Skip to main content
NCBI home page
The BMJ logoLink to The BMJ
. 2001 Jun 16;322(7300):1486.

Water fluoridation

Meta-analysis of fluoridation and fractures has been done

Graeme Jones 1
PMCID: PMC1120532  PMID: 11430366

Editor—The article by McDonagh et al is valuable but not necessarily new in the area of fractures.1 They conclude that the evidence relating fluoridation to fractures is weak and shows no significant effect either way. This is identical to the conclusion that we published in a 1999 meta-analysis not cited by McDonagh et al but listed on Medline.2 This study located 26 studies in this area, compared with their 20, and reported a relative risk for fracture of 1.02 (95% confidence interval 0.96 to 1.09). Interestingly, sex, urbanicity index, and the quality (but not duration) of the study explained 25% of the variation between studies in meta-regression.

Further work in this area will need to be done at an individual level to advance this field, but even this is problematic given the paper by Phipps et al in the same issue of the journal, which reports on a very large cohort of women.3 Nevertheless, it seems unlikely that water fluoridation to 1 ppm has any significant effect on the incidence of fractures at a population level based on the current (albeit weak) evidence.

Footnotes

Competing interests: None declared.

References

  • 1.McDonagh MS, Whiting PF, Wilson PM, Sutton AJ, Chestnutt I, Cooper J, et al. Systematic review of water fluoridation. BMJ. 2000;321:855–859. doi: 10.1136/bmj.321.7265.855. . (7 October.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Jones G, Riley M, Couper D, Dwyer T. Water fluoridation, bone mass and hip fracture: a quantitative overview of the literature. Aust N Z J Public Health. 1999;23:34–40. doi: 10.1111/j.1467-842x.1999.tb01202.x. [DOI] [PubMed] [Google Scholar]
  • 3.Phipps KR, Orwoll ES, Mason JD, Cauley JA. Community water fluoridation, bone mineral density, and fractures: prospective study of effects in older women. BMJ. 2000;321:860–864. doi: 10.1136/bmj.321.7265.860. . (7 October.) [DOI] [PMC free article] [PubMed] [Google Scholar]
BMJ. 2001 Jun 16;322(7300):1486.

Critical difference was overlooked

Paul Connett 1

Editor—I wish to comment on the paper by McDonagh et al on fluoridation and the study by Phipps et al on bone fracture.1-1,1-2 Although the paper by McDonagh et al deals a very serious blow to fluoridation on the basis of their findings of a much lower benefit for caries reduction than previously claimed and a much higher level of dental fluorosis than is acceptable, the authors do not make it clear that dental fluorosis is an indication of a toxic effect of fluoride, in a similar way that the blue line on gums is an indicator of lead poisoning. This raises the question of what other enzymes and proteins might be poisoned. The paper by McDonagh et al is only a partial scientific analysis. They did not consider work on the pineal or thyroid gland or many animal experiments showing an effect of fluoride on the brain.

On the basis of their meta-analysis, McDonagh et al claim that there is no evidence of a relation between hip fracture in elderly people and fluoridation but mischaracterise some of the key studies. For example, they indicated that Li et al showed no association.1-3 Li et al showed, however, that at above 1.5 ppm hip fracture rates were doubled and above 4 ppm they were tripled. Although I pointed this out to McDonagh et al twice, they still report no association.

The most disturbing aspect of the study by Phipps et al is how much attention is given to the decrease in hip fracture incidence and how little to the increase in wrist fracture. The ostensible reason for this is that the 32% increase in wrist fracture is deemed insignificant. However, the supposed insignificance is based upon the fact that the 95% confidence interval is given as 1.00 to 1.71. Had the interval been 1.01 to 1.71, the increase would have been deemed significant. Bearing in mind that this result is achieved after many adjustments, the claim that long term exposure to fluoridated drinking water does not increase the risk of fracture is cavalier. It is also intriguing that McDonagh et al reported the study by Phipps et al about wrist fractures as significant with a confidence interval of 1.02 to 1.71.

Overall, it is unfortunate that McDonagh et al have managed to convey the message that there is no evidence of harm from fluoridation, when in fact there is mixed evidence of harm—a critical difference.

Footnotes

Competing interests: PC is a founding member of the Fluoride Action Network, an international coalition of citizens and scientists seeking an end to fluoridation worldwide. With his son he runs the web page www.fluoridealert.org.

References

  • 1-1.McDonagh MS, Whiting PF, Wilson PM, Sutton AJ, Chestnutt I, Cooper J, et al. Systematic review of water fluoridation. BMJ. 2000;321:855–859. doi: 10.1136/bmj.321.7265.855. . (7 October.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 1-2.Phipps KR, Orwoll ES, Mason JD, Cauley JA. Community water fluoridation, bone mineral density, and fractures: prospective study of effects in older women. BMJ. 2000;321:860–864. doi: 10.1136/bmj.321.7265.860. . (7 October.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 1-3.Li Y, Liang C, Siemenda C, Ji R, Sun S, Gao J, et al. Effect of long term exposure to fluoridation in drinking water on risks of bone fractures. 1999. (Unpublished report, submitted by author.)
BMJ. 2001 Jun 16;322(7300):1486.

Clearer evidence of benefits and risks is needed

Sheila L M Gibson 1, Robin G Gibson 1

Editor—The systematic review by McDonagh et al indicates that the benefits and risks of water fluoridation balance each other in that a median of six people would need to receive fluoridated water for one to benefit while six people would have to be exposed to it for one additional person to develop dental fluorosis.2-1 It also shows that in artificially fluoridated areas on average 48% of the population has this condition. Since dental fluorosis, whether of aesthetic concern or not, is the first visible sign of poisoning by fluoride, which is as toxic as arsenic and lead, this shows that about half the population in these areas is suffering from some degree of harm. This suggests that water fluoridation is an unacceptable method of reducing dental decay.

The benefits may be overestimated whereas the risks may be grossly underestimated. A confounding factor that was repeatedly pointed out to the review panel but not taken into account is that fluoride delays tooth eruption. In comparing children of similar age, this biases the results in favour of fluoride.2-2

The review by McDonagh et al does point out the possibility of adverse effects of fluoride on the thyroid gland. Fluorine displaces iodine from its compounds and may be at least one of the factors, if not the major one, in causing the delay in tooth eruption. It may also be a factor in the increased neonatal death rate described by Schatz in a fluoridated area of Chile,2-3 a paper that was submitted to the review but does not seem to have been commented on. Birmingham and the West Midlands, fluoridated since the early 1960s, have one of the highest neonatal death rates in Britain,2-4 and an exceptionally high rate of diabetes.2-5

Since the established view is that water fluoridation is perfectly safe, little research into possible adverse effects has been carried out on human populations drinking artificially fluoridated water, and correlations are not made between increasing prevalences of illness and water fluoridation. There is, however, a large body of experimental and animal evidence, studies that fell outside the remit of the review by McDonagh et al, which attest to the toxic effects of fluoride on the stomach, kidneys, pancreas, thyroid, brain and immune system, as well as on bones and teeth. Since animal studies are the gold standard of drug safety testing, this evidence should also be considered before mass medicating any further sections of the community. If the York review shows anything, it is that there should be a moratorium on all current fluoridation schemes until clearer evidence of benefits and risks becomes available.

Footnotes

Competing interests: SLMG is medical adviser to the National Pure Water Association and member of an advisory panel to the systematic review of water fluoridation. RGG—none declared.

References

  • 2-1.McDonagh MS, Whiting PF, Wilson PM, Sutton AJ, Chestnutt I, Cooper J, et al. Systematic review of water fluoridation. BMJ. 2000;321:855–859. doi: 10.1136/bmj.321.7265.855. . (7 October.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2-2.Gibson SLM, Gibson RG. Water fluoridation and tooth decay in 5 year olds. BMJ. 1998;316:231. [PMC free article] [PubMed] [Google Scholar]
  • 2-3.Schatz A. Increased death rates in Chile associated with artificial fluoridation of drinking water, with implications for other countries. Journal of the Arts, Science and Humanities. 1976;2:1–17. [Google Scholar]
  • 2-4.Settatree R, Wyldes M, Tonks A. Stillbirth and neonatal death 1991-1994. Report of national, regional, district and unit mortality rates. Solihull: West Midlands Perinatal Audit; 1996. [Google Scholar]
  • 2-5.Regional Director of Public Health. Partners in health. Birmingham: West Midlands Regional Health Authority; 1992. pp. 21–35. . (Public health report.) [Google Scholar]
BMJ. 2001 Jun 16;322(7300):1486.

Government should meet commitment made in white paper

P J Holloway 1

Editor—The systematic review by McDonagh et al confirmed that fluoridation of the water supplies is effective in reducing dental caries and has no detrimental effects on public health.3-1 It is now time for the government to meet its commitment made in the white paper Saving Lives: Our Healthier Nation, published in July 2000, to require water companies to fluoridate where there is strong local support for doing so.

The method used, however, limits the information that should be considered by health and local authorities when deciding whether or not to implement this important pubic health measure. The stringent criteria used to measure effectiveness failed to evaluate the full benefits. The benefits of fluoridation are cumulative over a lifetime for those who consume fluoridated water over that period.3-2 To ignore the evidence of value to the adult population underestimates the true impact of the public health measure.

The only adverse effect that the review discovered was that of cosmetically unacceptable fluorosis. The estimates of the prevalences of cosmetically unacceptable fluorosis, while based on the international literature, appear too high for the situation in the United Kingdom. For example, the estimate of 10% of the population in fluoridated communities suffering from cosmetically unacceptable fluorosis runs against best current evidence for this country, which suggests that this affects only 3%.3-3

Those who will be called on to make decisions on this very important matter deserve to have the full picture, particularly as it is relevant to this country, on which to decide.

Footnotes

Competing interests: None declared.

References

  • 3-1.McDonagh MS, Whiting PF, Wilson PM, Sutton AJ, Chestnutt I, Cooper J, et al. Systematic review of water fluoridation. BMJ. 2000;321:855–859. doi: 10.1136/bmj.321.7265.855. . (7 October.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3-2.Thomas FD, Kassab JY. Fluoridation in Anglesey: a clinical study of dental caries in mothers at term. Brit Dent J. 1992;181:36–140. doi: 10.1038/sj.bdj.4807968. [DOI] [PubMed] [Google Scholar]
  • 3-3.Tabari ED, Ellwood R, Rugg-Gunn AJ, Evans DJ, Davies RM. Dental fluorosis in permanent incisor teeth in relation to water fluoridation, social deprivation and toothpaste use in infancy. Brit Dent J. 2000;189:216–220. doi: 10.1038/sj.bdj.4800726. [DOI] [PubMed] [Google Scholar]
BMJ. 2001 Jun 16;322(7300):1486.

Authors' reply

Marian S McDonagh 1,2, Jos Kleijnen 1,2

Editor—Our study was a review of primary studies; hence the meta-analysis by Jones et al was not cited. We were aware of their work, however, and did screen the six studies included in their analysis and not in ours. We excluded these studies because they did not meet our inclusion criteria. Both Connett and Gibson comment on our lack of describing fluorosis as an indication of a toxic effect of fluoride. In fact in our analysis, fluorosis is considered an individual negative outcome, as is, for example, bone fracture or cancer. Regarding Connett's comment about the lack of consideration of thyroid and pineal studies, we emphasise that any study meeting inclusion criteria was included (as were two thyroid studies). Animal studies were excluded because they provide less reliable estimates of effect and where human data are available we think these should not bear significant weight on decisions about interventions. Connett also indicates that we have misrepresented a study by Li et al.4-1 Our analysis of bone fracture relates to water fluoridation levels close to 1 ppm (the relevant concentration for artificial fluoridation). At 1 ppm, the risk of fracture found by Li et al was not increased.

We agree with Holloway that the effect of fluoridation should be examined over time, which is one reason we excluded cross sectional studies of positive effects. We did not exclude studies of adults, but only one met the inclusion criteria. Holloway further states that our estimates of fluorosis seem too high for the United Kingdom. The advantage of a systematic review is to bring together as much evidence as possible to get a clearer picture of the true effect. Rather than depend on a single study, we have more confidence in the pooled studies. In the meta-regression analysis, geographical location was not found to be significantly associated with fluorosis; neither were other geographically specific variables such as temperature and altitude.

Gibson, one of our advisory panel members, comments that delayed eruption of teeth potentially caused by fluoridation was ignored. This issue was not ignored in our review, but we found only one study that met inclusion criteria. This is an area that may merit further research. Counter to what Gibson writes, the Schatz study of neonatal death was indeed included in our review, and was commented on in our full report.4-2 Our review of water fluoridation is specific and appropriately narrow in focus. It is not the answer to the question of whether or not to fluoridate. Our review found a benefit in the form of reduction in caries, balanced against a dose related increase in fluorosis. We said that no clear evidence of other potential negative effects was found. We have been assiduous in our paper, our full report, and our contacts with the media, not to convey a message of no evidence of harm. We suggest caution against overinterpretation of our results and emphasise again that the quality of these data on benefit and harm is only low to moderate.

Footnotes

Competing interests: None declared.

References

  • 4-1.Li Y, Liang C, Siemenda C, Ji R, Sun S, Gao J, et al. Effect of long term exposure to fluoridation in drinking water on risks of bone fractures. 1999. (Unpublished report, submitted by author.)
  • 4-2.NHS Centre for Reviews and Dissemination. A systematic review of public water fluoridation. York: University of York; 2000. . (Report 18.) [Google Scholar]
BMJ. 2001 Jun 16;322(7300):1486.

Results should be viewed with concern rather than applause

T C Schmidt 1

Editor—Phipps et al state that a 95% confidence interval on relative risk for wrist fracture of 1.00-1.71 due to community water fluoridation is not significant and surmise that it may be one of the most cost effective methods for reducing the incidence of fractures related to osteoporosis, on the basis of reduced spine and hip fracture after adjustment for a dozen factors in addition to age.5-1 These are specific to bone mass density, and the changes in bone mass density in table 3 (due to and correlated with fluoride exposure) are congruent with changes associated with fluoride treatment.5-2–5-4 In the 1994 guidelines of the Food and Drugs Administration (published in 1997) for preclinical and clinical evaluation of agents used in preventing or treating postmenopausal osteoporosis, the relation between increased bone mass density and reduced fracture risk has been validated for patients receiving oestrogens, but not fluoride. It is now recommended that clinical trials be restricted to the axial skeleton only, provided that the patient has good peripheral bone density, renal function, and vitamin D status, factors not considered in the multivariate analysis of Phipps et al. Thus, the analysis is not entirely cogent regarding fracture. Rather, the two most significant trends for age adjusted only with continuous exposure in figure 5 are again increased wrist fracture (P=0.012) and maybe some decrease in spine fracture (P=0.079).

The benefit (if any) to the vertebra from increased bone mass is now recognised to be at the expense of increased risk of fracture to the parts of the skeleton that are subject to forces other than compressive loading. The new bone formed is weak and structurally abnormal because of fluoride's alteration of the normal remodelling process (FDA Consumer, April 1991), and this decrease in quality and strength presents both clinically and experimentally (with and without calcium supplementation). This osteogenic effect occurs at plasma fluoride concentrations associated with fluoridated areas, with total accumulation being directly proportional to total lifetime ingestion (unless exacerbated by the onset of chronic renal insufficiency). The net dosage increment associated with the age and weight adjusted increase in five year relative fracture risk comparing 1 mg/l with 4 mg/l fluoride in drinking wateris commensurate with 20 years of community water fluoridation and clinical experience has taught that denser bones are not necessarily better bones.5-4,5-5 Thus the 20 year changes in bone mass density reported by Phipps et al should be viewed with some concern rather than applause.

Footnotes

Competing interests: None declared. TCS agrees with Douglas Carnall (BMJ 2000;321:904) that those who wish to take fluoride should get it from toothpaste rather than the water supply.

References

  • 5-1.Phipps KR, Orwoll ES, Mason JD, Cauley JA. Community water fluoridation, bone mineral density, and fractures: prospective study of effects in older women. BMJ. 2000;321:860–864. doi: 10.1136/bmj.321.7265.860. . (7 October.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5-2.Bauer DC, Browner WS, Cauley JA, Orwoll ES, Scott JC, Black DM, et al. Factors associated with appendicular bone mass in older women. The Study of Osteoporotic Fractures Research Group. Ann Intern Med. 1993;118:657–665. doi: 10.7326/0003-4819-118-9-199305010-00001. [DOI] [PubMed] [Google Scholar]
  • 5-3.Riggs BL, Hodgson SF, O'Fallon WM, Chao EY, Wahner HW, Muhs JM, et al. Effect of fluoride treatment on the fracture rate in postmenopausal women with osteoporosis. N Engl J Med. 1990;322:802–809. doi: 10.1056/NEJM199003223221203. [DOI] [PubMed] [Google Scholar]
  • 5-4.Dequeker J, Declerck K. Fluor(ide)in the treatment of osteoporosis. An overview of thirty years clinical research. Schweiz Med Wochenschr. 1993;123:2228–2234. [PubMed] [Google Scholar]
  • 5-5.Sowers MF, Clark MK, Jannausch ML, Wallace RB. A prospective study of bone mineral content and fracture in communities with different fluoride exposure. Am J Epidemiol. 1991;133:649–660. doi: 10.1093/oxfordjournals.aje.a115940. [DOI] [PubMed] [Google Scholar]
BMJ. 2001 Jun 16;322(7300):1486.

Authors' reply

Kathy Phipps 1,2, Eric Orwoll 1,2, Jill Mason 1,2, Jane Cauley 1,2

Editor—Connett voiced his concern about our interpretation of fracture risk associated with long term exposure to community water fluoridation. After adjusting for known confounders, we found a slight, yet significant reduction in the risk of fracture of the hip and spine and a trend towards an increased risk of fracture of the wrist. Following standard statistical convention, we reported that the increase in the risk of wrist fracture was not significant because the 95% confidence interval included 1.0.

The decrease in the risk of fractures in both spine and hip outweighs the trend towards increased wrist fracture. Fractures of the hip account for the major burden of osteoporosis and can result in admission to nursing homes and death. Wrist fractures, however, tend to occur in healthy more active women and do not have the economic or social burden associated with hip fractures.6-1 Our conclusion that long term exposure to fluoridated drinking water does not increase the risk of fracture is also based on the fact that we saw no association when we evaluated all non-vertebral fractures (odds ratio 0.96, 95% confidence interval 0.83 to 1.10). Connett also refers to the fact that the fluoridation review of the University of York quotes our confidence interval for wrist fractures as 1.02 to 1.71. For their review, the York team used our draft results that were not fully adjusted for the known risk factors for fracture.

Schmidt focuses on the changes we found in bone mineral density and the association between bone mass density and fracture risk. He says that denser bones are not necessarily better bones. We disagree with this statement and cite the alendronate and raloxifene clinical trials as prime examples. These trials found that individuals taking raloxifene and alendronate had increased spinal bone mass density along with a decrease in vertebral fracture risk.6-2,6-3 In addition, trials testing time released or low dose fluoride for the treatment of osteoporosis also found both increased spinal bone mass density and fewer vertebral fractures.6-4,6-5 On the basis of the increasing literature documenting a significant correlation between increased bone mass density and reduced fracture risk, we stand by our claim that community water fluoridation does not increase the risk of osteoporotic fractures in older women and may actually slightly reduce the risk of hip and vertebral fractures.

Footnotes

Competing interests: None declared.

References

  • 6-1.Kelsey JL, Browner WS, Seeley DG, Nevitt MC, Cummings SR. Risk factors for fractures of the distal forearm and proximal humerus. The Study of Osteoporotic Fractures Research Group. Am J Epidemiol. 1992;135:477–489. doi: 10.1093/oxfordjournals.aje.a116314. [DOI] [PubMed] [Google Scholar]
  • 6-2.Hochberg MC, Ross PD, Black D, Cummings SR, Genant HK, Nevitt MC, et al. Larger increases in bone mineral density during alendronate therapy are associated with a lower risk of new vertebral fractures in women with postmenopausal osteoporosis. Fracture Intervention Trial Research Group. Arthritis Rheum. 1999;42:1246–1254. doi: 10.1002/1529-0131(199906)42:6<1246::AID-ANR22>3.0.CO;2-U. [DOI] [PubMed] [Google Scholar]
  • 6-3.Ettinger B, Black DM, Mitlak BH, Knickerbocker RK, Nickelsen T, Genant HK, et al. Reduction of vertebral fracture risk in postmenopausal women with osteoporosis treated with raloxifene: results from a 3-year randomized clinical trial. Multiple Outcomes of Raloxifene Evaluation (MORE) Investigators. JAMA. 1999;282:637–645. doi: 10.1001/jama.282.7.637. [DOI] [PubMed] [Google Scholar]
  • 6-4.Reginster JY, Meumans L, Zegels B, Rovati LC, Minne HW, Giacovelli G, et al. The effect of sodium monofluorophosphate plus calcium on vertebral fracture rate in postmenopausal women with moderate osteoporosis. A randomized, controlled trial. Ann Intern Med. 1998;129:1–8. doi: 10.7326/0003-4819-129-1-199807010-00001. [DOI] [PubMed] [Google Scholar]
  • 6-5.Pak CY, Adams-Huet B, Sakhaee K, Bell NH, Licata A, Johnston C, et al. Comparison of nonrandomized trials with slow-release sodium fluoride with a randomized placebo-controlled trial in postmenopausal osteoporosis. J Bone Miner Res. 1996;11:160–168. doi: 10.1002/jbmr.5650110204. [DOI] [PubMed] [Google Scholar]
BMJ. 2001 Jun 16;322(7300):1486.

Little is known about the safety of fluoridation

Baldwin of Bewdley 1

Editor—Your comments in “This week in the BMJ” and the editorial by Hausen treated the study by Phipps et al on community water fluoridation, bone mineral density, and fractures, which you published alongside the systematic review of water fluoridation by McDonagh et al, as introducing new evidence that should alleviate remaining concerns about the safety of fluoridation.7-1–7-3 The paper by Phipps et al, which seemed to show a slightly lower risk of fractures, had been fully accounted for in the systematic review by McDonagh et al, where it achieved the lowest rating (high risk of bias), in common with all but one of the studies on bone problems in the world literature. It is therefore premature to suggest that we know all the answers about fluoride and bones, let alone about the safety of fluoridation overall.

Footnotes

Competing interests: None declared. BB holds the honorary position of vice president of the National Pure Water Association

References

  • 7-1.Hausen HW. Fluoridation, fractures, and teeth. BMJ. 2000;321:844–845. doi: 10.1136/bmj.321.7265.844. . (7 October.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7-2.Phipps KR, Orwoll ES, Mason JD, Cauley JA. Community water fluoridation, bone mineral density, and fractures: prospective study of effects in older women. BMJ. 2000;321:860–864. doi: 10.1136/bmj.321.7265.860. . (7 October.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7-3.McDonagh MS, Whiting PF, Wilson PM, Sutton AJ, Chestnutt I, Cooper J, et al. Systematic review of water fluoridation. BMJ. 2000;321:855–859. doi: 10.1136/bmj.321.7265.855. . (7 October.) [DOI] [PMC free article] [PubMed] [Google Scholar]
BMJ. 2001 Jun 16;322(7300):1486.

Ireland has less decay in non-fluoridated communities

Don Mac Auley 1

Editor—I share Carnall's views on fluoridation.8-1 The Republic of Ireland has 73% water fluoridation. So what is its dental experience? In the most recent figures published by the World Health Organization for tooth decay in Ireland, the country lies in only sixth place for tooth quality.8-2 Of the five countries with better teeth, four do not fluoridate and the United Kingdom has only 10% fluoridation. In Ireland there are more decay free teeth in non-fluoridated communities.8-3

Fifty per cent of our population has dental fluorosis. I see patients daily in my surgery who are damaged by fluoride. They do not smile, they are teased at school, and they are traumatised by having “rotten” teeth. Is it an acceptable cost-benefit payoff to cause a dental disease in half of the population to reduce tooth decay by a supposed 15%? The idea of mass medicating an entire population is inherently flawed. When I prescribe drugs I do so with the knowledge of the patient's age, weight, and medical history. Water fluoridation is prescribed by thirst. The more you drink the more you get. Is this science?

References

Articles from BMJ : British Medical Journal are provided here courtesy of BMJ Publishing Group

RESOURCES