Is fluoride in drinking water bad for you?
Written by AI from saved sources. A script checks every quote word for word, and a second AI compares each claim with its source. How we work and corrections
The amount matters: higher fluoride exposure carries real risks, but evidence does not justify calling all fluoridated drinking water harmful. At the US target of 0.7 milligrams per litre (mg/L), harm to children's IQ (intelligence-test scores) remains uncertain. Adding fluoride to water may also slightly reduce decay in children's baby teeth.
Checked: NTP/Cochrane/WHO, October 2026.
Why people believe it
There are serious findings behind the worry. The US National Toxicology Program (NTP) links higher exposure with lower childhood IQ with moderate confidence. It also raises a practical concern. Some pregnant women and children get fluoride from several sources. Their combined intake may exceed safe amounts. Those are reasonable concerns to investigate.
“It lowers children's IQ”
NTP's 2024 review found this link with moderate confidence at higher exposures, such as drinking water above 1.5 mg/L. It assessed total fluoride exposure, rather than fluoridated water alone. It could not determine whether 0.7 mg/L in drinking water harms children's IQ.
A 2025 systematic review, which brings together study results, combined 74 studies. Of these, 52 were rated at high risk of bias: their methods might distort the findings. Three water studies rated at low risk of bias involved 879 children. Their result below 1.5 mg/L pointed towards lower IQ. But it was not statistically significant, meaning it did not meet the statistical test's threshold. The authors say the data on drinking-water fluoride below 1.5 mg/L and children's IQ are limited. Below 1.5 mg/L, uncertainty remains about how children's IQ relates to fluoride exposure estimated from drinking water alone.
“It damages teeth”
Dental fluorosis means changes in the appearance of tooth enamel. Swallowing fluoride over time while teeth form causes these changes. Cochrane's older evidence was low certainty, meaning confidence in the estimates was limited. At 0.7 parts per million, it estimated about 40% had fluorosis of any level. It estimated 12% had changes of cosmetic concern, meaning concerns about appearance. These are estimates of how common it was, not extra cases caused solely by water fluoridation. That evidence runs only to February 2015.
“It damages bones”
The World Health Organization's (WHO) 2004 review says skeletal fluorosis may occur at water concentrations of 3–6 mg/L. This means harmful changes in bone structure. It says the crippling form usually develops where water contains over 10 mg/L. These concentrations are much higher than the US fluoridation target.
“There is no benefit anymore”
Cochrane's 2024 review says benefits may be smaller since fluoride toothpaste became widespread. Two studies involved 2,908 children. After fluoridation began, they suggested an extra reduction over time in decayed, missing or filled baby teeth. This averaged 0.24 teeth per child, compared with control communities (communities used for comparison). The evidence was low certainty and allowed for no benefit. For changes in decayed, missing or filled permanent teeth and permanent tooth surfaces in children, the review was very uncertain.
Thinking tool: which dose, which population?
Before applying a finding to your tap water, check what was measured and in whom. The 2025 review found links with lower childhood IQ below 1.5 mg/L in urine. But urine and water measurements describe different exposure. WHO says how much water people drink and fluoride from other sources must also be considered. A finding about high exposure does not tell us how big the risk is at 0.7 mg/L. EFSA says the evidence is insufficient to establish a threshold for possible effects on an unborn baby's developing central nervous system.
Use this on other claims: ask whether the amount, people and outcome match the situation you are judging.
Official positions, with dates
The American Dental Association made a statement on 6 January 2025. It said the 2025 IQ review did not justify changing fluoridation practice. The American Academy of Pediatrics' position was updated on 3 November 2025. It says evidence does not support harm to the development of thinking and learning at the US target.
The European Food Safety Authority's assessment is dated 22 July 2025. It says evidence below 1.5 mg/L in water is inconsistent and insufficient for clear conclusions. This concerns possible effects on an unborn baby's developing central nervous system. The Environmental Protection Agency (EPA) released its assessment protocol on 3 August 2026. This plan sets out methods, not harmful exposure levels. Read these as dated positions and review steps, alongside the findings above.
What people also ask
Is there fluoride in tap water?
Some communities add fluoride to their drinking water. For US communities that do, the Public Health Service's 2015 recommended concentration is 0.7 mg/L.
Does fluoride stain teeth?
Swallowing fluoride over time while teeth are forming can change tooth enamel's appearance. This condition is called dental fluorosis.
Why is fluoride in tap water?
Community fluoridation programmes add it to help prevent tooth decay while limiting the chance of dental fluorosis. Cochrane's 2024 review found low-certainty evidence that starting fluoridation may slightly reduce decay in baby teeth. The results also allowed for no benefit.
What are the bad side effects of fluoride?
Fluoride swallowed over time while teeth form can cause dental fluorosis, which changes enamel's appearance. WHO's 2004 review says skeletal fluorosis may occur at drinking-water concentrations of 3–6 mg/L. This means harmful changes in bone structure. Those concentrations are above the US fluoridation target of 0.7 mg/L.
Is fluoride linked to low IQ?
NTP's 2024 review found, with moderate confidence, that higher fluoride exposure is associated with lower IQ in children. This includes exposure such as drinking water above 1.5 mg/L. It could not determine whether drinking water at the US target of 0.7 mg/L harms children's IQ.
Check it yourself (5 minutes)
- Open NTP's public page “Fluoride Exposure: Neurodevelopment and Cognition”.
- Use your browser's Find for “insufficient data”.
- Read that sentence: does it establish harm at 0.7 mg/L, establish safety, or leave the question unresolved?
What you should have found
“It is important to note that there were insufficient data to determine if the low fluoride level of 0.7 mg/L currently recommended for U.S. community water supplies has a negative effect on children’s IQ.”
Before you forward this: does the box at the top say exactly what the sources below say?