Handling deviations during analytical method validations

[v9991]

we have had an interesting discussion about the purpose of the deviation section and handling of the events noted during the method validation analysis

The contents of the validation report includes an an section for "deviations" and following are the contrasting intent/interpretation of it.

Argument 1 :- This section can be used to record the decision / events of the study.
viz., an decision to exclude an protocol criteria ( viz., excluding an pre-approved parameter of intermediate precision; reporting/recording the failure-event. )

Argument 2 :- all/any such scenarios to be addressed through formal QMS-Deviation/change-control , which appropriately resolves either through investigation/root-cause; or authorizes the scope of protocol/study. ; and Deviation section of report is primarily to record the summary / references to the events of the study;

Request your inputs to make the case for right argument.

 

[pziemlewicz]

I agree with Argument 1.

A regression analysis is completed with design change. This review tells you what needs to be revalidated. It is likely you can skip portions of the protocol because those aspects of design did not change. The rationale for this would be recorded in the deviation section of test report.

 

[indubioush]

Deviation section is to record any event that happened during the study that forced you to deviate from what the protocol required.

 

[v9991]

I agree with Argument 1.

A regression analysis is completed with design change. This review tells you what needs to be revalidated. It is likely you can skip portions of the protocol because those aspects of design did not change. The rationale for this would be recorded in the deviation section of test report.

to elaborate the question....the point is not about recording/citing the deviations; BUT key difference in the arguments is wrt the " Review(RCA/CAPA)-impact assessment and authorisation of such deviations." i.e., in argument-1, it infers that the decision can be made on the fly and recorded through the deviation section; and the argument-2 is to address the same formally through QMS/deviation management; and appropriately address it

just in case if weconsider argument-1 is right; : does it mean that., we set an acceptance criteria for process validation or qualification of equipment, and if and when it fails to meet the criteria, we simply take the decision to either drop the test, relax the critera etc and record same through deviation section.!!!

 

[pziemlewicz]

My misunderstanding. Thanks for clarification.

"(g) Design validation. Each manufacturer shall establish and maintain procedures for validating the device design. Design validation shall be performed under defined operating conditions on initial production units, lots, or batches, or their equivalents...."

You're required to formally establish validation procedures. That said, if you're changing those procedures it must also be done formally. Lowering the bar for acceptance criteria because you didn't meet the originally established threshold doesn't fly. If you do intend to modify/relax, there needs to be strong, documented rationale in the Change Order.