29 The toothpaste aisle
A large supermarket stocks around sixty toothpastes. They differ in price by a factor of ten and in fluoride content, which is the only thing with strong evidence behind it, hardly at all.
This chapter is a walk down that aisle with the method from Chapter 4 in hand. It is shorter and less formal than the audit chapters, because most of what is on sale has no guideline recommendation to audit and often no trial to appraise. Where I am relying on mechanism rather than evidence, I say so.
The rule that survives the whole aisle is the one from Chapter 8: buy a toothpaste with 1,350 to 1,500ppm fluoride and stop thinking about it. Everything below is a variation on why the other claims do not change that.
Charcoal toothpaste
The claim is whitening and detoxification. The second word is meaningless in this context and should be treated as a marketing signal.
Two real concerns. First, charcoal pastes are often abrasive, and abrasion removes enamel permanently. Chapter 23 explains why that matters: enamel does not grow back, and teeth get yellower as it thins, because the dentine underneath is yellow and starts to show through. A whitening product that thins enamel makes teeth darker in the long run.
Second, and more important: many charcoal toothpastes contain no fluoride. That is not a small trade-off. You would be giving up the single best-evidenced intervention in this book in exchange for a cosmetic claim with essentially no supporting trials.
Check the tube. If it says no fluoride, put it back.
Whitening toothpaste
Whitening toothpastes work by removing surface stains, generally through abrasives and sometimes through low concentrations of peroxide. They do not change the underlying color of your teeth, because the color of your teeth is mostly dentine seen through translucent enamel.
So they can return a stained tooth to its natural shade, which for a coffee drinker is a real if modest effect. They cannot make a naturally ivory tooth white, and any product promising several shades is describing something a toothpaste cannot do.
The relevant caution is again abrasivity, and the relevant question is whether the paste also contains 1,350 to 1,500ppm fluoride. Most mainstream ones do.
Hydroxyapatite toothpaste
This is the interesting one, and the only entry in the aisle where I think the science is genuinely open rather than merely absent.
Hydroxyapatite is the mineral teeth are made of. The proposal is that a paste containing nano-particles of it can deposit onto and remineralise the enamel surface, achieving what fluoride achieves without using fluoride. It is established in Japan and increasingly marketed in Europe, usually to people looking for a fluoride-free option.
The mechanism is coherent, which is more than can be said for charcoal. There is a growing trial literature, some of it reporting non-inferiority to fluoride toothpaste. I have not systematically reviewed it, and I want to flag two things that would need checking before anyone treats the question as settled.
Most of the trials are small, short, and industry-funded, which is the same configuration Chapter 16 found in the toothbrush literature and which (1) gives empirical reason to discount.
Non-inferiority is a specific and demanding claim. A non-inferiority trial requires a pre-specified margin, adequate power, and analysis conventions that differ from a superiority trial. A small trial that fails to find a difference has not shown non-inferiority; it has failed to find a difference, which Chapter 19 explains is a different thing.
Set against that: fluoride toothpaste has 55 randomized trials at high to moderate certainty behind it. Hydroxyapatite would need a great deal of evidence to displace that, and it does not have it yet. If you have a specific reason to avoid fluoride, this is the most plausible alternative on the shelf. If you do not, the case for switching is not there.
“SLS-free”, “natural”, “herbal”
Sodium lauryl sulphate is the detergent that makes toothpaste foam. Some people with recurrent mouth ulcers report fewer of them on SLS-free paste. The evidence is thin, the mechanism is plausible, the cost of trying is low, and it makes no difference to the fluoride. If you get ulcers, try it.
“Natural” and “herbal” are not regulated claims. The only question worth asking of such a tube is the same one: does it contain 1,350 to 1,500ppm fluoride? Many do not.
Sensitivity toothpaste
Sensitivity pastes contain potassium nitrate, which damps down nerve response, or stannous fluoride or arginine compounds, which physically occlude the open dentine tubules that cause the pain. These have a reasonable evidence base and they work for many people, though the effect builds over weeks rather than days.
One caveat worth stating: sensitivity is a symptom. It can be exposed dentine from gum recession or tooth wear, and it can be a cracked tooth or decay. A paste that removes the warning without addressing the cause is not always a good outcome. If it is new, localised, or severe, get it looked at rather than managing it.
The high-fluoride prescriptions
2,800ppm and 5,000ppm pastes are available on prescription, and Chapter 8 found the evidence for going above 1,500ppm to be thinner than most people assume: SMD 0.04 (95% CI −0.07 to 0.15) for 1,700 to 2,200ppm against 1,450 to 1,500ppm, from indirect evidence only, and −0.05 (−0.14 to 0.05) for 2,400 to 2,800ppm.
DBOH recommends them Conditionally, for people with active decay. Its evidence statement is worth quoting in full, because an earlier draft of this chapter cut it in the middle and the missing half is the important one:
Recommendation based on there being no reliable evidence of superior effectiveness but some evidence of dose-response relationship, although it may not extend to concentrations this high. Moderate-certainty evidence for effectiveness of 5,000ppm fluoride for root caries.
— Delivering Better Oral Health, chapter 13, table 7 (2)
So the picture is not “the trials are not there”. It is two different indications with two different evidence bases. For preventing decay on the crowns of teeth, the network comparisons above 1,500ppm are imprecise and do not establish an added benefit. For root caries, decay on the exposed root surface and largely a problem of older adults, the guideline names moderate-certainty evidence for 5,000ppm specifically.
Conditional is the right label across both, and my complaint here is only against my own earlier summary. If a dentist has prescribed one because you have active decay, use it, and if the decay is on the root surface the evidence behind that prescription is better than anything else in this chapter.
This chapter searched less than the audit chapters, and that limits what it can conclude rather than lowering the standard its conclusions have to meet.
Almost nothing in the toothpaste aisle corresponds to a DBOH recommendation, so there is nothing to trace and no strength label to check. I have not systematically reviewed the literature on charcoal, hydroxyapatite, whitening, SLS or sensitivity agents.
That has a consequence I did not draw in an earlier draft. Announcing a thinner search does not license a firmer answer than the search supports. Where this chapter says a product category has not been shown to do something, read it as “the sources consulted here do not establish it”, not as “it has been tested and failed”. The two claims are different and only the first is mine to make. Anywhere this chapter states a categorical verdict on a product class, that verdict is provisional in a way the Part II verdicts are not.
What I have done is apply the mechanisms established in earlier chapters, which are well grounded, and flag where a claim depends on evidence I have not appraised. The hydroxyapatite section in particular names what would need checking rather than reaching a verdict.
The one claim here carried over with full support is the fluoride concentration evidence from Chapter 8, which comes from a Cochrane review I read in detail.
Full record in appraisals/searches/ch29-toothpaste-aisle.md.
What this means for you
Turn the tube over and look for the fluoride concentration. If it says 1,350 to 1,500ppm, it will do. If it says nothing, or says fluoride-free, put it back unless you have a specific reason.
Everything else on the front of the packet is either a modest cosmetic effect, an unregulated word, or a claim that has not been tested to anything like the standard the fluoride claim has been.
The cheapest tube on the shelf with the right fluoride content is, on the evidence in this book, as good as the most expensive one.