For decades, a persistent worry has trailed the fluoride added to American tap water: What if it quietly damages children’s brains? A study spanning more than 50 years of data now offers one of the most direct answers yet. Researchers using the Wisconsin Longitudinal Study, a long-running research program at the University of Wisconsin-Madison, tracked participants from high school through older adulthood and found no link between community water fluoridation at standard U.S. levels and reduced IQ or diminished cognitive performance at any stage of life.
The findings, published in the Proceedings of the National Academy of Sciences, arrive during a period of intensified debate over fluoridation policy. A federal judge ruled in September 2024 under the Toxic Substances Control Act that fluoride at current recommended levels poses an “unreasonable risk” of harm to children’s neurodevelopment, a decision that energized calls to end the practice. As of spring 2025, the new leadership at the Department of Health and Human Services has signaled openness to revisiting fluoridation guidelines. Against that backdrop, the Wisconsin study adds a substantial data point to a question that has become as much political as scientific.
What the research actually shows
The Wisconsin Longitudinal Study has followed more than 10,000 people who graduated from Wisconsin high schools in 1957, collecting data on education, health, and cognition across multiple waves of interviews and testing. For the fluoride analysis, researchers compared participants’ IQ scores in adolescence and cognitive performance in later decades against their estimated lifetime exposure to fluoridated municipal water. The result: no measurable association between fluoride exposure and lower IQ, either in youth or in old age.
This is not an isolated finding. A separate U.S. life-course analysis by Warren and colleagues, published in Science Advances, drew on a different nationally representative cohort and reached the same conclusion. That study found no evidence that fluoride at typical American concentrations reduces cognitive or academic performance. Some of its results actually showed modestly higher adolescent test scores among those with fluoride exposure, though the authors cautioned that small advantages likely reflected confounding factors like household income or school quality rather than a direct benefit of fluoride.
International research reinforces the pattern. New Zealand’s Dunedin Multidisciplinary Health and Development Study, one of the world’s most cited longitudinal cohorts, compared IQ outcomes for people raised in fluoridated communities against those in non-fluoridated areas and found no meaningful difference. In Australia, a population-based longitudinal study used WAIS-IV IQ testing in young adults, combined with estimated early-life fluoride exposure and dental fluorosis as a biomarker, and reported no evidence of harm to brain development. Three countries, multiple independent cohorts, and the same conclusion: at the fluoride concentrations used in municipal water systems, there is no detectable IQ penalty.
Where the science gets more complicated
The picture changes at fluoride levels well above what American water systems deliver. The National Toxicology Program, part of the National Institutes of Health, completed a major synthesis on fluoride and neurodevelopment in 2024. The NTP concluded that evidence was insufficient to determine effects at the U.S. recommended concentration of 0.7 milligrams per liter. At higher exposures, however, the agency found more consistent associations with lower IQ in children.
Many of those higher-exposure studies come from regions in China, India, Iran, and Mexico where naturally occurring fluoride in groundwater reaches concentrations several times above what any U.S. municipality adds intentionally. Because those populations face different dietary patterns, environmental exposures, and socioeconomic conditions, translating their results directly to American fluoridation policy is not straightforward. The NTP report itself drew this distinction, and misreading it as a blanket condemnation of U.S. tap water overstates what the agency actually concluded.
The September 2024 federal court ruling added a new wrinkle. Judge Edward Chen, citing the NTP report and a Canadian study known as MIREC, found that fluoride at 0.7 mg/L presents an unreasonable risk under TSCA. That legal finding carries regulatory weight, but it is important to note that the legal standard for “unreasonable risk” under TSCA is not identical to the scientific standard for established causation. The ruling has been appealed, and the scientific community remains divided on whether the studies the judge relied upon adequately control for confounders at low exposure levels.
A separate gap involves subgroups who may face higher-than-average total fluoride intake even within the United States. Households relying on private wells with naturally elevated fluoride, or individuals consuming large quantities of fluoride-containing dental products and tea, could reach total exposures closer to those seen in the international studies flagged by the NTP. Current large-scale cohort research is not always designed to isolate these niche exposure patterns, so the evidence base is thinner for such scenarios than for typical city tap water.
Weighing the evidence clearly
The PNAS study and the Warren analysis in Science Advances are both primary research papers that tracked real individuals over time, measured cognitive performance with standardized instruments, and estimated fluoride exposure from municipal water records. That longitudinal design gives them strong internal validity for answering a specific question: does U.S.-level fluoridation harm IQ?
The NTP monograph serves a different purpose. It aggregates dozens of studies from around the world, many conducted at fluoride levels several times higher than the American standard. Its value lies in identifying a dose-dependent signal at elevated exposures. But its explicit qualifier about insufficient evidence at 0.7 mg/L means it cannot fairly be cited as proof that American tap water damages brains.
When multiple independent research teams working with different populations on different continents converge on the same null finding, the collective weight of that evidence is substantially stronger than any single paper. The Australian, New Zealand, and U.S. studies all used established birth or life-course cohorts and validated IQ instruments. Their agreement supports a straightforward reading: within the exposure range typical of municipal water systems, any neurodevelopmental effect of fluoride is either absent or too small to detect with current methods.
What this means for the roughly 200 million Americans on fluoridated water
About two-thirds of the U.S. population receives fluoridated tap water, according to CDC data. For those households, the best available longitudinal evidence from the United States, corroborated by comparable findings from New Zealand and Australia and framed by the NTP’s dose-focused review, indicates that drinking water fluoridated at 0.7 mg/L does not measurably lower IQ or impair cognitive function.
At the same time, the studies documenting potential harm at much higher concentrations justify continued monitoring of total fluoride exposure, particularly in regions with naturally elevated groundwater levels or in communities where multiple fluoride sources may combine. The ongoing legal and political battles over fluoridation are unlikely to be settled by any single study, but the Wisconsin findings add durable, decades-long evidence to a body of research that has consistently failed to find the cognitive damage that opponents of fluoridation describe.
Separating questions about extreme exposures from the realities of everyday tap water remains essential as this debate plays out in courtrooms, city councils, and social media feeds. On the narrow question of whether standard community water fluoridation harms intelligence, the research record through early 2025 points in a consistent direction: it does not.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
