Gosens et al. 2014 — Aggregate exposure to parabens in personal-care products for children 0-3 yr
This Journal of Exposure Science and Environmental Epidemiology paper compares deterministic (tier 1) and person-oriented probabilistic (tier 2) approaches for aggregating exposure to four parabens (methyl-, ethyl-, propyl-, butylparaben) in young children (0-3 yr) from twelve personal-care product types. The paper is a methodology paper on aggregate exposure assessment; it does not measure heavy metals. It is included in the Heavy Metal Index batch because the dermal-absorption fractions, product-use-frequency inputs, and per-product retention factors are useful infrastructure for HMI’s parallel heavy-metal exposure modeling in the same product categories.
Key numbers
- Modeled product set: 12 personal-care product types — toothpaste, shampoo, two-in-one shampoo, hair lotion, shower/bath soap, body lotion, bath oil, baby wipes, baby salve, sunscreen, aftersun, liquid soap (SI1 Table 1; categories named in Methods, p. 209).
- Body-weight default used in modeling: 11.1 kg (representative 1.5-year-old child) (p. 209).
- Tier 1 deterministic exposure outputs (Table 1, p. 209, mg/kg bw/day external; internal calculated by oral or dermal absorption fraction): methylparaben external 2.32, dermal absorption fraction 36%, internal 1.01 mg/kg bw/day, NOAEL 1000, MoE 991; ethylparaben external 0.36, dermal 55%, internal 0.20, NOAEL 1000, MoE 4966; propylparaben external 1.05, dermal 37%, internal 0.41, NOAEL 3.3, MoE 8; butylparaben external 0.47, dermal 42%, internal 0.20, NOAEL 2, MoE 10.
- Tier 2 probabilistic outputs: percentage of modeled population with exposure probability above the assumed “safe” MoE of 100 was 13% for propylparaben and 7% for butylparaben (Abstract, Results).
- Drivers of butylparaben aggregate exposure (Figure 3d, p. 212): baby wipes the dominant contributor (~60% of total population exposure); body lotion, baby salve, and sunscreen each contribute ~10-20%; remaining product types contribute <10% individually.
- Retention factors used: 0.01 for rinse-off products (shampoo, two-in-one shampoo, liquid soap, shower/bathsoap, bath oil); 1.0 for leave-on products (sunscreen, aftersun, lotion, salve); 0.1 for hair lotion (assumed 1/10 stays on); 0.1 for baby wipes (Engelen et al. assumption that 0.5 ml deposits on skin per event; 2.5 g/day baby-wipe-derived substance on skin) (p. 209).
- ConsExpo default product-application amounts and frequencies sourced from Cosmetics Fact Sheet (Bremmer et al. 2006). Child-vs-adult body-surface scaling factors used to extrapolate adult application amounts to a 1.5-year-old: 0.27 for sunscreen/aftersun/body lotion/shower-bath soap (4800 cm² child body application area / 17,500 cm² adult); 0.66 for hair lotion, shampoo, and two-in-one shampoo (768 cm² child head SA / 1155 cm² adult head); 0.29 for liquid soap (247.2 cm² child hands at 1.5 yr / 857.5 cm² adult hands) (p. 209).
Methods (brief)
The paper applies two tiered exposure-assessment frameworks. Tier 1 (deterministic, worst-case): maximum reported paraben weight-fraction in each product × default product use amount × default frequency × retention factor × dermal absorption fraction / body weight. Tier 2 (probabilistic, person-oriented): pilot online survey of 28 Dutch families (21.5% response rate from parents of children 0-3 yr) provided real product use amount and frequency by product type and age/gender; the survey data populated an MS Access database, and exposures were resampled via bootstrap simulation. Survey limitations stated: small sample (28 children); broad-range reporting of use amounts and frequencies introduces uncertainty spanning two orders of magnitude; 14-month-old outlier driver of butylparaben distribution may be a survey-reporting artifact. Dermal absorption set to a distribution 1-55% (lowest unmetabolized methylparaben = 1%; highest ethylparaben from in-vitro human-skin = 55%). Oral absorption = 100%. Paraben concentrations: methyl-, ethyl-, propyl-, butylparaben weight fractions from manufacturer listings and Netherlands Food and Consumer Product Safety Authority (NVWA; the paper text on p. 209 prints “NWWA” but reference 8 credits NVWA) 2006 measurements in 12 product types. R modeling and Access database available from authors.
Implications
- Certification (HMTc): Not directly relevant to heavy-metal threshold work. The product-use-frequency defaults (g/day applied), age-scaling factors (0.27 body-surface ratio), retention factors per product class (rinse-off 0.01, leave-on 1.0, wipes 0.1), and child body-weight default (11.1 kg) are useful infrastructure for any future HMI dermal/oral aggregate-exposure model that adapts the same product-use survey to metals.
- Courses: Methodology teaching reference for tiered (deterministic vs probabilistic) aggregate exposure assessment; demonstrates how aggregate exposure can produce MoE values an order of magnitude tighter than single-source assessments, and how a few products (baby wipes) can dominate aggregate exposure for the highest-use children.
- App: Not relevant to ingredient contamination_profile. The product-use defaults for shampoo, body lotion, sunscreen, baby wipes, etc. could seed an “expected daily product use by age” annex for a future dermal-route exposure feature.
Wiki pages this source may touch
- baby-wipes
- baby-lotion-cream
- baby-shampoo-body-wash
- baby-sunscreen-mineral
- baby-sunscreen-chemical
- toothpaste
- children-personal-care
Verification notes
- This is a parabens (preservative) paper, not a heavy-metals paper. No metals analyte data are reported. Frontmatter
metals: []reflects this. Included in the batch as a methodological reference for aggregate dermal-exposure modeling of personal-care products. - The 0.27 body-surface scaling factor (1.5-yr child vs adult) and the per-product retention factors are reusable for heavy-metal exposure modeling on identical product categories.
- “baby-salve” is not in the current product taxonomy; closest existing slug is baby-lotion-cream or diaper-cream-zno / diaper-cream-non-zno depending on intended use. Flagging for Part 10 review.
- 2026-05-17 merge-enhance: corrected license to CC BY-NC-ND 3.0 (paper is open-access per the back-cover Creative Commons notice, not “All rights reserved”); fixed transposed body-surface-area attribution in the scaling-factor bullet (4800 cm² is the child application area, 17,500 cm² is the adult — the prior wording reversed them); replaced “plus a generic leave-on” placeholder with “two-in-one shampoo” (the actual 12th product type per the Methods description on p. 209).
- 2026-05-17 audit pass (fresh-context subagent, PROMOTE verdict with 2 ⚠️ minor concerns): narrowed butylparaben body-lotion contribution from “~15-25%” to “~10-20%” to match Figure 3d more accurately; corrected agency name in Methods from “NWWA” (a typo printed in the paper text on p. 209) to “NVWA” (Netherlands Food and Consumer Product Safety Authority, as credited in reference 8 of the paper), with an inline annotation preserving the source-text discrepancy.
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.