The Quiet Chemistry of Childhood

A new U.S. study made headlines—but the deeper questions remain unanswered

Every few months, a new fluoride study drops—another twist in a debate that has stretched across generations. Some studies praise fluoride for protecting teeth. Others warn about possible harm to the developing brain. And a few arrive with the promise that this one has finally settled the issue.

The latest comes from Robert John Warren and a team of sociologists who analyzed data from the U.S. High School and Beyond cohort. Their question was straightforward: Do children who grow up in fluoridated communities perform differently on cognitive tests in high school or later life?

It’s an attention-grabbing question in a country where two-thirds of the population drinks fluoridated water and where concern about possible impacts on early brain development has intensified over the past decade. Headlines quickly followed: “New study exposes truth about fluoride exposure and kids’ test scores.” “Fluoride does not negatively affect cognitive ability—and may actually provide benefit.”

Warren’s team used a large, nationally representative cohort, tracking people from adolescence into their 60s. They divided people into three groups—those who grew up entirely, partly, or not at all in a fluoridated community—and linked those histories to standardized tests in 12th grade and cognitive assessments decades later.

Their result: people who grew up in fluoridated communities scored a bit higher—roughly one point on a typical standardized high-school test. By adulthood, the associations were smaller and statistically inconclusive but still positive. They adjusted for residential mobility, socioeconomic factors, and community characteristics.

Their conclusion: “Despite these limitations, our results provide strong evidence that exposure to fluoride—at levels ordinarily seen in the United States…has benefits for adolescent cognition and is, at worst, not harmful for later-life cognitive functioning.”

Before taking that conclusion at face value, it’s useful to look closely at how fluoride exposure was measured.

Why Exposure Matters

In environmental-health research, exposure is the first link in the causal chain. Without knowing who was exposed, how much, and when, it becomes difficult to interpret health effects reliably. Precise exposure data sharpens the dose–response curve; crude exposure data flattens it, blurs it, or even sends it in the wrong direction. No matter how large the study or how sophisticated the statistics, poor exposure measurement weakens every conclusion built on top of it.

The Beyond Study had no personal fluoride measurements—not during pregnancy, infancy, childhood, or adolescence. Exposure was assigned solely based on whether a community fluoridated their water or the fluoride content of ground water when the participants were adolescents—long after the period when the developing brain is most vulnerable to toxic chemicals.

We don’t know how often the participants drank tap water or how much fluoride-containing toothpaste they swallowed. We don’t know whether they received fluoride supplements, introduced in the 1950s. We don’t know how many had private wells. In a representative sample, roughly 25% of households would rely on private wells. And while USGS fluoride maps are useful for describing regional patterns, they are a blunt tool when used to assign exposure to individuals.

We also don’t know what their mothers drank during pregnancy. Another major source of fluoride—often overlooked—is tea. Tea plants hyper-accumulate fluoride in their leaves, with the highest concentrations found in the oldest leaves used for black teas and inexpensive blends. Yet the cohort included no information on tea consumption of study participant’s mothers.

Infant feeding adds another layer. Breastmilk contains almost no fluoride. But formula mixed with fluoridated water can expose infants to fluoride levels up to 70-times higher. The Beyond Study had no data on breastfeeding or formula use—an enormous gap when evaluating early-life exposure to fluoride or arsenic.

In short, the study tells us too little about the fluoride these children received during the years when their brains were wiring, pruning, and developing at astonishing speed.

Why the Beyond Study Raises Questions, Not Answers

The Beyond Study is inventive and thought-provoking. But its exposure metric is simply too crude to answer the question it raises. At best, it offers an interesting signal—not a conclusion.

To understand fluoride’s effects, we need studies that measure exposure at the right time. And we have some of those.

What a Stronger Exposure-Based Study Shows

Take the Canadian birth cohort study by Rivka Green, then a doctoral student at York University. Unlike the Beyond Study, this study reconstructed water-fluoride levels and tea consumption across the entire pregnancy and directly measured fluoride exposure using three maternal urinary fluoride samples—one from each trimester. Importantly, the water-fluoride levels in this cohort (0.6 mg/L) mirrors levels common in U.S. fluoridated cities.

Their findings are striking: a 3.7-point decrease in IQ for every 1 mg/L increase in fluoride intake and, among boys, a 4.5-point decrease in IQ per 1 mg/L increase in maternal urinary fluoride. These effects persisted even after adjusting for socioeconomic status, maternal education, ethnicity, prenatal smoking and alcohol use, lead exposure, and other environmental and pregnancy-related factors.

Green’s work echoes well-designed, prospective cohort studies from Mexico City and Bangladesh that also measured urinary fluoride in mothers. These studies likewise found lower IQ scores in children from in-utero exposures at levels found in fluoridated U.S. communities.

Nature Doesn’t Care About Passports

Some have argued that results from Canada, Mexico, or Bangladesh may not apply to the U.S. Yes, biology has modifiers—iodine deficiency, for example, increases susceptibility to fluoride—but pollutants do not spare one nationality and target another. Lead does not weaken Mexicans more than Americans. Arsenic does not discriminate between Bangladeshis and Canadians. Air pollution does not give the British or Japanese a pass. Fluoride has shown no such selectivity.

The developing brain is a universal organ.

Where Do We Go from Here?

David Savitz, in his editorial on the Beyond Study, offered a familiar reassurance: “Until clear evidence exists that water fluoridation lacks public health benefit or compelling evidence of harm… it seems foolhardy to interfere with a long-established public health success.”

It’s a comforting sentence. But it sits uneasily beside what the evidence shows.

A new 2024 review by the Cochrane Oral Health team—one of the most respected groups in evidence appraisal—reached a very different conclusion: studies of fluoridation conducted since 1975 show little, if any, measurable benefit in reducing dental caries. And unlike most population-level interventions we rely on, such as vaccines, no randomized controlled trials show that fluoridation reduces tooth decay.

In other words, the “long-established success” Savitz invokes rests on surprisingly shaky ground.

Meanwhile, the evidence on the risks has grown stronger, not weaker. A landmark meta-analysis and many high-quality studies find consistent links between early-life fluoride exposure and lower IQ in children, even at levels common in U.S. fluoridated cities. These aren’t fringe studies; they’re rigorous, prospective cohort studies using internal dosimeters of fluoride exposure.

A federal court has now weighed in as well. “The Court finds that fluoridation of water at the level presently considered ‘optimal’ in the United States poses an unreasonable risk of reduced IQ in children,” wrote Judge Edward Chen. That risk, he added, is significant enough to require EPA action.

When you put all this together, the picture becomes clearer: one side of the ledger—the benefits—is thinner than we’ve long assumed. The other side—the risks—is thicker than many expected. And caught between them is a practice begun in the 1940s, carried into the 21st century, and rarely examined with modern scientific tools.

This is exactly the kind of issue the National Academies of Sciences, Engineering, and Medicine was created to unpack: a complex, cross-disciplinary question that touches public health, ethics, environmental exposure, early-life development, and the responsibilities of government.

A rigorous, independent review should ask:

• Do today’s fluoridation levels meaningfully reduce tooth decay?

• What are the risks—fluorosis, IQ loss, ADHD, bone fractures, hypothyroidism?

• Where, exactly, are pregnant women, infants, and children getting most of their fluoride?

• How do we weigh benefits and risks when exposure is delivered to everyone, without individual choice?

Until we have clearer answers on both benefit and harm, a measure of humility is warranted. Re-evaluating long-standing practices is not a setback; it is how science moves forward.