Dental Watch Home Page
In 1997, the journal Perspectives in Biology and Medicine published an opinion piece, "Why I changed my mind about water fluoridation," by John Colquhoun . Although the journal's stated purpose is to convey new ideas or stimulate original thought in biological and medical sciences, Colquhoun presented no new data. His paper rehashed earlier criticisms of water fluoridation, using selective and highly biased citations of the scientific and nonscientific literature [2-10].
Colquhoun, who died in March 1999, was a dental officer in New Zealand during the 1970s and early 1980s. His writings display blatant bias in evaluating the published literature on fluoridation, criticizing any study supporting fluoridation for poor design (e.g. non-blind examinations, non-random selection of subjects and communities), while ignoring gross defects in the methods used in studies that found no caries reduction in fluoridated communities. He was the Editor of Fluoride, a magazine published by the International Society for Fluoride Research, for which he was also the Treasurer. Despite its propitious title, Fluoride is primarily a vehicle for printing articles that decry the benefits of communal water fluoridation (13 of Colquhoun's 73 citations are from this anti-fluoride publication). Colquhoun's paper is in the same genre. In it, Colquhoun states that water fluoridation is ineffective in reducing caries and that the decline in caries observed in most Western industrialized countries is not due to fluorides but rather to a vaguely described "improved nutrition." He also contends that water fluoridation is harmful to teeth (causes fluorosis) and to general health (causes bone cancer, weakens bones, affects behavior) and lacks universal endorsement. In general, Colquhoun ignores the overwhelming findings of the efficacy of water fluoridation in reducing caries prevalence, not only by failing to cite recent individual studies from various parts of the world, including Australia, Britain, Canada, Ireland, and New Zealand as well as the United States, that have reached this conclusion, but also by omitting major reviews that have cited these studies [11,12]. Instead, Colquhoun cites his own and Diesendorf's publications, which have been discredited because of significant errors, misquotations, and the use of questionable data . Other reports that Colquhoun cites that supposedly do not show benefits of fluoridation failed to establish residence histories of the sample populations, used crude measurements of caries prevalence, and were based on intake of natural fluoride, the level of which is neither controlled nor monitored [14,15].
Claims that fluoride is harmful have been amply reviewed by international, national, state, and local authorities, including the following:
The groups that wrote these reports included independent eminent experts in a variety of different fields, such as medicine, epidemiology, pathology, pharmacology, and toxicology, as well as scientists and water engineers. The conclusions have been remarkably similar, namely that the benefits of water fluoridation far outweigh any potential health risks. Consumer Reports magazine has aptly summarized the situation:
The simple truth is that there is no 'scientific controversy' over the safety of fluoridation. The practice is safe, economical, and beneficial. The survival of this fake controversy represents, in our opinion, one of the major triumphs of quackery over science in our generation .
Although it is not practical to refute each error of commission or omission in Colquhoun's paper, the following comments are intended to set the record straight.
Colquhoun served as the Principal Dental Officer of the city of Auckland, New Zealand, during which time he accepted the findings that water fluoridation is effective in reducing caries. He writes that he became troubled by the declining caries prevalence in non-fluoridated areas as well as in fluoridated areas. Then in 1984 and 1985 he published his own evaluations of records collected on children in Auckland, which is fluoridated, and neighboring Onehunga, which is not fluoridated, and reported that the dental health in Auckland was only "slightly better" than in Onehunga. All of this led him to question the efficacy of water fluoridation.
The opponents of fluoridation are a heterogeneous group and cannot easily be categorized by any single characteristic. Among their number are right-wing extremists, misguided environmentalists ("Greens"), chiropractors, persons concerned about the costs of fluoridation, food faddists, and antiscience "naturalists." Other opponents have emerged, including the self-proclaimed "neutrals" who try to portray an image of dispassionate open-mindedness but clearly have accepted the opposition's arguments irrespective of whether they have been adequately tested and answered [6,31,32]. Others have been described as the "born-again antifluoridationists" , who previously accepted the mainstream belief in the benefits of fluoridation but have experienced an epiphany so that the scales have fallen from their eyes and they have seen the light [2,34,35]. Clearly Colquhoun falls in this latter category. However, it is important to recognize that simply by claiming to be a former advocate and now clearly being a dedicated opponent of fluoridation in no way validates his judgment nor excuses his distortion of the literature.
Teeth Better Without Fluoridation?
Colquhoun cites several studies, including his own findings [2,3], which purportedly show that water fluoridation has no significant beneficial effect in reducing caries prevalence. When comparisons are made between fluoridated and non-fluoridated communities in a properly designed research study, examiners are standardized and calibrated to agreed-upon criteria for diagnosis. However, Colquhoun's data were not recorded for this purpose, as he simply took from school dental clinics the treatment records, in which caries diagnoses were not standardized. The examined children were treated by different school dental nurses, who were untrained in epidemiological survey methods, not standardized in their interpretation of diagnostic criteria, and not calibrated to an expert examiner in their examining techniques. The examiners were likely to have had their own criteria for what constituted a cavity and what teeth required restorative care. Worse still, Colquhoun made no effort to ensure that the children actually resided in either the fluoridated and non-fluoridated communities, but analyzed the data simply on the basis of the district in which the school was located, not where the child lived. Subsequently, it was shown "that 38 percent of the 12-13 year-old children who attended school in Onehunga, the low social class area described by Colquhoun , actually lived in high social rank areas in Auckland, districts that received fluoridated drinking water. This confounding of residence and fluoride histories produced serious misclassifications and severely compromised Colquhoun's data, and hence his conclusions" .
A similar criticism applies to what Colquhoun describes as a large-scale survey in Tucson, Arizona, conducted by an anthropologist, a retired chemistry professor, and a public school nurse . This survey has never been published other than in abstract form, so details of the methods have to be surmised. However, inasmuch as none of the authors was qualified to conduct dental examinations, they presumably used school records of dental examinations, which were non-standardized. Moreover, there is no assurance that the children resided in the district in which the school was located, nor any indication as to how long they had lived there. Furthermore, this was not a comparison between a controlled fluoridated community and a non-fluoridated one, since parts of Tucson have natural fluoride in the water supply and other parts of the city do not. In naturally fluoridated communities the fluoride concentration may fluctuate with the season, depending on the rainfall. It is surprising that Colquhoun gives such findings credibility considering his stringent criticism of studies that have demonstrated beneficial effects, which he characterized as "flawed studies".
Without a doubt, residential histories have become more important since the 1940s, when the original fluoridation studies were conducted, because the US population has become increasingly transient. The importance of obtaining adequate residential histories of children when studying the effects of water fluoridation is illustrated in the large-scale 1986-87 national survey of US schoolchildren that Colquhoun cites . Of the 39,206 schoolchildren examined in this survey, only 16,398, less than 42 percent, qualified as having either a lifetime history of fluoride exposure or no fluoride exposure (drinking water or supplements). Yiamouyiannis' analyses  of these data, obtained by the Freedom of Information Act before the government's analyses were completed, are crude and ignore residential histories. He uses the index of decayed, missing or filled teeth (DMFT), which is less sensitive than the index of decayed, missing or filled surfaces (DMFS), and the emphasis on the percentage of caries-free children for analysis is questionable for statistical groupings. The fact remains that children who had continuously been exposed to community water fluoridation had an 18 percent lower caries prevalence (based on DMFS scores) than those who had never lived in a fluoridated community. When some of the background effect of topical fluorides was controlled for, this difference increased to 25 percent .
A study in India that Colquhoun claims showed no relation between water fluoridation and tooth decay involved 400,300 children who lived in communities with either less than 1 ppm F or more than 1 ppm F of natural fluoride . This multitude of children was seen by a single examiner, who scored caries as either present or absent. Data were expressed as the percentage of children who were caries free. Undoubtedly the investigator was overwhelmed by the number of subjects that had to be examined; nevertheless, this method is a crude measure of caries prevalence. A more sensitive scoring system involves measurement of decayed, missing and filled teeth or tooth surfaces (DMFT/S). In addition, data were expressed as "decayed teeth per 100 children," and children were separated into those who drank water with an average fluoride concentration of 0.60 ± 0.24 ppm and those who drank water with an average fluoride concentration of 4.18 ± 2.03 ppm. In countries with very hot, dry climates where air conditioning is a rare luxury and the population performs hard physical work, and consequently has increased sweat loss and greater water intake, lower maximal fluoride levels may be required . It is likely, therefore, that in India 0.60 ppm F is closer to an optimal concentration than 4.18 ppm F; certainly this is not a comparison between no fluoride and optimal fluoride levels, as Colquhoun claims.
Causes of Decline in Tooth Decay
Opponents of water fluoridation often argue that caries is caused simply by poor nutrition and that a good diet alone can impede dental decay. Of course, diets that severely restrict sugar intake can limit decay. The relationship between sugar intake and caries in humans has not been ignored by supporters of water fluoridation . However, dietary control, particularly the restriction of sugary foods, is not a practical public health method for caries prevention, because it can only work in populations that are institutionalized. There are no controlled studies that have demonstrated that caries is a result of calcium deficiency, nor is tooth decay a problem of some vague "general nutritional" inadequacy, as Colquhoun asserts. These ideas were proposed by Mellanby  but have not been sustained by rigorous research. The relationship of dietary sugar to caries has been amply documented . A figure compiled by Colquhoun that shows a continuous decline of decayed teeth per child among 5-year olds in New Zealand from 1930 to 1990, based on Health Department records, is highly suspect: the New Zealand school dental nurse program did not reach all children until the early 1960s, so earlier records are based on very limited data. Casting further doubt on Colquhoun's compilation is the fact that such a continuous caries decline, starting as early as the 1930s, runs contrary to the experience of most developed countries.
In comparing caries prevalence among young United States adults over a 40-year period (1930s to 1970s), Burt concluded that caries prevalence had not changed much during that time . In most countries the prevalence of decay in schoolchildren held steady from the 1930s until about 1940, when wartime rationing of sugar caused a decline in Europe and Japan [42-46]. In the 1960s dental decay was ubiquitous among American schoolchildren, causing widespread premature tooth loss and pain. For example, the mean caries score of all children who attended a rural school in Pennsylvania in 1958 was 13.51 DMFS . This high decay rate did not decline until the mid- to late 1970s , by which time topical use of fluoride-containing dentifrices was widespread and other fluoride products had become available. The 1986-87 national survey of US schoolchildren showed that the DMFS had dropped to 3.07 . This caries decline has occurred despite fairly stable national sugar consumption or sugar disappearance data .
Colquhoun tries repeatedly to make the point that dental caries prevalence has dropped considerably in non-fluoridated areas of New Zealand. It has dropped in many other countries of the world as well, during the past 25 years or so . He briefly mentions the dilution effects in measuring the effectiveness of water fluoridation from the widespread use of fluoride-containing products, primarily fluoride dentifrices, in non-fluoridated areas. But he barely acknowledges the well-known diffusion effect of community water fluoridation, wherein persons in non-fluoridated communities benefit from consuming products processed in fluoridated communities. New Zealand, with widespread community water fluoridation benefiting 66% of the population, particularly in larger cities where foods and beverages tend to be processed, undoubtedly experiences a profound diffusion effect from water fluoridation.
Colquhoun concedes that many published studies have shown a benefit of less caries in fluoridated communities than in non-fluoridated ones, since to deny so would be difficult. He then proceeds to impugn one such study  as biased in the selection of study sites, stating that the communities were not picked at random. However, the aim of that study was to obtain follow-up data in certain communities that had been compared previously , one of which had stopped fluoridating its water supply, and to include communities with continuous water fluoridation. Obviously, such communities could not be "randomly" selected, and only those children whose parents consented to the examination were included. Colquhoun states that when he compared all the children there were "virtually no differences in tooth decay rates." He omits explaining that his comparison used school dental service records, which, unlike the planned study that he impugns, made no attempt to standardize the therapist/examiners, nor did he consider issues of lifelong residency. For the record, the study by Treasure and Dever  found a mean prevalence of decayed, missing and filled deciduous teeth (dmft) for 5-year-old children of 2.91 in the non-fluoridated community, compared to 2.00 in the community that had stopped fluoridating, and 1.06 in the fluoridated communities, a pronounced difference of 64 percent less decay.
Colquhoun criticizes studies that have demonstrated reductions in caries prevalence following water fluoridation because examiners were not "blind" to the residence of the children being scored for caries. His prejudice is evident when he terms such studies "pro-fluoridation," whereas in fact the examiners were neither pro- nor anti-fluoridation; they simply were trying to determine if an effect could be found. The issues of blind examinations and examiner bias are not new to the fluoridation debate, having been raised by Groth in his thesis  and subsequently publicized by Jack Anderson in his syndicated columns. Sutton also claims that examiners are biased and that studies need to be conducted with blind examinations . In fact, this issue has been avoided in several studies in which patients or radiographs were examined under blind conditions [54-58]. In all of these studies in which the examiners were not aware of the residence site of the subjects, the findings were consistent: lower caries prevalence in fluoridated communities than in non-fluoridated ones. The whole question of blind studies has been more than amply answered in the literature , which Colquhoun chooses to ignore.
Harm to Teeth: Enamel Fluorosis
Colquhoun's statement that "Fluoridationists would not at first admit that fluoridated water contributed to the unsightly mottling" is untrue. It has been known since Dean's early studies that excessive levels of fluoride in the drinking water are associated with enamel fluorosis, but that optimal levels cause minimal fluorosis while providing significant protection against caries. The prevalence of fluorosis, mostly of the milder forms, in optimally fluoridated communities was reported to be about 14 percent in 1939-40 surveys and has increased to about 22 percent in 1986-87, since fluoride dentifrices came into widespread use and dietary fluoride supplements became readily available . It is proponents of water fluoridation who have been in the forefront of efforts to educate the public and to induce dentifrice manufacturers to use warning labels advising consumers that children under 6 years of age should use only small, pea-sized quantities and avoid unintentional swallowing of fluoride-containing dentifrices, since this is a major factor in the increasing prevalence of enamel fluorosis [60, 61]. Most fluorosis associated with controlled water fluoridation is of the milder forms and therefore a limited cosmetic problem, whereas caries is a pathological process, the repair of which often causes its own cosmetic problems; therefore, the benefits outweigh any harm.
Opponents of water fluoridation, including Colquhoun, have drawn attention to studies that report an association between fluoridated water supplies and hip fractures. Ecological studies (those in which the unit of analysis is the community rather than the individual) present conflicting results, with two showing a protective association [62,63], four showing no significant relationship [64-67], and three reporting an increased relative risk of hip fracture incidence with water fluoridation [68-70]. Some consider that the conclusions reached by each of these studies are limited, because associations found at the ecological level do not guarantee similar associations in studies of individuals ("ecological fallacy") . These conflicting findings may be due to the multifactorial pathogenesis of osteoporotic fractures and may prove impossible to resolve by current epidemiological-ecological methods .
In 1991, the National Institutes of Health (NIH) sponsored a workshop to examine historic and contemporary research on fluoride exposure and bone health in humans. The summary and recommendations of this workshop merit quoting: "Taken together, the results of these six contemporary studies fail to establish an adequate basis for making firm conclusions relating fluoride levels in drinking water to hip fracture and bone health. Most of the studies have important limitations that restrict generalization of their results either to the population as a whole or to determining risks for individuals. In general, the results yielded relatively small clinical impacts and/or weak statistical power. There is no basis for altering current public health policy" . An expert committee of the World Health Organization came to the same conclusion, stating: "With respect to hip fracture and bone health, there is no scientific evidence for altering current public health policy on the use of fluorides for caries prevention" . Recently, a very thorough study of the association between water-borne fluoride and bone mineral density (BMD), using both "ecological" and individual exposure, found no difference in the BMD of populations between optimally and suboptimally fluoridated communities . Lifelong exposure (> 20 years) to higher fluoride levels (2.5 ppm F) tended to increase lumbar spine and proximal femur BMD.
Apart from the issue of fluoride exposure at relatively low levels of 1 ppm F in the drinking water, physicians have used high daily fluoride supplements of 45 to 80 mg to treat osteoporosis. Colquhoun points to studies published in 1989-90 that used such doses as having "actually caused more hip fractures." However, a more recent regimen has involved intermittent use of a slow-release fluoride in a honeycomb wax tablet of 25 mg F taken twice a day, together with continuous ingestion of a calcium citrate supplement to assure adequate mineralization of bone matrix . This protocol administered for about 4 years inhibited new vertebral fractures and was safe to use [76-79]. Of course, other treatments besides fluoride, such as estrogen replacement therapy, biphosphonates such as etidronate and alendronate, and calcitonin, have also been successful in reducing bone fractures in women due to osteoporosis.
Colquhoun refers to a study that he claims reported "a fluoride-related incidence of a rare bone cancer, called osteosarcoma, in young male rats" . The actual conclusions of this study were: "NaF did not alter the incidence of preneoplastic and neoplastic lesions at any site in rats of either sex. Results from this study indicate that NaF is not carcinogenic in Sprague-Dawley rats." The National Toxicology Program (NTP) reported that in male rats, but not female rats or mice of either sex, fed high-dose fluoride, osteosarcomas occurred with a dose-response trend . The NTP study concluded that there was "equivocal" evidence of carcinogenic activity in male rats. Subsequently, an ad hoc subcommittee was appointed by the United States Public Health Service to review not only animal studies but also human studies on the relationship of cancer and fluoride exposure. It concluded that: "Optimal fluoridation of the drinking water does not pose a detectable cancer risk to humans as evidenced by extensive human epidemiological data" . Furthermore, it concluded that "animal studies fail to establish an association between fluoride and cancer."
One case-control study of the relationship, if any, between fluoride and childhood osteosarcoma  analyzed lifetime fluoride exposure, not only from water intake, but also from fluoride drops, tablets, and vitamins, fluoride mouth rinses and toothpastes, and fluoride topical applications in the dental office, in order to establish a fluoride exposure index. One cannot fail to be impressed by the thoroughness of data collection and rigorous inquiry into residential histories in this study, which concluded that there was no evidence that fluoride exposure increases the risk of childhood osteosarcoma. Because there clearly was not a dose-response relationship between fluoride and childhood osteosarcoma, anti-fluoridationists have suggested that perhaps persons with osteosarcoma are "more susceptible to equal fluoride exposure" . In other words, when research findings fail to support the preconceived opinions of anti-fluoridationists, they invoke hypersusceptibility to fluoride, which has never been demonstrated under controlled conditions.
Hoover, Devesa and Fraumeni, expert epidemiologists and scientists at the National Cancer Institute of the NIH, have also responded to the charge that fluoride causes cancer. They write: "Opponents to fluoridation have spent enormous amounts of time attempting to link adverse effects with fluoridation. Given enough attempts, someone should be able to link some condition with fluoridation in some group using some methodology. Responding to each such attempt would seem less useful than periodic reviews of all the evidence by qualified panels of experts that can assess the quality of the work and the resultant weight of the evidence. The last such review dealing with cancer issues reached the same conclusion as those preceding it, that optimal fluoridation of the drinking water does not pose a detectable cancer risk to humans as evidenced by extensive human epidemiological data".
Other Evidence of Harm?
Colquhoun warns of "chilling evidence" that children with dental fluorosis in China have lower than average intelligence. Yet the only statistically significant finding in this report was the rather obvious observation that children of parents with higher education showed higher IQ scores than did children of parents with less education . He cites a study on rats that were fed water containing 75, 100, 125 or 175 ppm fluoride and were reported to have behavioral deficits . At 175 ppm fluoride, 50% of the rats died, and body weights were affected at 125 ppm doses, indicating that these levels induced overt toxicity and considerable physiological perturbations. It is not clear whether the reported behavioral changes at these levels were due to fluoride's action on central nervous system function or to the illness induced by extremely high levels of fluoride. The largest and most consistent behavioral change was a reduction in grooming, which is a generally recognized adaptation to illness in animals . Because these questions were not addressed by the research design, behavioral changes found at these dosage levels cannot be considered fluoride-induced. This rat study reported some highly questionable differences in fluoride plasma levels between males and females fed fluoride at identical levels, as well as highly variable and improbably high brain-to-plasma fluoride ratios. Expert review of this study (G. Whitford, personal communication) concluded that the fluoride concentration data are inaccurate and not representative of values that exist in vivo. Thus, the claimed relationship between these fluoride concentrations and the behavioral effects are uncertain. Surprisingly, Colquhoun, who lives in New Zealand, omitted citing a New Zealand study of the relationship between duration of exposure to fluoridated water supplies and measures of child health and behavior . In this seven-year study, no association was found between duration of exposure to fluoridated water supplies and a large range of child health and behavioral measures.
Endorsements of Fluoridation
Colquhoun and other anti-fluoridationists assert that certain countries have "banned" water fluoridation for health and safety reasons. This is an egregious distortion, far from the truth, as the actual reasons for not adopting water fluoridation have been political, legal, or technological. In 1962 an enabling Water Fluoridation Act was passed by the Swedish parliament which provided that a community could fluoridate its water supply if it met very stringent requirements of the National Board of Health . In 1971, because of vocal opposition by the anti-fluoridation lobby, the Swedish parliament repealed this Water Fluoridation Act on a narrow vote of 137 to 126, in which most Conservative, Liberal, Center, and Communist members voted for repeal while most Social Democrats voted to retain the Act (22% of members had already left for the weekend). This vote by a strange combination of political parties has been interpreted as being directed against the Social Democrat government rather than fluoridation per se, since the Parliament's Social Insurance Committee had given a favorable report on fluoridation. The action was a political decision that was not supported by Sweden's leading health experts . In 1992, water fluoridation was stopped by a majority vote of the City Council in Kuopio, Finland, again on a political basis. An expert committee consisting of professors from the University of Kuopio had concluded that there was not any medical, toxicological, or ethical reason for stopping fluoridation . In the Netherlands, water fluoridation was stopped in Culemborg in 1973 because the High Court ruled that the Water Supply Act was not a sufficient legal basis for fluoridation. Thus, the decision to discontinue fluoridation was on legal grounds, not for safety reasons.
Colquhoun's statement that fluoridation is practiced only in America and in countries under strong American influence is patently absurd. Singapore (100% fluoridated), Hong Kong (100% fluoridated), Malaysia (44% fluoridated), Brazil (25% fluoridated), Canada (50% fluoridated), Chile (40% fluoridated), Colombia (30% fluoridated), Ireland (60% fluoridated), Australia (66% fluoridated), and New Zealand (66% fluoridated) are hardly all under the hegemony of the United States. He also fails to point out that several countries such as Germany, France, Spain, Mexico, Jamaica, and Costa Rica have chosen to use salt fluoridation for practical, technical, or other reasons. In Switzerland, both water and salt fluoridation are in use in different regions. A wide spectrum of health-related organizations, including the World Health Organization (since 1958) , the health advisory committee of the European Union, and numerous national health authorities have supported water fluoridation for caries prevention. Colquhoun obfuscates the position of the Environmental Protection Agency (EPA), which has set a maximum contaminant level of 4 ppm for natural fluoride and historically has supported water fluoridation at optimal levels. A small clique of EPA employees have controlled the policy of their union and have used it as a vehicle for opposing water fluoridation.
Conspiracy Theory and Paranoia of the Anti-fluoridationists
Colquhoun, like many opponents of fluoridation, subscribes to the conspiracy theory, according to which the government, health authorities, and the dental profession are trying to foist water fluoridation on an unsuspecting public. He did not receive a response to his inquiries about the US survey because the data were still being analyzed and pre-publication data were not available. Results were first presented at a meeting of the International Association for Dental Research in 1989  and were fully published in 1990 . Colquhoun implies there was a conspiracy of silence by dental researchers to withhold information from the public and to bias findings .
The assertion by Colquhoun of intimidation and pressure on opponents of fluoridation is without basis. For example, it is paranoiac of Martin  to state that unpublished critiques of the writings by Diesendorf and Groth were "circulated". Murray and Rugg-Gunn prepared a detailed point-by-point response to Diesendorf's article  which Nature refused to accept; subsequently it was made generally available (Fluoridation and Declining Decay: a Reply to Diesendorf, British Fluoridation Society, London, Dec. 1987). Similarly, comments on Groth's argument for the need to perform blind examinations were readily available in the literature .
Debating the Issue
Opponents of fluoridation like nothing more than to have public debates on the radio, television, or in the press because it makes fluoridation seem a "controversial " issue and gives them free publicity. In such debates with an equal number of speakers pro and contra, it appears as if the health science community is evenly divided on this issue. In fact, the overwhelming majority, probably well over 90%, of scientists, physicians, dentists, nurses, veterinarians and public health professionals fully support community water fluoridation. Indeed, according to a recent Gallup poll, 70% of the American public believes community water should be fluoridated, 18% is opposed and 12% has no opinion . The proportion of the public favoring fluoridation in such surveys has changed very little over the years, with 60-75% perceiving it as desirable since 1952 . Colquhoun's plea for examining and discussing new and old evidence in a scholarly debate with proponents of fluoridation sounds good, but is frustratingly difficult in practice. Colquhoun, Diesendorf and their colleagues are unequivocally unwilling to accept the validity of any of the hundreds of published studies on the effectiveness of community water fluoridation because the communities have not been selected randomly. In other words, they will accept only studies in which a randomly selected community fluoridated its water (whether it wants to or not) and be paired with a randomly selected control community. The purity of random selection sounds great, but it is unfeasible when it comes to selecting sites for the implementation and evaluation of water fluoridation. Colquhoun and his colleagues do not engage in scholarly debate, but rather engage in polemical arguments.
Dr. Newbrun is Professor Emeritus of Oral Biology and Periodontology, Department of Stomatology, University of California San Francisco. Dr. Horowitz is a consultant in dental research and public health. This article,which appears here with their kind permission, is slightly modified from its original format in Perspectives in Biology and Medicine 42:526-541, 1999.