“Should evidence always dictate policy?”

This was the key question at a recent debate at the Royal Society of Medicine organised as part of the Battle of Ideas, a festival that debates the “core issues of the day” and encourages free thinking and lively exchanges of views.

The panel and audience at this satellite event were ruminating over the role of scientific evidence in making public health policies. Is it vital that only policies based on the most robust evidence available should be implemented? Or can politically or morally driven policies still go ahead when there is little evidence behind them or they’re contradicted by the available evidence? Are politicians getting away with fudging then citing “evidence” to offload responsibility and avoid debate over contentious policies?

First of the panel to speak was Nigel Hawkes, director of Straight Statistics and a columnist for the BMJ. He argued that evidence is necessary for policy making in public health, but not sufficient. It would be foolish to implement a policy that had absolutely no evidence supporting it, but policies are also based on political, moral, and ideological views.

Next up was Dr Amanda Killoran, a public health analyst for the National Institute for Health and Clinical Excellence and co-editor of Evidence-based Public Health: Effectiveness and Efficiency. She agreed with Hawkes that evidence alone is not sufficient, but argued that it does have an important role in legitimising policy.

She is also a strong believer in the importance of government preventative public health initiatives, but not through an authoritarian approach. Although we know many deaths are preventable through changing behaviour, we also know how hard it is to change behaviour. Instead the government should facilitate the environment in which people can make healthy choices.

Timandra Harkness, former director of Cheltenham Science Festival’s FameLab, and co-writer and performer of Your Days Are Numbered: The Maths Of Death, argued that the government should use evidence in an honest way. She pointed out that most unhealthy habits are not entirely awful (bar smoking) but are in fact beneficial to a degree—moderate drinking, for example, can protect against heart disease. However, this point doesn’t appear in public health messages, and often evidence is selectively dished out with a specific agenda.

Dr Jon Rohrer, a member of the Royal Society of Medicine council and a neuroscience researcher and trainee, pointed out that as a doctor he wouldn’t start a treatment without evidence on its efficacy, so why would a public health “treatment” be instituted without evidence? Policy without evidence is simply opinion based policy. However, he suggests, governments hate to be well informed in case the evidence interferes with the political aim of the policy they’re trying to press through.

Last to speak was Liberal Democrat politician and former MP for Oxford West and Abingdon, Evan Harris. He agreed with Harkness on the issue that politicians shouldn’t claim a policy is evidence based when it’s not, citing as an example the recent sacking of government drug adviser David Nutt over his reports on the low risks of various illegal drugs. Rather than saying that Nutt’s policy advice simply wasn’t politically palatable, the government tried to make out that the evidence was wrong in order to withhold their policies and serve the moral views of large portions of the public.

Various topics were discussed when the floor was opened up for questions, one of the main ones being cases in which evidence is “ignored” when it contradicts the political or moral message the government wants to put across—that is, when the evidence says one thing but the government does another

Harkness cited the example of drinking in pregnancy. The evidence suggests that drinking a small amount in pregnancy is not harmful to the foetus, but the public health advice is that women shouldn’t drink during pregnancy at all.

Harris countered that a black and white approach is easier for people to understand and more effective. If you say one or two units of alcohol is OK, you will get more problem drinkers than if you advise that women should not drink during pregnancy. Furthermore, stopping women drinking in pregnancy is for benefit of the baby not the civil liberties of the mother.

Harris called this simplification or well intended exaggeration of the evidence, whereas Harkness called it overriding the evidence because the public can’t be trusted to make their own decisions about their health.

This discussion spiralled into the second main question of the night, which was, essentially, “To what degree should the government try to influence our health behaviours?” Are preventative health initiatives a responsible way of stopping people harming themselves, or do they risk becoming prurient moralism in disguise?

Harkness was a strong believer in giving people the evidence to make up their own minds, as anything else would constitute telling people what to do. Harris, however, believed that one person’s “nanny state” is another’s “nurturing state,” and that the government has a responsibility to protect the vulnerable against the unhealthy behaviour of others (such as via the smoking ban).

As Killoran pointed out, there is a value judgment in doing nothing. The available evidence for a policy is often not equivocal, or doesn’t necessarily apply to every single person on an individual level, but the greater good from such a policy will probably outweigh any negative effects on people’s perception of evidence based policy making or on civil liberties. Governments have to make that call.

As someone who believes strongly in evidence based medicine, it did initially seem absurd to me that people could argue against evidence based public health. This debate did make me think about how politicians might be justified in some cases in skirting over the evidence somewhat in order to make a simple to understand policy or, to a lesser degree, a morally palatable policy. But I’ll still always choose evidence based policy over opinion based policy.

• Read more about this debate in BMJ news at: http://bit.ly/aQLRT0

Helen Jaques is a technical editor, BMJ