Page 51 - QA and QC

GMP Training – Quality Assurance and Quality Control by www.gmpsop.com

GOOD TO KNOW – OUT-OF-SPECIFICATIONS (OOS) CONDITIONS

If a single assay result does not meet the agreed specification, a laboratory
investigation is required before any repeat assays are conducted.

The laboratory must determine, if possible, whether a laboratory error or a

sample or batch failure is the cause of the out-of-specification result.

The Investigation should be documented according to a written procedure.

GOOD TO KNOW – TEST METHOD VALIDATION

The laboratory is required to demonstrate that all test methods are reliable,

this means they must be validated by verifying the following attributes:

precision (amount of variation in the assay)

accuracy (difference between the average results and the "true" value)

selectivity (ability of the method to measure the analyte in the presence of
interfering compounds)

sensitivity (limit of quantitation or limit of detection of the method)

linearity and range (over what range the method measures the analyte in
direct proportion)

ruggedness (how the test method Is affected by reasonable changes)

If a method is not validated, it is difficult to be assured that the results are
reliable. GMP requires that all critical steps of manufacture are reliable or
validated. This naturally includes laboratory test methods.

Copyright©www.gmpsop.com. All rights reserved
Unauthorized copying, publishing, transmission and distribution of any part of the content by electronic means are strictly prohibited.
Page 51 of 64