Evidence is most likely to influence public health if it fits the right problem at the right time for the right people

Vittal Katikireddi charts the multiple roles that evidence is playing in transforming minimum unit pricing of alcohol into a policy imperative.

 Discovering a proven solution doesn’t mean we’ll change in a hurry. For a start, we may not care much unless the problem is big enough to warrant the effort and the solution is reasonably straightforward.  Even then we might not bother. Change is awkward. Generally, it has to fit in with everything else we want – and are able – to do.

So, if I showed you how swimming transforms physical and mental health, you still might not bother if you were already running every day and the swimming pool was five miles away. But, if you had just strained your Achilles tendon moving home to a flat next to the pool, you might act differently.

The relationship between evidence and public health policy likewise works in a complex and unpredictable way. Evidence is most likely to be used when it happens to fit the right problem at the right time according to the right people, based on our new conceptual model, developed by studying how the Scottish Government came to legislate for minimum unit pricing of alcohol.

Here is how evidence made a difference. Researchers from Sheffield University produced lots of strong evidence that a 50p minimum price per unit would improve health. They showed that around 300 lives would be saved annually in Scotland and predicted that alcohol consumption would fall by 5.7 per cent, resulting in an estimated reduction of 6,500 hospital admissions each year, north of the border. They reckoned that, over 10 years, accumulated benefits such as reduced absenteeism, increased working, reduced crime and lower health costs, would amount to nearly £942m.

That all sounds impressive. Minimum unit pricing sounds like it has clear health benefits. But it still might not have caught on. Why? Because, for some, it was a solution to something that was not perceived as a big problem. In the 1990s and early 2000s, policy makers were not particularly concerned about general drinking.  They were more interested in more narrowly-focussed solutions tackling public order issues, mainly involving young people, related to alcohol consumption.

But evidence successfully shifted opinion of what the ‘big problem’ was and, at the same time, made a minimum pricing solution more attractive. This vital evidence came from epidemiological research demonstrating that levels of drinking by the population overall are closely linked to the burden of alcohol-related problems. It also showed that both general consumption and ill-health are reduced as prices rise. In short, if prices go up, general consumption typically falls and health improves across the board. This evidence helped reframe the goal of Scottish policy away from encouraging responsible consumption towards reducing general alcohol-related harms.

The knowledge that everyone – not just those affected by binge-drinking youths – can benefit from this wider policy goal also made a broad spectrum policy prescription, such as minimum pricing, more acceptable.

All of this shows how evidence helped to reconceptualise ‘the problem’ and offered highly credible proofs for an appropriate solution.  But turning minimum unit pricing into a policy imperative required an additional factor in its favour. Let’s call it ‘the Scottish context’.

The Scottish context has two elements. First, epidemiological evidence showed not only that that general drinking was a big problem. It was an especially big problem in Britain – liver cirrhosis deaths were falling in western Europe, but rising in England. The increase was even faster in Scotland.

No government run by the Scottish National Party wants to be shamed by Europe and certainly not by England. That’s especially true of a government keen to demonstrate its readiness for independence by heralding its innovative leadership of public health policy (Scotland was the first UK country to ban smoking in public).

Secondly, the powers of the Scottish government are heavily limited. It is not entitled to increase taxes on alcohol. However, it can legislate for a minimum unit price. So, minimum unit pricing fits the broader SNP aim of showing that independence could offer gains for all voters. It supports the SNP’s wish to demonstrate its capacity to use financial instruments effectively for the public good. And minimum unit pricing is probably more effective than taxation, in this case, because it focusses on the biggest part of the problem – heavy drinking of very cheap alcohol.

The laws for minimum unit pricing of alcohol were passed in 2012. The measure has yet to be implemented, because of legal challenges from the drinks’ lobby. Nevertheless, evidence has played vital and multiple roles in shifting public health policy. It has provided the knowledge for what’s most likely to work. At least as important, it has supported advocacy reframing the nature of the alcohol policy problem and linking it to appropriate solutions. In the process, it has aligned minimum unit pricing with the broader political objectives and capacities of the Scottish government.

This complex interplay of evidence, policy making and serendipitous national politics helps to explain why this innovation is likely to be implemented faster than one might expect, and certainly more quickly than in England. The process also highlights some of the many hurdles that must be overcome if one seeks speedy adoption of public health innovations that seem so plainly to benefit health.

Dr. S Vittal Katikireddi is a Clinical Lecturer in Public Health at the MRC/CSO Social and Public Health Sciences Unit, University of Glasgow. His presentation, ‘Minimum unit pricing of alcohol in Scotland: Lessons for evidence-informed policy,’ will be made on Thursday, 20th March 2014, 5:15 pm at Faculty Meeting Room (G9), LSHTM, 15-17 Tavistock Place, London WC1H 9SH. It is part of a series of presentations hosted by the School for Public Health Research @ LSHTM (SPHR@L).