Does tongue scraping actually work for bad breath?

It reduces the sulfur compounds responsible for the smell, but not for long. The Cochrane review on the subject found reductions of 42% with a tongue cleaner, 40% with a scraper and 33% with a toothbrush, and reported that reduced levels could not be detected more than 30 minutes after cleaning in any group. That review was withdrawn in 2016 and has not been replaced.

Is a tongue scraper better than a toothbrush?

Probably not by much. The 2006 review found a small edge for dedicated cleaners; a 2021 comparison of a toothbrush, a scraper and both together found all three reduced measured malodour with no difference between the groups. Where you clean matters more than what you clean with.

How far back should you scrape your tongue?

As far back as you can tolerate. The coating that produces odour sits on the posterior third of the tongue, near where it curves towards the throat. Cleaning only the front two-thirds is comfortable and largely pointless.

Will tongue scraping cure my bad breath?

Not on its own, and not if the cause is elsewhere. It does nothing for tonsil stones, dry mouth or gum disease, all of which produce breath odour by other routes. If odour persists after two weeks of cleaning teeth, between teeth and tongue, that is a reason to be examined.

Tongue Scraping: What the Evidence Shows

Fig. 1 — The percentages are the reductions in volatile sulfur compounds one trial measured with a portable monitor. The bars are drawn to the review’s own limit: reduced levels “could not be detected for more than 30 minutes after the intervention in any of the groups”. The mechanism is real; the duration is the part nobody quotes.

The systematic review that this practice’s reputation rests on contained two trials and forty people, reported that neither of them measured what the review had set out to measure, and was withdrawn by Cochrane in 2016. The strongest finding it did contain was a reduction in odour compounds that could not be detected half an hour later.

Tongue cleaning is still worth doing. This page is about the difference between that sentence and the ones you will read elsewhere.

What the review actually found

Outhouse and colleagues published Tongue scraping for treating halitosis in the Cochrane Database of Systematic Reviews in 2006. It searched for randomised trials comparing methods of tongue cleaning in adults with halitosis, and found two, with forty participants between them.

Its Main results section contains a sentence in the same family as the one that governs the flossing literature: the trials “were methodologically sound but included no data for the primary outcomes specified in this review”. In other words, the outcomes the reviewers cared about were not collected, so everything reported is a secondary measure.

That secondary measure was volatile sulfur compounds — the gases responsible for the smell — read off a portable sulfide monitor. One trial reported reductions of 42% with a tongue cleaner, 40% with a tongue scraper and 33% with a toothbrush. And then, verbatim: reduced levels “persisted longer with the tongue cleaner than the toothbrush and could not be detected for more than 30 minutes after the intervention in any of the groups”.

The authors’ own conclusion was that there is “weak and unreliable evidence to show that there is a small but statistically significant difference in reduction of VSC levels when tongue scrapers or cleaners rather than toothbrushes are used”, and that they found no high-level evidence comparing mechanical with other forms of tongue cleaning.

And then it was withdrawn

In May 2016 the review was republished under the title WITHDRAWN: Tongue scraping for treating halitosis. Cochrane withdraws reviews for several reasons — most commonly because they are out of date and nobody has updated them — and we could not open a record stating which reason applied here, so we are not going to guess at one.

What matters practically is that the best-known evidence summary on tongue scraping is no longer a live Cochrane review, and has not been replaced by one. Articles still cite it as current. It is not.

What has been published since

Two small studies from 2021 are worth knowing about, because they are the closest thing to current evidence and neither says what the internet says.

A comparison of mechanical methods. Toothbrush alone, tongue scraper alone, and both together were each measured for their effect on breath odour and tongue coating. All three significantly reduced measured malodour and tongue-coating scores, and there was no difference between the groups. The only between-group difference was that the scraper group’s hydrogen sulfide reading fell immediately after cleaning where the others did not.
A crossover pilot with a zinc lactate mouthwash. Sixty volunteers followed two 14-day protocols — brushing alone, and brushing plus a zinc lactate rinse and tongue scraping. The highest odour readings, both organoleptic and by volatile sulfur compounds, were in the brushing-only protocol. Note the design: the rinse and the scraping were bundled into one arm, so the study cannot tell you which of the two did the work.

That is the whole modern evidence base: small, short, and mostly unable to separate the tongue cleaner from everything else in the routine.

Correction, 9 September 2026

This page used to say that tongue scraping helps breath “more than almost anything else you could add”. That was too strong. The measured effect is real but small, the trials behind it are two studies in forty people, the review summarising them has been withdrawn, and the most recent comparison found no advantage over a toothbrush. The recommendation below has not changed — the reasoning behind it has, and our corrections policy is to say so on the page.

Why the tongue is still the right target

Because the mechanism is not in dispute, and mechanism is what is carrying this recommendation.

The back of the tongue is a rough, low-oxygen, food-trapping surface, and the bacteria that live in it break down proteins into volatile sulfur compounds — hydrogen sulfide, methyl mercaptan, dimethyl sulfide. Those gases are what bad breath mostly is, in the large majority of cases where the cause is in the mouth rather than the stomach or the sinuses. Disturbing that coating removes some of the substrate and some of the organisms.

So the argument for tongue cleaning has the same shape as the argument for cleaning between your teeth: a direct mechanism, negligible cost, negligible risk, and a body of trials nobody is proud of. That is a good enough reason to do it. It is not a good enough reason for the confident claims made about it.

Scraper, brush, or the ribbed back of a toothbrush?

On the evidence: probably it does not matter much. The 2006 review found a small edge for dedicated cleaners over a toothbrush; the 2021 comparison found no difference between a toothbrush, a scraper, and both. Given the sizes of these studies, “no clear difference” is the honest summary.

A dedicated scraper is a thin strip of metal or plastic. It clears a wider area per pass, tends to provoke less gagging than a brush head, and is easy to rinse.
A toothbrush works, and the trials say so. If it is what you have, use it.
The ribbed pad on the back of some toothbrushes has no separate evidence at all. It is not a reason to buy a particular brush.
Where you clean matters more than what you clean with. Almost everyone stops too far forward.

How to do it without gagging

Start further back than feels natural The coating that produces the smell sits on the posterior third of the tongue, near where it curves down towards the throat. Cleaning the front two-thirds is pleasant and largely pointless.
Reach back once, then pull forward Place the scraper as far back as you can tolerate, then draw it forward in one steady stroke. Repeated pushing backwards is what triggers the reflex.
Rinse between strokes Four or five passes, rinsing the scraper each time. Then rinse your mouth.
Breathe out through your mouth as you go It genuinely suppresses the gag reflex, and it is easier than any of the other tricks people suggest.
Do not scrub You are removing a soft coating, not sanding a surface. Hard scraping makes the tongue sore and adds nothing.

Thirty seconds, once a day. Morning suits most people, because the coating builds overnight when saliva flow drops.

What it will not do

It will not last all day. That is the finding at the top of this page. Do it for the same reason you brush in the morning — because the first few hours matter to you.
It will not fix tonsil stones. Those sit in the crypts of the tonsils, behind the tongue, and no amount of scraping reaches them.
It will not fix dry mouth. If saliva flow is the problem, removing coating addresses a symptom of it. See dry mouth.
It will not treat gum disease. Breath that comes from periodontal pockets needs the pockets treated; see the early signs of gum disease.
It will not permanently change the colour of your tongue. A white tongue that returns within a day is a coating; one that does not shift is worth showing to someone.

See a dentist or doctor if

Bad breath persists despite cleaning your teeth, between them, and your tongue for a fortnight.
A white or red patch on the tongue does not wipe or scrape off, or has been there more than two weeks.
The tongue is sore, burning, or has changed texture.
Your mouth is persistently dry, or you are on medication that dries it.

Fig. 1 is free to reuse under CC BY 4.0 — see the diagrams page.

References

Outhouse TL, Al-Alawi R, Fedorowicz Z, Keenan JV. Tongue scraping for treating halitosis. Cochrane Database Syst Rev 2006;2:CD005519. DOI · PubMed Withdrawn by Cochrane in May 2016 (withdrawal record, DOI); not replaced. · Funding: see ledger
Choi HN, Cho YS, Koo JW. The effect of mechanical tongue cleaning on oral malodor and tongue coating. Int J Environ Res Public Health 2021;19(1):108. DOI · PubMed · Funding: see ledger
Dudzik A, et al. Efficacy of a zinc lactate mouthwash and tongue scraping in the reduction of intra-oral halitosis: a single-blind, controlled, crossover clinical trial — a pilot study. J Clin Med 2021;10(23):5532. DOI · PubMed · Funding: see ledger

Tongue scraping: what the evidence actually shows