Evidence

Treatments marketed far beyond their evidence

How cosmetic treatments with limited evidence are marketed as proven in the UK, the recurring patterns, and how to test a claim yourself.

The sell· Reviewed 2026-08-01·Published by Northbank Media
Archive shelving. Supply chains and records, not products.
Archive shelving. Supply chains and records, not products.
The short answer

The recurring pattern is a real mechanism, a small quantity of human evidence, and marketing that describes the mechanism as though it were the outcome. Test any claim by asking four questions: what specifically is claimed, in whom was it studied, compared with what, and who funded it. Clinically proven is not a regulated phrase, and under the CAP Code the advertiser must hold evidence for objective claims. If a claim cannot be traced to a source you can read, treat it as marketing rather than as evidence.

This publication does not name treatments as good or bad, and it will not tell you that a particular procedure does not work. That would require an evidence review we have not carried out. What we can do, and what is more useful anyway, is describe the pattern by which a thin evidence base is presented as a thick one, so that you can recognise it applied to anything.

The pattern

It runs in four steps and it is remarkably consistent.

First, a real mechanism. Something genuinely happens at a cellular or physical level, and it can be described in a way that sounds explanatory. Mechanistic plausibility is where every claim starts, and it is not misleading in itself.

Second, laboratory or animal evidence. Effects observed in cell culture or in animals are real findings, and they are also a very early stage. The proportion of promising preclinical findings that translate into demonstrable clinical benefit is low across all of medicine.

Third, small human studies. Often uncontrolled, often small, often short, often measuring satisfaction rather than an objective endpoint, and frequently funded by a party with a commercial interest.

Fourth, marketing that skips to the conclusion. The mechanism is presented as the outcome: because the mechanism exists, the result is asserted. This is where the gap opens, and it is the step at which a patient can intervene.

CheckpointTesting any efficacy claim
What exactly is claimed: outcome, magnitude, timeframeImproves skin quality is not a claim. Ask for one that could be measured.
In whom was it studiedAge, sex, skin type and baseline. Results do not transfer automatically between groups.
Compared with whatNo control group means the study cannot separate the treatment from time, attention and expectation.
Who funded and conducted itIndustry funding is normal. Absence of independent replication is the thing to weigh.
Where can I read itIf a claim cannot be traced to something you could read, it is marketing.
If the answer is not thisIf clinically proven cannot be turned into a named source, walk away from the claim.

Four questions that test any claim

What exactly is claimed? Improves skin quality is not a claim, it is a mood. A claim has an outcome, a magnitude and a timeframe. Ask for one.

In whom was it studied? Age, skin type, sex, baseline condition. Results in one group do not transfer automatically to another, and skin type in particular matters for energy based treatments.

Compared with what? This is the question that does the most work. Compared with nothing, compared with the same treatment without the active element, or compared with an established alternative. Many studies in this field have no control group at all, which means they cannot separate the treatment from time, attention, aftercare or expectation.

Who funded it, and who conducted it? Industry funding is normal and does not invalidate research. It is one input into how much weight a single study carries, particularly where independent replication is absent.

Your rights

What you are entitled to at this stage

  • Advertising that does not mislead, with the advertiser holding evidence for objective claims under the CAP Code.
  • To ask what the evidence for a claim is and where you can read it.
  • To be told when a medicine is being used off label and why.
  • To be told plainly when a treatment is new and long term data does not yet exist.
  • To report an unsubstantiated claim to the ASA, free, whether or not you are a customer.

A thin evidence base is not proof that something does not work. It is a reason to be told that nobody yet knows.

Phrases that carry no fixed meaning

Several terms appear constantly and are worth translating.

Clinically proven. Not a regulated phrase in this context. Under the CAP Code an advertiser must hold adequate evidence for objective claims, so the correct response is to ask what the evidence is. If it cannot be named, the claim is unsupported.

Medical grade. Not a regulatory category for skincare. It generally signals a distribution channel rather than a standard.

FDA approved or CE marked. These relate to regulatory clearance for marketing, which is not the same as evidence of a cosmetic benefit, and a mark for one indication does not extend to another.

Natural and biocompatible. Statements about origin, not about safety or effect.

Regenerative. A description of an intended mechanism. It is not a result.

Red flag

A treatment described as new, advanced or next generation without a single named study, and offered at an introductory price. New and evidenced are frequently in tension: a treatment that is genuinely new cannot yet have long term data, and that is a fact worth stating rather than obscuring.

Off label is not the same as unevidenced

Using a licensed medicine outside its licensed indication is lawful, common and often reasonable. It is not a scandal and it is not the same thing as a treatment being unsupported. What it does require is that the prescriber is satisfied it is appropriate, that the reasoning is recorded, and that you are told. See our guide to prescription only medicines.

Where to look yourself

You do not need to read primary research to check a claim. Where a treatment has been assessed by NICE, that assessment is public and is written to be read. The NHS publishes patient facing information on many procedures. The British Association of Dermatologists publishes patient information leaflets. Where a Cochrane review exists on a question, it is the most rigorous summary available and its plain language summaries are readable.

Absence from those sources is not proof that a treatment does not work. It usually means the evidence base has not reached the point where anyone has systematically reviewed it, which is itself a useful thing to know.

What to ask the practitioner

One question, asked plainly, does most of the work: what is the best evidence that this works, and where can I read it. A practitioner who is enthusiastic about a treatment and well informed will enjoy the question. One who cannot answer it beyond describing the mechanism has told you where the treatment sits.

A second question is worth adding: what would make you stop offering this. It reveals whether the treatment is held as a clinical position that could be revised, or as a product line.

The fair version of all this

A thin evidence base does not mean a treatment does not work. It means nobody yet knows how well it works, in whom, or for how long. That is a reasonable thing to be told and a reasonable basis on which to decide, and some patients will decide to proceed anyway. The objection is not to the treatment. It is to being sold uncertainty as though it were certainty, at a price that reflects certainty.

No commercial links on this page

This article contains no commercial links of any kind. No affiliate links, no sponsored placements, and no links to any clinic, practitioner, brand, product or retailer. Nobody paid for it, nobody previewed it and nobody outside the editorial team saw it before publication.

This publication does not name, rank, rate or review any clinic or practitioner, because it has assessed none of them. Our funding is set out in full on the about page and in our editorial standards.

Nothing here is medical or legal advice. Speak to a qualified clinician about your own circumstances, and to a solicitor about a claim.

Sources

We cite regulators, legislation and clinical institutions, and we link them so you can check the current position yourself. We do not link to clinics or retailers. Regulation in this field changes, so the primary source is always better than our summary of it.

Frequently asked questions

Is clinically proven a regulated term?

Not in this context. Under the CAP Code, advertisers must hold adequate evidence for objective claims, so the useful response is to ask what that evidence is and where it can be read.

Does medical grade mean anything?

It is not a regulatory category for skincare. In practice it generally describes a distribution channel rather than a verified standard.

Does CE marking or FDA clearance prove a treatment works?

No. Those relate to regulatory clearance for marketing a product, which is a different question from whether a particular cosmetic benefit has been demonstrated, and clearance for one indication does not extend to another.

Is off label use a warning sign?

No. Off label prescribing is lawful and common. What matters is that the prescriber is satisfied it is appropriate, records the reasoning, and tells you.

Where can I check a treatment claim myself?

NICE assessments, NHS patient information, British Association of Dermatologists leaflets and Cochrane reviews where they exist. Absence from those sources usually means the evidence base has not yet been systematically reviewed.

When the rules change, we will tell you

One email when a law or a regulator's guidance changes in a way that alters what you should check, and when a new article is published. No treatment offers, no product recommendations and no clinic suggestions, because we do not make any.

Sponsor lineThe newsletter may carry one clearly labelled sponsor line, placed after the editorial content. Sponsors see nothing before it is sent and cannot suggest, approve or veto any item.