Does potassium nitrate work within a week? Every published time point, from day three to week twelve.

Potassium nitrate is measured in weeks, and its published record is thinnest exactly where this question sits. The 2019 systematic review that sorts desensitising treatments by how soon each one reaches significance puts potassium nitrate in the long-term group and keeps only in-chair treatments in its up-to-seven-days window¹. Its earliest reading in a randomised trial is day three, where an industry-funded study found it ahead of the fluoride control on the cold-air score and not on the tactile threshold, and the one reading earlier than that belongs to a paste which also carried strontium chloride²⁷. Three studies have examined a potassium toothpaste at one week and none publishes a figure for what it did on its own, because in every one of them the week-one comparison belongs to the active it was tested against, and two were written by employees of the company whose paste won³⁴⁵. If the deadline behind your search is real, the NHS answer to pain that has lasted more than two days is a dentist rather than a different tube²³. S3 Sensitivity Science™ is judged at two to four weeks, and says so.

What was checked22 peer-reviewed studies, NHS and Oral Health Foundation guidance

Is there a published reading for potassium nitrate at one week?

There is, in three studies, and none of them tells you what you want to know. Four PubMed searches run for this page in September 2026 returned seventy-nine records between them, and screening every abstract left three trials in which a potassium toothpaste was examined at seven days, two of them industry-funded and written by the makers of the pastes it was measured against³⁴⁵.

The largest is an industry-funded phase III trial of 120 completers in which a stabilised stannous fluoride toothpaste was tested against a marketed potassium nitrate dentifrice as its positive control, with readings at baseline, day one, day three and day seven³. At the seven-day reading the stannous paste had reduced tactile hypersensitivity 79.8% more than the potassium comparator and air-blast hypersensitivity 47.1% more; seven of the ten authors are employees of the company that makes the stannous paste, and the paper appeared in a journal supplement³. What the abstract never gives is the number this reader wants: how much the potassium arm changed from its own baseline.

The second is an eight-week trial of 80 adults from Rome, whose authors include the manufacturer's scientists who developed the arginine technology it was compared against, and it names its potassium comparator precisely: a marketed paste carrying 2% potassium ion, dosed as 5.0% potassium nitrate⁴. Its examination points are baseline, one week, two weeks, four weeks and eight weeks. Its published results begin at two weeks — the one-week examination was carried out and its number is not in the paper's abstract. The third is a small three-group clinical study of 45 adults, fifteen per group, in which an arginine paste beat a potassium-based one at every interval including one week; it published no figures at all, ran no placebo arm, and never states how much potassium its test tube carried⁵.

So the honest answer to the search that brought you here is this: a potassium toothpaste has been measured at seven days, three times, and on all three occasions it was the thing something else was measured against. No published trial reports what a 5% potassium nitrate toothpaste does on its own at the end of week one, and two of the three that took the reading were industry-funded³⁴⁵.

What has been measured, and when?

Everything that has been published, in one table. Two instruments do most of the work: the Schiff scale, which scores how much a blast of cold air hurts, and the Yeaple probe, which raises a measured force against the tooth until you feel it. They can disagree, and on potassium nitrate they have. A cell marked no published reading is not an oversight; those cells are the answer to the question at the top of this page.

The two instruments matter more here than they look. In an eight-week trial of 120 adults, an arginine regimen improved the tactile threshold significantly at two, four and eight weeks while the potassium regimen never did at any point, even though both improved on the air blast¹⁶. In a four-week trial of 30 adults, a bioactive glass paste was significantly ahead of 5% potassium nitrate at two weeks and at four, and the potassium paste separated from its control only at four weeks¹⁵. "It worked at day three" can therefore mean two different things, and where both instruments are read the answer is often a split.

The pattern holds whichever row you start from. Nothing placebo-controlled arrives before four weeks^11121314; everything earlier is a comparison potassium loses, or a reading nobody published. That is not the same as saying nothing happens in week one. An untested claim and a disproved one are different things, and this page will not blur them.

Why do the actives that work on the tubule show up sooner?

Because they change something you can photograph, and potassium changes something you cannot. A tubule-occluding active — a bioactive glass, a stannous salt, arginine with calcium carbonate — deposits material in the openings on the dentine surface, and the deposit is there the moment brushing stops. Potassium is thought to work on the nerve endings inside the tooth instead, by raising the potassium concentration around them until they are less ready to fire, and that is a slow accumulation rather than a plug. It is also a proposed mechanism rather than a confirmed one, which the page on how potassium nitrate is thought to calm a sensitive tooth sets out at length: nobody has yet confirmed it inside an intact human tooth. The Sensitivity Journal's explainer on how sensitive toothpastes work walks through both routes end to end.

The published clocks follow the mechanisms. In the systematic review of 74 randomised trials, stannous fluoride and hydroxyapatite reach significance in the medium-term window of up to a month, potassium nitrate and arginine only in the long-term one, and the up-to-seven-days window contains no toothpaste at all¹. In the twelve-week head-to-head, the bioactive glass paste was significantly ahead of potassium nitrate at two weeks and four weeks, and by twelve weeks the two were level¹⁰. In a three-day trial whose funding is not stated in the record, an arginine paste beat a potassium paste on a single fingertip application, before any brushing had happened⁸. And in the newest and largest synthesis — a network meta-analysis of 93 randomised trials and 9,548 participants — the network is built at two weeks, not one, because that is where the trials put their earliest common measurement; potassium with or without fluoride comes out at a small effect on low confidence, and stannous fluoride is the only node rated high²¹.

So, plainly. If your question is really within a week, potassium nitrate is the wrong active to be asking about — and potassium nitrate is the nerve-calming active in S3 Sensitivity Science™. The useful next question is not which tube is quickest, but what a week can realistically buy you, and the answer to that involves a dentist.

Why is your own impression at day seven unreliable?

Because the trials measured what happens when a person brushes with a paste that does nothing, and the answer was: quite a lot¹⁸¹⁹. In a six-week trial of 120 adults comparing strontium acetate, potassium nitrate and a plain fluoride paste, no product differed significantly from any other, and the fluoride-only group improved significantly on cold air from its own baseline¹⁸. A network meta-analysis of 30 randomised trials reported a significant placebo effect across the literature and found no significant difference between potassium and placebo¹⁹. The Cochrane review of potassium toothpastes, still the most-quoted source on the subject, found a significant effect on air-blast and tactile measures at six to eight weeks and none at all on what patients said about their own teeth²⁰.

Add to that what happens outside a trial. In a cohort of 1,862 patients treated by 171 community dentists and asked to rate their own pain at one, four and eight weeks, roughly 60% reported improvement over the whole period — with treatment chosen by the dentist, no control group, and no figures published by time point²². Neither a good week nor a flat one tells you what the paste did. A trial controls for the weather, for the tooth you stopped chewing on, for the fact that being watched changes how you brush. Your bathroom controls for none of it.

What should the first two weeks with S3 feel like?

Mostly like brushing your teeth. The judgement point on the pack is two to four weeks, which is where the published trials of this active put their first clean readings, and the honest version of that promise is a date to check in rather than a deadline to hit. Two brushings a day, the full two minutes, spit and do not rinse, and no swapping on day eight because nothing has happened yet: swapping at day seven restarts a clock that nobody in the published literature has run for less than a fortnight.

One thing worth knowing is that the trials disagree with the pack about stopping. The Oral Health Foundation's advice is to keep using a sensitivity toothpaste to maintain the effect²⁴. The one trial that stopped the paste and measured what happened — industry-funded, and the only washout arm in this literature — found most of the benefit still there three weeks later². Both statements are on this page because both are true, and the practical reading is the boring one: keep brushing.

Ranked by when each active first has a published reading against a control, the picture does not flatter the one S3 uses, and the combination S3 carries has no published time-course of its own.

What if you genuinely need relief in a week?

Then the answer is an appointment, not a tube, and the evidence says so plainly. The systematic review's up-to-seven-days window contains three things and a dentist applies all of them: glutaraldehyde with HEMA, glass ionomer cements and lasers¹. Those are procedures with their own indications, costs and failure rates, and this page is not the place to choose between them — a dentist is.

There is a second reason to go, and it is the better one. The NHS threshold for toothache is short: pain that has lasted more than two days is a question for a dentist²³. The Oral Health Foundation adds four flags of its own — severe pain, sensitivity lasting more than a few weeks, only one tooth affected, or pain that arrived suddenly²⁴. A wedding on Saturday and a tooth that has been screaming at cold water since Tuesday is not something you can shop your way out of. Start the paste today, by all means, because the clock has to start somewhere. But book the appointment first.

Frequently asked questions

Can any toothpaste work in a week?

One class has published week-one numbers: stannous fluoride. In a four-week trial of 96 participants, a stabilised stannous fluoride paste improved electrical-stimulation readings over a plain fluoride control by 15.1% at three days and 34.2% at one week, rising to 111.6% at four weeks; the trial was funded by the manufacturer and five of its authors are that company's employees¹⁷. The systematic review of 74 randomised trials is stricter still, and places no toothpaste of any kind in its up-to-seven-days window¹. Read that as a genuine difference between actives, not as a reason to expect a tooth to be fixed by Friday.

How long should I give S3 before deciding it is not working?

Two to four weeks of twice-daily brushing, which is the judgement point on the pack and the window in which the trials of its nerve active took their first clean readings. If nothing at all has shifted by four weeks, the useful move is not a different tube but a dentist: a single tooth that stays sharp, pain that lingers after the cold has gone, or pain at night are all signs of something a toothpaste was never designed to treat²⁴.

Why did I feel better on day two?

Possibly the paste; possibly not. In the trials, people given a plain fluoride toothpaste with no desensitising active in it improved significantly from their own baseline, and a network meta-analysis of 30 randomised trials found the placebo response large enough to swamp the differences between products¹⁸¹⁹. There is also the ordinary business of a sore tooth settling down, and the fact that you started paying attention to it. None of this means your day-two improvement is imaginary. It means it is not evidence.

Does using more paste make it work sooner?

Nothing published says so. The trials that produced every number on this page used a pea-sized amount twice a day, and the one variable they changed was time, not quantity. Two of them changed the base around the potassium instead — a copolymer in one, aluminium lactate in the other — and both combinations beat the same 5% potassium nitrate on its own, which suggests the interesting variable is the rest of the formula rather than how much you squeeze out; both were manufacturer-run single trials⁶⁹.