Fluoride is the ion form of fluorine, an element found in rock, soil and water. In the body it settles in bones and teeth. In the mouth, even at low concentrations, it helps tooth enamel resist acid.
Water is the main source for most people. The US Public Health Service says drinking water supplies 40% to 70% of the fluoride that children aged 6 months to 14 years take in. For adults it is about 60%, and food, tea and swallowed toothpaste add the rest.
Here is what most people do not know: fluoride is not officially an essential nutrient. EFSA, the EU’s food safety agency, concluded in 2013 that it is not essential. The WHO (World Health Organization) says it may be essential, but that this has not been shown unequivocally. Both still recognize its benefit for teeth, and the US and EU set intake targets for that reason.
The other surprise is tea. The WHO puts the fluoride in dry tea leaves at 100 mg per kg on average. In EFSA’s 2025 assessment of European food surveys, tea provided more than 10% of dietary fluoride in 66 surveys. In one group of very elderly people it provided up to 63%.
Fluoride is also one of the most argued-over minerals. This page reports what each source says and no more. That covers Cochrane reviews of toothpaste and water, the dose-related marks fluoride can leave on teeth and bones, and children’s IQ. On IQ, the US National Toxicology Program, EFSA and its critics read the same studies differently.
It is general information, not medical advice. Dosing for children is a decision for a dentist or doctor.
Health benefits of fluoride
The top findings from 30 studies, strongest evidence first
What the stars mean
- Strong. Several randomized trials, pooled in a meta-analysis, point the same way, or regulators (FDA or EFSA) have approved the claim.
- Moderate. Good randomized trials support it, but there are fewer of them, or their results vary.
- Limited. Small, short or single trials, or mixed results. Promising, but it needs confirmation.
Population and lab studies get no stars: they show a link or a mechanism, not proof.
How we rate evidence →- ★★★Fewer cavities in children from fluoride toothpasteIn children, fluoride toothpaste cut new decay on permanent teeth by 24% against paste without fluoride (70 trials). Paste of 1450 to 1500 ppm (parts per million) did slightly better than 1000 to 1250; 1700 to 2800 did no better than that.
- ★★★EU health claim: fluoride maintains tooth mineralizationThe EU allows one fluoride claim: it contributes to maintaining tooth mineralization (the minerals in teeth). It does not say more fluoride helps, and the same panel found no cause-and-effect link with normal bone.
- ★★★Fluoride drug: denser spine, fractures not clearly fewerIn postmenopausal women taking it as a drug for osteoporosis, spine bone density rose 8.1% after 2 years. New spine fractures were not clearly lower, and non-spine fractures were more common at 4 years.
- ★★Less decay with fluoride varnish and rinsesIn children, varnish put on by a dental professional cut decay by 43% in permanent teeth and 37% in baby teeth. Supervised fluoride mouthrinse cut decay in permanent teeth by 27%.
- ★★Lower-fluoride toothpaste, less fluorosis in young childrenFluorosis means marks on teeth. In two trials of 1,968 children, 550 ppm paste had 0.75 times the risk of fluorosis of 1000 ppm paste.
All 27 studies, strongest evidence first
The EU allows one fluoride claim: it helps maintain tooth mineralisation
EFSA’s 2009 panel found a cause-and-effect link between dietary fluoride and maintenance of tooth mineralisation, and Regulation (EU) No 432/2012 lists the claim. A food may carry it only if it is at least a source of fluoride (Regulation (EC) No 1924/2006). As a rule, that is 15% of the 3.5 mg EU Reference Intake per 100 g (0.525 mg). For a drink it is 7.5% per 100 ml (about 0.26 mg), under Regulation (EU) No 1169/2011. The same panel found no established cause-and-effect link with maintaining normal bone. The claim says fluoride contributes to the function, not that more fluoride helps.
Fluoride toothpaste cut new decay by 24% against non-fluoride paste
This Cochrane review pooled 70 randomized and quasi-randomized trials with 42,300 children up to 16, each lasting at least a year. New decay on permanent tooth surfaces was 24% lower, with clear differences between trials. The effect was larger with more starting decay, stronger paste and more frequent or supervised brushing, and water fluoridation did not change it. The review found little information on baby teeth or on fluorosis.
1000 to 1250 ppm toothpaste cut decay in children; stronger added little
This Cochrane network meta-analysis pooled 96 randomized trials of at least a year. Against non-fluoride paste, 1000 to 1250 ppm (parts per million) gave a standardized mean difference of -0.28 in 55 studies. Pastes of 1450 to 1500 ppm gave -0.36 in four. Both results were high- or moderate-certainty, and 1450 to 1500 ppm did slightly better than 1000 to 1250 ppm. Pastes of 1700 to 2800 ppm did no better than 1450 to 1500 ppm. In adults, 1000 or 1100 ppm cut decayed surfaces by 0.53 in three studies of 2,162 people. The authors say strength for young children should be balanced against the risk of fluorosis.
As a drug, fluoride thickened spine bone but did not prevent fractures
This Cochrane review (searched to December 1998) covered 11 trials in 1,429 postmenopausal women taking fluoride for osteoporosis. Lumbar spine density rose 8.1% after 2 years and 16.1% after 4. New spine fractures were not clearly lower (0.87 times at 2 years, 0.9 at 4). Non-spine fractures were more common at 4 years (1.85 times). So were stomach and bowel side effects, especially at high doses and in non slow-release form. These were drug doses for osteoporosis, not the amounts in water or food.
Fluoride varnish cut decay 43% in adult teeth, 37% in baby teeth
This Cochrane review of 22 trials randomized 12,455 children to varnish applied by a dental professional, or to placebo or no treatment. Using results nearest three years, decay fell 43% on permanent tooth surfaces (13 trials) and 37% on baby teeth (10 trials). Results varied a lot between trials (heterogeneity 75% for permanent teeth). The authors rate the evidence moderate quality because most trials were at high risk of bias, and side effects were little reported.
Supervised fluoride mouthrinse cut decay in permanent teeth by 27%
This Cochrane review of 37 trials with 15,813 children looked at rinsing, mostly in school programs. Decay fell 27% (35 trials, 15,305 children), on moderate-quality evidence. Almost all children used a sodium fluoride rinse, daily at 230 ppm or weekly to every two weeks at 900 ppm. Most trials (28) were at high risk of bias, and few reported side effects.
Lower-fluoride toothpaste lowered fluorosis risk in children
This Cochrane review covered 43 studies (three randomized trials) of fluoride on the teeth of children under six and fluorosis of the permanent teeth. In two of these trials (1,968 children), 550 ppm against 1000 ppm gave 0.75 times the risk, and 440 ppm against 1450 ppm 0.72. Brushing started at 24 months in one trial and 12 months in the other. The authors judge that brushing with 1000 ppm or more from age one to two until five to six probably raises the chance. Evidence on when to start, how often and how much paste was low or very low certainty, and most evidence points to mild fluorosis.
Fluoride drug studies: denser bones, no overall drop in fractures
This meta-analysis of 25 studies of fluoride as an osteoporosis drug found spine density up 7.9% and hip density up 2.1%. Fractures were 0.8 times the odds for spine and 0.8 for non-spine, gaps that could be chance. In a subgroup, doses of 20 mg a day of fluoride equivalents or less went with fewer spine (0.3) and non-spine fractures (0.5). Higher doses did not (1.3 and 1.5, which could be chance), and this subgroup result needs confirming. For comparison, the US adequate intake for adults is 3 to 4 mg a day.
In adults, fluoride prevented about 0.29 decayed crown surfaces a year
This meta-analysis covered 20 studies in adults aged 20 and over, published in 1980 or later. It mixed self-applied fluoride, professionally applied fluoride and water fluoridation, and also found about 0.22 fewer decayed root surfaces a year. For water fluoridation alone, decay fell 27%. The 2024 Cochrane review of water fluoridation found no eligible studies in adults.
Fluoride tablets, drops or lozenges went with 24% less decay
This Cochrane review of 11 trials in 7,196 children compared supplements with none, for permanent teeth (three studies for this comparison). The effect on baby teeth was unclear, and there was no difference against topical fluorides or other preventive measures. The authors rate the trials as weak evidence, with one at high risk of bias and ten at unclear risk. They found limited information on side effects.
Fluoride tablets in pregnancy did not prevent baby-tooth decay
This Cochrane review found only one eligible trial of fluoride supplements taken by pregnant women. Decay risk was 1.46 times at age 3 (938 children) and 0.84 at age 5 (798), both very low-quality evidence that could be chance. Fluorosis at age 5 was similar with tablets and with placebo. The authors conclude there is no evidence that supplements in pregnancy prevent decay in the offspring.
Cochrane 2024: water fluoridation’s modern benefit is small and uncertain
This update of the 2015 review below included 157 non-randomized studies, so all evidence was downgraded for risk of bias. In the two studies after 1975 (2,908 children), starting fluoridation went with 0.24 fewer decayed baby teeth per child, about a quarter of a tooth. This was low-certainty evidence whose range includes no benefit, and caries-free children rose by 4 percentage points. Older evidence was very low certainty, the review could not judge stopping fluoridation, and it found no studies of decay in adults. At 0.7 ppm, about 12% had fluorosis of aesthetic concern and about 40% any level. One author is co-director of the Colgate-Palmolive Dental Health Unit at the University of Manchester.
Cochrane 2015: water fluoridation cut decayed baby teeth by 35%
This review of 155 non-randomized studies (107 with usable data) found 1.81 fewer decayed baby teeth per child and 1.16 fewer permanent ones (26%). The two main analyses covered 44,268 and 78,764 children, and caries-free children rose 15% for baby teeth and 14% for permanent teeth. Of the studies, 71% were done before 1975, when fluoride toothpaste became widely used. The authors limit their confidence because of high risk of bias and how well old findings apply today, and call the fluorosis evidence limited. The 2024 update above replaced it.
After Calgary stopped fluoridation, more 7-year-olds had decay than in Edmonton
This before-and-after study compared Calgary, seven years after it stopped fluoridating, with Edmonton, which kept fluoridating. Grade 2 school exams in 2018 and 2019 covered 2,649 children in Calgary and 2,600 in Edmonton. Unadjusted, decay in baby teeth affected 64.8% against 55.1%, about 65% against 55%. The gap held after adjustment and had widened over time, while results for permanent teeth were less consistent at that age. The authors call the findings consistent with harm from stopping, not proof of one.
In US teens, fluorosis rose from 29.7% to 61.3% in a decade
Two national NHANES surveys of 16- and 17-year-olds (875 in all) compared 2001 to 2002 with 2011 to 2012. Very mild or worse dental fluorosis on the modified Dean index rose, with 3.85 times the odds after adjustment. Very mild is the mildest recorded category, and the study cannot say which fluoride source drove the rise. The Cochrane review above separates fluorosis of aesthetic concern from fluorosis of any level.
Lifetime water fluoride was not linked to hip fracture
This case-control study, comparing people with and without hip fracture, covered people aged 50 and over in Cleveland, England. Of 914 cases and 1,196 controls identified, 514 and 527 were interviewed. Estimated lifetime exposure ranged from 0.15 to 1.79 ppm, and 0.9 ppm or more gave odds of 1.0 after adjustment, meaning no difference. Fractures were strongly linked to low body-mass index and inactivity, and the authors see a low hip fracture risk at about 1 ppm. The EFSA 2025 review discusses a later Swedish cohort that linked higher total intake to hip fracture.
Skeletal fluorosis was about twice as common at the highest exposure
This dose-response meta-analysis of 23 studies found 2.05 times the risk at the highest fluoride exposure, and 2.73 times for drinking water. Risk rose almost in a straight line up to 5 mg/L in water and 2.5 mg/L in urine, with no clear rise above. The risk of moderate to severe forms increased from 5.00 mg/L in water. The authors ask for better-quality studies.
A woman drinking heavy tea for decades developed skeletal fluorosis
This case report describes a 48-year-old American woman who drank 1 to 2 gallons of brewed tea a day for more than 30 years. She had very dense bones (spine Z-score +9.9), severe bone and joint pain and a stooped spine, but normal kidney function. Fluoride was high in her blood, urine, nails and tea, and her pain resolved over months after she stopped. The authors note that about 20 mg of fluoride a day for 20 years is expected to cause symptoms. It is a single case and does not show how common this is.
NTP: water above 1.5 mg/L is linked to lower IQ
The US National Toxicology Program (NTP) concluded with moderate confidence that higher fluoride exposure is linked to lower IQ in children. It reviewed 72 IQ studies in children, and 18 of the 19 high-quality ones reported an inverse link. These were three prospective cohorts and 15 cross-sectional (snapshot) studies in five countries, and forty-six of 53 low-quality studies also did. It rates evidence in adults as low confidence and the animal and mechanism data as unclear. It says more studies are needed at lower exposure, does not address dental benefits, and shows an association, not proof of cause.
Meta-analysis: 1.63 fewer IQ points per 1 mg/L more urinary fluoride
This 2025 analysis is by NTP staff and a contractor who wrote the monograph above. Of its 74 studies (64 cross-sectional, 10 cohort), 52 had high risk of bias. Group-level studies showed lower IQ (standardized mean difference -0.45; n = 20,932), and 31 water studies gave a dose-response of -0.15. The link was null below 1.5 mg/L overall, and stayed inverse in the low-risk-of-bias studies. In 13 individual-level studies, IQ fell 1.63 points per 1 mg/L of urinary fluoride, and 1.14 points in low-risk studies. The authors say data are limited and uncertain when exposure is estimated from water alone below 1.5 mg/L.
More fluoride in pregnancy urine went with lower child IQ in Mexico
This Mexico City cohort followed 299 mother-child pairs, and the mothers’ mean urinary fluoride was 0.90 mg/L. Each 0.5 mg/L more in pregnancy went with 3.15 lower cognitive scores at age 4 and 2.50 lower IQ at ages 6 to 12. The authors say these exposures were in the general range of other samples of pregnant women and non-pregnant adults.
In Canada, higher maternal urinary fluoride went with lower IQ in boys
This birth cohort followed 512 mother-child pairs in six Canadian cities, where 41% lived with fluoridated water, and tested IQ at 3 to 4 years. Each 1 mg/L rise in mothers’ urinary fluoride went with 4.49 lower IQ in boys, with no clear link in girls (2.40). Women in fluoridated areas had higher urinary fluoride (0.69 against 0.40 mg/L). A 1 mg higher daily fluoride intake from water and drinks went with 3.66 lower IQ in boys and girls together.
More tap-water fluoride went with lower non-verbal IQ in Canadian children
In 398 children from the same Canadian cohort, water fluoride came from municipal reports and IQ was tested at 3 to 4 years. Each 0.5 mg/L more fluoride went with 9.3 lower performance (non-verbal) IQ in formula-fed babies and 6.2 lower in breastfed babies. The link held after accounting for fluoride in the womb, and 0.5 mg more fluoride from formula went with 8.8 points lower.
In Spain, higher maternal urinary fluoride went with higher scores in boys
This INMA birth cohort tested 316 to 248 mother-child pairs with Bayley scales at age 1 and McCarthy scales at age 4. There was no link at age 1. For boys at age 4, each unit of urinary fluoride went with a higher General Cognitive Index (beta 15.4), and girls showed none. The authors say this conflicts with some earlier studies and call for confirmation at low water fluoride.
In Denmark, mothers’ urinary fluoride was not clearly linked to IQ
The Odense Child Cohort followed 837 mother-child pairs, with urinary fluoride of 0.08 to 3.04 mg/L (median 0.52), and tested IQ at age 7. Per doubling of fluoride, the IQ estimate was 0.08 points, a result that could be chance. Pooled with two earlier cohorts from Mexico and Canada (more than 1,500 pairs), the link was there. That gave a benchmark concentration of 0.45 mg/L (lower limit 0.28), which the authors read as an approximate threshold. Authors include investigators of the Mexican and Canadian cohorts, so the pooled result is not an independent check.
In New Zealand, early-life fluoride showed no clear IQ differences
This Dunedin study followed people born in 1972 and 1973 to age 38, with 95.4% retention. It looked at water fluoridation, fluoride toothpaste and fluoride tablets. IQ was tested repeatedly between ages 7 and 13 and at 38, and results held after adjusting for sex, social class, breastfeeding and birth weight. The authors suggest earlier links between very high fluoride and low IQ may reflect confounding (other factors mixed in), such as urban or rural living.
German review: evidence does not support brain harm at European levels
This narrative review covered 23 studies since 2012, 21 of which linked high fluoride to lower intelligence, plus animal and cell work. The authors point to confounding in most of them: social class, residence, breastfeeding, low birth weight, mothers’ intelligence and other toxicants. They report a margin of exposure of 50 to 210 between animal no-effect levels and the adequate intake. The review predates the NTP monograph and the 2025 meta-analysis, and the NTP authors answered common critiques in a 2025 viewpoint.
Foods high in fluoride
For most people, water and tea matter more than any solid food. EFSA found measurable fluoride in tea, fish, vegetables, cereals and foods for infants. In its 2025 assessment, grains, milk and dairy products, tea and drinking water each gave over 10% of dietary fluoride in many European surveys. Tea reached up to 63% in one group of very elderly people.
The amount in a cup depends on the water. A Sri Lankan survey measured an average of 2.68 mg/L in 2% infusions of loose black tea. Over half of the fluoride leached out in the first brew. In the Azores, green tea and black tea were both tested. The fluoride in the infusions was clearly linked to the fluoride in the water used, and brewing longer raised it.
Bottled water varies widely. In a survey of 105 bottled waters, distilled water averaged under 0.01 ppm and spring water 0.08 ppm. Mineral water averaged 0.32 ppm and water with added fluoride 0.63 ppm. The FDA limits bottled water with added fluoride to 0.7 mg/L.
Bottled water with no added fluoride may not exceed 1.4 to 2.4 mg/L if packaged in the US, depending on local air temperature. If imported, the limit is 1.4 mg/L.
Food varies too. A US study of foods eaten by 2-year-olds found some oils and dairy products with no detectable fluoride. Some processed meats, fish and fruits had more than 3.0 µg per g. It estimated 412 µg a day from food and drink, without toothpaste.
Our USDA ranking is dominated by drinks, and the values probably follow the water used to make them. Per serving, coffee leads with 215 µg in 1 cup (237 g). Beer follows with 157 µg per can and red wine with 154 µg per 5 fl oz glass. Grapes (11.8 µg per cup) and semi-skimmed milk (8.4 µg per cup) come far behind.
Per 100 g, red wine has 105 µg and coffee 90.7 µg. A cup of coffee is about 6% of the 3.5 mg EU Reference Intake, and no food on our list reaches 20%. Brewed tea is not in the top ten, so use the published measurements above for it.
Per serving. No daily amount is set, so bars compare each food with the top food.
Per 100 g. No daily amount is set, so bars compare each food with the top food.
Per serving. Bars show the share of the EU Reference Intake (3,500 µg a day).
Per 100 g. Bars show the share of the EU Reference Intake (3,500 µg a day).
Per 100 kcal: the same calories of each food, for a fair comparison. Bars compare each food with the top food.
- 1.
- 2.
- 3.
- 4.
- 5.
- 6.
- 7.
- 8.
- 9.
- 10.
- 1.
- 2.
- 3.
- 4.
- 5.
- 6.
- 7.
- 8.
- 9.
- 10.
- 1.
- 2.
- 3.
- 4.
- 5.
- 6.
- 7.
- 8.
- 9.
- 10.
Sources
- FoodData Central. US Department of Agriculture, Agricultural Research Service. Values as shown in the ranking above.
- Updated consumer risk assessment of fluoride in food and drinking water including the contribution from other sources of oral exposure. EFSA Scientific Committee. EFSA J. 2025.
- Exposure assessment of fluoride intake through commercially available black tea from areas with high incidences of chronic kidney disease with undetermined origin in Sri Lanka. Chandrajith R, Bhagya S, Diyabalanage S, et al. Biol Trace Elem Res. 2022.
- Safety evaluation of fluoride content in tea infusions consumed in the Azores. Linhares DPS, Garcia PV, Amaral L, et al. Biol Trace Elem Res. 2017.
- Fluoride content of bottled water and its implications for the general dentist. Quock RL, Chan JT. Gen Dent. 2009.
- Fluoride in the diet of 2-years-old children. Martinez-Mier EA, Spencer KL, Sanders BJ, et al. Community Dent Oral Epidemiol. 2017.
- 21 CFR 165.110: bottled water. US Food and Drug Administration.
- Fluoride: chemical fact sheet. World Health Organization, Guidelines for drinking-water quality.
How fluoride works
The biochemistry below comes from reviews, official reports and lab and physiology work, not from trials of people taking extra fluoride. Four ideas cover most of it, and then comes what has been seen in people.
Fluoride works mostly on the surface of the tooth. The body absorbs it fast and stores it in bones and teeth. Too much while teeth form, or for years in bone, does harm. And the dose matters more than the word “fluoride”.
It works mostly on the surface of the tooth
For decades it was unclear whether fluoride had to be swallowed and built into tooth crystals. A history by ten Cate and Buzalaf describes how, between 1960 and 1990, researchers agreed that the effect is mainly topical (on the surface). It comes from fluoride in the fluids of the mouth, not from fluoride inside the enamel. Low, steady levels below 1 mg/L are enough.
During an acid attack, fluoride sticks to the surface of the enamel crystals and slows dissolving. When the acid fades, the fluid is highly saturated with respect to fluorhydroxyapatite, which speeds the re-forming of mineral. The new mineral includes fluoride and leaves out carbonate, which makes enamel more resistant.
EFSA’s claim opinion puts it plainly: fluoride-substituted apatite is less soluble, so it resists acid better. Like calcium and phosphorus, fluoride is part of the same mineral crystal.
Two side notes: in test tubes, fluoride at plaque levels slows acid production and the building of sticky glucan in the decay bacterium S. mutans. But Buzalaf and colleagues say what this means in the mouth is still unclear. Second, fluoride built into the pits of permanent first molars before they come through the gums does seem to protect them. But the main route is topical.
How fluoride protects enamel, in three steps. A simplified diagram, not to scale.
The body absorbs it fast and stores it in bones and teeth
After a dose, plasma fluoride rises quickly because the stomach absorbs it. Most of what is not absorbed there is taken up in the small intestine. The blood level peaks within 20 to 60 minutes, then falls as calcified tissues take fluoride up and the urine removes it.
Unlike some minerals, the blood level is not held steady by the body. Plasma fluoride varies with intake, deposition in hard tissues and excretion.
The WHO adds that water-soluble fluorides are rapidly and almost completely absorbed. It says calcium, magnesium, phosphorus and aluminum can bind them and reduce this. It also finds no difference in absorption between natural and added fluoride in water, and virtually no storage in soft tissues. Fluoride in teeth and bone can be released after exposure ends or falls.
Where swallowed fluoride goes. A simplified diagram, not to scale.
Too much while teeth form leaves marks, and too much for years affects bone
Dental fluorosis can only happen while teeth are forming, so it is a childhood exposure problem. DenBesten and Li explain that the lower front permanent teeth finish hardening at about 2 to 3 years of age. The last teeth finish much later. Mild fluorosis shows as white, opaque enamel from a poorly mineralized layer under the surface.
In severe fluorosis the surface pits and picks up brown stains. Lab and animal work suggests that fluoride delays the removal of an enamel protein, amelogenin, in a dose-related way. But the authors say more research is needed to know which cell changes matter.
In bone and teeth, fluoride can take the place of hydroxyl groups in the mineral crystal. EFSA’s claim opinion notes that fluoride is built into bone and can raise bone density. But there is evidence for a biphasic effect on bone strength and fracture risk, with no way to define a clear dose-response.
The WHO says skeletal fluorosis may be seen where drinking water has 3 to 6 mg of fluoride per liter, particularly with high water intake. The crippling form usually develops only above 10 mg/L. It cites an IPCS (International Programme on Chemical Safety) conclusion of clear excess risk of skeletal effects at 14 mg a day. That body saw suggestive evidence from about 6 mg a day.
The numbers for water, side by side
Water numbers get mixed up, so here are the main ones. The US Public Health Service recommends 0.7 mg/L for water systems that add fluoride. That replaced a 1962 range of 0.7 to 1.2 mg/L that depended on air temperature.
The WHO guideline value of 1.5 mg/L is a health-based ceiling for any water, not a target. The WHO notes it is higher than the 0.5 to 1.0 mg/L usually used for fluoridation.
In the US, the EPA has a legal maximum (MCL) of 4.0 mg/L. Its secondary level of 2.0 mg/L is described as a goal for water quality. The EU limit in its drinking water directive is 1.5 mg/L, and EFSA’s 2025 risk assessment suggests that this limit is not protective enough.
Fluoride in drinking water, in mg/L, drawn to scale. A recommendation, a guideline and a legal limit are different things. The gray part of the skeletal fluorosis bar shows the start of the scale.
What has been seen in people
The mechanisms fit the human evidence on teeth. Toothpaste, varnish and rinses, which put fluoride on the tooth surface, cut decay in children in trials by about 24% to 43%. Water fluoridation also works partly through the surface, but its measured benefit is smaller in modern studies, when toothpaste is common.
The dose-related harms fit the mechanisms too. Fluorosis in children grows with intake and with toothpaste strength. Skeletal fluorosis appears at the high intakes seen with natural fluoride in water or very heavy tea drinking.
The brain-development findings have no agreed mechanism. The NTP says the animal and mechanism data are unclear, so that part rests on population studies. Those can show an association but not a cause. All of these studies are among the cards above.
Sources
- Mechanisms of action of fluoride for caries control. Buzalaf MAR, Pessan JP, Honório HM, ten Cate JM. Monogr Oral Sci. 2011.
- Fluoride mode of action: once there was an observant dentist. ten Cate JM, Buzalaf MAR. J Dent Res. 2019.
- Fluoride metabolism. Buzalaf MAR, Whitford GM. Monogr Oral Sci. 2011.
- Chronic fluoride toxicity: dental fluorosis. DenBesten P, Li W. Monogr Oral Sci. 2011.
- EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA). Scientific opinion on the substantiation of health claims related to fluoride and maintenance of tooth mineralisation and maintenance of bone. EFSA J. 2009.
- Fluoride: chemical fact sheet. World Health Organization, Guidelines for drinking-water quality.
- U.S. Public Health Service recommendation for fluoride concentration in drinking water for the prevention of dental caries. U.S. Department of Health and Human Services Federal Panel on Community Water Fluoridation. Public Health Rep. 2015.
- 40 CFR 141.62: maximum contaminant levels for inorganic contaminants and 40 CFR 143.3: secondary maximum contaminant levels. US Environmental Protection Agency.
- Updated consumer risk assessment of fluoride in food and drinking water including the contribution from other sources of oral exposure. EFSA Scientific Committee. EFSA J. 2025.
- Fluoride exposure and neurodevelopment and cognition. National Toxicology Program, US National Institute of Environmental Health Sciences.
How much fluoride do you need?
Adults need 3 to 4 mg of fluoride a day, according to the US adequate intake: 4 mg for men and 3 mg for women. EFSA’s adequate intake is 0.05 mg per kg of body weight per day from all sources, for children from age 6 months and for adults. The table below shows the daily amounts.
Both bodies set an adequate intake, a target used when data are too thin for a firmer target, not a recommended allowance. The US panel said the data were not enough to set an average requirement. It based its value on intakes that reduce tooth decay without causing moderate fluorosis.
EFSA said that fluoride is not an essential nutrient, so no requirement can be defined. It based its value on a water level of 1 mg/L. At that level, decay prevention neared its maximum and only 10% of people had mild fluorosis. These values count fluoride from all sources, including toothpaste that is swallowed.
In pregnancy and breastfeeding, the US adequate intake is 3 mg a day at ages 14 to 50, the same as for other women. EFSA applies the same 0.05 mg per kg. In 2025 EFSA also set a “safe level of intake” of 3.3 mg a day from all oral sources to protect the fetus. It extended this to other adults and to children over 8, and it is a ceiling, not a target (see the safety section).
The US has no separate value for older adults: the adequate intake is the same from age 19 on. For babies it is 0.01 mg a day up to 6 months, the amount in human milk, and 0.5 mg from 7 to 12 months. For children it is 0.7 mg at ages 1 to 3 and 1 mg at ages 4 to 8. It is 2 mg at ages 9 to 13 and 3 mg at ages 14 to 18.
The US Daily Value list has no entry for fluoride, and food labels may declare it voluntarily. The EU Reference Intake for labels is 3.5 mg.
Fluoride: recommended intake per day
US: Adequate Intake (NASEM). EU: Adequate Intake (EFSA).
| Age and sex | US (AI) | EU (AI) |
|---|---|---|
| Men, 31–50 years | 4,000 µg | 3,400 µg |
| Women, 31–50 years | 3,000 µg | 2,900 µg |
All ages, pregnancy and breastfeeding
| Age and sex | US (AI) | EU (AI) |
|---|---|---|
| Infants 0–6 months | 10 µg | None set |
| Infants 7–11 months | None set | 400 µg |
| Infants 7–12 months | 500 µg | None set |
| Children 1–3 years | 700 µg | 600 µg |
| Children 4–6 years, female | None set | 900 µg |
| Children 4–6 years, male | None set | 1,000 µg |
| Children 4–8 years | 1,000 µg | None set |
| Children 7–8 years, female | None set | 1,400 µg |
| Children 7–8 years, male | None set | 1,500 µg |
| Children 9–10 years, female | None set | 1,400 µg |
| Children 9–10 years, male | None set | 1,500 µg |
| Females 9–13 years | 2,000 µg | None set |
| Males 9–13 years | 2,000 µg | None set |
| Children 11–14 years, female | None set | 2,300 µg |
| Children 11–14 years, male | None set | 2,200 µg |
| Females 14–18 years | 3,000 µg | None set |
| Males 14–18 years | 3,000 µg | None set |
| Children 15–17 years, female | None set | 2,800 µg |
| Children 15–17 years, male | None set | 3,200 µg |
| Adults 18+ years, female | None set | 2,900 µg |
| Adults 18+ years, male | None set | 3,400 µg |
| Females 19–30 years | 3,000 µg | None set |
| Males 19–30 years | 4,000 µg | None set |
| Females 31–50 years | 3,000 µg | None set |
| Males 31–50 years | 4,000 µg | None set |
| Females 51–70 years | 3,000 µg | None set |
| Males 51–70 years | 4,000 µg | None set |
| Females >70 years | 3,000 µg | None set |
| Males >70 years | 4,000 µg | None set |
| Pregnancy 14–18 years | 3,000 µg | None set |
| Pregnancy | None set | 2,900 µg |
| Pregnancy 19–30 years | 3,000 µg | None set |
| Pregnancy 31–50 years | 3,000 µg | None set |
| Lactation 14–18 years | 3,000 µg | None set |
| Lactation | None set | 2,900 µg |
| Lactation 19–30 years | 3,000 µg | None set |
| Lactation 31–50 years | 3,000 µg | None set |
EU and US age bands differ; "None set" means that body sets no value for the group. Label reference: EU Reference Intake 3,500 µg.
Fluoride deficiency: signs and who is at risk
Fluoride deficiency is not a disease with a list of symptoms. The US panel could not set an average requirement, and EFSA says fluoride is not essential. So the “shortfall” that matters is more tooth decay.
In older studies in the Cochrane reviews of water fluoridation, children in unfluoridated areas had more decay than those in fluoridated areas. The modern gap is smaller. The WHO says low concentrations protect against decay in children and adults, with the minimum water level for this effect around 0.5 mg/L.
Who is at risk of low exposure? People whose water has little fluoride and who do not use fluoride toothpaste. In EFSA’s European water data, more than 86% of samples held less than 0.3 mg/L, and more than 97% less than 0.7 mg/L.
The ADA (American Dental Association) panel names the group that may need a supplement. It is children at high risk of decay whose main drinking water is low in fluoride. We found no blood test or intake cutoff for fluoride deficiency. A dentist judges decay risk and asks about local water, and a doctor or dentist decides about extra fluoride.
Sources
- Dietary Reference Intakes for calcium, phosphorus, magnesium, vitamin D, and fluoride. Institute of Medicine, National Academies Press. 1997. Chapter 8: Fluoride.
- Updated consumer risk assessment of fluoride in food and drinking water including the contribution from other sources of oral exposure. EFSA Scientific Committee. EFSA J. 2025. It also describes the 2013 EFSA opinion on dietary reference values for fluoride (EFSA J 2013;11(8):3332).
- Fluoride: chemical fact sheet. World Health Organization, Guidelines for drinking-water quality.
- U.S. Public Health Service recommendation for fluoride concentration in drinking water for the prevention of dental caries. U.S. Department of Health and Human Services Federal Panel on Community Water Fluoridation. Public Health Rep. 2015.
- Evidence-based clinical recommendations on the prescription of dietary fluoride supplements for caries prevention. Rozier RG, Adair S, Graham F, et al. J Am Dent Assoc. 2010.
- 21 CFR 101.9(c)(5): nutrition labeling of food, fluoride. US Food and Drug Administration.
- Regulation (EU) No 1169/2011, Annex XIII. Nutrient reference values.
Forms of fluoride: absorption and effects
Fluoride reaches you in water, food, toothpaste, rinses and tablets, and the compound behind it differs. In water, the WHO says, the fluoride in the final water is always present as fluoride ions. That holds whether it comes from rock or from added chemicals. The US Public Health Service describes the breakdown (hydrolysis) of fluorosilicic acid at drinking-water pH as essentially 100%.
Some commenters on the 2015 US recommendation noted a lack of specific safety data on silicofluoride compounds. The panel answered that they dissociate virtually completely and are comparable to sodium fluoride. In toothpaste and rinses, the FDA lists three active ingredients: sodium fluoride, sodium monofluorophosphate and stannous fluoride.
The honest bottom line: the cavity trials mostly pool fluoride products by strength in ppm, not by compound. We found no head-to-head trial of sodium fluoride, monofluorophosphate and stannous fluoride on cavities. The mouthrinse trials in Health benefits used sodium fluoride almost entirely, at 230 or 900 ppm. The Cochrane toothpaste, varnish and supplement reviews do not name the compound in their abstracts.
What the studies below compare is how much fluoride is absorbed, and, for stannous fluoride, gum health. So claims that one compound prevents more decay than another have no trial behind them.
| Form | Share of the mineral by weight | What studies show | Source |
|---|---|---|---|
| Sodium fluoride (toothpaste, rinse, varnish, tablets) | 25 mg of sodium fluoride supplied 11.3 mg of fluoride (about 45%) | The FDA lists 0.188% to 0.254% sodium fluoride for toothpaste with 850 to 1,150 ppm fluorine and 0.05% as a treatment rinse. Nearly all rinse trials used it (27% less decay). In the paste trial below it was fully soluble. Taken fasting by postmenopausal women it was absorbed as fully as monofluorophosphate with calcium. | 1, 2, 3, 7 |
| Sodium monofluorophosphate (toothpaste) | 100 mg supplied 13.2 mg of fluoride (about 13%) | The FDA lists 0.654% to 0.884% for 850 to 1,150 ppm and 1.153% for 1,500 ppm pastes. With chalk (calcium carbonate), about 20% of its fluoride was insoluble and bioavailability was lower than from sodium fluoride and silica paste, and lower again after aging. With calcium carbonate and breakfast, it was equally bioavailable as fasting sodium fluoride. | 1, 2, 3 |
| Stannous fluoride (toothpaste, gel, rinse) | Not stated | The FDA lists 0.351% to 0.474% for toothpaste with 850 to 1,150 ppm and a 0.4% gel. In six trials it gave slightly better gum scores than other fluoride pastes. We found no cavity comparison. | 1, 4 |
| Fluorosilicic acid and sodium fluorosilicate (added to water) | Not stated | In water the fluoride is present as fluoride ions, and the hydrolysis of fluorosilicic acid is essentially 100%. The recommended level for added fluoride is 0.7 mg/L. Critics said safety data on these compounds were lacking. | 5, 6 |
| Fluoride naturally in water, tea and food | Not stated | The WHO finds no difference in absorption between natural and added fluoride in drinking water. Calcium, magnesium, phosphorus and aluminum can bind fluoride and reduce absorption. | 5 |
| Professional varnish and supervised rinse | Not stated | Varnish: 43% less decay in permanent teeth and 37% in baby teeth (22 trials). Rinse: 27% less decay (37 trials). The compound is not named for varnish in the abstract. | 7, 8 |
| Tablets, drops and lozenges (supplements) | Not stated | A 2011 Cochrane review found 24% less decay in permanent teeth against no supplement, on weak evidence. The ADA says supplements should go only to children at high risk whose main water is low in fluoride. The FDA calls these products unapproved and recommends limiting them to children aged three and older at high risk. | 9, 10, 11 |
Here are the head-to-head and form-specific studies, strongest evidence first, 3 in all. We found no trial of the compounds against each other on tooth decay. The side effects of the forms differ mainly by dose. A swallowed paste delivers the fluoride that is soluble in it. The fluorosis risk follows the amount swallowed, as the cards in Health benefits show.
All 3 studies, strongest evidence first
Stannous fluoride toothpaste gave slightly better gum scores in six trials
This systematic review found 16 randomized trials, of which six could be pooled, comparing stannous with other fluoride toothpastes. The Gingival Index difference was -0.14, but the Modified Gingival Index difference (-0.30) could be chance. The authors note a lack of consistent research to compare stannous fluoride formulas with other fluoride toothpastes. This review is about gums, and we found no trial comparing stannous with sodium fluoride on cavities.
Partly insoluble fluoride paste delivered less fluoride to the body
In this blind crossover trial, 20 adults swallowed 49.5 µg of fluoride per kg of body weight from each paste or from water. The sodium monofluorophosphate and chalk paste, labeled 1,450 µg F/g, had about 20% insoluble fluoride (40% after accelerated aging). The sodium fluoride and silica paste, labeled 1,100 µg F/g, was fully soluble. Absorption was lower for the first paste, and fell further after aging. The authors say fluorosis risk from swallowed paste should be estimated from soluble, not total, fluoride.
In 10 women, monofluorophosphate and sodium fluoride were absorbed equally
In this crossover trial, postmenopausal women took 25 mg of sodium fluoride (11.3 mg of fluoride) fasting. They also took 100 mg of sodium monofluorophosphate with 1,250 mg of calcium carbonate (13.2 mg of fluoride) with breakfast. The monofluorophosphate peaked sooner (1.4 against 2.5 hours) and higher (260 against 200 ng/ml). But total exposure (1,711 against 1,202 µg/liter/hour) did not clearly differ, and this drug-dose study in a small group says nothing about caries.
Sources
- 21 CFR 355.10: anticaries active ingredients. US Food and Drug Administration.
- Bioavailability of fluoride in postmenopausal women: comparative study between sodium fluoride and disodium monofluorophosphate-calcium carbonate. Lioté F, Bardin C, Liou A, et al. Calcif Tissue Int. 1992.
- Fluoride gastrointestinal absorption from Na2FPO3/CaCO3- and NaF/SiO2-based toothpastes. Falcão A, Tenuta LM, Cury JA. Caries Res. 2013.
- Comparison of new formulas of stannous fluoride toothpastes with other commercially available fluoridated toothpastes. Clark-Perry D, Levin L. Int Dent J. 2020.
- Fluoride: chemical fact sheet. World Health Organization, Guidelines for drinking-water quality.
- U.S. Public Health Service recommendation for fluoride concentration in drinking water for the prevention of dental caries. U.S. Department of Health and Human Services Federal Panel on Community Water Fluoridation. Public Health Rep. 2015.
- Fluoride mouthrinses for preventing dental caries in children and adolescents. Marinho VC, Chong LY, Worthington HV, Walsh T. Cochrane Database Syst Rev. 2016.
- Fluoride varnishes for preventing dental caries in children and adolescents. Marinho VC, Worthington HV, Walsh T, Clarkson JE. Cochrane Database Syst Rev. 2013.
- Fluoride supplements (tablets, drops, lozenges or chewing gums) for preventing dental caries in children. Tubert-Jeannin S, Auclair C, Amsallem E, et al. Cochrane Database Syst Rev. 2011.
- Ingestible fluoride drug products. US Food and Drug Administration. Guidance dated October 31, 2025.
- Evidence-based clinical recommendations on the prescription of dietary fluoride supplements for caries prevention. Rozier RG, Adair S, Graham F, et al. J Am Dent Assoc. 2010.
Is fluoride safe? Supplements, dose and interactions
At the amounts in water, food and toothpaste used as directed, fluoride is part of ordinary life. But the margin is narrower for young children than for adults.
The US tolerable upper intake level (UL), which covers all sources, is 10 mg a day from age 9, including pregnancy and breastfeeding. It is 0.7 mg for babies up to 6 months and 0.9 mg for 7 to 12 months. It is 1.3 mg at ages 1 to 3 and 2.2 mg at ages 4 to 8. For children it is set to avoid enamel fluorosis, and for older people to avoid skeletal fluorosis.
EFSA’s 2025 opinion sets fluorosis-based ULs of 1.0 mg a day for infants and 1.6 mg for toddlers. For children aged 4 to 8 it is 2.0 mg, and for adults and children over 8 it could not set a UL. Instead it gave a safe level of intake of 3.3 mg a day, based on brain-development effects seen with water above 1.5 mg/L. It says this also protects thyroid and bone.
EFSA found that exposure from average EU water was below these values for all age groups. But at the 95th percentile of intake with high-end water, exposure was above them for all groups except adolescents.
- Dental fluorosis. It follows the dose during tooth formation. At 0.7 ppm in water, Cochrane estimates about 12% have fluorosis of aesthetic concern and about 40% any level. The WHO gives 12% to 33% for mild fluorosis at 0.9 to 1.2 mg/L. It notes that mild forms may only be seen by a specialist. Toothpaste matters: brushing with 1000 ppm or more from age one or two probably raises the chance, and lower-strength paste lowered it in two trials. The FDA label for fluoride toothpaste says to supervise children under 6 so they swallow little. For children under 2, it says to ask a dentist or doctor.
- Bone and thyroid. Skeletal fluorosis is tied to long, high intake, such as naturally high-fluoride well water or, rarely, gallons of tea a day. EFSA says links with thyroid function and bone mineralisation have been seen at water concentrations above 1.5 mg/L. We found no trial showing harm to bone at the 0.7 mg/L recommended for fluoridation. The 2024 Cochrane update called the evidence on bone very low certainty.
- Children’s brain development. The US National Toxicology Program concluded with moderate confidence that exposure above 1.5 mg/L in drinking water is linked to lower IQ. It says there are insufficient data to determine whether 0.7 mg/L is. A 2025 meta-analysis by the same team found an inverse link in urine-based studies, and null results for water alone below 1.5 mg/L overall. EFSA reached a matching split: adverse effects above 1.5 mg/L, and evidence below it not consistent enough for risk assessment. Critics point to confounding. A New Zealand cohort found no IQ difference, and a Spanish cohort found higher scores in boys. All of this is association, and none of it shows cause.
- Supplements and swallowed products. The ADA recommends fluoride tablets only for children at high decay risk whose main water is low in fluoride. It prescribes by age and local water level, and we do not reprint the schedule. In October 2025 the FDA called ingestible fluoride products for children unapproved and recommended limiting them to children aged three and older at high risk. It cited lack of proven benefit on baby teeth and emerging safety concerns. If a child swallows more toothpaste than used for brushing, the label says to get medical help or call a poison control center. The WHO gives about 1 mg of fluoride per kg of body weight as the lowest dose seen to cause acute poisoning.
- Interactions and who should ask. Calcium, magnesium, phosphorus and aluminum reduce fluoride absorption. We found no drug interaction. Ask a dentist or pediatrician if you feed a baby formula made with fluoridated water, or if a child under 6 uses fluoride products. Ask too if your well water is high in fluoride or you drink large amounts of tea.
How to get the right amount of fluoride
- Brush with fluoride toothpaste. In 96 trials, 1000 to 1250 ppm paste cut decay in children, and the 2019 review says to balance strength against fluorosis for young children. Supervise small children so they spit.
- Know your water. Ask your water supplier or dentist for the local fluoride level, because decay prevention and fluorosis both depend on it. The US recommendation is 0.7 mg/L, and most EU samples in EFSA’s data held less than 0.3 mg/L.
- Watch tea and formula. A cup is a small dose, but gallons a day over decades led to skeletal fluorosis in one case. For infant formula, the ADA panel said mixing with optimally fluoridated water can continue while keeping the fluorosis risk in mind.
Sources
- Dietary Reference Intakes for calcium, phosphorus, magnesium, vitamin D, and fluoride. Institute of Medicine, National Academies Press. 1997. Chapter 8: Fluoride.
- Updated consumer risk assessment of fluoride in food and drinking water including the contribution from other sources of oral exposure. EFSA Scientific Committee. EFSA J. 2025.
- U.S. Public Health Service recommendation for fluoride concentration in drinking water for the prevention of dental caries. U.S. Department of Health and Human Services Federal Panel on Community Water Fluoridation. Public Health Rep. 2015.
- Fluoride: chemical fact sheet. World Health Organization, Guidelines for drinking-water quality.
- Fluoride exposure and neurodevelopment and cognition. National Toxicology Program, US National Institute of Environmental Health Sciences.
- Addressing critiques of the evidence linking fluoride and children’s IQ. Taylor KW, Eftim SE, Sibrizzi CA, et al. Ann Glob Health. 2025.
- 21 CFR 355.50: labeling of anticaries drug products. US Food and Drug Administration.
- Ingestible fluoride drug products. US Food and Drug Administration. Guidance dated October 31, 2025.
- Evidence-based clinical recommendations on the prescription of dietary fluoride supplements for caries prevention. Rozier RG, Adair S, Graham F, et al. J Am Dent Assoc. 2010.
- Evidence-based clinical recommendations regarding fluoride intake from reconstituted infant formula and enamel fluorosis. Berg J, Gerweck C, Hujoel PP, et al. J Am Dent Assoc. 2011.
- Chronic fluoride toxicity: dental fluorosis. DenBesten P, Li W. Monogr Oral Sci. 2011.
History of fluoride
The story begins with stained teeth. In the early 20th century, dentists in Colorado saw brown mottled enamel they could not explain. Ten Cate and Buzalaf write that it took 30 years to show that fluoride caused the mottling. It was also linked to fewer cavities in people who drank it.
The dentist Frederick McKay spent decades on the “Colorado brown stain”. In the 1930s and early 1940s, H. Trendley Dean’s team at the US National Institutes of Health published studies suggesting about 1 mg/L of fluoride went with far fewer cavities.
Grand Rapids, Michigan, began adding fluoride to its water on January 25, 1945, in a controlled experiment. The Cochrane review dates the start of community water fluoridation to 1945 in the USA. By 1975 about 49% of the US population drank fluoridated water, and by 2012 about 67%, according to the US Public Health Service.
Today community water fluoridation is practiced in about 25 countries, Cochrane says. Researchers took another 30 years to agree how fluoride works, from 1960 to 1990.
The policy has been revised as the evidence changed. In 2015 the US Public Health Service replaced a temperature-based range with a single 0.7 mg/L. The Cochrane review was updated in 2024 with a smaller modern effect. The US NTP published its monograph on neurodevelopment in 2024, and EFSA issued new safe levels in 2025.
Sources
- Fluoride mode of action: once there was an observant dentist. ten Cate JM, Buzalaf MAR. J Dent Res. 2019.
- U.S. Public Health Service recommendation for fluoride concentration in drinking water for the prevention of dental caries. U.S. Department of Health and Human Services Federal Panel on Community Water Fluoridation. Public Health Rep. 2015.
- Water fluoridation for the prevention of dental caries. Iheozor-Ejiofor Z, Walsh T, Lewis SR, et al. Cochrane Database Syst Rev. 2024.
- Water fluoridation. Wikipedia, history section.
Frequently asked questions about fluoride
How much fluoride is too much?
For adults the US upper limit is 10 mg a day from all sources, set to avoid skeletal fluorosis. Young children have far lower limits, mainly against dental fluorosis: 1.3 mg at ages 1 to 3 and 2.2 mg at 4 to 8. EFSA’s 2025 opinion sets 1.6 and 2.0 mg for those ages. For water, the WHO guideline is 1.5 mg/L and the US legal maximum is 4.0 mg/L. The US recommends 0.7 mg/L for added fluoride, and EFSA’s 2025 safe level for adults is 3.3 mg a day.
Does water fluoridation prevent cavities?
In older studies it did: the 2015 Cochrane review found 35% fewer decayed baby teeth, mostly in studies from before 1975. The 2024 update, which looks at modern studies, finds a smaller and uncertain benefit of about a quarter of a tooth per child. Its range includes no benefit. A Calgary study found more decay in 7-year-olds seven years after fluoridation stopped. All of this evidence is non-randomized, and the Cochrane authors say decisions should weigh toothpaste use, diet and other factors.
Does fluoride lower children’s IQ?
The US National Toxicology Program concluded, with moderate confidence, that exposure above 1.5 mg/L in drinking water is linked to lower IQ in children. It says there are insufficient data to judge 0.7 mg/L, the US level for fluoridation. A 2025 JAMA Pediatrics meta-analysis by the same team found 1.63 fewer IQ points per 1 mg/L more urinary fluoride. It found a null result for water alone below 1.5 mg/L overall. A New Zealand cohort found no difference, and critics cite confounding. These are associations, not proof of cause.
Is fluoride an essential nutrient?
Not by EFSA’s view. Its 2013 opinion concluded that fluoride is not essential, but it still set an adequate intake because of the benefit to teeth. The WHO says it may be essential, but that this has not been demonstrated unequivocally. The US sets an adequate intake of 4 mg for men and 3 mg for women, based on preventing decay. We found no other deficiency syndrome described.
Which foods and drinks contain fluoride?
Water is the biggest source, then tea, which averages about 100 mg per kg in the dry leaf according to the WHO. Fish, grains, vegetables and dairy contribute too. In our USDA ranking, coffee has 215 µg in a cup, beer 157 µg per can and red wine 154 µg per glass. These values follow the fluoride in the water used. Most foods contain much less.
Do fluoride supplements have side effects?
The main risk for children is dental fluorosis if they get more than they need. A Cochrane review found limited information on side effects. Stomach and bowel side effects were more common in adults taking fluoride tablets as an osteoporosis drug. The ADA says tablets are for children at high decay risk whose main water is low in fluoride. The FDA recommends limiting them to children aged three and older at high risk. Ask a dentist or pediatrician about dosing.
Is stannous fluoride better than sodium fluoride?
We found no trial that compared them on cavities. A 2020 review of six trials found slightly better Gingival Index scores with stannous fluoride toothpaste (-0.14), but not on the Modified Gingival Index. The FDA lists both, along with sodium monofluorophosphate, as active ingredients for toothpaste against cavities.
Can I mix infant formula with fluoridated tap water?
An ADA expert panel suggested in 2011 that powdered or liquid concentrate formula can still be mixed with optimally fluoridated water. The panel added that parents should be aware of the risk of enamel fluorosis. The US Public Health Service notes a possible increased chance of mild fluorosis for babies fed only formula made with fluoridated water. It says parents may sometimes use low-fluoride bottled water. In one Canadian cohort, formula-fed children had lower performance IQ with more fluoride in tap water, an association that does not prove cause. Ask your pediatrician or dentist.