Performance improvement and single laboratory validation of classical qualitative methods for the detection of adulterants in milk: starch, chlorides and sucrose

Abstract

Simple modifications in classical methods for detection of milk adulteration were validated with significant performance improvement.

Assessing limits of detection in qualitative methods: A simple implementation of logistic regression in a web‐based R Shiny application environment and its potential in toxicology and doping control

“…The likelihood in the last column describes how accurately the model predicts each individual result, and therefore, a very efficient way of determining the correct coefficients in the model is by maximizing this likelihood. Most of the literature on the LOD for qualitative methods rely on the calculation of positive response rates for several concentration points and fitting these response rates on a sigmoid curve by a minimum squares method 1–4 . The use of a logistic regression with a maximum likelihood method as proposed here is advantageous because it offers the calculation of the probabilities of detection directly from the raw data.…”

“…The use of appropriate probability functions for this estimation has been reported but has never been widely applied in the doping control and toxicology fields 2–4 . Although there are mentions to the use of 95% detection rates and response curves in the field, 10 there are no clear descriptions of the procedures that could be used by laboratories, which may be one of the main reasons why a harmonized approach is not yet widely used.…”

“…There are several compendia and extensive literature regarding quantitative method validation and most labs are able to perform such validation studies and understand all the performance characteristics. 1 When it comes to qualitative methods, however, although there is some literature available, [2][3][4] the harmonization of the validation procedures adopted in practice for toxicology and doping control analysis is poor.…”

“…The use of appropriate probability functions for this estimation has been reported but has never been widely applied in the doping control and toxicology fields. [2][3][4] Although there are mentions to the use of 95% detection rates and response curves in the field, 10 there are no clear descriptions of the procedures that could be used by laboratories, which may be one of the main reasons why a harmonized approach is not yet widely used. The logistic model has been used as one of the main models to describe the probabilities of occurrence of a binary event such as pass/fail and alive/dead or in social sciences for subjective responses such as agree/disagree.…”

Detecting food authenticity and integrity

“…Of course, many other approaches in addition to spectroscopy are as expected very much in evidence within the collection, such as mass spectrometry and/or chromatographic techniques; [11][12][13] genomics, [14][15][16][17] with at least one of these articles concerning the detection of horse DNA in meats, 16 ELISA, [18][19][20] as well as molecular imprinted polymer-based chemiluminescence. 21 As this collection is dedicated to showcasing detection methods for food authenticity and integrity, it covers a broad range of equally important areas in addition to those dedicated to food fraud/adulteration, such as the measurement of compounds as food quality indicators, as in the case of a cyclic aldehyde for honey for example, 22 whereas others focus on contamination by unwanted compounds, whether this contamination is intentional or inadvertent, and this is exemplied via an especially relevant, timely, and important Critical Review on supply chain risk and the urgent analytical needs required for food allergens 23 Other articles concerned with chemical contaminants include, perhaps not surprisingly, those on the detection of the now infamous plasticizer melamine, 9 pesticides, 13 hazardous dyes, 11 fungicides, 19 drug and preservative residues, 18,21 multiple adulterants of milk, including chloride, starch and sucrose, 24 as well as heterocyclic aromatic amines formed when heating protein-rich foods, 12 the latter having received considerable attention and debate of late due to the mutagenic and carcinogenic potential of the compounds concerned. Of course foods can be subject to bacterial contamination (as well as spoilage) and the detection of perhaps two of the most well-known foodborne pathogens, Salmonella species 20 and Campylobacter spp.…”