Page 77 - 2013 ISPAM Mid-Year Mtg Book

For validated qualitative methods, the relevant performance characteristic for verification is:
- Level of detection (LOD).
For validated quantitative methods (colony count or MPN), the relevant performance
characteristics for verification are:
-Intralaboratory reproducibility (intermediate precision).
-Measurement uncertainty (U).
-Repeatability (sr).
-Trueness (bias; if the reference method is available).
5 Verification
The objective of method verification is to assess the performance of a method in the user
laboratory against the specifications of the method established during the primary validation
and to assess the method performance on non-validated matrices.
It is not necessary to determine all performance characteristics again, only those that are
dependent on the user laboratory, for example sample types, operators, equipment, media,
etc.
The laboratory shall demonstrate that it is capable of performing the method.
The laboratory shall demonstrate that the method performs satisfactorily on the matrices
tested.
Table 1 summarizes the relevant performance characteristics of microbiological methods for
validation and verification.
Performance Qualitative Methods Quantitative Methods
Characteristic Validation Verification Validation Verification
Trueness (bias) + +
Levelof Detection + +
Limit of
Quantification + (if relevant)
Repeatability + +
Reproducibility + + (intralaboratory
reproducibility)
Inclusivity + +
Exclusivity + +
Measurement
uncertainty + + + +
6 Selection of sample types and number of samples and replicates tested
The choice of matrices depends on the sample types routinely tested in the laboratory.
A laboratory shall demonstrate that the selected method is suitable for all their sample types,
including any sample types not included in the validation study.