Which products need food contact testing before market access: part 4
Borderline food contact products need testing decisions based on real use rather than product names. Bundled accessories, dual-use decorations, children's ware, and reheatable containers each deserve scoping by their harshest foreseeable food contact, with representative sampling tying every result to a defined lot. Buyers who apply this method avoid both untested surprises and unnecessary blanket testing.
Key takeaways
- Scope testing by the harshest foreseeable food contact shown on packaging or listings.
- Bundled accessories that touch food need their own parameters and results.
- Representative sampling across variants and cartons is what links a report to a lot. [1]
- Factory acceptance logic releases a defined lot only when every required parameter passes. [2]
- Dual-use and decorative articles are scoped by actual food use, not by marketing labels.
Some food contact products sit clearly inside testing scope, such as baby bottles and cooking utensils. Others sit at the boundary: decorative pieces that may hold food, bundles mixing food contact and non-contact items, accessories tucked into packaging, or articles whose uses span gentle storage and harsh reheating. These borderline cases cause a disproportionate share of market access problems because buyers guess instead of scoping.
The remedy is a use-first method. Describe every foreseeable food contact for the product as sold and pictured, identify the harshest combination of food type, temperature, time, and repetition, and scope testing to that combination. Components that never touch food are documented as out of scope with reasons; everything else enters the programme with its own parameters.
This fourth part of the market-access series applies that method to bundled sets, accessories, children's and reheatable ware, decorative dual-use articles, and refurbished or changed variants. Sampling discipline and factory acceptance logic run through every case so that results stay tied to defined lots and release decisions rest on complete evidence.
Bundled sets are the most common borderline case. A storage set may combine containers, lids, seals, and serving spoons from different materials and suppliers; a tea set may mix cups, pots, filters, and trays; a lunch kit may bundle boxes, bottles, cutlery, and ice packs. Each food contact piece needs its own line in the scope table with material, contact type, and parameters, because one component's pass never covers another's chemistry.
Non-contact outer items still need a documented decision. An outer gift box, a carry sleeve, or a display tray that never touches food can be recorded as out of scope with a brief reason, which prevents later arguments about whether it was forgotten or deliberately excluded. Where set-off or vapour transfer is plausible, for example printed tissue inside a food box, either test the pathway or change the packing so the pathway no longer exists.
Variant discipline matters as much as component discipline. Colour variants with different glazes or inks, size variants with different surface-to-volume ratios, and supplier variants with different grades should be sampled and reported so the coverage is explicit. Testing one quiet variant while shipping the full range leaves most of the range undescribed, and reviewers increasingly ask for the variant mapping by name. Sellers who want help structuring multi-variant sampling can ask UTS to review the scope table before the laboratory programme starts.
Accessories cause failures out of proportion to their size. Straws, valves, spouts, seals, gaskets, filters, desiccant sachets that sit inside food packs, and measuring scoops packed with powders all touch food or food vapour directly. Their materials often differ entirely from the main body: silicone seals in plastic boxes, plastic valves in steel bottles, paper filters in plastic housings. Each needs assessment as its own food contact component.
The scoping rule is simple: if normal use brings it into contact with food, drink, or vapour that condenses back into food, it enters the programme. State the contact type, the material, and the applicable parameters for each accessory, and submit production-state accessories rather than generic equivalents. When accessory suppliers change, which happens frequently for seals and straws, treat the product as a new state and verify before shipment.
Small-part handling during sampling deserves attention. Accessories packed loose inside cartons can be missed by samplers who focus on main bodies, so the sampling instruction should name them explicitly with quantities. The laboratory submission lists them as separate test items where the laboratory recommends it. A product file that names every accessory and shows its result reads as credible; one that never mentions them reads as incomplete.
Sampling across mixed lots needs explicit planning because borderline assortments are rarely uniform. Acceptance sampling concepts frame the task: define the lot, state how many articles are drawn and by which method, and apply a clear decision rule to the result. Random or systematic drawing across cartons and variants gives every article a fair chance of inclusion, while convenience drawing from the nearest carton does not. [1]
For multi-variant sets, stratify the drawing so each variant and each food contact component is represented. Record carton numbers, variant identities, sampler identity, and seal numbers so the laboratory report can be tied back to a defined population. Where production spans several material batches, treat each batch subgroup as its own lot rather than blending results across different inputs.
Keep the sampling pages with the report permanently. Months later, when a customer or authority asks what a report covers, the sampling record answers the question in one page. Without it, even a technically excellent laboratory programme floats free of any defined goods and cannot support a release decision. Sampling documentation is the cheapest insurance in the whole compliance file.
| Borderline case | Scoping rule | Sampling focus | Release condition |
|---|---|---|---|
| Borderline case | Scoping rule | Sampling focus | Release condition |
| Gift and storage sets | Each food contact piece scoped separately | Stratify across components and colours | Every component meets its limits |
| Straws, valves, seals | Direct or vapour contact enters scope | Name accessories in sampling instruction | Each accessory reported |
| Children's feeding ware | Scope by most sensitive foreseeable use | Represent production lots fully | Harsher-condition programme passed |
| Reheatable and microwaveable | Test under reheating conditions | Include reheated-state samples | Claims match tested conditions |
| Decorative dual-use articles | Actual food use governs, not labels | Sample decorated finished state | Decoration and body both pass |
| Changed variants | New material or process is a new state | Sample the changed lot anew | New lot reported before shipment |
Children's feeding articles, reheatable containers, and decorative dual-use pieces justify the broadest programmes. Sensitivity, heat, acidity, repetition, and decoration converge in these groups, so scope by the harshest foreseeable use shown anywhere on the packaging, listing, or instructions. A bowl pictured with hot soup is a hot-use bowl whatever the fine print says; a decorative platter described as serving use is a food contact article whatever the marketing theme claims.
Factory acceptance logic governs the ending: laboratory results feed a lot-level release decision, and release follows only when every required parameter for every in-scope component meets its limit for the tested lot. Failures trigger investigation into materials and process, corrective action, and resampling rather than selective retesting. The disposition, with reasons, is filed alongside the report and sampling record so the decision stays auditable. [2]
Changed variants restart the cycle rather than inheriting old approvals. New colours, new decoration systems, new material grades, new accessory suppliers, and changed firing or curing conditions each create a new product state that needs its own sampling and testing before market access claims continue. Buyers who treat change as a retest trigger spend a little on verification and avoid the far larger cost of withdrawing non-covered goods.
Sources
- Wikipedia: Acceptance sampling —— Overview of acceptance sampling including AQL concepts used in product inspection.
- Wikipedia: Factory acceptance test —— Overview of factory acceptance testing performed before shipment or installation.
Frequently asked questions
Do accessories and bundled utensils need their own testing?
Accessories that touch food or vapour during normal use, such as straws, valves, seals, spoons packed with containers, and infusers, should be treated as food contact articles in their own right. Include each accessory in the scope table with its material and test parameters. If an accessory is genuinely never food contact, document that reasoning instead of leaving it unmentioned.
How should sampling work for mixed-material or multi-variant lots?
Acceptance sampling theory explains how a defined plan links tested articles to a lot-level decision, including the risks of wrong decisions in either direction. For buyers, this means stating the lot, the sample size, and the selection method, then drawing representatively across cartons. A documented plan is what makes the report defensible later.
Which product uses justify broader or harsher test programmes?
Children's feeding articles, reheatable containers, and long-contact storage ware face harsher or more sensitive uses, so their programmes are typically broader and their conditions more severe. Scope by the most demanding foreseeable use shown on the packaging or listing, not by the mildest possible interpretation. When in doubt, test the harsher case.
How do test results translate into a release decision?
Factory acceptance practice treats test results as release inputs for a defined lot: release follows only when every required parameter meets its limit, and failures trigger investigation and corrective action. Testing one variant or lot never releases unrelated variants or later lots. Keep the disposition record with the report and sampling pages.