Is Fluoride in Tap Water Safe? 157 Studies

Reviewed 2 September 2026. Twenty-eight records, each pulled and checked against Europe PMC before publication: two Cochrane reviews, five further systematic reviews (four of them with pooled estimates), four additional reviews including a scoping review of nineteen cessation studies, three narrative updates, one EFSA consumer risk assessment, one national longitudinal cohort plus a commentary on it, three critique and history pieces, three cessation studies, three infant-feeding exposure studies and two surveys of public knowledge. Every number below appears in the abstract of the record cited next to it; if a number is not there, it is not in this article. Written for a parent reading a water-quality report at the kitchen table and for a clinician answering the same question in a nine-minute visit. Not medical or dental advice.

The short answers

• How big is the practice? Community water fluoridation started in 1945 and over the past twenty years has reached over 400 million people; the 2024 Cochrane review counts it as practised in about 25 countries (Rugg-Gunn et al., Front Oral Health 2026-01-01, PMID 41958764) (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658). The countries named as covering a high proportion of their population are the USA, Australia, New Zealand, Ireland, Singapore, Hong Kong, Brunei, Malaysia, Oman, Chile, Gabon and Brazil — Ukraine is not among them, so for a patient arriving from there the question is usually where fluoride comes from rather than whether the tap adds it (Rugg-Gunn et al., Front Oral Health 2026-01-01, PMID 41958764).
• Does it reduce cavities? In the strongest contemporary measurement, yes but modestly: initiating fluoridation changed the primary-tooth decay index by a mean difference of 0.24 in favour of fluoridation (95% CI −0.03 to 0.52; P = 0.09; 2 studies, 2,908 children; low-certainty evidence), which Cochrane translates as roughly a quarter of a tooth and explicitly notes that the estimate “includes the possibility of benefit and no benefit” (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658). The proportion of caries-free children moved by 4 percentage points in the primary dentition (MD −0.04, 95% CI −0.09 to 0.01) and 3 points in the permanent dentition (MD −0.03, 95% CI −0.07 to 0.01) (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658).
• Why do other syntheses report a quarter of all cavities? Because they pool different comparisons. A 2026 meta-analysis of 74 studies (32 with usable data) found a standardised mean difference of −0.32 for DMFT and dmf (95% CI −0.48 to −0.17, I² = 96%) and lower caries prevalence in exposed people: OR 0.52 (95% CI 0.43 to 0.63) for permanent teeth and OR 0.60 (0.48 to 0.76) for deciduous teeth (Nascimento et al., JDR Clin Trans Res 2026-04-01, PMID 40574418). The JADA update covering 2014 to 2025 says most studies find roughly 25% less caries (Warren et al., J Am Dent Assoc 2026-04-01, PMID 41746238). A before-and-after change at the moment of initiation is a harder, cleaner question than a comparison of two communities that differ in many ways besides water — and Cochrane found that design difference plus a 1975 toothpaste boundary explain most of the gap (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658).
• What does it cost the enamel? At 0.7 ppm, Cochrane pooled data from 135 studies report fluorosis of aesthetic concern in approximately 12% of participants (95% CI 8% to 17%; 40 studies, 59,630 participants) and fluorosis of any level in approximately 40% (95% CI 35% to 44%; 90 studies, 180,530 participants) (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658). The clinical review of the same situation adds the part the number hides: prevalence is higher in fluoridated communities, but severity is mild to very mild and of little aesthetic concern (Rugg-Gunn et al., Front Oral Health 2026-01-01, PMID 41958764).
• Does it lower IQ? This is where the literature genuinely splits, and the split is about dose range, not about goodwill. The 2025 JAMA Pediatrics meta-analysis of 74 studies found an inverse association overall — pooled SMD −0.45 (95% CI −0.57 to −0.33) across 59 studies and 20,932 children — while stating that data and certainty are limited when exposure is estimated from drinking water alone below 1.5 mg/L (Taylor et al., JAMA Pediatr 2025-03-01, PMID 39761023). Meta-analyses restricted to fluoridated populations find nothing: pooled SMD 0.04 (95% CI −0.06 to 0.14; I² = 0%) in five studies conducted in fluoridated areas (Kumar et al., Community Dent Oral Epidemiol 2026-02-01, PMID 41174838), a regression coefficient of 1.01 with low heterogeneity in three studies on community fluoridation specifically (Marques et al., Cien Saude Colet 2026-02-01, PMID 41779589), and no negative association with adolescent IQ or adult cognition in a US state cohort (Warren et al., Proc Natl Acad Sci U S A 2026-04-01, PMID 41973917).
• Where is the line? Not at “fluoride yes or no” but at a concentration. The European Food Safety Authority’s 2025 assessment found human evidence of effects on the developing brain at drinking-water concentrations above 1.5 mg/L and concluded that evidence below 1.5 mg/L was not sufficiently consistent to draw conclusions; it set a safe total intake of 3.3 mg/day for pregnant women, and tolerable upper intake levels of 1.0 mg/day for infants, 1.6 for toddlers and 2.0 for children aged 4 to 8 (EFSA Scientific Committee et al., EFSA J 2025-07-01, PMID 40698337). A rapid review of 58 studies found no evidence of adverse effects below 1.0 mg/L and no convincing harm between 0.7 and 1.0 mg/L, while noting that study designs precluded causal inference (Zohoori et al., Adv Nutr 2025-12-01, PMID 41135657). EFSA’s sharpest sentence is not about fluoridation but about the ceiling: the current EU legal limit of 1.5 mg/L for drinking water is, in its view, not sufficiently protective (EFSA Scientific Committee et al., EFSA J 2025-07-01, PMID 40698337).
• What actually matters at home? The dose that parents control. Cochrane’s review of topical fluoride and fluorosis found moderate-certainty evidence that toothpaste concentration matters: 550 ppm versus 1000 ppm gave RR 0.75 (95% CI 0.57 to 0.99) and 440 ppm versus 1450 ppm gave RR 0.72 (95% CI 0.58 to 0.89) in children under six (Wong et al., Cochrane Database Syst Rev 2024-06-01, PMID 38899538). Its conclusion is blunt: children who brushed with 1000 ppm or more from one to two years until five to six years probably had an increased chance of fluorosis (Wong et al., Cochrane Database Syst Rev 2024-06-01, PMID 38899538). The water review reaches the same practical advice from the other direction — advice should focus on early use of fluoride toothpaste to reduce ingestion in infancy and early childhood (Rugg-Gunn et al., Front Oral Health 2026-01-01, PMID 41958764).
• What has nobody settled? Adults, cost with local numbers, and inequality. The Cochrane review states plainly that it found no eligible studies reporting caries outcomes in adults (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658). Only one post-1975 study in it reported on socioeconomic disparities, finding no evidence that deprivation modified the relationship — one study, not an answer (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658). The economic review could not pool its fifteen evaluations at all, only synthesise them qualitatively (Almutairi et al., Int Dent J 2026-08-01, PMID 42142407).

What fluoride does in the mouth, and in the body

Two mechanisms get conflated in almost every argument about this topic, and separating them settles half the dispute. The first is topical: fluoride present at the tooth surface during the day-to-day acid attacks, which is why the review of water fluoridation today states that both water fluoridation and fluoride toothpaste are effective and each adds to the effect of the other (Rugg-Gunn et al., Front Oral Health 2026-01-01, PMID 41958764). The second is systemic: fluoride absorbed and circulating while enamel is still forming inside the jaw. That second pathway is the one that produces fluorosis, and the mechanistic description in a 2026 systematic review of exposure and fluorosis is that sustained systemic levels inhibit the proteases MMP-20 and KLK4 and induce endoplasmic reticulum stress during enamel maturation, causing hypomineralisation (Funcuza et al., Int J Mol Sci 2026-06-01, PMID 42353337).

The reason the two mechanisms matter for the argument is that they have different dose-response behaviour. Topical benefit does not require swallowing anything; the water route delivers a low continuous level to the surface plus whatever is ingested. When an authoritative body says fluoride at 0.7 ppm is safe for the brain and a toothpaste label says do not swallow, both statements can be true at once. The systematic review of chronic exposure in people aged 0 to 30 (34 studies selected from 1,020) puts the same boundary: low fluoride concentrations of 0.1 to 1.5 mg/L effectively prevent dental caries, while prolonged high-dose exposure is associated with adverse outcomes including fluorosis, reproductive disturbances and cognitive effects (Obot et al., J Environ Sci Health C Toxicol Carcinog 2026-07-01, PMID 42418311).

For the neurodevelopmental question, one further distinction has to be held in mind: hazard identification versus risk characterisation. The 2024 National Toxicology Program Monograph concluded with moderate confidence that higher fluoride exposure is associated with lower IQ in children, as the JAMA Pediatrics authors summarise it (Taylor et al., Ann Glob Health 2025-01-01, PMID 41393310) (Taylor et al., JAMA Pediatr 2025-03-01, PMID 39761023). The counter-analysis in a dental journal states that there is no established biological mechanism to explain an effect on IQ, which is exactly what makes dose-response modelling so contentious (Kumar et al., J Am Dent Assoc 2026-05-01, PMID 41941356). Where a biological mechanism is described in the literature we verified, it is for enamel, not for neurons (Funcuza et al., Int J Mol Sci 2026-06-01, PMID 42353337).

How big is the caries effect, really

What was measured	Verified estimate	Design and certainty
Change in dmft after starting fluoridation (primary teeth)	MD 0.24 in favour of fluoridation (95% CI −0.03 to 0.52; P = 0.09)	2 contemporary studies, 2,908 children; low-certainty; non-randomised (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658)
Change in the proportion of caries-free children	MD −0.04 primary (95% CI −0.09 to 0.01; P = 0.12); MD −0.03 permanent (95% CI −0.07 to 0.01; P = 0.14)	2 studies each, 2,908 and 2,348 children; low-certainty; 4 and 3 percentage points (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658)
Change in permanent-tooth indices at initiation	Uncertain: DMFT 4 studies, 2,856 children; DMFS 1 study, 343 children	Contemporary evidence; “very uncertain of these findings” (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658)
Same questions using pre-1975 evidence	Not quantifiable with confidence: dmft 5 studies (5,709 children), DMFT 3 studies (5,623), caries-free 5 and 4 studies (6,278 and 6,219)	Very low certainty; downgraded for indirectness because societies changed (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658)
Fluoridated versus non-fluoridated populations, pooled	SMD −0.32 for DMFT and dmf (95% CI −0.48 to −0.17); caries prevalence OR 0.52 permanent (0.43 to 0.63), OR 0.60 deciduous (0.48 to 0.76)	74 studies, 32 pooled; I² = 96%; design, continent and decade explained the heterogeneity (Nascimento et al., JDR Clin Trans Res 2026-04-01, PMID 40574418)
Caries reduction reported in the recent narrative update	Approximately 25% in most studies, 2014 to 2025 literature	Narrative review of observational studies plus national panel reports; no pooled estimate offered (Warren et al., J Am Dent Assoc 2026-04-01, PMID 41746238)
Effect in adults	None quantified by Cochrane: no eligible studies reported adult caries outcomes	Gap, not a null result (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658)

Read the first two rows as a patient-facing number and the fifth row as a population-facing number, and do not let anyone sell you the second as if it were the first. A quarter of a tooth per child, averaged across communities, is the honest contemporary figure for what starting fluoridation does on its own in a toothpaste-using population (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658). The pooled difference between fluoridated and non-fluoridated communities is of high magnitude in both children and adults according to its authors (Nascimento et al., JDR Clin Trans Res 2026-04-01, PMID 40574418) — but 96% heterogeneity means those communities differ in ways the model cannot separate, and most of those studies are cross-sectional, which is why Cochrane downgraded everything for confounding and lack of blinding (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658).

One more number belongs in the same table because it is the one that gets quoted at public meetings: caries in children worldwide, unaffected by any single water decision, sits at 60% to 90% of school children in most industrialised countries according to the Cochrane background (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658). That is the denominator against which a quarter of a tooth should be judged — and it is also why no water policy replaces a toothbrush, a varnish and a low-sugar feeding pattern (Rugg-Gunn et al., Front Oral Health 2026-01-01, PMID 41958764).

Fluorosis: the number attached to 0.7 ppm

The 12% and the 40% are not competing numbers; they answer different questions. Forty per cent of people have some trace of fluorosis that a researcher with a Dean-scale definition can see; twelve per cent have a level that a person would call a cosmetic problem. The gap between them is the whole reason the fluorosis debate talks past itself (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658).

Two further findings belong in the same section. First, the 2026 systematic review of fluorosis and water fluoride levels (20 observational studies, 21,780 participants, Newcastle-Ottawa quality assessment) reports mild fluorosis prevalence of 15% to 20% in cohorts living in optimally fluoridated areas at 0.7 mg/L but carrying additional exposure it calls the “Halo Effect” — infant formula and processed beverages — describing that as a twofold increase over historical benchmarks, and noting that high altitude above 2000 m and arid climates appear to worsen retention by altering renal clearance (Funcuza et al., Int J Mol Sci 2026-06-01, PMID 42353337). Its policy conclusion is uncomfortable for everyone: monitoring water alone is insufficient for modern risk assessment, and total daily intake should be modelled instead (Funcuza et al., Int J Mol Sci 2026-06-01, PMID 42353337).

Second, in a Mexican birth cohort where fluoride arrives through fluoridated salt rather than water, median intake rose from 0.56 mg/day at age 1 to 1.14 mg/day at age 5, all adolescents had some fluorosis and 62% were at TFI 2, predominantly mild; each additional 0.1 mg of daily fluoride intake at age 1 carried OR 1.08 (95% CI 1.00 to 1.17) for a higher score on the upper central incisors, with a marginally significant OR 1.07 (1.00 to 1.16) at age 2 (Castiblanco-Rubio et al., Int J Environ Res Public Health 2025-04-01, PMID 40427806). A community with no fluoridated water at all still produced fluoride intake that exceeded recommendations — which is the argument for treating toothpaste, formula and diet as the exposure, not the tap alone.

The IQ question: what each side actually measured

Finding	Estimate as printed	What it was measured on
Inverse association between fluoride exposure and children’s IQ (Taylor et al., JAMA Pediatr 2025-03-01, PMID 39761023)	Pooled SMD −0.45 (95% CI −0.57 to −0.33); 59 studies, 20,932 children; 64 of 74 studies reported inverse associations; 52 studies rated high risk of bias	Mostly high-exposure settings: 45 of 74 studies in China, 12 in India; exposure from water, urine, dental fluorosis and other measures; authors state data are limited and uncertain below 1.5 mg/L when exposure comes from drinking water alone
Moderate confidence that higher exposure is associated with lower IQ (Taylor et al., Ann Glob Health 2025-01-01, PMID 41393310)	2024 National Toxicology Program Monograph conclusion, as summarised by its authors; the meta-analysis synthesised more than 70 epidemiological studies, with the inverse association persisting in the high-quality subset and consistent across sex, age, country, outcome method, timing and exposure matrix, including below 1.5 mg/L	All exposure routes combined — drinking water, food, beverages, urinary biomarkers
Concerns about the data and analyses used (Antoon et al., JAMA Pediatr 2025-05-01, PMID 40354082)	Correspondence published in the same journal raising methodological objections to the exposure and IQ analysis	Documented at title level in the record; no pooled figure attached
No association at fluoridation-relevant concentrations (Kumar et al., Community Dent Oral Epidemiol 2026-02-01, PMID 41174838)	Pooled SMD 0.04 (95% CI −0.06 to 0.14; p = 0.42; I² = 0%) across five studies in fluoridated areas, favouring higher fluoride; for the maternal urinary fluoride studies the pooled β was −1.06 (95% CI −3.63 to 1.50; p = 0.42; I² = 62%) but all four failed the EPA data-quality criteria, and benchmark dose models showed no functional relationship	Formal data-quality assessment plus hazard-modelling criteria; the paper opens by noting a National Academies consensus report concluded the evidence did not support assessing fluoride as a neurodevelopmental hazard
No association for community water fluoridation specifically (Marques et al., Cien Saude Colet 2026-02-01, PMID 41779589)	Pooled regression coefficient β 1.01, low heterogeneity, three studies included out of 941 identified; no pooled analysis possible for adults (one study only)	Observational studies of communities with versus without fluoridation, IQ as outcome; the authors read it as underlining safety for IQ at fluoridation levels
No negative association across the life course in a US state cohort (Warren et al., Proc Natl Acad Sci U S A 2026-04-01, PMID 41973917)	No evidence that community water fluoridation is negatively associated with adolescent IQ or adult cognitive functioning; exposure inferred from historical community fluoridation records, adolescent IQ from state testing records	Wisconsin Longitudinal Study; the paper states its result is in contrast to studies cited in recent decisions to end fluoridation in Utah, Florida and elsewhere
No causal link supported by the evidence base (Kumar et al., J Am Dent Assoc 2026-05-01, PMID 41941356)	No association at concentrations relevant to fluoridation; associations reported in endemic areas above 1.5 mg/L rest on studies with numerous methodological weaknesses — few high-quality studies, weak designs, inadequate exposure and outcome measurement, inappropriate statistics — and there is no established biological mechanism	Grounded in 4 government reports, 3 independent reports and 7 meta-analyses and systematic reviews, plus a reanalysis of a urinary-fluoride meta-analysis

The structure of the disagreement is worth stating plainly, because it is the useful part. Nobody in this literature has measured IQ in randomised communities assigned to 0.7 ppm versus none; nobody can. What exists is a large body of studies in populations with exposure often far above fluoridation levels, a smaller body of studies inside fluoridated areas, and a national cohort in the United States. The first group shows inverse associations with pooled effects around half a standard deviation (Taylor et al., JAMA Pediatr 2025-03-01, PMID 39761023). The second and third show nothing, with intervals tight enough to exclude an effect of that size in fluoridated areas (Kumar et al., Community Dent Oral Epidemiol 2026-02-01, PMID 41174838) (Marques et al., Cien Saude Colet 2026-02-01, PMID 41779589) (Warren et al., Proc Natl Acad Sci U S A 2026-04-01, PMID 41973917). A commentary on the national study states the policy reading: adverse neurodevelopmental effects do not result from municipal fluoridation (Savitz et al., Sci Adv 2025-11-01, PMID 41259519).

Neither side can be quoted as “proven”, and two honest caveats belong to the pro-fluoridation columns. The fluoridated-area meta-analyses are built from few studies — three and five respectively (Kumar et al., Community Dent Oral Epidemiol 2026-02-01, PMID 41174838) (Marques et al., Cien Saude Colet 2026-02-01, PMID 41779589) — and the state cohort infers exposure from historical records of whether a community fluoridated, not from measured intake (Warren et al., Proc Natl Acad Sci U S A 2026-04-01, PMID 41973917). Meanwhile the opposing camp’s own abstract concedes the exposure-specific gap: limited data and uncertainty below 1.5 mg/L when exposure is estimated from drinking water alone (Taylor et al., JAMA Pediatr 2025-03-01, PMID 39761023). That is the exact region where a family lives, so the answer to “is the tap safe for my child’s brain” is not “certainly safe” but “no consistent effect has been demonstrated at these concentrations, and the burden of the studies suggesting harm sits above them” (Kumar et al., J Am Dent Assoc 2026-05-01, PMID 41941356) (Zohoori et al., Adv Nutr 2025-12-01, PMID 41135657).

One more piece of evidence matters for how much weight to give any of it: an archival history of ninety years of industry documents, held at the University of California San Francisco, concludes that records from the 1930s onward show the sugar industry, sometimes together with dental interests, exaggerating fluoride’s effectiveness and downplaying safety concerns, and that some reviewers with undisclosed conflicts produced biased reviews (Neurath et al., Environ Health 2025-09-01, PMID 41016930). This does not settle any dose question. It does explain why both camps should be read for their conflicts of interest as well as their confidence intervals.

Formula, bottled water and the other fluoride you are not counting

The question “is fluoride in tap water safe for babies” almost never means the tap. It means the mixture that goes into a bottle several times a day, and the toothpaste on the brush. In a study of four infant formula brands marketed in Spain, prepared with four different waters and analysed with an ion-selective electrode, fluoride concentrations ranged from 0.04 to 1.02 mg/L; no formula exceeded the daily dose limit associated with fluorosis risk (0.07 mg F per kg body weight per day) when reconstituted with water containing 0.1 mg/L, and every one of them exceeded it when the same powders were mixed with water containing 0.99 mg/L (Gallego-Reyes et al., J Clin Pediatr Dent 2024-01-01, PMID 38239163). The authors’ recommendation is that formula packaging should carry a warning, because the risk track belongs to the water, not the powder (Gallego-Reyes et al., J Clin Pediatr Dent 2024-01-01, PMID 38239163). A review of 17 studies of fluoride in infant formula gives the same variability from the other side: milk-based formulas measured 0.01 to 0.92 ppm with only two studies above 1.30 ppm, soy-based 0.13 to 1.11 ppm, and the variation is attributed to the water used to reconstitute them (Velez-León et al., Children (Basel) 2023-12-01, PMID 38136098).

This is why EFSA’s 2025 assessment is expressed as a daily intake rather than a water concentration: its aggregate exposure model includes food, drinking water, discretionary salt and ingested dental care products, and it found that exposure based on the mean fluoride concentration in EU drinking water stayed below its health-based guidance values for all age groups, while at the 95th percentile of water concentration the guidance values were exceeded for every age group except adolescents (EFSA Scientific Committee et al., EFSA J 2025-07-01, PMID 40698337). Its explicit note is also worth quoting to anyone using it as a pro- or anti-fluoridation document: benefit assessment was not included in that risk assessment (EFSA Scientific Committee et al., EFSA J 2025-07-01, PMID 40698337).

And the toothpaste half is where the parent has direct control, with the trial-grade evidence behind it: moderate-certainty evidence that a lower concentration reduces fluorosis risk in the permanent teeth of children under six (550 ppm vs 1000 ppm RR 0.75, 440 ppm vs 1450 ppm RR 0.72), very low-certainty evidence for the amount squeezed on the brush (less than half a brush vs half or more: OR 0.77, 95% CI 0.41 to 1.46) and for how often a toddler is brushed (less than once a day vs once or more: OR 0.62, 95% CI 0.53 to 0.74, low-certainty) (Wong et al., Cochrane Database Syst Rev 2024-06-01, PMID 38899538). A survey of 8,011 adults in Canada and the United States found that most parents report using more fluoride toothpaste for young children than recommended (Till et al., Curr Probl Pediatr Adolesc Health Care 2025-06-01, PMID 40628571). For a comparison of what the same fluoride does when a professional applies it, see our review of fluoride varnish in early childhood.

What happens when a city stops

Cochrane can answer this question only weakly: one study after 1975 reported the outcome it could use, and for 2,994 children the change in DMFS could not be determined — so from the strictest evidence base, cessation has not been quantified (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658). The narrative update covering 2014 to 2025 is much stronger in its wording: cessation consistently results in higher caries levels and greater restorative need (Warren et al., J Am Dent Assoc 2026-04-01, PMID 41746238). Between those two poles sits a scoping review of nineteen studies from 1970 to 2025 in Europe, North America, Australia and Israel, which found that jurisdictions without replacement fluoride programmes more often reported increases in caries — up to a 75% increase in mean DMFT — and increased treatment utilisation, particularly restorations, while settings with alternative fluoride strategies or broader prevention reported stable outcomes or continued improvement; several studies reported greater burden and cost, including a 32% increase in caries-related expenditure, concentrated among younger children, socioeconomically disadvantaged groups and publicly insured children (Hung et al., BMC Oral Health 2026-06-01, PMID 42337497).

The three country-level studies are the ones worth reading as numbers rather than rhetoric. In Alberta, among 2,659 children who received caries-related treatment under general anaesthesia between 2010 and 2019, 65% lived in the non-fluoridated area, and cessation was significantly associated with an increased rate of general-anaesthesia events per 10,000 children in both the 0 to 5 and 6 to 11 age groups, with a more pronounced effect in the younger group, rising with post-cessation time (Yazdanbakhsh et al., Can J Public Health 2024-04-01, PMID 38389035). The mean age of those children was 4.8 years (SD 2.3), median 4 (IQR 3 to 6) — an age at which a filling decision is really an anaesthesia decision (Yazdanbakhsh et al., Can J Public Health 2024-04-01, PMID 38389035). In Israel, where fluoridation ran from 1981 until the Ministry of Health stopped it in 2014, records for children aged 3 to 5 treated in one clinic network showed a statistically significant increase in the mean number of treatments after cessation, counting restorations of all types and stainless steel crowns (Nezihovski et al., Isr J Health Policy Res 2024-09-01, PMID 39304948). In Saudi Arabia, a retrospective examination of 568 people aged 6 to 50 after cessation in one city found DMFT of 5.62 in adults versus 3.98 in children, and its author describes the result as the dual-edged nature of fluoride exposure through the water supply (Alrashdi et al., Front Oral Health 2025-01-01, PMID 40260429).

Note what those three studies measure: operations under anaesthesia, restorations, crowns. They are not measuring a child’s feeling about white patches. A cessation decision that is defended on cosmetic fluorosis grounds and produces more four-year-olds in general anaesthesia has traded a mild enamel finding for a procedural risk, and that trade is visible in the data even though the studies were not designed to make it (Yazdanbakhsh et al., Can J Public Health 2024-04-01, PMID 38389035) (Nezihovski et al., Isr J Health Policy Res 2024-09-01, PMID 39304948). The corollary matters too: where a community that stops fluoridation simultaneously ramps up varnish, sealant and toothpaste programmes, the scoping review finds outcomes that stay flat (Hung et al., BMC Oral Health 2026-06-01, PMID 42337497) — which is the honest way to state the position of every country that never fluoridated and instead put fluoride in toothpaste and in the school-nursery chair (Rugg-Gunn et al., Front Oral Health 2026-01-01, PMID 41958764). For the individual version of that substitution, see silver diamine fluoride for arresting cavities and the sealant evidence.

How to read this like a clinician

• Ask for the source, not the opinion. Municipal treated water with a published concentration, untreated private well, bottled, or “we use a filter” — these are four different exposures. The whole exposure science reviewed above is dose-based, not source-based (EFSA Scientific Committee et al., EFSA J 2025-07-01, PMID 40698337) (Funcuza et al., Int J Mol Sci 2026-06-01, PMID 42353337).
• Convert the patient’s number into mg per day. EFSA’s intake ceilings are 1.0 mg/day for infants, 1.6 for toddlers and 2.0 for children 4 to 8 years, with 3.3 mg/day established from a 1.5 mg/L water reference point for pregnant women, other adults and children over 8 (EFSA Scientific Committee et al., EFSA J 2025-07-01, PMID 40698337). A bottle-fed infant on 0.99 mg/L water is the documented way to cross that line on a non-fluoridated diet (Gallego-Reyes et al., J Clin Pediatr Dent 2024-01-01, PMID 38239163).
• Separate the aesthetic endpoint from the neurological one. At 0.7 ppm the pooled aesthetic-concern prevalence is 12% and any-level prevalence 40% (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658); the brain-effect signal in the reviewed human evidence appears above 1.5 mg/L, with inconsistency below it (EFSA Scientific Committee et al., EFSA J 2025-07-01, PMID 40698337) (Zohoori et al., Adv Nutr 2025-12-01, PMID 41135657).
• Quote the interval, not the point. “A quarter of a tooth” with a CI of −0.03 to 0.52 is a small expected benefit with a real possibility of none; a chart note that says “reduces caries 25%” without the design that produced it is not defensible (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658) (Warren et al., J Am Dent Assoc 2026-04-01, PMID 41746238).
• Flag the I² before quoting any pooled caries figure. The 74-study meta-analysis reports I² = 96% for DMFT and dmf, and its authors found that study design, continent and decade of publication explained the heterogeneity — which is a statement that context drives the number more than fluoride does (Nascimento et al., JDR Clin Trans Res 2026-04-01, PMID 40574418).
• For adults, document the absence. Cochrane found no eligible studies reporting adult caries outcomes, so advice to a 45-year-old about the tap is an extrapolation from children plus root-surface biology, not a measured effect (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658). The adult numbers that exist come from before-and-after and cross-sectional work (Nascimento et al., JDR Clin Trans Res 2026-04-01, PMID 40574418) (Alrashdi et al., Front Oral Health 2025-01-01, PMID 40260429).
• Prescribe the topical part regardless of the water answer, in age-appropriate doses, and say why: the concentration effect is the only fluorosis-related comparison in the Cochrane topical review graded moderate certainty (Wong et al., Cochrane Database Syst Rev 2024-06-01, PMID 38899538). Our brushing frequency and duration review is the companion for the instruction sheet.
• Where fluoridation is being ended locally, expect the paediatric anaesthesia list to grow and plan the replacement: varnish intervals, sealants on newly erupted molars, SDF where behaviour limits care (Hung et al., BMC Oral Health 2026-06-01, PMID 42337497) (Yazdanbakhsh et al., Can J Public Health 2024-04-01, PMID 38389035). The same logic applies to a baby tooth decision; see whether baby teeth should be filled.

Cost, coverage and the Utah angle

The economic case is real but soft in exactly the way dentists should be able to describe. Fifteen economic evaluations from high- and middle-income settings were synthesised without pooling because of heterogeneity in settings, methods and outcomes; most reported fluoridation as cost-effective or cost-saving compared with no fluoridation, with more favourable results in larger populations, in populations with higher baseline caries risk, and in studies focused on children (Almutairi et al., Int Dent J 2026-08-01, PMID 42142407). The JADA update puts the same finding in one line: the cost of fluoridation is consistently a small fraction of the increased treatment costs expected without it (Warren et al., J Am Dent Assoc 2026-04-01, PMID 41746238). The clinical review adds the qualitative virtues — equitable coverage so that those at greatest risk benefit most, low environmental impact, low cost — and notes that WHO urges member states to consider implementing it (Rugg-Gunn et al., Front Oral Health 2026-01-01, PMID 41958764).

Utah belongs to this argument twice. The Wisconsin cohort paper states that its null finding is in contrast to studies cited in recent decisions to end community water fluoridation in Utah, Florida and elsewhere — which is a documented fact about the policy debate, not about caries (Warren et al., Proc Natl Acad Sci U S A 2026-04-01, PMID 41973917). And for a patient in Utah the practical consequence of a non-fluoridated or private-well water supply is not a policy problem but a scheduling one, because the public children’s benefit in this state already pays for the individual substitutes: fluoride varnish up to four times per calendar year (below age 5 a physician may apply it at a well-child visit, from age 5 it is done in the dental office), cleanings twice a year, sealants once every two years per eligible first or second permanent molar or premolar, and silver diamine fluoride every six months per tooth on baby teeth as a non-invasive alternative to fillings — the document lists all four as covered children’s services without attaching a prior-authorisation requirement, in contrast with orthodontics, for which Utah Medicaid requires prior authorisation for all treatment scored on the state IOTN necessity sheet, and with retainers, covered once per lifetime at the end of treatment (InsureKidsNow summary of benefits for Utah Medicaid, data as of 4 February 2026; see Additional documents below). Nitrous oxide analgesia is explicitly not covered, and general anaesthesia or intravenous sedation only where local anaesthesia cannot be used safely, with the reason documented — a detail that matters when the comparison is “tap fluoride” versus “a four-year-old in an anaesthesia chair”, which is what the Alberta and Israeli cessation data actually describe (Yazdanbakhsh et al., Can J Public Health 2024-04-01, PMID 38389035) (Nezihovski et al., Isr J Health Policy Res 2024-09-01, PMID 39304948).

For a family paying cash the arithmetic is simpler than the epidemiology: population fluoridation is cheap per person and its benefit per child is a quarter of a tooth (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658), while varnish, toothpaste and a first visit are individually priced and individually schedulable. Neither is a substitute for sugar-frequency control, which none of these sources can fix for you.

What the evidence does not support

• That fluoridation “prevents 25% of cavities” as a contemporary, individual-level promise. The ~25% figure comes from a narrative review of observational studies, and the contemporary pooled estimate at initiation is a quarter of a tooth per child with an interval crossing zero (Warren et al., J Am Dent Assoc 2026-04-01, PMID 41746238) (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658).
• That fluoridation at 0.7 ppm is proven harmless to cognition for all time. What is documented is an absence of consistent association at fluoridation-relevant concentrations in a small number of studies, and acknowledged uncertainty in that exposure region (Kumar et al., Community Dent Oral Epidemiol 2026-02-01, PMID 41174838) (Marques et al., Cien Saude Colet 2026-02-01, PMID 41779589) (Taylor et al., JAMA Pediatr 2025-03-01, PMID 39761023).
• That fluoride is an established neurodevelopmental toxin at drinking-water concentrations. The fluoridation-relevant analyses find no association, the hazard-modelling assessment states the data do not meet quality criteria and show no functional concentration-response relationship, and one government-anchored review records that a National Academies consensus study concluded the evidence did not support treating fluoride as a neurodevelopmental hazard (Kumar et al., Community Dent Oral Epidemiol 2026-02-01, PMID 41174838) (Kumar et al., J Am Dent Assoc 2026-05-01, PMID 41941356).
• That stopping fluoridation is cost-neutral. Cessation increases caries and restorative need in the narrative review and is associated with a 32% rise in caries-related expenditure in the scoping review, concentrated in the youngest and most publicly insured children (Warren et al., J Am Dent Assoc 2026-04-01, PMID 41746238) (Hung et al., BMC Oral Health 2026-06-01, PMID 42337497).
• That fluorosis is only cosmetic in every case or severe in none. Prevalence at 0.7 ppm reaches 40% for any level and 12% for aesthetic concern; severity at that level is described as mild to very mild (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658) (Rugg-Gunn et al., Front Oral Health 2026-01-01, PMID 41958764).
• That the toothpaste decision is trivial. It is the comparison with moderate-certainty trial evidence attached, and it is the one exposure a household sets itself (Wong et al., Cochrane Database Syst Rev 2024-06-01, PMID 38899538) (Till et al., Curr Probl Pediatr Adolesc Health Care 2025-06-01, PMID 40628571).
• That boiling, filtering or buying bottled water solves the problem. None of the records verified for this article evaluated boiling or a consumer filter, so no claim is made here — the measurable step is reading your supplier’s reported concentration and comparing it with the intake ceilings above (EFSA Scientific Committee et al., EFSA J 2025-07-01, PMID 40698337).
• That the 1.5 mg/L figure is a safety line for fluoridated communities. It is EFSA’s reference point, chosen because that is where human evidence of brain effects starts to appear, and the same assessment states that the EU legal limit of 1.5 mg/L is not sufficiently protective for aggregate intake (EFSA Scientific Committee et al., EFSA J 2025-07-01, PMID 40698337).

A plan, if you want one

• Find your number. Get the published fluoride concentration of the water you actually drink — municipal report, well test, or the label on the bottled brand. Everything else in this article is calibrated by that one value against the 1.0 mg/L zone where a rapid review found no evidence of adverse effects and the 1.5 mg/L line where human associations begin (Zohoori et al., Adv Nutr 2025-12-01, PMID 41135657) (EFSA Scientific Committee et al., EFSA J 2025-07-01, PMID 40698337).
• If you are at or below about 0.7 mg/L and pregnant or feeding an infant: nothing in the verified record argues for changing the tap. Do the toothpaste arithmetic instead, and if formula is used, look at the water it is mixed with (Gallego-Reyes et al., J Clin Pediatr Dent 2024-01-01, PMID 38239163) (Velez-León et al., Children (Basel) 2023-12-01, PMID 38136098).
• If you are above 1.5 mg/L from a private well: that is the range in which the human evidence for brain effects and for fluorosis clusters, and it is a household exposure question worth acting on medically, not a political one (EFSA Scientific Committee et al., EFSA J 2025-07-01, PMID 40698337) (Zohoori et al., Adv Nutr 2025-12-01, PMID 41135657) (Obot et al., J Environ Sci Health C Toxicol Carcinog 2026-07-01, PMID 42418311).
• Whatever the water does, do the three individually controlled things: brush twice a day with an age-appropriate amount of toothpaste and spit rather than rinse; keep free sugars in episodes rather than all day; and keep the professional schedule — varnish and sealants on the interval your plan pays for (Wong et al., Cochrane Database Syst Rev 2024-06-01, PMID 38899538) (Rugg-Gunn et al., Front Oral Health 2026-01-01, PMID 41958764).
• For a child already showing white or brown marks: ask for the severity graded, not just “fluorosis”, because the decision-relevant line is aesthetic concern versus any level, and mild enamel changes do not need the same response as the 12% group (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658) (Rugg-Gunn et al., Front Oral Health 2026-01-01, PMID 41958764). Our page on white spots and whether they need a filling covers that conversation.
• If your community is voting on it: ask two questions in public — what the replacement strategy will be for the highest-risk children, and who will measure the outcome after the change. Both are the parts the evidence says have been missing (Hung et al., BMC Oral Health 2026-06-01, PMID 42337497) (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658).

Frequently asked questions

Is fluoride in tap water safe for babies? The verified evidence does not say “safe” or “unsafe” at 0.7 mg/L; it says the aggregate intake is what matters, and the ceilings for the first years are 1.0 mg/day for infants and 1.6 mg/day for toddlers (EFSA Scientific Committee et al., EFSA J 2025-07-01, PMID 40698337). A formula-mixed diet can approach that on its own: with water at 0.1 mg/L no tested formula exceeded the fluorosis-risk dose, while with water at 0.99 mg/L all of them did (Gallego-Reyes et al., J Clin Pediatr Dent 2024-01-01, PMID 38239163).

Does boiling the water remove fluoride? Not answered by anything verified here. None of the twenty-eight records evaluated boiling, so this article makes no claim about it. What is documented is the concentration range and the total-intake ceilings, and those are the numbers to test your own water against (EFSA Scientific Committee et al., EFSA J 2025-07-01, PMID 40698337).

Is it safe for dogs or cats? The exposure literature above is human: doses in mg per day, effects in children aged 0 to 30, and a risk assessment built on human pregnancy and childhood (Obot et al., J Environ Sci Health C Toxicol Carcinog 2026-07-01, PMID 42418311) (EFSA Scientific Committee et al., EFSA J 2025-07-01, PMID 40698337). Nothing verified here supports transferring those conclusions to pets in either direction; that question belongs to a veterinarian, and the honest answer from this evidence base is that we do not know.

Does fluoridated water help adults? Cochrane found no eligible studies reporting caries outcomes in adults, so a specific adult effect size from initiation does not exist (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658). What does exist: pooled population comparisons whose authors describe high-magnitude differences in children and adults, and a post-cessation city where adult DMFT was 5.62 (Nascimento et al., JDR Clin Trans Res 2026-04-01, PMID 40574418) (Alrashdi et al., Front Oral Health 2025-01-01, PMID 40260429).

Is it worth doing if toothpaste exists? The reviews describe them as additive — water fluoridation and fluoride toothpaste each add to the effect of the other — and note the property that makes fluoridation different: coverage is equitable, so the highest-risk children benefit without anyone scheduling anything (Rugg-Gunn et al., Front Oral Health 2026-01-01, PMID 41958764). The counterweight is that the contemporary effect size shrank once toothpaste became universal (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658).

Why do people say fluoridation was invented to keep industry waste out of the air? A ninety-year archival study of industry documents records the sugar industry’s role in promoting fluoride as the solution to tooth decay and in exaggerating its effectiveness while downplaying safety concerns, and warns that researchers with undisclosed conflicts produced biased reviews (Neurath et al., Environ Health 2025-09-01, PMID 41016930). That history is a reason to distrust sponsored claims in both directions; it is not a measurement of caries or IQ.

Should I switch to bottled water to avoid fluoride? Measured bottled and formula-reconstitution water in the study above spanned 0.04 to 1.02 mg/L, so the label, not the category, is what matters (Gallego-Reyes et al., J Clin Pediatr Dent 2024-01-01, PMID 38239163). And replacing fluoridated municipal water with an unmonitored supply is the one change that reliably increases uncertainty about total daily intake (EFSA Scientific Committee et al., EFSA J 2025-07-01, PMID 40698337) (Funcuza et al., Int J Mol Sci 2026-06-01, PMID 42353337).

What should a dentist tell a frightened parent? The public data are usable: among 8,011 adults in Canada and the United States, 60% correctly identified why fluoride is added to drinking water, and among those familiar with fluoridation 51% supported it, 27% opposed and 25% were neutral, with trust in public health officials at 87% among supporters and 52.1% among non-supporters; when shown hypothetical risk-benefit scenarios, participants prioritised avoiding a possible reduction in IQ over preventing one cavity (Till et al., Curr Probl Pediatr Adolesc Health Care 2025-06-01, PMID 40628571). That is a values trade-off, not an arithmetic error, and it is the reason to lead with the toothpaste dose rather than the pipeline.

Glossary: kitchen-table words ↔ chart words

What you say at home	What is in the notes	How it is measured in the studies above
“Is the water fluoridated?”	Community water fluoridation (CWF) status; mg F per L or ppm	Concentration in the public supply, or inferred from community records (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658) (Warren et al., Proc Natl Acad Sci U S A 2026-04-01, PMID 41973917)
“How much fluoride does my child actually get?”	Total daily intake; aggregate exposure	Food, drinking water, discretionary salt and ingested dentifrice modelled together (EFSA Scientific Committee et al., EFSA J 2025-07-01, PMID 40698337); intake measured in mg per day in a birth cohort (Castiblanco-Rubio et al., Int J Environ Res Public Health 2025-04-01, PMID 40427806)
“Chalky spots from the water”	Dental fluorosis; enamel hypomineralisation	Dean’s index or TFI scores; prevalence with 95% CI across studies (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658) (Castiblanco-Rubio et al., Int J Environ Res Public Health 2025-04-01, PMID 40427806)
“A little bit of decay”	dmft or DMFT (decayed, missing, filled teeth); surfaces as dmfs and DMFS	Change from baseline around a policy change, or difference between exposed and unexposed (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658) (Nascimento et al., JDR Clin Trans Res 2026-04-01, PMID 40574418)
“A third of a tooth”	Mean difference; standardised mean difference (SMD)	Random-effects pooling with 95% CI and I² reported (Taylor et al., JAMA Pediatr 2025-03-01, PMID 39761023) (Kumar et al., Community Dent Oral Epidemiol 2026-02-01, PMID 41174838)
“How sure are we?”	Certainty of evidence (GRADE: high, moderate, low, very low)	Downgraded for non-randomised design, confounding, unblinded assessors, imprecision, indirectness (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658) (Wong et al., Cochrane Database Syst Rev 2024-06-01, PMID 38899538)
“The safe level”	Tolerable upper intake level (UL); health-based guidance value (HBGV); reference point	1.0, 1.6 and 2.0 mg per day by age; 3.3 mg per day derived from a 1.5 mg/L water reference point (EFSA Scientific Committee et al., EFSA J 2025-07-01, PMID 40698337)
“Does the dose matter?”	Concentration-response; benchmark dose modelling	Models tested against EPA data-quality criteria; no functional relationship in the assessed dataset (Kumar et al., Community Dent Oral Epidemiol 2026-02-01, PMID 41174838)
“Other fluoride in the house”	Halo effect / co-exposure: infant formula, processed beverages, salt, well water at altitude	Confounder-selected systematic review; mild fluorosis 15-20% in high-exposure cohorts at 0.7 mg/L (Funcuza et al., Int J Mol Sci 2026-06-01, PMID 42353337)
“They turned it off”	CWF cessation; replacement fluoride strategy	Before-and-after service utilisation, restorations, crowns, treatment under general anaesthesia, expenditure (Hung et al., BMC Oral Health 2026-06-01, PMID 42337497) (Yazdanbakhsh et al., Can J Public Health 2024-04-01, PMID 38389035) (Nezihovski et al., Isr J Health Policy Res 2024-09-01, PMID 39304948)

Related reading on this site: whether you actually need that filling, why night feeding drives toddler cavities and whether to remove wisdom teeth or monitor them.

How this page was built, and what it cannot tell you

Method, briefly: candidate papers were found through Europe PMC with queries matching what people actually type, and every record used here — authors, journal, volume, pages, DOI, PMID, open-access flag and citation count — was re-fetched from Europe PMC on the day of publication and is reproduced in the Sources list with its DOI. Numbers in the body were copied from the retrieved abstract text; where an abstract reports an interval, the interval is reproduced here. A figure was generated only when a single source reported a single unit, so the chart carries one study’s prevalences and its confidence limits rather than a mixture. Coverage statements were read from the CMS InsureKidsNow benefit summary for Utah Medicaid (data as of 4 February 2026) and each claim in this article was checked against the text of that document. Public-opinion numbers come from the two surveys named in the body (Till et al., Curr Probl Pediatr Adolesc Health Care 2025-06-01, PMID 40628571) (Saveanu et al., Epidemiologia (Basel) 2026-05-01, PMID 42201212), which found familiarity high and understanding shallow: in a Romanian sample of 200 adults, 94% knew what fluoride was and 91% knew it protects teeth, but only 34% knew fluoridation methods and 53% had heard of dental fluorosis (Saveanu et al., Epidemiologia (Basel) 2026-05-01, PMID 42201212).

What this page cannot tell you: your water. Nothing verified above measures the concentration in any particular household, and the entire risk assessment depends on that value (EFSA Scientific Committee et al., EFSA J 2025-07-01, PMID 40698337). It cannot tell you whether your child’s fluorosis is in the 12% aesthetic-concern group or the wider 40%, because that distinction is made by looking at teeth with a severity index, not by reading (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658). It cannot tell you what happens to an adult’s root surfaces when a supply changes, because Cochrane found no eligible adult caries studies (Iheozor-Ejiofor et al., Cochrane Database Syst Rev 2024-10-01, PMID 39362658). And it cannot settle the IQ dispute, because the studies that would settle it cannot be done: as the Cochrane topical review puts it, proposing new randomised trials of a fluorosis outcome would be unethical (Wong et al., Cochrane Database Syst Rev 2024-06-01, PMID 38899538). What can be done is the part this article is for — the concentration you have, the dose you choose, and the interval you keep.

Sources

Peer-reviewed evidence

• Iheozor-Ejiofor Z, Walsh T, Lewis SR, Riley P, Boyers D, Clarkson JE, Worthington HV, Glenny AM, O’Malley L. Water fluoridation for the prevention of dental caries. Cochrane Database Syst Rev 2024-10-01;10():CD010856. doi:10.1002/14651858.cd010856.pub3 · PMID 39362658 · PMCID PMC11449566 · cited by 33 (Europe PMC)
• Wong MCM, Zhang R, Luo BW, Glenny AM, Worthington HV, Lo ECM. Topical fluoride as a cause of dental fluorosis in children. Cochrane Database Syst Rev 2024-06-01;6():CD007693. doi:10.1002/14651858.cd007693.pub3 · PMID 38899538 · PMCID PMC11187792 · cited by 10 (Europe PMC)
• Warren JJ, Levy SM, Kumar JV. An update on community water fluoridation, part 1: Mechanism of action, effectiveness, and cost savings. J Am Dent Assoc 2026-04-01;157(4):334-343. doi:10.1016/j.adaj.2026.01.009 · PMID 41746238
• Kumar JV, Levy SM, Warren JJ. An update on community water fluoridation, part 2: Fluoride exposure and children’s intelligence (IQ) scores-relevance of systematic reviews and meta-analyses to community water fluoridation. J Am Dent Assoc 2026-05-01;157(5):471-486. doi:10.1016/j.adaj.2026.02.010 · PMID 41941356
• Warren JR, Rumore G, Sicinski K, Herd P, Engelman M. Municipal water fluoridation, adolescent IQ, and cognition across the life course: Evidence from the Wisconsin Longitudinal Study. Proc Natl Acad Sci U S A 2026-04-01;123(16):e2536005123. doi:10.1073/pnas.2536005123 · PMID 41973917 · PMCID PMC13099618 · open access
• Kumar JV, Moss ME, Liu H, Fisher-Owens S, Rugg-Gunn A, Kuring J. Fluoride and Neurodevelopmental Hazard Modelling: An Assessment of Concentration-Response Analysis. Community Dent Oral Epidemiol 2026-02-01;54(1):101-111. doi:10.1111/cdoe.70027 · PMID 41174838 · PMCID PMC12808860 · open access · cited by 3 (Europe PMC)
• Taylor KW, Eftim SE, Sibrizzi CA, Blain RB, Magnuson K, Hartman PA, Bucher JR, Rooney AA. Addressing Critiques of the Evidence Linking Fluoride and Children’s IQ. Ann Glob Health 2025-01-01;91(1):83. doi:10.5334/aogh.4853 · PMID 41393310 · PMCID PMC12700148 · open access · cited by 1 (Europe PMC)
• Taylor KW, Eftim SE, Sibrizzi CA, Blain RB, Magnuson K, Hartman PA, Rooney AA, Bucher JR. Fluoride Exposure and Children’s IQ Scores: A Systematic Review and Meta-Analysis. JAMA Pediatr 2025-03-01;179(3):282-292. doi:10.1001/jamapediatrics.2024.5542 · PMID 39761023 · PMCID PMC11877182 · open access · cited by 54 (Europe PMC)
• Antoon JW, Kumar JV. Concerns About Data and Analyses Used in Assessing Fluoride Exposure and Children’s IQ Scores. JAMA Pediatr 2025-05-01;. doi:10.1001/jamapediatrics.2025.0932 · PMID 40354082 · PMCID PMC12353652 · cited by 1 (Europe PMC)
• Nascimento CFD, Gindri LADS, de Oliveira MN, Paranhos LR, Hugo FN. Water Fluoridation and Dental Caries Prevention Globally: A Systematic Review and Meta-Analysis. JDR Clin Trans Res 2026-04-01;11(2):107-116. doi:10.1177/23800844251342804 · PMID 40574418
• Hung M, Ward C, Marx J, Nelson M, Chriss H, Smit Z, Shah A, Newman J, Parry A, Mohajeri A. The impact of community water fluoridation cessation on health equity and service utilization: a global scoping review. BMC Oral Health 2026-06-01;. doi:10.1186/s12903-026-09007-w · PMID 42337497
• Almutairi D, Al Humaidi M, Alshahrani SM, Albalawi A. Cost-Effectiveness of Population-Level Fluoride Policies: A Systematic Review of Health and Economic Outcomes. Int Dent J 2026-08-01;76(4):109619. doi:10.1016/j.identj.2026.109619 · PMID 42142407 · PMCID PMC13200109 · open access
• Rugg-Gunn A, Lowry R, Cockcroft B, Walmsley AD. Water fluoridation today: – benefits and challenges. Front Oral Health 2026-01-01;7():1745916. doi:10.3389/froh.2026.1745916 · PMID 41958764 · PMCID PMC13057279 · open access
• Savitz DA. Evidence-based water fluoridation policy. Sci Adv 2025-11-01;11(47):eaed4503. doi:10.1126/sciadv.aed4503 · PMID 41259519 · PMCID PMC12629195 · open access · cited by 3 (Europe PMC)
• Zohoori FV, Kumah EA, Kronic J, Drinnan M, Morris AJ. Public Health Impacts of Water Fluorides: Current Evidence from a Rapid Systematic Review. Adv Nutr 2025-12-01;16(12):100547. doi:10.1016/j.advnut.2025.100547 · PMID 41135657 · PMCID PMC12662105 · open access · cited by 1 (Europe PMC)
• Obot DN, Udom GJ, Aziakpono OM, Obilor OF, Onyeukwu NJ, Aturamu A, Agbana RD, Omole JG, Olusola AJ, Okon IA, Udom NG, Orisakwe OE. Health implications of chronic fluoride exposure in children and young adults: a systematic review. J Environ Sci Health C Toxicol Carcinog 2026-07-01;:1-25. doi:10.1080/26896583.2026.2697147 · PMID 42418311
• EFSA Scientific Committee, Bennekou SH, Allende A, Bearth A, Casacuberta J, Castle L, Coja T, Crépet A, Hoogenboom R, Knutsen H, Lambré C, Nielsen SS, Turck D, Civera AV, Villa R, Zorn H, Castenmiller J, Cheyns K, Darney K, Gilbert M, Leblanc JC, Meyer H, Ntzani E, Paparella M, Vinceti M, Wallace H, Anastassiadou M, Bastaki M, Cattaneo I, Greco L, Lanzoni A, Riolo F, Mosbach-Schulz O, Terron A, Halldorsson T, Halldorsson T. Updated consumer risk assessment of fluoride in food and drinking water including the contribution from other sources of oral exposure. EFSA J 2025-07-01;23(7):e9478. doi:10.2903/j.efsa.2025.9478 · PMID 40698337 · PMCID PMC12280829 · open access · cited by 3 (Europe PMC)
• Funcuza M, Magunga BT, Rathebe PC, Mbonane TP. A Systematic Review on the Association Between Water Fluoride Levels and Dental Fluorosis: Exploring the ‘Halo Effect’ and Confounding Environmental Factors. Int J Mol Sci 2026-06-01;27(12):5623. doi:10.3390/ijms27125623 · PMID 42353337
• Velez-León E, Pacheco-Quito EM, Díaz-Dosque M, Tobar-Almache D. Worldwide Variations in Fluoride Content in Beverages for Infants. Children (Basel) 2023-12-01;10(12):1896. doi:10.3390/children10121896 · PMID 38136098 · PMCID PMC10741400 · open access · cited by 7 (Europe PMC)
• Gallego-Reyes SM, Cury JA, Pérez-Silva A, Serna-Muñoz C, Fernández-Pizarro I, Martínez-Beneyto Y, Ortiz-Ruiz AJ. Potential risk of dental fluorosis associated with different baby formulas and water brands marketed in Spain. J Clin Pediatr Dent 2024-01-01;48(1):111-119. doi:10.22514/jocpd.2024.013 · PMID 38239163 · cited by 1 (Europe PMC)
• Castiblanco-Rubio GA, Hector EC, Urena-Cirett J, Cantoral A, Hu H, Peterson KE, Tellez-Rojo MM, Martinez-Mier EA. Dietary Fluoride Exposure During Early Childhood and Its Association with Dental Fluorosis in a Sample of Mexican Adolescents. Int J Environ Res Public Health 2025-04-01;22(5):689. doi:10.3390/ijerph22050689 · PMID 40427806 · PMCID PMC12111587 · open access · cited by 2 (Europe PMC)
• Yazdanbakhsh E, Bohlouli B, Patterson S, Amin M. Community water fluoride cessation and rate of caries-related pediatric dental treatments under general anesthesia in Alberta, Canada. Can J Public Health 2024-04-01;115(2):305-314. doi:10.17269/s41997-024-00858-w · PMID 38389035 · PMCID PMC11027763 · cited by 6 (Europe PMC)
• Nezihovski SS, Findler M, Chackartchi T, Mann J, Haim D, Tobias G. The effect of cessation of drinking water fluoridation on dental restorations and crowns in children aged 3-5 years in Israel – a retrospective study. Isr J Health Policy Res 2024-09-01;13(1):50. doi:10.1186/s13584-024-00637-5 · PMID 39304948 · PMCID PMC11414298 · open access · cited by 7 (Europe PMC)
• Alrashdi M. Associations between community water fluoridation cessation and the prevalence of dental caries and fluorosis in Alrass city, Saudi Arabia. Front Oral Health 2025-01-01;6():1508466. doi:10.3389/froh.2025.1508466 · PMID 40260429 · PMCID PMC12009762 · open access
• Marques RB, Bastos LF, Magalhães EIDS, Ribeiro CCC, Hugo FN. Community water fluoridation and intelligence quotient: systematic review and meta-analysis of observational studies. Cien Saude Colet 2026-02-01;31(2):e19682023. doi:10.1590/1413-81232026312.19682023 · PMID 41779589
• Neurath C. The sugar industry’s efforts to manipulate research on fluoride effectiveness and toxicity: a ninety-year history. Environ Health 2025-09-01;24(1):62. doi:10.1186/s12940-025-01154-x · PMID 41016930 · PMCID PMC12477810 · open access
• Saveanu CI, Roxana H, Condrea BI, Anistoroaei D, Saveanu AE, Saveanu MS, Golovcencu L. Sociodemographic Determinants of Knowledge and Risk Perception Regarding Community Fluoridation: A Cross-Sectional Study in Iași, Romania. Epidemiologia (Basel) 2026-05-01;7(3):68. doi:10.3390/epidemiologia7030068 · PMID 42201212 · PMCID PMC13214882 · open access
• Till C, El-Sabbagh J, Goodman C, Subiza-Pérez M, Hall M. How the public’s knowledge, attitudes, and practice intersect with scientific evidence about fluoride. Curr Probl Pediatr Adolesc Health Care 2025-06-01;55(6):101768. doi:10.1016/j.cppeds.2025.101768 · PMID 40628571 · PMCID PMC13147203 · open access · cited by 2 (Europe PMC)

Additional documents

• InsureKidsNow (CMS). Summary of Benefits Report for Utah, Medicaid — dental (data as of 04 Feb 2026, print date 03/02/2026) — preventive package the plan pays for when the water carries no fluoride: cleanings twice per calendar year (one additional cleaning may be billed in the same calendar year as a comprehensive oral evaluation), fluoride treatments including fluoride varnish up to four times per calendar year (below age 5 a physician may apply it during a well-child exam; from age 5 it must be done in a dental office to be covered), sealants once every two years per tooth on first and second permanent molars and premolars provided the tooth has no existing decay or fillings, and silver diamine fluoride once every six months per tooth on baby teeth as a non-invasive alternative to fillings (PDF)
• InsureKidsNow (CMS), same report — nitrous oxide analgesia is listed as a non-covered service under the dental plan: it may be used to help a patient relax but is not a separately reimbursable benefit and Medicaid does not cover its cost; behaviour management is not covered; orally administered sedation medications are covered under the Medicaid pharmacy program by prescription only; general anaesthesia and IV conscious sedation are covered only when a child cannot be treated safely under local anaesthesia because of disability or another complex medical condition, documented in the record; inpatient hospital costs are billed to the medical plan (PDF)
• InsureKidsNow (CMS), same report — prior authorisation: Utah Medicaid requires it for all orthodontic treatment, decided with the Utah Medicaid Index of Orthodontic Treatment Need (IOTN) Medical Necessity Score Sheet, the only accepted form, and a child qualifies with at least one Automatic Qualifying Condition or at least two Other Qualifying Conditions; retainers are covered once per lifetime at the end of comprehensive treatment and replacements for lost or broken retainers generally are not; scaling and root planing is limited to one per quadrant per rolling year with prior authorisation, and periodontal maintenance to once every six months with prior authorisation (PDF)

Bibliographic records above were retrieved and verified programmatically from Europe PMC (authors, journal, volume, pages, DOI, PMID, citation count). Coverage, guideline and regulatory statements are quoted from the cited public documents with their dates; verify against the current version before relying on them. This article is not medical or dental advice.