What If Fluoride in Drinking Water Was Actually Harmful? What 80 Years of Studies Show

Water fluoridation has been called everything from a public health triumph to a covert poisoning, and the evidence supports neither slogan. At the 0.7 milligrams per liter (mg/L) used in the United States, fluoride lowers tooth decay and causes mostly mild, cosmetic enamel changes. At several times that concentration, found naturally in some regions, it damages teeth and bones. Whether the dose in between affects children's brains is the live scientific question. This article walks through 80 years of data, what each study design can and cannot show, and where the uncertainty actually sits.

Thought experiment. This article explores a hypothetical scenario for educational and speculative purposes. The events described are not predictions and should not be interpreted as warnings, advocacy, or instructions. It describes general research findings, not medical advice.

Key takeaways

• The CDC estimates that community water fluoridation reduces tooth decay by about 25% in children and adults. A 2024 Cochrane review found a smaller benefit in studies run after fluoride toothpaste became common, and rated the evidence low certainty.
• At 0.7 mg/L, roughly 12% of people develop dental fluorosis of aesthetic concern and about 40% show it at any level, according to Cochrane. Most cases are the mildest form.
• Skeletal damage appears at far higher exposures, mostly in regions with naturally fluoride-rich groundwater.
• A 2024 National Toxicology Program (NTP) review found moderate confidence that exposure above 1.5 mg/L is linked to lower IQ in children, and said the data were insufficient to judge 0.7 mg/L.
• Whether and how to fluoridate is a policy decision made differently around the world, not a question the science alone settles.

What fluoride is, and where you already get it

Fluoride is the ionic form of fluorine, a highly reactive element. The CDC describes it as found naturally in "almost all soil and water and many rocks," which is why nearly every water supply contains some. A river or aquifer in granite country may carry a lot; a supply drawn from a low-fluoride source may carry almost none.

Fluoridating a public water supply means topping it up to a target concentration. In the United States the most common additive is fluorosilicic acid, followed by sodium fluorosilicate and sodium fluoride (Wikipedia: Water fluoridation). A concentration of 1 mg/L is the same as one part per million (ppm), the unit older documents use.

Water is only one route into the body. Fluoride also comes from toothpaste, mouth rinses, and processed foods and drinks manufactured with fluoridated water, a spillover the CDC's 2001 recommendations call the "halo effect" (CDC MMWR, 2001). That matters for everything below, because a person's total dose is what counts, and studies that measure only tap water can miss most of it.

Three numbers frame the policy debate, and they are easy to mix up:

Benchmark	Value	What it means	Source
U.S. recommended level for fluoridation	0.7 mg/L	Target concentration, set by the U.S. Public Health Service in 2015	CDC timeline
EPA secondary level	2.0 mg/L	Advisory ceiling to help children avoid dental fluorosis	CDC FAQ
EPA enforceable limit (MCL)	4.0 mg/L	Legal maximum in public water systems, set in 1986	EPA

The World Health Organization (WHO) uses a guideline value of 1.5 mg/L as an upper limit, and the WHO does not itself set a fluoridation target (Wikipedia: Water fluoridation). The gap between 0.7 and 4.0 is the whole argument in miniature: the recommended dose sits well below the legal limit, and the open question is how safe the space between them is.

How a brown-stain mystery became a public health program

The story starts with discolored teeth. In 1901, dentist Frederick McKay arrived in Colorado Springs and noticed that many locals had brown-mottled enamel, which he called the "Colorado brown stain." He also noticed these residents seemed to have unusually little decay. G.V. Black joined the investigation in 1915, and water samples from the affected communities showed nothing unusual (CDC timeline).

The answer came in 1931, when researcher H. Trendley Dean determined that the mottling came from excess fluoride and renamed it dental fluorosis. Dean's 21 Cities Study, reported in 1942, found that water at about 1.0 ppm reduced decay with negligible risk of fluorosis. That was the pivot from "fluoride stains teeth" to "the right amount of fluoride protects them."

The first real-world test began on January 25, 1945, when Grand Rapids, Michigan, became the first city to add fluoride to its water. A parallel trial paired Newburgh, New York, with nearby Kingston as an unfluoridated control. By 1950 the trials suggested that adding fluoride produced "the same effects as natural fluoridation," in the CDC timeline's words.

Policy then moved in steps. In 1962 federal guidance set an optimal range of 0.7 to 1.2 ppm, depending on local climate. The Safe Drinking Water Act of 1974 made fluoridation a state responsibility rather than a federal one. In 2015, the Public Health Service replaced the range with a single figure of 0.7 mg/L, a change that, per the Wikipedia summary of the guidance, aimed to reduce the risk of dental fluorosis.

That history shows something. The recommended dose was lowered once already, in response to evidence that people now get fluoride from more sources than they did in 1945. Public health guidance on fluoride has revised itself before, which is how a healthy evidence process behaves.

How fluoride protects enamel

Think of tooth enamel as a brick wall made of tiny mineral crystals, mostly a calcium phosphate called hydroxyapatite. Every time you eat sugar or starch, bacteria in dental plaque turn it into acid, and the acid pulls minerals out of the wall. Between meals, saliva pushes minerals back in. Decay is what happens when the withdrawals outrun the deposits for long enough.

Fluoride acts on both sides of that ledger. The CDC's 2001 report says fluoride "enhances the remineralization (i.e., recovery) of demineralized enamel," and that enamel which takes up fluoride becomes "more acid resistant and contains more fluoride and less carbonate." In effect, some of the mortar is swapped for a harder, less acid-soluble kind.

A second effect works on the bacteria. The same report says that as fluoride concentrates in plaque, "it inhibits the process by which cariogenic bacteria metabolize carbohydrates," which slows acid production. The report also cautions that whether this actually lowers the cavity-causing power of plaque bacteria in people is unclear.

Does swallowing it matter, or only touching it?

The CDC's 2001 report concluded that fluoride's "predominant effect is posteruptive and topical," meaning it works mainly by bathing teeth that have already emerged, not by being built into them during development. A preeruptive effect, from fluoride incorporated while teeth form, was hypothesized and has some supporting evidence, but it is hard to separate from the topical effect.

This is why toothpaste can do much of fluoridation's job. Water reaches teeth as you drink and, to a lesser degree, through saliva after swallowing. Toothpaste delivers a far higher concentration directly to the surface, twice a day. The anatomy of a tooth explains why the outer enamel layer, which has no living cells to repair it, depends on this chemistry.

How much does fluoridation actually prevent?

The honest answer is "less than the first studies suggested, and measuring it has become harder." Four reference points:

Source	Finding	Caveat
CDC early-studies summary (2001)	About 50% to 60% less childhood decay in early studies, 18% to 40% in more recent estimates	Drop attributed mainly to fluoride from other sources, especially toothpaste
Cochrane review, 2015	About 35% fewer decayed, missing, or filled baby teeth and about 26% in permanent teeth; 14% to 15% more caries-free children	Over 97% of studies at high risk of bias
Cochrane review, 2024 update	Difference of about a quarter of a tooth in baby teeth (0.24, 95% CI -0.03 to 0.52); 3 to 4 percentage points more caries-free children	157 non-randomized studies; certainty low to very low; permanent-teeth effect unclear
CDC scientific statement, May 15, 2024	"About 25% in children and adults" less tooth decay	Summarizes expert reviews rather than a new meta-analysis

The two Cochrane reviews are worth reading side by side. The 2015 version estimated large reductions in decay. The 2024 update, which added newer studies through August 2023, found that in studies run after 1975, the benefit shrank to roughly a quarter of a tooth per child, with a confidence interval that includes zero. The authors note that pre-1975 studies predate widespread fluoride toothpaste, so their results may not apply today (Cochrane, 2024).

Two things can both be true. Fluoride remains protective, and the extra protection from water, on top of toothpaste, is smaller than it was in 1945. The Cochrane authors advise weighing local factors such as baseline oral health, diet, how much tap water people drink, and other prevention options. A community with high decay rates and limited dental care has more to gain than one with universal dental coverage.

This is also a lesson in how to read a research paper: a headline percentage depends on the era, the comparison group, and the study design. All 157 studies in the Cochrane review were non-randomized, because assigning whole cities to fluoridated or unfluoridated water at random is rarely feasible. When randomized trials are off the table, researchers lean on comparisons between communities that differ in other ways, which weakens certainty.

What dental fluorosis is, and how common it is

Dental fluorosis is a change in enamel caused by taking in too much fluoride while teeth are forming. The CDC notes it develops only in children whose teeth are still developing under the gums, up to about age eight. In its mildest form it shows as faint white flecks that most people never notice. In severe cases, from high exposures, enamel can be brown-stained and pitted.

The CDC's May 2024 statement says the documented risks of fluoridation are limited to dental fluorosis, "a change in dental enamel that is cosmetic in its most common form." The 2001 CDC report adds that most fluorosis today is of the mildest form, "which affects neither cosmetic appearance nor dental function," and that its prevalence has risen in both fluoridated and unfluoridated areas.

Cochrane puts numbers on it. At 0.7 ppm, about 12% of people (95% CI 8% to 17%) had fluorosis of aesthetic concern, and about 40% (95% CI 35% to 44%) had fluorosis of any level, though the certainty of that evidence is low. "Of aesthetic concern" is a judgment about appearance, not a measure of damage to the tooth.

A 2016 photographic survey in four English cities gives a feel for the scale. Among 1,904 children aged 11 to 14, fluorosis above the second severity grade (TF2) appeared in about 10% of children in the two fluoridated cities and 2% in the two unfluoridated ones. The mildest grade, TF1, was common everywhere: 41% versus 32%. The authors concluded that the level of aesthetic concern was low and stable (Pretty et al., 2016).

What about infant formula?

Because fluorosis depends on intake during tooth development, infants who get most of their calories from formula mixed with fluoridated water get a larger share of their dose from the tap. The CDC says powdered or liquid-concentrate formula mixed with fluoridated water may raise the chance of dental fluorosis if formula is the child's main food. Ready-to-feed formula contains little fluoride, and the CDC notes that parents can use low-fluoride bottled water some of the time to lower the risk. A pediatrician or dentist is the right person to weigh this for an individual child.

What happens at high doses

Everything about fluoride follows the old rule that the dose makes the poison, and the high end of the dose range is not hypothetical. In parts of India, China, and Africa, groundwater is naturally rich in fluoride, and people who drink it for decades can develop skeletal fluorosis, in which bones thicken and joints stiffen and ache.

A 2020 review of the global picture found fluorosis to be a notable medical problem in 24 nations, including India, and estimated that endemic fluorosis affects approximately 100 million people, mostly in parts of Africa and Asia (Srivastava and Flora, 2020). In the rural areas the review sampled, 80% of villages had fluoride above WHO limits. Remediation there means defluoridation, switching sources, or both, which is the exact opposite of adding fluoride.

The U.S. legal limit of 4.0 mg/L exists for this reason. For private wells or supplies at or above that level, the CDC advises considering a filter that removes fluoride from water used for drinking and cooking, and contacting the local health department. The asymmetry is striking: the same element, at 0.7 mg/L in one place and above 4.0 mg/L in another, is a prevention measure in the first and a contamination problem in the second.

The dose line, in one sentence. Fluoride at 0.7 mg/L is a deliberate, regulated addition. Fluoride at 4.0 mg/L is the legal ceiling the EPA has set because of the risks, and in some regions natural levels run higher still.

The IQ question: what the NTP and the 2025 meta-analysis found

This is where the current scientific argument lives, and it deserves care because both sides overstate it. The short version: there is consistent evidence of an association between higher fluoride exposure and lower IQ in children, mostly at exposures above those in typical U.S. fluoridated water, and unresolved uncertainty about whether the association holds, or matters, at 0.7 mg/L.

The NTP review

The National Toxicology Program, a U.S. federal research program, published its final monograph on August 21, 2024. It concluded "with moderate confidence" that higher fluoride exposure is associated with lower IQ in children, with the clearest evidence at drinking-water levels above 1.5 mg/L. For the low level used in fluoridation, it said there were "insufficient data to determine if the low fluoride level of 0.7 mg/L" affects children's IQ. The review covered studies published through October 2023 (NTP).

The JAMA Pediatrics meta-analysis

The NTP's reviewers published a companion meta-analysis in JAMA Pediatrics in March 2025 (Taylor et al., 2025). It pooled 74 studies, most of them cross-sectional and 45 of them from China. Fifty-two of the 74 were rated at high risk of bias, and none came from the United States.

Its main results, in plain terms:

• Across 59 group-level studies covering 20,932 children, higher fluoride exposure was associated with lower IQ (pooled standardized mean difference of -0.45).
• In 31 drinking-water studies there was an inverse dose-response association, but it was null below 1.5 mg/L.
• In 13 studies with individual-level urine measurements, each 1 mg/L increase in urinary fluoride was associated with a 1.63-point lower IQ (1.14 points among the low-risk-of-bias studies).

The authors wrote that evidence below 1.5 mg/L from drinking water alone was limited and uncertain. In May 2025, JAMA Pediatrics published letters raising concerns about the data and analyses, along with the authors' reply, which shows the interpretation is still being argued out in print.

Two birth cohorts

Cohort studies follow people forward in time and measure exposure before the outcome, which makes them stronger than cross-sectional snapshots. Two are central:

Study	Population	Exposure measure	Finding
Bashash et al., 2017	299 mother-child pairs, Mexico City	Maternal urinary fluoride in pregnancy (mean 0.90 mg/L)	A 0.5 mg/L rise was linked to 3.15 points lower on a cognitive index at age 4 and 2.50 points lower IQ at ages 6 to 12
Green et al., 2019	601 mother-child pairs, six Canadian cities (512 with complete urine data)	Maternal urinary fluoride and self-reported intake	A 1 mg/L rise was linked to 4.49 points lower IQ in boys (95% CI -8.38 to -0.60); the association in girls was not significant

Both studies measured fluoride in urine, which reflects everything a person takes in, not just tap water. That makes them informative about total exposure but harder to translate into a statement about a water concentration. The Canadian study also found a sex difference whose cause nobody has explained, a pattern that needs replication before anyone draws conclusions from it.

Why an association is not yet a verdict

The NTP and meta-analysis findings are associations, and the gap between "linked to" and "caused" is wide here. Most of the underlying studies compare communities that differ in many ways at once, such as other water contaminants, nutrition, schooling, and family income. The review itself rated most of its studies at high risk of bias. A simple primer on correlation versus causation explains why that distinction matters.

The CDC's May 2024 scientific statement, written before the final NTP monograph, said expert panels had found no convincing evidence linking community water fluoridation to adverse outcomes including low intelligence. The NTP and the court case below came later. Taken together, the fair summary is that the data at higher exposures have strengthened, and the data at 0.7 mg/L remain inconclusive in both directions.

The scale of the reported effect also deserves context. A 1.63-point difference per 1 mg/L of urinary fluoride is a group average. It is not a prediction for any individual child, and it is small relative to the normal spread of IQ scores, which is why the studies rely on large samples to detect it.

What if fluoridation really were harmful? How we would find out

Now the thought experiment. Suppose the low-dose IQ association turned out to be real and causal at 0.7 mg/L. What would the evidence have to look like, and what would change?

First, the signal would have to replicate. A single well-run cohort is a lead, not a conclusion. The strongest confirmation would be several prospective cohorts, in countries with different baseline exposures, that measure water, urine, and diet, adjust for other exposures, and find the same dose-response curve. The existing studies already hint at this design, but their ranges overlap only partly with 0.7 mg/L.

Second, natural experiments would help. When Calgary stopped fluoridating in May 2011 while Edmonton continued, researchers compared children's teeth in the two cities and found more primary-tooth decay in Calgary than in Edmonton afterward, consistent with an adverse effect of stopping (McLaren et al., 2016). The same logic could be applied to cognitive outcomes if cities stop or start fluoridating and children are tracked afterward. Randomly assigning water supplies is not realistic, but such cessation and start-up events create comparisons that cohorts cannot.

Third, the policy response would follow the evidence in steps, as it did in 2015. Regulators could lower the target concentration again, shift toward topical delivery such as toothpaste and varnish, or let communities choose. The WHO's own oral-health guidance already encourages twice-daily brushing with fluoride toothpaste at 1,000 to 1,500 ppm (WHO fact sheet), which shows that fluoride delivery does not depend on the tap.

Fourth, the cost-benefit math would shift with the size of the effect. The benefit side, per Cochrane's 2024 estimate, is modest in the era of fluoride toothpaste and larger in places with poor dental access. The risk side is unknown at 0.7 mg/L. A policy decision needs both numbers, and today only one of them is firm. That is the real state of play, and it explains why serious people reach different conclusions.

The law and the regulators are moving

Several recent events have put the debate in front of regulators and courts. In September 2024, U.S. District Judge Edward Chen ruled in Food & Water Watch v. EPA that the EPA must take regulatory action, writing that there is "substantial and scientifically credible evidence" that fluoride poses a health risk at currently recommended levels. The ruling also stated that it "does not conclude with certainty that fluoridated water is injurious to public health," and it left the remedy to the EPA, with options ranging from a warning label to tighter restrictions (CBS News).

The EPA's own review has been moving since. The agency says it committed on April 7, 2025, to review new science on fluoride, released a preliminary assessment plan on January 22, 2026, and published a protocol for its human health toxicity assessment on August 3, 2026. As of that page, it had not said when the assessment will finish or whether the 4.0 mg/L limit will change (EPA).

At the state level, Utah's HB 81, signed on March 27, 2025, and effective May 7, 2025, prohibits communities from adding fluoride to public water supplies and allows pharmacists to prescribe fluoride supplements. Utah was the first state to enact such a ban, and about two in five Utah residents were served by fluoridated systems before the law. The bill's House sponsor said it "isn't anti-fluoride legislation" and described it as pro-informed consent (CBS News).

As of 2022, CDC data show 72.3% of Americans on community water systems, about 209 million people, received fluoridated water (CDC). That is a large installed base, and it means any change in U.S. policy would reach most of the population.

Different countries, different choices

Water fluoridation is the dominant approach in a handful of countries and rare in most of the rest, and the split does not track dental health in any simple way. These figures come from the compilation in the Wikipedia article on water fluoridation by country, which cites national agencies; years differ by row, so treat them as approximate snapshots.

Country	Share of population on fluoridated water	Year or note
United States	72.3% of people on community water systems	2022, CDC
Australia	70% or more in every state and territory	Not stated in source
Ireland	Almost 71%	2002
Canada	About 45% (Quebec about 6%)	2007
Brazil	41%	2008
United Kingdom	About 14%	Not stated in source
Spain	About 10%	Not stated in source
Singapore	100%	Program began 1956
Israel	None in practice	Stopped August 26, 2014

Other places started and then stopped. These examples show that the decision is local policy, not a coordinated plan:

Place	What happened
Netherlands	Fluoridated 1960 to 1973, when the High Council ruled it unauthorized
Sweden	Norrköping began in 1952; parliament prohibited water fluoridation in 1971
Czech Republic	Prague stopped in 1988; no water fluoridated since 2008
Finland	Kuopio, the only city of 70,000 or more ever fluoridated, stopped in 1992
Basel, Switzerland	Water fluoridation stopped in 2003; fluoridated salt used instead
Waterloo Region, Canada	Ended after a 2010 referendum
Calgary, Canada	Started 1991, stopped May 2011

Some countries deliver fluoride another way. Switzerland has fluoridated salt since 1962, with about 83% of domestic salt sold containing it, and Mexico, France, and Germany also use fluoridated salt programs. In those countries the policy goal is the same, preventing decay, with a different delivery mechanism and a different consent model. A person can choose not to buy fluoridated salt; a person on a municipal water supply cannot easily opt out.

Claims people make about fluoride, checked

A number of claims travel with this topic. Some are testable, some are about values, and a few have been checked directly.

"Fluoridation chemicals are industrial waste"

The origin is real: fluorosilicic acid, the most common additive in the United States, is a by-product of phosphate fertilizer manufacturing. That fact alone says nothing about safety, because many ordinary water treatment chemicals come from industrial processes. The relevant question is what is in the finished water and at what concentration, which is the 0.7 mg/L target and the enforceable limits above.

"Fluoride causes cancer"

The bone cancer osteosarcoma is the case most often cited. A 2006 case-control study of 103 cases under age 20 and 215 controls found a link in males exposed at age seven (adjusted odds ratio 5.46, 95% CI 1.50 to 19.90) and none in females, and its own authors called it exploratory and said further research was required (Bassin et al., 2006). A 2011 study measured fluoride directly in bone from 137 osteosarcoma cases and 51 tumor controls and found no significant association (adjusted odds ratio 1.33, 95% CI 0.56 to 3.15) (Kim et al., 2011). The CDC's 2024 statement says expert panels have not found convincing evidence linking fluoridation to cancer, heart disease, or immune disorders.

"It is mass medication without consent"

This is an ethical argument, not an empirical one, and it is legitimate. Societies weigh public health interventions against individual choice differently, as the table above shows. Evidence can inform that trade-off but cannot settle it. Utah's lawmakers framed their ban in these terms, while the CDC frames fluoridation as a population-level prevention measure.

"There is a global conspiracy"

The record shows the opposite of coordination. Different countries have started, stopped, and reversed fluoridation for their own reasons, and scientific bodies keep arguing about it in public, as the NTP review, the Cochrane updates, and the JAMA Pediatrics letters all show. A conspiracy would not produce this much open disagreement. The same pattern is why people studying how scientific consensus forms keep returning to why scientists disagree and why that is a feature of the process.

Common misconceptions

"Fluoride is toxic at any dose." Almost every substance, including water and salt, is harmful at a high enough dose. Fluoride at 0.7 mg/L has decades of use, a well-documented cosmetic effect, and an unresolved question about neurodevelopment. At 4 mg/L and above it is regulated as a hazard.

"Fluoride toothpaste makes water fluoridation pointless." Toothpaste has reduced the marginal benefit, which is the main reason Cochrane's 2024 numbers are smaller than its 2015 numbers. The extra benefit is not necessarily zero, and it is larger where people have less access to dental care.

"New research overturns 80 years of consensus." The new research raises a real question about neurodevelopment at higher exposures. It has not overturned the evidence that fluoride at low concentrations protects teeth, and it has not shown harm at 0.7 mg/L either.

"A finding in China or India applies directly to a U.S. city." Many of the IQ studies come from regions with high natural fluoride and different baseline nutrition, which is exactly why the meta-analysis finds the evidence below 1.5 mg/L uncertain. Dose, population, and co-exposures all matter.

"Skeptics and defenders are each simply wrong." Both camps hold pieces of the picture. A person who says fluoridation prevents decay is supported by the CDC and by Cochrane, though the 2024 Cochrane effect is smaller than the old figures. A person who says the neurodevelopment data deserve scrutiny is supported by the NTP, though the NTP drew no conclusion at 0.7 mg/L. Other what-if scenarios about contested health claims follow the same pattern: an extreme claim fails, and a narrow, measurable question survives.

Frequently asked questions

Q: Does fluoridated water help adults or only children? A: The CDC's 2024 statement estimates a reduction of about 25% in decay for both children and adults. Cochrane's 2015 review found the evidence for adults inconclusive, so the adult benefit is less firmly measured than the benefit for children.

Q: Is fluoride safe for infants? A: Dental fluorosis is the documented concern, because it can develop while teeth form. The CDC says formula made from powder or liquid concentrate with fluoridated water may raise fluorosis risk if formula is the main food, that ready-to-feed formula contains little fluoride, and that low-fluoride bottled water can be used some of the time. Parents should ask their pediatrician or dentist about their own situation.

Q: Do water filters remove fluoride? A: Filters differ, and many common ones are not designed to remove it. For water at 4.0 mg/L of fluoride or higher, the CDC advises considering a filter that is specifically designed to remove fluoride and contacting the local health department for specific steps.

Q: What did the National Toxicology Program conclude? A: In its August 21, 2024 monograph, the NTP concluded with moderate confidence that fluoride exposure above 1.5 mg/L is associated with lower IQ in children. It said there were insufficient data to determine whether 0.7 mg/L affects IQ.

Q: Why do the Cochrane reviews disagree with the CDC's 25% figure? A: They do not exactly disagree; they measure different things at different times. Cochrane's 2015 review found larger reductions, its 2024 update found a smaller benefit in studies after 1975, and the CDC summarizes expert reviews. The differences mostly reflect when the studies were done and how much fluoride people now get from toothpaste.

Q: Is dental fluorosis dangerous? A: In its most common, mildest form it is cosmetic. The CDC says it affects neither appearance nor function in most cases today. Severe fluorosis, with pitting and brown staining, is associated with high exposures.

Q: What is the EPA doing about fluoride right now? A: According to the EPA, it committed to a review in April 2025, released a preliminary assessment plan in January 2026, and published a toxicity assessment protocol in August 2026. It has not announced whether the 4.0 mg/L limit will change.

The takeaway

The fluoride debate is not "safe versus poison." It is a dose-response problem with three zones. At 0.7 mg/L, the benefit is real but smaller than the first studies suggested, mild dental fluorosis is common, and the effect on children's brains is unresolved. Above roughly 1.5 mg/L, the NTP found moderate confidence that exposure is associated with lower IQ. Above 4 mg/L, regulators treat fluoride as a hazard, and natural exposures in some regions cause skeletal disease. A better question than "should we fluoridate?" is "what does the next round of cohort and natural-experiment data show at 0.7 mg/L, and what would we change if it showed harm?" Anyone claiming certainty in either direction is ahead of the evidence.

Sources

• NTP: Fluoride exposure and neurodevelopment and cognition (monograph page)
• Cochrane, Water fluoridation for the prevention of dental caries (2024 update)
• CDC: Timeline for community water fluoridation
• CDC: Community water fluoridation FAQs
• CDC: Scientific statement on safety and effectiveness of community water fluoridation (May 15, 2024)
• CDC: 2022 water fluoridation statistics
• CDC MMWR: Recommendations for using fluoride to prevent and control dental caries (2001)
• EPA: Fluoride in drinking water
• Taylor et al., JAMA Pediatrics (2025)
• WHO: Oral health fact sheet

---

Read next:

• How We Know Vaccines Are Safe: The Evidence
• What If GMOs Were as Dangerous as Critics Claim? What 30 Years of Evidence Show
• What If 5G Really Did Cause Health Problems? Here's What Science Found
• What If Humans Could Live to 1,000 Years? The Science of Radical Longevity