Three tests, and what each one is for
China’s National Institutes for Food and Drug Control has put forward three human testing methods for cosmetic safety assessment: a Human Skin Patch Test, a Cosmetic Safety In-use Test, and a Cosmetic Safety in Long Use Test.
They are not redundant. The patch test is a controlled-exposure screen for irritation and allergens — it catches the obvious failures early and cheaply. The in-use test puts the product into normal application conditions, which is where problems that a patch never reproduces tend to surface. The long use test looks for cumulative effects, and for an abrasive category that is the one that matters.
Why abrasive formats are exposed here
A physical scrub’s active mechanism is mechanical damage, controlled and superficial. A single application is fine. Ninety applications, on a user who scrubs harder than the instructions suggest, is a different question — and it is precisely the question a long use protocol is built to ask.
Barrier disruption from over-exfoliation does not show up in a two-day patch test. It shows up over weeks, as increasing sensitivity and transepidermal water loss. A category whose failure mode is cumulative has been assessed until now largely by methods that measure the acute.
The classification detail worth reading twice
Alongside the testing methods, the guidance addresses how microcrystalline and microneedle formats should be categorised and labelled. That sounds like housekeeping. It is not.
These formats sit on a boundary — abrasive enough that some jurisdictions have questioned whether they belong in cosmetics at all. Getting an explicit classification means products in that space know which rulebook applies, rather than discovering it during a customs hold.
It also runs alongside the long shadow of the microbead bans, which pushed the whole physical exfoliant category toward alternative particles without settling how those alternatives should be assessed.
What changes in the development timeline
A long use protocol is measured in weeks of subject time. That is not something to schedule after the formula is locked and the packaging is printed.
- Test readiness moves upstream. Grain size, particle hardness and use frequency on the label all affect whether a formula clears a long use assessment. Those are early decisions, not late ones.
- Chemical exfoliants are not automatically the safe route. AHA and BHA formats avoid the abrasion question but bring their own cumulative-irritation profile, and long use testing will find it.
- Instructions become part of the dossier. If a product is tested at twice-weekly use, a label implying daily use is a mismatch a reviewer can see.
A wider direction of travel
China is not moving alone. The European SCCS has continued issuing opinions setting concentration ceilings for widely used cosmetic ingredients, and regulators across markets have tightened expectations on what a safety file has to contain. What differs is method: China’s approach leans on human in-use data where others have leaned on ingredient-level toxicology.
For a manufacturer supplying multiple markets, that means one formula may need two evidence packages. Building toward the stricter one is usually cheaper than retrofitting, and it is why formulation consistency across production runs now gets scrutiny it did not used to.
“China’s new testing requirements are a good example of how regulation is starting to shape sourcing conversations earlier than it used to. Buyers exporting into China now ask upfront whether a scrub formula can realistically pass patch, in-use, and long-use testing before they commit, rather than treating documentation as a final step before shipment. From where we sit, that’s pushed the OEM/ODM relationship earlier into product development — testing readiness has become part of the pitch, not just paperwork at the end.” — Hyejin Cho, Digital Marketing Specialist, Marketing and Communication Team at SPSCOS