Fluoride in Drinking Water: A Scientific Review of EPA's Standards

National Research Council


News Conference
March 22, 2006

Opening Statement
by

John Doull

Professor Emeritus of Pharmacology and Toxicology
University of Kansas Medical Center, Kansas City
and
Chair, Committee on Fluoride in Drinking Water

Good morning. On behalf of my colleagues on the committee, I would like to welcome you to the public release of the National Research Council's new report on fluoride in drinking water. Let me begin with some background on fluoride and the origin of this study, and then I will summarize our key findings and recommendations.

Fluoride is regulated as a drinking-water contaminant by the Environmental Protection Agency. Water can become contaminated with fluoride from natural sources, including runoff and leaching from rocks and soils that contain fluoride. Industrial pollution can also contribute to fluoride levels in water. Because high amounts of fluoride can be toxic, EPA places a cap, or maximum contaminant level, on fluoride concentrations in drinking water in order to prevent the public from being exposed to harmful levels. Our committee was charged with evaluating the adequacy of EPA's guidelines for protecting public health.

Most people associate fluoride in water with fluoridation -- the addition of fluoride into public drinking-water supplies to help prevent tooth decay in communities where natural fluoride levels are very low. Thus, it is important to make the distinction here that EPA's standards are not guidelines for that practice. Water fluoridation guidelines were established by the Public Health Service more than 40 years ago. The levels used for that practice are lower than EPA's drinking water standards, and an evaluation of the safety or efficacy of those lower concentrations were outside the charge to the committee.

EPA's health-based guideline is the maximum contaminant level goal, upon which a standard is based. The goal is a concentration at which no adverse health effects are expected to occur and the margins of safety are judged adequate. For fluoride, the maximum contaminant level goal and corresponding standard are 4 mg/L. EPA also has established a secondary maximum contaminant level, which is a guideline for reducing the occurrence and severity of a cosmetic condition of the teeth called enamel fluorosis. The secondary level is 2 mg/L.

As part of the requirement under the Safe Drinking Water Act to periodically reassess its drinking water guidelines, EPA requested that the National Research Council conduct an independent study on whether the EPA guidelines for fluoride are adequate to protect public health. To address this task, the Committee on Fluoride in Drinking Water was established. We held six meetings between August 2003 and June 2005, two of which included information-gathering sessions that were open to the public. At these sessions, the committee heard presentations and input from federal agencies, scientists conducting fluoride research, citizens' groups, and members of the public. We reviewed a large body of scientific literature on fluoride, particularly new studies conducted within the last decade.

We found that drinking water is the largest contributor to an individual's total exposure to fluoride. When concentrations in water are 4 mg/L, the contribution ranges from 72 percent to 94 percent. Dental products, food, and other sources contribute the balance. Highly exposed groups include people who have high concentrations of fluoride in their drinking water or who drink more water than the average person because of exercise, outdoor work, or a medical condition. Relative to their body weight, infants and young children are exposed to three to four times as much fluoride as adults. Children also may use more toothpaste than is advised or swallow it, and many receive fluoride treatments from their dentists.

After reviewing the collective evidence on adverse health effects associated with fluoride, our committee concluded unanimously that EPA should lower the maximum contaminant level goal for fluoride. Ongoing exposure to water fluoride levels at 4 mg/L puts children at risk of developing severe enamel fluorosis, which is characterized by discoloration, enamel loss, and pitting of the teeth. This condition occurs during tooth development in children, before the teeth erupt in the mouth. Once it develops, it is a permanent condition.

On average, approximately 10 percent of children in communities with water fluoride concentrations at or near 4 mg/L develop severe enamel fluorosis. Previous assessments have considered all cases of enamel fluorosis, including serious ones, to be aesthetically displeasing because of the yellow and brown staining of teeth that occurs, but not adverse to health. However, the majority of the committee concluded that severe cases of enamel loss constitute an adverse health effect because one function of enamel is to protect the teeth and underlying dental tissue. There is some evidence that pitting of tooth enamel could increase the risk of cavities because it allows plaque and bacteria to become trapped in the enamel. The damage to teeth caused by severe enamel fluorosis is a toxic effect that is consistent with prevailing risk assessment definitions of an adverse health effect. However, two of the 12 members of the committee did not agree that enamel defects alone are sufficient to consider severe enamel fluorosis an adverse health effect, as opposed to a cosmetic one. Although these members disagreed over the characterization of severe enamel fluorosis, they concurred with our recommendation that EPA's maximum contaminant level goal should prevent the occurrence of this unwanted condition.

Studies indicate that up to 15 percent of children in communities exposed at the secondary guideline level of 2 mg/L have moderate enamel fluorosis, a less severe form of enamel fluorosis that does not involve enamel loss or pitting. Although this moderate condition can also lead to tooth discoloration that may be aesthetically objectionable, there is inadequate data to categorize it as an adverse health effect.

In addition to effects on teeth, the committee reviewed new studies of fluoride's effects on bone. Fluoride is readily incorporated and accumulates in the bone. Concerns about fluoride's effects on the musculoskeletal system historically have been and continue to be focused on bone fracture and skeletal fluorosis. Studies with laboratory animals show that fluoride incorporation can weaken bone and increase the risk of fracture. Evidence for this effect in humans was found in several new studies of populations exposed to fluoride at 4 mg/L or higher in drinking water, as well as in clinical studies of fluoride as a therapeutic agent. Overall, there was consensus among the committee that under certain conditions fluoride can weaken bone and increase the risk of fractures. The majority of the committee concluded that a population with lifetime exposure to fluoride in water at concentrations of 4 mg/L or higher is likely to experience more fractures than a group exposed to 1 mg/L. Three of the 12 committee members, however, supported a conclusion that EPA's 4 mg/L standard might not be protective against bone fractures. They said more evidence was needed that bone fractures occur at an appreciable frequency in human populations exposed to fluoride at 4 mg/L before drawing a firm conclusion about the risks at that level.

We also looked at the risk of skeletal fluorosis, a bone and joint condition associated with prolonged exposure to fluoride. It involves increases in bone density and effects on the bone that lead to joint stiffness, pain, and, in severe stages, impaired mobility. The committee judged stage II and stage III of the condition, the most severe stages, to be adverse to health. There have been few reported cases of skeletal fluorosis in the past decade, and no studies of the incidence in populations exposed at high concentrations in their drinking water. To determine whether EPA's maximum contaminant level goal protects the public from stage II and stage III skeletal fluorosis, the committee compared modeled predictions of bone fluoride concentrations and historical data on bone fluoride levels associated with different stages of skeletal fluorosis. The committee found that lifetime exposure to fluoride at 4 mg/L can lead to levels in the bone that fall within or exceed the ranges associated with stages II and III of the condition. However, stage III skeletal fluorosis appears to be a rare condition in the United States, and the existing data are insufficient for determining whether stage II is occurring in U.S. residents. Thus, more research on skeletal fluorosis is needed before any conclusions can be drawn.

The evidence to date regarding fluoride's potential to cause cancer, particularly of the bone, is tentative and mixed. Animal studies have suggested the possibility of increased risk of bone cancers in male rats, but no new tests in animals have been performed to adequately evaluate this possibility. New epidemiologic investigations of cancer in relation to fluoride were mixed, with some reporting a positive association and others no association. A study under way at the Harvard School of Dental Medicine will be published this summer and may help identify future research that would be useful for studying fluoride's carcinogenic potential.

In addition to new scientific information now available on fluoride, new risk assessment practices have been developed since EPA established its drinking water guidelines for fluoride in 1986. These include updated approaches to quantifying risk, considering susceptible subpopulations, and characterizing uncertainties and variability. We recommend that EPA take advantage of these new practices to conduct a new health risk assessment of fluoride and determine how much lower the maximum contaminant level goal should be set to protect public health.

My colleagues and I are glad to take your questions now, including those sent by reporters via e-mail. Please remember to identify yourself and your news organization when asking a question. Thank you.