Specifications

Agilent ChemStation Plus Method Validation Pack
analytical problem to address the
quantitative question.
Eventually it requires an answer
to the question "does this method
provide good results based on
independent requirements?"
For step 1, ICH and FDA used the
definition of general good quality
criteria (along with many other
contributors). These organizations
defined a number of criteria that a
separation method must fulfill to
be classified as "good quality."
These criteria are
• accuracy
• precision
• robustness
• selectivity
• limit of detection/quantification
• linearity
• calibration function
Based on the compound type, all
or a subset of these criteria must
be met. Most common compound
types are
• main compound
• side compound
• known impurities
• unknown impurities
It is obvious that it is not neces-
sary to determine the limit of
detection for the main compound
as the amount of the main com-
pound will always be closer to sat-
uration than to limit of detection.
A detailed list of compound types
and appropriate test criteria can
be found in ICH and USP litera-
ture. Method validation pack uses
built-in templates to automatically
configure the method validation
according to the guideline under
consideration.
Based on these test criteria, step
2 can be executed. This is an indi-
vidual definition by the responsi-
ble validation person of require-
ments (quantitative limits) for the
statistical results of the tests. This
step must be repeated for each
new validation and will provide
different limits for each validation
experiment. Method validation
pack maps this step with a set of
advanced statistical calculations.
The calculations offer simple
summary statistics (RSD, %RSD
and linear regression statistics) as
well as a set of advanced calcula-
tions for outlier detection, trend
tests and many more. For each
criterion, a different set of statisti-
cal calculations on the result val-
ues is performed based on the
checkpoint requirement. The
administrative user defines the
calculation limits, transferring
analytical requirements into quan-
titative result criteria. Method val-
idation pack offers tests such as
Neumann trend tests and Outlier
tests (e.g. according to Dixon)
and many more to provide and
document an assessment of the
quality of analytical separations.
32
Introduction
A comprehensive understanding
of the functionality of Agilent
ChemStation Plus Method
Validation Pack requires a brief
introduction to the general
aspects of method validation.
The goal of method validation is
“to provide documented evidence
that a specific process will consis-
tently provide results meeting the
predetermined specifications.”
This definition is taken from one
of the FDA's method validation
guidance documents. The analyti-
cal purpose of the method valida-
tion experiments is to provide a
master method with a master
chromatogram for all consecutive
separations of this particular sam-
ple. Method validation testing
must compare results of multiple
runs (it is inter-chromatographic)
in order to answer the question "is
this method suitable for the sepa-
ration task?" The comparison
must give a qualitative and quanti-
tative answer to this question
based on the analytical results. A
comparison typically involves a
human judgement, so how can a
comparison provide a user-inde-
pendent and quantitative result?
This is the most difficult task in
the method validation process,
because proper execution
requires:
1. The definition of general good
quality criteria for a method to
address the qualitative require-
ments, and
2. The definition of specific
requirements for the individual
4.