Starska et al. 2011 — Pb, Cd, tHg, tAs in 483 Polish milk and dairy samples (2006–2007 NIH monitoring)

National monitoring of Pb, Cd, total Hg, and total As in 483 milk and dairy product samples drawn from all 16 Polish voivodships in 2006–2007, coordinated by the National Institute of Public Health — National Institute of Hygiene (NIH) and executed through the integrated voivodship sanitary-epidemiological station laboratories. Products covered: liquid milk, fermented liquid products (kefir, yogurt, buttermilk, curds), non-fermented liquid products (cream, coffee cream, condensed milk, milk desserts), maturing cheeses, cottage cheeses, butter, and ice cream. Analytical methods met the performance criteria of EU Commission Regulation 333/2007 and were accredited to EN ISO/IEC 17025. Across all formats, contamination was low and on average well below the 0.020 mg Pb/kg liquid-milk limit then in force (Reg. 1881/2006); two milk samples did exceed the limit (highest 0.05 mg Pb/kg), and a domestic salami cheese reached 0.220 mg Pb/kg (the corpus maximum). Estimated total daily intake from milk and dairy for a 60-kg adult contributed only 1–2% of the then-applicable PTWI values for each metal but, against the newer JECFA/EFSA reference points adopted in 2010, the 90th-percentile lead intake from this product group reached 13.4% of BMDL₁₀ for renal effects in adults — and for a 20-kg child reached approximately 50% of the BMDL for neurotoxic action, which the authors flag as the principal residual concern.

Key numbers

All values in mg/kg wet weight unless noted. Total n = 483 (2006: 227; 2007: 256) across all 16 voivodships. Below-detection results were treated at one-half of the range per the SCOOP convention (e.g., a result of <0.01 mg/kg was assigned 0.005 mg/kg).

Headline averages reported in the abstract (mean across all milk samples vs. all milk-product samples):

• Pb: milk 0.008 mg/kg; milk products 0.017 mg/kg.
• Cd: milk 0.001 mg/kg; milk products 0.002 mg/kg.
• tAs: milk 0.005 mg/kg; milk products 0.009 mg/kg.
• tHg: milk 0.001 mg/kg; milk products 0.002 mg/kg.

Milk and liquid milk-product detail (Results, pp. 1045–1046):

• Milk (n=75): average Pb 0.008 mg/kg; 90th percentile 0.017 mg/kg; two samples exceeded the 0.020 mg Pb/kg legal threshold; highest single result 0.05 mg/kg.
• Liquid processed milk products, domestic vs. imported from other EU member states: average Pb 0.013 vs. 0.015 mg/kg; 90th percentile 0.033 vs. 0.031 mg/kg; highest 0.072 vs. 0.059 mg/kg.
• Cd in domestic milk and liquid milk products: average 0.001 mg/kg; 90th percentile 0.003 mg/kg; highest milk result 0.007 mg/kg. Highest Cd in processed liquid products: 0.020 mg/kg (domestic) and 0.039 mg/kg (other EU).
• As: average 0.008 mg/kg; 90th percentile 0.019 mg/kg; highest 0.08 mg/kg (across milk and liquid products combined).
• Hg: average 0.001 mg/kg; 90th percentile 0.002 mg/kg; highest 0.010 mg/kg.
• No significant differences in element content between fermented and non-fermented liquid milk-product groups.

Solid milk-product detail — cottage and maturing cheeses, butter, and ice cream (Results, p. 1046):

• Across the four formats combined: average Pb 0.020 mg/kg; 90th percentile 0.043 mg/kg; highest 0.220 mg Pb/kg in a domestic salami cheese; 0.147 mg Pb/kg in a rural cottage cheese and in an ice cream sample.
• Cd average 0.002 mg/kg; 90% of results below 0.004 mg/kg.
• tAs average 0.009 mg/kg; 90% of results below 0.020 mg/kg.
• tHg average 0.001 mg/kg; 90% of results below 0.003 mg/kg.
• Maturing cheeses carried the highest values across all four analytes: average and 90th percentile for Pb 0.026 and 0.054 mg/kg; Cd 0.002 and 0.003 mg/kg; As 0.011 and 0.025 mg/kg; Hg 0.002 and 0.004 mg/kg.
• Butter carried the lowest values: average Pb 0.014 mg/kg, average Cd 0.001 mg/kg.

Estimated daily intake (60-kg adult; average with 90th-percentile in parentheses), µg/person/day:

• Milk: Pb 0.936 (1.989); Cd 0.117 (0.351); As 0.585 (1.170); Hg 0.117 (0.176).
• Liquid fermented and non-fermented milk products: Pb 0.468 (1.188); Cd 0.036 (0.108); As 0.324 (0.792); Hg 0.040 (0.068).
• Cheese: Pb 0.713 (1.488); Cd 0.062 (0.155); As 0.310 (0.682); Hg 0.047 (0.093).
• Butter: Pb 0.140 (0.310); Cd 0.10 (0.20); As 0.070 (0.170); Hg 0.010 (0.030).
• Total daily intake from milk and milk products combined: Pb 2.257 (4.975); Cd 0.225 (0.634); As 1.289 (2.814); Hg 0.214 (0.367).

Intake expressed against reference values (adult, 60 kg):

• Pb: 1.1% (2.3%) of the then-applicable PTWI; 6.1% (13.4%) of BMDL₁₀ for renal effects in adults. More than 40% of the dairy-derived Pb intake is contributed by liquid milk alone.
• Cd: 0.4% (1.1%) of the JECFA PTWI, 0.4% (1.3%) of the JECFA PTMI, and 1.1% (3.0%) of the EFSA TWI. More than half is contributed by milk.
• As: 1.0% (2.2%) of the then-applicable PTWI; 0.7% (1.6%) of BMDL₀.₅ for lung cancer. About 45% contributed by milk.
• Hg: 0.5% (0.9%) of the PTWI for total mercury, 0.6% (1.1%) of the PTWI for inorganic mercury, and 1.6% (2.7%) of the PTWI for methylmercury. More than 55% contributed by milk.
• Across all four metals, milk and dairy contribute 6.1% (Pb), 1.4% (Cd), 6.2% (As), and 4.5% (Hg) of the modeled Polish total dietary intake.

Child exposure flagged by the authors. Assuming a 20-kg child body weight, intake of lead at the 90th-percentile contamination level via milk and milk products would reach approximately 50% of the BMDL for neurotoxic action — the principal residual concern in the authors’ conclusions despite all aggregate intakes remaining well below adult PTWIs.

Regulatory context as of 2006–2007: Pb in liquid milk capped at 0.020 mg/kg under Reg. (EC) 1881/2006; Hg in foodstuffs generally capped at 0.01 mg/kg (analytical-method limit of determination) under Reg. (EC) 396/2005; no EU-wide maximum levels for Cd or As in milk or dairy at the time. The JECFA Pb PTWI of 0.025 mg/kg b.w. was withdrawn in mid-2010 (after this survey was complete but before publication); the JECFA Cd PTWI was replaced with a PTMI of 0.025 mg/kg b.w., and EFSA had set a 2.8× tighter Cd TWI of 2.5 µg/kg b.w. on 30 January 2009.

Methods (brief)

Sampling per Commission Directive 2001/22/EC (through May 2007) and Commission Regulation (EC) 333/2007 (from 1 June 2007), executed via the voivodship sanitary-epidemiological stations under sampling plans prepared by the NIH. Analytical methods accredited under Reg. (EC) 882/2004 and EN ISO/IEC 17025, with performance criteria meeting Commission Regulation (EC) 333/2007: LOD < 1/10 of the maximum level and LOQ < 1/5 of the maximum level for the analyte in question (for Pb in liquid milk, the required LOD < 0.004 mg/kg and LOQ < 0.008 mg/kg given the 0.020 mg/kg legal limit). Pb and Cd determined by flame atomic absorption spectrometry (FAAS) or graphite-furnace AAS (GFAAS) depending on matrix and concentration; As by hydride-generation AAS (HGAAS); Hg by cold-vapour AAS (CVAAS). Internal quality control with certified reference materials including CRM 8435 Whole Milk Powder (NIST) and CRM BCR-150 Spiked Skim Milk Powder. Proficiency tested through NIH-organised inter-laboratory rounds for voivodship stations and through FAPAS, the Swedish National Food Administration trace-elements scheme, and the JRC-IRMM Community Reference Laboratory for the NIH reference laboratory itself. Speciation was not measured — arsenic and mercury totals only, no iAs/MeHg fractionation. Below-detection results were treated at one-half of the range per the SCOOP convention to enable cross-country comparison.

Implications

• Certification (HMTc): Largest available national survey of Pb, Cd, tHg, and tAs in Polish dairy, suitable for the European clean-market dairy occurrence baseline. Mean Pb in Polish liquid milk (0.008 mg/kg) sits at roughly 40% of the 0.020 mg/kg EU limit in force at the time, with 90th-percentile values at 0.017 mg/kg; two of 75 milk samples nonetheless exceeded the limit (highest 0.05 mg/kg), and the maturing-cheese sub-format carried the highest mean Pb (0.026 mg/kg) among the seven product formats studied. Butter consistently sat at the low end of the Pb and Cd distributions for solid dairy products, complementing rather than contradicting the Pankiewicz (2012) finding that butter and milk powder concentrate total mercury (this study reports only that average and 90th-percentile tHg values across the cheese/butter/ice-cream group were 0.001 and below 0.003 mg/kg respectively). Useful as European-baseline corroboration for the dairy occurrence pool.
• Courses: Reference example of a national-monitoring design under EU Reg. 882/2004 and 333/2007 with full chain-of-custody from voivodship sampling through accredited regional laboratory through NIH reference laboratory through inter-laboratory proficiency testing. Useful as a worked example of below-detection handling under the SCOOP “one-half of the range” convention.
• App: Contributes to the contamination profiles for milk-and-dairy, whole-milk, yogurt, and butter in a European (Polish) context, with sub-format granularity across liquid milk, fermented and non-fermented liquid products, cottage cheese, maturing cheese, butter, and ice cream. Values land in the lower tail of the European dairy distribution and contribute occurrence weight rather than threshold-setting weight; the child-exposure flag (50% of BMDL for neurotoxic action at the 90th percentile, 20-kg child) is the load-bearing finding for downstream lead-in-dairy synthesis on the metals/lead page.

Wiki pages this source may touch

• lead
• cadmium
• mercury
• arsenic
• milk-and-dairy
• whole-milk
• yogurt
• butter
• butter-and-ghee
• eu-1881-2006-contaminants-superseded
• jecfa-lead-ptwi-withdrawn
• jecfa-cadmium-ptmi
• efsa-cadmium-twi
• efsa-lead-contam-2010
• efsa-arsenic-contam-2009

Verification notes

2026-05-19 (Claude Code, merge-enhance against the source PDF). The previous version of this page (dated 2026-05-14) had four classes of defect that this revision corrects:

1. Invented or wrong slugs in frontmatter. Previous ingredients: listed ingredients/milk and ingredients/cheese, neither of which exists in the current taxonomy. Replaced with [[ingredients/milk-and-dairy]] and [[ingredients/whole-milk]], plus retained [[ingredients/yogurt]] and [[ingredients/butter]]. Cheese is carried in matrices: (cottage-cheese, maturing-cheese) rather than as an ingredient slug because no cheese ingredient page exists; only cheddar-cheese does, and this paper does not measure cheddar. Previous products: listed products/mixed-meals-non-rice, which is not applicable to a paper that measures milk, yogurt, cheese, butter, and ice cream; removed. [[products/butter-and-ghee]] is retained because the paper measures 61 butter samples explicitly.

2. Wrong compliance statement. Previous wording read “No exceedances of the 0.020 mg/kg Pb limit for liquid milk were detected.” Source page 4 explicitly states: “two samples of milk did not meet the 0.020 mg Pb/kg threshold provided for in the legislation; the highest result was 0.05 mg/kg.” Corrected accordingly in both the opening prose and the Key numbers / Implications sections.

3. Missing exposure numbers. Previous Key numbers section contained only the headline averages and the regulatory context; the source’s full estimated-daily-intake breakdown by product format and analyte (source pp. 1048–1049), the % PTWI / PTMI / BMDL₁₀ / TWI EFSA framing, and the child-exposure flag (50% of BMDL for neurotoxic action at the 90th-percentile, 20-kg child, source p. 1049) were all absent. All added in this revision.

4. Missing maximum-value context. Previous version did not report the corpus-maximum Pb of 0.220 mg/kg (domestic salami cheese), the 0.147 mg/kg in a rural cottage cheese and in an ice-cream sample, the maturing-cheese sub-format averages and 90th-percentiles for all four analytes, or the per-format breakdown of average and 90th-percentile values across milk, liquid milk products, cottage cheese, maturing cheese, butter, and ice cream. All added.

Speciation discipline. The paper measures arsenic by HGAAS and mercury by CVAAS without speciation. metals: carries tAs and tHg (not iAs or MeHg). The Methods section makes this explicit.

Brand firewall (Part 12, strict 2026-05-17 reading). The source does not name any brands. Voivodship-level geographic descriptors (Upper Silesia, Lower Silesia, Lublin, Podlasie, Beskid Średni, Bieszczady, Wrocław region) and the regulator-event identifier “RASFF notifications” are retained as scientific-method / regulatory-event references, not brand identifiers. Comparative cross-country results from other published surveys (UK, France, Germany, Spain, etc., referenced in the source’s Discussion) are not reproduced here in tabular form because they are properly part of metals/lead, metals/cadmium, etc. synthesis pages and not part of this single-source Key numbers section.

Wiki/HMTc firewall (Part 2). No HMTc threshold proposals or certification-aligned framing in ## Implications; the section limits itself to what the paper’s findings contribute to the European dairy occurrence baseline. Direction of edit: toward the literature (correcting the wrong compliance statement, populating the missing exposure numbers, and removing invented slugs all make the wiki more accurately reflect the source); no movement toward HMTc convenience.

Routing. products: [[[products/butter-and-ghee]]] is the only locked HMTc row this paper feeds, via the 61 butter samples (n=61 in 2007 cohort). Milk, yogurt, cheese, and ice-cream sub-formats do not yet have locked HMTc product pages and route through ingredient pages (milk-and-dairy, whole-milk, yogurt) and the matrices: field instead.

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.