OECD 2019 — Considerations when assessing children’s exposure to chemicals from products (Series on Testing and Assessment No. 310)

This OECD Environment Directorate / Joint Meeting of the Chemicals Committee and Working Party on Chemicals, Pesticides and Biotechnology document presents a five-question decision tree (Q0–Q5) for determining whether a child-specific exposure assessment is needed for a chemical in a consumer product, and a checklist of child-specific physiological and behavioural factors (Table 1) that influence inhalation, oral and dermal exposure routes. Three modelled case studies illustrate the framework: two hair sprays (non-VOC and VOC), a kitchen cleaner spray, and a post-application MCPA herbicide scenario in a public park. The focus is “industrial chemicals” in “consumer products”; exposure via food or the environment is explicitly treated as background and is out of scope. No analytical occurrence data on heavy metals are reported; the document is methodology and exposure-modelling guidance, not an occurrence study.

Key numbers

• WHO age sub-classifications used (Section 1.3, p. 11, citing Cohen Hubal et al. 2014): neonate <1 month; infant 1 month–1 year; toddler 1–2 years; early childhood 2–6 years; middle childhood 6–11 years; early adolescence 11–16 years; late adolescence 16–21 years. “Child” spans birth to 21 years per WHO. No earlier life stages (in utero) or pregnant women are within scope of “child-specific” in this document.
• Decision-tree outcomes (Figure 1, p. 17; Section 3, p. 19): four possible end-states — (1) “only exposure assessment for adult occupational use”; (2) “only exposure assessment for children” (e.g., children’s clothing, toys, baby wipes, children’s toothpaste); (3) “only exposure assessment for adults (which covers child exposure as well, when relevant)”; (4) “exposure assessment both for children and adults”. Conservative default: when “Yes/No” is unknown at any decision node, “Yes” is chosen to trigger child-specific assessment.
• Child air intake ratio (Section 2 Decision Tree discussion, p. 16, citing Armstrong et al. 2002): air intake of a child <1 year is 2.3× the adult intake rate; passive child exposure scales accordingly if air concentration is the same.
• Case 1 — hair spray, integrated dose for toddler vs adult (Table 2, p. 21, µg/kg bw/day):
  • Non-VOC hair spray: toddler 0.88, adult 142, toddler/adult factor 0.01 (adult exposure ~160× higher; driver = dermal exposure during direct application).
  • VOC hair spray: toddler 22, adult 11, toddler/adult factor 2.00 (toddler ~2× higher; driver = inhalation; bolded in source table because the toddler exceeds the adult).
  • Modelled with ConsExpo 4.1 spray model (Delmaar & Bremmer 2009), US EPA residential SOPs (US EPA 2012), and ConsExpo cosmetics factsheet (Bremmer et al. 2006). 100 % bioavailability assumed for all routes; for VOCs 100 % evaporation assumed (no dermal contribution). Toddlers modelled as passively present during application and post-application.
• Case 2 — kitchen cleaner spray, integrated dose for toddler vs adult (Table 3, p. 22, µg/kg bw/day):
  • Toddler 161, adult 0.78, toddler/adult factor 206 (toddler ~200× higher; driver = dermal exposure from table-top contact post-application).
  • Modelled with ConsExpo 4.1 spray model + cleaning-products factsheet (Prud’homme de Lodder et al. 2006). Toddler exposure pathway = passive post-application hand-to-mouth + dermal table-top contact during crawling and exploration; adult exposure pathway = active spray application only.
• Case 3 — MCPA herbicide (2-methyl-4-chlorophenoxyacetic acid) on public-park lawn:
  • Application rate 1.8 kg active substance (AS) per hectare; dislodgeable fraction 5 %; dermal absorption 2.5 % (EC 2008); default oral absorption 100 % (refined values 55–78 % from EC 2008 noted as alternative).
  • Risk-assessment defaults (Table 4, p. 24): infant 10.5 months — body weight 8.69 kg, transfer coefficient 0.7 m²/hr, contact duration 1.5 hr/day; child 4 years — 15 kg, 1 m²/hr, 3 hr/day; adult lying in grass — 63 kg, 1.3 m²/hr, 3 hr/day. Source: Prud’homme de Lodder et al. 2006 RIVM secondary-exposure-to-lawn-pesticides methodology.
  • Internal-dose results (Table 5, p. 24): infant total 1.16 mg AS/day = 133 µg AS/kg bw/day (oral 0.95 mg/day = 109 µg/kg bw/day + dermal 0.21 mg/day = 24 µg/kg bw/day); child 4 years total 0.68 mg AS/day = 45 µg AS/kg bw/day (dermal only, no mouthing assumed at age 4); adult lying in grass total 0.88 mg AS/day = 14 µg AS/kg bw/day (dermal only).
  • Infant/adult internal-dose ratio: ~9.5× higher in infants; child/adult ratio ~3.2× higher in 4-year-olds. Even when refined oral absorption (55–78 %) is substituted for the default 100 %, infant internal dose remains ~6× adult.
  • Oral exposure for infants modelled as 10 % of dermal exposure (Prud’homme de Lodder et al. 2006 default) to account for hand-to-mouth and mouthing behaviour.
• Child-specific exposure factors checklist (Table 1, p. 18, by route):
  • Inhalation: higher breathing rate / surface area for absorption per kg bw (drives higher internal exposure when vapour pressure is high); crawling or lower body height results in a lower personal breathing zone if SVOC stratification occurs at lower air positions; longer time spent indoors at home/day-care/school; smoking in older children.
  • Oral: higher intake per kg bw; oral exploration and mouthing → leaching from product → additional oral route absent in adults.
  • Dermal: crawling, larger surface area per kg bw and relatively bigger head → migration of chemicals from textile/clothing/flooring; lying on grass/ground, longer sleep time in bed, high use of cosmetics and personal-care products in older children → migration from mattresses, grass, fabrics.
• Section 1.2 lead remark (p. 11, illustrative only): lead is named as an example of a chemical “strongly absorbed by soil [that] can be ingested by children during child-specific hand-to-mouth behaviour, while adult exposure via this route is considered less relevant” (citing Landrigan and Garg 2002). Lead exposure via food and environment is treated as background exposure and is therefore not the subject of any decision-tree application or case study in this document. No quantitative lead values are reported.
• Products singled out (Section 2.4, p. 15) as triggering child-specific assessment even when not specifically marketed for children: toothpaste, sunscreen, body lotions (direct exposure); paint, cleaning agents, building materials (indirect exposure). Listed as illustrative regulatory practice, not case-studied.
• Substances flagged (Discussion, p. 27) as examples where child-specific exposure assessment has been performed in scientific literature: polybrominated diphenyl ethers (PBDEs) via mouthing toys (Ionas et al. 2016); Bisphenol A (Healy et al. 2015); diethylhexyl phthalate (DEHP) (Ginsberg et al. 2016); parabens (Gosens et al. 2011, 2014). No heavy-metal substance is named as a worked-example case in the document.

Methods (brief)

OECD WPEA-led methodology and decision-tree development. The decision tree (Figure 1, p. 17) was first drafted at a 2013 Utrecht workshop on children’s exposure to chemicals (ENV/JM/MONO(2014)29; OECD 2014), revised at the 6th TFEA meeting in Tokyo (2014), and finalised by a Netherlands-led sub-group with members from Australia, Canada, Denmark, France, Germany, Italy, Japan, Sweden, the US, and BIAC. Case studies modelled with: ConsExpo 4.1 spray model (Delmaar & Bremmer 2009); ConsExpo 4.1 cosmetics factsheet (Bremmer et al. 2006); ConsExpo 4.1 cleaning-products factsheet (Prud’homme de Lodder et al. 2006); US EPA residential Standard Operating Procedures (US EPA 2012); RIVM secondary-exposure-to-lawn-pesticides methodology (Prud’homme de Lodder et al. 2006). Reference defaults drawn from US EPA Exposure Factors Handbook 2011 (US EPA 2011) and Child-Specific Exposure Factors Handbook 2008 (US EPA 2008), HEEG default human factor values for biocidal products (HEEG 2013), and EFSA dermal-absorption guidance (EFSA 2017). Conservative assumptions across all three cases: 100 % bioavailability for dermal/oral/inhalation routes; 100 % evaporation for VOCs (no dermal route); well-mixed-box model for inhalation during application; no residue depletion during exposure duration. Limitations stated explicitly: default values for behavioural parameters (mouthing duration and frequency, crawling time) are highly variable and substance- and product-specific; the decision tree is a first-tier awareness-raising tool, not a fixed protocol; toxicokinetic and toxicodynamic differences between adults and children are flagged as relevant but out of scope.

Implications

• Certification (HMTc): Decision-tree and child-specific physiological/behavioural defaults reference for any HMTc category covering products children can passively encounter (cleaning agents, household sprays, lawn-care chemicals, hair sprays used by adults in proximity to infants). The Q3/Q4 framing — direct vs indirect exposure — maps onto HMTc per-category exposure-pathway analyses. No threshold values are derivable from this document because no occurrence data on heavy metals are reported.
• Threshold-setting workbench: Not directly threshold-anchoring (no occurrence values). Supplies modelling-default precedents: infant/toddler/child body-weight defaults consistent with US EPA EFH; the 10 % dermal-to-oral conversion factor (Prud’homme de Lodder et al. 2006) and the 2.5 % MCPA dermal-absorption value (EC 2008) are substance-specific and not HMI-relevant. The body-weight defaults (infant 8.69 kg, child-4-year 15 kg, adult 63 kg) and contact-duration defaults (1.5 hr/day infant, 3 hr/day child and adult) are reusable for any consumer-product exposure model.
• Courses: Foundational regulatory-affairs and exposure-assessment training reference. The decision tree is a classroom-ready tool for teaching the child-specific exposure question. The three case studies demonstrate how the same product type can produce a child/adult exposure ratio ranging from 0.01× (non-VOC hair spray, adults higher) to 206× (kitchen cleaner, toddlers higher), depending on volatility and use pattern.
• App: Not relevant to ingredient contamination_profile. Provides per-age-group body-weight and exposure-route default-value seed for any future per-product child-exposure estimator the app may layer onto occurrence data.

Wiki pages this source may touch

• lead — only via Section 1.2 illustrative mention of lead’s soil-adsorption / hand-to-mouth pathway as background-exposure example; no occurrence values reported.

Verification notes

• 2026-06-01 fresh NEW-path ingest (Claude Opus 4.7, /ingest-next-manual-fetch-pdf v2.0 skill). PDF read in full: pp. 1–10 (front matter, Foreword, Executive Summary, Introduction §1.1), pp. 11–20 (§1.2–1.5 sources/definition/physiological/behavioural; §2 decision-tree Q0–Q5; §3 case-study intro and Case 1 scenario/decision-tree), pp. 21–32 (Case 1 exposure-assessment Table 2; Case 2 kitchen cleaner Table 3; Case 3 MCPA Table 4 risk-assessment factors and Table 5 internal doses; §4 Discussion and Conclusion; References).
• DOI: none. OECD ENV/JM/MONO(2019)29 series documents are not DOI-minted; no_doi_assigned: true. Document identifier ENV/JM/MONO(2019)29 + Series No. 310 carried in publication: field.
• License: “OECD copyright” rather than CC BY 4.0. The 2019 copyright page (p. 5) reads ”© OECD 2019. Applications for permission to reproduce or translate all or part of this material should be made to: Head of Publications Service, RIGHTS@oecd.org”. No Creative Commons designation is present in this 2019 document (in contrast to the OECD 2024 Crafts & Toys document oecd2024-crafts-toys-children-exposure, which carries CC BY 4.0). Conservative attribution chosen.
• metals: [], ingredients: [], products: [], matrices: [] retained. The document is exposure-assessment methodology guidance, not an occurrence study; its three case studies (hair spray, kitchen cleaner, MCPA herbicide) are illustrative modelled exposures and do not correspond to HMI heavy-metal product slugs. Section 1.2 explicitly states that exposure via food and environment is treated as background exposure and is out of scope. Section 2.4 lists toothpaste, sunscreen, body lotions, paint, cleaning agents, and building materials as products triggering child-specific assessment under various national regulatory frameworks, but these are illustrative regulatory practice — not case-studied here — and assigning routing on illustrative mentions would over-attribute the document.
• jurisdictions: [international] (precedent: oecd2024-crafts-toys-children-exposure uses the same code for an OECD intergovernmental publication endorsed by member countries across multiple continents).
• Brand-firewall clean. The document cites: agencies (RIVM, US EPA, EFSA, ECHA, SCCS, WHO, EC, OECD itself), tools (ConsExpo 4.1, ConsExpo Web, E-FAST), test standards (EN 71-3 not mentioned in this document; the only EN/CEN reference is CEN 2015 on mouthing behaviour in contact with toys, CEN/TR 16918:2015), and academic references. No commercial brand names are attributed to contamination values anywhere in the document body. The Case 1 hair-spray and Case 2 kitchen-cleaner case studies are modelled exposures using generic product factsheets, not branded products.
• HMTc-firewall clean. The document does not advocate, propose, or harmonise any HMTc threshold; it presents a child-specific exposure decision tree at a methodology level. No claims softened or strengthened in this page beyond the source’s own framing. The lead mention in Section 1.2 is preserved as the source presents it (background-exposure illustrative example), not promoted to a synthesis claim.
• Three illustrative case studies (hair spray, kitchen cleaner, MCPA herbicide on lawn) modelled the OECD decision tree on consumer products that do not map to HMI product slugs. The case studies are summarised under Key numbers for methodology-documentation purposes; they do not constitute occurrence evidence and do not feed any ingredient or product contamination_profile.
• Wiki pages this source may touch limited to lead based on the single illustrative mention in Section 1.2. The mention is qualitative (“lead is strongly absorbed by soil and can be ingested by children during child-specific hand-to-mouth behaviour”), not quantitative; routing is informational, not contamination-data-bearing.
• 2026-06-01 audit subagent (fresh-context Agent, general-purpose) verdict PROMOTE. One ⚠️ concern in Check 1: child-air-intake bullet was attributed to “Section 3, Discussion p. 16,” but the passage sits in Section 2 (Decision Tree) prose on p. 16, not Section 3 (Case Studies, begins p. 19) or Section 4 (Discussion and Conclusion, begins p. 25). Verified independently against source pp. 14–20; finding correct. Corrected attribution to “Section 2 Decision Tree discussion, p. 16”. One ⚠️ concern in Check 2 noted that international is not strictly in the taxonomy snapshot’s jurisdiction vocabulary; precedent established by sibling oecd2024-crafts-toys-children-exposure; left as-is and surfaced for next taxonomy-snapshot refresh. 0 ❌ definite-error findings.

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.