Dental Sealants: Do They Prevent Cavities?

Reviewed 31 August 2026. Numbers come from three Cochrane reviews, one real-world electronic-health-record study of 7,299 teeth, four systematic reviews/meta-analyses, three trials, two material-science studies, two exposure studies on the bisphenol-A question, and one US payer document — each listed with its PMID or reference at the end. Written for parents deciding about a toothbrush-and-floss visit, and for clinicians writing the note. Not medical advice.

The short answers

• A pit-and-fissure sealant is a thin resin or glass-ionomer layer painted into the chewing grooves of back teeth, where most childhood cavities start. It is one of the few dental interventions with a Cochrane review that reports an effect clearly away from zero: over 24 months, resin-based sealants versus no sealant gave an odds ratio of 0.12 (95% CI 0.08 to 0.19) in children aged 5 to 10, based on 7 trials, 1,548 children randomised and 1,322 evaluated, graded moderate-quality evidence (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120).
• The same review translates that into teeth you can picture: if 160 per 1,000 control tooth surfaces went decayed over 24 months, sealing would bring it down to 52 per 1,000 (95% CI 31 to 74); at a 40% baseline it goes to 62.5 per 1,000 (95% CI 38.4 to 96.3); at a 70% baseline, to 19% (95% CI 12.3 to 27.2) (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120). The benefit is real and it is larger where risk is higher — which is the single most useful fact for deciding who needs them.
• Protection does not vanish at two years, but the evidence gets thinner: at 48 to 54 months the odds ratio was 0.21 (95% CI 0.16 to 0.28; 4 trials, 482 children evaluated), and the risk ratio 0.24 (95% CI 0.12 to 0.45; 203 children) (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120).
• Sealants versus fluoride varnish is a genuinely unresolved comparison, and anyone who tells you otherwise is selling something. Across 11 trials with 3,374 participants aged five to ten, resin sealant compared with varnish gave OR 0.67 (95% CI 0.37 to 1.19; I² = 84%; 4 studies, 1,683 children) — the interval crosses zero and the heterogeneity is severe. The reviewers’ own conclusion: “it has not been possible in this review to reach reliable conclusions about which one is better to apply” (Kashbour et al., The Cochrane database of 2020-11-01, PMID 33142363).
• Doing both beats doing one — in the one trial that tested it: sealant plus varnish versus varnish alone, OR 0.30 (95% CI 0.17 to 0.55) at two years in 92 children, described by the authors as a clinically meaningful 77% reduction, but rated very-low-certainty evidence (Kashbour et al., The Cochrane database of 2020-11-01, PMID 33142363).
• On baby teeth, the evidence is not equivalent, and this matters because paediatric practices routinely sell sealants for primary molars. The 2022 Cochrane review of sealants in primary teeth found only nine studies (1,120 children randomised, ages 18 months to eight years); for the fluoride-releasing resin versus no sealant comparison the result was a non-significant Becker-Balagtas OR 0.76 (95% CI 0.41 to 1.42; one study, 85 children, 255 tooth surfaces), and for glass-ionomer sealants one study found nothing (OR 0.97, 95% CI 0.63 to 1.49; 449 children) while another found a dramatic benefit (OR 0.03, 95% CI 0.01 to 0.15; 107 children). The reviewers rated the certainty low or very low and described the evidence base as fragile (Ramamurthy et al., The Cochrane database of 2022-02-01, PMID 35146744).
• The practical failure mode is not biology, it is moisture. In a study of 160 extracted human molars assigned to eight groups, saliva contamination “markedly reduced bond strength and increased leakage”, while re-etching effectively restored enamel receptivity; resin and flowable composite showed less microleakage and higher shear bond strength than glass-ionomer materials (p < 0.05) (Gok et al., PloS one 2026-01-01, PMID 42384668). If your child’s sealant goes on while the tooth is drooling, you are buying a shorter service life.
• Adverse events: in the sealant-versus-varnish review, five trials with 1,801 participants reported no adverse events over one to nine years of follow-up (Kashbour et al., The Cochrane database of 2020-11-01, PMID 33142363), and in the permanent-teeth review only four of 38 trials assessed adverse events at all, and none was reported (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120). The one adverse event recorded in the primary-teeth review was a gag reflex during placement (Ramamurthy et al., The Cochrane database of 2022-02-01, PMID 35146744). Absence of harm here is mostly absence of measurement, and that is worth naming.
• On the bisphenol-A question — the thing most likely to keep a parent awake — the honest answer is “tiny, measurable, within limits, long-term unknown”. In a prospective cohort of 67 Thai schoolchildren (mean age 9.9 ± 1.3 years, 2.9 ± 1.9 teeth sealed), median adjusted urinary BPA rose from 0.01 µg/g creatinine before placement to 0.03 on day 1, 0.19 on day 7 and 0.23 on day 14, with a significant association between the number of sealed teeth and BPA level at every visit; the authors call it low-level exposure accepted by international authorities, but “a silent, chronic, and persistent systemic event” whose long-term implications are not yet deciphered (Supornsilchai et al., JDR clinical and transla 2026-01-01, PMID 40077867).
• In the US, expect coverage with rules. Utah’s Medicaid children’s dental benefit covers sealants once every two years per tooth, limited to first and second permanent molars and premolars, and only where the tooth has no existing decay or filling (data as of 04 February 2026; see Additional documents). That last clause is the one that generates surprise bills.

What a sealant is, and why the groove is the whole argument

Cavities in children are not evenly distributed over the tooth surface. They cluster in pits and fissures — the grooves where the cusps of a molar meet — because a toothbrush filament is wider than many of those grooves, so the biofilm inside them is not reachable by brushing alone. Sealants are a liquid resin or glass-ionomer flowed into those grooves and set, leaving a smooth, cleansable surface with the anatomy buried underneath. That is the entire mechanism, and it is why the intervention works on chewing surfaces and not on the flat sides between teeth, where the evidence for sealants is simply not applicable.

Two consequences follow that people usually miss. First, a sealant is a physical barrier with no biological activity of its own (glass-ionomer variants also release fluoride, which is a real but secondary effect). Second, the amount of benefit depends on how much decay was going to happen without it — which is exactly what the Cochrane absolute-risk arithmetic shows: the same odds ratio of 0.12 turns a 16% two-year risk into 5.2%, and a 70% risk into 19% (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120). A sealant on a low-risk tooth prevents little; on a high-risk tooth it prevents a lot. This is why “seal all eight molars in every child” and “sealants are overrated” are both wrong at the same time.

The Cochrane evidence, with the caveats attached

The permanent-teeth review is the backbone. It is an update of reviews published in 2004, 2008 and 2013; searches ran to 3 August 2016 across the Cochrane Oral Health register, CENTRAL, MEDLINE and Embase with no language or date limits; eligible trials had at least 12 months of follow-up and were in people under 20. Thirty-eight trials with 7,924 children qualified, of which seven (1,693 participants) were new to this update (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120). The structure of the evidence is worth knowing because it explains the certainty rating:

• 15 trials tested resin-based sealant against no sealant (3,620 participants across 14 studies, plus 575 tooth pairs in one); three tested glass-ionomer against no sealant (905 participants); and 24 compared one sealant with another (4,146 participants).
• Children were 5 to 16 years old. Trials “rarely reported background fluoride exposure or baseline caries prevalence” — a gap that makes every one of these estimates harder to transport to your child, because fluoride exposure is the main driver of the baseline.
• Blinding of outcome assessment was rated at high risk of bias for all trials, for a physical reason: the assessor can see the sealant. So “caries present or absent” is recorded by someone who knows which group the tooth belonged to. That is the biggest single threat to the headline effect, and it applies in the direction that usually favours the intervention.

The glass-ionomer numbers deserve a special caution, because glass-ionomer sealants are popular in paediatric practice for exactly the reason the data are weak — they tolerate moisture better. Against no sealant at 24 months, results were “inconclusive”, very-low-quality evidence; and the comparison of one sealant type against another had “insufficient evidence” overall, across 24 trials with wildly varying materials, outcomes and follow-ups (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120).

Sealant or varnish? The question the literature cannot settle

Both work; which is better is open. The dedicated Cochrane comparison (an update of reviews from 2006, 2010 and 2016) included 11 trials with 3,374 participants aged five to ten when trials started; two trials contributed no data (Kashbour et al., The Cochrane database of 2020-11-01, PMID 33142363).

• Resin sealant versus fluoride varnish: seven trials assessed it, five contributed data. OR 0.67 (95% CI 0.37 to 1.19), I² = 84%, 4 studies, 1,683 children — “uncertain if resin-based sealants may be better than fluoride varnish, or vice versa”. Very low certainty.
• One trial that measured surfaces rather than teeth found a small advantage for sealant: DMFS mean difference −0.09 (95% CI −0.15 to −0.03) and DMFT −0.08 (95% CI −0.14 to −0.02) at two years, 542 participants — statistically away from zero, and the reviewers say it “may not be clinically significant”. A tenth of a surface over two years is a real number that means almost nothing for one child.
• One small study, rated high risk of bias, reported a clear sealant advantage at four years (RR 0.42, 95% CI 0.21 to 0.84) and at nine years (RR 0.48, 95% CI 0.29 to 0.79) in 75 children — the longest follow-up in the review, and the least trustworthy design.
• Glass-ionomer-based sealant versus varnish: three trials, too clinically diverse to pool; in general no benefit of one over the other at one, two or three years, with one study (which also gave oral health education) suggesting sealants help high-risk children more.

Read that list as a whole and a fair conclusion emerges: for a child at average risk, choosing sealant over varnish or varnish over sealant is a choice between two roughly comparable things, and the strongest defensible position in the review is that doing both is better than either alone (Kashbour et al., The Cochrane database of 2020-11-01, PMID 33142363). That also means: if a clinic offers you “sealants instead of fluoride varnish” as an either/or upsell, the literature does not support the “instead”.

The comparison with our fluoride-varnish evidence is instructive. Varnish, applied on all surfaces, has a permanent-tooth prevented fraction of 43% (95% CI 30 to 57%) in its own Cochrane review of 22 trials and 12,455 randomised children, and that estimate also comes with real limitations (our fluoride varnish review). Sealants, applied only on occlusal surfaces, have a much bigger relative effect in a smaller, unblinder-able trial base. They are not substitutes; on a high-risk molar they stack.

Primary (baby) teeth: a weaker claim, said carefully

Sealing baby molars is widely offered and it is where the evidence is thinnest. The 2022 Cochrane review set out to test it directly and ended up with nine studies, 1,120 randomised children aged 18 months to eight years, of which only three provided usable data for the sealant-versus-nothing comparison — and the reviewers chose not to pool them because ages and follow-up differed (Ramamurthy et al., The Cochrane database of 2022-02-01, PMID 35146744).

The individual numbers tell the story of fragility better than any summary: fluoride-releasing resin versus no sealant at 24 months, Becker-Balagtas OR 0.76 (95% CI 0.41 to 1.42) — no significant difference. Glass-ionomer versus no sealant: one study, 449 children, OR 0.97 (95% CI 0.63 to 1.49) — nothing; another, 12 months, 107 children, OR 0.03 (95% CI 0.01 to 0.15) — an almost total effect that no one should believe from one trial. Material comparisons (six trials, 411 children, only 221 with data) showed low caries incidence across all arms and were graded low or very low. Only one study assessed adverse events at all, and reported gag reflex during placement (Ramamurthy et al., The Cochrane database of 2022-02-01, PMID 35146744).

So: for primary molars the reasonable clinical position is “may help, especially when the tooth is deep-pitted and the child is high-risk; not an evidence-based guarantee; not a reason to postpone anything else.” Where the child already has a cavitated lesion, a sealant is the wrong tool — see the section below.

What actually makes a sealant fail, and how to check yours

Sealants fail by coming off or by leaking at the margin. The evidence points at three controllable things.

Moisture control at placement. In a bench study of 160 extracted human molars in eight groups (three material types, with and without adhesive, with and without saliva contamination), saliva contamination markedly reduced bond strength and increased microleakage; re-etching restored enamel receptivity and sealing ability; and scanning electron microscopy showed continuous margins for resin and flowable composite but irregular interfaces with micro-gaps in the contaminated groups. Adhesive application increased bond strength but slightly increased microleakage — a good example of a material “improvement” that is not a clean win (Gok et al., PloS one 2026-01-01, PMID 42384668). In practice: rubber dam or good isolation with a cooperative child is part of the procedure, not a nicety.

Material choice. Resin sealants keep better than glass-ionomer: across 13 randomised trials comparing them in permanent molars, resin showed “consistently higher retention rates” while caries-preventive efficacy was “generally comparable” — with heterogeneity of I² = 86.6%, and no small-study effect on funnel plot or Egger’s test. The authors note explicitly that certainty was not graded with GRADE (Kaur et al., Journal of the Indian So 2025-10-01, PMID 41235550). A double-blind randomised trial of 400 glass-ionomer-based sealants (three compositions on mandibular first molars, reviewed at 3, 6, 12, 18 and 24 months) found two of the three formulations had significantly better retention than the third (p < 0.001), with mean survival 20.47 and 19.53 months, and no occlusal caries development or upward ICDAS-II transitions in any group over 24 months — which is the kind of null that should be read as “under these conditions, retention is not the same as protection” (Kucukyilmaz Izgi et al., Scientific reports 2026-04-01, PMID 42045411).

Hydrophilic versus hydrophobic resin. This is sold as a difference that matters. A meta-analysis of randomised trials retrieved 20,945 records, kept 14 for quantitative synthesis (five at high risk of bias, five with some concerns, four low), and found a significant difference for caries development (OR 0.490, 95% CI 0.277 to 0.867; p = 0.014) but not for retention (OR 0.859, 95% CI 0.596 to 1.237; p = 0.414) — with both outcomes graded very low quality. The authors’ own conclusion is that the two are “approximately equal”, with a practical note that the hydrophilic version may be preferable when isolation is difficult (Gheidari et al., Dental research journal 2026-01-01, PMID 41777750).

How to check your own: sealants should be examined at each recall. Ask the dentist to state retention (complete / partial / lost) and marginal status, because “sealant intact” is the finding that makes the whole prevention claim hold over time, and the trial literature shows retention is where the variation is (Kaur et al., Journal of the Indian So 2025-10-01, PMID 41235550) (Kucukyilmaz Izgi et al., Scientific reports 2026-04-01, PMID 42045411).

Special cases the searches actually ask about

White spots after braces. A white spot after orthodontics is usually demineralisation, and the sealant question there is the wrong shape: what the literature supports is fluoride and plaque control, and sealants are for intact grooves. What we do have for the adjacent problem — hypomineralised molars (MIH) — is one 18-month randomised trial in 136 children aged 6 to 9 with affected first molars, comparing a glass-ionomer sealant (Fuji Triage) with Er,Cr:YSGG laser therapy. Caries progression rose slightly overall (22%) with no significant difference between groups (RR 1.22, p = 0.210); the sealant gave more sustained protection against post-eruptive breakdown (Baraka et al., Journal of dentistry 2026-03-01, PMID 41577163). If your child has MIH molars, the useful question is not “sealant or laser” but “how do we keep this tooth from breaking down”.

Sealant versus filling. A sealant is prevention for a surface that has not cavitated; a filling is treatment for a surface that has. That is not a philosophical distinction, it is the inclusion criterion: the primary-teeth review excluded studies that used sealants in cavitated lesions outright (Ramamurthy et al., The Cochrane database of 2022-02-01, PMID 35146744). Once there is a hole, “seal it and watch” is a different intervention with different evidence, and the relevant Cochrane comparison is about treating dentine lesions, where sealed restorations and semi-permanent restorations behave differently from conventional ones (Schwendicke et al., The Cochrane database of 2021-07-01, PMID 34280957). Our article on silver diamine fluoride covers the arrest-not-repair option.

Adults. “Do adults benefit?” is a fair question with a thin answer: the trial populations were aged 5 to 16 (permanent-teeth review) and under 20 by inclusion criterion (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120). Extrapolation to a 35-year-old with deep grooves and no decay is reasonable mechanistically and unproven clinically — and for an adult whose molars are already restored, the evidence base simply does not apply.

Why good sealants are still under-used — measured

This is the part of the story that has nothing to do with efficacy. In a study of routine electronic health records from a large US dental network, researchers followed 7,299 teeth diagnosed with an occlusal non-cavitated carious lesion for two years — the exact indication the ADA recommends sealing. Dentists restored 591 of those teeth and sealed 164. The sealant application rate among eligible teeth was 2.2%. Sealed teeth did better: 8.2% of unsealed teeth progressed to a restoration versus 3.0% of sealed teeth (RR 0.37, 95% CI 0.16 to 0.88; p = 0.02), and multilevel survival analysis found unsealed teeth were restored sooner (adjusted HR 0.11, 95% CI 0.03 to 0.36 for sealed teeth) (Shah et al., Caries research 2025-01-01, PMID 39154643). It also found the practice was concentrated: 1.9% of providers placed more than half of all the sealants.

That is an adoption gap inside a guideline-supported indication, and it is why the “before and after” photos you can find online show so few sealed teeth. And the clinician-side explanation is documented too: a systematic review and meta-analysis of dentists’ knowledge, attitudes and practices toward sealants found high knowledge proportions (0.75 knowing how to use them, 0.75 performing caries risk assessment) but inconsistent behaviour — routine sealant use at 0.55, belief that they prevent caries at 0.69 with a prediction interval of 0.08 to 1.00 (Vaja et al., Journal of the American 2026-07-01, PMID 42446467). When the prediction interval spans 0.08 to 1.00, “the profession is undecided” is the accurate summary.

Payer design changes use, and a natural experiment shows it: with data from the Korean Youth Risk Behavior Survey 2005–2024 and the national nutrition/oral examination survey 2007–2015 among 12- to 18-year-olds, sealant use rose until 2018 (annual percent change +1.5%) then declined slightly, DMFT and filled-tooth rates fell over 2007–2015, and inclusion of sealants under insurance coverage was significantly associated with higher sealant use and lower DMFT and filled-teeth rates (Kim et al., Journal of public health 2026-06-01, PMID 41807286). A systematic evaluation of the economics, covering 19 studies selected from 874 records, adds why the arithmetic is hard: cost-effectiveness conclusions depended on baseline caries prevalence, willingness to pay, follow-up length, delivery setting, sealant retention rate and varnish reapplication interval — and the authors concluded the existing evidence was too limited to declare either method more cost-effective (Zhang et al., The Journal of clinical 2023-09-01, PMID 37732430).

Bisphenol A: what is actually measured in children

Resin-based sealants contain Bis-GMA, a BPA derivative, and “does the sealant leach estrogen-mimicking chemical into my kid” is a real search query. Two sources, read together, give the defensible answer.

The exposure study: 67 Thai primary-school children (50.7% boys, mean age 9.9 years, mean 2.9 ± 1.9 sealed teeth, range 1–11) in a voluntary school sealant programme, with urinary BPA measured before placement and on days 1, 7 and 14 by HPLC, adjusted for creatinine. Medians: 0.01, 0.03, 0.19 and 0.23 µg/g creatinine. More sealed teeth meant higher levels at every visit. The authors’ framing is the one to keep: exposure was low and within limits approved by international authorities, yet they describe it as “a silent, chronic, and persistent systemic event, the long-term implications of which are yet to be deciphered”, and their Knowledge Transfer statement says this should inform dental practice and policy assessment (Supornsilchai et al., JDR clinical and transla 2026-01-01, PMID 40077867).

The context: a narrative review of BPA release from dental materials notes that the amalgam phaseout has pushed use of resin materials up, that food-contact materials remain the dominant exposure route with dental contribution minor, and that risks cannot simply be dismissed because BPA shows a non-monotonic dose–response. It also records why the debate is unsettled: in 2023 the European Food Safety Authority proposed a 20,000-fold reduction of the tolerable daily intake to 0.2 ng per kg body weight on the basis of immune effects at extremely low doses — a limit that would effectively ban BPA in food contact materials — and it says plainly that data interpretation is compromised by methodological inconsistency across studies (material composition, specimen preparation, extraction media, duration, analytical methods) and by incompatible reporting (Tichy et al., Journal of dental resear 2025-09-01, PMID 40524375).

Translated into a decision: if you want to minimise measured BPA exposure, the levers are (1) fewer sealed teeth — seal only what the risk analysis justifies, which is also what the efficacy arithmetic says (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120); (2) sealed teeth that are actually indicated, since an adult with already-restored molars gains nothing to justify any exposure (Ramamurthy et al., The Cochrane database of 2022-02-01, PMID 35146744); and (3) not treating a fresh sealant as a reason to skip fluoride, brushing and diet changes, which carry no such question at all. What you should not do is decline a sealant on a high-risk first permanent molar on BPA grounds: the trial evidence for the benefit is moderate-quality and the exposure study found levels below the accepted limits.

How to read this like a clinician

• Indicate by risk, not by age. The effect is a constant odds ratio against a variable baseline: 0.12 at any baseline, but that means very different absolute numbers (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120). Record the baseline judgement (deep anatomy, lesion activity, fluoride exposure, caries in siblings) so the decision can be audited.
• Isolation is the procedure. Report it in the note. Contamination is the one mechanism with a bench study behind it (Gok et al., PloS one 2026-01-01, PMID 42384668).
• Resin first for retention; glass-ionomer where moisture control is impossible, and know that the evidence for GIC versus nothing is inconclusive (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120) (Kaur et al., Journal of the Indian So 2025-10-01, PMID 41235550).
• Do not present sealant and varnish as competitors: the comparative review is undecided, and the combined arm is the only one that clearly favoured an intervention, at very low certainty (Kashbour et al., The Cochrane database of 2020-11-01, PMID 33142363).
• Do not seal cavitated lesions and call it prevention — outside the inclusion criteria of every review here (Ramamurthy et al., The Cochrane database of 2022-02-01, PMID 35146744) (Schwendicke et al., The Cochrane database of 2021-07-01, PMID 34280957).
• In primary molars, document that this is an uncertain indication with low-certainty evidence; the two GIC trials point in opposite directions (Ramamurthy et al., The Cochrane database of 2022-02-01, PMID 35146744).
• Reassess retention at every recall; survival of the better materials in a 400-sealant trial was around 19.5 to 20.5 months (Kucukyilmaz Izgi et al., Scientific reports 2026-04-01, PMID 42045411).
• If asked about BPA, quote the numbers rather than the folklore: 0.01 → 0.19/0.23 µg/g creatinine at day 7/14 in 67 children, within approved limits, long-term unknown (Supornsilchai et al., JDR clinical and transla 2026-01-01, PMID 40077867) (Tichy et al., Journal of dental resear 2025-09-01, PMID 40524375).

Evidence at a glance

Question	Best estimate	Certainty as rated by the source
Resin sealant versus nothing, permanent molars, 24 months	OR 0.12 (95% CI 0.08 to 0.19); 7 trials, 1,548 randomised (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120)	Moderate
Same, 48–54 months	OR 0.21 (0.16 to 0.28); RR 0.24 (0.12 to 0.45) (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120)	Lower quantity and quality
Glass-ionomer versus nothing, permanent teeth	Inconclusive at 24 months (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120)	Very low
Sealant versus fluoride varnish	OR 0.67 (0.37 to 1.19), I² 84%, 1,683 children (Kashbour et al., The Cochrane database of 2020-11-01, PMID 33142363)	Very low; undecided
Sealant plus varnish versus varnish	OR 0.30 (0.17 to 0.55), 92 children, 2 years (Kashbour et al., The Cochrane database of 2020-11-01, PMID 33142363)	Very low
Sealants in primary molars	BB OR 0.76 (0.41 to 1.42); conflicting GIC studies (Ramamurthy et al., The Cochrane database of 2022-02-01, PMID 35146744)	Low to very low; “fragile”
Real-world effect on progression to filling	3.0% vs 8.2%; RR 0.37 (0.16 to 0.88); aHR 0.11 (0.03 to 0.36) (Shah et al., Caries research 2025-01-01, PMID 39154643)	Observational EHR, 7,299 teeth
Resin versus glass-ionomer retention	Resin consistently higher; comparable caries effect; I² 86.6% (Kaur et al., Journal of the Indian So 2025-10-01, PMID 41235550)	Not graded with GRADE by authors
Hydrophilic versus hydrophobic	Caries OR 0.490 (0.277 to 0.867); retention null (Gheidari et al., Dental research journal 2026-01-01, PMID 41777750)	Very low
Adverse events	None reported where sought: 1,801 participants, 1–9 years (Kashbour et al., The Cochrane database of 2020-11-01, PMID 33142363); gag reflex in primary-teeth review (Ramamurthy et al., The Cochrane database of 2022-02-01, PMID 35146744)	Only 4 of 38 trials assessed them (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120)
BPA exposure after sealing	Median urinary BPA 0.01 → 0.19/0.23 µg/g creatinine at days 7/14, n = 67 (Supornsilchai et al., JDR clinical and transla 2026-01-01, PMID 40077867)	Cohort without control group
Use despite guideline support	2.2% of eligible teeth sealed; 1.9% of providers placed over half (Shah et al., Caries research 2025-01-01, PMID 39154643)	Direct measurement

Cost and coverage, with the local numbers

Sealants are cheap relative to the rest of dentistry and still a source of billing surprises, so here is what the public documents actually say. Utah’s Medicaid children’s benefit covers sealants, with a frequency of once every two years per tooth, restricted to the first and second permanent molars and premolars, and only where the tooth is caries-free and unrestored (Summary of Benefits Report for Utah, Medicaid; data as of 04 February 2026). The last clause is the one that surprises families: a sealant on a tooth that already has decay or a filling is a different code and a different decision, and the plan simply does not buy it. In the same document, fluoride treatments including varnish are covered up to four times per calendar year, cleanings twice a year, and silver diamine fluoride every six months per tooth on primary teeth.

Two practical consequences. (1) Ask for the sealant on the indication the payer recognises — a deep-pitted, unrestored permanent molar in a child — because that is also the indication with moderate-quality evidence (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120). (2) Ask what the plan pays for replacement before you assume a partly lost sealant gets repaired free: the frequency rules are per tooth and per interval, not per clinical need. Coverage in other states and in private plans varies; nothing in this page substitutes for the actual benefit booklet.

What the evidence does not support

• That sealants and varnish are interchangeable, or that one is better. The head-to-head review says the question is unresolved, with I² = 84% across its pooled comparison (Kashbour et al., The Cochrane database of 2020-11-01, PMID 33142363).
• That sealants on baby molars are evidence-based in the same sense as on permanent first molars — the underlying evidence was graded low or very low and described as fragile (Ramamurthy et al., The Cochrane database of 2022-02-01, PMID 35146744).
• That glass-ionomer sealants prevent more caries than nothing in permanent teeth: “inconclusive”, very-low-quality evidence (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120).
• That a specific brand or “hydrophilic” formulation is clinically superior: retention differences were not significant and both outcomes were very low quality (Gheidari et al., Dental research journal 2026-01-01, PMID 41777750).
• That sealants are “proven safe” in the way the phrase is used in marketing. They are un-reported-on: only four of 38 trials assessed adverse events (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120), and only one in the primary-teeth review (Ramamurthy et al., The Cochrane database of 2022-02-01, PMID 35146744). Nobody has shown harm; the honest sentence is that systematic harm measurement in this literature is almost absent.
• That sealants replace brushing, fluoride toothpaste, diet change or recall visits. They act on grooves; the majority of the rest of the tooth surface is out of scope, and the varnish evidence covers a different set of surfaces (Kashbour et al., The Cochrane database of 2020-11-01, PMID 33142363).
• That a sealant can be painted over an open cavity to “stop” it. Studies of cavitated lesions were excluded from the preventive reviews, and treating dentine lesions is a separate evidence question with different failure rates (Ramamurthy et al., The Cochrane database of 2022-02-01, PMID 35146744) (Schwendicke et al., The Cochrane database of 2021-07-01, PMID 34280957).
• That adult sealing has trial support: the evidence base is 5- to 16-year-olds (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120).

A plan for the dental visit, if you want one

• Ask which grooves and why. “All eight molars, because that is our policy” is a weaker answer than “the two first permanent molars are deep-pitted, stain-holding and uncleansable; the others are self-cleansing”. The absolute benefit depends on that judgement (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120).
• Ask how isolation was achieved and whether the tooth was caries-free and unrestored at the time — the latter also decides coverage (Gok et al., PloS one 2026-01-01, PMID 42384668).
• Ask for varnish as well, on schedule (Utah Medicaid allows up to four times a year), not instead (Kashbour et al., The Cochrane database of 2020-11-01, PMID 33142363).
• Ask for retention to be checked and recorded at each recall, and ask what the plan covers if a sealant partly falls off (Kaur et al., Journal of the Indian So 2025-10-01, PMID 41235550).
• If you are worried about BPA, say so and negotiate the number of sealed teeth rather than refusing the concept — and ask about glass-ionomer or BPA-free formulations as a trade-off you are knowingly making on retention data, not on safety data (Supornsilchai et al., JDR clinical and transla 2026-01-01, PMID 40077867) (Gheidari et al., Dental research journal 2026-01-01, PMID 41777750).
• For the child’s whole mouth, the ranking does not change: brushing with fluoride toothpaste twice daily, keeping free sugars between meals, and regular recall are the things with effect on every surface — see our pieces on fluoride varnish, xylitol and SDF.

Frequently asked questions

How long do sealants last? Measured retention in a randomised trial of glass-ionomer-based materials was about 19.5 to 20.5 months on average, with material-dependent differences; resin sealants retain better than glass-ionomer across 13 trials (Kucukyilmaz Izgi et al., Scientific reports 2026-04-01, PMID 42045411) (Kaur et al., Journal of the Indian So 2025-10-01, PMID 41235550). The caries-preventive effect in trials is reported at 24 and 48–54 months (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120). “Lasts for years” is a sales phrase; “check retention at recall” is the evidence.

Do they hurt? Does my child need freezing? No injection is involved: etching, rinsing, drying, painting, light-curing. The only procedural adverse event recorded in the primary-teeth review was gag reflex during placement (Ramamurthy et al., The Cochrane database of 2022-02-01, PMID 35146744).

My child’s sealant fell off. Is that a failure? It is a retention event, and the trials show retention is the weak link, not the biology (Kaur et al., Journal of the Indian So 2025-10-01, PMID 41235550). Repair is a different coverage question — check the frequency rule.

Can sealants cause cavities under them? That is the classic fear. What the verified trials support: in the 400-sealant trial, no occlusal caries development and no upward ICDAS-II transitions occurred in any group over 24 months (Kucukyilmaz Izgi et al., Scientific reports 2026-04-01, PMID 42045411); and the permanent-teeth review found lower caries in sealed teeth, not a hidden reservoir of it (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120). A sealant over an active, cavitated lesion is a different situation and outside the evidence base of these reviews (Ramamurthy et al., The Cochrane database of 2022-02-01, PMID 35146744).

Are sealants better than fillings? They prevent rather than repair, so the comparison is about the decision to seal early versus restore later. The one real-world dataset that measures the outcome directly: 3.0% of sealed teeth needed a restoration within two years versus 8.2% of unsealed ones (Shah et al., Caries research 2025-01-01, PMID 39154643).

Should an adult get them? Mechanistically plausible, unproven in trials: the trial populations were children and adolescents up to 20 (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120).

Do they contain BPA? They contain a Bis-GMA resin, and measurable BPA appears in urine afterwards at low levels within approved limits, with unknown long-term implications (Supornsilchai et al., JDR clinical and transla 2026-01-01, PMID 40077867) (Tichy et al., Journal of dental resear 2025-09-01, PMID 40524375).

Glossary: parent words ↔ clinical words

What you hear at home	What is in the chart	How it is measured in studies
“Sealants”	Pit-and-fissure sealant (PFS), resin-based or glass-ionomer-based	Presence/absence of occlusal caries on sealed vs unsealed surfaces; OR or risk ratio with 95% CI (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120)
“Chewing grooves”	Pits and fissures of occlusal surfaces	Anatomy is a baseline variable trials rarely reported (Ahovuo-Saloranta et al., The Cochrane database of 2017-07-01, PMID 28759120)
“The white spot on my tooth”	Non-cavitated carious lesion (NCCL) / initial lesion	ICDA-II codes; DIAGNOdent; lesion transitions (Kucukyilmaz Izgi et al., Scientific reports 2026-04-01, PMID 42045411) (Shah et al., Caries research 2025-01-01, PMID 39154643)
“Sealant fell off”	Loss of retention; marginal deficiency	Modified USPHS criteria (retention, marginal integrity, discoloration, secondary caries) (Kucukyilmaz Izgi et al., Scientific reports 2026-04-01, PMID 42045411)
“Baby teeth molars”	Primary molars	dmfs/dmft; low-certainty evidence base for sealing (Ramamurthy et al., The Cochrane database of 2022-02-01, PMID 35146744)
“Hypomineralised molars”	Molar incisor hypomineralisation (MIH), post-eruptive breakdown (PEB)	18-month trial: sealant vs laser, RR 1.22 (p = 0.210) for caries; sealant better for PEB (Baraka et al., Journal of dentistry 2026-03-01, PMID 41577163)
“Fluoride painting”	Topical fluoride varnish (D1206)	Prevented fraction; head-to-head with sealants unresolved (Kashbour et al., The Cochrane database of 2020-11-01, PMID 33142363)
“Chemical in the plastic”	Bis-GMA; bisphenol A release	Urinary BPA µg/g creatinine, HPLC, repeated measures (Supornsilchai et al., JDR clinical and transla 2026-01-01, PMID 40077867)

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

We searched Europe PMC for systematic reviews and randomised trials on sealants, then pulled each record programmatically — authors, journal, volume, issue, pages, DOI, open-access status and citation count — and read the abstracts in full, quoting only numbers that appear in them. Absolute-risk examples (160/1,000 → 52/1,000 etc.) are the Cochrane authors’ own arithmetic, reproduced, not ours. Where a confidence interval crosses zero we say so in the same sentence as the number, and where a review grades its evidence (Cochrane uses GRADE) we report the grade rather than our opinion. Payer statements come from the cited public document with its date — check the current version before relying on them.

What this page cannot tell you: whether your child’s specific molars are high- or low-risk (that needs a visual and tactile examination, and possibly radiographs); your local price; whether your plan will pay for a replacement sealant; whether a particular brand is the right one; or what the long-term meaning of a 0.2 µg/g creatinine urinary BPA value is, because the exposure study explicitly says that is not yet deciphered (Supornsilchai et al., JDR clinical and transla 2026-01-01, PMID 40077867).

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)
• Kashbour W, Gupta P, Worthington HV, Boyers D. Pit and fissure sealants versus fluoride varnishes for preventing dental decay in the permanent teeth of children and adolescents. The Cochrane database of 2020-11-01;11():CD003067. doi:10.1002/14651858.cd003067.pub5 · PMID 33142363 · PMCID PMC9308902 · cited by 59 (Europe PMC)
• Schwendicke F, Walsh T, Lamont T, Al-Yaseen W, Bjørndal L, Clarkson JE, Fontana M, Gomez Rossi J, Göstemeyer G, Levey C, Müller A, Ricketts D, Robertson M, Santamaria RM, Innes NP. Interventions for treating cavitated or dentine carious lesions. The Cochrane database of 2021-07-01;7():CD013039. doi:10.1002/14651858.cd013039.pub2 · PMID 34280957 · PMCID PMC8406990 · cited by 63 (Europe PMC)
• Ramamurthy P, Rath A, Sidhu P, Fernandes B, Nettem S, Fee PA, Zaror C, Walsh T. Sealants for preventing dental caries in primary teeth. The Cochrane database of 2022-02-01;2():CD012981. doi:10.1002/14651858.cd012981.pub2 · PMID 35146744 · PMCID PMC8832104 · cited by 29 (Europe PMC)
• Zhang B, Zhao M, Duan S, Tian J, Lei L, Huang R. An economic evaluation of pit and fissure sealants and fluoride varnishes in preventing dental caries: a systematic review. The Journal of clinical 2023-09-01;47(5):4-11. doi:10.22514/jocpd.2023.048 · PMID 37732430 · cited by 4 (Europe PMC)
• Shah NH, Fellows JL, Polk DE. Adoption and Effect of Sealants for Occlusal Noncavitated Caries in a Large Dental Network in the USA. Caries research 2025-01-01;59(1):11-21. doi:10.1159/000540884 · PMID 39154643 · PMCID PMC11790370 · cited by 3 (Europe PMC)
• Supornsilchai V, Sutthirat L, Kaewkamnerdpong I, Jantarat C, Sakorn N, Nosoongnoen W, Chaiboonyarak T, Samaranayake L, Wacharasindhu S, Porntaveetus T. Bisphenol A Biomonitoring after Sealant Placement: A Prospective Cohort Study in Schoolchildren. JDR clinical and transla 2026-01-01;11(1):53-61. doi:10.1177/23800844251320009 · PMID 40077867
• Tichy A, Srolerova T, Schwendicke F. Release of Bisphenol A from Dental Materials: Risks and Future Perspectives. Journal of dental resear 2025-09-01;104(10):1051-1060. doi:10.1177/00220345251337728 · PMID 40524375 · PMCID PMC12301515 · open access · cited by 11 (Europe PMC)
• Kaur N, Srivastava N, Rana V, Kaushik N, Pruthi T, Sirohi A. Retention of resin-based versus glass ionomer pit and fissure sealants in permanent molars: A systematic review of randomized clinical trials. Journal of the Indian So 2025-10-01;43(4):457-467. doi:10.4103/jisppd.jisppd_371_25 · PMID 41235550 · cited by 1 (Europe PMC)
• Baraka M, Elwardani G, Ogwo CE. Sealant vs laser for caries and post-eruptive breakdown prevention in MIH molars: 18-month randomized controlled clinical trial. Journal of dentistry 2026-03-01;166():106519. doi:10.1016/j.jdent.2026.106519 · PMID 41577163
• Gheidari A, Shirazi AS, Parisay I, Mowji M. Hydrophobic or hydrophilic fissure sealants: A systematic review and meta-analysis. Dental research journal 2026-01-01;23():6. doi:10.4103/drj.drj_358_25 · PMID 41777750 · PMCID PMC12952550 · open access · cited by 1 (Europe PMC)
• Kim M, Mun SJ, Han SY, Kim HN, Kang JY, Noh H. Trends in Pit and Fissure Sealant Use and Decayed, Missing, and Filled Teeth Rates Among Korean Adolescents After Including Dental Sealants Under Insurance Coverage. Journal of public health 2026-06-01;86(2):213-219. doi:10.1111/jphd.70052 · PMID 41807286 · PMCID PMC13241914 · open access
• Kucukyilmaz Izgi E, Bolukbasi B, Kavrik F, Timarci I, Savas S. Retention and caries-preventive outcomes of glass ionomer-based fissure sealants in children: a 24-month split-mouth randomized controlled trial. Scientific reports 2026-04-01;16(1):19403. doi:10.1038/s41598-026-49037-6 · PMID 42045411 · PMCID PMC13287736 · open access
• Gok A, Bilge K, Gok T. The effects of material type, salivary contamination and adhesive application on the performance of pit and fissure sealants. PloS one 2026-01-01;21(7):e0352985. doi:10.1371/journal.pone.0352985 · PMID 42384668 · PMCID PMC13322517 · open access
• Vaja D, Bhattacharjee A, Singh RK, Khanna R, Ansari AA. Global knowledge, attitudes, and practices toward pit-and-fissure sealants: A systematic review, meta-analysis, and meta-regression. Journal of the American 2026-07-01;:S0002-8177(26)00188-1. doi:10.1016/j.adaj.2026.03.019 · PMID 42446467

Additional documents

• InsureKidsNow (CMS). Summary of Benefits Report for Utah, Medicaid — dental (data as of 04 Feb 2026) — sealants covered, frequency one every two years per tooth, limited to first and second permanent molars and premolars, only on teeth without decay or restorations; fluoride treatments including varnish up to four times per calendar year; cleanings twice per year; silver diamine fluoride every six months per tooth on primary teeth (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.