If finalized, the new regulation would lower allowable levels of fluoride in domestically packaged and imported bottled water to 0.7 milligrams per liter, a slight reduction from the current standard of 0.8 milligrams per liter allowed by the FDA. The proposed standard would apply only to bottled water with added fluoride. It would not affect allowable levels of fluoride in bottled water that may contain fluoride from source water. 

The FDA's proposed rule aligns with a 2015 recommendation from the US Public Health Service, part of the US Department of Health and Human Services, that suggests 0.7 milligrams per liter is the optimal fluoride concentration for community water systems that add fluoride. The new rule "is based on findings from evolving research on optimal concentrations of fluoride that balances fluoride's benefits in preventing tooth decay with its risk of causing dental fluorosis, a condition most often characterized by white patches on teeth," the FDA said in its statement. Dental fluorosis is caused by taking in too much fluoride over a long period when adult teeth are forming under the gums. But some scientists' concerns extend far beyond fluorosis. 

"Given that fluoride can damage brain development, I would recommend that the maximum fluoride concentration in bottled water be kept at a lower level than 0.7 mg/L," Dr. Philippe Grandjean, an adjunct professor of environmental health at the Harvard TH Chan School of Public Health, wrote in an email. Christopher Neurath, research director of the American Environmental Health Studies Project, which is connected to the Flouride Action Network, an environmental advocacy group, said "currently, there are rapidly increasing scientific studies showing neurotoxicity to fluoride," with research showing a direct link between children's IQ and their level of fluoride exposure in the womb: "That is our largest concern." 

Morteza Bashash, an assistant professor in the Dalla Lana School of Public Health at the University of Toronto, found higher fluoride levels as measured in urine samples of pregnant women are associated with both lower IQ and increased risk of ADHD among children in Mexico. Specifically, Bashash found a drop in children's' scores on intelligence tests for every 0.5 milligram-per-liter increase in fluoride exposure beyond 0.8 milligrams per liter detected in a pregnant mother's urine. It is not clear whether this is research applicable to the US population.

In Mexico, for example, the government delivers cavity-reducing fluoride by adding it to salt, not water (since many people avoid drinking tap water). Still, his research findings were "based on the true measurement of fluoride absorbed in the body." And a Canadian study presented at a conference last year and studies conducted in China showed IQ losses as related to fluoride levels within a similar order of magnitude. Due to similar fluoride sources, regulations and diet, Canada's findings of urine levels are likely similar to American urine levels, said Bashash. Neurath trusts that both the Mexican and Canadian study results would generally apply to the US since "urine fluoride is best to measure of total fluoride intake." 

Canadian data from the past 15 years has shown that women living in cities with fluoridated water supplies had "almost double" urine fluoride concentrations levels as women living in non-fluorinated cities. "Drinking water fluoride is the major source of fluoride for these women," he said. The effect of prenatal exposure to fluoride on IQ is "very large," Neurath believes. "And on a population basis, that's very concerning."