. "Appendix B Measures of Exposure to Fluoride in the United States: Supplementary Information." Fluoride in Drinking Water: A Scientific Review of EPA's Standards. Washington, DC: The National Academies Press, 2006.
aBased on temperature data obtained for a minimum of 5 years.
bBased on 4.5 times the optimal fluoride level for communities. School water fluoridation is recommended only when the school has its own source of water and is not connected to a community water system. Several other criteria are also considered; for example, if >25% of the children attending the school already receive optimally fluoridated water at home, the school’s water should not be fluoridated.
SOURCE: CDC 1995.
are still served by water systems with fluoride exceeding 4 mg/L (Hirschauer 2004).
Miller-Ihli et al. (2003) reported on fluoride concentrations in water samples collected in 1999 from 24 locations nationwide; these locations were expected to provide nationally representative samples for the National Food and Nutrient Analysis Program.2 Not unexpectedly, their findings indicate a bimodal distribution of fluoride concentrations in public drinking water: either water was fluoridated at approximately 1 mg/L or it was not fluoridated, with concentrations bordering on undetectable.
WATER INGESTION AND FLUORIDE INTAKES
Tables B-4 to B-7 summarize recent estimates by the U.S. Environmental Protection Agency (EPA) of the mean and selected percentiles of water ingestion by source (community supplies, bottled water, “other” sources, and all sources combined) and subpopulation (EPA 2000a); Tables B-8 and B-9
Miller-Ihli et al. (2003) reported that 40% of the samples were fluoridated and suggested that, rather than using an average fluoride concentration for the country, an individual should be assumed to have a 40% probability of ingesting fluoridated water and a 60% probability of ingesting nonfluoridated water. However, CDC (2002a) estimates that about two-thirds of the U.S. population served by public water supplies receives fluoridated water. Thus, the sampling reported by Miller-Ihli et al. was probably not sufficiently representative on a population-weighted basis.