Adding fluoride to community drinking water is widely ranked among the top public health achievements of the 20th century, thanks to the protection it provides against tooth decay.
Water fluoridation is also widely considered safe, with little evidence of risk emerging from generations of research and real-world application around the world. At the levels found in most public water supplies, experts generally agree we have nothing to fear from fluoride.
While concerns about water fluoridation typically arise from myths, research does hint at dangers from exceptionally high levels of fluoride well above those found in most community water supplies.
According to a new study, long-term consumption of such water has concerning links with cognitive impairments in children.
This was a relatively small pilot study, and it remains unclear if there is any causal relationship between fluoride exposure and neurotoxicity, says lead author Tewodros Godebo, an environmental geochemist at Tulane University.
But given the stakes, he adds, the issue warrants ongoing scrutiny by scientists.
"Though further epidemiological studies are needed to validate the findings, these results add to the growing concern about the potential neurotoxic effects of fluoride, especially during early brain development and childhood," Godebo says.
"These tests affirmed a clear association between high fluoride and cognitive impairment," he adds.
Godebo and his colleagues conducted the study in rural Ethiopia, where many farming communities rely on well water with varying levels of naturally occurring fluoride, which ranges from 0.4 to 15.5 milligrams per liter, the researchers note.
For context, the World Health Organization (WHO) recommends levels below 1.5 mg of fluoride for every liter of drinking water, while the US Environmental Protection Agency enforces a maximum level of 4.0 mg/L across the country.
The study's authors recruited 74 children between the ages of 5 and 14, used a combination of methods to estimate their cognitive ability, and looked for correlation between their scores and the level of fluoride in their community's drinking water.
One cognitive test relied on figure-drawing criteria, which the researchers note has been used in previous studies to assess child cognition. They asked the children to draw a familiar subject, like a house or a donkey, and scored them based on the inclusion or omission of certain details.
The children also took a computerized memory test, considered language- and culturally-neutral, that's sensitive to spatial learning and memory. The brain's medial temporal lobe is involved with those tasks, the researchers explain, and is thought to be highly affected by fluoride toxicity.
Higher exposure to fluoride in drinking water was associated with more errors on both the drawing and memory tests, the study found, although as the researchers acknowledge, that doesn't necessarily mean the higher exposure is responsible for the lower scores.
Public water fluoridation began in the 1940s in the United States, and has since become a widespread public health measure around the world. But in communities that rely on well water, research suggests variation in naturally occurring fluoride levels may pose a danger.
Well water from areas rich in fluoride-containing minerals often contain up to 10 mg of fluoride per liter, according to the WHO, though in extreme cases natural fluoride levels have been measured in excess of 2,000 mg/L.
Previous research has reported links between excess fluoride consumption and cognitive impairment, the authors of the new study point out, while research on nonhuman animals suggests fluoride can cross the placental and blood-brain barriers.
Millions of people globally may be exposed to high fluoride levels in their drinking water, the researchers add, citing Ethiopia's Rift Valley as an optimal research area for studying the possibility of health effects.
The region is relatively safe from confounding variables, they explain, due to low variability in lifestyle factors and consistent exposure to stable levels of naturally occurring fluoride.
"We have a unique opportunity to study low-fluoride communities in the same setting as high-fluoride communities, so we can determine if fluoride is a neurotoxicant at low levels," Godebo says.
"Such studies are important to the public and government agencies to determine the safety and risk of water fluoridation in drinking water supply systems."
The study was published in Neurotoxicology and Teratology.