CFS 2013 — First Hong Kong Total Diet Study: Metallic Contaminants

The Centre for Food Safety (CFS) of Hong Kong conducted the First Hong Kong Total Diet Study (1st HKTDS) to estimate dietary exposures of the Hong Kong general population and population subgroups to metallic contaminants and to assess associated health risks. This report (Report No. 5 of the TDS series, January 2013) covers aluminium (Al), antimony (Sb), cadmium (Cd), lead (Pb), methylmercury (MeHg), nickel (Ni), and tin (Sn); vanadium (V) was additionally reported because it was co-analysed in the same HR ICP-MS run. The dietary exposures to Al, Sb, Cd, Ni, and Sn were all below their respective health-based guidance values for average consumers, and average lead exposure (0.21 µg/kg bw/day) was below the JECFA 1.2 µg/kg bw/day level for systolic-blood-pressure risk in adults. About 11% of women of childbearing age (20-49 years) had methylmercury exposure exceeding the PTWI of 1.6 µg/kg bw protective of the developing foetus, which the report identifies as a public health concern. The general adult population was assessed as unlikely to experience major adverse health effects from these contaminants.

Inorganic arsenic from this same TDS series was reported separately in Report No. 2 (February 2012), ingested as cfs2012-hktds-inorganic-arsenic. Arsenic speciation data are not in the present report.

Key numbers

Study design. 1,800 samples of 150 TDS food items, three samples per item on each of four occasions (March 2010-February 2011), combined into 600 composite samples. Methylmercury was analysed in 51 food items (204 composites) selected as predominantly food of animal origin and seafood. Analysis by HR ICP-MS (Al, Sb, Cd, Pb, Ni, Sn, V) after concentrated HNO3 microwave digestion in Teflon high-pressure vessels; MeHg by enzymatic hydrolysis, HCl extraction, tetraphenylborate derivatisation, iso-octane extraction, and GC-ICP-MS with propylmercury as internal standard.

Limits of detection and quantification (general food; water and tea LODs are 5× lower):

Non-detects treated per WHO GEMS/Food-EURO guidance: ≤60% < LOD → medium-bound (½ LOD); 60-80% < LOD → lower and upper bounds presented separately. Dietary exposure estimated via the EASY (Exposure Assessment System) tool; mean and 95th percentile represent average and high consumers respectively.

Aluminium contents (mg/kg) by food group (Table 3.1):

Highest food-item means (per Appendix I, Table A): deep-fried dough 250 mg/kg (range 50-450), steamed barbecued pork bun 170 mg/kg, oyster 62 mg/kg, Chinese spinach 35 mg/kg (range 25-45 per Appendix I Table A; chapter prose §3.8 abbreviates this to “range 25-35”), ear fungus 23 mg/kg (range 14-36), cake 19 mg/kg (range 2.0-51), fermented bean products 18 mg/kg (range 3.3-31). The high deep-fried-dough and steamed-bun values are attributed to aluminium-containing raising agents. Non-alcoholic beverages contributed 33% of total dietary aluminium exposure despite moderate concentrations (milk tea 11 mg/kg, Chinese tea and malt drinks 3.2 mg/kg each), driven by high consumption (1,625 g/person/day in the FCS).

Cadmium contents (µg/kg) by food group (Table 5.1):

Highest food-item means: oyster 1,300 µg/kg, scallop 730 µg/kg, crab 540 µg/kg. Vegetables, fish/seafood, and cereals were the top three food-group contributors to dietary Cd at 36%, 26%, and 21% respectively.

Lead contents (µg/kg) by food group (Table 6.1):

Highest food-item means: oyster 230 µg/kg, ear fungus 100 µg/kg, watercress 96 µg/kg. Vegetables contributed 30% of dietary Pb exposure; non-alcoholic beverages 16%, mixed dishes 14%, cereals 13%.

Methylmercury contents (µg/kg, as Hg) by food group (Table 7.1; 51 food items, 204 composites):

All MeHg concentrations in fish were below the Codex guideline levels (1,000 µg/kg for large predatory fish; 500 µg/kg for other fish). Highest food-item mean: tuna 330 µg/kg. In a 2007 CFS Mercury in Fish study (referenced in §7.6), only 3 of 280 samples of imported alfonsino were detected with methylmercury above 500 µg/kg.

Antimony, nickel, tin, and vanadium contents (µg/kg unless noted) by food group (Tables 4.1, 8.1, 9.1, 10.1):

Nickel food-item highs: peanut 5,300 µg/kg, peanut butter 3,800 µg/kg, chocolate 1,400 µg/kg, fermented bean products 890 µg/kg. Tin food-item highs (driven by can coatings): pineapple 37 mg/kg, mushroom 32 mg/kg, peach 31 mg/kg. Vanadium individual high: oyster 190 µg/kg.

Dietary exposure estimates and reference benchmarks:

For aluminium, the high consumer of male aged 70-84 slightly exceeded the PTWI (2.1 mg/kg bw/week). For cadmium, the high consumer of male aged 30-39 slightly exceeded the PTMI (25 µg/kg bw/month). For lead, the report notes MOEs of 6 (average) and 3 (high) against the JECFA 1.2 µg/kg bw/day systolic-BP reference; MOE >1 indicates acceptably low adverse-effect risk. JECFA has not established a safety reference dose for vanadium.

Methods (brief)

Total Diet Study methodology: 150 food items selected from the 2005-2007 Hong Kong Population-based Food Consumption Survey, purchased as consumed, prepared as normally eaten, homogenised into composites (3 purchases per item × 4 occasions = 12 individual samples reduced to 4 composites per item, 600 composites total). All 600 composites analysed for Al, Sb, Cd, Pb, Ni, Sn, and V by HR ICP-MS after microwave-assisted concentrated-nitric-acid digestion in Teflon high-pressure closed vessels. MeHg analysed in 204 composites covering 51 food items predominantly of animal origin and seafood: composites washed with acetone and toluene, hydrolysed enzymatically with pancreatin, extracted with 50% v/v HCl, buffered to pH 4.1-5.0, derivatised using sodium tetraphenylborate, the derivative extracted by iso-octane and quantified by GC-ICP-MS with propylmercury as internal standard. MeHg quantified as mercury; a 1.075 molar-mass conversion factor (MeHg/Hg) was applied for the dietary-exposure step so the estimate is on a MeHg basis.

Sub-LOD treatment: when ≤60% of results were below LOD, the medium-bound estimate (½ LOD assigned to non-detects) was used for the overall population. When 60-80% of results were below LOD, both lower-bound (0) and upper-bound (LOD) exposure estimates were reported. Dietary exposure was estimated via the in-house web-based EASY system, mapping analytical results onto consumption data; mean and 95th percentile represent average and high consumers respectively. The cumulative consumption universe was the adult population aged 20-84 stratified by age and sex.

Implications

The 1st HKTDS provides validated population-level dietary exposure benchmarks for an East Asian high-seafood-consuming jurisdiction across seven metallic contaminants plus co-analysed vanadium. Cadmium at 33% of PTMI for average adult consumers and 75% for high consumers, with vegetables (36%), fish/seafood (26%), and cereals (21%) as top contributors, supplies a baseline for any product category whose ingredients draw from these groups. Oyster, scallop, and crab cadmium concentrations (1,300, 730, 540 µg/kg) are useful anchors for shellfish-derived ingredients. The MeHg distribution — 11% of women aged 20-49 exceeding the more protective PTWI of 1.6 µg/kg bw with high-consumer exposures of 2.1-2.5 µg/kg bw/week — identifies a specific vulnerable-population exceedance that is policy-relevant in any jurisdiction with comparable fish-consumption patterns.

For courses: a worked example of the Total Diet Study methodology, including the differential treatment of sub-LOD data (medium-bound vs lower/upper-bound), the molar-mass conversion between MeHg-as-Hg and MeHg-as-MeHg, and the use of population-based food-consumption surveys to convert per-food-item concentrations into per-person dietary exposures.

For the app: this source supplies aggregate dietary-exposure data and per-food-group concentration ranges, not per-ingredient ppb concentrations suitable for the contamination_profile per-cell format. The TDS food-item level (e.g., oyster, deep-fried dough, peanut, milk tea) maps to ingredient-level estimates with caveats about cooking/preparation as consumed.

Verification notes

Merge-enhance pass 2026-05-20 from the original 2026-05-13 ingest:

• Added vanadium (V) to metals and to Key numbers; the original page omitted V despite the source explicitly reporting Table 10.1 and noting V was co-analysed in the same HR ICP-MS run.
• Removed invented product slug products/mixed-dishes (not in the current product taxonomy); removed products/fruit-juices-non-apple because the source’s beverages-non-alcoholic group does not isolate fruit juices and naming this specific product over-routes the source.
• Corrected ingredients/dairy to ingredients/milk-and-dairy (matches the current ingredients taxonomy slug); added ingredients/eggs because the eggs food group is a distinct TDS analyte group with its own per-group means.
• Expanded Key numbers from a bulleted list to per-analyte tables for Al, Cd, Pb, MeHg, plus a combined Sb/Ni/Sn/V table, plus dietary-exposure benchmarks. The 2026-05-13 page had Al food-group means in prose and omitted the other six analytes’ food-group tables.
• Added near_duplicates and related_sources cross-link to cfs2012-hktds-inorganic-arsenic (Report No. 2 of the same TDS series, ingested separately).
• Numerical fidelity rechecked: all PTWI/PTMI/TDI percentages, LOD/LOQ values, food-group means and ranges, top food-item means (oyster 1,300 µg/kg Cd; deep-fried dough 250 mg/kg Al; tuna 330 µg/kg MeHg; peanut 5,300 µg/kg Ni; oyster 190 µg/kg V), and the 11%-of-childbearing-age-women-exceeding-1.6-PTWI exposure figure verified against Tables 3.1, 4.1, 5.1, 6.1, 7.1, 8.1, 9.1, 10.1, and the Executive Summary paragraphs 4-7.
• Brand-firewall scan: the source contains no brand-name product attribution; only generic food-item descriptors (“deep-fried dough”, “steamed barbecued pork bun”, “oyster”, “ear fungus”). No Part 12 violations to strip.
• Audit subagent (2026-05-20) flagged three ⚠️ items under Check 1 about item-level Al ranges (Chinese spinach, deep-fried dough, ear fungus/cake/fermented bean products) not visible in §3.7-3.8 chapter prose. Independently verified: all three sets of values are present in Appendix I Table A of the source PDF, which the auditor did not read. A genuine source-internal discrepancy was surfaced for Chinese spinach: §3.8 chapter prose says “range 25-35 mg/kg” while Appendix I Table A row says “mean 35, range 25-45 mg/kg”. Resolution: the page now cites both numbers and tags item-level ranges as sourced from Appendix I Table A explicitly. No values changed; only sourcing made explicit. Audit verdict: PROMOTE (with minor wording revision applied).
• 2026-06-02 byte-identical filesystem-copy enhancement: added duplicate_filesystem_copies block recording the second filesystem location of the canonical 1st HKTDS Metallic Contaminants report PDF at raw/manual-fetch/Kimi_Agent_Download Corruption Issue/condiments2_papers/05_Snacks_Canned_Prepared/HongKong_CFS_1st_HKTDS_Metallic_Contaminants.pdf (sha256 f87da014f81ceae5a964f8df6b5b7274bb2da1c805a1f7466cc27359b51d1f6c, byte-identical to the canonical raw_path under raw/Manual Fetch Kimi /02_Vegetables_and_Vegetable_Products/). Kimi agent filed the same report PDF in two locations; both now resolve back to this canonical source page. No claim, value, slug, exposure number, key-numbers, or HMTc-firewall change. No new audit cycle spawned because no body or evidence-bearing frontmatter changed.
• 2026-06-03 byte-identical filesystem-copy enhancement: appended a third filesystem location to duplicate_filesystem_copies at raw/manual-fetch/Kimi_Agent_Download Corruption Issue/seafood_papers/02_Marine_Nonpredatory/The First Hong Kong Total Diet Study_ Metallic Contaminants.pdf (same sha256 f87da014f81ceae5a964f8df6b5b7274bb2da1c805a1f7466cc27359b51d1f6c, verified byte-identical via shasum -a 256). The Kimi agent filed the same 1st HKTDS Metallic Contaminants report PDF a third time, this time in the seafood/marine-nonpredatory subfolder — likely because the report’s Chapter 7 (Methylmercury) and Chapter 5 (Cadmium) include fish/seafood food-group tables (oyster Cd 1,300 µg/kg, tuna MeHg 330 µg/kg, etc.). All three filesystem locations now resolve to this canonical source page so future manual-fetch identity-check cycles in any of the three Kimi-corruption subfolders skip re-ingest. No claim, value, slug, exposure number, key-numbers, or HMTc-firewall change. No new audit cycle spawned because no body or evidence-bearing frontmatter changed.
• 2026-06-08 byte-identical filesystem-copy enhancement: appended a fourth filesystem location to duplicate_filesystem_copies at raw/Manual Fetch Kimi /June 8/Kimi_Agent_Download Corruption Issue/_extracted_02_Vegetables_and_Vegetable_Products/02_Vegetables_and_Vegetable_Products/The First Hong Kong Total Diet Study_ Metallic Contaminants.pdf (same sha256 f87da014f81ceae5a964f8df6b5b7274bb2da1c805a1f7466cc27359b51d1f6c, verified byte-identical via shasum -a 256 against the canonical raw_path copy). Surfaced via a June 8 re-extraction batch placing the report PDF a fourth time under the vegetables-and-vegetable-products subfolder of the corruption-issue tree. All four filesystem locations now resolve to this canonical source page so future manual-fetch identity-check cycles in the June 8 batch skip re-ingest. No claim, value, slug, exposure number, key-numbers, or HMTc-firewall change. No new audit cycle spawned because no body or evidence-bearing frontmatter changed.

Wiki pages updated on ingest

• aluminum
• antimony
• cadmium
• lead
• mercury-methyl
• nickel
• tin
• vanadium
• cereals
• leafy-vegetables
• vegetables
• legumes
• fruits
• meat
• seafood
• eggs
• milk-and-dairy
• beverages
• seafood

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.