Method validation and

verification

W. W. Wong
Senior Accreditation Officer
HKAS
8 September 2009

1
 Method Validation and
verification
• What is it?
• When is it required?
• Why is it necessary?
• What are required?
• How much is adequate?
• How should it be done?
• Any questions on the questions?

2
 Method Validation – what is it
Definition –
• validation is the confirmation by
examination and the provision of objective
evidence that the particular requirements for
a specific intended use are fulfilled (ISO/IEC
17025:2005 cl. 5.4.5.1)
• verification, where the specified
requirements are adequate for an intended
use (ISO/IEC Guide 99:2007)

3
 Method Verification – what it is

• Verification – provision of objective

evidence that a given item fulfils specified
requirements (ISO/IEC Guide 99:2007)

4
 Selection of methods
(ISO/IEC 17025:2005 cl. 5.4.2)
• Methods published in international, regional
or national standard shall preferably be used.
• Laboratory developed methods or methods
adopted by the laboratory may also be used if
they are appropriate for the intended use and
if they are validated (cl. 5.4.2).

5
 Non-standard methods
(ISO/IEC 17025:2005 cl. 5.4.4)
• Verification is applicable only for standard
methods which have been validated.
• (Non-standard methods) The method
developed shall have been validated
appropriate before use (cl. 5.4.4).

6
 Laboratory-developed methods
(ISO/IEC 17025:2005 cl 5.4.3)
• …shall be a planned activity and shall be
assigned to qualified personnel equipped
with adequate resources.
• Plans shall be updated as development
proceeds ….

7
 Validation of methods
(ISO/IEC 17025: 2005 cl. 5.4.5)
cl. 5.4.5.2
• The laboratory shall validate
non-standard methods,
laboratory designed/developed methods,
standard methods used outside their
intended scope, and
amplifications of standard methods
to confirm that the methods are fit for the
intended use.

8
 Laboratory internal validation and
verification
Laboratory
Existing information requirement

Fully validated standard methods Verification

(have been studied in a collaborative (Secondary
trail) validation)

Standard methods – amplifications and Validation

modifications e.g. new instrument
Standard methods – outside their Validation
intended scope
Laboratory – developed and non- Validation
standard methods (Primary validation)9
 Validation of methods
(ISO/IEC 17025: 2005 cl. 5.4.5)
cl. 5.4.5.2
• The validation shall be as extensive as is
necessary to meet the need of the given
application or field of application.

10
 Validation of methods
(ISO/IEC 17025: 2005 cl. 5.4.5)
cl. 5.4.5.2
• The laboratory shall record the results
obtained, the procedure used for the validation,
and a statement as to whether the method is fit
for the intended use.

11
 Method Validation – what are required

Validation of methods (cl. 5.4.5.2 Note 1)

• May include procedure for sampling,

handling and transportation

12
 Method Validation – what are required

Validation of methods (cl. 5.4.5.2 Note 2)

• techniques for method performance
determination include
– Calibration using reference standards and Reference
Materials
– Comparison of results achieved with other methods
– Interlaboratory comparisons
– Systematic assessment of the factors influencing the result
– Assessment of uncertainty of results based on scientific
understanding of the theoretical principles of the method and
practical experience

13
 Method Validation – what are required

Validation of methods (cl. 5.4.5.2 Note

3)
• When changes are made in the validated non-standard
methods, the influence of such changes should be
documented and, if appropriate, a new validation
should be carried out.

14
 Method Validation – what are required

Validation of methods (cl. 5.4.5.3 )

Examples:
• Uncertainty of results
• Detection limit
• Selectivity
• Linearity
• Repeatability
• Reproducibility
• Robustness/cross-sensitivity

15
 Method Validation – what are required

Validation of methods (cl. 5.4.5.3 Note 1 )

• Validation includes
– Specification of the requirements
– Determination of characteristics of method
– Check requirement fulfilled by method
– Statement on validity

16
 Method Validation – what are required

Validation of methods (cl. 5.4.5.3 Note 2)

• Regular review as method development
proceeds to verify customer needs
fulfilled
• Modifications to development plan due
to change in requirements should be
approved and authorised

17
 Method Validation – what are required

Validation of methods (cl. 5.4.5.3 Note 3)

• Validation is a balance between
costs, risks and technical
possibilities

18
 Method Validation – what are required

Validation of methods
(5.4 H HOKLAS Policy (c))
• Requirements differ significantly from
one technical discipline to another
• Should commensurate with intended use
• Confirm fit for intended use

19
 Method Validation – what are required

Validation of methods
(5.4 H HOKLAS Policy (d))
• HOKLAS classifies test methods into 3
– standard method
– standard method with modifications
– in-house method

20
 Method Validation – what are required

Validation of methods
(5.4 H HOKLAS Policy (d))
• standard method
–Conforms exactly to the standard

21
 Method Validation – what are required
Validation of methods
(5.4 H HOKLAS Policy (d))
• standard method with modifications
– Differs from standard but deviations unlikely
to affect test results
– Need supporting evidence, i.e. validation
required
– Modifications stated in scope of accreditation
and test reports

22
 Method Validation – what are required

Validation of methods
(5.4 H HOKLAS Policy (d))
• In-house method
– Methods other than the above two
– Validation required
– Cannot make reference to other
standard method, i.e. cannot claim
“based on” a standard method

23
 Method Validation – what are required

Validation of methods
(5.4 H HOKLAS Policy (f))
• Non-standard methods shall be fully
documented and validated

24
 Method Validation – what are required
Validation of methods
(5.4 H HOKLAS Policy (g))
• Shall have policy and procedure for design,
development and subsequent validation of
– in-house methods,
– laboratory designed/developed methods,
– standard methods used outside their intended scope
– amplifications and modifications of standard
methods

25
 Method Validation – what are required

HOKLAS Supplementary Criteria No.

20 “Chemical Testing”, “Chinese
Medicine”, “Construction Materials”,
“Food”, Toys and Children’s Products”
– Chemical Testing
Section 5.3 Validation of methods

26
 Method Validation – what are required

HOKLAS SC No. 20 Section 5.3

• Concentration range
• Sample matrices
• “more advanced” techniques may
be a deviation
• Confirmation of identity

27
 Method Validation – what are required

HOKLAS SC No. 20 Section 5.3

• LoD
• Precision and trueness
• Applicable concentration range
• Applicable sample matrices

28
 Method Validation – what are required

HOKLAS SC No. 20 Section 5.3

• Method bias assessed by appropriate matrix
CRMs
– Levels of analytes
– Matrix matches intended sample matrix
– Uncertainty of assigned values suitable

29
 Method Validation – what are required

HOKLAS SC No. 20 Section 5.3

• Method performance characteristics
review, and revised, regularly

30
 Method Validation – what are required

HOKLAS SC No. 20 Section 5.3

• For food analysis, the method validation
required depends very much on the
analytes of interest and the matrices.
Common food matrices include those
rich in protein, carbohydrate, oil, dietary
fibre, liquid, etc.

31
 Method Validation – what are required

HOKLAS SC No. 20 Section 5.3

• If a method is to be accredited for
general food, satisfactory validation
data shall be obtained for at least
five different food matrices with at
least three kinds of food for each
food matrix.

32
 Method Validation – what are required

HOKLAS SC No. 20 Section 5.3

• The range of matrices shall also in line
with those listed in relevant regulations.
Due consideration shall also be taken
for the food matrices with potential
interferences, e.g. high chloride effect
on the ICP-MS determination.

33
 Method Validation – what are required

HOKLAS SC No. 20 Section 5.5

• Estimation of uncertainty of
measurement

34
Method Validation – how much is adequate

Depends on
• the criticality of the measurement
• The scope of the method

35
Method Validation – how much is adequate

• How far can the light shine?

• How far can the validation data on some
matrices be extended to other matrices?
• Need professional judgment based on technical
knowledge of the limitations of the
methodology used.

36
Method Validation – how much is adequate

• Example: General foodstuff

• Can never validate using all possible foods
• Classify food according to matrix from an
analytical chemistry point of view
• Main components of food – fat, protein,
carbohydrate, (fibre, water)
• To obtain reliable data, 3 food types for each
matrix
• Performance at specification limits should be
available
37
Method Validation – how much is adequate

0%
0%
10

%P
RO
AT

TE
%F

IN

10
0%
0%

100% 0%
%CARBOHYDRATES
J. AOAC, 83 (2), 413 (2000) 38
“The Referee”, AOAC Int’l, July 1993
 Validation of methods –
Accuracy and trueness
• Definition (ISO3534-1)
bias – the difference between the expectation
of the test results and an accept reference
value

39
 Validation of methods –
Accuracy and trueness
• Definition (ISO5725-1)
laboratory bias - the difference between the
expectation of test results from a particular
laboratory and an accepted reference value.

method bias – the difference between the

expectation of test results obtained from all
laboratories using that method and an accepted
reference value.

40
 Validation of methods –
Accuracy and trueness
• Assessment of trueness (ISO Guide 33)
The trueness of a measurement process is
checked by comparing the average x with
the certified value, µ, of a Certified
Reference Material (CRM)

41
 Validation of methods –
Accuracy and trueness
• Two factors contributing to the difference
between the certified value and the
measurement results
1. The uncertainty of the certified value;
2. The uncertainty of the results of the
measurement process

42
 Validation of methods –
Accuracy and trueness
-a2-2σD ≤ x-µ ≤ a1+2 σD
a1 and a2 are adjustment values chosen in advance
by the experimenter according to economic and
technical limitation

σD 2 = σLm 2 + sw2/n
n = number of replicate determinations
σLm = between labs fluctuation ≈ σ1 (intermediate
precision) or σL supplied by CRM

43
 Validation of methods –
Accuracy and trueness

For many measurement, sw is small in

comparison with σLm ; consequently when
n>10, σD can be equate with σLm

-a2-2σLM ≤ x-µ ≤ a1+2 σLM

44
Method Validation – how much is adequate

• Document the food matrix that have been

validated in the test procedure.
• Labs should consider additional validation
when they receive a food matrix not included
in the initial validation.

45
 Common deficiencies
• Number of food matrices used not adequate
• Number of food types for each matrix not
adequate
• Concentration levels used not adequate
• Reporting limits estimated but not verified
• Food samples chosen not representative of the
food matrices

46
 Common deficiencies

• Confirmation of identity technique not available

or adequate
• Validation/verification data analyses not done
correctly
• Measurement uncertainty not available/not
estimated correctly
• Equipment and/or test procedures not
conforming to the test std requirements

47
 Common deficiencies

• Lack of or inappropriate sampling procedure

• QC plan acceptance limits not appropriate

48
Thank you

49