"Fluoride is a neurotoxin" is either obviously true or scaremongering, depending on who you ask. The honest answer depends almost entirely on dose, and on which studies you trust.
Where the "neurotoxin" label comes from
In 2014, researchers Philippe Grandjean and Philip Landrigan published a review in The Lancet Neurology adding fluoride to a list of developmental neurotoxicants, alongside lead, mercury and arsenic. That was an academic review, mostly based on Chinese studies of high natural fluoride. We found no government that formally classifies fluoride at water-fluoridation levels as a neurotoxin.
The high-dose evidence
- US National Toxicology Program, August 2024: "moderate confidence" that drinking water above 1.5 mg/L is linked to lower IQ in children. Of 19 high-quality studies, 18 found that link. It said the data were insufficient to judge the 0.7 mg/L level used in US fluoridation.
- Taylor et al., JAMA Pediatrics, January 2025: a meta-analysis of 74 studies by the same team. The link held when looking only at water under 2 mg/L, but there was no association below 1.5 mg/L. An accompanying editorial argued the analysis overstated the impact.
The studies at lower levels
These are the ones that matter most for Australia, and they point in different directions.
- Green et al. 2019 (Canada, pregnancy): 512 mother-child pairs. Each 1 mg/L rise in the mother's urinary fluoride was linked to a 4.49-point lower IQ in boys, with no link in girls. The journal took the unusual step of publishing an editor's note explaining why it ran the paper.
- Till et al. 2020 (Canada, infant formula): formula-fed babies in fluoridated areas had lower non-verbal IQ. Formula fluoride wasn't measured directly.
- Bashash et al. 2017 (Mexico): 299 mother-child pairs. Higher fluoride in pregnancy was linked to lower scores at ages 4 and 6 to 12. Mexico fluoridates salt, not water.
- Karolinska Institutet 2025 (Bangladesh): about 500 mother-child pairs with natural fluoride. Higher exposure was linked to lower scores at age 10, not at age 5. The authors stress it can't prove cause.
The studies that found nothing
- Do et al. 2022 (Australia): 2,682 children from the National Child Oral Health Study. Early-life fluoridated water had no link to behaviour or executive function.
- Do et al. 2024-25 (Australia): 357 young adults given a full IQ test. No difference between those with full early-life fluoridation and none. The small sample means wide margins of error.
- Broadbent et al. 2015 (New Zealand): the Dunedin cohort, followed to age 38. No IQ difference.
- England, March 2026: the UK government's monitoring report found no relationship between water fluoride and test scores at ages 5 and 7.
The US court case
In September 2024, US federal judge Edward Chen found fluoridation at 0.7 mg/L posed an "unreasonable risk" of lower IQ for babies exposed in the womb, and ordered the EPA to act. In May 2026 an appeals court vacated that order on procedural grounds, without ruling on the science.
So, is it a neurotoxin?
At high exposures, the evidence of harm to developing brains is fairly consistent. At the levels used in Australian water, the best Australian and New Zealand studies have found no effect, while some Canadian and Mexican pregnancy studies have found one. The group with the most uncertainty is pregnant women and formula-fed babies, which is exactly what the NHMRC's 2026 review is expected to examine.