Skin Biology ·

Beauty Science 03 — Ingredient Evidence

03 — EVIDENCE

What does the science actually support?

Skincare is surrounded by scientific language.

Clinically tested. Dermatologist tested. Proven. Scientifically advanced. Clinically proven. Visible results.

These phrases can sound authoritative while conveying surprisingly little about the quality of the underlying evidence.

Science does not begin with a claim and search for information that makes the claim sound convincing. It begins with a question, establishes a method for testing that question, examines the results, considers uncertainty and determines how strongly the evidence supports a conclusion.

For skincare, this distinction is essential because there are several different levels of evidence:

A molecule may have a plausible mechanism.
An ingredient may have laboratory evidence.
A formulation may have human clinical data.
A finished product may have product-specific evidence.
And a marketing claim may still go beyond what any of those data demonstrate.

Understanding evidence means learning to separate these levels.

CLINICAL CLAIMS VS MARKETING CLAIMS

A claim is a statement about what a product, ingredient or treatment does.

A marketing claim may be technically worded, emotionally persuasive or scientifically framed without necessarily representing the strength of the evidence behind it.

Consider the difference between:

“Contains niacinamide.”

and

“Clinically proven to transform skin.”

The first is primarily a compositional statement.

The second makes a biological and consumer-outcome claim that requires evidence.

The important question is therefore:

What exactly is being claimed, and what evidence would be required to support it?

A meaningful claim should specify, directly or indirectly:

what outcome is being measured;
in whom;
under what conditions;
over what period;
compared with what;
using what measurement method;
and with what magnitude of effect.

A laboratory experiment cannot automatically substantiate a consumer claim. Evidence that an ingredient influences a cellular pathway does not establish that a finished cream produces a visible clinical improvement.

The closer the claim is to the actual consumer outcome, the more important human, product-specific evidence becomes.

HOW TO EVALUATE SKINCARE STUDIES

A study should never be judged solely by the words clinical study.

Start with the question:

What was actually tested?

Then examine the study design.

Important considerations include:

Population — Who participated? Age, sex, skin characteristics, baseline severity and relevant exclusions can affect applicability.

Sample size — How many participants were studied? A small study may detect a large effect, but estimates can be imprecise and vulnerable to chance findings.

Control group — Was there a meaningful comparator?

Randomization — Were participants randomly assigned to treatment groups?

Blinding — Did participants, investigators or assessors know which treatment was being received?

Comparator — Was the product compared with placebo, vehicle, no treatment, an established treatment or another active?

Duration — Was the study long enough to measure the claimed outcome?

Outcome measures — Were results based on validated instruments, instrumental measurements, standardized photography, expert grading or subjective participant reports?

Statistical analysis — Were the methods appropriate? Were confidence intervals reported? Were multiple outcomes or multiple comparisons considered?

Magnitude of effect — Was the difference statistically significant, and was it large enough to matter to a person using the product?

Reproducibility — Have similar findings been observed independently?

A scientifically interesting result is not automatically a clinically meaningful result.

WHAT DOES “CLINICALLY TESTED” MEAN?

“Clinically tested” is not, by itself, a measure of efficacy.

The phrase generally indicates that some form of testing involving human participants has been performed, but the scientific value of that testing depends on its design.

A study could involve a small number of people, have no control group, use subjective assessments and produce a positive result.

That is still technically human testing.

But it is very different from a well-designed randomized controlled trial with an appropriate comparator, validated outcome measures, adequate sample size and transparent statistical analysis.

When encountering clinically tested, ask:

Tested on whom?
How many people?
For how long?
Against what?
Measuring what?
By which method?
And what was the size of the observed effect?

The phrase alone cannot answer those questions.

WHAT DOES “DERMATOLOGIST TESTED” MEAN?

“Dermatologist tested” does not necessarily mean “dermatologist proven effective.”

It generally indicates dermatologist involvement in some form of testing or assessment, but the exact protocol can vary.

The dermatologist may have participated in:

safety or irritation assessment;
patch testing;
clinical evaluation;
tolerability assessment;
grading of visible changes;
or another study protocol.

The phrase does not automatically establish:

efficacy;
superiority over another product;
mechanism of action;
long-term benefit;
or clinical significance.

The scientifically useful question is therefore not:

Was a dermatologist involved?

It is:

What did the dermatologist test, using what protocol, in how many people, and what was actually demonstrated?

WHAT DOES “PROVEN” ACTUALLY MEAN?

In ordinary language, proven sounds absolute.

Scientific evidence rarely operates that way.

Scientific conclusions exist along a spectrum of confidence. Evidence may be:

plausible → preliminary → suggestive → supported → strongly supported

depending on the quantity, quality, consistency and relevance of the available data.

An ingredient may be well supported for one function but poorly supported for another.

For example, evidence that a substance improves hydration does not establish that it reduces wrinkles, repairs collagen, treats pigmentation and reverses aging.

Evidence is claim-specific.

The stronger the word proven, the stronger the evidence should be.

A scientifically responsible interpretation therefore asks:

Proven to do what, under what conditions, compared with what, and supported by which evidence?

INGREDIENT EVIDENCE

Ingredient evidence concerns the scientific literature surrounding a particular substance.

This evidence can exist at several levels.

In vitro evidence examines isolated cells, tissues, enzymes or biochemical systems.

Ex vivo evidence examines biological material outside the living organism.

In vivo animal evidence examines effects in living organisms, but animal findings do not automatically translate to human skin.

Human observational evidence identifies associations in people but generally cannot establish causation on its own.

Human intervention studies provide stronger evidence when appropriately designed.

Systematic reviews and meta-analyses can synthesize multiple studies, although their conclusions depend on the quality and comparability of the underlying research.

Ingredient evidence should therefore be viewed as a hierarchy rather than a single yes/no label.

PRODUCT EVIDENCE

Ingredient evidence is not the same as product evidence.

This distinction is fundamental.

Suppose an ingredient has strong evidence demonstrating a particular effect under specific experimental conditions.

That does not automatically prove that a finished cosmetic containing the ingredient produces the same effect.

The finished product may differ in:

concentration;
chemical form;
pH;
vehicle;
stability;
release characteristics;
packaging;
application frequency;
exposure time;
and interactions with other ingredients.

Therefore:

Ingredient evidence tells us what the ingredient may be capable of doing.

Product evidence tells us what the finished formulation actually demonstrated under defined conditions.

The second is more directly relevant to a product claim.

DOES THIS INGREDIENT WORK?

An editorial series for the URIBHO Evidence library

This question should be deceptively simple:

Does this ingredient work?

But the scientific answer is rarely simply yes or no.

The better question is:

Does this ingredient produce the claimed effect, at a relevant concentration, in the appropriate formulation, when used on human skin under realistic conditions?

Every Does This Ingredient Work? investigation should examine the evidence through several layers:

01 — MECHANISM

What biological or chemical mechanism has been proposed?

Does the mechanism have credible support?

A plausible mechanism is useful, but mechanistic plausibility is not proof of clinical efficacy.

02 — INGREDIENT IDENTITY

Which chemical form was actually studied?

The name used in marketing may conceal important differences in:

molecular structure;
purity;
extraction method;
chemical derivatives;
salts;
stereochemistry;
standardization;
or biological preparation.

03 — CONCENTRATION

At what concentration was the effect demonstrated?

Was that concentration used in the finished product?

An ingredient studied at one concentration cannot automatically be assumed to produce the same effect at another.

04 — FORMULATION

Was the ingredient studied alone, in a laboratory solution, or within a finished formulation?

Vehicle composition can influence stability, release, penetration and exposure.

05 — HUMAN EVIDENCE

Has the ingredient been tested on people?

If so:

how many?
who were they?
for how long?
what was the comparator?
what was measured?
and how large was the effect?

06 — CLINICAL RELEVANCE

Was the measured difference actually meaningful?

A statistically significant result can be too small to produce a noticeable or practically important improvement.

07 — CONSISTENCY

Do multiple studies point in the same direction?

A single positive study is evidence.

A consistent body of well-designed evidence is much stronger evidence.

08 — LIMITATIONS

What remains uncertain?

A scientifically honest assessment must identify what the evidence does not establish.

The final verdict should therefore never be merely:

WORKS / DOESN'T WORK

Instead, URIBHO can classify evidence according to the strength and relevance of support:

Strongly supported
Well supported
Promising but limited
Mixed evidence
Insufficient evidence
Evidence does not support the claim

This transforms Does This Ingredient Work? from a product-rating exercise into a genuine scientific literacy series.

BOTANICAL INGREDIENTS: TRADITION VS EVIDENCE

Botanical ingredients require particular scientific care.

Traditional use can provide an important historical record of how a plant has been used, prepared and valued.

But traditional use and clinical efficacy are different forms of evidence.

A plant may contain hundreds of chemical constituents. Its composition can vary according to:

species;
cultivar;
geographical origin;
soil;
climate;
harvest time;
plant part;
extraction solvent;
extraction temperature;
processing;
storage;
and standardization.

Consequently, “contains botanical X” does not necessarily define a reproducible biological preparation.

Evidence should distinguish between:

traditional use → phytochemical evidence → laboratory evidence → human evidence → product-specific evidence.

A botanical can be scientifically interesting without having sufficient clinical evidence for every claim made about it.

Tradition deserves respect.

Evidence determines what we can responsibly conclude.

CONCENTRATION VS EFFICACY

More is not necessarily better.

An ingredient may exhibit:

a threshold effect;
a concentration-dependent response;
a plateau;
diminishing returns;
or an adverse-effect threshold.

Biological systems are rarely simple linear switches.

The relationship between dose and response may resemble a curve rather than a straight line.

Concentration must also be interpreted alongside exposure.

Two formulations containing the same nominal concentration can produce different effective exposure because their vehicles differ in release, partitioning, evaporation, retention and penetration.

Therefore:

percentage on the label ≠ biological dose at the target site.

This is one of the most important concepts in cosmetic evidence.

CORRELATION VS CAUSATION IN SKINCARE

Correlation means that two variables are associated.

Causation means that changing one variable produces a change in another.

This distinction is critical in skincare research.

Suppose people who regularly use a particular skincare product have better-looking skin.

That observation does not prove that the product caused the improvement.

Those users may also:

spend less time in the sun;
use sunscreen;
have different socioeconomic circumstances;
follow different diets;
have different healthcare access;
or use multiple skincare products.

These are potential confounding variables.

Similarly, an ingredient that is more common among people with healthy skin is not necessarily responsible for their healthy skin.

Well-designed intervention studies, randomization, appropriate controls and careful statistical analysis help reduce these problems.

Association can generate a hypothesis. Controlled evidence is needed to establish causation.

READING THE EVIDENCE LIKE A SCIENTIST

When evaluating a skincare claim, move through five questions:

1. WHAT IS THE CLAIM?
Define exactly what is being promised.

2. WHAT WAS ACTUALLY TESTED?
Ingredient, formulation, treatment protocol or finished product?

3. HOW WAS IT TESTED?
Laboratory experiment, observational study, controlled trial, instrumental measurement or subjective assessment?

4. HOW LARGE AND CONSISTENT IS THE EFFECT?
Statistical significance is not the same as practical significance.

5. HOW CERTAIN SHOULD WE BE?
Consider study quality, sample size, replication, limitations and relevance to real-world use.

The result is not simply true or false.

The scientifically useful conclusion is:

How strongly does the available evidence support the specific claim?

THE URIBHO EVIDENCE STANDARD

URIBHO's approach to skincare science should be neither cynical nor credulous.

Not everything marketed as science is scientifically strong.

But equally:

Not everything lacking a large clinical trial is ineffective.

Scientific judgment requires proportion.

Strong evidence should be described as strong.

Preliminary evidence should be described as preliminary.

Traditional knowledge should be distinguished from clinical evidence.

Laboratory findings should not be inflated into human outcomes.

Ingredient evidence should not automatically be presented as product evidence.

And uncertainty should not be disguised as certainty.

The purpose of evidence literacy is not to make skincare less exciting.

It is to make the excitement deserved.

SCIENTIFIC REFERENCES

Sackett, D. L., et al. Evidence based medicine: what it is and what it isn't. BMJ.
Guyatt, G. H., et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ.
Higgins, J. P. T., et al. Cochrane Handbook for Systematic Reviews of Interventions. Cochrane.
Schulz, K. F., Altman, D. G., Moher, D. CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials. BMJ.
Sterne, J. A. C., et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ.
Page, M. J., et al. PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ.
U.S. Food and Drug Administration. Guidance for Industry: Cosmetic Labeling Guide.
European Commission. Regulation (EC) No 1223/2009 on Cosmetic Products.
International Council for Harmonisation. ICH Harmonised Guideline E6: Good Clinical Practice.
World Health Organization. WHO Guidelines for Good Clinical Practice for Trials on Pharmaceutical Products.
Evans, W. C. Trease and Evans' Pharmacognosy. Elsevier.
Williamson, E. M., Driver, S., Baxter, K. Stockley's Herbal Medicines Interactions. Pharmaceutical Press.