Meharg et al. 2008 - Rice arsenic literature review

Meharg and colleagues prepared this Food Standards Agency contract review to summarize total and inorganic arsenic in rice, rice products, and rice-containing infant foods. The report mixes published literature, FSA survey tables, and University of Aberdeen data, so this page treats it as B-tier review and agency-commissioned evidence rather than as one primary market survey. Total arsenic and inorganic arsenic are kept separate throughout; DMA(V) is reported in the source but is not collapsed into inorganic arsenic.

Key numbers

• Market rice total arsenic by origin: Table 3.1 reports mean total arsenic from 0.05 mg/kg in Egyptian rice (n=110) to 0.26 mg/kg in French rice (n=39) across the whole white/brown/red rice dataset. The same table reports USA rice n=203, mean 0.25 mg/kg, S.D. 0.1 mg/kg.
• White rice total arsenic by origin: Table 3.2 reports mean total arsenic from 0.05 mg/kg in Egyptian white rice (n=110) to 0.25 mg/kg in USA white rice (n=174). White rice means for India, Bangladesh, Thailand, China, Italy, Japan, Spain, and France are 0.08, 0.13, 0.14, 0.15, 0.16, 0.19, 0.19, and 0.19 mg/kg, respectively.
• UK-purchased rice: Table 3.3 reports UK-purchased USA-origin rice n=26, mean 0.25 mg/kg, min 0.09 mg/kg, max 0.4 mg/kg; UK-purchased India-origin rice n=10, mean 0.05 mg/kg, min 0.03 mg/kg, max 0.07 mg/kg; and UK-purchased Thailand-origin rice n=13, mean 0.16 mg/kg, min 0.02 mg/kg, max 0.39 mg/kg.
• Published rice surveys: Table 3.4 reports total arsenic ranges including USA rice 0.22 to 0.46 mg/kg (n=4, mean 0.30 mg/kg), Vietnam white rice 0.03 to 0.46 mg/kg (n=31, mean 0.21 mg/kg), Vietnam brown rice 0.08 to 0.70 mg/kg (n=25, mean 0.29 mg/kg), and Australian white rice 0.12 to 0.78 mg/kg (n=11, mean 0.26 mg/kg).
• Rice speciation: Table 3.7 reports total inorganic arsenic averages of 0.08 mg/kg for Bangladesh-origin rice, 0.14 mg/kg for China-origin rice, 0.03 mg/kg for UK-purchased India-origin rice, 0.12 mg/kg for UK-purchased Italy-origin rice, 0.16 mg/kg for Japan-purchased Japan-origin rice, and 0.09 mg/kg for UK-purchased USA-origin rice. The table reports inorganic arsenic percentages from 37% for UK-purchased USA-origin rice to 60% for Bangladesh-origin rice on the average rows.
• Rice drinks and other liquids: Table 3.8 reports rice vinegar total arsenic values from 0.006 to 0.042 mg/L, rice wine values from 0.001 to 0.011 mg/L, Japanese rice mirin 0.032 mg/L, and rice milk products from 0.012 to 0.033 mg/L.
• Baby rice unpublished table: Table 3.9 reports baby rice total arsenic values from 0.128 to 0.494 mg/kg, with group means of 0.197, 0.168, 0.360, and 0.245 mg/kg for the four anonymized baby-rice product groups. Manufacturer names from the table are omitted under the brand firewall.
• FSA baby-food survey total arsenic: Table 3.10 reports pure/organic baby rice dry products from 0.150 to 0.276 mg/kg, organic rice porridge 0.217 mg/kg, organic rice cakes orange 0.404 mg/kg, organic rice cakes 0.272 mg/kg, chocolate rice sandwiches 0.106 mg/kg, and organic rice cakes apple 0.220 mg/kg.
• FSA baby-food descriptive statistics: Table 3.11 reports total arsenic means of 0.200 mg/kg for 7 rice products, 0.250 mg/kg for 4 rice bars, 0.217 mg/kg for 1 rice porridge, 0.069 mg/kg for 1 rice + milk product, and 0.003 to 0.006 mg/kg for the two infant-formula groups.
• Weaning-food speciation: Table 3.12 reports 13 rice-labelled infant/weaning foods with average total arsenic 0.197 mg/kg, average inorganic arsenic 0.135 mg/kg, and average percent inorganic arsenic 63%. Individual rice-labelled rows span total arsenic 0.098 to 0.318 mg/kg and inorganic arsenic 0.067 to 0.230 mg/kg.
• Exposure modeling: Table 5.5 (from the 1999 Total Diet Study, FSA report 51/04, combined with NDNS rice consumption data) estimates inorganic arsenic intake from rice using concentrations from 0.03 mg/kg for Indian rice to 0.12 mg/kg for Italian rice purchased in the UK. At the 50th percentile: adults 0.009 - 0.035 µg/kg/d, young people (4–18y) 0.018 - 0.071 µg/kg/d, toddlers (1.5–4.5y) 0.030 - 0.090 µg/kg/d. At the 95th percentile: adults 0.030 - 0.119 µg/kg/d, young people 0.065 - 0.260 µg/kg/d, toddlers 0.089 - 0.358 µg/kg/d.

Methods (brief)

This is a literature review and agency-commissioned synthesis rather than a single new laboratory survey. The report tabulates published rice arsenic surveys, FSA reports FSIS 17/06 and Fd 06/12, and University of Aberdeen data on total arsenic, DMA(V), and total inorganic arsenic in rice and rice-containing foods. Speciation of the Aberdeen market-basket dataset (Tables 3.7, 3.12) was by HPLC-ICP-MS (per Figure 3.4 caption). The PDF metadata creation date is Thu Jun 5 18:22:52 2008 EEST, and no DOI is printed in the extracted report. The source does not provide one harmonized sample frame or one analytical method for all values, so routeable numbers should retain their table provenance and should not be treated as one pooled primary dataset.

Implications

Certification (HMTc): The report is useful B-tier context for rice-bulk-grain, rice milk, infant rice cereal, rice cakes/snacks, and rice-containing infant-food families, especially because it keeps inorganic arsenic separate from total arsenic for the speciation tables. Values transported from primary papers or FSA survey reports should be de-duplicated against those primary sources before pooled benchmarking.

Courses: The report is a compact example of why rice evidence requires species labels, product-form labels, and provenance: total arsenic, inorganic arsenic, DMA(V), dry infant cereal, rice milk, and exposure-model estimates cannot be swapped into one undifferentiated arsenic row.

App: The source can support UK rice-arsenic context, including category-level warnings for rice drinks, baby rice, rice cakes, and rice-containing infant foods, while avoiding manufacturer-level display.

Wiki pages this source may touch

• arsenic-total
• arsenic-inorganic
• rice
• rice-milk
• rice-cakes
• infant-rice-cereal
• baby-food
• rice-bulk-grain
• rice-beverages-rice-milk
• plant-milks-rice-based
• baby-cereals-dry-rice-based
• teething-and-snacks-rice-based
• rice-snacks-crackers
• mixed-meals-rice-containing
• infant-food-general

Verification notes

• PDF text was extracted with pdftotext -layout to /tmp/mfk_june8_553.txt; the title page, contents, Tables 3.1-3.4, Table 3.7, Table 3.8, Tables 3.9-3.12, Table 5.5, references, and PDF metadata were checked against this page.
• No DOI is printed in the extracted report. Title, contract number C101045, raw handle MFK_levels-of-arsenic-in-rice-literature-review, and candidate cite-key path wiki/sources/meharg2008-rice-arsenic-literature-review.md were checked before creation; no existing source page was found.
• Units are copied exactly as printed: rice and dry infant-food values use mg/kg, rice-drink values use mg/L, and exposure-model outputs use µg/kg/d. No unit conversions were performed.
• Speciation: Table 3.7 and Table 3.12 report total arsenic and total inorganic arsenic separately. DMA(V) values are retained as organic/speciation context in prose and are not treated as inorganic arsenic.
• Brand firewall: Table 3.8 and Table 3.9 include manufacturer/product labels. This page reports product-form/category values and omits manufacturer names.
• Frontmatter slugs were checked against docs/gpt-collaboration/taxonomy-snapshot.md. The taxonomy lacks a broad resolved product row for unspecified infant formula in this review, so the formula means in Table 3.11 remain in Key numbers but are not routed to a formula product row.
• Audit subagent (2026-06-10), Check 1, Table 3.9: flagged “four manufacturers” vs three; verified against PDF p.27 (“four different makes of rice were examined from three different manufacturers”) and Table 3.9 — wiki says “four anonymized baby-rice product groups” (accurate), not “four manufacturers”; finding was a false positive; no change made.
• Audit subagent (2026-06-10), Check 1, Table 5.5: flagged missing 50th-percentile toddler (0.030–0.090 µg/kg/d) and young-people (0.018–0.071 µg/kg/d) rows and absent provenance note; verified against PDF Table 5.5 (p.46) — rows confirmed present in source; all three missing rows and provenance (1999 TDS, FSA report 51/04 + NDNS) added to Key numbers.
• Audit subagent (2026-06-10), Check 3: flagged missing HPLC-ICP-MS attribution for Aberdeen speciation dataset; verified against PDF Figure 3.4 caption (p.24) — “by HPLC-ICP-ms” confirmed; one sentence added to Methods section.

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.