Evidence review
Fluoridated water versus toothpaste
Fluoride from drinking water can reach the tooth surface through saliva and direct contact. A 2024 review found that newer studies showed a smaller added benefit than older studies, but rated the modern estimate uncertain. Toothpaste use is one possible reason the additional benefit is harder to measure today.23
How swallowing fluoride ends up on your teeth
This part is genuinely elegant and it is often told wrong. Fluoride you swallow is absorbed into your blood. Your salivary glands then pull it back out of the blood and put it into saliva. Saliva bathes your teeth constantly. So a swallowed dose comes back around and reaches the tooth surface from the outside.3
That helps explain why water fluoridation can have a topical effect even though fluoride's main anticavity action happens at the tooth surface. The size of the added benefit today remains uncertain.23
The amounts are wildly different
Here is where the comparison gets uncomfortable for water. Saliva coming straight from the glands carries about 0.016 parts per million of fluoride where water is fluoridated, against 0.006 where it is not. Real, measurable, and very small.13
Toothpaste is around 1450 parts per million, applied directly to the tooth. That is roughly a hundred thousand times the concentration, sitting exactly where you want it.4
Water does get one thing back. Every drink briefly spikes the fluoride in your mouth by 100 to 1000 times, falling back over an hour or two. Sipping water through the day means repeated small top-ups. That is a real mechanism, and it is still much less fluoride than two minutes of brushing delivers.3
What changed in 1975
Fluoride toothpaste became much more common during the 1970s. Water fluoridation was then being measured alongside other fluoride exposures and changes in dental care, which makes its added effect harder to isolate in newer studies.2
The 2024 Cochrane review separated studies begun before and after 1975. Their estimates differ, although the newer estimate is less certain.2
What the 2024 review found
Cochrane updated its assessment of water fluoridation in 2024. Studies from before 1975 showed what it calls a clear and important effect: roughly two fewer decayed baby teeth per child.2
Studies since 1975 show about a quarter of one tooth. And the range around that estimate includes the possibility of no benefit at all. Cochrane rated the certainty of this evidence low, and for permanent teeth very low.2
Widespread fluoride toothpaste use is one plausible reason that the added benefit of fluoridated water is harder to detect in newer studies. The review cannot isolate that explanation from other changes in dental care and study design.
What that does and does not mean
Older studies found a substantial benefit. The newer estimate is smaller and uncertain. The review cannot tell us how much of that difference comes from toothpaste use, baseline cavity rates, other changes in care or the studies themselves.2
It also does not settle the policy question. Toothpaste only helps people who use it. Water reaches everyone on the supply, including people who do not brush regularly or cannot afford a dentist, which is the strongest remaining argument for it.
The short version
- Swallowed fluoride does reach your teeth, through saliva. The mechanism is real.
- The dose that arrives that way is tiny next to toothpaste, by a factor of thousands.
- Water fluoridation had a large effect when it was the only source, and a small and uncertain one now that almost everyone brushes with fluoride.
- Fluoride toothpaste has strong evidence for cavity prevention. The additional benefit from fluoridated water in today's setting is less certain.
Read the full evidence review The salivary route in numbers, the 1975 stratification, and what the 2024 Cochrane review actually reported.
Once fluoride's action was recognized as post-eruptive and topical, water fluoridation needed a new explanation, and it has a good one: ingested fluoride is secreted back into saliva and reaches the enamel surface from outside. That mechanism is sound. What has changed is its magnitude. Since fluoride toothpaste became near-universal around 1975, the measured marginal benefit of fluoridating water has fallen by roughly an order of magnitude, and the most recent Cochrane review puts the contemporary effect at about a quarter of a tooth with a confidence interval that crosses zero.
Key points
- Ingested fluoride is secreted into saliva by the salivary glands, so water fluoridation delivers fluoride to the tooth surface topically rather than structurally.
- Ductal saliva carries roughly 0.016 ppm fluoride in fluoridated areas against 0.006 ppm in non-fluoridated ones. Toothpaste is applied at around 1450 ppm.
- Each fluoride exposure transiently raises intraoral fluoride 100- to 1000-fold, returning to baseline within one to two hours.
- Cochrane 2024 stratified the evidence at 1975. Pre-1975 studies showed a clear and important effect; post-1975 studies show MD 0.24 dmft, 95% CI -0.03 to 0.52, low certainty.
- The review attributes the difference to the increased availability of fluoride toothpaste. This is diminishing returns on a successful intervention, not evidence that it never worked.
The salivary route
As covered in how fluoride works, the systemic model of fluoride action was abandoned by the end of the 1990s: the concentration achievable in developing enamel through ingestion is far too low to account for the observed caries reduction, and benefit tracks ongoing topical exposure rather than developmental history.3
That conclusion appeared to leave water fluoridation without a mechanism, and it did not. Fluoride absorbed from the gut enters plasma, and the salivary glands secrete it into ductal saliva at a concentration proportional to plasma fluoride. Saliva then bathes the dentition continuously, as described in saliva and oral health. Ingested fluoride therefore arrives at the enamel surface from the outside, which is exactly where the mechanism requires it.
There is a second, more direct route that gets less attention. Fluoridated water passes over the teeth while being drunk, which is a straightforward topical application, briefly and at low concentration, several times a day. The distinction between topical and systemic delivery is set out in the ADA's own summary of the two.5
The magnitudes, side by side
| Source | Concentration | Route | Duration |
|---|---|---|---|
| Ductal saliva, non-fluoridated area | about 0.006 ppm | Continuous baseline | Constant |
| Ductal saliva, fluoridated area | about 0.016 ppm | Continuous baseline | Constant |
| Any fluoride exposure, transient peak | 100 to 1000 times baseline | Intraoral | Returns to baseline in 1 to 2 hours |
| OTC fluoride toothpaste | 1000 to 1500 ppm | Direct topical application | Two minutes, plus reservoir effect |
Two things follow. The continuous baseline elevation from fluoridated water is small in absolute terms, though it is present at all times including overnight, when salivary flow and every other defense is at its lowest. And the transient peaks matter more than the baseline: drinking fluoridated water repeatedly through the day produces repeated small excursions, which is the same frequency-over-intensity logic that governs the reservoir effect described in how fluoride works.
Neither consideration closes a gap of this size. A dentifrice delivers vastly more fluoride, directly to the surface that needs it, and leaves a calcium fluoride-like reservoir behind. On mechanism alone one would predict toothpaste to dominate, and that is what the outcome data now show.
Why 1975 is the dividing line
Fluoride dentifrice moved from a minority product to the market default across the late 1960s and early 1970s. Cochrane's reviewers stratify their meta-analyses at 1975 for this reason: studies conducted before that date compare fluoridated against non-fluoridated water in populations with little other fluoride exposure, while studies after it compare the same thing in populations where nearly everyone is already brushing with fluoride.
These are not the same question. The first measures the effect of fluoride. The second measures the marginal effect of adding water fluoridation to fluoride toothpaste. Failing to separate them is the most common error in public argument on this subject, in both directions.
What Cochrane 2024 reported
The 2024 update of the Cochrane review of water fluoridation for the prevention of dental caries is the current best synthesis.2 Its contemporary findings, stated precisely:
| Outcome, studies after 1975 | Effect | Certainty |
|---|---|---|
| Decayed, missing, filled primary teeth (dmft) | MD 0.24 (95% CI -0.03 to 0.52) | Low |
| Permanent teeth (DMFT) | Reviewers "very uncertain of these findings" | Very low |
| Caries-free, primary dentition | 4 percentage point difference | Low |
| Caries-free, permanent dentition | 3 percentage point difference | Low |
Pre-1975 studies, by contrast, showed what the reviewers describe as a clear and important effect on the prevention of tooth decay, with an analysis of 5,708 children estimating around 2.1 fewer decayed primary teeth per child.2 The certainty attached to the older evidence is itself very low by modern standards, since those studies were largely uncontrolled before-and-after comparisons, but the effect size is an order of magnitude larger and not seriously disputed.
The reviewers attribute the contemporary attenuation to the increased availability of fluoride in toothpaste since 1975.
Reading this honestly
This finding is quoted in both directions by people who have not read it, so it is worth being exact about what it supports.
It does not show that water fluoridation never worked. The pre-1975 evidence points the other way, and the mechanism is sound. A public health measure whose marginal benefit falls as the underlying disease becomes less common and as a second intervention covers the same ground is exhibiting diminishing returns, which is what success looks like from the inside.
It does not show that fluoride is ineffective. Fluoride toothpaste remains one of the better-evidenced interventions in preventive dentistry, with a clear dose-response relationship above 1000 ppm,4 as covered in how fluoride works. If anything this review strengthens that conclusion, since the most plausible explanation for water's shrinking effect is that toothpaste is already doing the work.
It does not settle the policy question. Toothpaste benefits only those who use it consistently, and use is not evenly distributed. Water reaches everyone connected to the supply regardless of behavior, income or access to dental care, which is the strongest argument that remains for it and the one least addressed by an effect size averaged over a whole population. As noted in how fluoride works, average benefit and individual benefit are different quantities, and that applies to this review too.
Our interest here is obvious and worth naming. A company selling fluoride-free toothpaste benefits from a reader concluding that fluoride matters less than they thought. We would rather state the finding accurately and let it be less useful to us than overstate it. The measured contemporary effect of water fluoridation is small and uncertain. The measured effect of fluoride toothpaste is not.
Where the evidence is strong and where it is not
Strong. That fluoride is secreted into saliva in proportion to plasma concentration. That fluoride's caries-preventive action is topical and post-eruptive. The concentration figures, which are direct measurements.
Weak by the reviewers' own assessment. Essentially all of the contemporary water fluoridation outcome data, rated low to very low certainty. The pre-1975 evidence is weaker still in design terms, though larger in effect. Nobody has run a randomized trial of water fluoridation and nobody is going to.
Not addressed here. Fluoride's safety profile, which is covered in fluoride in drinking water. Cost effectiveness. And whether reduced marginal efficacy justifies changing policy, which depends on distributional questions this page does not attempt to weigh.
References
- US Department of Health and Human Services Federal Panel on Community Water Fluoridation. U.S. Public Health Service Recommendation for Fluoride Concentration in Drinking Water for the Prevention of Dental Caries. Public Health Reports. 2015;130(4):318-331. PMC4547570.
- Iheozor-Ejiofor Z, Walsh T, Lewis SR, et al. Water fluoridation for the prevention of dental caries. Cochrane Database of Systematic Reviews. 2024;10(10):CD010856. PMID 39362658; Cochrane Library.
- ten Cate JM. Current concepts on the theories of the mechanism of action of fluoride. Acta Odontologica Scandinavica. 1999;57(6):325-329. PMID 10777135.
- Walsh T, Worthington HV, Glenny AM, Marinho VCC, Jeroncic A. Fluoride toothpastes of different concentrations for preventing dental caries. Cochrane Database of Systematic Reviews. 2019;3(3):CD007868. PMID 30829399.
- American Dental Association. Fluoride: Topical and Systemic Supplements. ADA.
This article is educational and describes published research. It is not medical or dental advice, not a diagnosis, and not a claim about any product. We make a fluoride-free toothpaste and benefit from readers concluding fluoride matters less than they thought, which is a reason to read this page critically. Nothing here argues that fluoride toothpaste is ineffective.
Keep reading
Athena Naturals
Seven ingredients. Fluoride-free, no water, no preservatives, no surfactants. Made in small batches in Seattle.