IJCA - Volume 3 - Flipbook - Page 15
2024 | Volume 3, Issue 1
There are many studies in the literature about
sampling for food safety. Besides the difficulties
in determining the uncertainty of sampling, new
statistical approaches and proposals for guidelines
are available [8, 9, 10, 11, 12, 13]. Information about
validation and/or conformity assessment of sampling
is quite rare. Conformity testing means a systematic
examination of the extent to which an entity conforms
to a specified criterion [14]. In the case of sampling
for pesticide residue determination in crops, the
analytically determined residue concentration of a
sample, evaluated according to a given specification,
shall be valid for the entirety, i.e., according to food
law, not exceeded by any item.
No procedure was found in the literature for real
product populations that allows testing this claim.
There are two main reasons concerning real product
populations:
• A multivariable dependency of the distribution of
analytes
• No reference target for pesticide residue samples.
This leads to the approach for conformity assessment
of sampling to demonstrate that during the sampling
process the following requirements are fulfilled:
• Evaluation of a very probable distribution of analytes
from the main influencing parameters.
• Adaptation of the sampling plan to the distribution.
• Taking single items under consideration of their
characteristics.
• Correct documentation of the sampling.
• Value-preserving transport to the laboratory.
The experience of the sampler has a considerable
influence on the result due to the evaluation of the
multivariable distribution dependency on-site. This
allows a reference sample to be generated, which is
only valid for the respective product at the location
and time of sampling. Other samplers or sampling
methods can be tested against the reference. The
analytical information about pesticide residues and
their concentrations in the sample is an important
source for assessment and shall be included. Many
studies are available for two further statistical
measures that are necessary for the assessment: the
unit-to-unit variability of pesticide residues in fruits
and vegetables [15, 16, 17, 18] and the uncertainty of
sampling crops [7, 8, 19, 20]. For the former there is
a requirement in the EU regulation; for the latter, only
case studies so far.
13
This study presents a conformity assessment for
sampling with some practical results.
Materials and Methods
The underlying idea is to carry out sampling and
evaluate it for the suitability of the selected sampling
plans and the plausibility of the analytical results. A
reference sample and a test sample are taken from
the same basic population and at the same time. The
test sample is evaluated regarding the results of
the reference.
The head of the test selects the population to be
sampled according to the examination requirements,
particularly related to parameters or risks that
might affect the distribution of the analytes. Such
risks include applied spraying technique, natural or
artificial hindrances, edge zones, plant physiognomy,
or vegetative characteristics of plants. The current
spraying plan for the population can also be
used for evaluating the plausibility of the residue
concentrations in the samples.
The reference sampling is planned according to
the sampling order, executed according to the
corresponding Standard Operating Procedure (SOP),
and documented by an authorized sampler. The head
of the test is present, evaluates the sampling, and
countersigns the sampling protocol. Critical nonconformities during the sampling lead to aborting the
sampling procedure and repeating the aptitude test.
Non-critical non-conformities are taken into account
in the evaluation of the conformity test. The sample
is transported to the laboratory, registered, and
analyzed. The result is delivered to the head of the
test as a complete laboratory report.
The checklist contains general data (report number,
sampler name, according to SOP) and evaluates
the sampling and documentation (the sample is
clearly identified and traceable, the scope is defined
including recognized risk parameters, there is
documentary evidence, the quantity of the sample
is adequate) and the analytical results (analytes
were found qualitatively and quantitatively within the
uncertainty of measurement).
The combined expanded uncertainty of the measured
residue (UE) was used when deciding compliance
of the residue concentration of the test sample (CT)
with the reference sample (CR) [21]: C(T,R) ± (2 x
UE x C(T,R). The default value of UE is 25% in the
European Union [22]. By comparison, the test was
also assessed by a simple laboratory uncertainty (UL)
