Total Diet Study of metals and other elements in food

Baxter M; Brereton N

FERA / FSA 2015 - UK Total Diet Study of metals and other elements in food

This UK Food Standards Agency total diet study measured metals and other elements in 3312 retail food samples collected from 24 UK locations. Foods were prepared as for consumption, composited into 138 food categories and 28 food groups, and analyzed primarily by ICP-MS. The report is valuable because it gives as-consumed UK baseline concentrations for broad food groups and category composites, and because it reports inorganic arsenic separately from total arsenic. All values below are mg/kg as received or prepared unless otherwise noted; the source uses a leading ~ for values above LoD but below LoQ.

Key numbers

Selected food-group composites (Table 5)

Food group	Al	Cd	tAs	iAs	tHg	Pb	Notes
Bread	4.01	0.021	<0.006	<0.012	<0.001	~0.006	Group 1
Miscellaneous cereals	3.67	0.016	~0.009	~0.013	<0.0005	~0.004	Group 2; Sn 1.26 mg/kg
Offal	~0.29	0.067	<0.006	<0.012	0.0035	0.042	Group 4
Fish	1.43	0.014	2.0	<0.012	0.0497	~0.004	Group 7; high tAs but iAs below LoQ
Green vegetables	1.28	~0.006	<0.003	<0.012	<0.0005	0.017	Group 11
Potatoes	0.99	~0.020	<0.006	~0.015	<0.001	<0.004	Group 12
Nuts	2.31	0.020	~0.005	<0.012	<0.0005	<0.002	Group 20; Ni 2.14 mg/kg
Snacks	3.97	0.059	0.011	<0.012	~0.0005	~0.005	Group 23
Condiments	4.06	0.008	0.011	<0.012	<0.0002	0.009	Group 26

Cereal and grain category composites (Table 7)

Category	Al	Cd	tAs	iAs	tHg	Pb	Sn
Flour	5.19	0.030	0.008	<0.012	<0.0002	0.006	<0.004
Buns, cakes and pastries	5.04	0.012	~0.005	<0.012	<0.0005	~0.006	~0.02
Savoury biscuits	3.02	0.024	~0.004	<0.012	<0.0005	~0.005	<0.01
Sweet biscuits	2.83	0.015	~0.003	<0.012	<0.0005	~0.005	<0.01
Chocolate biscuits	6.90	~0.017	<0.006	<0.012	<0.001	~0.010	<0.02
Breakfast cereals	2.42	0.026	~0.010	<0.012	<0.001	<0.004	<0.02
Rice	0.59	~0.006	0.036	~0.028	~0.0012	<0.004	<0.02
Other cereal products	2.32	0.014	~0.005	<0.012	<0.0005	~0.005	<0.01
Pasta	1.04	~0.009	<0.003	<0.012	<0.0005	<0.002	13.3
Pizza	3.72	0.015	~0.005	<0.012	<0.0005	~0.005	<0.01

The rice composite reports iAs at ~0.028 mg/kg and total arsenic at 0.036 mg/kg as consumed, while the fish group reports total arsenic at 2.0 mg/kg but inorganic arsenic below LoQ. That contrast is one of the report’s clearest demonstrations that total arsenic and inorganic arsenic cannot be interchanged across food matrices.

Bread category composites (Table 6)

Category	Al	Cd	tAs	iAs	tHg	Pb
White sliced bread	5.12	0.021	~0.003	<0.012	~0.0003	0.008
White unsliced bread	4.84	0.019	~0.004	<0.012	<0.0005	~0.005
Brown bread	4.27	0.026	<0.006	<0.012	<0.001	~0.006
Wholemeal and granary bread	2.72	0.025	<0.006	<0.012	<0.001	~0.004
Other bread	4.02	0.016	0.007	<0.012	<0.0002	0.005

Fish category composites (Table 12)

Category	Cd	tAs	iAs	tHg	Pb
White fish	<0.003	4.78	~0.016	0.0719	~0.004
Fatty fish	<0.003	1.08	~0.014	0.0698	~0.003
Shellfish	0.085	1.77	~0.035	0.0567	0.017
Canned salmon	<0.003	0.237	<0.012	0.0302	<0.002
Other canned/bottled fish	0.019	0.831	<0.012	0.207	<0.002
Fish-based ready meals/products	~0.007	1.35	<0.012	0.0212	~0.002
Takeaway fish-based meals	~0.006	2.51	<0.012	0.0537	~0.005

Methods (brief)

Retail foods were collected by HallMark Meat Hygiene Ltd from 24 UK locations and sent to Fera. Samples were prepared as for consumption following FSA instructions: inedible portions were discarded where appropriate, foods were cooked where appropriate, and potable tap water was used when water was needed for preparation. Prepared samples were homogenized with a Buchi B-400 mixer or Robot Coupe food processor. Food-category composites were then pooled into 28 food groups according to UK household food-purchase proportions from the Family Food Survey.

Multi-element analysis used UKAS-accredited ISO 17025 methods. Aliquots of 0.5 g dry material or up to 3 g fresh material were digested in concentrated nitric/hydrochloric acid at 4:1 using an UltraWAVE single-reaction-chamber microwave digestion system, diluted, and measured on an Agilent 7700x ICP-MS with yttrium and holmium internal standards. Iodine used a separate enzyme/TMAH extraction and ICP-MS method. Inorganic arsenic used a separate method based on Munoz et al. 1999: hydrochloric-acid standing, hydrobromic acid and hydrazinium sulfate, chloroform extraction/back-extraction, and ICP-MS measurement with gallium and rhodium internal standards.

QA/QC included a 10% duplicate audit, procedural blanks, spike recoveries, certified or in-house reference materials in each analytical batch, instrument-stability checks, FAPAS proficiency testing, and replicate-agreement criteria. For the 0.5 g LoD/LoQ row in Table 3, the LoD/LoQ grouping is 0.006/0.020 ug/kg for As, iAs, Cd, Pb, and Tl and 0.001/0.0033 ug/kg for Hg, In, Rh, and Sb. Measurement uncertainty was 21% for iAs, 17% for As, 18% for Cd, 16% for Hg, 14% for Pb, 15% for Al/Cu/Zn, 25% for Cr, and 8% for Sn.

Speciation and methods caveats

Inorganic arsenic is separately measured iAs, not inferred from total arsenic.
Total arsenic is not a proxy for inorganic arsenic in fish: the fish group has tAs 2.0 mg/kg while iAs is <0.012 mg/kg.
Mercury is total mercury; methylmercury was not speciated.
Results are composite concentrations, not sample-level distributions. They are useful for as-consumed baseline context but cannot support within-category percentile estimates.
All values are as received or prepared, not dry weight. Dry-weight comparison requires moisture conversion not supplied in this report.

Implications

Standards work: This source provides A-tier UK as-consumed composite baselines for grains, rice, bread, fish, nuts, snacks, condiments, and other food groups. It is strongest for source-route context and regulatory cross-checking, not percentile math, because the underlying retail samples are pooled before analysis.

Courses: The report is a compact teaching example for total-diet-study design, prepared-as-consumed compositing, matrix-specific arsenic speciation, and the consequences of mixing total-As and iAs endpoints.

App: Values can support UK baseline context displays for broad product groups, with prominent labels that they are composite means and not individual-product measurements.

Microbiome: Not addressed in this study.

Wiki pages this source may touch

Verification notes

Merge-enhanced 2026-05-18 from the full manual-fetch PDF path and SHA-256 recorded in frontmatter.
Corrected an earlier table-column alignment error: Table 5’s iodine column had been misread as Hg/Pb in several group rows. The page now uses the report’s actual Cd, Hg, and Pb columns from the second half of each table.
Replaced product-empty routing and non-vocabulary matrices with controlled product routes and broad matrices.
Replaced invalid wiki links such as ingredients/wheat-flour, ingredients/breakfast-cereals, ingredients/milk, metals/mercury, and regulations/uk-fsa-tds.
Strict brand-firewall check: no sampled consumer brands are named; contractor, instrument, utensil/material, reagent, and reference-material names are retained only as Methods details under the Part 12 scientific-method exception.
Fresh-context audit subagent (2026-05-18) verdict PROMOTE. Spot-checked ~80 numerical values across Tables 5/6/7/12 against the PDF — all correct. Two false-positive concerns recorded: (a) matrices: [dietary-intake, cereal-grain] flagged as possibly off-vocabulary — verified against docs/gpt-collaboration/system-prompt.md matrices list; both terms are explicitly listed, finding rejected. (b) “HallMark Meat Hygiene Ltd” spelling flagged as drift from PDF’s “Hall Mark” — verified against PDF; the report itself uses both spellings (Summary page 2 “HallMark”, Methods page 7 “Hall Mark”), so wiki’s one-word form is faithful to one of the source’s own uses, finding rejected.

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.

Commit	Date	Description
b0f3d38	2026-06-12	batch \| corpus rescreen b04 old terminal skips

Heavy Metal Index

Index