Guidance Document on Measurement Uncertainty for Laboratories performing PCDD/F and PCB analysis using Isotope Dilution Mass Spectrometry

This document on measurement uncertainty was developed within the network of the European Union Reference Laboratory (EURL) for Dioxins and PCBs in Feed and Food and the respective National Reference Laboratories (NRLs) of member states. Detailed guidance is given on the evaluation of measurement uncertainty in the quantitative analysis of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) and polychlorinated biphenyls (PCBs), to assist laboratories performing official feed and food control within the European Union, especially National Reference Laboratories (NRLs) and Official Laboratories (OFLs). It provides useful key elements contributing to further harmonization of compliance assessment and outlines practical aspects related to measurement uncertainty estimation.
A new concept placing special emphasis on the inclusion of current method performance data is presented. The concept covers the full analytical process from sample receipt at the laboratory through sample storage, preparation and analysis, to data processing and reporting. In particular, it focuses on the role of analytical variability generally known as "measurement uncertainty" (MU) in the interpretation of analytical results for assessment of their compliance with a specification. Effects from sampling [EURACHEM/CITAC 2007, /18/] and transport also contributing to MU are acknowledged but not treated within the scope of this document.
Two selected approaches for measurement uncertainty estimation are proposed for the determination of PCDD/Fs and PCBs in food and feed by gas chromatography-mass spectrometry (GC-MS) using internal standard stable isotope labelled analogues. An empirical, or “top-down”, approach combines contributions from intermediate (intra-laboratory) precision and trueness (expressed as bias) to estimate measurement uncertainty, both for individual congeners and for sum parameters. The working group recommends the use of the empirical approach as described in this document as the main option for MU estimation, because it is designed and developed to cover the whole analytical process and also includes the opportunity to reassess or update MU on a regular basis.
However, an alternative methodology based on a semi-empirical approach following the EURACHEM/CITAC guide [EURACHEM/CITAC 2012, /12/] is also presented. It has been designed for laboratories new to this type of analysis that have generated data from initial validation studies. In this case the semi-empirical approach may be a good starting point, however the authors recommend implementing the empirical or top-down approach once enough data have been gathered.