Skip to main content

Personal care · product guide

Toothpaste: does it work?

Fluoride is the ingredient that matters: decades of trials show fluoride toothpaste prevents decay. Most other claims are secondary.

Strong evidenceReviewed September 2026

What it is

Toothpaste combines a mild abrasive (hydrated silica, calcium carbonate) that helps the brush remove plaque and surface stain, a detergent, flavor, and in most products fluoride as sodium fluoride, stannous fluoride or sodium monofluorophosphate. Fluoride helps enamel resist acid from mouth bacteria and repair early damage.

Specialty pastes add a second job: potassium nitrate or stannous fluoride for sensitive teeth, stannous fluoride or zinc for gum health and breath, extra abrasives or low-strength peroxide for whitening, and nano-hydroxyapatite as a fluoride alternative. The actives pages cover fluoride, stannous fluoride, potassium nitrate and the abrasives.

How strong is the evidence?

Strong evidence

A Cochrane review of 70 trials in 42,300 children found fluoride toothpaste prevented about 24% of decayed, missing or filled tooth surfaces versus non-fluoride paste, with bigger effects at higher concentration and more frequent brushing. A later review found 1,450 to 1,500 ppm slightly outperforms 1,000 to 1,250 ppm, and 1,000 to 1,100 ppm reduces decay in adults. Evidence for sensitivity pastes is weaker, and fluoride-free alternatives such as hydroxyapatite have far fewer trials.

The claims you'll see, checked

Fluoride toothpaste prevents cavities

Holds up

One of the best-established findings in dentistry, across children, adolescents and adults.

Sensitivity toothpaste relieves sensitive teeth

Mixed

Potassium-nitrate pastes reduced measured sensitivity at 6 to 8 weeks in small trials, but self-rated relief was not clearly better. Stannous fluoride pastes are another option.

Whitening toothpaste whitens teeth

Mixed

Abrasives and polishing agents remove surface stain; they don't change the tooth's inner color like peroxide bleaching.

Charcoal toothpaste whitens and detoxifies

Unproven

A review found no good evidence for its claims, and some charcoal pastes are abrasive and lack fluoride.

Fluoride-free is safer

False

At toothpaste doses used as directed, fluoride's benefit is clear. The real risk is young children swallowing too much, which is why dose guidance differs by age.

Hydroxyapatite works as well as fluoride

Unproven

A few promising trials, several industry-funded; far less evidence than fluoride.

What to look for

  • Fluoride at 1,350 to 1,500 ppm for adults (the UK standard); 1,000 ppm and up still protects.
  • Stannous fluoride if you have bleeding gums, sensitivity or enamel erosion.
  • Potassium nitrate or stannous fluoride for sensitive teeth; give it 2 to 4 weeks of twice-daily use.
  • Low-abrasive pastes if you have exposed roots, receding gums or enamel wear.
  • Age-appropriate amounts for children rather than fluoride-free paste.

What to skip

  • Charcoal pastes and powders, especially fluoride-free ones.
  • Fluoride-free 'natural' pastes as your only paste unless a dentist advises otherwise.
  • Paying extra for 'whitening' if your aim is a real shade change; peroxide products do that.
  • Triclosan pastes; most brands have dropped it and plain fluoride pastes work.

How to use it

  • Brush twice a day for about two minutes, including last thing at night.
  • Spit, don't rinse, so fluoride stays on your teeth.
  • Use a soft brush; a powered brush helps most people remove more plaque.
  • Replace the brush or head every 3 to 4 months, or sooner if bristles splay.

Who should be careful

  • Children: a smear or rice-grain amount under age 3, a pea-sized amount from 3 to 6, and supervise brushing so they spit rather than swallow.
  • Swallowing large amounts of fluoride toothpaste can upset a child's stomach; contact poison control if a young child eats a lot of it.
  • High-fluoride prescription pastes (2,800 to 5,000 ppm) are for adults at high decay risk, under a dentist's advice.
  • See a dentist for sensitivity lasting more than a few weeks, which can signal decay, a cracked tooth or gum recession, and for any mouth ulcer, patch or lump lasting 3 weeks.

The studies

Every citation is checked against its PubMed record.

  1. Fluoride toothpastes for preventing dental caries in children and adolescents
    Marinho VC, Higgins JP, Sheiham A, Logan S · Cochrane Database of Systematic Reviews · 2003 · Systematic review and meta-analysis

    70 trials (42,300 children): fluoride toothpaste prevented 24% of decayed, missing or filled tooth surfaces versus non-fluoride paste. The effect grew with higher fluoride concentration, more frequent brushing and supervised brushing.

  2. Fluoride toothpastes of different concentrations for preventing dental caries
    Walsh T, Worthington HV, Glenny AM, Marinho VC, Jeroncic A · Cochrane Database of Systematic Reviews · 2019 · Systematic review and network meta-analysis

    96 trials: 1,000 to 1,250 ppm and 1,450 to 1,500 ppm fluoride pastes reduced decay versus non-fluoride paste, 1,450 to 1,500 ppm slightly more than 1,000 to 1,250 ppm, and 1,000 to 1,100 ppm reduced decay in adults. Higher strengths in young children should be weighed against fluorosis risk.

  3. Powered versus manual toothbrushing for oral health
    Yaacob M, Worthington HV, Deacon SA, Deery C, Walmsley AD, Robinson PG, Glenny AM · Cochrane Database of Systematic Reviews · 2014 · Systematic review and meta-analysis

    56 trials: powered brushes reduced plaque by 11% at 1 to 3 months and 21% beyond 3 months, and gingivitis by 6% and 11%, compared with manual brushing (moderate-quality evidence). Clinical importance was uncertain.

  4. Potassium containing toothpastes for dentine hypersensitivity
    Poulsen S, Errboe M, Lescay Mevil Y, Glenny AM · Cochrane Database of Systematic Reviews · 2006 · Systematic review and meta-analysis

    Six trials: potassium nitrate toothpaste reduced sensitivity to air blasts and touch at 6 to 8 weeks, but participants' own ratings did not improve significantly. The authors found no clear evidence overall, based on few participants.