Fluoride

WTR

2.0 mg/L where EPA requires your utility to warn parents

The only fluoride line that changes what we tell you. Below it the number is a design setting, and what we say depends on where in that range you sit.

Contested, tiered honestly

Stasis Home Research Every claim on this page is graded against peer-reviewed health research and linked to its primary source.Updated August 2026

Fluoride is the one item in this library put into water on purpose, for a documented dental benefit, at a level agencies choose. So there is no lower-is-better frame here, and what is true of your water depends on the number. At or below 1.0 mg/L you are at or near the level public-health agencies target and well under every safety limit. Between 1.0 and 2.0 mg/L we tell you how far above the target you are. Above 2.0 mg/L, EPA requires your utility to tell parents that children under nine should be given another source of drinking water.

Fluoride: what is it, and where does it come from?

Fluoride is the anion of the element fluorine, and it reaches a tap two ways, often blended. Some is natural background dissolving out of fluoride-bearing minerals in local groundwater. The rest is added on purpose: community water fluoridation, started in the U.S. in 1945, doses public supplies toward a set concentration to reduce tooth decay, and roughly 200 million Americans receive water treated that way. In one Southern California region, for example, the imported surface water most districts buy is fluoridated to about 0.7 mg/L while local groundwater runs closer to 0.3 to 0.5 mg/L, so blended tap fluoride lands a little under the target. That decision was made for a whole community by a state or local government, and nobody asked you. That is a plain description of how the policy works, not an accusation, and it is part of why this topic carries the charge it does.

Fluoride: why does it matter?

The benefit is real, and its modern size is smaller than the numbers most people have heard. The Community Preventive Services Task Force recommends community water fluoridation on strong evidence, and its pooled result is expressed in percentage points rather than as a percent reduction: a median decrease of 15.2 percentage points in caries after fluoridation began, and a median increase of 25.1 percentage points in the share of people free of caries. Those studies mostly predate universal fluoride toothpaste. The 2024 Cochrane update found that starting fluoridation today may produce a slightly greater reduction in decayed teeth and a slightly greater share of caries-free children, with smaller effect sizes than the pre-1975 studies and low-certainty evidence. Low certainty and a smaller effect still mean an effect. The dose-banded harm is dental fluorosis, a change in enamel formed while teeth are developing. EPA's own rulemaking record found no moderate to severe fluorosis at or below 0.6 mg/L and severe fluorosis consistently observed at or above 2.5 mg/L. The number at the top of the range comes from a different source: the 2006 National Research Council review found severe enamel fluorosis in roughly 10 percent of children at or near 4.0 mg/L, which is why that review concluded the 4.0 mg/L goal is not adequately protective. At the levels fluoridated systems actually deliver, fluorosis occurs but is predominantly very mild or mild, which is faint white flecking. At high chronic dose, fluoride accumulating in bone produces skeletal fluorosis, which is the harm the federal 4.0 mg/L limit exists to prevent and which is extremely rare in the United States. Then there is neurodevelopment, which is where the noise is, and burying it would be as dishonest as leading with it. The federal government's own toxicology program concluded in 2024, with moderate confidence, that higher fluoride exposures are consistently associated with lower IQ in children, at drinking-water concentrations above the WHO guideline value of 1.5 mg/L. Its qualifiers travel with it every time: that confidence rests primarily on exposures higher than what optimally fluoridated U.S. water delivers, no study of fluoride and IQ has been done in children in the United States, and below that 1.5 mg/L anchor the agency says the association remains unclear. The honest summary, and it is as far as we go: at the concentrations most American households actually drink, nobody has shown a cognitive effect, and nobody has shown there isn't one either, because the study that would settle it in a U.S. population has not been done.

The Community Preventive Services Task Force recommends community water fluoridation on strong evidence, reporting a median decrease of 15.2 percentage points in caries after fluoridation began and a median increase of 25.1 percentage points in the share of people free of caries. Those are percentage points, not a percent reduction, and they are not the same quantity.Community Preventive Services Task Force, Dental Caries: Community Water Fluoridation (recommendation April 2013; median decrease of 15.2 percentage points in caries, median increase of 25.1 percentage points caries-free)
The 2024 Cochrane review found that starting community water fluoridation today may lead to a slightly greater reduction in decayed teeth and a slightly greater share of caries-free children, with smaller effect sizes than pre-1975 studies and low-certainty evidence.Iheozor-Ejiofor et al., Cochrane Database of Systematic Reviews 2024;10(10):CD010856 (contemporary fluoridation effect sizes smaller than pre-1975 studies; post-1975 dmft mean difference 0.24, low certainty)
EPA's rulemaking record found no moderate to severe dental fluorosis at or below 0.6 mg/L and severe dental fluorosis consistently observed at or above 2.5 mg/L. Above 2 mg/L and up to 4 mg/L, a community water system must notify its customers that children drinking the water may develop cosmetic discoloration of their permanent teeth and that children under nine should be provided with an alternative source.40 CFR 141.208, special notice for exceedance of the fluoride secondary standard (notice required above 2 mg/L and not above 4 mg/L; children under nine should be provided alternative sources), with the dose bands from EPA's 1985 rulemaking record at 50 FR 20170
The U.S. National Toxicology Program concluded in 2024, with moderate confidence, that higher fluoride exposures are consistently associated with lower IQ in children at drinking-water concentrations above the WHO guideline of 1.5 mg/L. That confidence rests primarily on exposures higher than optimally fluoridated U.S. water, no such study has been conducted in children in the United States, and NTP states the association at lower exposures remains unclear.U.S. National Toxicology Program, Monograph 08: Fluoride Exposure, Neurodevelopment and Cognition, August 2024 (moderate confidence above the WHO 1.5 mg/L guideline; no U.S. studies; associations at lower exposure remain unclear)
Drinking water accounts for roughly 40 to 70 percent of total fluoride intake for children between six months and fourteen years, so removing it at the tap does not remove it from toothpaste or food.U.S. Public Health Service recommendation, Public Health Reports 2015;130(4):318-331 (0.7 mg/L recommended concentration; caries reduction plateaued at 0.7 to 1.2 mg/L; drinking water is 40 to 70 percent of total fluoride intake for children)
Activated alumina is the fluoride-specific adsorptive medium, with removal reported up to 99 percent or more, and reverse osmosis rejects dissolved inorganic ions including fluoride. Granular activated carbon is for organics, volatile compounds, and taste and odor, not for small inorganic ions.US EPA, Overview of Drinking Water Treatment Technologies (activated alumina for fluoride removal, up to 99 percent or more; reverse osmosis removes many inorganics; granular activated carbon is for organics and taste and odor)

Fluoride: what do we grade it against?

Fluoride has a health-based level but no enforceable legal limit. The health-based level for Fluoride, U.S. optimal target (mg/L) is 0.7 (the concentration U.S. public-health agencies set as optimal in 2015 for systems that fluoridate; a target, not a limit, and caries reduction plateaued between 0.7 and 1.2 mg/L), from PHS 2015, Public Health Reports 130(4):318-331. The table below lists every level on this page next to the source that set it.

Where the health research draws the line, versus the legal limit.
Contaminant	Health-based level	Legal limit	Source (health-based)
Fluoride, U.S. optimal target (mg/L)	0.7 the concentration U.S. public-health agencies set as optimal in 2015 for systems that fluoridate; a target, not a limit, and caries reduction plateaued between 0.7 and 1.2 mg/L	none set	PHS 2015, Public Health Reports 130(4):318-331
Fluoride, California Public Health Goal (mg/L)	1.0 a health-based goal set with no feasibility considerations. It is also where our copy stops reassuring and starts naming the distance, because above it the reassurance sentence is no longer true at the top of the band	none set	California SWRCB, MCL/DLR/PHG table, September 2025 (PHG set 1997)
Fluoride, WHO guideline value (mg/L)	1.5 WHO's health-based guideline value. It is also the concentration the U.S. National Toxicology Program anchored its neurodevelopment review at	none set	WHO, Guidelines for Drinking-water Quality, fluoride fact sheet
Fluoride, dental-fluorosis line (mg/L)	2.0 EPA's secondary standard, set on dental fluorosis in children. The level itself is not enforceable, but above it a community water system is federally required to notify customers and to tell them children under nine should be given an alternative source	2.0 California's enforceable MCL, effective 1998-04 (22 CCR 64431). The federal 2.0 is a notice trigger, not a legal limit	40 CFR 143.3 and 40 CFR 141.208 (health basis and notice); California SWRCB MCL table (state limit)
Fluoride, federal enforceable MCL (mg/L)	4.0 EPA's health goal sits at the same 4.0 and exists to prevent crippling skeletal fluorosis. The 2006 National Research Council review concluded unanimously that it is not adequately protective	4.0 federal MCL, effective 1987-10-02. It applies only to community water systems, so it is not a limit on a school or workplace system	40 CFR 141.62(b)(1) and 40 CFR 141.51
Measured example: Southern California tap (mg/L)	~0.5 to 0.7 verified Orange County districts land at or just under the optimal target: imported surface water fluoridated to about 0.7, local groundwater about 0.3 to 0.5, blended	none set	Orange County district Consumer Confidence Reports, 2026

Health-based levels come from peer-reviewed research and government risk scientists working without cost constraints, while legal limits are enforceable compromises. Your report grades to the health column.

What helps against Fluoride?

4 approaches below reduce Fluoride in a home, ordered from the most direct fix to the smallest change in habit. The most direct is Reverse osmosis (point-of-use, NSF/ANSI 58). Each entry also names what it cannot do, because no single fix covers every exposure route.

Direct fixes

Reverse osmosis (point-of-use, NSF/ANSI 58)
Rejects fluoride at the kitchen tap, reported at 80 to 90 percent for cellulose acetate membranes and up to 95 percent for thin-film composite. It is the same under-sink unit already used for PFAS, chromium-6, nitrate and radionuclides, so if you want fluoride out and need RO anyway you get it in one device.
Fluoride reduction is an optional claim under NSF/ANSI 58, where the mandatory test is total dissolved solids. Check that the specific unit lists fluoride.

Bigger retrofits

Activated alumina (NSF/ANSI 53 or WQA S-200)
The fluoride-specific adsorptive medium and the standard municipal answer, relatively undisturbed by competing sulfate, nitrate or chloride. The targeted option if you want fluoride reduced without a full RO system.
It needs a narrow pH window, roughly 5.5 to 6.5, and regeneration or cartridge exchange, which in a home means a service relationship rather than a set-and-forget filter.

Free and behavioral

Activated carbon (pitcher or block)
Improves taste and odor by adsorbing organic chemicals and chlorine.
Carbon does not remove fluoride. Fluoride is a small dissolved inorganic ion that passes straight through a carbon block, the same limit that applies to chromium-6. A pitcher filter is not reducing your fluoride, whatever you may assume.
Read your Consumer Confidence Report before deciding anything
On a fluoridated system the tap number sits near the target because that is where it is dosed, and your utility prints it every year. Knowing whether you are at 0.6 mg/L or 1.8 mg/L is the whole decision here, because what is true of your water is different in each case.

Fluoride is the only water item we do not anchor to a lower-is-better frame, and the only one where the failure we are most worried about is over-alarm rather than under-alarm. Calling a system's intended, in-range dose a finding would be a fear we manufactured. So we band it, and we say a different true thing in each band. At or below 1.0 mg/L, California's health goal: your water is at or near the level public-health agencies target to reduce tooth decay and well under every safety limit. Removal there is a preference we will fulfill, not a health upgrade, and it is worth knowing that drinking water is only part of total fluoride intake, so taking it out of the tap does not take it out of toothpaste or food. Above 1.0 and up to 2.0 mg/L, the reassurance sentence stops being true at the top of that band, so we retire it and name the distance instead: above the 0.7 mg/L optimal concentration and above California's 1.0 mg/L health goal, below the 2.0 mg/L level where EPA requires a utility to notify parents. That is not a violation and we do not call it a finding. Above 2.0 mg/L, removal is a real recommendation, and it is EPA's instruction rather than our opinion: children under nine should be given an alternative source of drinking water or water treated to remove the fluoride. The rest of the honest answer is the part people actually come here for. The benefit is real and its modern size is measured with low certainty. The findings on higher exposures and children's cognition are real too, and they sit above the levels most U.S. systems deliver, where the evidence gets thin rather than reassuring. Fluoride is in most public water because a state or local government decided to add it for the population it serves, and nobody asked you. If you would rather not drink it, that is a legitimate preference and we will fulfill it. We are not going to tell you your water is dangerous to make the sale, and we are not going to argue the public-health question at you either. Naturally occurring or deliberately added, there is no villain in this.

Common questions about Fluoride

Is the fluoride in my tap water harmful?

It depends on the number, and a single blanket answer here would be wrong. At or below 1.0 mg/L you are at or near the level public-health agencies target and well under every safety limit; nothing here is a health finding. Between 1.0 and 2.0 mg/L, that reassurance stops being accurate at the top of the range, so the honest framing is the distance: above the 0.7 mg/L optimal and above California's 1.0 mg/L goal, below the 2.0 mg/L notice level. Above 2.0 mg/L, EPA requires your utility to tell you that children under nine should be given an alternative source. Your annual Consumer Confidence Report prints your number.

Why is fluoride in my water on purpose, and who decided that?

A state or local government did, for the population it serves. Community water fluoridation began in the U.S. in 1945 and is administered as a public-health program delivered through a utility, so the decision was made at the community level rather than household by household, and there is no consent step at the tap. That is worth saying plainly rather than pretending fluoride arrived the way every other analyte in this library did. The reason it was added is a documented dental benefit: fluoride incorporates into enamel and promotes remineralization, and the Community Preventive Services Task Force recommends fluoridation on strong evidence.

How much does fluoridation actually reduce cavities?

Less than the number most people have heard, and measured with less certainty. The Task Force's pooled result is a median decrease of 15.2 percentage points in caries after fluoridation began and a median increase of 25.1 percentage points in the share of people free of caries, from studies that mostly predate universal fluoride toothpaste. Those are percentage points, which are a different quantity from a percent reduction, and converting one into the other would be a fabrication. The 2024 Cochrane review found that starting fluoridation in a population that already brushes with fluoride toothpaste produces a smaller effect, rated low-certainty. A smaller, less certain effect is still not zero.

What does the research actually say about fluoride and children's IQ?

The federal government's own toxicology program concluded in 2024, with moderate confidence, that higher fluoride exposures are consistently associated with lower IQ in children, at water concentrations above the WHO guideline value of 1.5 mg/L. That is a real finding from a serious institution and we are not going to bury it. The qualifiers are equally real and travel with it: the confidence rests primarily on exposures higher than optimally fluoridated U.S. water delivers, no study of fluoride and IQ has been performed in children in the United States, and below 1.5 mg/L the agency says the picture remains unclear. So at the levels most American households actually drink, nobody has shown a cognitive effect, and no study that would settle the question in a U.S. population has been done. It is also not a reason we will ever give you to buy a filter.

What is dental fluorosis, and should I be worried about it?

It is a change in tooth enamel formed while permanent teeth are developing in childhood, and it is not reversible and not progressive once the teeth have formed. The dose bands are the best-documented numbers in this subject: EPA's record found no moderate to severe fluorosis at or below 0.6 mg/L, and severe fluorosis consistently observed at or above 2.5 mg/L. At the levels fluoridated systems deliver, fluorosis occurs but is predominantly very mild or mild, which looks like faint white flecking rather than staining. The 2.0 mg/L federal notice level exists for exactly this endpoint, which is why crossing it triggers a warning aimed at parents of young children.

Will my carbon pitcher or block filter remove fluoride?

No. Fluoride is a small dissolved inorganic ion and it passes straight through a carbon block, the same limitation that applies to chromium-6. If you want fluoride out, reverse osmosis or activated alumina will do it; carbon will not, and you should not assume a carbon filter is reducing it.

How do I find out how much fluoride is in my water?

Your annual Consumer Confidence Report prints the number. On a fluoridated system it lands near the target by design, so expect something near 0.7 mg/L, and where a district blends fluoridated surface water with lower-fluoride groundwater the blended tap comes in a little under that. A private well is a different case: no utility report covers it, and naturally high fluoride is a groundwater story, so if you have a well you have to test to know.