The Evidence for Evidence-Based Policy

James Gibson

As I was scrolling through my Facebook feed (an experience which I’m sure every student has shared) I stumbled upon an article that piqued my interest, from Vice.[1] The article was an interview with former Chief Drug Advisor to the UK government David Nutt. Nutt was dismissed as Chair of the Council for the Misuse of Drugs on the grounds that Alan Johnson MP, then Home Secretary, claimed “he cannot be both a government adviser and a campaigner against government policy”.[2] Nutt had expressed some controversial views, including that taking ecstasy was “no worse than horse riding”.[3] The Vice article was sparked by the recent resignation of Norman Baker MP from the Home Office, citing a lack of “support for ‘rational evidence-based policy’”.[4]

In the article, Nutt outlines what he calls an “evidence-based policy for every drug you’re ever likely to take”. There were some expected facts in there – I had long suspected that alcohol would be criticised as dangerous by an outraged Daily Mail*, if it wasn’t an ancient invention – but others that surprised me, chiefly a “personal allowance of about 50 doses per year” for MDMA, or ecstasy.

A successful experiment?

Many people have heard about Portugal’s recent attempt to reduce the possession and abuse of chemical substances [5] by near-total legalisation. A question too rarely asked, however, is “why isn’t the rest of the world following in their footsteps?” Here there is a case where drug legalisation, coupled with an increase in support funding for addicts, led not to an explosion of gang violence and drug tourism,[6] but a stabilisation in abuse and an increasing acceptance of drug users as needing help. So far, so pro-legalisation.

But, as always, it’s not as black and white as that. The architect of Portugal’s policy, Dr Manuel Pinto Coelho, claims that the policy led to an immediate increase in deaths,[7] citing White House statistics.[8] This conflict, described as “considerable”[9] brings to light the greatest weakness of evidence-based policy:, it doesn’t really exist. Any evidence found can be twisted to fit some agenda by partisan politicians, and policies made are usually based more on ideology than evidence.

A gold standard

The “gold standard of medicine”, randomised controlled trials (RCT), are held up as the epitome of what government policy could look like by many, [10] but care must be taken when deciding on their appropriateness. The advantages of RCT are many and obvious to scientists – the removal of effects other than the one being studied and the absence of observer bias to list a few – but the process does have its limitations.

Unlike in medicine, there can be no placebo in most circumstances, as political interventions are, by necessity, seen by the public. The group separations must be geographically localised, so as to ensure boundaries to the citizens affected. Perhaps most importantly, there is a public misconception that RCTs are in some way “unethical” [10]. Shaking this stigma would be necessary for further expansion of RCT policy-testing. As an RCT involves a control group who receive no intervention at all, it can be seen as callous, when instead we could implement a new technique to improve, say, waiting times in NHS general practices. Why should one group reap benefits when the others are left in an old system we have already described as unfit for purpose? Well, because the new system is not guaranteed to be better, and it is only by comparing the effectiveness with no (or a “known”) intervention that we can accurately rank the actions of government. Widely and blindly applying a single policy to a large swathe of the populace without consideration of local variance, or the advantages of alternatives, can in fact be seen as the irresponsible approach.

In addition, there is a concept, within political science, known as the tyranny of the majority. Whilst normally applied to the oppression of minorities within a pluralist society (to protect the interests of those not belonging to that minority) the same argument can be made for important topics of scientific consensus, such as climate change. Since many people within a country are unfamiliar with the level of evidence for these things, their priorities may be skewed towards more immediate problems that they are more familiar with. Compare the 97.2% of climate scientist publications assuming humanity is the cause of climate change to the 57% of Americans who believe such a thing, and you may begin to see a barrier to meaningful action to counteract high greenhouse gas emissions.[11][12]

Scientific dictatorship?

The main problem with evidence-based policy, I feel, is the threat it may pose to democracy in a non-scientifically minded populace. There exists another element – irrational idealism, cynicism, biases, and misconceptions held by all people. In a democracy, it is required that those we vote for share these beliefs with the general populace. Do we really want to live in a society where politicians make decisions with no regard as to how the voters will feel, and purely base their ideology on facts?

While it perhaps is too extreme to say that evidence-based policy would lead to a scientific dictatorship, voters expect to elect an individual to make decisions for them. Truly impartial policy-making would lead to no difference between political parties – in many ways, a pluralist democracy demands that the citizens make a decision, for better or for worse. At some point, we have to let rational human beings choose their poison – to do otherwise would, after a fashion, be a dictatorship.

In addition, gathering the evidence required to fully analyse the situations addressed every day by politicians is a time-consuming process – one that takes considerably longer than the timescale on which most politicians operate . There is something to be said for snap decisions, as they at least reassure the populace that their concerns are being addressed. Any action is better than no action at all. Sometimes.

When we are presented with the facts surrounding evidence-based policy, it is easy to think of it as a binary choice. This is not so, and there is no contradiction in allowing the public to choose on important matters and informing minutiae by a more evidence-based approach, or vice versa. Another approach is to have a single evidence-based party, to allow the people to voice their concerns over issues that they feel demand scientific input. Whilst science absolutely should inform public opinion, it is harder to claim it should shape it entirely. Therefore, we must choose carefully the situations in which we turn to science for answers in the messy world of politics.

* Nutt specifically outlines that alcohol should be a “controlled drug”, available from licensed premises, such as pubs, but not in supermarkets or other convenience stores.


[2] The Guardian, Monday 2nd November 2009



[5] Sources include Forbes
Germany’s Der Spiegel

[6] and The Guardian







The Evidence for Evidence-Based Policy

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s