[ Pobierz całość w formacie PDF ]

or imperfect recovery. Estimation of uncertainty must form part of this validation
process and in addition to covering the above factors, should address issues such as
sample homogeneity and sample stability. Advice on method validation is given in
Section 18.
17.5 Documentation of methods shall include validation data, limitations of applicability,
procedures for quality control, calibration and document control. A laboratory
documenting methods may find it convenient to adopt a common format, such as ISO 78-
2: (Ref C10), which provides a useful model. In addition, advice on documentation of
methods is available from other sources such as national standardisation bodies and
accreditation bodies.
17.6 Developments in methodology and techniques will require methods to be changed from
time to time and therefore method documentation must be subject to adequate document
control. Each copy of the method should show issue number/date, issuing authority, and
copy number. It must be possible to determine from records which is the most up-to-date
version of each method is authorised for use.
Edition 2002 28
GUIDE TO QUALITY IN ANALYTICAL CHEMISTRY CITAC/Eurachem Guide
17.7 Obsolete methods should be withdrawn but must be retained for archive purposes and
clearly labelled as obsolete. The difference in performance between revised and obsolete
methods should be established so that it is possible to compare new and old data.
17.8 When methods are revised then the validation also needs to be updated. The revision may
be of a minor nature, involving different sample sizes, different reagents etc.
Alternatively, it may involve significant changes, such as the use of radically different
technology or methodology. The level of revalidation required increases with the scale of
the changes made to the method.
18. METHOD VALIDATION
18.1 Checks need to be carried out to ensure that the performance characteristics of a method
are understood and to demonstrate that the method is scientifically sound under the
conditions in which it is to be applied. These checks are collectively known as
validation. Validation of a method establishes, by systematic laboratory studies that the
method is fit-for-purpose, i.e. its performance characteristics are capable of producing
results in line with the needs of the analytical problem. The important performance
characteristics include:
" Selectivity & specificity (Description of the measurand),
" Measurement range
" Calibration and traceability,
" Bias *
" Linearity,
" Limit of detection/ Limit of quantitation,
" Ruggedness,
" Precision.
* In some fields of chemical measurement, the term recovery is used to describe the total
bias, in other fields, recovery is used in relation to certain elements of bias.
The above characteristics are interrelated, many of these contribute to the overall
measurement uncertainty and the data generated may be used to evaluate the
measurement uncertainty (see Section 16 and refer C13) during validation.
Good practice in method validation is described in a EURACHEM Guide (Ref A3). Note
that there is no unanimous agreement on the interpretation of some of the above terms
nor the conventions used in their determination. Thus, when stating validation data, it is
advisable to state any conventions followed.
18.2 The extent of validation must be clearly stated in the documented method so that the user
can assess the suitability of the method for their particular needs.
18.3 Standard methods will have been developed and validated collaboratively by a group of
experts (ref C14-C19). This development should include consideration of all of the
necessary aspects of validation and related uncertainty. However, the responsibility
remains firmly with the user to ensure that the validation documented in the method is
sufficiently complete to fully meet their needs. Even if the validation is complete, the
Edition 2002 29
GUIDE TO QUALITY IN ANALYTICAL CHEMISTRY CITAC/Eurachem Guide
user will still need to verify that the documented performance characteristics (eg trueness
and precision) can be met in their own laboratory.
18.4 As indicated above, there are different opinions concerning the terminology and the
process of method validation. The following explanations supplement those in other parts
of this guide and are intended as a guide rather than a definitive view.
18.5 Selectivity of a method refers to the extent to which it can determine particular analyte(s)
in a complex mixture without interference from the other components in the mixture. A
method which is selective for an analyte or group of analytes is said to be specific. The
applicability of the method should be studied using various samples, ranging from pure
measurement standards to mixtures with complex matrices. In each case the recovery of
the analyte(s) of interest should be determined and the influences of suspected
interferences duly stated. Any restrictions in the applicability of the technique should be
documented in the method. This work will allow a clear description of the measurand to
be made.
18.6 Range: For quantitative analysis the working range for a method is determined by
examining samples with different analyte concentrations and determining the
concentration range for which acceptable uncertainty can be achieved. The working range
is generally more extensive than the linear range, which is determined by the analysis of a
number of samples of varying analyte concentrations and calculating the regression from
the results, usually using the method of least squares. The relationship of analyte
response to concentration does not have to be perfectly linear for a method to be
effective. For methods showing good linearity it is usually sufficient to plot a calibration
curve using measurement standards at 5 different concentration levels (+ blank). More
measurement standards will be required where linearity is poor. In qualitative analysis, it
is common place to examine replicate samples and measurement standards over a range
of concentrations to establish at what concentration a reliable cut-off point can be drawn
between detection and non-detection (also see section 18.8).
18.7 Linearity for quantitative methods is determined by the measurement of samples with
analyte concentrations spanning the claimed range of the method. The results are used to
calculate a regression line against analyte calculation using the least squares method. It is
convenient if a method is linear over a particular range but it is not an absolute
requirement. Where linearity is unattainable for a particular procedure, a suitable
algorithm for calculations should be determined. [ Pobierz całość w formacie PDF ]