Do water flossers actually work?

A water flosser is a pump and a nozzle: a pulsed jet of water aimed along the gum line and between the teeth. The adverts say it is clinically proven to be more effective than string floss, and there are trials behind that sentence. The trials are real. What they measured, how long they ran, and who paid for them are the whole story, and it fits on one page.

The short answer

In the short trials that exist — almost all four weeks, one of them a single use — adding a water flosser to brushing reduced gum bleeding and gingival inflammation scores, and in three comparisons that share a co-author listed at Water Pik, Inc. it reduced them more than string floss did (Barnes, 2005; Rosema, 2011; Goyal, 2018). Cochrane pooled the independent-of-any-one-trial view and found less: against brushing alone, irrigators may reduce the gingival index at one month, at very low certainty, and not at three or six months; they did not reduce the proportion of bleeding sites, or plaque, at any time point (Worthington, 2019). No trial measured decay between the teeth, and most did not measure periodontitis — the two things you would actually buy the machine to prevent. So: it does something measurable to inflamed gums over a month. Whether that turns into fewer fillings or fewer lost teeth over years has not been tested by anyone.

What the trials measured — and what they never measured

Every trial below scored plaque, bleeding or a gingival index, or all three, over two to eight weeks. None measured interproximal caries. One followed people for six months. The column that matters most is the last one.

Two things stand out. The follow-up column is measured in weeks; the one six-month trial is on people with a bonded retainer, not the general public. And “plaque only, single use” is the design behind the most-quoted percentage in the category: the water flosser removed 74.4% of whole-mouth plaque against 57.7% for string floss, in 70 adults, after one supervised session (Goyal, 2013). That is a measurement of one clean, not of what happens to a mouth over a year.

Who paid for the studies

Stated plainly, because the pattern is plain, and stated exactly, because it is easy to overstate. Of the eight trials in the table, two record a Water Pik, Inc. employee among their authors, and the ledger classes those two, and only those two, as manufacturer-employed: the 2018 paper by its affiliation line on PubMed (Goyal, 2018), and the 2025 paper by its own funding statement, which reads: “The first author, Mancinelli-Lyle, was an employee of Water Pik, Inc. when the study was conducted … ACTA Research B.V. received financial support from Water Pik, Inc.” (Mancinelli-Lyle, 2025). That is also the only one of the eight whose own words say manufacturer money was involved.

Four earlier papers — 2005, 2008, 2011 and 2013 — share a co-author with the 2018 paper. Their journals are not open access and we could not read a funding statement at source, so we do not quote one and the ledger records each funder as not stated. What PubMed does show is the affiliation: a contract research firm, BioSci Research Canada Ltd., on the 2008 and 2013 papers; a university dental college on the 2005 one; ACTA in Amsterdam on the 2011 one. None of those four is recorded as manufacturer-funded or manufacturer-employed, here or in the ledger. We report the shared author and leave it there.

That is not an accusation. Manufacturers fund and staff trials of their own products in every category, the trials are registered and examiner-blinded, and nothing here suggests the numbers are wrong. It is a reason to notice what those trials were designed to find — short-term bleeding and plaque, against comparators chosen by the people running them — and to give more weight to the three with a named public funder and no shared author: a university-funded trial in Karachi with no competing interests that ran six months (Masood, 2026), a periodontal-society-funded trial in Riyadh (AlMoharib, 2024), and a university-grant trial in Samsun (Şahin, 2026). Two of those three found the irrigator ahead of an interdental brush; the one that compared it with floss found a difference too small to be distinguished from chance.

Corrected 21 September 2026. Earlier versions of this page called the trials that share this co-author “paid for by the manufacturer”, “manufacturer-funded” and “sponsor-funded”. That was more than the papers say. Only the 2025 paper states manufacturer money, and only the 2018 and 2025 papers record a manufacturer’s employee as an author; for the other four the funder is not stated. Our own funding ledger had it right and this page did not, so the page now says what the ledger says.

Water flosser vs string floss: what “as good as” actually meant

The claim on the box comes from three comparisons. Barnes (2005) found the water-jet groups reduced a bleeding index more than brush-and-floss at 14 and 28 days, with “no statistical difference” on plaque by day 28. Rosema (2011) found lower bleeding scores in both irrigator groups than in the floss group at four weeks and, for plaque, “no significant difference among groups”. Goyal (2013) is the single-use plaque measurement above. So “as good as floss” — usually written “up to 2× as effective” or, from the 2018 trial, “3×” — means: on bleeding scores at four weeks, in trials that share the one co-author, against people flossing once a day. It does not mean it cleans the contact point, because it does not reach it (Fig. 1), and it does not mean fewer cavities, because that was never measured.

Cochrane’s reading of the same two-trial comparison is the sober version: irrigation “may reduce gingivitis at one month compared to flossing” at low to very low certainty, with no difference for plaque (Worthington, 2019). And in orthodontic patients, where the comparison has been made most often since, a 2026 meta-analysis of four trials found no significant difference in plaque between irrigator and floss (SMD −0.88, 95% CI −2.25 to 0.50, p = 0.10), with the trials disagreeing almost completely (I² = 98%) (Yiamwattana, 2026). There is a separate point the flossing evidence makes about floss itself: the Cochrane review found no trial measured decay between the teeth for any device. The bar the water flosser is being compared against is lower than the adverts on either side imply.

Where it genuinely helps

• Braces. The most-studied use, and the least settled. The earliest trial, run by a contract research firm with its funder not stated, found the jet reduced plaque and bleeding in 106 adolescents over four weeks (Sharma, 2008). Two independent reviews pooled what came after: seven trials showed no significant difference on gingival or plaque index for a jet added to brushing, at low to very low quality (Zarei, 2024), and four trials against floss showed none for plaque (Yiamwattana, 2026). One independent 2026 trial did find lower gingival index and bleeding than an interdental brush at eight weeks, in 30 patients (Şahin, 2026). Honest grade: plausible, small trials, results in both directions.
• Bonded retainers. The best single piece of evidence on this page: 70 patients, six months, no manufacturer money, plaque, gingival and calculus indices all lower with the water flosser than with an interdental brush, and no difference in attachment level or pocket depth (Masood, 2026). Grade: one good trial, one specific situation.
• Bridges and implants. The geometry argument is the same as for retainers — floss cannot get under a fixed span, a jet can — and there are systematic reviews of implant maintenance from 2025. We have not read them at source and do not summarise them here.
• Large gaps. If the space takes an interdental brush, the brush has the better evidence: Cochrane found interdental brushes may reduce gingivitis and plaque more than floss, at low certainty (Worthington, 2019). Nothing shows a jet beats a brush that fits.
• People who will not floss. The strongest practical case, and the one to be careful with. Something done daily beats something done never, and the four-week trials say the jet reduces bleeding. But Cochrane’s pooled finding against brushing alone is that irrigators did not reduce bleeding sites or plaque at any time point (Worthington, 2019). For what bleeding means in the first place, see the early signs of gum disease.

What it cannot do

It does not remove tartar. Once plaque has mineralised — roughly 24 to 72 hours undisturbed — no jet, brush, rinse or floss shifts it; only an instrument does, which is what a scaling appointment is for and why the two-day window matters more than the tool. It does not open the contact point between two teeth, so it does not clean the surface where interproximal decay starts; whether floss prevents that decay is itself unmeasured, but at least floss reaches it. And it cannot put mineral back into an early lesion — that is saliva’s job and fluoride’s, covered in can teeth remineralise? A water flosser is a gum-margin tool. Judged as that, the evidence is modest and real. Judged as a decay-prevention device, there is no evidence, because nobody has looked.

If you buy one, what the evidence says about using it

Only what was actually studied, which is less than the instruction leaflet implies.

• Once a day. Every trial used it once daily alongside twice-daily brushing (Barnes, 2005; Rosema, 2011; Goyal, 2018; Mancinelli-Lyle, 2025). Nobody has tested twice.
• Medium pressure, lukewarm water, a full reservoir. The 2005 protocol: “a medium setting, irrigating once per day with 500 ml of luke warm water” (Barnes, 2005). Higher settings have not been shown to do more; the 2025 trial found no difference in gingival abrasion between a jet and an interdental brush, so the medium setting does not appear to injure the gum (Mancinelli-Lyle, 2025).
• Tip. The braces trial used an orthodontic tip (Sharma, 2008). The 2011 trial found a prototype jet tip and a standard one both beat floss on bleeding, with no advantage of one tip over the other for plaque (Rosema, 2011).
• Before or after brushing. Not tested. The trials say “once per day” and leave it there; one instructed use at night (AlMoharib, 2024), the rest do not say. Do not rinse fluoride toothpaste off straight after brushing, which is a separate and better-founded rule.

A sensible way to decide

1. If you clean between your teeth already — floss or brushes, daily — the trials give you no strong reason to switch. Keep what you do.
2. If you have braces, a bonded retainer, a bridge or implants, the jet reaches places the alternatives do not, and the one six-month independent trial favours it. Reasonable buy.
3. If you will not floss and will use a jet, use the jet: once a day, medium, lukewarm. Know that the pooled evidence for bleeding against brushing alone is nil, and that your gums bleeding is the thing to watch either way.
4. Whatever you buy, it does not touch tartar and it has never been shown to prevent decay. The cleaning appointment and the fluoride are not optional because of it.

References

1. Worthington HV, MacDonald L, Poklepovic Pericic T, et al. Home use of interdental cleaning devices, in addition to toothbrushing, for preventing and controlling periodontal diseases and dental caries. Cochrane Database Syst Rev 2019;4:CD012018. DOI · PubMed · Funding: see ledger
2. Barnes CM, Russell CM, Reinhardt RA, et al. Comparison of irrigation to floss as an adjunct to tooth brushing: effect on bleeding, gingivitis, and supragingival plaque. J Clin Dent 2005;16(3):71–77. PubMed · Funding: see ledger
3. Sharma NC, Lyle DM, Qaqish JG, et al. Effect of a dental water jet with orthodontic tip on plaque and bleeding in adolescent patients with fixed orthodontic appliances. Am J Orthod Dentofacial Orthop 2008;133(4):565–571. DOI · PubMed · Funding: see ledger
4. Rosema NA, Hennequin-Hoenderdos NL, Berchier CE, et al. The effect of different interdental cleaning devices on gingival bleeding. J Int Acad Periodontol 2011;13(1):2–10. PubMed · Funding: see ledger
5. Goyal CR, Lyle DM, Qaqish JG, Schuller R. Evaluation of the plaque removal efficacy of a water flosser compared to string floss in adults after a single use. J Clin Dent 2013;24(2):37–42. PubMed · Funding: see ledger
6. Goyal CR, Qaqish JG, Schuller R, Lyle DM. Evaluation of the addition of a water flosser to manual brushing on gingival health. J Clin Dent 2018;29(4):81–86. PubMed · Funding: see ledger
7. Mancinelli-Lyle D, Van der Weijden FGA, Slot DE. Efficacy of a water flosser compared to an interdental brush on gingival bleeding: a 4-week RCT. Int J Dent Hyg 2025;23(1):86–96. DOI · PubMed · Funding: see ledger
8. AlMoharib HS, Alqasem A, Almusfer G, Aldosari MA, Almadhoon HW. The effectiveness of water jet flossing and interdental flossing for oral hygiene in orthodontic patients with fixed appliances: a randomized clinical trial. BMC Oral Health 2024;24:496. DOI · PMC11055227Funding: Saudi Society of Periodontology; no competing interests · Funding: see ledger
9. Şahin E, Cetinkaya BO, Avci B, Kurt-Bayrakdar S, Elekdag-Turk S. Clinical and biochemical evaluation of oral irrigator effectiveness in fixed orthodontic treatment: a randomised clinical trial. Int J Dent Hyg 2026. DOI · PubMed 42120952Grant: Ondokuz Mayıs University · Funding: see ledger
10. Masood Z, Fida M, Sukhia RH. The effects of using water flosser versus interdental brush on gingival health in patients with mandibular bonded retainer — a randomized controlled trial. BMC Oral Health 2026. DOI · PMC13348601Funding: Aga Khan University Hospital; no competing interests · Funding: see ledger
11. Zarei Z, Yazdi M, Sadeghalbanaei L, Tahamtan S. The effectiveness of oral irrigators on periodontal health status and oral hygiene of orthodontic patients: a systematic review and meta-analysis. BMC Oral Health 2024;24:1479. DOI · PMC11619677“No financial or non-financial support” · Funding: see ledger
12. Yiamwattana I, Kaomongkolgit R, Jaikumpun P, Sang-Ngoen T, Teerawong C, Tosasukul J, Sadaeng W. Oral irrigator versus dental floss in orthodontic patients: a systematic review and meta-analysis. Int J Dent Hyg 2026. DOI · PubMed 41331619Naresuan University · Funding: see ledger

Our illustrations and tools are free to reuse under CC BY 4.0 — see the diagrams page.