So they highlight problems that should be confronted and emphasise the considerable uncertainties in the evidence in the hope of furthering constructive public consultation and debate.
Worldwide, almost 6% of people receive water containing fluoride to around 1 mg/l, write Sir Iain Chalmers and colleagues. In England and Wales, about 9-10% of water supplies contain 0.5-1 mg/l fluoride, either naturally or as an additive.
Plans to add fluoride to water supplies are often contentious. Problems include difficulties in identifying benefits and harms, whether fluoride is a medicine, and the ethics of a mass intervention.
So, what are the known benefits of adding fluoride to water, they ask? In 1999, the Department of Health commissioned the Centre for Reviews and Disseminations at York to conduct a systematic review on the potential benefits and harms of adding fluoride to water. Despite identifying 3,200 research papers, the reviewers were surprised by the poor quality of the evidence and the uncertainty surrounding the benefits and harms of fluoridation.
There is no such thing as absolute certainty on safety, say the authors. But in the case of fluoridation, people should be aware of the limitations of evidence about its potential harms and that most studies are not sufficiently well designed and large enough to detect small but important risks (especially for chronic conditions) after introducing fluoridation.
They point out that the evidence for alternative ways of preventing dental caries (mainly toothpastes containing fluorides) is strong. However, the use of toothpastes depends on individual behaviour, which has important implications for reducing inequality.
Another important issue is whether fluoride added to water supplies is a medicine, they add. Some people regard it as a form of mass medication, while others point out that it occurs naturally at concentrations comparable to those in fluoridation programmes and is therefore not a medicine.
If fluoride is a medicine, evidence on its effects should be subject to the standards of proof expected of drugs, say the authors. If used as a mass preventive measure in well people, the evidence of net benefit should be greater than that needed for drugs to treat illness. Yet, to date, there have been no randomised trials of water fluoridation.
The ethical implications of water fluoridation must also be considered, they say. Under the principle of informed consent, anyone can refuse treatment with a drug or other intervention. Potential benefit must therefore be balanced against uncertainty about harms, the lower overall prevalence of dental caries now than a few decades ago, the availability of other effective methods of prevention, and people's autonomy.
Finally, they believe that trust in the dissemination of evidence is vital. Those opposing fluoridation often claim that it does not reduce caries and they also overstate the evidence on harm. On the other hand, the Department of Health's objectivity is questionable - it funded the British Fluoridation Society and, along with many supporters of fluoridation, it used the review findings selectively to give an overoptimistic assessment of the evidence in favour of fluoridation.
They conclude: "Against this backdrop of one sided handling of the evidence, public distrust in the information it receives is understandable. We hope this article helps provide professionals and the public with a framework for engaging constructively in public consultations."
In a Personal View article, Rod Griffiths, Regional Director of Public Health for the West Midlands believes that, on balance, fluoridation of water supplies is a good thing.
He points out that Birmingham has had fluoride for 40 years and dental health is among the best in the country, even though other indicators, such as obesity, heart disease, and life expectancy are far from best. And asks, why do some people become so passionate about fluoride, when other regulations about harmful chemicals to which we are all exposed through agriculture and industry attract so much less attention?