Interlaboratory studies and Proficiency tests


The European Union Reference Laboratory (EURL) for halogenated POPs in Feed and Food organizes interlaboratory studies and proficiency tests (PTs) on the determination of PCDD/Fs and PCBs, PBDEs, HBCDDs, PFASs and CPs in food and feed matrices for National Reference Laboratories (NRLs) of EU member states regularly twice a year. These interlaboratory studies and PTs are also open for official laboratories of these member states, NRLs from other countries and in certain cases also for commercial laboratories performing analysis for self-control of industry. Since 2006, numerous interlaboratory studies and proficiency tests covering various food and feed matrices were organized by the EURL, some of these in cooperation with the Norwegian Institute of Public Health (2007) and the RIKILT – Institute of Food Safety (2010). Objective of the interlaboratory tests is to assess the analytical performance of participating laboratories and the interlaboratory comparability of results from analyses of the relevant parameters. For assessment of the analytical performance, the determination of the assigned value and the scoring of the results are of profound importance and are therefore based on the requirements of ISO/IEC 17043, ISO 13528 and the IUPAC technical report on proficiency testing. Beyond, the network of EURL and NRLs of EU Member States for halogenated POPs in Feed and Food developed additional criteria for an overall assessment of PT test samples (for PCDD/Fs and CPBs) and for results from application of bioanalytical screening methods for PCDD/Fs and DL-PCBs.

Structure of the proficiency tests

EURL proficiency tests comprise the determination of the analytes of interest in feed and food samples applying at least one of the following methods for PCDD/Fs and PCBs and all kinds of methods for other analytes:
- GC-HRMS-methods for PCDD/Fs and dioxin-like PCBs
- GC-MS/MS (or other alternative methods for GC/HRMS) for PCDD/Fs and dioxin-like PCBs
- Bioanalytical screening methods for PCDD/Fs and dioxin-like PCBs
- any kind of method for indicator PCBs

For reporting of results, laboratories applying physico-chemical methods are asked to report, besides the analytes of interest and the sum parameters, if or if not the test sample exceeds respective EU maximum or action levels layed down as sum parameters for regulated feed and food matrices beyond reasonable doubt, taking into account the measurement uncertainty and the applied expanded measurement uncertainty. Laboratories applying bioanalytical screening methods are requested to report if the sample is suspected to be noncompliant with EU legal limits and confirmation is required, and, if applicable, PCDD/F and/or PCB results in bioanalytical equivalents (BEQ).

Test material

Test materials are prepared from regular market food/feed or samples from contamination incidents are used. In some cases test samples are fortified with the analytes of interest. Selection and/or fortification of the test materials is performed in a way that the concentration of analytes of interest cover the range of the level of interest. Tests for sufficient homogeneity are performed for sum parameters and individual congeners/substances.

Calculation of assigned values

Statistical evaluation of the PT results is performed according to ISO 13528 and the IUPAC protocol. Assigned values for the test samples are determined by estimating the consensus value of participants’ results (including LOQ of individual congeners). The Huber robust mean is taken as assigned value after excluding extreme outliers (outside the range of ± 50 % of the median of all reported results) and examination of the distribution of the remaining results using histogram and kernel density estimation, if necessary. The proportion of participants’ results contributing to the assigned value is calculated and must be higher than 2/3 of all reported results. The Huber robust mean is additionally compared with the median of all values. For individual congeners (including LOQs) assigned values are only then calculated according to the above mentioned procedure, if less than 1/3 of all reported concentrations (including LOQ) are outside the range of ± 50 % of the median of all reported results. For other congeners only the median of all reported values (including LOQs) is calculated. Robust standard deviation and standard uncertainty on the assigned value are calculated according to IUPAC. Assigned values are calculated for sum parameters and individual congeners/substances. In addition, TEQ-based results are re-calculated using the WHO-TEFs of 2005.

Scoring of results

Physico-chemical methods:
Criteria for successful participation of laboratories using physico-chemical methods are based on the evaluation of the results of sum parameters and evaluated individual congeners/substances. The criteria are applicable for sum parameter concentrations in the range (about 0.5 to 4 times) of the level of interest (maximum or action level). For evaluation of results, z-scores are calculated. For WHO-PCDD/F-TEQ, WHO-PCB-TEQ and WHO-PCDD/F-PCB-TEQ the standard deviation for proficiency assessment is defined as being 10 %, for the sum of six indicator PCBs (PCB #28, 52, 101, 138, 153, 180) as 15 % and for evaluated individual congeners/substances as 20 %. Acceptable z-scores are between - 2 and + 2; not acceptable are z-scores outside the range of - 3 to + 3.

A “positive scoring” system including results for PCDD/F and PCB sum parameters and congeners has been developed within the EURL/NRL network. This new scoring system yields an assessment for one PT sample covering all relevant sum parameters and congeners and will be included in the evaluation in future PTs. The total score is calculated according to some general principles:
• Calculation of z-scores for sum parameters and evaluated individual congeners
• Calculation of the positive scores for individual congeners based on their contribution to the sum parameters (TEQ, sum indicator PCBs) and z-score

Bioanalytical screening methods:
According to current EU regulations, “a screening method in principle classifies a sample as compliant or suspected to be non-compliant. For this, the calculated BEQ level is compared to the cut-off value […]. Samples below the cut-off value are declared compliant, samples equal or above the cut-off value as suspected to be non-compliant, requiring analysis by a confirmatory method.” Therefore, the main criterion for evaluation of results from bioanalytical screening methods is their ability to reliably identify compliant samples and samples suspected to be non-compliant with established legal limits. For further evaluation of the performance of bioanalytical screening methods, bioassay-scores were applied: The reported BEQ-values derived from bioanalytical screening methods are compared to the WHO-TEQ consensus values calculated on basis of the results of physical-chemical methods for the concentration range of 0.5 to 2 times the level of interest. Due to the focus of bioanalytical screening methods on the decision over compliance or potential non-compliance of a sample, direct comparison of bioassay-scores and z-scores is not possible. However, bioassay scores may serve as a tool to assess method performance within the scope of external quality control measures of the respective laboratory. For PCDD/F-BEQ, PCB-BEQ and PCDD/F-PCB-BEQ, the bioassay target deviation is defined as being 20 %.