How do you actually prevent cavities in kids? What the trials rank highest

Parents are sold a lot of things to keep children's teeth healthy, but when the measures are ranked by the strength and length of their trial record, the list is short and unglamorous: brush with a fluoride toothpaste at home, and let the dentist apply fluoride varnish and, on the back teeth, resin sealants [s1] [s2] [s3]. Each of those has randomised evidence behind it. Most of the extras do not, and the honest caveat throughout is that even the best of this evidence is graded moderate at best [s2] [s3].

The home foundation: fluoride toothpaste

The single most-studied preventive measure is also the cheapest. A Cochrane review that pooled 70 trials involving 42,300 children put the benefit of fluoride toothpaste at a prevented fraction of 24% (95% confidence interval 21 to 28%) — meaning it cut the increase in decayed, missing and filled tooth surfaces by roughly a quarter compared with a non-fluoride paste [s1]. That figure, built from decades of trials, is the benchmark against which everything else is measured, and nothing sold over the counter has evidence to rival it [s1]. It is the reason "brush twice a day with fluoride toothpaste" is the one instruction every guideline shares. It also frames how to read everything that follows: the professional treatments below are not replacements for daily brushing but additions on top of it, tested in children who were, for the most part, already using toothpaste at home [s1].

At the dentist: fluoride varnish

Fluoride varnish is a concentrated coating a dentist or hygienist paints onto the teeth a few times a year, and it has been used this way for over three decades [s2]. A Cochrane review of 22 trials, with 12,455 children randomised, found a substantial effect: in permanent teeth the pooled prevented fraction was 43% (95% CI 30% to 57%) across 13 trials, and in the primary (baby) teeth it was 37% (95% CI 24% to 51%) across 10 trials [s2]. Those are large numbers, but the reviewers were careful to grade the evidence only moderate quality, because it drew mainly on studies at high risk of bias and showed considerable heterogeneity — the trials did not all agree on the size of the benefit [s2]. The direction, though, was consistent: varnish inhibits decay in both sets of teeth [s2].

On the chewing surfaces: sealants

The back teeth carry deep pits and fissures that a toothbrush bristle cannot clean, and this is where sealants earn their place. A Cochrane review of 38 trials involving 7,924 children found that resin-based sealants applied to the biting surfaces of permanent molars prevented caries compared with no sealant — at 24 months the odds ratio was 0.12 (95% CI 0.08 to 0.19) across seven trials [s3]. Translated into whole mouths, the reviewers estimated that resin sealants reduced caries by between 11% and 51% at two years, with the benefit larger where a child's baseline decay risk was higher [s3]. For example, in a group where 40% of unsealed surfaces would decay, sealing brought that down to about 6.25% (95% CI 3.84% to 9.63%) [s3]. The evidence for the older glass-ionomer sealants was inconclusive, and which sealant material is best remains unknown [s3].

What it means

The pattern across all three reviews is that prevention is layered, not singular: a fluoride toothpaste used at home, reinforced by varnish and — on the vulnerable molars — sealants applied professionally, each adds protection through a different route [s1] [s2] [s3]. What is striking is how much of the marketed extra — the special rinses, the supplements, the gadgets — sits outside this evidence base entirely. It is also worth holding the caveats in view. Sealant trials cannot blind the assessor, who can see the sealant, and varnish trials leaned on studies at high risk of bias, so the precise percentages should be read as directional rather than exact [s2] [s3]. None of this is medical advice about any individual child; risk varies, and how often varnish or sealants are worth applying is a judgement a dentist makes case by case [s2] [s3]. But the ranking is clear enough: the measures with real trial evidence are few, familiar and, for the most part, already free or cheap [s1].