100
7 Validation of the Measurement Procedure
7.10 Criteria for Selecting the Measurement Procedure
Before selecting the most appropriate measurement procedure, tailor-made for the
intended research the analyst should participate in a discussion on the aim of performing measurements. Although the terms of reference should be established by
client, the analyst should be aware of the use of the result. Formally, customers of
the laboratory can be of external origin or internal origin, the latter meaning from the
same organization. In both cases, it is essential to evaluate whether the laboratory
has the necessary resources, including the competent personnel to implement ordered
tests or calibrations. Arrangements can be of a more or less formal nature; the less
formal approach applies primarily to internal clients, for which the contract review
can be performed in a simplified way. It is worth remembering that the ‘client’ may
also be the analyst; this applies to those cases in which research studies requiring
measurements are conducted, for example.
It is relevant to recall the part of Clause 5.4.2 (ISO/IEC 17025:2005), stating
that, “The laboratory shall use test and/or calibration methods, including methods
for sampling, which meet the needs of the customer and which are appropriate for
the test and/or calibrations it undertakes.”
Therefore, it is clear that the analysts should orient themselves in issues related to
the conduct of measurements, which allows for the evaluation of the problem from
the point of view of a particular choice of the analytical process. The responsibilities
of the analyst include discussing with the client the possible measurement capability,
so as to avoid improper actions or erroneous conclusions.
The selection of the measurement procedure requires a decision stemming from
the knowledge of the object (type of matrix and analyte) as well as the intended use
of the results. Thus, the analyst must collect and evaluate a range of information
relating to the expected testing and quality of results (Table 7.5).
7.11 Validation Parameters
Validation of the measurement procedure can be carried out within the laboratory
(single laboratory validation) and by interlaboratory comparisons in which the participating laboratories use the same procedure. In any case, validation includes the
specification of analytical requirements; the evaluation of selected analytical parameters; assessment of the extent to which the expected requirements can be met; and
finally the preparation of the report, with a clear statement about the suitability of
the procedures for the intended purpose (the validation claim). Measures of this
assessment are assigned numerical values for the individual parameters. The most
important parameters are selectivity, the range of calibration, sensitivity, limit of
detection and determination, robustness, accuracy and precision.
Selectivity defines the possibility of determining a compound of interest in the presence of other components in the sample (i.e. the matrix). A method that is perfectly
7 Validation of the Measurement Procedure
7.10 Criteria for Selecting the Measurement Procedure
Before selecting the most appropriate measurement procedure, tailor-made for the
intended research the analyst should participate in a discussion on the aim of performing measurements. Although the terms of reference should be established by
client, the analyst should be aware of the use of the result. Formally, customers of
the laboratory can be of external origin or internal origin, the latter meaning from the
same organization. In both cases, it is essential to evaluate whether the laboratory
has the necessary resources, including the competent personnel to implement ordered
tests or calibrations. Arrangements can be of a more or less formal nature; the less
formal approach applies primarily to internal clients, for which the contract review
can be performed in a simplified way. It is worth remembering that the ‘client’ may
also be the analyst; this applies to those cases in which research studies requiring
measurements are conducted, for example.
It is relevant to recall the part of Clause 5.4.2 (ISO/IEC 17025:2005), stating
that, “The laboratory shall use test and/or calibration methods, including methods
for sampling, which meet the needs of the customer and which are appropriate for
the test and/or calibrations it undertakes.”
Therefore, it is clear that the analysts should orient themselves in issues related to
the conduct of measurements, which allows for the evaluation of the problem from
the point of view of a particular choice of the analytical process. The responsibilities
of the analyst include discussing with the client the possible measurement capability,
so as to avoid improper actions or erroneous conclusions.
The selection of the measurement procedure requires a decision stemming from
the knowledge of the object (type of matrix and analyte) as well as the intended use
of the results. Thus, the analyst must collect and evaluate a range of information
relating to the expected testing and quality of results (Table 7.5).
7.11 Validation Parameters
Validation of the measurement procedure can be carried out within the laboratory
(single laboratory validation) and by interlaboratory comparisons in which the participating laboratories use the same procedure. In any case, validation includes the
specification of analytical requirements; the evaluation of selected analytical parameters; assessment of the extent to which the expected requirements can be met; and
finally the preparation of the report, with a clear statement about the suitability of
the procedures for the intended purpose (the validation claim). Measures of this
assessment are assigned numerical values for the individual parameters. The most
important parameters are selectivity, the range of calibration, sensitivity, limit of
detection and determination, robustness, accuracy and precision.
Selectivity defines the possibility of determining a compound of interest in the presence of other components in the sample (i.e. the matrix). A method that is perfectly