Alcohol policy has, for too long, spurned evidence or evaluation. Scotland is showing England and Wales the way forward, prioritising monitoring and evaluation of new policies to help generate better evidence, argues David Humphreys

Government policies to control alcohol have, over the years, set public health researchers a series of daunting challenges. Recent trends in alcohol policy have witnessed policies with strong empirically-based potential for harm reduction, like minimum unit pricing (MUP), withdrawn on grounds of insufficient practical evidence of effectiveness. Yet, over the last 15 years, we have seen a raft of different local policies and initiatives implemented that lack strong evidence of effectiveness (e.g. Alcohol disorder zones, designated public places orders, early morning restriction orders, ‘drunk tanks’ etc). The implementation of poorly evidenced policies is problematic, but not uncommon. However, failure then to create evidence from unproven alcohol policies is a pressing concern and is failing to contribute valuable evidence to improve the knowledge base.

The Licensing Act (2003), the last major liberalisation of alcohol policy, is a good example of such failure.  While there was little evidence to support this legislation in the first place, there has been scant provision to evaluate and understand its impact. What research has been completed was largely planned and conducted retrospectively, limiting the ability of researchers to disentangle the complex impacts of policy on the availability of alcohol and on related social and physical harms.

The 2003 Licensing Act overturned more than 150 years of blanket restrictions on the opening hours to control the harm done by alcohol. The new and controversial legislation enabled licensed premises to apply for licenses to trade at whatever time they thought fit, instead of them all being set largely the same opening hours. This policy shift was driven by the imperatives of public order and crime reduction, rather than by health considerations. It was based on plausible but unproven assumptions. The Government believed the change would naturally stagger pub closing times, reducing incentives for fast ‘drinking up’ and diffusing crowd dispersal from pubs. As a result, there would be fewer groups of drunken people in the streets at once, reducing the environmental trigger points for violence and disorder. However, in contradiction of established public health principles governing alcohol control, the deregulation of closing times would potentially serve to increase the availability of alcohol – a recipe for greater harm.

In an ideal world, a policy change such as this might have been informed by better evidence. But failing this, researchers might have been involved in planning the legislation’s implementation, making evaluation easier. But when it comes to policy on alcohol, at both the local and national level, Britain tends neither to draw on evidence nor think about generating evidence from it. Further legislation, introduced by the Coalition Government serves to empower local authorities to take action on alcohol-related harm. This is likely to create increased local policy experimentation in some areas and provide further opportunities for evidence generation. What is public health research to do in such circumstances?

My goal is to demonstrate how local-level evaluation might be approached using a previous example evaluating the Licensing Act (2003) and to outline what could be done to generate more reliable knowledge for policy in this field. I suggest that it will require greater policy openness to evaluation. There will need to be better pooling and standardisation of locally held data so that it is accessible and useful to research. Improved knowledge will also require better communication among public sector agencies, such as the police and the NHS, so we can better establish links between actions against consumption of alcohol and the outcomes for individual well-being.

In Scotland, there is at least a plan to improve the situation for the future. The Alcohol (Minimum Pricing) (Scotland) Act 2012 was passed in June 2012, though it has not yet been implemented due to a legal challenge led by the Scotch Whisky Association. However, based on the assumption that minimum pricing per unit of alcohol will be introduced, a programme, ‘Monitoring and Evaluation of Scottish Alcohol Strategy’ (MESAS), has been set up to provide ongoing monitoring and evaluation. England and Wales could learn from such research initiatives. We need policy makers to keep more open minds and not to be afraid to gather evidence that might require them to recast policies already implemented.

We could also make it easier to access data that is available. For example, local authorities routinely collect data about licensing but each has a different way of collecting the data about the location of alcohol outlets, their opening times, what activities they promote, whether they have entertainment, food, etc. Every local authority tells the story in a different way. Gathering data to understand the dynamics of alcohol availability is difficult to understand across local government boundaries. If standardised, we could use this information to understand the nature of alcohol availability, how it changes over time and how it is influenced by different policies, allowing a national picture of alcohol availability to be observed at high resolutions.

There is research suggesting that the availability of alcohol is strongly related to harm, but this relationship is complex and there are many unanswered questions, requiring further research. We need ways to find out how the availability of alcohol is distributed across the population, how it changes over time in response to different policies and most importantly, how changing availability impacts on the health and well-being of different groups of people as well as on crime and disorder.  In short, we need to develop better surveillance systems that capture data on both alcohol availability (as exposure) and on related harm outcomes. There are pockets of good practice across the country from which important lessons could be learned. In Cardiff, for example, the police and A&E hospital units share data on alcohol harm, related to crime and to physical injury. This type of public sector collaboration is vital if we are to gain a full picture of the impact of policy on alcohol harm. As researchers, if we are going to secure this type of collaboration, we must also help practitioners by making our evaluations more accessible, so they see and gain the benefits of proper evaluations.

There are hopeful signs of progress. The shift in responsibility for public health to local authorities means that a single tier of government now has an interest in reducing both harms to social order and harms to health caused by alcohol. Further integrations would be helpful. For example, moves to decentralise the NHS and integrate it with local authority services in Greater Manchester offer additional potential synergies. We have a lot yet to understand about the mechanisms that cause alcohol harms, including availability, price and deprivation. Tackling the fragmentation of the public sector may be an important step towards understanding intricate causal pathways and creating more focussed, better evidenced strategies.

David Humphreys is Associate Professor of Evidence-Based Intervention and Policy Evaluation, University of Oxford. He will present ‘Preventing harm with alcohol licensing: can we generate evidence from policy?’ at a SPHR@L seminar on Wednesday, 11th March 2015, 17.15-18.30 at LSHTM (Room G9, 15-17 Tavistock Place, London WC1H 9SH)