Qualitative case study of local alcohol policy implementation in Scotland

Background: Alcohol use is a major risk factor for ill-health and a contributor to health inequalities. Scottish alcohol-related mortality rates are the highest in the UK, with the greatest burden of alcohol-related harm falling on people from the most deprived areas. The Scottish Government has been ambitious in their policy and legislative efforts to address the problem, collectively known as Scotland’s ‘alcohol strategy’. Although evaluations exist which focus on population-level outcomes and national-level overviews of single aspects of the strategy’s implementation, much less is known about how the broad strategy has been implemented at local level. The focus of this thesis is the strategy in place from 2009-2018, which has been recognised internationally for taking a progressive, ‘whole population approach’ to tackling alcohol-related harm. This thesis explores how local implementation of the alcohol strategy occurred, investigating how the strategy was implemented in practice, and examining challenges facing the local policy actors and partnerships tasked with this implementation work. Methods: To set the scene for data analysis in this thesis, a systematic review of empirical studies on alcohol policy implementation in high-income contexts was conducted to identify factors affecting local implementation of alcohol policies. The substantive part of the thesis then offers a qualitative, embedded case study of alcohol policy implementation in three purposefully selected local areas of Scotland. Data were generated from semi-structured interviews with nine national-level alcohol policy stakeholders and 54 local alcohol policy implementers (63 interviewees in total). Local interviewees were recruited from two key bodies: (i) Alcohol and Drug Partnerships (ADPs), which are dedicated partnerships tasked with local alcohol policy implementation; and (ii) local Licensing Boards, which are independent regulatory bodies comprised of elected Local Councillors and which preside over the local alcohol licensing system, thus controlling alcohol availability. In order to supplement my understanding about the content of the alcohol strategy, aspects of context, roles of relevant actors, and representations of policy decisions, I also undertook documentary analysis of 16 relevant national policies, legislation, and reports. These included documents comprising the alcohol strategy, as well as supporting guidance or more contextual documents. Results: The systematic review identified a range of factors that available research suggest influence alcohol policy implementation, and these were grouped into the following three themes: accountability and governance; evidence use; and context and resources. The first two themes directly informed the research questions, while the third is discussed as a cross-cutting theme of the thesis. Multiple accountability relationships within and across ADPs and Licensing Boards were explored using Hupe and Hill’s (2007) public accountability typology, refined to capture how local implementers experience multiple ‘directional’ (vertical top-down, horizontal, and vertical bottom-up) accountabilities. The findings show that, for ADPs, these accountabilities are characterised by complexity, confusion, and miscommunication, with horizontal accountability within ADPs’ partnership structures appearing particularly challenging. With regards to Licensing Boards, the findings demonstrate that their current accountability system almost exclusively relies on legal accountability arrangements. This makes Licensing Boards’ accountability system distinct from the system for ADPs; the data suggest that these contrasting accountability systems are posing challenges to the implementation of Scotland’s alcohol strategy since the system for Licensing Boards does not appear to impose sufficient obligations on Licensing Boards to prioritise public health in their decision-making. In relation to evidence use, the findings suggest that evidence plays an important but complex role in the implementation of Scotland’s alcohol strategy. Drawing on Lorenc and colleagues’ (2014) idea of ‘cultures of evidence’, the data demonstrate that ADPs and Licensing Boards each have a distinct identifiable ‘culture of evidence’, which is characterised by the varying ways in which actors within these organisations perceive and use evidence. These two cultures of evidence overlapped insofar as members of both organisations reported valuing locally contextualised evidence and each culture was characterised by having a diversity of perspectives within it, reflecting members’ diverse professional backgrounds. However, important differences between these two cultures of evidence were also apparent and potentially help explain why ADP members described struggling to influence Licensing Board decision-making. In addition, the findings indicate that evidence constructed in traditional public health settings (e.g. the NHS and universities) was not perceived to be translating well into either ADPs or LBs. Both the accountability issues, outlined above, and the application of ‘cultures of evidence’, help explain why seemingly relevant public health evidence may have limited traction within Scottish alcohol policy implementation. Each of the results chapters also demonstrates the significance of context and resource constraints in shaping implementation. This included impacting on the capacity of ADPs to fulfil national-level expectations, and reinforcing apprehension among LB members about potentially costly appeals against their decisions. Conclusion: This thesis provides important insights into the implementation of Scotland’s national alcohol strategy, demonstrating that accountability, evidence use, context and resources all shape implementation in important ways. Notably, it suggests that effective implementation is being hampered by inadequate governance arrangements in a complex context characterised by a commitment to multi-sectoral collaboration, in which key challenges to partnership working and limited policy coherence are identified