Why would fluoride reduce IQ?

Sterling Hayden as Colonel Jack Ripper in Dr Strangelove This Stanley Kubrick movie stigmatized research on fluoride toxicity for over fifty years

luoride isn't just for conspiracy theorists, mad Air Force colonels, and believers in chemtrails anymore. The US EPA is now planning to “expeditiously review new scientific information” on fluoride, an inorganic ion that has been added to our drinking water for decades, saying there is “moderate confidence that fluoride exposure above 1.5 milligrams per liter is associated with lower IQ in children.”

Even though the news media try to make it political by spinning it as President Trump trying to wreck children's health, the possibility that we've been reducing the collective intelligence of the population by our attempts to reduce dental cavities is too serious to be dismissed.

There are several points to keep in mind.

• Fluoride over 0.7 mg/L adversely affects learning and memory in animals, according to the 2016 National Toxicology Program.^[1]

• Fluoride anion doesn't participate in free radical or oxidation-reduction reactions. Chemists regard it as the most gentle counterion in ionic reactions, as it's easily displaced from anion exchange columns by other anions such as chloride and even hydroxide. Therefore, it's unlikely to react directly with human cells, though it could still interfere in biochemical processes, possibly in the form of a complex with aluminum or calcium.

• When acidified, fluoride becomes hydrogen fluoride (HF), aka hydrofluoric acid, an acid that is so corrosive that it etches glass. Ironically, the toxicity of HF is thought to be due more to fluoride ion (F⁻) than to its corrosive effect. HF is said to be cytotoxic due to the ability of F⁻ to form insoluble chelates CaF₂ and MgF₂ with calcium and magnesium ions, causing local calcium depletion and increasing permeability of the cell membrane to potassium.^[2] This differs from other strong acids, which cause coagulative necrosis that slows penetration. Again, this is only an effect of concentrated HF, not fluoride at the levels in water, but it means that the possibility of direct chemical effects can't be peremptorily ruled out.

  Hydrofluoric acid isn't as strong an acid as people think. (A ‘strong’ acid is defined as one that is highly dissociated, in this case into H⁺ and F⁻.) Its pKa is only 3.15–3.18, which means its Ka (acid dissociation constant) is 6.61×10⁻⁴, making it a ‘weak’ acid roughly comparable to formic acid. This means that in acidic cell compartments such as lysosomes (pH 4.7–4.8), secretory vesicles (pH 5.8), and trans-Golgi (pH 6.0), some percentage of the fluoride would be in the form of hydrofluoric acid. For example, as much as 3% of any fluoride in a lysosome would be in the form of HF. It should be noted that some of the pKa tables on the Internet have many typos.

• Other dietary components, especially milk, inhibit fluoride bioavailability by forming insoluble calcium precipitates. Calcium in hard water could also in principle affect fluoride levels. Removing fluoride by ion exchange, as used in home water purifiers, would be difficult because, as mentioned, fluoride has low affinity for anion exchange resins. Most water purifiers only remove cations such as mercury and calcium and make no attempt to remove anions like F⁻.

• Much current thinking presupposes a neurodevelopmental effect, but this may be premature. If it lowers IQ in children, it would cause lifelong cognitive impairment. And it doesn't rule out the possibility of effects on adults.

• To be scientifically credible, a mechanism is needed. Everyone remembers the hysteria about power lines and radio waves, which turned out not to be a real effect. One reason scientists were skeptical about risks from 5G was that there was no physical mechanism that anybody could think of for how it could harm human cells.

• Measuring small amounts of cognitive impairment is extremely challenging, especially in children.

• Focusing exclusively on drinking water may overlook the possibility that other sources, such as fluoridated toothpaste, endogenous fluoride in water supplies, mouthwash, and foods and beverages could push the levels high enough to induce toxicity. Thus, actually measuring fluoride levels is essential. (It's a bad reflection on science that this needs to be said.)

What does the research say

Margin of safety Guth et al.^[3] say the margin of safety (margin of exposure, MoE) is high, ranging from 50 to 210. They dismiss the epidemiological evidence, saying many studies are done in regions without community fluoridation, so any effects could be due to confounding factors such as socioeconomic status, low birth rate, maternal intelligence, and exposure to other neurotoxic chemicals. It is unclear, however, how socioeconomic status and parental IQ would covary with fluoride, even in places without community fluoridation.

So, if you dismiss human studies, where does the 50 MoE come from? Animal studies (mice, rats, and rabbits), which could be orders of magnitude different in sensitivity than humans, and where subtle declines in intelligence are even harder to measure. The abstract says:

The results show that the margin of exposure (MoE) between no observed adverse effect levels (NOAELs) in animal studies and the current adequate intake (AI) of fluoride (50 μg/kg b.w./day) in humans ranges between 50 and 210, depending on the specific animal experiment used as reference.

In terms of concentration, if an average two year old male baby (13 kg, 85 cm) drinks 0.9 liters of water a day, that comes out to needing 0.72 mg/L in the water supply. So the standard concentra­tion of 4 mg/L would be 5.54 times more than the baby needs. It seems that they're all unknow­ing­ly in agreement that fluoride levels can be safely reduced. So why is everybody arguing?

How many rats would you need to detect a 2 or 3 point drop in their IQ? From the experiments my colleagues and I have done, the number would be astronomical if it could be done at all. Yet 3 points could mean the difference between a civilization that thrives and explores the universe and one that sinks into bureaucracy and cultural stagnation.

(A rat usually scores a 0 on the Stanford-Binet test. Sometimes they just eat the form instead of filling it out.)

Small amounts of fluoride come from tea and fish, but the primary source is water. The authors say that in Germany, Switzerland, and parts of Latin America, fluoride is added to salt. They cite research saying that normal adults have 7.6–45 μg/L plasma while adults drinking fluoridated water may have 190 μg/L.

However, their claim that addition of fluoride causes water acidification that could lead to contamination of water with aluminum, zinc, arsenic, lead, or mercury due to leaching is questionable. If it were a simple textbook reaction, your tap water would have a pH around 4, when in fact natural processes turn it closer to 6.0–6.5. If it were true that it's acidic enough to leach heavy metals, it would be a strong argument for banning fluoridation altogether.

Epidemiological studies A high-profile epidemiological meta-review in JAMA Pediatrics by Taylor et al. of the NIEHS found inverse correlations between children's IQ and fluoride exposure above 1.5 mg/L across a “large multicountry epidemiological literature ^[4] but did not propose a mechanism of action.

Grandjean et al. reviewed 14 studies and found that 13 showed a statistically significant connection between elevated fluoride exposure and reduced intelligence. These authors say early exposure to fluoride can result in IQ deficits that may be considerable. They cite a study that found that after adjustment for socioeconomic status, each 1% increase in fluoridation was associated with 67,000–131,000 additional ADHD diagnoses. However, epidemiological studies are ‘soft’ almost by definition, meaning they're subject to stray correlations. For instance, in the ADHD study, correcting for mean elevation eliminated the effect. Of course, that doesn't mean it is or is not real. The authors say prospective studies are needed, but that there are challenges in measuring early exposure. Because fluoride is a neuro­develop­mental toxicant, they advise that the Maximum Containment Level Goal (MCLG) should be reduced by at least ten-fold.^[5]

DNA theory Adkins et al. proposed that fluoride induces mitochondrial damage and affects mitochondrial DNA.^[6] Cai et al. found evidence that Bcl2 DNA hydroxymethylation could be involved.^[7]

Free radical production A paper by Russell Blaylock ^[8] published by an anti-fluoride organization called the International Society for Fluoride Research in New Zealand speculated that fluoride acts by free radical production and excitotoxicity. This is inconsistent with the chemical properties of fluoride, which doesn't participate in oxidation-reduction chemistry. The author proposes that fluoride acts by forming a complex with aluminum and speculates that it activates protein kinase C (PKC) and also increases lead levels. This isn't as crazy as it sounds: there's a paper claiming that PKC mediates fluoride-induced apoptosis and one claiming that PKC-θ mediates fluoride-induced immune system abnormalities. Overall, though, this article is insufficiently objective.

Agalakova and Nadei said more research is needed. This is a paywalled article, so we can ignore this one.^[9]

Autism A multi-decade, large population-based, hypothesis-testing statistical study ^[13] found that the rates of autism spectrum disorder (ASD) were increased by 6.25× in the fluoride treated group (N=25662) vs the unexposed group (n=2509). This was highly statistically significant. However, there were also huge differences in health status of the two groups, with malnutrition being 31.75x higher, disturbances in tooth eruption 12.7x higher, and lead toxicity being 2.8x higher, indicating substantial differences between the groups. According to their estimate, adjusting for these differences reduced the odds ratio from 6.25 to 5.55. Differences in IQ, which they call intellectual disability (ID), were smaller but also statistically significant. The odds ratio increased from 2.02 to 3.84 when corrected for all the other changes.

The finding of so many other big differences in the two populations suggests that there could be many more differences that were not measured and not looked for. If so, it would be impossible to adjust the statistics adequately. It's a conundrum: if everything is different in the two groups, what should you adjust for? You have to assume ID and ASD could be related, so you can't adjust for that. But what about all the other things?

The authors are noted for finding other improbable correlations, all with big odds ratios: ASD correlating with precocious puberty, asthma correlating with dental amalgam, and MMR vaccine correlating with seizure disorder, so it may be advisable to wait for confirmation on the fluoride study.

Other effects One group hypothesized that fluoride could induce glioblastoma.^[10] A variety of reproductive effects have also been claimed.

Summary

Hydrofluoric acid is well known to be highly toxic even in small quantities. Its toxic effect comes mainly from depletion of calcium and magnesium and from hyperkalemia arising from release of fluoride, which affects calcium-dependent potassium channels. This is why treatment often involves administering calcium. Symptoms typically include cardiac abnormalities ^[11].

Cell culture treatment of embryonic mouse cortical neurons showed evidence of oxidative stress when NaF was added at 50 mg/L.^[12] Offspring of pregnant mice given the same concentration in drinking water showed reduced frequency and amplitude of miniature excitatory post-synaptic currents (mEPSCs), indicating impaired synaptic function.^[12] At this dose—33 times higher than the proposed limits in drinking water and 12.5 times higher than the current standard of 4 mg/L—it was possible to see not just subtle learning impairment but measurable electrophysiological effects, conflicting with claims that the MoE is over 50.

So far the evidence for the levels found in treated water suggests a low but real risk, but the research is inconclusive with some articles dismissing risk using tendentious arguments and some overstating it. Obviously, more research is needed, but what is also needed is more understanding of inorganic chemistry by both researchers and the public. Ironically, Cotton and Wilkinson's 1350-page Advanced Inorganic Chemistry, though well worth reading, says almost nothing about fluoride ion. Maybe that will soon change.

apr 10 2025, 9:09 am. updated apr 11 2025, 4:07 am