Where's that page gone? Search us...

Now this may well be the case - but some researchers - and in particular statisticians - are concerned that such scientific reports often end up confusing - rather than informing the public.



Take the alleged links between passive smoking and cancer - or perhaps whether drinking red wine is good for you - even esteemed medical journals like the Lancet and the British Medical Journal have published conflicting reports.



It sent a shiver down parents' spines. The science journal, Nature reported that children who slept with a night light would become short-sighted. The lights went off across the nation. Yet just a few months later two further studies crushed the findings. They found instead that short-sighted parents like night lights so they can see the babies - and myopic parents have myopic children. It was all in the genes.



It's a numbers game - statistics and what they mean to we the public. Should you eat butter? Is jogging bad for you? Does the pill cause cancer? Now there is a move to increase the understanding of this science because of concern that too few people including medical researchers themselves know how to interpret them and this leads to confusion.



As the incoming president of the Royal Society of Statisticians, Andy Grieve believes that now is the time to address the problem.



The interpretation of statistics is at the very heart of the raging passive smoking debate. Last week an article in the British Medical Journal claimed it may not after all be so dangerous - that it is not significantly associated with coronary heart disease or lung cancer. And yet last November we were told that children were at risk from their parents' smoking and that even short-term exposure to second-hand smoke could increase the chances of heart disease and strokes.



Last October women who were moderate drinkers were told their risk of breast cancer was not increased, but if they smoked young there was a risk of breast cancer. By November, just a month later - after a few more glasses of wine had been imbibed another study found alcohol does increase breast cancer but smoking is innocent.



Part of the controversy is over how big does a trial need to be to show whether a treatment does or does not work.



Currently raging in the pages of one medical journal is a debate over a study of arnica a popular homeopathic remedy for bruising. This study found that when taken orally following surgery it had no effect. But only 64 people took part in the study.



There has been an explosion in the numbers of medical and science journals and according to Dr Grieve the editors have a responsibility to ensure that they are publishing research that is statistically robust and not just headline grabbing.



The Lancet sends all research to be vetted by a statistician. But even then, the editor says, mistakes do happen.



Only today a new report claims that chlorine in swimming pools could be causing the rise in childhood asthma. It may well be true, but statistically it's based on a small sample and has not taken into account other environmental hazards like passive smoking for instance.



So maybe it is not true. But what is certain is there will now be a lot of worried parents and probably some kids kept on dry land.