Improving the nutritional status of populations remains an urgent priority for addressing the growing burden of obesity and chronic disease worldwide( Reference Roberto, Swinburn and Hawkes 1 ). Recent papers advocate that the most effective way of achieving this involves the promotion and implementation of public policy change, focusing on regulation, legislation and public awareness campaigns designed to make healthy food choices the preferred choice( Reference Roberto, Swinburn and Hawkes 1 , Reference Hawkes, Smith and Jewell 2 ). However, relevant political science theory highlights that changing policy at a national, state or local level of government is a slow and often challenging process due to the complexity of the policy change process and the conservative nature of political systems, which favour the status quo( Reference True, Jones and Baumgartner 3 ). Research from policy process scholars suggests this preference towards the status quo is due to: institutions’ resistance to change; constraints on the rationality of decision makers due to limited time and the cognitive overload they face; and the power of vested interests in maintaining their privileged position( Reference True, Jones and Baumgartner 3 , Reference Kingdon 4 ). For change to occur, therefore, the power of vested interests needs to be challenged and the policy problem, the policy solution and the political climate all need to align( Reference True, Jones and Baumgartner 3 , Reference Kingdon 4 ).
Nutrition advocates from non-government organisations, academia and government bodies have taken up the challenge by attempting to influence policy positively through a range of advocacy strategies, with varied success( Reference Swinburn and Wood 5 ). Advocacy is defined as ‘active interventions by organisations on behalf of the collective interests they represent, that has the explicit goal of influencing public policy or the decisions of any institutional elite’ (p. 643)( Reference Onyx, Dalton and Melville 6 ). Some interest groups have greater power to influence public policy and decision making than others due to their organisational or personal resources, including knowledge of effective advocacy strategies and how to apply them. For groups in positions of low power and influence, improving knowledge of effective advocacy strategies and how to apply them, particularly under resource-constrained conditions, could change the power dynamic between policy makers and the people affected by their decisions( Reference Cohen 7 ).
Internationally, a limited range of peer-reviewed literature documents effective advocacy strategies for influencing public health policy in general, with even less literature on influencing public health nutrition policy( Reference Jenkin, Signal and Thomson 8 ). Political science is one field where a large volume of research has occurred in the study of the policy process and influencing policy change( Reference True, Jones and Baumgartner 3 , Reference Kingdon 4 , Reference Sabatier 9 ); see Table 1 for a summary of influential policy process theories. While theoretical models and frameworks from political science can guide practitioners in influencing the policy-making process, only limited numbers of nutrition policy advocates utilise these theories when championing policy change( Reference Cullerton, Donnet and Lee 10 ). One criticism of these theories, which may explain their limited use, is that they have been designed for an academic audience, rather than a practitioner audience, with concepts that are often difficult to understand and translate to the real world( Reference Cairney 11 ).
Summary of influential theories of the policy process( Reference Cullerton, Donnet and Lee 10 )
To progress nutrition policy change and develop more effective advocates, it is useful to consider the real-world factors and practical experiences of past advocacy efforts to determine the barriers to and enablers of nutrition policy change. Understanding these barriers and enablers may ensure future advocacy efforts of health professionals are more targeted with interventions tailored to address specific barriers and to make use of specific enablers. To date, there has been only one review examining advocacy in nutrition policy undertaken and it focused specifically on obesity prevention( Reference Lyn, Aytur and Davis 12 ). No systematic review has been conducted of the barriers to and enablers of nutrition policy change. To address this gap in knowledge, the current paper presents a systematic review of the barriers to and enablers of nutrition policy change, with a specific focus on high-income, democratic countries.
Methodology
The current review reanalyses the results of a previous systematic literature review( Reference Cullerton, Donnet and Lee 10 ) to answer a new question. These data were first used to determine whether policy process theory had been used in nutrition policy papers. As the majority did not use policy process theory, the data were reassessed via interpretive synthesis( Reference Dixon-Woods, Bonas and Booth 13 ) to determine the enablers of and barriers to public health nutrition policy change. The search strategy has been reported elsewhere( Reference Cullerton, Donnet and Lee 10 ) and is summarised below.
The systematic review was conducted based on the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement (Fig. 1) and a review protocol developed by Breton and De Leeuw( Reference Breton and De Leeuw 14 ). Five electronic databases (PubMed, Informit, CINAHL, PsycINFO and Scopus) were searched using the inclusion and exclusion criteria specified in Table 2. The application of the criteria resulted in 147 eligible papers progressing to the next stage in the screening process.
PRISMA 2009 flow diagram
Inclusion and exclusion criteria for the identification of journal articles reporting on policy research
These papers were read fully by the first author (K.C.) and assessed as to whether barriers to and enablers of influencing public health nutrition policy were identified. When there was uncertainty as to whether a paper fit the inclusion criteria, it was reviewed by a second author (D.G.). Disagreements were resolved by discussion until consensus was reached. As a means of controlling for comparable systems of policy making, a second screening excluded all papers focused on low-income and/or partial or non-democratic countries as defined by the Democracy Index 2014 (n 19)( 15 ).
Analysis
The Framework Method( Reference Ritchie and Lewis 16 ) was adapted to guide the analysis of the data. This is a systematic thematic analysis process which identifies commonalities and differences in qualitative data before focusing on relationships between different parts of the data( Reference Gale, Heath and Cameron 17 ). Papers were coded line by line by one investigator (K.C.). Initial codes were a mix of open and predefined codes informed by the following political science policy process theories: the Advocacy Coalition Framework; the Punctuated-Equilibrium Theory; and Multiple Streams Theory( Reference True, Jones and Baumgartner 3 , Reference Kingdon 4 , Reference Sabatier 9 ). The predefined codes from the literature included: policy entrepreneur; policy window; and coalition. The remaining codes were deduced from the data. After coding six papers a working analytical framework, which incorporated the codes, was applied to each paper. However, there was flexibility to add new codes if needed. The qualitative data analysis software NVivo 10 (QSR International; 2012) was used to manage the data and to generate the analytical framework.
Once the analytical framework had been applied, the characteristics of and differences between the data were identified. This was followed by mapping the connections between these categories and identifying the central themes that emerged which characterised and represented the key enablers and barriers. Throughout this process, the range of attitudes and experiences, including deviant cases for each theme, was considered. As this was an interpretive review( Reference Dixon-Woods, Bonas and Booth 13 ), no attempt was made to undertake a content analysis in order to differentiate the themes in terms of dialogue frequency and/or importance.
Results and discussion
Screening resulted in sixty-three studies included for analysis. The characteristics of these studies can be found in a previous paper( Reference Cullerton, Donnet and Lee 10 ). Numerous themes were identified explaining the barriers to and enablers of policy change, all of which fell under the overarching category of ‘political will’. Political will reflects whether or not decision makers (politicians or senior bureaucrats) were supportive of policy change. This was underpinned by a second major category, ‘public will’, which reflects the mood and policy preferences of voters. Public will was a major component of political will, as politicians were often not motivated to act on an issue that had little public interest or could cause a backlash from wider society( Reference Kersh and Morone 18 ).
Four themes emerged as barriers and six as enablers for political will, while three themes emerged as barriers and four as enablers for public will. There was high concordance between the themes connected to influencing public and political will (see Fig. 2).
The barriers to and enablers of political and public will
Barriers to change in policy making
The barriers to policy change included: the rise of neoliberal ideology; pressure from industry; lack of knowledge, skills and resources from health advocates; and government silos. These themes all fell under the overarching theme of political will and all, except government silos, were also applicable to the second overarching theme of public will.
The rise of neoliberal ideology
The influence of neoliberal ideology in policy decisions was evident throughout the literature. Several papers referred to the prioritising of economic prosperity as the main barrier to policy change( Reference Kersh and Morone 18 – Reference Ceccarelli 23 ). This was demonstrated by a lack of support for regulatory intervention, with the justification that it could interfere with market-driven economies( Reference Shill, Mavoa and Allender 24 ). Furthermore, the cost of implementing proposed solutions and monitoring compliance was also cited as a concern to governments( Reference Vogel, Burt and Church 20 , Reference Shill, Mavoa and Allender 24 , Reference Crammond, Van and Allender 25 ). Specifically, many countries now require government departments proposing regulation to demonstrate that they are cost-saving; that is, the health benefits directly attributed to the regulation will exceed the direct or indirect costs to business( Reference Shill, Mavoa and Allender 24 , Reference Crammond, Van and Allender 25 ). This was problematic for many public health nutrition initiatives as evidence around cost-effectiveness did not exist or was difficult to determine due to the complexity and multifactorial nature of nutrition problems( Reference Crammond, Van and Allender 25 – Reference Nathan, Develin and Grove 27 ).
Prioritisation of government policies that resulted in short-term economic benefit, over policies related to longer-term health outcomes, was also evident in the literature. This prioritisation was apparent when agricultural policies based on subsidies raised the relative price of healthy foods and lowered the price of unhealthy foods( Reference Kersh and Morone 18 , Reference Ceccarelli 23 , Reference Shill, Mavoa and Allender 24 , Reference Nestle 28 , Reference Milio 29 ). In addition, conflicted decision making was seen when the government department responsible for protecting agricultural producers was the same department advising the public about dietary intake( Reference Nestle 28 ). This conflict was evident when the US Department of Agriculture, which had a dual mandate for health and agriculture, under pressure from meat producers, changed federal dietary advice from ‘decrease consumption of meat’ to ‘have two or three (daily) servings’( Reference Nestle 28 ). Even within health departments, there were competing agendas which prioritised certain nutrition issues over others, for example food safety over food security( Reference Seed, Lang and Caraher 30 ).
Some authors identified the overriding desire by governments to ensure personal freedoms was a driver in eschewing regulation or ‘nanny state’ options. As Campbell( Reference Campbell 31 ) pointed out, proposing legislation as a solution creates tension within nations with a liberal tradition of personal responsibility, individual choice and free markets. This ideological conflict was identified by several authors who documented a change in government resulting in food and nutrition policies being watered down or rescinded on the basis of the revised policy being more industry friendly, or because of personal responsibility arguments( Reference Swinburn and Wood 5 , Reference Caraher, Carey and McConell 22 , Reference Corvalán, Reyes and Garmendia 26 ). It should be noted that the ideology of personal responsibility is an entity in its own right albeit strongly influenced by industry as a strategy to increase consumption( Reference Brownell and Warner 32 ).
Pressure from industry
Pressure from industry was one of the main barriers cited for policy change not occurring. This came in a number of forms, including intense lobbying; creation of scientific uncertainty; industry–government or professional body partnerships; and influencing cultural norms.
Intense lobbying. The most noticeable form of pressure from industry was ‘intense lobbying’( Reference Swinburn and Wood 5 , Reference Caraher, Carey and McConell 22 , Reference Ceccarelli 23 , Reference Corvalán, Reyes and Garmendia 26 , Reference Nestle 28 , Reference Campbell 31 , Reference Gilson Sistrom 33 ). Although it should be noted that several authors referred to industry ‘using their influence’, there was no elaboration on what this involved( Reference Caraher, Carey and McConell 22 , Reference Ceccarelli 23 ). Dodson et al.( Reference Dodson, Fleming and Boehmer 34 ) elaborated by stating that lobbyists were ever present and that they ‘watch for issues a lot more carefully than public interest groups’. Gilson Sistrom( Reference Gilson Sistrom 33 ) concurred with this and stated that through their higher capacity and resources, industry lobbyists were able to ‘dog’ legislators every day. In addition, it was noted that industry had a lot more points of interaction with government than health organisations as providers of tax revenue, major employers, through international linkages, and as holders of specialised knowledge( Reference Field and Gauld 35 ). Other food industry strategies mentioned included contributing directly to political parties or individual politicians( Reference Nestle 28 , Reference Freudenberg 36 , Reference Hobbs, Ricketts and Dodds 37 ) and utilising high-level contacts( Reference Shill, Mavoa and Allender 24 , Reference Brownell and Warner 32 ).
Creation of scientific uncertainty. Another form of pressure evident in the literature was industry claiming there was insufficient evidence on effective interventions and that more evidence was needed before further action occurred( Reference Nathan, Develin and Grove 27 , Reference Brownell and Warner 32 , Reference Field and Gauld 35 , Reference Nestle 38 ). This strategy generated a level of uncertainty regarding evidence and tended to reduce the authority of health professionals advocating for action( Reference Nathan, Develin and Grove 27 , Reference Brownell and Warner 32 , Reference Field and Gauld 35 ). Another strategy was for industry to fund scientific research to support its viewpoint or to spread unsubstantiated misinformation( Reference Brownell and Warner 32 , Reference Nestle 38 , Reference Shelley 39 ). Consequently, this allowed industry to control how the issue was portrayed to policy makers and the general public.
Industry–government or professional body partnerships. Forming associations with professional bodies and government agencies allowed industry to build citizens’ trust in systems they perceived to be ethically flawed or puzzling( Reference Dixon, Sindall and Banwell 40 ). Badging or developing a formal relationship with a nutrition organisation was seen as a third-party endorsement for a company and gave it legitimacy to intervene in the regulatory environment( Reference Brownell and Warner 32 , Reference Dixon, Sindall and Banwell 40 ). Industry–government partnerships allowed industry to shape a supportive regulatory environment that did not impact on corporate profits( Reference Brownell and Warner 32 , Reference Field and Gauld 35 , Reference Nestle 38 , Reference Dixon, Sindall and Banwell 40 , Reference Panjwani and Caraher 41 ). A more concerning aspect of this was where public policy was not simply influenced but actually co-created and delivered by the private sector, for example the ‘Change4Life’ campaign in the UK( Reference Panjwani and Caraher 41 – Reference Norum 43 ).
Influencing cultural norms. Industry’s pursuit of competitive advantage requires the active shaping of the cultural environment in which food choices are made( Reference Dixon, Sindall and Banwell 40 , Reference Miller and Harkins 42 ). Consequently, industry invested considerable resources to control the frame around nutrition problems and solutions. Framing is how an issue is portrayed, often using media to influence popular and elite opinion( Reference Shelley 39 ). One effective way issues were framed and reported in the media was that regulation violates freedom of expression and limits personal responsibility( Reference Brownell and Warner 32 , Reference Field and Gauld 35 ). This effectively links in with the pursuit of the neoliberal agenda and increases the value of industry as a government ally. A further strategy to influence cultural norms was manufacturing public sentiment through the use of ‘fake’ citizen groups funded by the food industry, for example ‘The Obesity Awareness and Solutions Trust’( Reference Miller and Harkins 42 ). Miller and Harkins( Reference Miller and Harkins 42 ) concluded that this strategy was effective because it plays on the fears of politicians of a backlash from wider society.
Lack of resources, skills and knowledge from health advocates
The limited resources, money and time of health advocates were cited as significant barriers to policy change( Reference Vogel, Burt and Church 20 , Reference Dodson, Fleming and Boehmer 34 , Reference Hobbs, Ricketts and Dodds 37 ). The volunteer advocates in the Oregon Healthy Foods Schools Bill felt themselves to be at a distinct disadvantage against paid, full-time lobbyists. They did not have the time or money to directly lobby decision makers and adding to this disadvantage was that they did not understand the policy and political processes( Reference Gilson Sistrom 33 ).
Confusion over the policy-making process was a key barrier for health advocates and components of this included: not knowing the entry point or responsibility of government departments for certain nutrition issues( Reference Allender, Gleeson and Crammond 21 ); not understanding that politicians usually decide on whether policy goes ahead, not the bureaucrats( Reference Crammond, Van and Allender 25 ); the importance of a uniform voice( Reference Gilson Sistrom 33 ); the dogged persistence that is required( Reference Gilson Sistrom 33 ); and the hierarchy of power within government departments( Reference Shill, Mavoa and Allender 24 ). Also, some advocates’ inability to compromise meant that they missed out on opportunities for policy progression( Reference Gilson Sistrom 33 , Reference Hobbs, Ricketts and Dodds 37 ).
Poor communication of the solution. The most commonly identified communication issue in the literature was advocates not comprehending how little time policy makers have to understand an issue( Reference Vogel, Burt and Church 20 , Reference Allender, Gleeson and Crammond 21 , Reference Sadler, Gilliland and Arku 44 , Reference Craig, Felix and Walker 45 ). Food and nutrition policy is complex and multifactorial; however, policy makers are often required to make decisions with limited time. Providing reams of evidence or using complex language was not helpful for policy makers, as most do not have the time required to summarise and understand best practice solutions( Reference Vogel, Burt and Church 20 , Reference Allender, Gleeson and Crammond 21 , Reference Craig, Felix and Walker 45 , Reference Gollust, Kite and Benning 46 ). Furthermore, ineffective framing of the issue weakened arguments and shifted the focus away from interventions or policies of worth( Reference Shelley 39 ).
Government silos
Dietary intake is influenced by a complex set of factors and as a result the responsibility is spread across different government departments and different levels of government( Reference Ceccarelli 23 , Reference Crammond, Van and Allender 25 ). Coordinating various departments to work together can be challenging as they tend to work in ‘silos’ and prioritise their own objectives( Reference Vogel, Burt and Church 20 , Reference Shill, Mavoa and Allender 24 ). Furthermore, some government departments can take a narrow perspective of their responsibilities, for example: local government identifying its sole role in food policy as the regulation of food hygiene and safety( Reference Allender, Gleeson and Crammond 21 ); and Departments of Agriculture seeing food only as a product to grow, sell and export( Reference Caraher, Carey and McConell 22 , Reference Milio 29 , Reference Kugelberg, Jönsson and Yngve 47 ). These narrow perspectives were also reinforced by a broader societal mindset around a biomedical approach to health; that is, that treatment and management take priority over prevention( Reference Allender, Gleeson and Crammond 21 ).
Enablers of policy change
As highlighted in Fig. 2, the literature pointed to a wide array of enablers that can be harnessed to improve the likelihood of positive policy change. All were focused on building political and/or public will, as detailed below.
Develop a well thought-through solution
A key enabler of policy change that emerged from the literature was providing policy makers, and often the general public, with a well thought-through solution( Reference Lyn, Aytur and Davis 12 , Reference Crammond, Van and Allender 25 , Reference Blay-Palmer 48 , Reference Johnson, Payne and McNeese 49 ). The most successful solutions had several components: clear, costed strategies on how to solve the problem; single or incremental strategies; and a local focus.
Clear, costed strategies on how to solve the problem (technical feasibility). Many of the included studies described the importance of evidence in the policy-making process. However, several caveats were provided regarding the need to ensure evidence is useful for policy makers (see Table 3).
Making evidence useful for policy makers
Offer single or incremental solutions. A potential source of confusion for advocates in the complex field of public health nutrition is whether to put forward a package of interventions which will have a far greater impact on health status, or put forward one, manageable intervention which would have a more limited impact( Reference Crammond, Van and Allender 25 ). All but one of the studies that examined this issue suggested one single intervention or incrementalism had a far greater chance of enactment( Reference Dodson, Fleming and Boehmer 34 , Reference Eyler, Nguyen and Kong 50 – Reference He, Brinsden and Macgregor 52 ). Opposing this view was one study that suggested providing a package of interventions was more effective, as the government requirements of cost-effectiveness could not be demonstrated with just one intervention( Reference Crammond, Van and Allender 25 ). It was also suggested that advocates should always have a list of backup ideas to allow flexibility during negotiations or when a different window of opportunity opens( Reference Lyn, Aytur and Davis 12 ).
Start locally. Starting locally was found to be another way to present a more manageable proposal to policy makers. Local advocacy initiatives were more effective in achieving their objectives than national ones( Reference Yeatman 51 , Reference Freudenberg, Bradley and Serrano 53 ). This success may be because they set more modest goals, defined a more manageable scope of activities, or the institutional structures of local government are not as complex with fewer vested interests than at the state or national level. Compared with state- and nationally focused initiatives, local initiatives were able to leverage funding and support through different, often less competitive routes( Reference Bedore 54 ). However, while this strategy can be successful at a local level, there was limited evidence of success at ‘scaling up’ to the national policy level( Reference Kjærnes 55 ).
Build relationships with key stakeholders
Gaining support from a wide range of policy makers and key stakeholders increased the chance of legislation passing or a policy solution being supported( Reference Lyn, Aytur and Davis 12 , Reference Dodson, Fleming and Boehmer 34 , Reference Eyler, Nguyen and Kong 50 , Reference Bedore 54 , Reference Bedore 54 ). Stakeholder support to which decision makers responded positively included their constituencies and colleagues, including public officials, community leaders and interest groups( Reference Lyn, Aytur and Davis 12 , Reference Eyler, Nguyen and Kong 50 , Reference Bedore 54 , Reference Shelley, Ogedegbe and Elbel 56 ). Consequently, those advocates who did not engage with key stakeholders found their position was weakened( Reference Field and Gauld 35 ). By building relationships with key stakeholders, advocates developed a greater understanding of the wider environment and were able to identify policy opportunities, tailor policy options to the existing political environment, and gauge the likelihood of success( Reference Lyn, Aytur and Davis 12 , Reference Field, Gauld and Lawrence 57 ).
Form diverse and well-connected coalitions. Forming coalitions encompasses formally bringing together interest groups and organisations, and sometimes politicians, to strengthen the voice on an issue and influence policy change. Establishing broad engagement and commitment across a mix of stakeholders, particularly non-traditional alliances, for example agricultural and producer interests, contributed to the building of a strong coalition which was able to exert prevailing influence on policy decisions( Reference Lyn, Aytur and Davis 12 , Reference Seed, Lang and Caraher 30 , Reference Dodson, Fleming and Boehmer 34 , Reference Kugelberg, Jönsson and Yngve 47 , Reference Eyler, Nguyen and Kong 50 , Reference Yeatman 51 , Reference Freudenberg, Bradley and Serrano 53 , Reference Shelley, Ogedegbe and Elbel 56 , Reference Johnson, Cheadle and Podrabsky 58 ). One example of this was the Washington Nutrition Obesity Policy Research and Evaluation Network which included researchers, public health practitioners, and advocates from nutrition, agriculture, law, economics, public policy, epidemiology and urban design( Reference Johnson, Cheadle and Podrabsky 58 ). Another advantage of fostering diversity was that all parties can provide complementary skills, resources, and new strategies and contacts to move towards the common policy goal( Reference Lyn, Aytur and Davis 12 , Reference Vogel, Burt and Church 20 , Reference Yeatman 51 , Reference Freudenberg, Bradley and Serrano 53 , Reference Johnson, Cheadle and Podrabsky 58 , Reference Freudenberg, Mcdonough and Tsui 59 ). Furthermore, coalitions that included groups with access to key points of government and decision making were more likely to have their issue placed on an agenda( Reference Lyn, Aytur and Davis 12 ). Others highlighted the advantages of the often contentious strategy of forming a coalition with industry to reach mutually agreeable solutions on nutrition issues( Reference Johnson, Payne and McNeese 49 , Reference Pollard, Lewis and Binns 60 ).
Choose one common goal. When issues become linked based on a common goal, this creates synergy rather than competition and increases the likelihood of policy change( Reference Freudenberg, Mcdonough and Tsui 59 ). Conversely, disunity among interest groups creates confusion and weakens each interest group’s position. Such disunity, and the ensuing lack of change, was evident in two papers: the debate on a proposed tax on sugar-sweetened beverages, where hunger activists said the tax unfairly hurt the poor( Reference Freudenberg, Mcdonough and Tsui 59 ); and legislation to improve school meals in Oregon, which was met with opposition from parents, teachers and food-service professionals as they wanted to protect profits and keep local control( Reference Gilson Sistrom 33 , Reference Dodson, Fleming and Boehmer 34 ). Disunity through a lack of coordination was identified as a barrier by Field and Gauld( Reference Field and Gauld 35 ) in a government inquiry into food marketing to children where multiple advocacy groups participated. However, there was little coordination in their advocacy efforts, which weakened their overall position, resulting in failure to influence the outcome of the inquiry.
Use emotions and values
While using evidence was cited as important, other factors, such as using stories and emotion to engage values, can be equally, if not more, compelling( Reference Lyn, Aytur and Davis 12 , Reference Shelley 39 , Reference Gollust, Kite and Benning 46 ). A popular technique for building support and political will among policy makers and the general community by way of emotions and values was framing. Small changes in how an issue is presented can significantly alter how an issue is understood, which can result in a change in public opinion as well as the policy-making environment( Reference Shelley 39 , Reference Gollust, Kite and Benning 46 , Reference Bedore 54 , Reference Jou, Niederdeppe and Barry 61 ). The most effective frames relied on simple messages that were easy for the general public and politicians to grasp( Reference Norum 43 , Reference Craig, Felix and Walker 45 ). Frames that advocates successfully used to build policy support included: protecting the health of children; truth and honesty; fairness and social justice; and urgent crisis or a looming evil( Reference Kersh and Morone 18 , Reference Freudenberg, Bradley and Serrano 53 , Reference Kurzer and Cooper 62 , Reference Hawkes, Ahern and Jebb 63 ). Safeguarding children seemed particularly effective with neoliberal governments where there was concern about excessive government control over individuals and food choices( Reference Hawkes, Ahern and Jebb 63 ).
Be visible
To be effective, advocates must first have their issue acknowledged by policy makers as a problem worthy of attention. Being visible is one way to get your issue acknowledged. Visibility can come in a variety of forms but often requires high-level communication skills and creative solutions( Reference Kersh 64 ). The various techniques highlighted in the literature are presented below.
High-profile event or report. A large number of studies found that attendance by policy makers at a high-profile event or policy makers reading a prominent, salient report was the trigger point for initiating policy action on an issue( Reference Vogel, Burt and Church 20 , Reference Corvalán, Reyes and Garmendia 26 , Reference Nestle 28 , Reference Craig, Felix and Walker 45 , Reference Kugelberg, Jönsson and Yngve 47 , Reference Yeatman 51 , Reference Kurzer and Cooper 62 , Reference Hawkes, Ahern and Jebb 63 ). Creating documents or events that either raised awareness of an issue of which policy makers previously were unaware or/and clearly outlined steps to solve the problem were considered most effective. Effectively framing the problem and utilising international data or targets as a comparison also came up as important ways to attract attention and encourage action to be taken( Reference Lyn, Aytur and Davis 12 , Reference Craig, Felix and Walker 45 , Reference Kugelberg, Jönsson and Yngve 47 , Reference Kurzer and Cooper 62 , Reference Hawkes, Ahern and Jebb 63 ).
Media engagement or campaign. Advocacy coalitions that engaged with the media frequently were more likely to achieve policy change( Reference Lyn, Aytur and Davis 12 , Reference Dodson, Fleming and Boehmer 34 ). However, it was noted that engaging the media can be challenging for health advocates due to a lack of media skills and experience( Reference Lyn, Aytur and Davis 12 ). Specific strategies that were successful included leveraging media coverage off national and international stories( Reference Bedore 54 ) and using high-profile organisations or individuals to attract media attention( Reference Kersh and Morone 18 , Reference Kurzer and Cooper 62 ). Using high-profile organisations or individuals also provided credibility to issues with policy makers and the general public( Reference Kersh and Morone 18 , Reference Kjærnes 55 , Reference Kurzer and Cooper 62 ).
Public mobilisation. Mobilising wider society to demonstrate support for an issue led to increased public and political will. Historically, this has been a frequently used strategy of health advocates in the form of rallies, mass email communications and boycotts( Reference Freudenberg 36 , Reference Freudenberg, Bradley and Serrano 53 ). This is a very visible way of demonstrating public support to policy makers. Alternatively, public surveys and interviews with key opinion leaders can also demonstrate the support of the public( Reference Lyn, Aytur and Davis 12 ). Providing public education or community workshops about the topic at hand and policy solutions may be required before the public engages and becomes mobilised( Reference Lyn, Aytur and Davis 12 , Reference Vogel, Burt and Church 20 , Reference Dodson, Fleming and Boehmer 34 , Reference Kugelberg, Jönsson and Yngve 47 , Reference Jou, Niederdeppe and Barry 61 ). Conversely, failing to mobilise community support was identified as a key reason for a policy issue not getting policy makers’ support( Reference Gilson Sistrom 33 ).
Engage a policy entrepreneur or develop skills of advocates
The value of a ‘policy entrepreneur’, described as an ‘advocate for proposals or for the prominence of an idea, in progressing policy change’( Reference Kingdon 4 ), was highlighted in many studies. Often the policy entrepreneur was in a position of power, for example a local government chief executive officer or a politician( Reference Allender, Gleeson and Crammond 21 , Reference Seed, Lang and Caraher 30 , Reference Craig, Felix and Walker 45 , Reference Bedore 54 , Reference Hawkes, Ahern and Jebb 63 , Reference Loureiro and Freudenberg 65 ), but not always( Reference Vogel, Burt and Church 20 , Reference Dodson, Fleming and Boehmer 34 , Reference Yeatman 51 , Reference Powles, Wahlqvist and Robbins 66 ). Regardless of position, policy entrepreneurs were seen as effective conduits for disseminating evidence to legislators and for promoting recommended strategies( Reference Vogel, Burt and Church 20 , Reference Kugelberg, Jönsson and Yngve 47 , Reference Eyler, Nguyen and Kong 50 , Reference Yeatman 51 , Reference Kurzer and Cooper 62 ). The skills of the policy entrepreneur were often extolled in the literature, these included: their analytical, strategic and political policy skills; high-level communication skills which enabled them to negotiate with, influence and encourage consensus among high-level stakeholders; and their general passion, enthusiasm and vision for the issue at stake( Reference Vogel, Burt and Church 20 , Reference Kugelberg, Jönsson and Yngve 47 , Reference Yeatman 51 , Reference Kurzer and Cooper 62 , Reference Loureiro and Freudenberg 65 , Reference Good, Hammond and Martin 67 ).
Other attributes of a successful policy entrepreneur were the need to be flexible, adaptive and persistent. Practically, this meant being able to refine policy solutions in line with the politics of the time( Reference Caraher, Carey and McConell 22 , Reference Kugelberg, Jönsson and Yngve 47 ) and being open to compromise by offering alternative solutions that were mutually agreeable between interested parties( Reference Gilson Sistrom 33 , Reference Hobbs, Ricketts and Dodds 37 ). Moreover, this required policy entrepreneurs to be nimble enough to take advantage of new political opportunities when they arose and to understand the strategies and end goal of their opposition( Reference Vogel, Burt and Church 20 , Reference Gilson Sistrom 33 , Reference Blay-Palmer 48 ). Finally, policy making is a long process and so persistence was required( Reference Kugelberg, Jönsson and Yngve 47 , Reference Good, Hammond and Martin 67 ).
Understand the policy-making process
Lack of understanding of the policy-making process emerged as a barrier to policy change in the literature; however, key strategies were identified to overcome this.
Ensure solutions are politically palatable. For a solution to be politically palatable it needed to align with at least one goal of the government as well as the prevailing political ideology( Reference Crammond, Van and Allender 25 , Reference Dixon, Sindall and Banwell 40 , Reference Blay-Palmer 48 ). If a policy does not align with the prevailing neoliberal view, it is unlikely to make it on the agenda despite the advocate’s best efforts. Additionally, interventions that are ‘announceable’ – that is, they are not contentious and the government receives favourable press coverage – needed to be offered( Reference Nathan, Develin and Grove 27 , Reference Bedore 54 ). Clearly demonstrating community benefit increased the political palatability of a policy solution( Reference Allender, Gleeson and Crammond 21 , Reference Nathan, Develin and Grove 27 , Reference Bedore 54 ). Finally, avoiding all conflict with, or criticism of, the policies of government ensured an advocate or organisation remained politically neutral( Reference Bedore 54 ). However, as pointed out by Bedore( Reference Bedore 54 ) this strategy required advocates to relinquish the opportunity to be critical of the ethics of decision making and the existing power structures that impact on the food system, social welfare and health.
Responding to the desire to keep wider society ‘on board’, policy makers tended to choose interventions which were the least intrusive and thereby least likely to cause disquiet among target populations( Reference Allender, Gleeson and Crammond 21 ). There was a preference for voluntary action solutions to avoid battles between stakeholders and long regulatory processes to enact such laws. In the studies reviewed, policy makers were more supportive of non-legislative solutions, despite the effectiveness of these being limited( Reference Ceccarelli 23 , Reference Shill, Mavoa and Allender 24 , Reference Nathan, Develin and Grove 27 , Reference Field and Gauld 35 , Reference Craig, Felix and Walker 45 , Reference He, Brinsden and Macgregor 52 , Reference Good, Hammond and Martin 67 , Reference Cawley and Liu 68 ). Particularly popular solutions were: health and nutrition education in schools( Reference Craig, Felix and Walker 45 , Reference Eyler, Nguyen and Kong 50 , Reference Cawley and Liu 68 ); developing and promoting nutrition guidelines( Reference Ceccarelli 23 ) or taskforces( Reference Swinburn and Wood 5 , Reference Nathan, Develin and Grove 27 , Reference Craig, Felix and Walker 45 , Reference Eyler, Nguyen and Kong 50 ); cooking classes and community vegetable gardens( Reference Good, Hammond and Martin 67 ); school canteen recommendations/toolkits( Reference Nathan, Develin and Grove 27 , Reference Craig, Felix and Walker 45 ); voluntary salt reduction( Reference He, Brinsden and Macgregor 52 ); and collaborations with industry( Reference Shill, Mavoa and Allender 24 , Reference Field and Gauld 35 , Reference Panjwani and Caraher 41 ).
Strategically target decision makers. Advocates who did not understand the policy-making process often did not know who to target. Several strategies were identified to overcome this (see Table 4).
Strategies to strategically target decision makers
Be nimble (ready for the policy window). Successful policy change often hinges on understanding how and when the processes of problem recognition, policy development and political decision making converge. The coupling of these domains is commonly referred to as the opening of a policy window, representing an advantageous time for policy change( Reference Lyn, Aytur and Davis 12 ). A theme throughout the papers was that advocates need to be nimble and ready for the policy window( Reference Kersh and Morone 18 , Reference Caraher, Carey and McConell 22 , Reference Shelley 39 , Reference Gilson Sistrom 33 , Reference Freudenberg, Bradley and Serrano 53 ). Identifying and being prepared to take advantage of new political opportunities when they arise is crucial as missing that window will mean the policy solution will fall on deaf ears( Reference Kersh and Morone 18 , Reference Caraher, Carey and McConell 22 , Reference Dodson, Fleming and Boehmer 34 ). Alternatively, it was important to be realistic when the timing was not right. In times of budget crisis at a local or national level, there seemed to be little support for providing funding for nutrition solutions( Reference Ceccarelli 23 , Reference Milio 29 , Reference Freudenberg, Bradley and Serrano 53 ).
Threaten litigation. If all else fails, three papers identified threatening corporations with litigation as a powerful strategy to influence policy change( Reference Kersh and Morone 18 , Reference Freudenberg 36 , Reference Freudenberg, Bradley and Serrano 53 ). However, threat of litigation usually results in policy change from private industry rather than mandated government change.
Study limitations
As with any systematic review, the scope of the initial literature search and publication bias may have resulted in relevant papers being missed. In addition, only papers that were peer-reviewed from high-income democratic countries were included in the review. The authors acknowledge important work in policy advocacy has occurred in developing countries and the grey literature; however, it was not within the scope of the review to incorporate those findings.
Despite these limitations, a rapid appraisal of such alternative literature suggests that, had the search been broadened, the results would not have altered in any significant way. In addition we believe that the synthesised data present a compelling guide for nutrition advocates who wish to better influence nutrition policy making in the future.
Conclusion
The frustration expressed by advocates around lack of policy change in nutrition frequently stems from their belief that policy making is a rational process in which evidence is used to assess the relative costs and benefits of options( Reference Field, Gauld and Lawrence 57 ). The present synthesis of the literature confirms that the reality is far from this. Consistent with the perspectives of policy scholars, the findings from the current review illustrate that evidence is only one component of many influencing policy change. For policy change to occur there needs to be political will, often underpinned by public will, for the proposed policy problem and solution. This requires policy entrepreneurs to work with coalitions of stakeholders to frame policy problems and communicate politically palatable solutions both to decision makers and the general public.
Many of the barriers identified in the review are consistent with those identified in political science theories, particularly Multiple Streams Theory, Punctuated-Equilibrium Theory and the Advocacy Coalition Framework. Most of the barriers were macro or structural issues, including the rise of neoliberal ideology, pressure from industry and government silos. These are issues that an advocate generally cannot change; however, it is imperative to acknowledge and understand these issues and consider how to work with them.
The suite of enablers presented in the current review highlights several areas where advocates can attempt to influence political and public will and ultimately change policy. These include framing problems based on values and emotions, and effectively coupling problems with well-considered solutions. It is important to appreciate that no one method will give results on its own and equally that undertaking all the documented strategies will not necessarily result in policy change, as policy making is often a complex and irrational process. However, the literature highlighted that the greater the variety of enabling strategies undertaken, the more likely advocates will be able to effect policy change. Further research into the effectiveness and impact of these strategies, particularly around using values and emotions in the context of public health nutrition policy, is warranted.
Despite the fact that few nutrition policy studies utilised political science theory in their analyses( Reference Cullerton, Donnet and Lee 10 ), the enablers and barriers identified in the present review are consistent with political science theories which have been empirically tested in other fields of policy. The review has demonstrated that these theories are relevant to public health nutrition policy; therefore researchers and advocates are encouraged to incorporate the use of these theories to assist in refining their advocacy practice.
Advocacy is frequently said to be a combination of science and art. The current review has considered the ‘science’ behind what barriers to and enablers of policy change have been identified in the nutrition policy literature. However, it has also identified that effectively influencing policy change relies heavily on the ‘art’ of advocacy where nimbleness, flexibility, persistence, communication skills and the ability to identify the policy window opening come to the fore. The present review provides health professionals with a greater understanding of the barriers to and enablers of policy change, so that future advocacy efforts are more targeted with interventions tailored to address specific barriers and to make use of specific enablers.
Acknowledgements
Acknowledgements: The authors wish to acknowledge the guidance provided by Jennifer Browne, Karen Topp and Martin Reece in preparation of the manuscript. Financial support: K.C. is supported by an Australian National Health and Medical Research Council (NHMRC) Postgraduate Scholarship. Conflict of interest: None. Authorship: K.C. searched and reviewed the literature and drafted the manuscript. D.G. assisted in reviewing the literature. A.L., T.D. and D.G. offered valuable input and feedback on the manuscript. Ethics of human subject participation: Not applicable.
