CFS 2012 — Safety Issues of Baby Bottles and Children’s Tableware (RAS Report No. 47)

The Hong Kong Centre for Food Safety (CFS) review summarises the food-contact-material safety landscape for baby bottles, sippy cups, and children’s tableware as of January 2012. The bulk of the document covers two organic migrants — bisphenol A (BPA) from polycarbonate (PC) baby bottles and formaldehyde from melamine-ware — and their regulatory status across the EU, US, mainland China, Australia, and Hong Kong. Heavy-metal relevance for HMI is concentrated in Annex II (Overview of Common Non-Plastics): the table identifies lead and cadmium as the migrants of concern from ceramic glazes and decorative paints, nickel and chromium (including chromium VI) as the components of concern from stainless steel tableware, and aluminium as the component of concern from aluminium drinking bottles. The body narrative additionally notes that migration of heavy metals from metal-ware increases when the article contacts highly acidic foods, and the public advice section recommends avoiding holding highly acidic foods in metal tableware for long periods. No original heavy-metal occurrence or migration data are reported; the document is a qualitative regulatory review with IARC classifications and pointers to the EU, US, and Chinese specific migration limits.

Key numbers

The document is a narrative literature review, not a measurement study. No heavy-metal concentrations, migration values, sample sizes, LODs, or percentiles are reported for any metal. The numbers below are the regulatory thresholds, IARC classifications, and reference values the report quotes from third-party sources.

Heavy-metal-relevant migration limits and regulatory references (Annex II, p. 49–50):

Heavy-metal-relevant narrative claims (Background §3-4, p. 6–7; Executive Summary §4, p. 3):

• “The major concern on metals and ceramics is the migration of heavy metals such as lead and cadmium.” (Background §3, p. 6.)
• “The migration of heavy metals in metal-ware may increase when the article comes into contact with highly acidic foods.” (Executive Summary §4, p. 3; restated in Background §3, p. 6–7.)
• General principle (Executive Summary §4, p. 3; Background §3, p. 7): chemical migration from food contact materials in general increases with temperature, fat content, acidity of food, and contact time. Plastics are particularly sensitive to fat and temperature; metals and ceramics are particularly sensitive to acidity.

Public advice on metal tableware (Summary and Recommendations §“Use”, p. 33–34):

• “Avoid using metal tableware to hold highly acidic foods for long period of time.”
• “Avoid using tableware that is broken or damaged on its surface.” (Damaged glaze on ceramic tableware is a recognised driver of lead/cadmium migration; the report does not explicitly link the breakage-avoidance advice to heavy-metal migration but the connection is in the cited toxicology references.)

Adverse-health-effect summaries (Annex II, p. 49–50):

• Lead from ceramic glazes: “Chronic exposure caused developmental neurotoxic effects in humans.” (Citing WHO Food Additives Series No. 44, Lead, 2000, footnote 12.)
• Cadmium from ceramic glazes: “Chronic exposure affected affect [sic] kidney in humans.” (Citing WHO Food Additives Series No. 52, sixty-first JECFA meeting, 2004, footnote 13. The doubled “affected affect” is a typo in the source Annex II row as printed on p. 49 and is preserved here so the quotation is verbatim.)
• Nickel from stainless steel: “Not poisonous in small quantities but can provoke a reaction in people allergic to nickel.” (Citing Health Canada, The Safe Use of Cookware, 2011, footnote 14.)
• Chromium from stainless steel: “Chronic excessive exposure to chromium (VI) caused development toxicity in experimental animals. Some people are allergic to chromium (VI) and (III).” (Citing ATSDR ToxFAQs for Chromium, September 2008, footnote 15.)
• Aluminium from aluminium drinking bottles: “Chronic exposure affected the reproductive system and developing nervous system in experimental animals.” (Citing WHO/JECFA sixty-seventh meeting, WHO Technical Report Series 940, 2007, footnote 16.)

Regulatory regime citations relevant to heavy metals from food contact materials:

• EU Directive 84/500/EEC: ceramic articles intended to come into contact with foodstuffs; sets specific migration limits for lead and cadmium from ceramic-ware. Amended by Commission Directive 2005/31/EC concerning the declaration of compliance and performance criteria of the analytical method (Annex II, p. 49).
• USFDA Compliance Policy Guide Sec. 545.400 and Sec. 545.450: Pottery (Ceramics); Imported and Domestic — lead contamination action levels (Annex II, p. 49).
• Mainland China GB 13121-91: hygienic standard for ceramic food containers; GB 9684-2011: hygienic standard for stainless steel food containers; GB 11333-1989: hygienic standard for aluminium food utensils; GB 9685-2008: hygienic standard for additives in food containers and packaging materials (Annex II, p. 49–50).
• Hong Kong Consumer Goods Safety Ordinance (CGSO), Cap 456: enforced by the Customs and Excise Department; sets a general safety requirement applicable to baby bottles and tableware supplied for private use. CFS does not derive heavy-metal-specific migration limits at the Hong Kong level; compliance is referenced against EU, US, and mainland China standards (Background §12–14, p. 10–11).

Non-heavy-metal regulatory numbers carried for completeness (not HMI-scope but referenced in the same document):

• BPA tolerable daily intake (TDI): 0.05 mg/kg bw, EFSA 2006, reconfirmed 2008 and 2010 (§29, p. 17). EU Reg. 10/2011 specific migration limit for BPA from food contact materials: 0.6 mg/kg food (§30, p. 17). Joint FAO/WHO 2010 worst-case realistic BPA migration estimate from PC baby bottles: 15 µg/L (§31, p. 17).
• Melamine TDI: 0.2 mg/kg bw, WHO/FAO 2008 (§55, p. 26). EU specific migration limits for plastics: melamine 30 mg/kg food (5.0 mg/dm²); formaldehyde 15 mg/kg food (2.5 mg/dm²). Mainland China migration limits: melamine 2.5 mg/dm²; formaldehyde 0.2 mg/dm² (note the report’s wording suggests these may be transposed between the two analytes; the source text reads “the migration limits for melamine and formaldehyde is 0.2 mg/dm² and 2.5 mg/dm², respectively” but the EU rows in the same paragraph have melamine 30 mg/kg / 5.0 mg/dm² and formaldehyde 15 mg/kg / 2.5 mg/dm², so the China assignment in the source as printed appears to assign 0.2 mg/dm² to melamine and 2.5 mg/dm² to formaldehyde — preserved as-printed because this is a regulatory-citation field and the wiki should report what the source says, not what the wiki thinks the source should have said) (§56, p. 27).

Methods (brief)

Narrative literature review. Search strategy (§16–18, p. 12): academic resources via Google Scholar, EBSCO, informaworld, and Wanfang databases; internet resources via Google. Major keywords: “baby bottles and safety”, “tableware and safety”, “children’s tableware and safety”, “bisphenol A”, “polycarbonate”, “formaldehyde”, “melamine-ware”. Search dates limited to 1980–2011 inclusive, English or Chinese language. Additional references obtained from WHO, FAO, EFSA, USFDA, FSANZ, Health Canada, UK FSA, BfR Germany, ILSI, and trade-association websites (Plastics New Zealand, British Plastics Federation, PlasticsEurope, British Stainless Steel Association, American Ceramic Society, Aluminum Association). Reference lists of retrieved documents were hand-searched.

No analytical measurements of heavy metals are reported. The single piece of original CFS testing referenced (61-sample melamine-ware survey, 2010) covered melamine and formaldehyde migration only and is not heavy-metal data.

Implications

Certification. The report contributes regulatory-context and qualitative-toxicology reference material for the infant-bottle, sippy-cup, tableware-ceramicware, tableware-glassware, tableware-metal-bottles-tumblers, and aluminium-bottle product categories. It does not contribute occurrence data for any HMI heavy-metals percentile calculation. The Annex II regulatory pointers (84/500/EEC, 2005/31/EC, USFDA CPG 545.400 and 545.450, GB 13121-91, GB 9684-2011, GB 11333-1989) are the canonical entry points for the ceramic-ware lead/cadmium and stainless-steel/aluminium contamination-platform pages and would be the appropriate reference inputs when those product-category pages are scaffolded under Step 0 Lock.

App. The acidity-drives-migration framing (metals leach more from glaze and stainless steel into acidic foods) is the kind of qualitative use-condition guidance that the consumer app can attach to ceramic-ware and metal-ware product entries.

Courses. The IARC classifications (Pb 2A, Cd 1, Cr(VI) 1, Ni compounds 1) and the WHO/JECFA reference-value pointers are usable as introductory framing for course modules on heavy-metal toxicology and food-contact-material risk, with the caveat that the underlying toxicology references (WHO Food Additives Series No. 44 for Pb, No. 52 for Cd, JECFA sixty-seventh meeting for Al, ATSDR ToxFAQs for Cr) should be cited directly rather than through this 2012 review.

Verification notes

• 2026-05-31 fresh ingest (Claude Opus 4.7, autonomous v2.0 manual-fetch skill, NEW path). Three identity checks against wiki/sources/: DOI grep returned no matches (paper has no DOI); raw_handle grep MFK_06-2012-programme-rafs-fcm-01-04-report-e returned no matches; cite-key grep cfs2012-baby-bottles-tableware returned no matches. The sibling page cfs2012-hktds-inorganic-arsenic.md exists and is a different CFS 2012 product (1st HKTDS Report No. 2 on inorganic arsenic dietary exposure); not a duplicate. PDF was read in 5–12 page chunks across pages 1–60. Heavy-metal content is concentrated in Annex II (p. 49–50) plus brief narrative passages in Executive Summary §4 (p. 3), Background §3 (p. 6–7), and Summary §“Use” (p. 33–34). The bulk of the body (p. 13–32) covers BPA toxicology and melamine/formaldehyde migration and is out of HMI scope; key BPA and melamine reference numbers retained in Key numbers for completeness because the report is one document.
• evidence_tier: B reflects literature-review status with no original heavy-metal measurements. Sibling government-agency review reports in the corpus use B-tier for review-and-summary documents; A-tier is reserved for primary measurement studies (e.g., the HKTDS papers cfs2012-hktds-inorganic-arsenic and cfs2013-hktds-metallic-contaminants).
• metals: [Pb, Cd, Ni, Cr, Cr-VI, Al]. Each is named explicitly in Annex II as a migrant of concern for one of the food-contact materials. Cr is included as the unspeciated marker and Cr-VI is included separately because the report quotes the Cr(VI)-specific IARC Group 1 classification and the Cr(VI)-specific developmental-toxicity reference (ATSDR ToxFAQs for Chromium). The strict Cr-VI speciation discipline is preserved: when the report makes statements specifically about Cr(VI), they are tagged as Cr-VI in this page; when the source talks about “chromium” as a stainless-steel component generically, it falls under Cr. tHg, MeHg, iAs, tAs, Sn, U are not in metals: because the report does not discuss them.
• ingredients: []. This is a food-contact-material report, not a food-commodity report. No ingredient-level findings are attributable.
• products: lists the ten product slugs that map to the food-contact-material categories the report discusses: PC baby bottles and alternatives → infant-bottles; rubber and silicone teats → infant-bottle-nipples; sippy/training cups → sippy-cups-toddler-drinkware; ceramic tableware → tableware-ceramicware; glassware → tableware-glassware; aluminium drinking bottles and stainless-steel containers → tableware-metal-bottles-tumblers; plastic tableware (PP, melamine, etc.) → tableware-plastic-drinkware; ceramic decorative decals → tableware-decorative-decals-coatings; cutlery and chopsticks → utensils-metal (stainless steel) and utensils-plastic-silicone-tools (plastic, melamine, silicone). All slugs are present in the current taxonomy snapshot (verified against docs/gpt-collaboration/taxonomy-snapshot.md, generated 2026-05-18).
• matrices: uses the bare-string vocabulary established across food-contact-material source pages: food-contact-material (top-level), infant-feeding-bottle, tableware, ceramic-ware, stainless-steel-ware, aluminium-ware. The matrices vocabulary is not enumerated in the taxonomy snapshot (that snapshot covers ingredients/products/metals/regulations only); the bare-string convention follows the precedent from kaya2024-milk-packaging-heavy-metals.md (food-packaging, packaging-material), kim2024-food-container-metal-migration-korea.md (food-packaging-migration, food-simulant), and onghena2016-baby-bottle-migration-duration-tests.md (which uses an empty matrices: [] because that paper measures organic migrants only).
• jurisdictions: [HK, EU, US, CN]. The report is a Hong Kong agency publication; its regulatory survey covers EU (84/500/EEC, 2005/31/EC, 10/2011, etc.), US (USFDA CPG 545.x, 21 CFR 177.x), and mainland China (GB standards) frameworks. Australia and New Zealand are mentioned but only as references to EU/US/Australian standards, not as primary regulatory contributors.
• Brand firewall (Part 12 strict reading). No brand names are reported in this page. Annex III of the source PDF (p. 55–57) consists of photographs of baby bottles and tableware grouped by material; the photo captions name only the material type (Polycarbonate, Melamine, Polypropylene, Wood and bamboo, Ceramic and glass, Stainless steel, Aluminium, Silicone, Rubber, etc.), never a brand. Some plates do depict visually identifiable licensed-character merchandise (e.g., a Hello Kitty aluminium bottle, a SpongeBob SquarePants stainless-steel thermos) but no brand or character text crosses into the wiki page, and no contamination value is attached to any of the depicted items. The melamine-ware survey (§61, p. 28) is summarised in aggregate (“61 samples, 7 categories, 9 brands”) without naming any brand; this is a regulatory-event-style summary (the survey is a public CFS testing programme), not a brand-by-brand ranking.
• Wiki/HMTc firewall (Part 2). No HMTc threshold suggestions, comparisons, or rationale claims are made on this page. The Implications section names the product categories the regulatory pointers feed into without proposing certification values. No cross-source synthesis claims (“consistent with the consensus that…”) are introduced; the page reports only what this one CFS 2012 review states.
• Possible source-text inconsistency carried verbatim. §56 (p. 27) lists the mainland China migration limits as “the migration limits for melamine and formaldehyde is 0.2 mg/dm² and 2.5 mg/dm², respectively”, which numerically inverts the typical EU ordering (the EU row in the same paragraph reads melamine 30 mg/kg / 5.0 mg/dm² and formaldehyde 15 mg/kg / 2.5 mg/dm²). The wording is reported verbatim in Key numbers with a note flagging the possible transposition; the wiki does not silently “correct” what the source prints. This is not an HMI-scope value and was preserved only because it falls within the same Key numbers section.
• 2026-05-31 audit-application (fresh-context subagent, verdict PROMOTE). Subagent ran all five checks against the PDF. Checks 1, 2, 3, 5 were clean. Two ⚠️ items applied here:
  • CHECK 1 ⚠️ — the cadmium adverse-health-effect quote on the wiki page rendered “Chronic exposure affected kidney in humans.” inside quotation marks, but the source Annex II row as printed on p. 49 contains a doubled word (“Chronic exposure affected affect kidney in humans”). Verified against the PDF independently; the source typo is real. Corrected the wiki quote to “Chronic exposure affected affect [sic] kidney in humans.” and added a one-line in-Key-numbers note documenting that the doubled word is in the source.
  • CHECK 4 ⚠️ — Annex III photographs (p. 56–57) include visually identifiable licensed-character merchandise. Verified against the PDF; the captions name only the material type, not the brand or character. No brand or character text crosses into the wiki and no contamination value is attached to any depicted item. Tightened the brand-firewall verification note above to acknowledge the licensed-character merchandise explicitly rather than asserting the photographs are unbranded.

Page history

The five most recent substantive edits to this page. The full version history lives in git; when DOI minting comes online (see schema docs), each entry below will also link to a version-pinned DataCite DOI.