edrington Group – Gin and Vodka Association – Heineken – SABMiller
Public sector	– West Dunbartonshire Licensing Forum – Association of Directors of Social Work – Dumbarton East and Central Community Council – North Aberdeenshire Licensing Forum – Scottish Commissioner for Children and Young People – Silverton and Overtoun Community Council – South Aberdeenshire Licensing Forum
+ 7 Licensing Boards
+ 7 Local authorities	– City of Edinburgh Licensing Standards – Glasgow City Council Licensing Board – Convention of Scottish Local Authorities – East Lothian Licensing Board – Moray Licensing Board – North Lanarkshire Council, Sports – West Dunbartonshire Council – West Lothian Licensing Board	– Alcohol and Gaming Commission of Ontario – Liquor Control Board of Ontario – Alberta Gaming and Liquor Commission – Highland Licensing Board – Aberdeen City Council	– Clackmannanshire Licensing Board – East Renfrewshire Licensing Board
Supermarket		– Tesco		– Asda – Sainsburys – Cooperative supermarket – Morrisons
Trade representative	– Association of the Chief Police Officers of Scotland – NUS Scotland – BMA Scotland – Campaign for Real Ale – Scottish Police Federation	– British Hospitality Association – Federation of Small Businesses – General Medical Council	– Counselling and Psychotherapy in Scotland – Unison Scotland	– CBI Scotland – Scottish Centre for Development and Industry
Voluntary	– Salvation Army – Children in Scotland – Aberlour Child Care Trust – Youth Link Scotland – Alcohol Focus Scotland – Consumer Focus Scotland – Alcohol Concern – Barnardo’s Scotland – Breakthrough Breast Cancer – Children in Scotland – Church of Scotland – Institute for Alcohol Studies – Poverty Truth Commission – Church of Scotland, Presbytery of Edinburgh – Scottish Episcopal Church – Scottish Youth Commission on Alcohol – Scottish Youth Parliament	– Home Safety Scotland – Scottish Women’s Convention	– Action for Children Scotland	– The International Coalition Against Prohibition

The following criteria were used to categorise the position of actors, based on the manifest content of their evidence submissions to the Scottish Parliament’s Health and Sport Committee: Supportive = explicitly states that the stakeholder favours minimum unit pricing; Against = explicitly states that the stakeholder opposes minimum unit pricing; Neutral = presents both positive and negative statements regarding minimum unit pricing and does not explicitly adopt either of the above positions; Not explicit =presents no explicit statements regarding supportiveness and therefore does not reveal the stakeholder’s position on minimum unit pricing. In addition, some actors were classified as exempt if they were unable to express an explicit opinion because of the statutory nature of their organisation, e.g. some civil service actors.

Declaration of interest

This study received no dedicated funding. At the time the study was conducted, SVK was funded by the Chief Scientist Office at the Scottish Health Directorates as part of the Evaluating Social Interventions programme at the MRC/CSO Social and Public Health Sciences Unit (MC_U130059812 and MC_UU_12017/4) and SH was funded by the Medical Research Council as part of the understandings and uses of Public Health Research programme (MC_U130085862 and MC_UU_12017/6). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

SVK and SH are involved in planning an evaluation of the effects of minimum unit pricing of alcohol in Scotland. They declare that they have no other conflicts of interest.

Acknowledgements

The authors would like to acknowledge our collaborators, Chris Patterson and Karen Wood at the MRC/CSO Social and Public Health Sciences Unit, who led the coding of newspaper articles which this work builds on. SVK would like to thank the interviewees for their participation.

Appendix