A risk-free environment is bad for our health

says Roger Bate, director of science watchdog the European Science and Environment Forum

Environmental organisations and health groups have fed on this concern to promote a particular interpretation of the EU Treaty of Rome's precautionary principle. This is a proposition that technologies and products should not be permitted until we know they pose no danger. The principle has become the categorical imperative of the new health and environmental fascists: thou shalt not tolerate even the risk of a risk.

A UK government review of the dangers from breast implants is due to be completed soon, at about the same time as a review of passive smoking. These reviews were motivated by surveys and precautionary rhetoric, from health lobbies including the Cancer Research Campaign, designed to alarm people about silicone implants and cigarettes.

According to the editor of the prestigious New England Journal of Medicine, Marcia Angell, the public has problems 'in thinking in terms of probabilities, or in acknowledging the possibility of coincidence'. Research, she says, has consistently failed to find a link between silicone and disease. Yes, women who have implants get sick, but, in a typical study, 'the implant group was no more likely to develop connective tissue disease than the group without implants'. Nevertheless, the public is willing to believe the media's simplified exaggerations. For example, in the USA jurors have sided with plaintiffs, and have put Dow Corning, a company producing gel implants, out of business. Now, if Dow Corning was producing defective products, of course it should be sued and potentially bankrupted, but on such flimsy evidence thousands of jobs have been lost.

For women who have had mastectomies there will probably be less chance of corrective surgery, because few companies would take up the poisoned chalice and try to develop and improve silicone gel implants after the destruction of Dow Corning. Baroness Jay, in charge of the implant review, should think about these women before succumbing to the urge to ban.

The picture with tobacco is even worse. A central argument behind designs to extend tobacco regulation is that smoking harms non-smokers. There would be wide support, from me included, for this regulation, if passive smoking were shown to be harmful and not just a nuisance. However, until that evidence is presented, the loss of freedom to the one in three people who smoke would be unacceptable.

The review of passive smoking being undertaken by the UK government is relying heavily on the work of professor Nicholas Wald from the Royal London School of Medicine. Wald and his colleagues recently published a paper in the British Medical Journal on the impact of passive smoking on lung cancer rates. They estimated that there was an increased risk of 24 per cent. The report was accompanied by a press release from minister Tessa Jowell, announcing that these 'shocking figures' had reaffirmed the government's commitment to 'tobacco control'. Segregation of smoking in pubs was immediately announced, and banning sales to under-18s soon followed. Further actions are likely, and, if the Wald figure is accurate, would be supportable. However, the figure is extremely misleading.

According to Robert Nilsson, a senior toxicologist at the Swedish National Chemicals Inspectorate and Professor of Toxicology at Stockholm University, the paper performs a 'statistical trick'. He says that there are so many biases in Wald's analysis that the alleged elevated risk figure is meaningless. For example, Wald relies on data from the memories of spouses as to how much their former (dead) partners used to smoke. Recall survey bias is often considerable (potentially far higher than the 24 per cent estimate of increased risk) but is not even mentioned by the authors. Nilsson also explains that Wald pooled studies, which were not comparable, to increase the sample size, and hence exaggerated the statistical significance of his results. His most stinging criticism, however, is left for the BMJ editorial board, whom he insists must be 'innocent of epidemiology' ever to have published the paper in its existing form.

Nevertheless, like breast implant damage, the media now believes in cancer-causing passive smoking. Consequently, the public accepts it as fact, and focus group responses (the backbone of New Labour policymaking) are broadly in favour of banning smoking in public places and in work environments. An ill-conceived combination of precautionary approach and media hype has led to this unwarranted state of affairs.

Precautionary action not based on sound scientific data could lead to product bans in other areas. For what starts with silicon and tobacco may swiftly move to alcohol, fatty foods, sexual practices or other favourite pastimes - fast cars perhaps.

In the environmental field a good illustration of the precautionary principle at work is the debate on pesticides. Many environmentalists point out that the adverse effects of a pesticide like DDT did not emerge until years after it was first used. Although not harmful to humans it accumulated up the food chain until it killed carnivorous predators such as otters and eagles. Because we did not foresee this problem, they argue, we should ban today's pesticides whose effects tomorrow we also cannot predict. Indeed, Danish officials, under constant lobbying from Denmark's powerful environmental movement, are discussing the possibility of banning all pesticides.

However, it should be remembered that DDT was cheap to produce and replaced older arsenic-based pesticides which were fatal to humans and toxic to the environment. It also rid large parts of Africa and Asia of mosquito-borne diseases like malaria and typhus. DDT now lays unused in less developed countries due to the adoption of the Western perception of risk, while the World Health Organisation reports that malaria is killing nearly three million people a year world-wide. Importantly, the effects of DDT have been reversible: otter and eagle populations have recovered. And, as we learn more, so products become more benign. The replacements for DDT were far less harmful to the environment.

According to Bruce Ames, Professor of Biochemistry at Berkeley University of California, without pesticides there would be pressure to cultivate more land, potentially reducing biodiversity and, crucially, food would be much more expensive. A poor diet (low in pesticide-protected fruits and vegetables) is a far greater and more immediate health hazard than ingesting pesticide residues ever could be, and the poor would suffer most from dearer fresh produce. Ames concludes that 'the risk from using pesticides is far lower than the risk from not using them'.

This balancing act cuts no ice with the environmentalists. All that appears to matter to them is whether a substance or technology may do harm. The campaign against genetic research for the bio-food markets bears these hallmarks. According to the Green precautionary position, foods such as melons and maize genetically modified to resist pests may have indefinable and irreversible consequences on ecosystems. Most scientists accept that there is a tiny risk of altered genes reaching the environment, and that a tiny percentage of these could be harmful. However, genetically altered crops need less pesticide as they are resistant to many pests. So advances in biotechnology should please environmentalists for reducing the perceived dangers from pesticides. Yet they seem reluctant to acknowledge this silver lining to the biotech cloud.

The public, on the other hand, has proved willing to acknowledge this trade-off. In a recent study, Richard Braun, Professor of Microbiology at the University of Berne, Switzerland, showed that Swiss public acceptance for genetically modified maize was very low until it was pointed out that less pesticides would be used on modified maize than on conventional maize. Once the unknown risks of genetic manipulation were compared with the perceived risks of pesticide use, the public was happier to accept the new technology along with the possibility of irreversible harm, rather than pay a high price for not having that technology.

Of course new technologies bring new hazards, but they also usually replace older technologies which were more harmful. Filter-tip low-tar cigarettes, are far less harmful than their predecessors. Toxic arsenic pesticides were replaced by non-toxic DDT, which was replaced by more environment-friendly pesticides - and now genetic resistant plants are slowly making even these pesticides redundant. A precautionary principle that balances costs and benefits is fine, but one that is a principle of immobility and technological stagnation is a real risk.

Roger Bate is the editor of What Risk?: Science, Politics and Public Health, published by Butterworth Heinemann