I have no hesitation in endorsing the view that everyone in the policymaking process should guard against the selective use of science or other analytical evidence to justify pre-chosen policies. The evidence is clear that policies firmly rooted in objective evidence are most likely to be effective in practice and to deliver the intended outcome.

But my light-hearted comment at this year's World Conference of Science Journalists, from which you develop your editorial thesis in The curse of policy-based evidence, was a snapshot in time from some two years ago — before I had been appointed as UK government chief scientific adviser and before I had experience of working directly with and for the UK government.

The process of driving better evidence into policymaking is a journey, not a destination — one on which we have made great strides in the UK.

For example, we now have (or are about to appoint) chief scientific advisors — mainly leading scientists and engineers — in all the main Whitehall departments, thus putting world-class science and engineering evidence at the heart of government decision-making in the UK. These are paralleled by roles of similarly senior colleagues from social research, operational research, economics and statistics who are in each case supported by professional communities within government.

We are also working to improve the flow of information between the academic world and policymakers in the UK. My ten-point action plan responds to the Council for Science and Technology's recommendations made in its report How Academia and Government can work together. Implementing these actions will improve links between government and academia and thus the flow of evidence and advice into policymaking.

Since the inquiry into Bovine Spongiform Encephalopathy (BSE) in the late 1990s, the government has greatly improved the openness and transparency with which it asks for and receives scientific advice, including communicating the risk and uncertainty associated with that advice — allowing science journalists and the public to make their own judgements on evidence-based policy as opposed to policy-based evidence.

How Whitehall departments use science and engineering evidence will remain a key element in my Science and Engineering Assurance reviews. The need for such reviews to provide external scrutiny and benchmarking of the use of science and scientific evidence in policy development and delivery across government was originally identified in the July 2002 Government Strategy for Science, Engineering and Technology.

By 2011 I will have completed reviews of all the main Whitehall departments and these will be followed up by further self-assessment with external scrutiny.

It would be unrealistic, and frankly naive, to imagine that scientific evidence is the only thing that politicians will or should take into account in their decision-making. Scientific evidence should be tempered by complementary views from other analytic disciplines — so, for example, the science must be balanced against the economics and social research.

Hypothetically, the science may point to the benefits of a particular medical intervention, while economic analyses may conclude that it is not cost effective and the social research may indicate that there will be patient resistance and uptake will be low anyway.

The real world situation is well illustrated by the case of cannabis classification in the United Kingdom. It was vitally important that the Home Office be provided with good scientific evidence to inform their policy decisions on the classification of drugs. My role was to ensure this and it is a matter of public record that I agreed with the majority view of the Advisory Council on the Misuse of Drugs that cannabis should not be reclassified.

But science is just one of the factors that ministers consider when reaching a decision. I am content that the Home Secretary fully considered the scientific advice, but in reaching a decision she also had to consider evidence on wider issues such as the uncertain long-term impacts on health, public perceptions and the needs and consequences for policing priorities.

John Beddington