Use of cut‐off values on the limits of quantification reported in datasets used to estimate dietary exposure to chemical contaminants

left-censored data, limit of quantification, analytical results, chemical contaminants, dietary exposure
First published in EFSA Supporting Publications
20. Juli 2018
Approved
4. Juni 2018
Type
Technical Report

Abstract

The analytical data on the presence of chemical contaminants in food and feed used in dietary exposure assessments in EFSA scientific outputs often contain a high number of left‐censored data, i.e. non‐quantified/non‐detected data; EFSA usually uses the so‐called substitution method to make use of this type of data in its assessments. However, the presence of left‐censored data together with a wide range of detection and/or quantification limits is often an important source of uncertainty associated with the dietary exposure estimates. This Technical Report describes a step‐wise approach to select the most appropriate cut‐off values on the provided censoring limits based on: 1) legal requirements; 2) typical expanded uncertainty levels; or 3) distributions of the quantified values and reported limits of quantification. This approach is intended to minimise the influence of left‐censored data on the uncertainty associated with the dietary exposure estimates.

Contact
DATA.Admin [at] efsa.europa.eu
doi
10.2903/sp.efsa.2018.EN-1452
Question Number
On request from
EFSA