Allwood et al. (2024) — Recently Recalled Children’s Products Due to Lead Hazards

This is a five-page review article (with one figure and one table) published in Pediatrics as part of a 2024 supplement (Suppl 2), authored by CDC Lead Poisoning Prevention and Surveillance Branch staff (Paul Allwood PhD; Perri Zeitz Ruckart DrPH; Qaiyim Harris). The paper describes the implementation of a CDC automated tool that monitors the US Consumer Product Safety Commission (CPSC) Really Simple Syndication (RSS) recall feed for the keyword “lead” and uses Microsoft Power Automate workflows to filter, log, and disseminate lead-related children’s-product recall notices to public health partners. The authors then descriptively summarise the 30 lead-related children’s-product recalls captured by the tool during June 2022 through April 2024, including product type, country of manufacture, time-on-market before recall, recall type (voluntary vs not), remedy offered, distribution scope, and injuries reported. The paper reports no primary contamination measurements; its contribution to the HMI corpus is as a contemporaneous CDC-authored regulatory-landscape paper documenting CPSC lead-recall activity for children’s products in the two-year window immediately following the 2022 wave of cinnamon-applesauce and toddler-cup CPSIA actions.

Key numbers

• Study period and capture mechanism (Abstract; Methods, p. 2): CDC staff reviewed CPSC lead-related recall notices for children’s products issued between June 2022 and April 2024 (a 23-month window). Notices were extracted from the CPSC RSS feed by a CDC-built automated tool that uses Microsoft Power Automate workflows to filter, clean, log, and provide near-real-time data notifications for content containing the keyword “lead.”
• Total recalls identified (Abstract; Results, p. 3): 30 lead-related recalls of children’s products. Combined number of units sold across the 30 recalls: 914,598.
• Units-sold distribution (Results, p. 3): per-recall unit counts ranged from 100 to 346,000; median 2,700 units.
• Product-type distribution (Results, p. 3; Table 1, p. 8): eight product-type categories captured. Toys n=11 (37%), utensils n=8 (27%), trinkets n=5 (17%), clothing n=3 (10%), accessories n=1 (3%), shoes n=1 (3%), cosmetics n=1 (3%), furniture n=1 (3%).
• Country of manufacture (Results, p. 3; Table 1, p. 8): country of origin was reported in 28 of 30 (93%) recalls (missing for 2). Of those 28: China n=24 (86%), Egypt n=1 (4%, Table 1 percentage; text on p. 3 states 3%), Germany n=1 (4%), India n=1 (4%), Italy n=1 (4%). (The percentages in Table 1 sum to 102% across the five named countries; the manuscript text on p. 3 reports each non-China country as “n=1 each, 3%“. The base for the Table 1 percentages is the 28 recalls with country reported, while the text’s “3%” reflects rounding off the 30-recall denominator. Both percentages reproduce the paper’s own arithmetic; the discrepancy is internal to the source.)
• Time on market before recall (Results, p. 3; Fig. 1, p. 7): overall range 1 to 73 months; mean 25 months. Per-product-category means from Fig. 1: Accessory 38.0, Toy 31.5, Utensil 25.6, Clothing 22.3, Trinket 16.2, Furniture 5.0, Shoes 1.0 months. (Fig. 1 omits Cosmetics from the per-category bars; the cosmetics recall is one of the 30 reported in the body text and Table 1.)
• Voluntary vs non-voluntary recalls (Results, p. 3): 23 of 30 (77%) recalls were indicated as voluntary. Of the 7 non-voluntary recalls, 2 were toys and 1 each were a trinket, accessory, shoes, utensil, and furniture.
• Remedy offered (Results, p. 3): 25 (83%) companies provided a refund; 4 (13%) provided replacement; 1 (3%) provided repair.
• Distribution scope (Results, p. 3): 28 of 30 (93%) recalled products were reported as marketed and sold nationwide. The manuscript does not give a percentage for the two non-nationwide products.
• Hazard mechanism reported (Results, p. 3): 1 of 30 recalls — described in the paper as “One (6%) product” — was associated with “a part breaking off and exposing lead solder.” The paper’s “6%” figure does not match 1/30 (3.3%); the source does not state the denominator the 6% reflects (1/17 non-toy recalls = 5.9% is the nearest integer-denominator match).
• Injuries (Abstract; Results, p. 3): no injuries or other incidents were reported across any of the 30 recalls.
• CPSIA Section 101 statutory limits cited as the regulatory backbone (Introduction, p. 2): total Pb must not exceed 100 ppm in accessible substrate parts of a children’s product, and Pb concentration in paint or surface coatings must not exceed 0.009% (by weight of dried paint film; equivalent to 90 ppm). Children’s-product manufacturers must obtain a Children’s Product Certificate demonstrating compliance-based testing from a CPSC-approved laboratory. The CPSIA defines a “children’s product” as one designed or intended primarily for children ≤12 years of age.
• Historical comparison context cited from prior literature (Introduction, p. 2): during 2001–2020, the cited Wagner & Johnson recall review reports 317 lead-related recalls of children’s products in the US (generally <20 per year, with a 2006–2010 spike). The 2008 CPSIA enactment is framed as the regulatory response to that 2006–2010 spike. The 2006 death of one child after ingesting a metallic charm containing 99.1% Pb by weight is cited from MMWR 2006;55(12):340–341. Of 103 lead-related recalls in children’s toys from 2007–2008, 30 (29%) were of jewelry.
• Discussion: prior recall-volume trend (Discussion, p. 4): a cited 2020 review (Niven et al. Inj Prev) of US and Australian children’s-product recalls from 2011–2017 found a 27% overall decline in the number of children’s products recalled for any reason in the US and a 41% decline in toys-and-games recalls. Of 176 children’s toys/games recalls in the US in that earlier window, 16 (9%) were caused by non-compliance with mandatory safety requirements, with the majority exceeding the federal lead paint standard.
• HACCP analogy proposed (Discussion, p. 4): the paper proposes the seven principles of the Food and Drug Administration’s Hazard Analysis and Critical Control Point (HACCP) framework — hazard analysis, critical control point identification, establishing critical limits, monitoring procedures, corrective actions, verification procedures, and record-keeping and documentation — as a potential model for a similar supply-chain quality-control system in children’s-product manufacturing to reduce Pb incorporation at source.

Methods (brief)

Surveillance review of secondary regulatory data. The CDC Childhood Lead Poisoning Prevention Program implemented an in-house automated tool that subscribes to the CPSC RSS recall feed and filters posts containing the keyword “lead.” The tool is built on Microsoft Power Automate workflows that clean, log, and provide near-real-time notifications of matching content. The authors manually reviewed each lead-related children’s-product recall captured by the tool during June 2022 through April 2024. For each recall, the following fields were extracted from the CPSC recall notice: recall source, product category (toys, clothing, furniture, food, skin care, other), source country, dates sold and date of recall, number of units sold, voluntary vs non-voluntary classification, remedy offered (repair, replacement, refund), distribution scope (nationwide or limited), injuries reported, and stated recommendations to consumers. Descriptive statistics are presented (counts, percentages, range, mean, median); no inferential statistical testing was performed. The paper reports no primary Pb concentration measurements, no laboratory analytical methods, no XRF or ICP-MS measurements of recalled products, no children’s blood lead level (BLL) data, and no exposure modelling. Limitations acknowledged by the authors: (i) the automated tool may not have captured all lead-related recalls; (ii) products contaminated with Pb but not identified or recalled by the manufacturer are not captured; (iii) data were too sparse to stratify time-on-market by product type; (iv) absence of reported injuries does not exclude unrecorded adverse health effects, including elevated BLLs from exposure to the recalled products.

Implications

• Certification (HMTc): Provides a contemporaneous CDC-authored count of CPSC lead-recall actions for children’s products in the 23-month window June 2022 – April 2024. Total recall volume (914,598 units across 30 recalls; median 2,700 units per recall) and the dominant origin pattern (86% of recalls with country reported were manufactured in China) document the regulatory-landscape denominator for any children’s-product certification work covering this period. The category distribution (toys 37%, utensils 27%, trinkets 17%, clothing 10%, plus accessories/shoes/cosmetics/furniture 3% each) identifies the eight CPSC-recognised children’s-product categories that produced Pb recalls in the 2022–2024 window; the corresponding HMTc product-category slugs in the current taxonomy that map to those CPSC categories are toys-painted, toys-substrate-materials, utensils-metal, infant-clothing, childrens-makeup, cribs-and-bassinets, high-chairs-and-booster-seats, car-seats, and the children-personal-care master. The single recall associated with “a part breaking off and exposing lead solder” parallels the solder-bead recall pattern documented in padoh2023-pahan-727-cpsc-lead-cup-recall for stainless-steel children’s drinkware (a separate product category from those summarised in Table 1, since the Pennsylvania PAHAN advisory cups would have entered the CPSC database under “utensils” if at all). The 25-month average time-on-market before recall, with per-category means as long as 38.0 months for accessories and 31.5 months for toys, documents the latency window in which Pb-contaminated children’s products are commercially available before CPSIA Section 101 enforcement removes them.
• Courses: Useful overview for the children’s-product Pb supply-chain module. The paper is well-suited for teaching the difference between CPSIA Section 101 (the 100 ppm substrate limit) and 16 CFR Part 1303 (the 0.009% paint limit), the 12-years-of-age “children’s product” definition, the Children’s Product Certificate compliance regime, and the recall mechanism (voluntary vs non-voluntary; refund, replacement, repair remedies). Pairs naturally with hatlelid2009-cpsc-cpsia-lead-childrens-products (the 2009 CPSC staff presentation that summarises the CPSIA Section 101 statutory scheme in detail) for a regulatory-history-to-current-enforcement teaching arc, and with padoh2023-pahan-727-cpsc-lead-cup-recall for the contemporaneous state-public-health relay layer that disseminates CPSC recall information through public-health channels. The HACCP analogy proposed by the authors (transposing the FDA’s seven-principle food-safety quality-control framework to children’s-product manufacturing) is teachable as a supply-chain-engineering proposal but is a discussion-section recommendation, not an established programme.
• App: Not directly relevant to ingredient contamination_profile data because the paper reports no food-matrix occurrence values and no per-product Pb concentrations. Relevant to a future children’s-feeding-accessory or children’s-product Pb-recall surfacing app surface that flags whether a product has been the subject of a CPSC Pb recall; the CDC RSS-feed-monitoring tool described in this paper is the upstream data source such an app would need to subscribe to. The paper does not provide the per-recall hazard mechanism in sufficient detail to support a product-attribute risk-prediction model; for that level of detail, the underlying CPSC recall notices (referenced but not reproduced in the manuscript) would be the appropriate primary source.

Wiki pages this source may touch

• lead
• toys-painted
• toys-substrate-materials
• utensils-metal
• infant-clothing
• childrens-makeup

Verification notes

• Source identification. Paul Allwood, Perri Zeitz Ruckart, Qaiyim Harris. “Recently Recalled Children’s Products Due to Lead Hazards.” Pediatrics. 2024 October 01; 154(Suppl 2). DOI 10.1542/peds.2024-067808P. HHS Public Access author manuscript, available in PMC 2025 March 11. All three authors are CDC employees (Lead Poisoning Prevention and Surveillance Branch, Division of Environmental Health Science and Practice, National Center for Environmental Health, Atlanta, Georgia). Corresponding author Perri Zeitz Ruckart (afp4@cdc.gov). Conflict-of-interest disclosure: all authors are CDC employees; the findings and conclusions are those of the authors and do not necessarily represent the official position of the CDC.

• Source-tier rationale. evidence_tier: A. The paper is a CDC government-agency authored review published in Pediatrics (American Academy of Pediatrics peer-reviewed journal). It documents federal regulatory enforcement data captured from the CPSC RSS feed, which is itself an A-tier primary regulatory source. Both the authoring agency (CDC) and the underlying data source (CPSC recall feed) are A-tier per Part 13 (“government reports (FDA, EFSA, EPA, WHO, Codex)”).

• Source-type rationale. source_type: regulatory. The paper is a descriptive review of US federal regulatory enforcement data (CPSC lead-related recall notices) authored by CDC staff. The regulatory source-type captures both the agency-authored character of the paper and the regulatory-enforcement-event subject matter, consistent with the handling of padoh2023-pahan-727-cpsc-lead-cup-recall (a state public-health relay of federal CPSC recall actions) and hatlelid2009-cpsc-cpsia-lead-childrens-products (a CPSC staff presentation on the CPSIA Section 101 regulatory scheme). An alternative source_type: agency-review would distinguish CDC’s descriptive review character from the underlying regulatory enforcement events, but agency-review is not currently a recognised value in the schema; regulatory is the closest available match and preserves the regulatory-event character of the substantive content.

• Frontmatter products: field. Five product slugs from the current taxonomy (2026-05-18 snapshot): toys-painted (the CPSIA 16 CFR Part 1303 paint-limit regime is one of the two regulatory hooks; toys are 37% of recalls); toys-substrate-materials (the CPSIA Section 101 100 ppm substrate limit is the other regulatory hook; substrate Pb is the dominant recall mechanism for toys); utensils-metal (utensils 27% of recalls; children’s eating utensils with metallic Pb in solder or alloy are the dominant utensil-recall pathway); infant-clothing (clothing 10% of recalls; children’s clothing with Pb in paint, applique, zipper-pull, or button hardware); childrens-makeup (cosmetics 3% of recalls, n=1; children’s makeup with Pb in pigment). Slugs not included although the source’s eight categories include them: “trinkets” (children’s jewelry; the corpus has no childrens-jewelry slug in the 2026-05-18 snapshot — sekerak2016 and nelson2023 use [[products/childrens-makeup]] for jewelry-adjacent routing or none at all; surfacing for Karen’s review as a possible new-page proposal), “accessories” (no clear children’s-accessories slug), “shoes” (no children’s-shoes slug), “furniture” (no children’s-furniture slug — cribs-and-bassinets, high-chairs-and-booster-seats, car-seats, and toy-chests exist as specific durable-goods slugs but the source’s “furniture” category is described only at the genus level, with n=1 recall). The five chosen slugs cover 87% (26/30) of the recalls in Table 1 by category mass.

• Frontmatter ingredients: [] is correct — the source is a regulatory-landscape paper about non-food children’s products; no food ingredients are involved.

• Frontmatter matrices: [] is correct — no measurements are reported in any food matrix. The source describes recall metadata (product category, country of manufacture, units sold, time on market, voluntary status, remedy) for non-food articles; there are no occurrence-in-food values.

• Frontmatter metals: [Pb] is correct — the CDC automated tool filters the CPSC RSS feed specifically for the keyword “lead”; all 30 recalls captured are Pb-related by construction of the surveillance method. No other metals are discussed in the paper. Speciation question is not applicable: CPSIA Section 101 and 16 CFR Part 1303 both regulate total Pb in substrate and paint respectively, and the metallic-charm exemplar (99.1% Pb by weight) and lead-solder hazard (1 recall) are both metallic/elemental Pb pathways. The Pb in pigments cited in the cosmetics recall and in the paint-coating recalls would in principle be Pb(II) salts; the paper does not speciate.

• Brand-firewall (Part 12). No brand names appear in the wiki page body. The source itself describes the 30 recalls in aggregate (counts, percentages, distributions) without naming the manufacturers, importers, or distributors of any specific recalled product. The only brand-adjacent identifier in the source is the cited 2006 metallic-charm death (MMWR 2006;55(12):340–341), which is a Reebok-branded shoe-charm but the source describes it only as “a metallic charm that contained 99.1% lead by weight,” consistent with Part 12 product-form-descriptor framing. No Part 12 issue. (The source is methodologically positioned to avoid brand naming because its unit of analysis is the recall action, not the recalled product; the manufacturer name appears only in the underlying CPSC recall notices that are referenced but not reproduced in the manuscript.)

• Wiki/HMTc firewall (Part 2). The source proposes no HMTc thresholds and is not compared to existing HMTc certification levels in the body. The Implications section describes how the paper’s findings size the regulatory-landscape denominator for HMTc children’s-product certification programmes and identifies the eight CPSC-recognised children’s-product categories that have produced Pb recalls; this is procedural pointer-language to the certification-coverage question, not an HMTc threshold proposal. The HACCP-analogy discussion-section recommendation by the authors is preserved in the Key numbers and Implications sections as a paper-content claim by the authors, not as a wiki-side endorsement.

• Speciation flag. Not applicable — CPSIA Section 101 regulates total Pb in the accessible substrate (100 ppm); 16 CFR Part 1303 regulates total Pb by weight of dried paint film (0.009% ≈ 90 ppm). Metallic charm exemplar is metallic Pb (99.1% by weight in a Pb alloy). Solder-bead exemplar (1 recall) is metallic Pb in a Pb-bearing solder alloy. No inorganic-vs-organic Pb speciation question arises.

• Internal arithmetic and date discrepancies in the source. Three are noted:

  • The Table 1 country-of-origin percentages (China 86%, Egypt 4%, Germany 4%, India 4%, Italy 4%; sum 102%) use the 28-recall denominator (recalls with country reported), while the manuscript text on p. 3 reports each non-China country as “3% (n=1 each)” using an unstated denominator. Both reproduce the paper’s own arithmetic.
  • The hazard-mechanism statement “1 (6%) product was associated with a part breaking off and exposing lead solder” uses an unstated denominator; 1/30 = 3.3% and 1/17 (the seventeen non-toy recalls) = 5.9%, so the 6% figure may reflect 1/17.
  • Table 1 title date discrepancy. Table 1 on p. 8 is captioned “Summary of Findings of Lead-Related Recalls of Children’s Products, June 2022 Through May 2023.” The Abstract, Methods (p. 2), and Results body text (p. 3) consistently state the study window as June 2022 through April 2024 (a 23-month window). The Table 1 caption appears to be a residual draft date from an earlier cut of the analysis; the body-text window is consistent with itself across three independent statements and is reproduced as authoritative in this source page. Flagged for transparency rather than silently corrected.

  The page reproduces the paper’s own arithmetic and date framings as published and flags the discrepancies here rather than silently correcting them.

• CPSIA-Section-101 statutory citation. The paper cites the CPSIA Section 101 100 ppm Pb substrate limit and the 16 CFR Part 1303 0.009% Pb paint limit but does not give the public-law number (110-314) or the 15 USC 1278a codification. Cross-reference: the more complete statutory citation chain is in hatlelid2009-cpsc-cpsia-lead-childrens-products (Hatlelid 2009 CPSC staff presentation), which cites Public Law 110-314, the 2009 → 2011 phased reduction from 300 ppm to 100 ppm, and the 16 CFR 1500.87, 1500.88, 1500.89, 1500.90, and 1500.91 implementing rules. The CPSIA-Section-101 regulation page is not yet in the wiki/regulations/ taxonomy (surfaced as a new-page proposal in the padoh2023-pahan-727-cpsc-lead-cup-recall source-page verification notes; this source’s regulatory-context routing reinforces that proposal but does not create the page).

• License. public-reference-only: the manuscript is an NIH/HHS Public Access author manuscript (CDC-authored work; US federal-government works are in the public domain under 17 USC 105 in the country of origin but the journal-typeset version is subject to the Pediatrics publisher’s copyright). The HHS Public Access author manuscript version is the version archived in PMC. Quotation for review and synthesis purposes is fair use; re-hosting the publisher-typeset PDF requires permission from the American Academy of Pediatrics. The public-reference-only value matches the handling of other CDC- and US-federal-agency-authored sources in the corpus (e.g., the MMWR and CPSC-staff sources).

• Related corpus pages. hatlelid2009-cpsc-cpsia-lead-childrens-products is the closest existing corpus page (CPSIA Section 101 regulatory-scheme reference; same regulatory backbone). padoh2023-pahan-727-cpsc-lead-cup-recall is the closest contemporaneous corpus page documenting specific 2023 CPSC Pb recall actions in children’s drinkware (one of the recall categories — utensils — that this Allwood et al. paper aggregates over). pirg2010-trouble-in-toyland, sekerak2016-wa-childrens-jewelry, nelson2023-wa-childrens-jewelry-followup, and digangi1997-greenpeace-vinyl-children-products cover specific children’s-product Pb-survey work in the pre-CPSIA, immediate-post-CPSIA, and Washington-state-jewelry-followup contexts and are landscape-comparison companions to the CDC paper.

• Near-duplicates. None identified. The CDC paper is a unique CDC-authored review of a specific 23-month CPSC recall window; no other source in the corpus aggregates over the same window or uses the same CDC surveillance-tool methodology.

• Raw-file note. The PDF is the HHS Public Access author manuscript version (8 pages, 219 KB), filed in the manual-fetch tracker as MFK_08-unknown and located in the Kimi-extraction subfolder _extracted_infantdurable_03_Carriers_HighChairs_CarSeats/03_Carriers_HighChairs_CarSeats/08_Unknown.pdf. The folder-name “03_Carriers_HighChairs_CarSeats” reflects the Kimi-extraction batch structure (the PDF was part of a batch tagged for the durable-children’s-equipment manual-fetch sweep), not the topic of the paper itself; the paper is a broader review across eight children’s-product categories (toys, utensils, trinkets, clothing, accessories, shoes, cosmetics, furniture) and is not specifically about carriers, high chairs, or car seats. Carriers and car seats are not among the eight recall categories in Table 1; high chairs would fall under the “furniture” category (n=1) but are not separately identified in the source. The raw_handle: MFK_08-unknown is shared by other 08_Unknown.pdf files in adjacent Kimi-extracted subfolders (per data/evidence/autonomy/manual-fetch-tracker.csv), but the SHA256 8bd47c37951dfc460fb471a03ecbb1f7f151d9195790b2243a668d922406724b uniquely identifies this specific file.

• Audit subagent application (2026-06-01). Fresh-context audit subagent (general-purpose Opus 4.7) returned verdict REVISE with four findings, all verified against the source and applied: (1) Distribution-scope “6%” cross-attribution corrected — the source’s “6%” attaches to the lead-solder hazard sentence (p. 3), not to the two non-nationwide products; the Distribution-scope key-numbers bullet now states only the 28/30 (93%) nationwide figure, and the hazard-mechanism bullet retains the “6%” with the unstated-denominator caveat. (2) Table 1 title date discrepancy added to the source-internal-arithmetic verification note — Table 1 caption reads “June 2022 Through May 2023” while body text consistently states June 2022 through April 2024. (3) Certification-implications paragraph softened from prescriptive (“HMTc product-category coverage should at minimum extend to…”) to descriptive (the paper documents recalls in CPSC categories that map to these HMTc taxonomy slugs). (4) The “HMTc programmes operating on a faster pre-market certification cadence … provide the missing front-of-pipeline filter” sentence removed — it was wiki-side HMTc advocacy not present in the source. Zero subagent findings were false positives; zero findings rejected. Check 2 (slug vocabulary), Check 3 (speciation/methods), and Check 4 (Part 12 brand firewall) returned clean. Audit-application commit appended after this revision.

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.