Intended for healthcare professionals


Fluoridation of water supplies

BMJ 1999; 319 doi: (Published 31 July 1999) Cite this as: BMJ 1999;319:269

Debate on the ethics must be informed by sound science

  1. David Coggon, professor of occupational and environmental medicine (dnc{at},
  2. Cyrus Cooper, professor of rheumatology
  1. MRC Environmental Epidemiology Unit, Southampton General Hospital, Southampton SO16 6YD

    The British government has recently announced a new expert review of fluoride and health with a view to strengthening the legislation that underpins fluoridation of water supplies if the findings are supportive.1 The epidemiological evidence that fluoride protects against dental caries is overwhelming, but in England and Wales less than 10% of the population receives fluoridated water,2 and proposals to extend the coverage provoke strong opposition from a vociferous minority who view the intervention as hazardous and unjustified. The debate raises both scientific and ethical questions.

    The scientific challenge is to quantify reliably the benefits and risks from fluoridation as compared with alternative strategies for preventing caries. Early surveys found that among children rates of decayed, missing, or filled teeth were some 50% lower in places with water fluoride concentrations of about 1 ppm than in those where they were 0.1 ppm or less.3 These differences were borne out by controlled comparisons before and after fluoridation was introduced for selected communities.4 Subsequently, improvements in diet and the incorporation of fluoride in toothpaste may have reduced the scope for gain from fluoridation of water. Nevertheless, recent research suggests that the benefits are still important, particularly among the poorer sections of society.5 In comparison, other policies such as the promotion of fluoride tablets and better oral hygiene are unlikely to be as effective.

    Among the potential hazards of fluoridation, the best established is dental fluorosis, a mottling of the teeth caused by excessive ingestion of fluoride when they are forming in the jaw. The effects are cosmetic rather than functional and cause problems only when the disorder is severe. Such cases would not normally be expected from fluoridation of water alone, but fluoridation might make a contribution where children were also heavily exposed to other sources of fluoride.

    Other possible adverse effects that have been proposed include cancer, arthritis, and fractures. Careful epidemiological studies have failed to support associations with either cancer6 or osteoarthritis,7 but concerns about fluoride and fracture have a stronger basis. About half of ingested fluoride is taken up by bone, where it replaces hydroxyl ions in the hydroxyapatite lattice, with possible implications for its mechanical properties. In addition, raised plasma fluoride concentrations increase osteoblast differentiation and activity. Most of the direct evidence on fluoride and fracture comes from ecological studies of hip fracture, and the results have been inconsistent.8 In part this may reflect limitations in design, such as an inability to take account of potential confounding factors or to assess fluoride exposures earlier in life as well as at the time of injury. An interim analysis of a continuing case-control study that addresses these deficiencies found no evidence of increased risk with higher exposures to fluoride, but the final results have yet to be published.9 Meanwhile, the balance of evidence suggests that any risks of fracture from fluoridation are small.

    Fluoridation raises ethical issues because it can be viewed as an infringement of personal liberty: individuals have no choice in the water that they use—unless they buy expensive bottled water. There are circumstances in which it is considered acceptable to impose measures on a population in the interests of public health (as with the use of seat belts), but to justify such action we must be more confident that benefits outweigh risks than when the individual is free to opt out. A further complication arises because the people who would benefit most from fluoridation (deprived children) are not necessarily those who would be put at most risk by the intervention (people with high intakes of fluoride from other sources). Again, it is accepted that sometimes one section of society may reasonably be disadvantaged for the benefit of another (as with restrictions on driving in people with epilepsy), but where the balance is drawn is arguable.

    Because of these ethical issues decisions about fluoridation must take account of public opinion, but before lay people can make a meaningful contribution to the debate they need an understanding of the science. This is difficult because inevitably there are uncertainties, and the evaluation of the scientific evidence is to some extent subjective. Often the media are more interested in highlighting disagreement and controversy than in establishing the extent of consensus, and this leads to confusion and distrust. The government's new review will again bring the arguments for and against fluoridation to the fore. The best hopes for constructive discussion will lie in clear communication from scientists and responsible journalism.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    View Abstract

    Log in

    Log in through your institution


    * For online subscription