Index | S1 | S2 | S3 | S4 | S5 | S6 | S7

Section 4.

4.10 Fluoride and Bone Health.

The one issue for which the FF authors did attempt a semi-independent review was Fluoride and Bone Health. However, this section is marred by an unwillingness, or inability, to use a weight of evidence approach to this issue and their selective use of the literature.

a) Selective studies.

The Forum authors discuss two recent studies in which no increase in hip fracture associated with fluoride in water was found (Hillier et al, 2000, and Phipps et al, 2000) but fail to discuss another recent study that did (Li et al, 2001). Moreover, in discussing Phipps's findings, the Forum authors point out that fluoride exposure was associated with a decrease in hip fracture and vertebra fracture, after correcting for 13 variables, but failed to point out that she also found an increase in wrist fracture, which was only a thin hair off significance. In the York report, the wrist fracture finding was reported as significant (see Connett 2001). Dr. Connett discussed this situation with the Forum in October, 2000 but they chose to ignore his comments. Why?

As the Forum authors point out, increased hip fracture in the elderly is a very serious -- even life-threatening -- situation and the number of hip fractures in Ireland has approximately doubled over the last ten years. It is critically important therefore to examine all the evidence to see if it is possible that exposure to fluoride over a whole lifetime could lead to weakening of the bones.

b) Omitted studies.

Li et al (2001).

It is inexplicable that the Forum authors failed to discuss the study by Li et al (2001), whose findings Dr. Connett presented to the panel in his presentation in October 2000. At the time, the study was unpublished, but it has since been published in the Journal of Bone and Mineral Research, several months before the Fluoridation Forum report was finalized.

This study is important in our view because it provides a very convincing dose-response relationship between hip fracture rates in the elderly and the level of fluoride in their drinking water. It is also important because it helps us to define a margin of safety for this very serious outcome. Li et al (2001) compared the hip fracture rates for the elderly in six Chinese villages with fluoride at levels in their well water which ranged from 0.25 ppm to 8 ppm. Using the village at 1 ppm as control, they found that the rates of hip fracture doubled when the levels went over 1.5 ppm, a result which was not statistically significant, and tripled when they went over 4.5 ppm, a result which was statistically significant. Unfortunately, the York review (McDonagh et al, 2000) concentrated on comparing the rates of hip fracture in the village at 1 ppm, with the villages with less than 1 ppm and in this comparison Li et al found no statistical difference. In our view, that is taking a rather myopic view of the significance of this data. The fact that rates doubled over 1.5 ppm and tripled over 4.5 ppm gives us a clear indication that we are dealing with a health problem with a very narrow margin of safety: possibly, as low as 1.5! Moroever, as the researchers found that the consumption of this well water was the dominant source of fluoride for these people, it raises the question of what happens to people with lifelong exposure to fluoride, not only from water but with other sources such as dental products; processed food and beverages prepared in fluoridated water; pesticide residues and air pollution.

c) Weight of evidence from bone studies.

Standing on its own, the study of Li et al (2001) is highly significant. It becomes even more so when considering the studies of Sowers et al (1991), which found an increase in hip fracture in a US community drinking water at 4 ppm fluoride, and Turner et al (1992), which found that an equivalent of 4 ppm fluoride in the drinking water of rats resulted in significantly weaker bones.

Meanwhile, it is now clear from numerous animal and clinical studies that fluoride decreases the strength of bone (Roholm 1937; Gedalia et al., 1964; Daley et al., 1967; Beary 1969; Wolinsky et al., 1972; Chan et al., 1973; Riggins et al., 1974; Inkovaara et al., 1975; Riggins et al., 1976; Gerster et al., 1983; Moskilde et al., 1987; Hedlund & Gallagher, 1989; Kragstrup et al., 1989b; Bayley et al., 1990; Gutteridge et al., 1990; Riggs et al., 1990; Schnitzler et al., 1990; Turner et al., 1992; Sogaard et al., 1994; Sogaard et al., 1995; Turner et al., 1996; Turner et al., 1997; Haguenauer et al., 2000; Gutteridge et al., 2002). (See appendix 4)

The question is not whether fluoride reduces the strength of bone, but, at what level? It is this question which the Forum should have addressed -- but didn't.

At least two things need to be investigated. First, the levels of fluoride that accumulate in the bones of people living, for various periods of time, in fluoridated communities, need to be determined. Second, the levels of fluoride in bone that have been found to decrease bone strength, in both animal and clinical trials, need to be assessed. Thereupon, if the concentration of fluoride in bone which has been found to decrease bone strength in animal and clinical studies equals or exceeds the level of fluoride now found in the bones of some people living their whole lives in fluoridated communities, then it is clear there is a potential problem

Unfortunately, the Forum gives no indication at all of how much fluoride is accumulating in the bones of the Irish people. While research into fluoride exposure has been undertaken, no official, comprehensive results have emerged in the open literature. This state of affairs can no longer be considered a mere oversight.

However, from the scraps of data that are available from other countries, there is definite cause for concern.

In 1980, Alhava et al (1980) measured the concentrations of fluoride in the bones of people who had lived in a fluoridated community in Finland for less than 20 years, and compared it to the levels of fluoride in people from a non-fluoridated area.

According to Alhava, the average level of fluoride found in the bones of women in the fluoridated area was 1,360 ppm (cortical) and 2,070 ppm (trabecular), with some of the women having as much as 4,000 ppm.

An earlier study by Parkins et al (1974), found a range of 1,295 to 5,745 ppm in the iliac crest bones of people living in a fluoridated area of the United States. The average level of fluoride was 2,824 ppm.

A more recent study by Richards et al (1994), from Denmark, found the level of fluoride in bone to range from 463 to 4,000 ppm, with the average level for women being 1,337 ppm and for men 1,181 ppm. What's striking about this study, is that the bones came from people living in a non-fluoridated area. Moreover, another study by the same team (Sogaard 1994), found fluoride concentrations in osteoporotic patients (before treatment) to be as high as 4,250 and 6,500 ppm!

Thus, based on the limited data currently available, fluoride concentrations in human bone from areas with 1 ppm fluoride or less in the water, ranges from as low as 500 to as high as 6,500 ppm.

To put these figures into perspective, the following is a list of fluoride concentrations which have been found to reduce the strength and/or quality of animal bone.

1,963 - 2,223 ppm in quail. (Chan et al., 1973)
1,704 ppm in pigs (cortical bone). (Kragstrup et al., 1989b)
2,826 ppm in pigs (trabecular bone). (Moskilde et al., 1987)
3,300-4,600 ppm in rats. (Sogaard et al., 1995)
4,500 ppm in rats. (Turner et al., 1993)

From these studies, it appears that the threshold at which fluoride reduces the strength and quality of bone is somewhere between 2,000 and 4,500 ppm.

However, whether the threshold is as low as 2,000 ppm or as high as 4,500 ppm, it is clear from recent human data that there will be people in the population exceeding this level. We find this deeply disturbing.

It is particularly disturbing when considering that 9 studies conducted since 1990 have found a significant association between fluoridated water and hip fracture (Jacobsen et al., 1990; Cooper et al., 1991; Keller 1991; Danielson et al., 1992; May & Wilson 1992; Jacobsen et al., 1992; Jacqmin Gadda et al., 1995; Kurttio et al., 1999; Hegmann et al., 2000).

An additional study also found an association between fluoridation and hip fracture (Suarez Almazor et al., 1993), although the authors dismiss the association since it was slight and only found in men.

Since it is important in public health policy to consider the worst-case scenario, it is essential that any discussion of fluoridation and bone, consider those individuals with inadequate nutrition, failing kidneys, and excessive thirst (and combinations thereof).

In the Forum report, there is no mention at all of how water fluoridation might affect the bones of people with any or all of these conditions. This despite the fact, now well established, that poor nutrition (particularly a deficiency of calcium) reduces the concentration at which fluoride reduces bone strength (Beary 1969; Riggins et al., 1974; Riggins et al., 1976); and that calcium deficiency, poor renal function, and excessive thirst (which often accompanies poor renal function) all serve to increase the bone's accumulation of fluoride -- sometimes dramatically so (Jackson 1955; Adams & Jowsey 1965; Call et al., 1965; Beary 1969; Juncos & Donadio 1972; Riggins et al., 1974; Spencer et al., 1980; Gerster et al., 1983; Noel et al., 1985; Welsch et al., 1990; Turner et al., 1996).

It is thus imperative for any analysis on how water fluoridation may affect bone to consider what lifetime exposure to fluoride might do to an individual with any or all of the above conditions.

The Forum's report doesn't come close to answering any of these questions. This despite the fact that many Irish people undoubtedly have some of these conditions.

Lastly, it is unfortunate that the Forum did not provide their own independent assessment of the numerous recent clinical trials which have examined how high-dose fluoride treatment affects bone. Instead, the Forum relies on yet another review (from the Australian Dental Journal) to serve as a substitute for their own analysis.

This represents yet another serious weakness and missed opportunity of the Forum report. For, in contrast to the ecological studies, the clinical trials provide the most thorough and scientific evidence on how fluoride affects human bone (see Inkovaara et al., 1975; Gerster et al., 1983; Vigorita & Suda 1983; Riggs 1984; Dambacher et al., 1986; O'Duffy et al., 1986; Kragstrup et al., 1989; Hedlund & Gallagher, 1989; Hodsman & Drost 1989; Bayley et al., 1990; Riggs, et al.; 1990; Kleerekoper et al., 1991; Fratzl et al, 1994; Schnitzler et al., 1986; Sogaard et al., 1994; Lundy et al., 1995; Pak et al., 1995; Patel et al., 1996; Balena et al., 1998; Haguenauer et al., 2000; Gutteridge et al., 2002). These studies provide an invaluable tool for guiding public health policy on the matter. (See appendix 4)

d) Cumulative dose versus daily dose.

The daily doses of fluoride used in the clinical trials are considerably higher than one would receive drinking fluoridated water.  However, this is no reason to dismiss the relevance of the findings, particularly considering that the high doses used in these trials were given over very short periods of times (e.g. 1-5 years) compared to lifelong exposure (70 years or more) to fluoridated water (as well as other sources of fluoride).

More useful from a public health point of view would be a careful analysis which compares the total, cumulative dose of fluoride delivered in these trials versus the total, cumulative dose of fluoride one could expect to receive living one's whole life in a fluoridated area.

Such an analysis is not difficult to do. For example, if one multiplies the daily dosage (33.75 mg/day) of fluoride used in the Riggs study (1990) by the number of days in the year (365), and by the number of years in the study (4), one will find that the total dosage given in the trial was roughly 49,275 mg of fluoride.

According to Riggs, patients receiving this dosage had an increased rate of hip fracture. To receive the same amount of fluoride as delivered in the Riggs study, a person would need to consume an average of 2.7 mg of fluoride a day for 50 years, 2.25 mg/day for 60 years, 1.8 mg/day for 75 years, or 1.5 mg/day for 90 years.

What's striking is that all of these dosages (1.5 - 2.7 mg/day) are well within the current estimates for how much fluoride people living in fluoridated areas are now receiving. For instance, according to a 1991 review by the US Public Health Service (DHHS 1991), the average daily ingestion of fluoride in a fluoridated community ranges from roughly 1.6 to 6.6 mg/day.

More troubling is that other clinical trials have found an increase in bone fracture at dosages considerably lower than the Riggs study. For instance, the Hedlund and Gallagher (1989) study found an increase in hip fracture in patients receiving just 22.5 mg fluoride per day for just two years, not four.

Thus, the total fluoride dosage used in the Hedlund study (16,425 mg) was approximately a third of the dosage used in the Riggs study. To receive this same total amount of fluoride, a person would need only to consume 0.9 mg of fluoride per day for 50 years, 0.75 mg/day for 60 years, 0.6 mg/day for 75 years, and 0.5 mg/day for 90 years. Such doses (0.5-0.9 mg/day) are routinely, and often grossly, exceeded in fluoridated areas.

Although there are likely other factors to be considered when making such comparisons (i.e the pre-treatment accumulation of fluoride in the osteoporotic patients and the potential differences between rapid and gradual accumulation of fluoride), the compatibility of dosages between the short-term trials and long-term real-life exposures is a definite cause for concern.

It is revealing to look closely at the words the Forum report authors chose to dismiss concerns about bone. They state:

"Trials have shown that high doses of sodium fluoride substantially increased vertebral bone density, but this effect was not associated with lower rates of spinal fractures (114). This effect has only been seen when intake has been substantially higher than would be expected from fluoridation of water. Sodium fluoride as an anabolic substance was used in the past in the management of osteoporosis, but is no longer licensed in Ireland and Europe. It prolongs bone remodelling if given in twice the therapeutic dose. Experimental studies have shown that fluorotic bone is more resistant to compressive forces, but more easily fractured by torsional strains. Moderate doses of fluoride have been shown to increase bone strength in experimental animals and high doses of continued exposure decrease strength."

And later in their conclusion to the section on bone, they state:

"The use of fluoride in the treatment of osteoporosis was referred to above. The use of high doses of fluoride in the treatment of osteoporosis is no longer a therapeutic option. However, the role of low doses of fluoride, as is obtained in drinking water, is the subject of a systematic review (118)."

This truncated analysis does not begin to do justice to this issue, either qualitatively or quantitatively. In the first sentence above, you will note that the Forum authors talk about fluoride treatments not being associated with "lower rates of spinal fractures". However, what they do not state is that the fluoride treatment of osteoporotic patients has actually led to increased hip fracture rates, i.e. the very opposite result to that intended. The forum states that fluoride was used in the management of osteoporosis "but is no longer licensed in Ireland and Europe", but they don't tell us why. What they don't stress or even acknowledge is that this treatment is not licensed because it did not decrease hip fracture rates but often increased them.

In their concluding remarks, the Forum authors are extremely misleading when they state: "The use of high doses of fluoride in the treatment of osteoporosis is no longer a therapeutic option. However, the role of low doses of fluoride, as is obtained in drinking water, is the subject of a systematic review."

Again, the authors do not stress that in clinical trials fluoride increased hip fracture rates; all we are told is that these treatments are "no longer a therapeutic option".

Instead of quantifying the comparison between the doses used in these clinical trials and the doses estimated for lifetime exposure to fluoride in optimally fluoridated communities, readers are simply told about the "high doses" used in the treatment of osteoporosis, and the "low doses of fluoride, as is obtained in drinking water". The juxtaposition of the words "high" and "low" in these two sentences is highly misleading. It obfuscates the fact that the "cumulative dose" is comparable, as we have demonstrated above.

In short, the Forum makes no attempt to analyze this clinical data in any toxicologically meaningful way. It simply implies that, since the dosages used in the trials were higher on a daily basis, that the trials have no relevance to water fluoridation. Such a conclusion -- especially without any supporting analysis -- is cavalier and crude.

e) Fluoride and bone damage in children.

There are two other studies not mentioned by the Forum, which point to the fact that fluoride might also damage the bones of children.

i) Schlessinger et al, (1956).

Schlessinger et al found a statistically significant increased incidence (13.5% versus 7.5%) of cortical bone defects in fluoridated Newburgh after ten years of fluoridation at 1 ppm (compared to children in unfluoridated Kingston). This was the first health study conducted in the US on artifically fluoridated water. According to a reviewer for a National Academy of Sciences (NAS) report published in 1977:

"Caffey (1955) noted that the age, sex and anatromical distribution of the bone defects are 'strikingly' similar to that of osteogenic sarcoma. While progression of cortical defects to malignancies has not been observed clinically, it would be important to have direct evidence that osteogenic sarcoma rates in males under 30 have not increased with fluoridation" (NAS, "Drinking Water and Health", 1977, p. 388-9).

This comment is serious on two fronts. First, it confirms that bone damage was observed in young children as a result of drinking fluoridated water for 10 years and second, it raises a red flag of concern (as early as 1955) that fluoride might cause bone cancer in young men. The fact that increases in osteosarcoma has been subsequently found in both male rats treated with fluoride (NTP, 1990) and higher rates of osteosarcoma have been found in young men living in fluoridated communities in at least two epidemiological surveys (SEER, 1991; Cohn, 1992), should not have been so cavalierly dismissed as it has been by American and other authorities (Hoover, 1990, 1991, NRC 1993), and in a second-hand fashion in this Forum report.

ii) Alarcon-Herrera et al (2001).

This study has already been referred to above in the discussion of dental fluorosis. Alarcon Herera et al (2001) found that the incidence of bone fracture in children in an area in Mexico (which had naturally high levels of fluoride in the water) increased in a linear fashion with the severity of dental fluorosis. It is well established that dental fluorosis is a bio-marker for fluoride exposure.

f) No monitoring of fluoride levels in bone.

As we have indicated above, fluoride may be damaging our bones at both the beginning and the end of our lives. What is particularly disturbing about the way governments of fluoridated countries have handled this issue, is that, despite the fact that they have known for many years that approximately 50% of the fluoride we ingest each day accumulates in our bones, there has been no systematic attempt in Ireland, or elsewhere, to track the level of fluoride in our bones as a function of age, geography, diet, health status and fluoridated water consumption. After nearly 40 years of fluoridation in Ireland, there should be a wealth of data; as the most recent international reviewers (York Review, 2000 and MRC, 2002) have indicated, there is none. While the Forum has recommended yet more studies on teeth, it has failed to recommend the collection of the most obvious and most basic data one would need to investigate the serious end point of bone damage. Why?

4.11 Failure to take into account total dose.

Another key weakness in the Forum report is the authors' failure to address the total dose of fluoride from all sources (See Stannard et al, 1991; Kritsky et al 1996 ; Turner et al 1998 ; Heilman et al 1999 ; Fein & Cerklewski 2001; Warnakulasuriya et al 2002.) . At times one gets the distinct impression that they see their task as exonerating water fluoridation of any harm, even though this means pointing the finger of blame at other sources of fluoride like toothpaste. For the average citizen, exactly which sources of fluoride is causing a problem is less relevant than the fact that many of our children are being over-exposed to fluoride from all sources combined. It is inexplicable that the Forum authors have not provided estimates of the total fluoride exposure to fluoride for both children and adults.

4.12 Failure to use weight of evidence approach.

In the section on bone strength (see 4.10) we stressed the failure of the Forum to combine the evidence that can be gleaned from animal, clinical and epidemiological studies. A similar failure to use "a weight of evidence" approach is revealed in many of the "Reviews" they have summarized. Two other examples are a repeated failure by government agencies to take a weight of evidence approach to a possible relationship between fluoride exposure and osteosarcoma in young men (see discussion in 4.10), and the impact of fluoride, in conjunction with aluminum and other ions, on the central nervous system (see Varner's work discussed in 4.9).

4.13 Failure to discuss the Precautionary Principle.

A definition of the Precautionary Principle was recently crafted by a group of scientists meeting in Racine, Wisconsin. They stated it this way: "When an activity raises threats of harm to the environment or human health, precautionary measures should be taken even if some cause and effect relationships are not fully established scientifically". Put more simply, it states, "If in doubt, leave it out."

If ever a policy should be forced to satisfy the precautionary principle it should be fluoridation, since this is the only time in human history (apart from a short experimentation with iodide) the public water supply has been used to deliver medication. If ever a policy screamed out for caution , this is it. What is being delivered to the whole population is a substance known to be highly toxic at moderate doses. The delivery system is incapable of monitoring individual response and doses cannot be controlled.  Furthermore, the individual's normal right to "informed consent" to medication is being over-ridden.

The necessity for this precautionary principle has emerged because in the past it has been very difficult to prove convincingly that a persistent chemical has caused harm to workers or citizens. This is because by their very nature it is extremely difficult in epidemiological studies to control for all the complex and confounding variables in society. As some scientists have jokingly observed, "An epidemic is a health problem that even an epidemiologist can spot."

With the wisdom of hindsight, scientists have realized that by the time scientific proof has been obtained, which is robust enough to resist the most entrenched invested interest, it is too late for thousands or even millions of people who have meanwhile been damaged by exposure to the chemical of concern. The precautionary principle is a principle designed to help officials protect the public from this kind of damage.

In our view, there are five questions which should help to avoid exposing people to unnecessary risks from chemicals for which the toxicological and epidemiological data base is incomplete, as is the case with fluoride and the other chemicals like hexafluorosilicic acid used to fluoridate the public water supply. They are:

First question. Is the public being exposed to a chemical for which there is plausible evidence of harm?

Second question. How serious is this harm if it is found that indeed this chemical causes it?

Third question. How significant is the benefit being pursued?

Fourth question. Are there alternative approaches to pursuing this benefit?

Fifth question. Have all the people being exposed to these risks agreed to the exposure?

In our view, water fluoridation fails on all five questions.

Unfortunately, the Fluoridation Forum authors seem totally oblivious to any notion of a precautionary principle approach. Essentially, they are insisting that the practice of water fluoridation should continue until there is absolute proof of harm.

4.14 Failure to address Paul Connett's "50 Reasons".

In October of 2000, the Forum was presented with "50 Reasons to Oppose Fluoridation" by Dr. Paul Connett, professor of chemistry at St. Lawrence University. At the time, the Forum stated that it would respond to the 50 reasons, and indeed, soon set up a sub-committee to do so. For approximately 10 months, the Fluoridation Forum website reported on several updates of the progress this subcommittee was making in responding to this list. In September 2001, however, it was announced on the website that the subcommittee did not have the time to complete the task -- this despite the fact that they had almost a year to do so.

The only apparent reference to this matter in the Forum report comes in one line in the section labeled "Presentations and Submissions" in which the authors write: "One presenter requested a response to his submission and the response of the Forum to this request will be presented on the Forum website. The final Forum Report has taken account of the issues raised in the submission." (p23)

If readers check the "50 Reasons" (available at http://www.fluoridealert.org/50Reasons.htm or in appendix 2 of this report), they will find that very few of the concerns are addressed in the Forum's report, and none adequately.

4.15 The use of hexafluorosilicic acid instead of sodium fluoride (Chapter 10).

In Chapter 10, the Forum authors claim that the reasons for the switch from sodium fluoride ( toxicologically tested) to hexafluorosilicic acid (not toxicologically tested) was made because,

"The sodium fluoride was very hygroscopic (water-absorbent) and as water treatment plants are by nature damp places there was a tendency for the powder to become solid, resulting in major difficulties measuring accurate amounts to add to the water. The dust from the powder was a serious health and safety threat to water plant workers."

According to Myron Coplan, an engineer with first-hand knowledge of fluoridating chemicals,

"This is an invalid, non-credible, specious argument. The dry powder, as delivered in bags, can easily be dissolved in plant water to a standard concentration (eg saturation), and stored in corrosion-resistant polyethylene tanks indefinitely without deterioration and used at any time, when called for.

"Such a solution, at known concentration, can easily be metered into the main water flow in a manner similar to the way hexafluorosilicic acid (HFSA) is metered into the main water flow. As a matter of fact, however, the saturated NaF solution would be far less hazardous to handle in the metering process. Spills and pump leaks, etc., would be easily collected and washed away without release of fluoride gases. The NaF solution itself would be far less corrosive at any concentration compared to the original HFSA and any dilution thereof. A saturated solution of NaF (4.2%) has a pH of 7.4, very slightly alkaline which can be handled easily in commonplace inexpensive available equipment such as pumps, valves and piping.

"Moreover, ANY prepared concentration of NaF could also be known precisely. In fact, the standard for calibrating the fluoride specific ion electrode used for quality control and other laboratory purposes is a weighed out amount of NaF added to a known volume of water. This is to be compared with the relatively imprecisely known HFSA concentration designated as "Not Less Than XX percent" of some total solids of some imprecisely known species of fluoride-bearing compounds. It is not and never could be used to prepare a standard solution of known fluoride concentration for laboratory test purposes.

"In short, the Fluoridation Forum Report in dealing with the fluoridating agent (HFSA) starts out with an unequivocal fabrication regarding an easily demonstrated fact. Was this the result of sheer ignorance by the authors of the Report? Were they deliberately misled by other so called "experts" intent on hiding the real motive behind disposing of fluosilicic acid in public water supplies, not only in Ireland, but also in Canada, Australia, the UK and US? In any case, given the fact that the entire validation of water fluoridation embodied in the Forum Report starts with this obfuscated, yet easily refuted basis for using HFSA, how can the rest of it be reliable?"

Index | S1 | S2 | S3 | S4 | S5 | S6 | S7