Baby Bottle Tooth Decay: Signs and Treatment

Reviewed 31 August 2026. The claims here rest on one Cochrane review of 23 randomised trials with 25,953 caregivers, a phase III placebo-controlled trial of 830 children, five systematic reviews and meta-analyses, two large prospective cohorts, one cluster-style behavioural trial and three payer/policy documents — all listed with PMIDs or references at the end. Written for parents and for clinicians. Not medical advice; a child with visible cavities, facial swelling or fever needs hands-on care, urgently.

The short answers

• “Baby bottle tooth decay” is not a bottle problem, it is a frequency problem. The pooled estimate from a meta-analysis of longitudinal cohorts is that children with early sugar exposure have an odds ratio of 1.59 (95% CI 1.50 to 1.68) for caries compared with those without — the search retrieved 718 studies, 59 went to full text, 17 were included and nine could be pooled, with low risk of bias as assessed by a Joanna Briggs Institute cohort checklist (Echeverria et al., Brazilian oral research 2025-01-01, PMID 41259577). Number of sips matters more than the drink’s prestige: milk, formula and juice all count as sugar exposure to a tooth.
• The single intervention with moderate-certainty trial evidence is boring and free: advice on the child’s diet and feeding practices, given to pregnant women, mothers and other caregivers in the first year. The 2024 Cochrane review of 23 randomised trials (25,953 caregivers and their children) found a probable 15% reduction in the risk of caries in primary teeth (RR 0.85, 95% CI 0.75 to 0.97; 3 trials, 782 participants), with a slightly lower mean dmfs (MD −0.29, 95% CI −0.58 to 0; 2 trials, 757 participants, low certainty) and a very uncertain dmft result (MD −0.90, 95% CI −1.85 to 0.05 — an interval that touches zero) (Gomersall et al., The Cochrane database of 2024-05-01, PMID 38753314).
• What did not work in that same review: breastfeeding promotion and support changed nothing measurable (RR 0.96, 95% CI 0.89 to 1.03; 2 trials, 1,148 participants; and mean dmft MD −0.12, 95% CI −0.59 to 0.36) (Gomersall et al., The Cochrane database of 2024-05-01, PMID 38753314). That is an important distinction — it is about a programme, not a verdict on breastfeeding.
• Brushing: start early, twice a day, and let the caregiver do it. A 2025 systematic review and meta-analysis of 32 studies (27 cross-sectional, 5 longitudinal) found late brushing initiation was associated with more ECC — pooled OR 1.40 (95% CI 0.91 to 2.16, p = 0.13) across all ten contributing studies with I² above 90%, which becomes OR 1.75 (95% CI 1.50 to 2.03) once one outlier is removed; the authors’ conclusion is that commencing brushing by 12 months and brushing at least twice daily may be protective (Al-Sharani et al., BMC oral health 2025-12-01, PMID 41469640). Read the two numbers as a warning about the literature’s consistency, not just the second one as a recommendation.
• Once a cavity is already open in a young child, there is now a US phase III randomised trial of what a non-invasive option can and cannot do: in 830 generally healthy children aged 12 to 71 months with severe early childhood caries and active cavitated dentine lesions (mean dmft 11.36, SD 4.56, range 1–23), 38% silver diamine fluoride arrested 57.5%, 54.0% and 50.2% of lesions at 3, 6 and 8 months versus placebo 18.8%, 22.5% and 17.4%; between-group differences were 38.7% (99.9% CI 28.7 to 48.6), 31.5% (21.5 to 41.6) and 32.8% (22.3 to 43.2). There were no differences in pain, and 70% of children completed follow-up (Fontana et al., JAMA pediatrics 2026-07-01, PMID 42507437).
• That trial also settles the safety question at scale: adverse events were reported by 196 of the SDF children (47.3%) and 180 of the placebo children (43.3%), mostly mild to moderate, with four severe events (0.6%) overall; treatment-related adverse events were nearly identical between arms (22.9% versus 22.2%), and discontinuation due to adverse events was 7.5% in both (Fontana et al., JAMA pediatrics 2026-07-01, PMID 42507437). “Arrests the cavity” is true; “changes nothing the child feels” is also true, which matters when someone promises your child will feel better.
• Population-scale data show what actually moves the needle at age three. In a linked administrative cohort of 60,404 children in Okinawa, Japan, 15,608 (25.8%) had caries at age 3, and of the 58,815 caries-free at 18 months, 14,269 (24.3%) developed it. Regular juice consumption carried a risk ratio of 1.32 (95% CI 1.28 to 1.35); caregiver-assisted toothbrushing was protective at 0.66 (95% CI 0.64 to 0.68); screen time above 2 hours a day was only 1.08 (1.04 to 1.12) (Yodoshi et al., Community dentistry and 2026-08-01, PMID 42661303).
• Paced feeding practices matter more than the specific fluid in the bottle. In a cross-sectional study of 438 children, feeding practices as such were not associated with early childhood caries, while bedtime bottle use, bedtime snacking, living in a rural area, older age, use of fluoridated toothpaste and having less-educated guardians were significantly associated (Szeto et al., Pediatric dentistry 2026-05-01, PMID 42271613). And in a Beijing prospective cohort of 919 children aged 1–2 years with 12–24 months of follow-up (76.9% retained), significant predictors of caries risk were snack frequency, candy frequency, bedtime brushing frequency, bedtime bottle-use frequency and baseline caries status (Miao et al., BDJ open 2026-05-01, PMID 42069747).
• In Utah Medicaid, the prevention is the covered part and it has explicit limits: cleanings twice a calendar year; fluoride treatments including varnish up to four times per calendar year (children under five can get these at a doctor’s well-child visit, from age five it must be in a dental office); sealants once every two years per tooth on first and second permanent molars and premolars only where the tooth has no decay or filling; and silver diamine fluoride covered for children with baby teeth once every six months per tooth (data as of 04 February 2026; see Additional documents).

What “baby bottle tooth decay” actually is

The clinical name is early childhood caries: the presence of one or more carious surfaces in a child aged six and under. The popular name — baby bottle tooth decay, nursing caries — names one exposure route, and this is where a great many parents get hurt by the label. The disease is not caused by a bottle. It is caused by fermentable carbohydrate sitting on a tooth surface often enough that the plaque bacteria can keep the pH low for long periods across the day, and in a toddler that exposure is usually liquid, usually at night, and usually not followed by cleaning. Breast milk, formula, cow’s milk, juice, and milk with cereal in it all supply the substrate. Water does not.

The bedtime logic is simple. Saliva flow falls during sleep, and saliva is the mouth’s buffer and remineralisation reservoir; a bottle at the end of the day therefore removes the defence and adds the substrate simultaneously. That is also why “brushed after the bottle, then sleep” changes the arithmetic and “falls asleep with the bottle” does not, even when the content is identical.

The teeth look different from adult cavities, which is why they get missed. Early childhood caries typically starts on the upper front teeth — the smooth surfaces behind the lip — and on molar grooves, and it can progress from a chalky white band to a soft brown cavity over months rather than years. “Black tooth in a child” is one of the commonest things parents actually search about; that colour usually means arrested or advanced dentine caries, and it is not a diagnosis of severity — it needs an examination.

The sugar question, with the number attached

The best synthesis available answers “does sugar in the first years cause cavities” using only cohort studies — the design that can look forward in time rather than ask a mother of a child with cavities what she fed her. The review searched five databases from December 2020 to May 2025, retrieved 718 studies, screened to 59 full texts, included 17, and pooled nine: OR 1.59 (95% CI 1.50 to 1.68), which the authors paraphrase as children who consumed sugar in early childhood being 59% more likely to develop caries. Their quality assessment found the included studies at low risk of bias (Echeverria et al., Brazilian oral research 2025-01-01, PMID 41259577).

Two limits should be stated alongside that. First, an odds ratio of 1.59 is a relative figure: on a population where most children get cavities, it means many cases; where almost none do, it means few. Second, “sugar consumption” in these cohorts is a questionnaire variable, and the frequency of exposure is usually what drives the association rather than grams per day. The dietary evidence overall points the same way: a systematic review of nutrition and caries prevention in children, built on 12 studies meeting its inclusion criteria, concludes that a balanced low-sugar diet is crucial and that the relationship between diet and caries frequency in children is consistent (Dipalma et al., BMC oral health 2026-02-01, PMID 41668017). And a scoping review of diet and oral-health prevention, guided by a four-concept framework and searching five databases from 2011 to 2022, identified 107 studies across behavioural practices (33), educational interventions (39) and dietary interventions (35) — the size of that scatter is itself a finding: there is no single dominant, well-tested dietary intervention for this disease (Chamut et al., International journal fo 2024-12-01, PMID 39623427).

What professional programmes have been shown to work

This is the section most advice columns skip, because the answer is unsatisfying and the evidence is genuinely graded. The Cochrane review set out to test everything offered to pregnant women, new mothers and other caregivers of infants in the first year, and found 23 randomised trials (five of them cluster-randomised) involving 25,953 caregivers and children, with at most five trials (1,326 children and 130 mothers) contributing data to any one comparison; many trials had unclear risk of bias due to missing methodological detail, and thirteen recruited socioeconomically disadvantaged participants (Gomersall et al., The Cochrane database of 2024-05-01, PMID 38753314).

• Diet and feeding practice advice for the child versus standard care: RR 0.85 (95% CI 0.75 to 0.97), 3 trials, 782 participants, moderate-certainty. dmfs MD −0.29 (95% CI −0.58 to 0), 2 trials, 757 participants, low certainty. dmft MD −0.90 (95% CI −1.85 to 0.05), 1 trial, 340 participants, very low certainty (Gomersall et al., The Cochrane database of 2024-05-01, PMID 38753314).
• Breastfeeding promotion and support versus standard care: caries RR 0.96 (95% CI 0.89 to 1.03), 2 trials, 1,148 participants, low certainty; dmft MD −0.12 (95% CI −0.59 to 0.36), 2 trials, 652 participants. Translation: these programmes changed how mothers fed, and did not change children’s cavities (Gomersall et al., The Cochrane database of 2024-05-01, PMID 38753314).
• Oral hygiene plus diet and feeding advice (the bundle) versus standard care: RR 0.73 (95% CI 0.50 to 1.07) — 5 trials, 1,326 participants, very low certainty, and the interval crosses zero; mean dmfs MD −0.87 (95% CI −2.18 to 0.43) and dmft MD −0.30 (95% CI −0.96 to 0.36) (Gomersall et al., The Cochrane database of 2024-05-01, PMID 38753314). Note the shape: bigger point estimates than the diet-only comparison, and much wider intervals. Bigger hope, less certainty.
• Diet and feeding advice for the child plus dental care for the mother: RR 0.44 (95% CI 0.05 to 3.95), 2 trials, 324 participants, very low certainty. That interval is so wide it is compatible with anything from a dramatic effect to no effect — and it illustrates how little evidence exists for the “treat mum to protect baby” idea (Gomersall et al., The Cochrane database of 2024-05-01, PMID 38753314).

The review’s own conclusion is the sentence to keep: there is moderate-certainty evidence that advice on diet and feeding to caregivers of children under one year probably leads to a slightly reduced risk of early childhood caries, while everything else is low to very-low certainty and “insufficient for determining which, if any, other intervention types and features may be effective, and in which settings”; the authors also list 13 ongoing studies (Gomersall et al., The Cochrane database of 2024-05-01, PMID 38753314).

Trials outside the review add texture. In the CHALO! randomised trial of 350 low-income Bangladeshi mothers of six-month-olds, six home visits and six phone calls from trained community health workers produced a smaller increase in bottle/sippy-cup use over 18 months than control (Poisson rate ratio 0.36, 95% CI 0.34 to 0.39 versus 0.58, 95% CI 0.56 to 0.61; p < 0.0001) (Debnath et al., International journal of 2026-06-01, PMID 42512142). That is a change in behaviour; the clinical outcome the programme was aiming at is a different question, and this is the recurring structure in ECC prevention — the middle steps move, and the tooth sometimes does not.

A digital version was tested too: a randomised care-based trial of an app-based instruction package for parents of children aged 6 to 72 months, with 20 in the test arm and 23 controls, found parents rated the app suitable (all criteria above 86%), the plaque index improved in the intervention arm (p = 0.01) with no change in controls (p = 0.72), and no difference in dmft was expected or found (Abdul Haq et al., Journal of clinical medi 2023-04-01, PMID 37048763). Forty-three children. Useful as a compliance tool; not a treatment.

Feeding: what the cohorts actually found

This is where the honest answer diverges furthest from the standard parenting advice, so it is worth reading slowly.

The Cochrane comparison of breastfeeding promotion programmes found no caries difference (Gomersall et al., The Cochrane database of 2024-05-01, PMID 38753314), which is consistent with a systematic review of observational and interventional literature on breastfeeding and ECC: 31 studies, 28,000 children; the review reports that exclusive breastfeeding under six months roughly halves the probability of ECC (OR 0.53 to 0.58), while breastfeeding beyond 12 months together with nocturnal feeds raises it by 60% to 86% (OR 2.35 to 7.14 for the risk factors they list, including high plaque levels, feeding to sleep and skipped post-feed cleaning, with parental factors at OR 8.51 to 75.6), and that interventions combining feeding counselling with home visits or visual aids reduced ECC incidence by 22% to 32% (RR 0.68 to 0.78) (Tănase et al., Children (Basel, Switzer 2026-01-01, PMID 41597110). Read those numbers as one coherent statement: exclusive breastfeeding in the first months is protective; it is the pattern of frequent night feeding on top of erupted teeth, without cleaning, that carries the risk — and the same review’s figures on parental factors show the cleaning and plaque part of the picture is bigger than the milk part.

Two longitudinal cohorts sharpen it further. In Toyonaka, Japan, 6,746 children were examined at 42 months and 5,161 had usable data: 13.3% had experienced caries between 18 and 42 months, and logistic regression identified four significant factors — twelve or fewer erupted teeth at 18 months, seventeen or more erupted teeth, breastfeeding, and combined breast-and-bottle feeding at 18 months (Mikasa et al., Scientific reports 2025-11-01, PMID 41309798). A companion analysis of the 1,210 children in the same dataset breastfed for at least 18 months found 24.3% caries experience at 42 months with birth order and caries-risk-test results as significant factors (Otsugu et al., Nutrients 2025-12-01, PMID 41470791). Note that the eruption variables come out as significant alongside feeding: the number of teeth exposed is part of the exposure. A cohort of long-breastfed children where a quarter of them had cavities by 42 months is a better argument for “clean the teeth after night feeds” than for “stop breastfeeding”.

And the Okinawa cohort of 60,404 children puts an effect size on the two most modifiable things: regular juice consumption aRR 1.32 (95% CI 1.28 to 1.35) and caregiver-assisted toothbrushing aRR 0.66 (95% CI 0.64 to 0.68), with screen time above two hours only 1.08 (1.04 to 1.12) and an interaction — the screen-time association was clearer among children who did not regularly drink juice (Yodoshi et al., Community dentistry and 2026-08-01, PMID 42661303). If you only change two things, change these two: the juice and who holds the toothbrush at night.

Brushing, toothpaste and fluoride

“From when?” is a real question with a soft answer. The 2025 meta-analysis described above found the pooled effect for late initiation did not reach significance on the full data set (OR 1.40, 95% CI 0.91 to 2.16, p = 0.13, I² > 90%) and became significant after excluding one outlier (OR 1.75, 95% CI 1.50 to 2.03), concluding in favour of starting by 12 months and brushing at least twice daily while stressing the between-study variability (Al-Sharani et al., BMC oral health 2025-12-01, PMID 41469640). A professional consensus process published in the International Dental Journal — built from an evidence map of systematic reviews and guidelines from 2000 to May 2022 and a survey of 22 draft recommendations among FDI committee members — is the document to read for method (brush twice daily, for two minutes, and the recommendation on how much toothpaste and who supervises) rather than a cohort of questionnaire studies (Glenny et al., International dental jou 2024-06-01, PMID 38052700).

The fluoride varnish arm of prevention has its own evidence, and it is better than most of the material in this article: our dedicated review summarises the Cochrane estimate of a 43% prevented fraction in permanent teeth (95% CI 30 to 57%) and the frequency data: see our fluoride varnish review. For a child already developing lesions, the professional options that have real numbers attached are varnish and SDF; for teeth with deep grooves but no decay, sealants have a Cochrane odds ratio of 0.12 (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120).

One randomised trial deserves mention because it is the kind of “natural alternative” claim parents are marketed, and because it was large and well-run rather than small: a double-blinded multicentre randomised trial enrolled 1,063 children aged 3 to 4 years in 12 kindergartens in Hubei, China, comparing a fluoride-free toothpaste with 7.5% bioactive glass against a fluoride toothpaste (800 ppm F) over 27 months. Caries increment and incidence increased progressively in both groups, with no statistically significant difference between them (Li et al., International dental jou 2026-02-01, PMID 41172677). Read carefully, this is a null result that does not prove equivalence — but it means a fluoride-free option was not better in the one large trial we could verify, and any decision to go fluoride-free in a high-risk toddler is being taken without trial support.

Once there are cavities: what the phase III trial showed

For years, the claim “SDF stops cavities without a drill” circulated on small studies. It now has a phase III, multisite, randomised, blinded, placebo-controlled trial with two parallel arms, recruiting October 2018 to April 2023, in 830 generally healthy children aged 12 to 71 months with severe early childhood caries and active cavitated dentine lesions (Fontana et al., JAMA pediatrics 2026-07-01, PMID 42507437).

• Baseline severity was high: mean dmft 11.36 (SD 4.56; range 1–23). This is a high-risk population, so absolute numbers here are not what a low-risk child’s dentist will see.
• Arrest: 57.5% of lesions at 3 months, 54.0% at 6 months and 50.2% at 8 months under 38% SDF, versus 18.8%, 22.5% and 17.4% under placebo; between-group differences with 99.9% confidence intervals of 38.7% (28.7 to 48.6), 31.5% (21.5 to 41.6) and 32.8% (22.3 to 43.2) (Fontana et al., JAMA pediatrics 2026-07-01, PMID 42507437).
• Pain: no difference between groups (Fontana et al., JAMA pediatrics 2026-07-01, PMID 42507437). This is worth stating plainly, because “child stopped complaining” is a common reason parents choose it.
• Adverse events: 47.3% in the SDF group and 43.3% in placebo reported them; most mild to moderate; four (0.6%) severe. Treatment-related events 22.9% versus 22.2%; discontinuation for adverse events 7.5% in both (Fontana et al., JAMA pediatrics 2026-07-01, PMID 42507437).
• Follow-up reality: 584 of 830 children (70%) completed the study; about 30% were lost, 81 of them (9.8%) because of COVID-19 (Fontana et al., JAMA pediatrics 2026-07-01, PMID 42507437).

Two conclusions follow. SDF arrests about half of active lesions over 8 months, roughly one in three more than placebo — a real effect with a tight interval, in a population selected for severity. And it is a disease-arresting measure, not a repair: the tooth stays visibly stained and physically broken until it is restored, which is the same limitation set out in our SDF review.

How to read this like a clinician

• Ask about frequency, not food names. The pooled cohort estimate for early sugar exposure is OR 1.59 (1.50 to 1.68) (Echeverria et al., Brazilian oral research 2025-01-01, PMID 41259577); the Beijing prospective cohort found snack frequency, candy frequency, bedtime brushing frequency, bedtime bottle-use frequency and baseline caries to be the significant predictors in a ZINB model, with parents’ caries status and saliva-sharing behaviour adding signal in the caries-free subgroup (incidence 29.8%, mean Δdmft 0.94 ± 1.94 over 12–24 months) (Miao et al., BDJ open 2026-05-01, PMID 42069747).
• Prescribe the caregiver advice, not a device. Diet and feeding advice to caregivers under one year has moderate-certainty evidence (RR 0.85, 0.75 to 0.97) (Gomersall et al., The Cochrane database of 2024-05-01, PMID 38753314); bundles with the same content are supported only by very-low-certainty intervals that cross zero.
• Do not weaponise breastfeeding. The programme-level comparison found no difference (RR 0.96, 0.89 to 1.03) (Gomersall et al., The Cochrane database of 2024-05-01, PMID 38753314), and cohort data are consistent with exclusive early breastfeeding being protective (OR 0.53 to 0.58) while night-feeding patterns on erupted teeth raise risk (Tănase et al., Children (Basel, Switzer 2026-01-01, PMID 41597110). The intervention is post-feed cleaning and frequency, not the feeding itself.
• Name the two behavioural levers with the largest verified effect sizes in population data: eliminate regular juice (aRR 1.32, 1.28 to 1.35) and ensure caregiver-assisted brushing (aRR 0.66, 0.64 to 0.68) (Yodoshi et al., Community dentistry and 2026-08-01, PMID 42661303).
• Set expectations for SDF using the trial numbers, not the brochure: about half of lesions arrested at 8 months, no pain difference, ~1 in 4 treatment-related adverse events similar to placebo, and a 30% loss to follow-up in the trial itself (Fontana et al., JAMA pediatrics 2026-07-01, PMID 42507437).
• Use the payer frequency rules as the care schedule where they align with the evidence: varnish up to four times per year and SDF every six months per tooth in Utah Medicaid children, sealants every two years per eligible tooth (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120).
• Document uncertainty honestly: most of this field’s evidence is graded low or very low, and a Cochrane review explicitly calls the ECC-prevention literature insufficient to say which interventions work (Gomersall et al., The Cochrane database of 2024-05-01, PMID 38753314).

Evidence at a glance

Question	Best estimate	Certainty as rated by the source
Early sugar exposure and caries	OR 1.59 (95% CI 1.50 to 1.68); 9 pooled cohorts (Echeverria et al., Brazilian oral research 2025-01-01, PMID 41259577)	Low risk of bias per JBI appraisals
Caregiver diet/feeding advice in year one	RR 0.85 (0.75 to 0.97), 3 trials, 782 (Gomersall et al., The Cochrane database of 2024-05-01, PMID 38753314)	Moderate
Breastfeeding promotion programmes	RR 0.96 (0.89 to 1.03), 1,148 (Gomersall et al., The Cochrane database of 2024-05-01, PMID 38753314)	Low
Bundled advice (hygiene + diet + feeding)	RR 0.73 (0.50 to 1.07), 5 trials, 1,326 (Gomersall et al., The Cochrane database of 2024-05-01, PMID 38753314)	Very low; interval crosses zero
Brushing started late	OR 1.40 (0.91 to 2.16) overall; 1.75 (1.50 to 2.03) after removing an outlier (Al-Sharani et al., BMC oral health 2025-12-01, PMID 41469640)	I² > 90%; consistency poor
Regular juice at age 3	aRR 1.32 (1.28 to 1.35), 60,404 children (Yodoshi et al., Community dentistry and 2026-08-01, PMID 42661303)	Population cohort, adjusted
Caregiver-assisted brushing	aRR 0.66 (0.64 to 0.68) (Yodoshi et al., Community dentistry and 2026-08-01, PMID 42661303)	Same cohort
38% SDF versus placebo in severe ECC	Arrest difference 32.8% (99.9% CI 22.3 to 43.2) at 8 months (Fontana et al., JAMA pediatrics 2026-07-01, PMID 42507437)	Phase III RCT, n = 830, 70% completed
Fluoride-free bioactive glass toothpaste	No significant difference versus 800 ppm fluoride over 27 months, n = 1,063 (Li et al., International dental jou 2026-02-01, PMID 41172677)	Double-blind multicentre RCT
Home-visit programme changing bottle use	Rate ratio 0.36 (0.34 to 0.39) versus 0.58 (0.56 to 0.61) (Debnath et al., International journal of 2026-06-01, PMID 42512142)	RCT; behavioural outcome

Cost and coverage, with the local numbers

For a child in Utah Medicaid, the published benefit summary (data as of 04 February 2026; see Additional documents) sets both the coverage and the ceiling, and the two align almost exactly with the evidence in this article. Covered without argument: dental exams and oral health screenings up to four times per calendar year; cleanings twice a year; fluoride treatments including varnish up to four times per calendar year, with the explicit design that for children under five a doctor may apply it during a well-child visit, while from age five it must happen in a dental office; sealants once every two years per tooth on first and second permanent molars and premolars with no decay or existing filling; silver diamine fluoride on primary teeth once every six months per tooth, listed as a non-invasive alternative to traditional fillings; stainless steel crowns every two years per tooth for teeth with large cavities or after a pulpotomy; simple extractions without prior authorisation when a tooth cannot be saved by a filling or root canal; and emergency care for an abscess — exam, x-rays and incision and drainage — with no prior authorisation at all.

Not covered, or covered only after paperwork: treatment of jaw joint problems, including therapy and occlusal appliances; metal and porcelain crowns on permanent teeth require prior authorisation for medical necessity, once every five years per tooth; root canals on permanent teeth explicitly do not require prior authorisation but are covered only where needed to save a tooth, and third molars are generally excluded; a pulpotomy on a baby tooth is covered without authorisation, once per tooth, and not on a tooth already loose; orthodontics requires the state IOTN score sheet, is limited to once per lifetime, and retainers are covered once per lifetime at the end of active treatment. General anaesthesia and IV sedation are covered only where a child cannot be treated safely under local anaesthesia because of a physical or mental disability or another complex medical condition, with documentation kept in the record.

What that means for a toddler with early cavities: the two things the plan will pay for repeatedly and that carry the best evidence in this article — varnish up to four times a year and, where a tooth is already cavitated, SDF every six months per tooth — are exactly the pair the trials support (Gomersall et al., The Cochrane database of 2024-05-01, PMID 38753314) (Fontana et al., JAMA pediatrics 2026-07-01, PMID 42507437) (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120). Sedation for a “one-visit fix everything” plan is the part that requires proof of necessity. If you are paying privately, the same logic prices out: preventive visits and varnish are the cheap line items, operating under sedation is the expensive one, and everything in the cohorts above (water instead of juice, caregiver brushing twice a day) costs nothing at all (Yodoshi et al., Community dentistry and 2026-08-01, PMID 42661303).

What the evidence does not support

• That the bottle is the cause and the contents are innocent — or the reverse. It is frequency and cleaning; milk and juice both count, and water at bedtime is not the risk anyone claims (Echeverria et al., Brazilian oral research 2025-01-01, PMID 41259577) (Szeto et al., Pediatric dentistry 2026-05-01, PMID 42271613).
• That stopping breastfeeding prevents cavities. The only graded evidence on breastfeeding programmes is that they did not change caries, and early exclusive breastfeeding was associated with lower risk in the review (Gomersall et al., The Cochrane database of 2024-05-01, PMID 38753314) (Tănase et al., Children (Basel, Switzer 2026-01-01, PMID 41597110).
• That juice “counts as fruit”. In a cohort of 60,404 children, regular juice consumption was the second-largest modifiable correlate of caries at age three (Yodoshi et al., Community dentistry and 2026-08-01, PMID 42661303).
• That an app replaces an examination: the app trial was 43 children, and its effect was on plaque index, not on caries (Abdul Haq et al., Journal of clinical medi 2023-04-01, PMID 37048763).
• That “first dental visit at age three” is safe practice: the Utah public benefit summary’s own framing is up to four oral health evaluations per calendar year in childhood, and the cohort data show a quarter of children developing caries between 18 months and three years (Yodoshi et al., Community dentistry and 2026-08-01, PMID 42661303).
• That SDF treats the child’s symptoms: the phase III trial found no difference in pain between arms (Fontana et al., JAMA pediatrics 2026-07-01, PMID 42507437).
• That fluoride-free “reminalising” toothpastes are an equivalent choice in high-risk toddlers: the one large trial we could verify found no advantage over 800 ppm fluoride toothpaste (Li et al., International dental jou 2026-02-01, PMID 41172677).
• That any of this substitutes for hands-on treatment of a cavitated lesion, swelling or fever in a young child. Those need urgent in-person care.

A plan, if you want one

• From the first tooth. Brush twice a day with a smear of fluoride toothpaste, done by the caregiver; the pooled association for late initiation points to starting by 12 months (Al-Sharani et al., BMC oral health 2025-12-01, PMID 41469640) (Glenny et al., International dental jou 2024-06-01, PMID 38052700).
• Then order the liquids. Water is the only thing that lives in a bottle or cup between meals and at night; juice is not a health drink for a toddler and has the strongest dietary signal in the population data (Yodoshi et al., Community dentistry and 2026-08-01, PMID 42661303).
• Break the feed-to-sleep link rather than the feed: finish the feed, then brush, then sleep (Tănase et al., Children (Basel, Switzer 2026-01-01, PMID 41597110).
• Book the first visit early. The covered schedule in Utah Medicaid is up to four evaluations per calendar year, cleanings twice a year and varnish up to four times a year — use them; that is where the professional fluoride goes in.
• Seal and arrest. As permanent molars erupt and are decay-free, seal them (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120). Where lesions are already cavitated and a young child cannot be treated operatively, discuss SDF with expectations set from the phase III numbers (Fontana et al., JAMA pediatrics 2026-07-01, PMID 42507437).
• Watch the two outcomes that matter: new white bands at the gumline, and any brown soft area. Photograph the teeth monthly in the same light — this is the cheapest surveillance that exists.
• If your child has pain, swelling, fever or is not drinking: that is not a prevention conversation. Seek same-day care.

Frequently asked questions

My child only has “milk teeth”. Why bother? Because the disease in primary teeth predicts the disease in permanent ones, and because the trials of caregiver advice reduce it: RR 0.85 (95% CI 0.75 to 0.97) (Gomersall et al., The Cochrane database of 2024-05-01, PMID 38753314). The Beijing cohort also found baseline caries status to be a significant predictor of subsequent caries (Miao et al., BDJ open 2026-05-01, PMID 42069747).

Is the black stain from SDF decay? No: arrested dentine stains dark. The trial that measured the outcome showed about half of lesions arrested and no change in pain (Fontana et al., JAMA pediatrics 2026-07-01, PMID 42507437).

Night feeding — must I stop? The verified evidence does not say stop; it says clean afterwards and mind the frequency, because the promotion trials that changed feeding behaviour did not change caries (Gomersall et al., The Cochrane database of 2024-05-01, PMID 38753314) (Tănase et al., Children (Basel, Switzer 2026-01-01, PMID 41597110).

Is juice with no added sugar fine? The association in a cohort of 60,404 children was with regular juice consumption as such (aRR 1.32), not with added sugar specifically (Yodoshi et al., Community dentistry and 2026-08-01, PMID 42661303).

My toddler refuses the brush. The population signal for caregiver-assisted brushing is the second-largest protective factor measured (aRR 0.66); behavioural tricks are the standard answer because there is no trial of them in this evidence pool (Yodoshi et al., Community dentistry and 2026-08-01, PMID 42661303).

When is the first dental visit? Common guidance says within six months of the first tooth or by age one; nothing in the material verified here supports waiting until three, and coverage in Utah Medicaid permits up to four evaluations per year from infancy.

Does water fluoridation matter here? It changes the baseline on which every one of these relative numbers operates; the trials here “rarely reported background fluoride exposure”, which is why a family in a fluoridated area and a family in a non-fluoridated area can read the same OR and act differently (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120).

Glossary: parent words ↔ clinical words

What you say at home	What is in the chart	How it is measured
“Baby bottle tooth decay”	Early childhood caries (ECC); severe ECC (S-ECC)	dmfs/dmft — decayed, missing, filled surfaces/teeth in primary dentition (Fontana et al., JAMA pediatrics 2026-07-01, PMID 42507437)
“Chalky white line near the gum”	White spot lesion / initial enamel demineralisation	ICDAS codes; visual drying of the surface (Miao et al., BDJ open 2026-05-01, PMID 42069747)
“Black tooth”	Arrested or advanced dentine caries; staining after SDF	Cavitation and activity assessment; arrest rates per lesion (Fontana et al., JAMA pediatrics 2026-07-01, PMID 42507437)
“Nursing to sleep”	Bedtime bottle use; nocturnal feeding	Frequency variables in cohorts: bedtime bottle, bedtime snacking (Miao et al., BDJ open 2026-05-01, PMID 42069747) (Szeto et al., Pediatric dentistry 2026-05-01, PMID 42271613)
“Baby bottle caries risk test”	Caries risk assessment (e.g. Cariostat, ADA CRA)	Questionnaire plus microbiological test in the Japanese cohorts (Otsugu et al., Nutrients 2025-12-01, PMID 41470791)
“The varnish”	Topical fluoride varnish (D1206)	Prevented fraction in permanent teeth; frequency limits in benefit documents fluoride varnish review
“The liquid that stops cavities”	Silver diamine fluoride 38% (D1354 interim arresting medicament)	Lesion-level arrest at 3/6/8 months vs placebo (Fontana et al., JAMA pediatrics 2026-07-01, PMID 42507437)
“Sippy cup”	Bottle/sippy-cup use frequency	Self-reported frequency; Poisson rate ratios in a behavioural RCT (Debnath et al., International journal of 2026-06-01, PMID 42512142)

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

We searched Europe PMC for randomised trials, Cochrane reviews and large prospective cohorts on this topic, then retrieved each record programmatically — authors, journal, volume, issue, pages, DOI, open-access status, citation count — and quoted only numbers present in the fetched abstracts, including the reviews’ own certainty grades. Where an interval touches or crosses zero, we say so where the number appears. Payer statements are quoted from the cited public documents with their dates; check the current version.

What this page cannot tell you: whether your individual child’s spots need a filling or can be managed by cleaning and fluoride (that requires an examination with a probe and possibly radiographs); what your local price or plan allows; whether the specific feeding pattern in your family is the driver; and whether a lesion is active. It also cannot capture the biggest practical constraint in this population, which the trial itself documents — a 30% loss to follow-up among 830 families, meaning that the intervention with the best numbers in this article still depends on getting back to the chair at month six (Fontana et al., JAMA pediatrics 2026-07-01, PMID 42507437).

Sources

Peer-reviewed evidence

• Ahovuo-Saloranta A, Forss H, Walsh T, Nordblad A, Mäkelä M, Worthington HV. Pit and fissure sealants for preventing dental decay in permanent teeth. The Cochrane database of 2017-07-01;7():CD001830. doi:10.1002/14651858.cd001830.pub5 · PMID 28759120 · PMCID PMC6483295 · cited by 204 (Europe PMC)
• Abdul Haq J, Splieth CH, Mourad MS, Vielhauer A, Abdulrahim R, Khole MR, Santamaría RM. Digital Application for Promoting Evidence-Based Children’s Oral Health to Control Early Childhood Caries: Randomized Control Trial on Parental Acceptance and Efficacy. Journal of clinical medi 2023-04-01;12(7):2680. doi:10.3390/jcm12072680 · PMID 37048763 · PMCID PMC10094947 · open access · cited by 9 (Europe PMC)
• Glenny AM, Walsh T, Iwasaki M, Kateeb E, Braga MM, Riley P, Melo P. Development of Tooth Brushing Recommendations Through Professional Consensus. International dental jou 2024-06-01;74(3):526-535. doi:10.1016/j.identj.2023.10.018 · PMID 38052700 · PMCID PMC11123540 · open access · cited by 33 (Europe PMC)
• Gomersall JC, Slack-Smith L, Kilpatrick N, Muthu MS, Riggs E. Interventions with pregnant women, new mothers and other primary caregivers for preventing early childhood caries. The Cochrane database of 2024-05-01;5():CD012155. doi:10.1002/14651858.cd012155.pub3 · PMID 38753314 · PMCID PMC11098061 · cited by 20 (Europe PMC)
• Chamut S, Alhassan M, Hameedaldeen A, Kaplish S, Yang AH, Wade CG, Alghamdi S, Chamut D, Novy BB, Chandel T. Every bite counts to achieve oral health: a scoping review on diet and oral health preventive practices. International journal fo 2024-12-01;23(1):261. doi:10.1186/s12939-024-02279-0 · PMID 39623427 · PMCID PMC11613938 · open access · cited by 12 (Europe PMC)
• Li T, Guo H, Liu C, Jiang H, Gao S, Du M. Effectiveness of Bioactive Glass-Based Toothpaste for Early Childhood Caries Prevention: A Randomized Controlled Trial. International dental jou 2026-02-01;76(1):103985. doi:10.1016/j.identj.2025.103985 · PMID 41172677 · PMCID PMC12615736 · open access · cited by 3 (Europe PMC)
• Echeverria MS, Mathias FB, Schuch HS, Cenci MS, Correa MB, Huysmans MC, Demarco FF. Sugar consumption and early childhood caries: a systematic review and meta-analysis of cohort studies. Brazilian oral research 2025-01-01;39():e122. doi:10.1590/1807-3107bor-2025.vol39.122 · PMID 41259577 · PMCID PMC12628725 · open access · cited by 1 (Europe PMC)
• Mikasa Y, Otsugu M, Kadono M, Matsunami K, Nakamura M, Ohno Y, Kato T, Nakano K. Breastfeeding and early tooth eruption as predictors of dental caries occurrence throughout childhood. Scientific reports 2025-11-01;15(1):42424. doi:10.1038/s41598-025-26515-x · PMID 41309798 · PMCID PMC12660914 · open access · cited by 1 (Europe PMC)
• Al-Sharani HM, Stormon N, Al-Hutbany N, Zhang Y, Zulfiqar T. Optimal tooth brushing initiation age and frequency for preventing early childhood caries: a systematic review and meta-analysis. BMC oral health 2025-12-01;25(1):2006. doi:10.1186/s12903-025-07179-5 · PMID 41469640 · PMCID PMC12754868 · open access · cited by 2 (Europe PMC)
• Otsugu M, Mikasa Y, Kadono M, Matsunami K, Nakamura M, Ohno Y, Kato T, Nakano K. A Longitudinal Study on Dental Caries Focusing on Long-Term Breastfed Children in Japan. Nutrients 2025-12-01;17(24):3846. doi:10.3390/nu17243846 · PMID 41470791 · PMCID PMC12735669 · open access · cited by 1 (Europe PMC)
• Tănase M, Pistol AM, Zmărăndache DDD, Stanciu IA, Munteanu A. Review Regarding the Impact of Breastfeeding on Early Childhood Caries. Children (Basel, Switzer 2026-01-01;13(1):102. doi:10.3390/children13010102 · PMID 41597110 · PMCID PMC12840144 · open access · cited by 3 (Europe PMC)
• Dipalma G, Inchingolo AM, Fiore A, Chieppa S, Carone C, Tartaglia FC, Palermo A, Inchingolo AD, Inchingolo F. Role of nutrition in prevention of dental caries in children and adolescents: a systematic review. BMC oral health 2026-02-01;26(1):495. doi:10.1186/s12903-026-07839-0 · PMID 41668017 · PMCID PMC12997934 · open access · cited by 1 (Europe PMC)
• Miao S, Zhao M, Ren W, Chen W, Zhang H, Liu M. Influence factors of early childhood caries risk among children aged 1-2 years in Beijing: a prospective cohort study. BDJ open 2026-05-01;12(1):44. doi:10.1038/s41405-026-00432-1 · PMID 42069747 · PMCID PMC13135498 · open access
• Szeto A, Khan MW, de Jesus VC, Balshaw R, Menon A, Mittermuller BA, Baltus TH, Chelikani P, Schroth RJ. The Association Between Exclusive Breastfeeding, Bottle-Feeding, and Oral Rothia on Early Childhood Caries. Pediatric dentistry 2026-05-01;48(3):169-176. · PMID 42271613
• Fontana M, Moursi A, Gonzalez-Cabezas C, Khera D, Levy SM, Eckert GJ, Katz BP, Tenuta LMA, Campos MS, Karl E, Boynton J, Chinn C, Sheen A, Kolker J, Warren J, Thorpe E. Efficacy of Silver Diamine Fluoride on Young Children With Severe Early Childhood Caries: A Randomized Clinical Trial. JAMA pediatrics 2026-07-01;. doi:10.1001/jamapediatrics.2026.2567 · PMID 42507437
• Debnath A, Bonuck K, Gao Q, Ramachandran U, Gaur S, Custodio-Lumsden CL, Kopycka-Kedzierawski DT, Kim M, Karasz A. The CHALO! Study Results of a Randomized Controlled Trial to Reduce Risk of Childhood Dental Caries and Obesity. International journal of 2026-06-01;23(7):837. doi:10.3390/ijerph23070837 · PMID 42512142
• Yodoshi T. Screen Time, Juice Consumption, Caregiver-Assisted Toothbrushing and Dental Caries at Age 3 Years: A Population-Based Cohort Study in Japan. Community dentistry and 2026-08-01;. doi:10.1111/cdoe.70112 · PMID 42661303

Additional documents

• InsureKidsNow (CMS). Summary of Benefits Report for Utah, Medicaid — dental (data as of 04 Feb 2026) — cleanings twice per calendar year; fluoride treatments including varnish up to four times per calendar year (by a doctor at a well-child visit under age 5, in a dental office from age 5); sealants once every two years per tooth on first and second permanent molars and premolars without decay or restorations; silver diamine fluoride every six months per tooth on primary teeth; stainless steel crowns every two years per tooth; emergency abscess treatment without prior authorisation; orthodontics once per lifetime via the Utah IOTN sheet (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.