Google Fluoride Alerts
Report of the Expert Panel for Water Fluoridation Review - City of Calgary - 1998
A review of the risks-benefits of fluoridation by University of Calgary Professors who were mostly Pro-FluorideEditorial Comment by Elke Babiuk: The one redeeming feature in this report below is the dissenting opinion where the author makes an honest attempt to fairly analyze some of the voluminous literature. The scientists and other professionals listed directly below, agree that there are serious problems in the Calgary report. It is also noteworthy to mention that unlike what happened here, in the town of Natick Massachusets, fluoridation review panelists were chosen for their neutrality and objectivity. The scientists there came to an entirely different conclusion as the City of Calgary's panel of Professors. See naticks.htm for the Executive Summary of their report.
Click on the individual names directly below for a review of the Majority members' report. The qualifications of these professionals to speak to this issue, or their Curriculum Vitaes, are also available on this site.
Note by Elke: It is important to note that our Calgary Panel of Professors was NOT (as claimed) appointed by the Standing Committee on Operations and Environment which consists of Council members. Council had NO input. It was appointed by Calgary Regional Health Authority and the City of Calgary Waterworks' department officials, despite our objections that several panel members were known to be staunch fluoridation proponents. Information as to the pre-existing biases of some of the City of Calgary's panelists can be found at calgbias.htm
EXTENDED HYPERLINKS within the document (not yet completed) indicate questionable statements. The links provide proper documentation, corrections, and/or relevant criticisms. See also comments by Elke M. Babiuk, author of calgaryb.htm and Editor of this Web Site
Parts of the Calgary report below are scanned and not yet proof-read so any typos may be mine.
REPORT OF THE EXPERT PANEL FOR WATER FLUORIDATION REVIEW
Resources: Dr. Read Seidner, Waterworks Division 35, Glenmore Water Treatment Plant, P.O. Box 2100, Calgary, Alberta T2P 2M5 Dr. Richard Musto, Calgary Regional Health Authority
TABLE OF CONTENTS
City of Calgary, Expert Panel on Water Fluoridation Review, March 1998
The City of Calgary Standing Policy Committee on Operations and Environment and the Calgary Regional Health Authority decided that a review of water fluoridation as a public health policy was needed. A Panel of five experts was appointed. Support was provided by the Waterworks Division and the Calgary Regional Health Authority.
Scientific information was sought through advertisements and was obtained through the efforts of the Panel members. The Terms of Reference for the Panel focussed the work on scientific literature produced since 1989. The Terms of Reference stated that all studies must undergo critical appraisal for design strengths and weaknesses. Results were to be synthesized to form a comprehensive but useable body of evidence.
The Panel held ten meetings between December 3, 1997 and March 18, 1998. Four members reached mutual agreement on the issues, but there were differences with the fifth member which could not be resolved by the deadline for the submission of the Panel's report. Therefore, the report contains a dissenting section. However, the Panel members were very close and the disagreements are largely form, emphasis and level of detail.
Four Panel members agreed that there was not sufficient new scientific evidence upon which to base a recommendation for substantial changes to water fluoridation policy in Calgary. The fifth member concluded that there was evidence supporting health concerns, but that the evidence was weak.
There was a need for further carefully designed studies of total intake of fluoride and health risks of bone fractures and dental fluorosis.
The Panel was unanimous in their support of the reduction of the recommended optimum level of water fluoridation. The Panel majority recommended that the City change from the old level of 1.0 ppm fluoride to 0.7 ppm, which is basically compatible with the recommendation of the other Panel member. That is slightly lower than the new standard recently set by Standards and Guidelines Branch of Alberta Environmental Protection.
BACKGROUND AND TERMS OF REFERENCE
Waterworks Division was directed by the Operations and Environment Committee to assume the role of coordinating and organizing an independent and unbiased panel of experts. Waterworks approached various Deans and Department Heads at the University of Calgary for recommendations. The expertise, maturity and independence of university researchers were considered important assets for prospective panel members. An adequate expertise profile to address technical issues of health and safety was sought. An effort was made to exclude mission oriented candidates (either strongly pro-fluoridation or strongly anti-fluoridation).
A copy of the letter of appointment is in the Appendix. List of panel members from the University of Calgary:
Professor Hossein Joe Moghadam, Professor Emeritus, Departments of Paediatrics and Community Health Sciences, Faculty of Medicine
Professor Sheldon Roth, Departments of Pharmacology & Therapeutics and Anaesthesia, Faculty of Medicine, Director Environmental Research Centrer, Pharmacology, Therapeutics, Toxicology
Professor Miloslav Nosal, Department of Mathematics & Statistics, Faculty of Science
Professor David Hanley, Head, Division of Endocrinology & Medicine, Department of Medicine, Faculty of Medicine Endocrinology, Bone disorders, Calcium metabolism
Panel resource contacts:
Work was to commence as soon as possible after the panel was "virtually ratified " on November 20, 1997. The Panel's work was to be completed by a presentation to the Operations and Environment Committee on April 8, 1998. Prior to that a written brief was to be submitted to Waterworks Division and Calgary Regional Health Authority by March 23, 1998.
The focus of effort was to be on literature produced since 1989. In addition to literature which had already been received, submissions were sought through newspaper ads or provided by expert panel members. Verbal presentations would be at the panel's invitation only. If such presentations were required, then an opportunity was provided to parties on both sides of the debate effort to recommend suitable experts. A list of the material received by the Panel is in the Appendix.
The scientific publications and other material reviewed by the Panel appears in the Bibliography. Where appropriate, references to the original source of information, which can be found in the bibliography, appear in the text.
The Panel held ten meetings: Dec 3, 1997, Jan 12 and 28, Feb 11, two meetings on Feb 25, March 9, 11, 16, and 18, 1998. Based on the tasks outlined in the Panel's Terms of Reference, the following ground rules were set, as reflected in the minutes of the meetings:
Given the very limited time available to the Panel, the Terms of Reference excluded a detailed examination of pre-1989 scientific articles, although some familiarity with that information was important to set the Panel's work in context and to understand the issues.
The Panel's task was not to search the available literature for information to support one side or the other in the debate. The Panel was not to use an adversarial approach and attempt to weigh the relative merits of arguments put forth by opposing sides. The Panel's task was to examine the scientific literature to see if there was significant credible evidence of increased health risk or environmental risk resulting from fluoridation of Calgary's water supply.
Following the Terms of Reference, in the opinion of the Panel, the fact that questions had been raised about the risks of fluoride was not sufficient to warrant a change in policy. Such questions have been around for decades (Hileman, 1998). Studies using extreme conditions (high concentrations or other unusual circumstances), which showed some adverse health effect, were not relevant to our task. Studies with obvious methodological flaws had to be rejected because they could not provide reliable evidence of risk. The Panel did not place much weight on results which raised questions by showing weak associations but lacked evidence of cause and effect.
Types of Studies and Submissions:
Membership on the Panel represented five different areas of expertise: Professor David Hanley, a bone specialist; Professor Joe Moghadam, an epidemiologist; Professor Miloslav Nosal, a biostatistician; Professor Sheldon Roth, a toxicologist; and Professor Dixon Thompson, a chemist and environmental scientist. Each Panel member reviewed most, if not all of the scientific literature but focused on the material related to their area of expertise. Each Panel member wrote a short summary of their findings. There will necessarily be some overlap and duplication.
Four of the authors reached mutual agreement on form, emphasis, content and conclusions. However, there were disagreements between those four and the fifth Panel member, which could not be resolved, in spite of more than a week of extra meetings and revisions, which were held in an effort to resolve the differences. Because of the unresolved points of contention between four Panel members and the fifth, his section is included as a dissenting section. However, The Panel is close to agreement.
(a) Biological Effect of Fluoride on Bone
The second action of fluoride on bone, and the one which has caused interest in its use as a treatment of osteoporosis, is its direct stimulation of the cells which synthesize new bone matrix (osteoblasts). This stimulation requires higher extracelluar fluid concentrations than those which would be expected from artificial water fluoridation at 1 mg/L or lower. At these higher doses, fluoride stimulation of the osteoblast and new bone formation appears to bypass the normal bone remodelling cycle. Normally, new bone formation by osteoblasts only occurs in an area of the skeleton after other bone absorbing cells, called osteoclasts, remove a small amount of older bone (resorption). By causing new bone formation without prior activation of bone resorption, high dose fluoride therapy tie. doses of 40-75 mg of sodium fluoride per day, which would be the equivalent of drinking 18-34 litres of Calgary's water per day) causes a dramatic increase in the production of new bone. The problem has been whether this new bone is qualitatively normal.
(b) Studies of Fluoride Therapy of Osteoporosis
More recently, clinical trials of a lower dose, slow-release formulation of sodium fluoride taken intermittently (12 months on 2 months off) and in combination with calcium citrate supplementation, have shown a significant improvement in bone mass, and a reduction in vertebral fractures (Pak et al, 1994, 1995, and 1996). These studies have been criticized because of the small number of subjects and (in particular) the small number of subjects in the control group. The significance of the fracture prevention has been questioned, but it is clear that the patients had an increase in bone mass with none of the problems which occurred in the Riggs and Kleerekoper studies.
Is the increased bone mass seen in high dose fluoride treatment of osteoporosis normal? Bone biopsy studies accompanying clinical trials of higher dose fluoride therapy of osteoporosis have given varying results. Studies from Europe (Aaron et al, 1991; Boivin et al, 1993; Sogaard et al, .1994) have suggested that the bone obtained from patients receiving long term high dose fluoride therapy is of inferior strength and mineral quality. However, bone biopsy analyses from a study of high dose slow-release fluoride therapy (Antich et al, 1993; Zerkeweh et al, 1996) have suggested that this method of fluoride treatment causes increased amounts of bone that appears to be of normal quality. One explanation for the problems seen in the European bone studies is the wide swings in serum fluoride concentrations which occur when high dose "raw" sodium fluoride or sodium monofluorophosphate is ingested. The rapid absorption of fluoride in these forms gives temporary levels of serum fluoride which are well into a range associated with toxicity. This problem is not seen with the slow-release compound.
(c) Skeletal Fluorosis
A study is currently in progress in Canada, comparing fluoride content of bone from older individuals in communities with (Toronto - 30 years) and without water fluoridation (Montreal). This study has just begun to report its findings from the Toronto area, but has not yet obtained the bone samples from Montreal. We await the results of this project with interest.
(d) Epidemiologic Studies of Fluoridation and Hip Fracture
Since 1989 a number of papers have examined the relationship between the exposure to fluoride in the drinking water in communities and attempted to relate this to the risk of hip fracture. Some have found a weakly positive association. Most have simply examined community health records in areas with higher fluoride content in the water and compared these statistics with communities with low levels of fluoride, with no attempt to look at individual subjects' exposure to fluoride. Virtually none of these studies have assessed the individuals who suffered hip fractures to determine the relationship of the fracture to other known risk factors. For these studies to raise a "red flag" of concern, the studies would have to demonstrate that the relationship between hip fractures and fluoride was still significant after the analysis has controlled for other known risk factors for fracture. Further, they would have to demonstrate that the fractures can be related to a quantity of water ingestion and not to other environmental sources of fluoride, or at least show that fracture rate increases with time of exposure tie. the longer people live in the area, the greater their fluoride intake and the more risk of fracture). None of the studies showing a potential fracture risk with fluoridated water have done all of this. My assessment of several of these studies follows.
In Alberta, a study of individuals aged 45 years or older, who were admitted to hospitals in Calgary and Edmonton with the diagnosis of hip fracture between January 1, 1981 and December 31, 1987 showed no overall statistically significant difference in the age-sex-standardized rates for the two cities (Suarez-Almazor et al, 1993). The hip fracture rate for males over age 65 in Edmonton was only slightly, but significantly, higher than those for Calgary: 4.52 fractures/1000 patient years vs 5.09 fractures/1000 patient years in Edmonton. However, no attempt was made to assess other risk factors for fracture, or differences in osteoporosis prevention and therapy practice in the two cities.
In a study of one fluoridated and two non-fluoridated communities in Utah (Danielson et al, 1992), the investigators examined hip fracture rates during a 7-year period. They found a small but statistically significant increase in the risk of hip fracture in 246 men and women, 65 years of age and older, exposed to artificial fluoridation at 1 PPM (20). Again, the authors did not control for osteoporosis therapy practice in the community, and the only risk factor which was assessed was smoking. In addition, the authors excluded patients who suffered a second hip fracture from the analysis. This excludes a group of patients who are at very high risk for hip fracture, and may have biased the study.
In 1990, Jacobsen and his colleagues reported the geographic distribution of primary hip fracture for women aged 65 years and older at the county level for the entire United States between 1984 and 1987. The age-adjusted rates of hip fractures showed a distinct north to south pattern with higher rates in the south and lower rates in the north. There was a positive association between hip fracture incidence and the percent of 65-year old and older below poverty level and the percent of land in farms. There was also a weak positive association between hip fracture rates and the percent of county residents who received fluoridated water. Again, no assessment of actual fluoride intake or county differences in other well-known risk factors for fracture was possible.
In another study reported by Jacobsen and his colleagues in 1992, hip fracture rates of person 65 years of age and older residing in 129 counties across the United States, considered to be exposed to fluoridated public water supplies were compared with those of the same age group residing in 194 non-fluoridated counties. There was a small positive association between fracture rates and fluoridation. The authors, however, state that confounding factors may be responsible for this observation. For example, boundaries of public water systems rarely coincide with those of counties. Within counties some people use private well water with markedly different fluoride level. One striking finding in this paper was that the incidence of hip fracture was highest in the communities which had been fluoridated for 5 years or less, with no increase in fracture incidence (actually a slight decline) as the duration of fluoridation increased up to 26 years or more. If fluoride has a cumulative deleterious effect on bone, it would be reasonable to expect an increasing fracture rate with increasing duration of water fluoridation.
Jacobsen and his colleagues (1993) examined hip fracture rates in Rochester, Minnesota for the 10-year period before and the 10-year period after the fluoridation of drinking water in 1960. In this time-trend study the authors found no difference in the rate of hip fracture for the two periods.
In 1991, a prospective study of 5 years duration in three demographically similar communities in Iowa was reported by Sowers and her colleagues. Two communities had fluoridated water at 1 mg/liter and one with a fluoride content of 4 mg/liter. Within these communities 827 women, aged from 20 to 80 years, were recruited to participate in the study. Bone mass examinations were conducted at the beginning of the study. Five years later, out of the original sample, 684 women were available for follow-up. Residence in the higher fluoride community was associated with a significantly lower bone mass in all women, an increased rate of radial bone mass loss in pre-menopausal women and significantly more fractures in post-menopausal women. The relative risk of fractures was 2.2 times greater in the high fluoride community compared with low fluoride communities.
From the Pittsburgh centre for the US Study of Osteoporotic Fractures, Cauley et al (1995), reported on an attempt to relate water fluoridation exposure to fracture incidence, in a prospective study of 2076 non-black women of 65 years of age and older who had been recruited to assess predictors of risk of fractures. Bone density was assessed at the spine and hip, and spinal X-rays were taken during the baseline determinations. Vertebral and other fractures were assessed every 4 months. Information on fluoride exposure was limited to community water supplies and was assessed using a residential history questionnaire covering the period from 1950 to 1990. Women exposed to residential fluoridated water for over 20 years had similar spine and hip bone density to women not exposed or women exposed for 20 or fewer years. There was no statistically significant difference of the risk for all fractures in women exposed to fluoridated water for over 20 years and those not exposed. The authors concluded that their results do not support the findings from recent ecological studies which showed an increased risk of hip fracture among individuals exposed to fluoridated public water supplies.
The committee reviewed a number of reports on fluoride effects as well as number of reviews which have analyzed the epidemiologic evidence for a ink between fluoridation and fractures. The vast majority did not support the position that the ecological studies have convincingly demonstrated a causal link between water fluoridation and fractures.
One study which has found an increased incidence of fractures in a high dose (4 mg/L) water fluoridation area (Sowers et al, 1991) has attempted to address some of the above-outlined weaknesses of these studies by recruiting a cohort of subjects from each of the different fluoridation communities and has continued to follow these subjects for further study. The project has recently been extended to allow a much more detailed analysis, and will attempt to assess actual fluoride intake in these communities. Dr. Sowers informs me that she has just completed the data collection and will begin the analysis later this spring. However, it should be noted that this study compares a centre with fluoridation at 4 mg/L to low fluoride communities which use fluoridation at 1 mg/L. The study may not have relevance to the Calgary situation, where the fluoridation level has never been higher than 1 mg/L.
Because of the potential for individuals to have an excessive intake of fluoride from other less readily controlled sources, and the nature of the relationship between fluoride ingestion and dental fluorosis, I would support a recommendation for an overall reduction in the amount of fluoride added to the water to the level of 0.7 mg/L. This level would appear to be adequate to provide dental protection, and would not be likely to be associated with any putative risk to bone health.
The available epidemiologic data would suggest further study is needed with respect to effects of water fluoridation on bone, and the city's public health agencies should be aware of studies that are currently being pursued in this area.
Definition of Epidemiology
Ecological (geographic) studies examine the entire population of fluoridated versus non-fluoridated communities to see if a statistically significant relationship can be found between fluoridation of water and possible adverse health effects. They do not question the afflicted individuals with respect to the source of their drinking water (communal water, well water, bottled water), the amount of daily water consumption, other sources of fluoride exposure such as dental products, occupational exposure to fluoride, significant exposure to radiation, familial predisposition to osteoporosis or cancers, migration into and out of fluoridated communities, socioeconomic status, etc. For this reason they do not have a great statistical power. They can provide information about where further studies with greater statistical power are needed.
Another form of reported epidemiological studies is the temporal or time-trend studies which analyses the trends in the occurrence of an adverse health effect over a period of time in the same communities. These studies have a statisticaI power somewhat greater than the ecological studies.
A still greater statistical power is obtained in case-control studies which attempt to assess exposure to the causative agent in both individuals with the disease and those free of it, or follow both the exposed and unexposed individuals to assess the relative occurrence of disease in both groups.
This report reviews briefly the epidemiological studies published since 1989 on possible adverse health effects of the fluoridation of communal water supplies.
Fluoridation of public water supplies has been controversial since its inception in the United States in 1940s. It has been the subject of numerous epidem iological studies during the past several decades. More recently, various health aspects of ingested fluoride have been reviewed by several panels of scientists in the United States, Great Britain, Australia and Canada.
Water fluoridation and cancers
The NTP studies were reviewed by a peer panel of experts who agreed with the conclusions of the majority of the original investigators. The dissent of one of the original investigators, however, prompted the agency to seek additionally an independent review of the NTP study. This review was conducted by other scientists of the United States' Public Health Service and, together with a review of the literature, was published in 1991 (Review of Fluoride, Benefits and Risks). Briefly stated, their conclusion was that water fluoridation was both effective and safe.
Another independent review of the NTP study and the relevant literature was conducted by the National Research Council in the United States in 1993 (Health Effects of Ingested Fluoride). These scientists also concluded that "the weight of evidence from more than 50 epidemiological studies does not support the hypothesis of an association between fluoride exposure and increased risk of cancer in humans".
Still another independent review of the NTP study was conducted in 1991 by Dr. Edward Calabrese, professor of toxicology, School of Public Health, University of Massachusetts at the request of the East Bay Municipal Utility District (water provider for Oakland and Environs, California). Dr. Calabrese, took issue with the word "equivocal" when a statistically significant dose-response was evident in male rats. He went on, however, to state that the toxicological and epidemiological evidence on human health effects of fluoride do not support changes in community fluoridation programs.
In 1990, the scientists at the National Cancer Institute in the United States reported the analysis of 36 years of cancer mortality data. This was an expansion of their earlier analysis reported in 1977. In addition, they analysed 15 years of cancer incidence data from two populatiori-based cancer registries in relation to the fluoridation status of the drinking water supplies in populations. Due to the controversy stemming from the NTP animal studies, osteosarcoma was singled out for detailed analysis. This survey covered 2,300,000 cancer deaths in fluoridated counties across the United States and over 125,000 incidences of cancers in fluoridated and non-fluoridated counties covered by the two population-based cancer registries. It identified no trend in cancer risk which could be ascribed to the consumption of fluoridated drinking water (Hoover, 1990).
In 1992, an international time-trend analysis of bone cancer covering the period 1958 to 1987 was reported by Freni et al. It showed increases in incidence among young men in Canada, the United States and Europe which were not related to fluoridation.
A few smaller epidemiological studies have also been reported during the past several years on the relationship between osteosarcoma and water fluoridation:
In Alberta, rates of osteosarcoma in Edmonton (fluoridated since 1967) were compared with those of Calgary (not fluoridated until 1991) for the period 1970 to 1988 and reported in 1990. No differences were observed (Hurdy et al., 1990).
In New Jersey, in a study of seven counties, reported by Cohn in 1992, a geographic association between osteosarcoma and water fluoridation was found. The author of this study, however, stated that the residence of patients with osteosarcoma was determined at the time of diagnosis. No interviews were conducted and data on residential history, individual's exposure to fluoride or radiation etc. were not available. The author stated that the findings, even when taken with the overall findings currently in the scientific literature, were not sufficient to recommend that fluoridation of communal water supplies be halted.
In a time-trend study, published in 1991, Mahoney et al. analysed the bone cancer rates in New York State. The average age-adjusted incidence rates of osteosarcoma from 1 976 to 1987 in areas supplied with fluoridated water was no different from the rates in the non-fluoridated areas. The data did not support an association between fluoridated drinking water and the occurrence of osteosarcoma.
In another case-control study of childhood osteosarcoma in New York State, published in 1995, the average annual rate of osteosarcoma from 1976 to 1987 was the same for fluoridated and non-fluoridated areas. In this study, information about exposure to fluoride from all sources was obtained through a telephone interview of subjects, their parents or both. The total life-time exposure to fluoride was not significantly associated with the incidence of osteosarcoma (Gelberg et al., 1995).
Another case-control study of osteosarcoma was reported from Wisconsin in 1995. That study did not find a statistically significant association between exposure to fluoride in the drinking water and osteosarcoma (Moss et al., 1995).
Although the majority of these studies suggest no association between fluoride exposure and osteosarcoma, virtually all of the investigators emphasize that with a rare tumor such as osteosarcoma, no epidemiological study can establish a statistically meaningful conclusion.
An ecological study (Tohyama, 1996) was conducted in communities in Okinawa, Japan, some of which had fluoridated water during the American Administration from 1945 to 1 972. The purpose of the study was to investigate the relationship of uterine cancer mortality rates to fluoridation of water in the communities. Multiple correlation analysis of the data indicated an association between fluoride content of the drinking water and the uterine cancer rates. However, like all other ecological studies, this investigation did not take into consideration other confounding factors at the individual level. These include sexual activity, smoking, and the presence of human papilloma virus. Therefore, it cannot be concluded that the association is one of cause and effect.
Water Fluoridation and Hip Fracture
In Alberta, a study of individuals aged 45 years or older, who were admitted to hospitals in Calgary and Edmonton with the diagnosis of hip fracture between January 1, 1981 and December 31, 1987 showed no overall statistically significant difference in the age-sex-standardized rates for the two cities. The rates for males in Calgary were slightly higher than those for Edmonton males (Suarez-Almazor et al., 1993).
In a study reported from Utah in 1992, the investigators examined hip fracture rates in one fluoridated and two non-fluoridated communities during a 7-year period. They found a small but statistically significant increase in the risk of hip fracture in 246 men and women, 65 years of age and older exposed to artificial fluoridation at 1 ppm (Danielson et al, 1993).
In 1990, Jacobsen and his colleagues reported the geographic distribution of primary hip fractures for women aged 85 years and older at the county level for the entire United States between 1984 and 1987. The age-adjusted rates of hip fractures showed a distinct north to south pattern with higher rates in the south and lower rates in the north. There was a positive association between hip fracture incidence and the percent of 65-year olds and older, below poverty level and the percent of land in farms. There was also a weak positive association between hip fracture rates and the percent of county residents who received fluoridated water.
In another study reported by Jacobsen and his colleagues in 1992, hip fracture rates of persons 65 years of age and older residing in 129 counties across the United States, considered to be exposed to fluoridated public water supplies were compared with those of the same age group residing in 194 non-fluoridated counties. There was a small positive association between fracture rates and fluoridation. The authors, however, state that confounding factors may be responsible for this observation. For example, boundaries of public water systems rarely coincide with those of counties. Within counties some people use private well water with markedly different fluoride levels. Finally, there is probably no true "non-exposed" population due to exposure to fluoride in foods and beverages.
In another study, Jacobsen and his colleagues in 1993 examined hip fracture rates in Rochester, Minnesota for the 10-year period before and the 19-year period after the fluoridation of drinking water in 1960. In this time-trend study the authors found no difference in the rate of hip fracture for the two periods.
In a geographic study of a 5% sample of the white US Medicare population, 65 to 89 year olds between 1986 and 1990, fractures of hip, distal forearm, proximal upper arm and ankles were analysed. There were variations of geographic distribution of these fractures which could not be explained by the exposure to fluoride at the group level. Individual exposure to fluoride was not available (Karagas et al, 1996).
In 1991, a prospective study of 5 years duration in three demographically similar communities in Iowa was reported by Sowers and her colleagues. Two communities had fluoridated water at 1 mg/L and one with a fluoride content of 4 mg/L. Within these communities 827 women, aged from 20 to 80 years, were recruited to participate in the study. Bone mass examinations were conducted at the beginning of the study. Five years later, out of the original sample, 684 women were available for follow-up. Residence in the higher fluoride community (4 mg/L) was associated with a significantly lower bone mass in all women, an increased rate of radial bone mass loss in pre-menopausal women and significantly more fractures in post-menopausal women. The relative risk of fractures was 2.2 times greater in the high fluoride community as compared with the low fluoride communities. In this study the effect of calcium content of water was also studied. It had no significant effect on fracture rates.
In a study of osteoporotic fractures at the Pittsburgh Clinic, reported by Cauley and her colleagues in 1995, 2,076 non-black women of 65 years of age and older were recruited. Bone mineral density was assessed at the spinal and hip levels and spinal radiographs were taken during the baseline determinations. Vertebral and other fractures were assessed every 4 months. Information on fluoride exposure was limited to community water supplies and was assessed using a residential history questionnaire covering the period from 1950 to 1990. Women exposed to residential fluoridated water for over 20 years had similar axial and appendicular bone mass to women not exposed or women exposed for 20 or fewer years. There was no statistically significant difference of the risk far all fractures in women exposed to fluoridated water for over 20 years and those not exposed. The authors concluded that their results do not support the findings from recent ecological studies which showed an increased risk of hip fracture among individuals exposed to fluoridated public water supplies.
In conclusion, the above summaries demonstrate that while ecological studies generally find a slightly higher risk of hip fracture among the elderly living in fluoridated communities, this finding is not supported by the case control studies. The latter studies have a stronger statistical power since they examine individuals rather than comparing communities. Epidemiological studies do not provide us with credible scientific evidence to change public policy on water fluoridation.
In North America the vast majority of dental fluorosis is of very mild to mild types. The moderate and severe types which over the years become discolored as a result of taking up stain from foods, beverages and tobacco smoke are not common.
The prevalence of dental fluorosis has increased during the past several decades in both fluoridated and non-fluoridated communities. In fact this increase has been more marked in some non-fluoridated areas and has been attributed to the inappropriate use of fluoride supplements (drops, tablets, dental products) as well as the consumption of beverages and foods prepared with fluoridated waters (Pendrys et al, 1990).
Pendrys and Stamm (1990) and Lewis and Banting (1994) have reviewed the literature on enamel fluorosis. Both studies concluded that there is a strong association between mild to moderate fluorosis and the use of fluoride supplements in early childhood. Whereas during the past several decades, there has been a 33% increase in the prevalence of enamel fluorosis in the fluoridated communities, the non-fluoridated communities have experienced a 1,000% increase during the same period. Obviously the fluorosis prevalence in the fluoridated communities has also been affected by the inappropriate use of fluoridated dentifrices.
In a more recent publication Pendrys (1995) reported the results of his well-designed retrospective case-control study of middle-school-aged children who grew up in "optimally" fluoridated communities. He calculated that about 25 percent of fluorosis in these communities can be attributed to the inappropriate use of supplements (drops and tablets) and about 71 percent to the inappropriate use of fluoridated dentifrices during the children's first 8 years of life.
It should not come as a surprise that there is a greater prevalence of dental fluorosis in some non-fluoridated communities since fluoride drops and tablets are more frequently used in those communities. When these children use (and swallow) excessive amounts of fluoridated tooth paste and consume beverages and food prepared with fluoridated water, their total fluoride ingestion becomes greater than that of children in the fluoridated communities (Pendrys et al, 1990).
Given the availability and indiscriminate use of fluoridated dental products, it is clear that, at the present time young children can be exposed to excessive amounts of fluoride, which is unnecessary for maintaining their dental health.
This report addresses the toxic or adverse effects of fluoride.
Toxicology is defined as the study of the adverse effects of chemicals on living organisms, and implies that a toxic substance is any agent capable of producing a detrimental response in a biological system. Most known substances have the potential of inducing injury or death if a sufficient amount is present. Paracelsus, a 16th century alchemist, is known for his statement that 'all substances are poisons; there is none that is not a poison. The right dose differentiates a poison and a remedy.'
In section II of the Canadian Environmental Protection Act (CEPA), it is stated that a substance is toxic if it is entering or may enter the environment in a quantity or concentration or under conditions a) having or that may have an immediate or long-term harmful effect on the environment; b) constituting or that may constitute a danger to the environment on which human life depends; or c) constituting or that may constitute a danger in Canada to human life or health. The inorganic fluorides (fluorides derived from inorganic compounds) have been assessed to be entering the environment in quantities or under conditions that may be harmful to the environment (including humans and other biota), and therefore placed on Schedule I of the Act (Canadian Environmental Protection Act, Health Canada, Government of Canada).
Properties and Sources of Fluoride in the Environment
Dietary Sources of Fluoride
Kinetics of Fluoride in the Body
Fluoride exists in the plasma as two forms - ionic( free) and nonionic (bound). It is the free form that is important in reference to both health benefits and toxic effects and is also the form that is measured with an ion-specific electrode. Plasma (blood) levels are proportional to intake, and can increase slowly over the years. Following fluoride ingestion, the levels in saliva increase corresponding to plasma levels (the normal concentration of fluoride in saliva is about 1 pm, slightly less than plasma). Fluorides are well distributed to most soft tissues, but approximately 50% is associated with calcified tissues (bones and teeth) within 24 hours of ingestion. Nearly 99% of the body burden (total amount of fluoride in the body) is associated with calcified tissues. The major route of removal or elimination of fluoride from the body is via the kidney (urine). Clearance via the kidneys can be influenced by a number of factors, including pH, diet, drugs, age, metabolic and respiratory disorders, altitude, exercise, and of course kidney function. Therefore all these factors have to be considered if studies are to be reliable.
Toxicity of Fluoride
Acute poisoning is ingestion of high enough doses that immediate effects are observed. The smaller the amount of substance that causes acute poisoning, the more toxic the material. Chronic effects are due to long term, low levels of exposure and effects are not seen during the beginning of that low level exposure.
There has been considerable information published on the acute toxicity of fluoride in the human body (see Whifford, 1996). Acute toxicity following ingestion of high doses can develop quite rapidly; initial symptoms usually include nausea, gastric distress and vomiting. With larger doses, muscle spasms, tetany and convulsions may develop. Toxicity may progress to include the cardiovascular system resulting in hypotension and arrhythmia. It has been estimated that the lethal dose for humans is 32-64 mg/Kg. This would be 2.25 to 4.5 gm of sodium fluoride for a person of 70 Kg. It would be impossible to drink enough at one time to absorb an acute amount of fluoride from fluoridated drinking water. Of more importance to the issue of water fluoridation is the potential for chronic (low dose) toxicity. The following segments address some of those concerns.
Effects of Fluoride on Target Organs
Gastrointestinal System: Fluoride can interact with gastric (stomach) acid to form hydrogen fluoride which may irritate the stomach mucosa. This effect is dose-dependent, and irritation only appears to occur at high (about 200 ppm) concentrations, well above normal levels ingested.
Reproductive System: Although some studies using animal models have demonstrated adverse effects on some reproductive functions, the concentrations of fluoride used in the studies were very high (above 100 mg/L). There is no association between fluoride intake and reproductive outcomes in human subjects.
Cancer (also see previous section)
There is no conclusive evidence that levels of fluoride found in fluoridated water are associated with greater incidence of cancer.
Mutagenicity: Fluoride is not mutagenic (causes mutations in cells) in standard bacterial systems, but has been shown to produce chromosome damage and gene mutations in cultured mammalian cells (Zeigler et al 1993). The same author, however, concluded that the matter is still unresolved since the effects have not been observed in vivo.
Neurotoxicity: The study published by Mullenix et al (1995) has received considerable attention. It demonstrated sex and dose-specific behavioral deficits in rats exposed to fluoride. The doses administered in this study were high (50-60 ppm). The authors justified these doses by the apparent similarity in plasma levels found in humans exposed to high levels. Other studies recently reported behavioral and morphological effects on developing rat brain, but also used high concentrations (60 ppm) in the tests. Although these data do demonstrate that high doses of fluoride may have the potential for adverse effects, there is no evidence for neurotoxicity at levels associated with water fluoridation.
Dental Fluorosis (see also previous section)
Very mild fluorosis is change in coloration that is just Visible to the trained eye. Mild fluorosis would not likely be noticed without careful scrutiny. More severe forms cause noticeable staining from foods and beverages.
There is evidence that the prevalence, and to some degree the severity of dental fluorosis, is increasing in both fluoridated and non-fluoridated areas. There are no data that this is occurring in Calgary, however it is of concern and should be addressed. Even low fluoride intake may result in a certain level of fluorosis. This was, in fact, predicted in the earlier work of Dean (1942) who estimated that 10% of the population would exhibit signs of mild fluorosis with water fluoridation at 1.0 ppm. The balance between the health benefits and adverse effects must be evaluated with any treatment. The concentrations of fluoride in drinking water have remained constant, therefore other sources of fluoride may be responsible for an increase in total fluoride intake and associated increase in dental fluorosis. The increase in total fluoride intake has resulted from the use of fluoride supplements and fluoridated products and food. It appears that the primary contributors are fluoridated toothpastes, fluoridated mouth washes, and beverages prepared with fluoridated water. Lewis and Banting (1994) have shown that 60% of the total prevalence of dental fluorosis may be attributed to fluoride sources other than fluoridated water. This has also been documented by others (e.g. Hiller et al 1997, Horowitz 1996, Pendrys and Stamm, 1990, Pendrys, 1995).
Fluoride concentrations measured by Alberta Environment (Water Sciences Branch, 1998) at Cochrane (upstream from Calgary) are slightly lower than those at Carseland (downstream from Calgary). It is worth noting that when fluoride concentrations measured downstream from Calgary between 1967 and 1974 are compared to measurements between 1987 and 1991, the difference is very small, in spite of the large population increase. The raw data may show a very small increase (of the order of 0.05 mg/L) between 1991 and 1994 (Standards and Guidelines Branch, 1998). Increases in fluoride concentration downstream from cities cannot be attributed solely to water fluoridation because of the many other sources of fluoride associated with diet and dental practices.
Fluoride is less toxic to aquatic organisms in hard water (water high in carbonate) such as in the Bow River (Environment Canada and Health Canada 1993).
Soils and Groundwater
No one on the Panel was an expert on tooth biology and biochemistry. We therefore submitted our questions about tooth biology and caries prevention to Dr. Colin Dawes of the Faculty of Dentistry at the University of Manitoba, who was recommended to us by the Faculty of Dentistry at the University of Alberta. To ensure balanced perspectives on the issue, the Panel asked for a recommendation from a group opposed to fluoridation, who recommended that we review the same questions with Dr. John Colquhoun, Professor Emeritus of the University of Auckland, New Zealand.
The Panel met with Dr. Dawes on February 11, 1998. He outlined the basics of enamel physiology, and the proposed biological means by which fluoride might protect enamel and prevent caries. He also provided commentary on some of the epidemiological studies of dental caries and water fluoridation. Dr. Dawes stated that an examination of dental benefits provided by fluoride showed that 0.7 ppm would be the recommended concentration in treated water. He provided the Panel with a short report and supporting documentation.
The Panel held a videoteleconference with Dr. Colquhoun for an hour on the evening of Feb 26, 1998. In Dr. Colquhoun's opinion, fluoridation has no beneficial effect on the incidence of caries, and may even have a detrimental effect. He noted that recommended concentrations of fluoride in treated water had been reduced in New Zealand. He provided the Panel with supporting documentation for his presentation.
The presentations by Dr. Dawes and Dr. Colquhoun were videotaped. The Panel was grateful to Drs. Dawes and Colquhoun for taking the time to assist us.
Dental Health is generally improving. There have been reports since 1989 questioning the impact of water fluoridation on dental health, suggesting that a number of factors, which were not present when water fluoridation was started, have had a greater impact on the decline in the incidence of caries. Most notably, these include better understanding of dental hygiene and use of new techniques in dentistry, changes in diet, other environmental sources of fluoride, and general improvement in the standard of living.
Given the declining incidence of caries in the population, some would suggest that water fluoridation is no longer needed, even if it is beneficial. That is, other sources of fluoride are effective so the argument is about mode of delivery, relative costs and effectiveness. Those favouring water fluoridation say that the low dose of fluoride in water protects those individuals who do not have access to individual fluoride treatment and/or dental care and instruction in dental hygiene. Given the concentrations of fluoride in toothpastes, tablets, and drops, and topical applications, the total amount administered is less uniform and the possibilities of excessive doses increase.
Dr. Colquhoun and Dr. Dawes both discussed the epidemiological evidence that suggested that fluoridation no longer appears to be having significant impact on the declining incidence of caries. Dr. Colquhoun felt that those studies showed that water fluoridation was no longer necessary. Dr. Dawes felt that the studies which showed a continued decline in dental caries in Germany following the cessation of water fluoridation may be explained by a number of other important variables affecting incidence and prevalence of dental caries: the method of defining and reporting caries; the overall changes in dental care including fluoridated dental products; and the increased use of fluoridated salt in Germany. Water fluoridation's effect on dental health may be obscured by other variables.
The Panel agreed that some studies showed that dental health was improving in areas not receiving
fluoridated water or that dental health continued to improve after fluoridation of water supplies was
stopped. However, we believe that to conclude that water fluoridation is ineffective or unnecessary is an
oversimplification. Much more detailed, statistically sound studies of total fluoride intake, dental
practices, etc. must be done to reach such a conclusion with confidence.
2. Human fluoride intake and its apportionment
Canadian Environmental Protection Act, Inorganic Fluorides Assessment Report, Table 3, gives the total fluoride intake as 3.28 - 4.07 mg/day with fluoride from drinking water being 1.00 - 1.56 mg/day. These figures seem to be slightly low compared to data reported elsewhere.
Dietary Reference Intakes, National Academy of Sciences, Table 8-2, at the bottom, reports an average daily fluoride intake from prepared foods served to adult hospital patients (excluding drinks) as 1.8 mg/day. Additional 2 l/day of drinking water recommended by nutritionists at 1 ppm fluoride yields 2 mg/day of fluoride.
Burgess in his Fluoride Ingestion from Dental Products estimates the total fluoride ingestion from various dental care products at .43 mg/day (dentifrice + rinse + topical gel). Thus the total fluoride ingestion is approximately 4.43 mg/day (depending on age), with 2 mg/day from the drinking water. These figures are very close to the fluoride ingestion data reported in the above paragraph.
Public Health Service, U.S. Department of Health and Human Services, in the report Review of Fluoride, Benefits and Risks, gives the total amount of ingested fluoride as 2.21 - 9.24 mg/day with fluoride from the drinking water being .84 - 4.48 mg/day. Even though the range of the ingested fluoride is rather wide, it is generally comparable to the ranges reported above.
The above given ranges have to be adjusted upwards for subpopulations such as tea drinkers, marine fish eaters, heavy labourers working in increased temperature environments, athletes and many others. It should be noted, that in all these reports the apportionment of fluoride from the fluoridated drinking water is not negligible. It can be estimated to be ranging approximately within 30 - 45% of the total daily fluoride intake by adults.
Historically, the fluoride intake in Canada used to be much lower. The Health Canada report titled Investigation of Inorganic Fluoride and its Effect on the Occurrence of Dental Caries and Dental Fluorosis in Canada estimates (see Table 1) that during 1940's (before the fluoridation of water was introduced) the total fluoride intake from food was about 1.05 mg/day. Other reports give a baseline fluoride intake during 1940's as .45 - .55 mg/day which is consistent with the Health Canada report.
Cauley (1995): Authors studied 2076 non-black women, all aged over 65 years, and incidence of osteoporotic fractures. The variable was the number of years of exposure to fluoridated water. They compared the relative risk (RR) of various types of fractures between fluoride treated and non-fluoride women. All the comparisons turned out as statistically non-significant. The mean number of years of fluoride exposure was too low at 12.7 years to be able to demonstrate any significant effect. Furthermore, the confidence intervals for RR were incredibly wide (e.g. 0.10 - 1.89) demonstrating a poor statistical quality of the data which resulted in the fact that the findings were statistically non-conclusive.
Danielson (1992): The authors tested the effect of 1 ppm fluoridated water on the incidence of hip fractures in the elderly population in Utah. The incidence of femoral neck fractures in patients 65 years of age or older was compared in three communities in Utah: one with and two without water fluoridated to 1 ppm. Authors excluded from the study spurious cases. The relative risk of hip fracture in the fluoridated area was 1.27 (95% confidence interval 1.08-1.46) for women and 1.41 (1.00-1.81) for men. This significant increase in the risk of hip fracture in both men and women exposed to artificial fluoridation at 1 ppm suggests that low level of fluoride may increase the risk of hip fracture in the elderly. The authors examined a population with a different life-style (Latter-Day Saints church) which has lower exposure to various risk factors for osteoporosis, such as smoking and alcohol. Thus the relative contribution of fluoride to hip fracture may have been more evident than in other studies, which dealt with a more heterogenous populations.
Jacobsen (90/92): In order to assess the association between water fluoridation and hip fractures, authors identified 129 counties across US considered to be exposed to public water fluoridation and 194 counties without exposure. Data from Health Care Financing Administration and Veterans Affairs were used to calculate the incidence of hip fractures among white persons aged 65 years of more. There was a statistically significant weak positive association between fracture rates and fluoridation. The corresponding 95% confidence interval for the relative risk for men was very narrow at 1.13 - 1.22 pointing to the statistically significantly increased risk of hip fracture in fluoridated areas compared to nonfluoridated areas.
Jacqmin-Gada (1995): Authors report results of a population study of the relationship between concentration of fluorine and calcium in drinking water and risk of hip fractures or fracture at any site. The study investigated 3777 subjects aged 65 years or older living at home in 75 parishes of southwest France. The mean time the individuals remained in the same parish was 41 years. Fluorine and calcium concentration in water was measured from each parish. Five covariates were investigated: age, sex, Quetelet index (weight/height squared), smoking status and sport activity. Authors performed multiple logistic regression for data explanation. The risk of hip fracture was significantly higher when water fluorine concentration was higher than 0.11 mg/litre (p=.04). Thus adjusting for major risk factors, this study suggests a possible deleterious effect of fluorine in drinking water on the risk of hip fractures, even for moderate concentrations of fluorine.
Sowers (1986): Bone mass of women in three rural communities with different mineral water content was studied. Mean fluoride and calcium were 4 ppm and 16 ppm respectively for the first community, 1 ppm / 375 ppm for the second and 1 ppm / 65 ppm for the third. Bone mass was measured by photon absorption and women were interviewed about fracture history, dietary intake and other important covariates. Authors observed no protective effect with high fluoride intake. Women in high fluoride area reported significantly increased incidence of fractures. This is a very important study because individual subjects were evaluated for various specific variables which have a possible effect on hip fractures.
Sowers (1991): This is a follow-up of the previous study by Sowers (1986). In 1989 there were 684 women still living in same communities as earlier. Residence in the higher fluoride community was associated with a significantly lower radial bone mass in pre-menopausal and post-menopausal women and significantly higher incidence of fractures among post-menopausal women. The corresponding 5-year relative risk for women in high fluoride community, compared with the control community, was significantly high at 2.2. Estimates of risk were adjusted for age and body size.
Suarez-Almazor (1993): The purpose of this study was to compare hip fracture hospitalization rates between Edmonton and Calgary. Edmonton had fluoridated water since 1967 and Calgary, which at the time of the study (1981-1987) did not have fluoridated water. Case subjects were individuals aged 45 years or older. The hip fracture rates were adjusted for age and sex. No statistically significant difference was found for the entire population or for women only. However the hip fracture rate for men 65 years and older in Edmonton was 5.09 while for Calgary the rate was 4.47. This increased hip fracture incidence is statistically significant and represents an increase of almost 14%.
Whitford (1996 page 138) as well as the National Research Council report on Health Effects of Ingested Fluoride (page 59) state: "Most estimates indicate that crippling skeletal fluorosis occurs when 10 - 20 mg of fluoride have been ingested on a daily basis for at least 10 years".
3.2 Effect of therapeutic dose fluoride
Fratzl (1994): Bone biopsies were taken from three patients with postmenopausal osteoporotic bone before and after fluoride therapy (60 mg NaF/day for 1-2 years) as well as from three normal controls. An increasing amount of new bone is laid down on the surface of preexisting bone. Its mineral structure is identical to that of heavy fluorosis and is characterized by the presence of additional large crystals, presumably located outside of the collagen fibrils. These large crystals, which were not present in the controls and before the fluoride treatment, contribute to increase the mineral density without significantly improving the biomechanical properties of the bone. As soon as significant amounts of fluoride are present, bone turnover leads to the replacement of old normal bone by new pathological bone. Decreasing biomechanical properties of the bone may even lead to deterioration in the overall mechanical stability of the skeleton.
Hedlund (1989): The authors compared the incidence of hip fractures of four groups of osteoporotic women. 22 treated with placebo, 17 with fluoride and calcium, 18 with fluoride and calcitrol, and 21 with calcitrol alone. Slow release sodium fluoride, 50 mg/day, for 12 months was used as fluoride treatment. The difference in fracture rates for fluoride treatment versus nonfluoridated treatment is highly significant (p=.006). Moreover, the six hip fractures occurring in patients receiving fluoride treatment for 72 patient-years of treatment is 10 times higher than would be expected in normal women.
Riggs (1990): To assess the effect of fluoride treatment on the fracture rate in osteoporosis, the authors concluded 4-year clinical trial with 202 postmenopausal women with osteoporosis who were randomly assigned to receive sodium fluoride (75 mg/day) or placebo. In total 66 in the fluoride group and 69 in the placebo group concluded the trial. The number of non-vertebral fractures was significantly higher in the treatment group than in the placebo group (72 versus 24).
Riggs (1994): Authors present observations on osteoporotic 50 women receiving 50 mg/day of NaF or placebo for up to 6 years. Multivariate statistical analysis revealed a very complex pattern in which vertebral fracture rate was influenced by several variables. There was a two fold statistically significant increase in hip fracture occurrence in a randomized controlled study when the women were administered a relatively low dose of only 50 mg of sodium fluoride per day.
Sogaard (1994): In order to evaluate the effect of sodium fluoride (40-60 mg/day) on bone mechanical competence, iliac crest biopsies were taken before and after one year of treatment in 12 osteoporotic patients, and before and after five years of treatment in 14 patients. Bone fluoride content increased after both treatments. A significant reduction of 45% was found in trabecular bone strength after 5 years treatment and 58% reduction was found in trabecular bone quality. The results of this study indicate that long term administration of NaF may be detrimental to bone quality.
4. Decrease in caries
Report of the Second Alberta Dental Health Survey, 1985, compared incidence of DMFT rates between various regions of Alberta. Of particular interest is the comparison between Calgary, which did not have fluoridated drinking water at that time and Edmonton, which provided fluoridated water since 1967. The report concentrated on the cohorts of 13 years old children. Edmonton Board of Health Annual Report for 1985 reported for these Edmonton children mean DMFT rate as 2.80. The corresponding DMFT rate for the Calgary cohort was 2.995 DMFT. This difference is not statistically significant. How can it be explained that Edmonton with fluoridated drinking water did not have statistically lower incidence of DMFT than Calgary with nonfluoridated water? The authors stated that all children in Calgary in 1985 were rated as "fluoridated" due to exposure to fluoride from other sources (toothpaste, dental treatments etc.) even though fluoride was not added to the water.
Heller (1997): The purpose of this study was to investigate the relationship between caries experience and dental fluorosis at different fluoride concentrations in drinking water. The impact of other fluoride products was also assessed. The use of fluoride tablets, drops, professional fluoride treatments and school fluoride rinses were ascertained from caregiver questionnaires. The sharpest declines in DMFS were associated with increase in water fluoridation between 0 and 7 ppm. F, with little additional decline between .7 and 1.2 ppm. Fluorosis prevalence was 13.5%, 21.7%, 29.9% and 41.4% for children who consumed <.3 ppm, .3 to .7 ppm, .7 to 1.2 ppm and over 1.2 ppm fluoridated water. The author states that a suitable trade between caries and fluorosis appears to be around .7 ppm F. Data from this study suggest that a reconsideration of the policies concerning most appropriate water fluoridation might be appropriate.
Hildebolt (1989): The objective was to determine how the prevalence of caries vary between regions of Missouri. In the total sample, there were no significant difference between those children drinking optimally fluoridated water and those drinking suboptimally fluoridated water.
Ismail (1993): An epidemiological assessment of differences in caries prevalence between children in Truro (<.1 ppm fluoride in water) and Kentville (1.1 ppm fluoride), Nova Scotia was completed. Among other findings, the percentage of all caries free children in non-fluoridated area (26.8%) is higher (even though not significantly) than the fluoridated area (23%). The percentage of very high caries (5 or more) is lower in the fluoridated area (10.3% versus 22%). However the overall mean number of all caries in the fluoridated area (3.5) does not significantly differ from the mean number of all caries in the non-fluoridated area (4.2) and the difference is only .7 caries.
Kunzel (1997): The authors investigate the rise and fall of caries prevalence (DMFT) and its relation to changing F concentration of drinking water and other health related factors based on findings of more 286,000 subjects of either sex. Water fluoridation at 1 ppm was implemented in Chemnitz (formerly East Germany) in 1959 and was operated till unification of Germany in 1990. A comparison is made with Plauen, city with F poor (.2 ppm) water. In 1959 the mean caries incidence in two cities was quite comparable at about 4. Since 1959 till 1971 Chemnitz water was fluoridated at .9 ppm and caries incidence significantly decreased to about 2 despite an increase in sugar consumption from 27 to 35 kg per capita/year. Over the same period, the caries incidence in Plauen (receiving F poor water) increased to about 5.
There were no caries-preventive facilities or oral hygiene program in either city during this period and number of topical dental F applications was very low. Water fluoridation was introduced in Plauen in 1972 and caries incidence started decreasing till about 3 in 1990. Due to technical difficulties in Chemnitz water fluoridation was interrupted in 1971 for a longer period of time and the caries incidence again rose to about 3 in 1990. After 1990 water fluoridation was terminated in both cities due to new laws after reunification of Germany and both cities received only F poor water. However the caries incidence continued to decrease in both cities to the lowest level ever at 2 in 1995. This is explained by completely different conditions for management of caries: F-enriched domestic salt was offered since 1992 and reached 15% of the market in 1995. The sugar consumption decreased from 42 kg to 35 kg in 1993 due to nutritional education. The use of fluoridated toothpaste increased from 18% to 88% after 1992. The dental care system was entirely restructured in this period and its efficiency greatly increased. The authors conclude that the relation between F concentration in drinking water and the caries level valid from 1959 to mid 1980's is no longer true. The main differences between 1959 - 1971 period and after 1990 period are broader F availability (mainly toothpaste) and increased awareness of oral hygiene. An argument that changes in diagnostical criteria may partially explain the decline may have some merit but the decline was evident even before the new diagnostic criteria were introduced in 1993.
Johnston (1986): Dental caries prevalence in Ontario health units from 1951 to 1956 and data from 1972 to 1984 for children aged 5, 7, 9, 11 and 13 years are analyzed to determine the rate of decline in caries over those years. The provincial average rate decline over the last two decades was 50% and there is no evidence that this is lessening. Decline in tooth decay cannot be fully explained by water fluoridation.
Newburn (1989): The efficacy of communal water fluoridation in reducing dental caries has been reviewed based on surveys conducted in the last decade in fluoridated and nonfluoridated communities in the US, Australia, New Zealand, Britain, Canada and Ireland. The efficacy is greatest for the deciduous dentition and decreases as the age increases. Effectiveness of water F has decreased as the benefits of other forms of F have spread to communities lacking optimal water F. Splieth (1996): Water fluoridation in East Germany was introduced in 1959 and continued until 1990 unification of Germany. Despite the discontinuation of water fluoridation, caries incidence is decreasing. Extensive use of F toothpaste after 1990 in East Germany led to caries reduction even though water fluoridation discontinued. Also the use of dental sealants and dental prevention programs were expanded.
The significant decrease in the incidence of dental caries over the past 20 years cannot be explained only as a consequence of water fluoridation. It was estimated that total fluoride intake in 1940's was about .5 - 1 mg/day. However, this is clearly not true any more: the total fluoride intake from various sources is estimated to be almost 10 times higher (3-9 mg/day). There are many other very important factors such as improved knowledge and availability of high quality nutrition, improved oral hygiene and care and mainly very pervasive fluoride availability in food (bottled water, salts, grains, tea, marine fish etc.) and in the form of dental products (toothpaste, rinses, gels and paints etc.) which now contribute to statistically
significant caries decline even in areas with nonfluoridated water. Many studies provide evidence that today water fluoridation is not the main reason of caries decline and such practice should be reconsidered. The 1985 Alberta Dental Health Survey provides data documenting that caries rate among 13 year old children in Edmonton (water fluoridated since 1967) and Calgary (water not fluoridated) was not statistically different. Another study by Kunzel demonstrates that during 1960 to mid 1980's, the increase or decrease in the incidence of dental caries in Germany was very closely associated with levels of fluoride in the water. However, after 1990 the fluoridation was terminated everywhere in Germany, yet caries incidence keeps declining to its lowest levels ever.
Some authors argue that specific cohorts of the general population, particularly those with lower socioeconomic status, may benefit from continued water fluoridation. However, these studies show either the effect on deciduous (first) teeth only but not on permanent teeth which are most important for human health or the difference is very small and equal to about .1 - .7 surface of a tooth. Other studies emphasize that cohorts with lower socioeconomic status are deprived of high quality nutrition and as a consequence are exposed to increased health risks from fluoride exposure. Furthermore, the lack of knowledge about importance of oral care and availability of proper dental care has a very significant effect in this regard.
The inference that "crippling skeletal fluorosis occurs when 10 - 20 mg of fluoride have been ingested on a daily basis for at least 10 years" (Whitford (1996 page 138); the National Research Council report on Health Effects of Ingested Fluoride (page 59)) is generally accepted in the scientific literature. The actual range of total ingested fluoride (as reported in section 2 above) in cases of 1 ppm fluoridated water is 3 - 9 mg/day. Compared to the above range given by Whitford, the documented total fluoride intake represents a potential risk of mild to moderate skeletal fluorosis in adult populations drinking water fluoridated at 1 ppm over long periods of time. Similarly increased risks are present for population cohorts with increased fluoride ingestion such as heavy tea drinkers, consumers of marine fish or for cohorts with impaired fluoride metabolism such as renal dysfunctions.
There is also evidence that an exposure to drinking water containing fluoride at 1 ppm over a long period of time may be associated with a statistically significant increase in the hip fracture incidence and corresponding relative risk. Possibility of a such effect on bone is further documented by studies using high doses of fluoride over a shorter period of time for therapeutic treatment. Of special interest is the study by Suarez-Almazor, which shows that the hip fracture rate for men 65 years and older in Edmonton (fluoridated water) was significantly higher than in Calgary before the introduction of water fluoridation.
Heller concluded that a suitable trade-off between a potential caries decrease and fluorosis increase appears to be around .7 ppm F and that there is not much improvement in caries decrease for fluoridation over .7 ppm. However, fluoridation over .7 ppm sharply increases prevalence of fluorosis.
Toxicological Profile for Fluorides (1993) issued by the Agency for Toxic Substances and Disease Registry unequivocally states that there are populations which are unusually susceptible to the toxic effects of fluoride. These populations include the elderly, people with deficiency of calcium, magnesium, and/or vtamin C, and people with cardiovascular and kidney problems. Because the fluoride is excreted through the kidneys, people with renal insufficiency would have impaired renal clearance of fluoride. Poor nutrition increases the incidence and severity of dental fluorosis and skeletal fluorosis.
Summarizing the above facts I can conclude that recently published literature provides biostatistical evidence of possible statistically significant association between fluoride intake from drinking water and increased hip fracture rates as well as skeletal fluorosis incidence. This association cannot be construed as clear cut deterministic cause - effect relationship and much more research will be needed to clarify very complex relationship between fluoride intake and human health.
However prudence dictates that potential risks cannot be disregarded either. Unfortunately there is no actual data available at this time to critically assess the effect of water fluoridation on general health in Calgary.
The Panel agrees that there are documented health benefits, in the control of tooth decay, from the fluoridation of water. Water fluoridation is a safe and inexpensive mechanism to improve the health of Calgarians, especially the 25% with lowest incomes, who can least afford alternatives. The Panel also notes that some other means of providing protection from tooth decay are more likely to lead to inadvertent ingestion of high fluoride doses, particularly when young children swallow toothpaste containing high concentrations of fluoride. The relative health benefits are now less than they were forty years ago, because of other sources of fluoride in the diet, better oral hygiene and better dental procedures, as well as a general improvements in overall health. However, these other improvements have not reduced the benefits of water fluoridation to the point where it is no longer needed.
Dental fluorosis occurs when total fluoride ingested is too high during the formation of enamel on children's teeth. Because the total intake of fluoride from all sources is increasing, more fluorosis is being observed although much of it is of the mild forms, which are only apparent to the trained eye or upon very close inspection. The Panel recommends that health authorities pay more attention to identifying uncontrolled sources of fluoride, especially intake due to children swallowing high fluoride toothpaste. Reducing uncontrolled sources of fluoride would be a more effective means of reducing dental fluorosis than eliminating fluoridation of water.
The scientific literature on fluoride and bone fractures, especially hip fractures in the elderly, did not provide evidence that would lead to substantial changes in water fluoridation policy. Questions about exercise and activity, calcium and vitamin D intake, overall health status, other sources of fluoride, use of other medication, and general standards of osteoporosis medical practice in the studied communities have to be addressed before the results of the inconclusive epidemiological studies can be confirmed.
From the perspective of epidemiology and toxicology, the available scientific literature has not substantiated the claims that water fluoridation was a factor in other adverse health effects. The results found in the literature have not eliminated the need for further research. Carefully designed studies, which take into account total fluoride intake and all other relevant factors, are required. There is no need for or value in further studies which attempt to relate water fluoridation per se to adverse health effects.
Although there is considerable literature on the effects of fluoride, many of these studies were related to high dose toxicity or to the effects of high therapeutic doses. Studies at these high levels were not considered relevant by the Panel.
The scientific literature has not revealed significant health concerns about fluoride at currently estimated levels of intake. However, there are legitimate health concerns about higher levels of total intake, which are not generally associated with fluoridated water but are due to other sources which increase the total intake above safe levels.
Fluoride is a natural component of soils and water. In the Calgary region, there are no large active industrial sources. Such industrial sources have produced environmental problems in other parts of Canada. Because of the carbonate base for our soils and the hard water in our surface and ground water, environmental concerns are lower relative to areas with soft water and acid soils.
The Panel is pleased to note that Alberta Environmental Protection has recommended that the optimum level of fluoride in treated water has been reduced from 1.0 mg/L (1 ppm) to 0.8 mg/L (0.8 ppm). A similar reduction has been made in other jurisdictions and reflects the recognition that total fluoride intake has increased since the standard was first established.
The fact that these issues have been articulated for more than 50 years, and have received a good deal of study, suggests that if there are cause and effect relationships between water fluoridation and adverse health effects, they are small and very complex. Otherwise, more substantial results would have been achieved by the studies undertaken to date. Although there are recommendations for further research, the results of which would have made this Panel's work easier, the absence of that work did not substantially affect the Panel's conclusions.
With respect to fluorosis, the Panel recommends that health authorities provide more advice on control of total intake in young children, especially from dental products.
The Panel recommends that carefully designed studies of the possible adverse health effects of
total fluoride intake, which consider other relevant factors (diet, health status, vitamin and mineral
intake, other medication, etc.), be undertaken. The Panel recommends that health authorities
undertake careful studies of total fluoride intake, with particular emphasis on identifying
significant possible sources high in fluoride, and effective means of reducing exposures.
TERMS OF REFERENCE - EXPERT PANEL FOR WATER FLUORIDATION REVIEW
Waterworks Division is involved in the fluoridation issue only as a purveyor, giving sole attention to the proper operation, monitoring and compliance of the fluoridation process. Thus, it has been directed by the Operations and Environment Committee to assume the role of coordinating and organizing an independent and unbiased panel of experts. Waterworks approached various maturity and independence of university researchers were considered important assets for prospective panel members. an adequate expertise profile to address technical issues of health and safety was sought. An effort was made to exclude mission oriented candidates (either strongly pro-fluoridation or strongly anti-fluoridation).
PANEL RESOURCE CONTACTS:
Dr. Read Seidner, Laboratory Superintendent, Waterworks Division
2. Work is to commence as soon as possible and completed by a presentation to Operations and Environment Committee. Prior to that a written brief is to be submitted to Waterworks Division and Calgary Regional Health Authority. The panel chairman is responsible to verbally present the panel's findings to the Standing Committee on Operations and Environment.
3. The focus of effort will be on literature produced since 1989. Literature additional to already received submissions will be sought through newspaper ads or may be provided by expert panel members. Verbal presentations at the panel's invitation only may be sought. If such are sought, then effort will be made to include equal opportunity from both sides.
4. Standard scientific criteria as to evaluating the pertinence of reports and studies will be employed. Methodology employed by the Canadian Task Force on the Periodic Health Examination will serve as a model. The panel will decide its own process of dividing up submitted literature.
5. All costs will be split equally by Calgary Regional Health Authority and Waterworks Division. Secretarial and library search assistance will be provided to the panel. The panel is to assess additional resources needed once all submissions are received.
SUMMARY OF MATERIAL RECEIVED FOR FLUORIDE REVIEW PANEL: March 9, 1998
Background briefs on Connett, Mullenix, Schatz, Foulkes, Hirzy and Burgstahler