Using electronic lab notebooks in the design control process

duff999

Quite Involved in Discussions
Hi Folks

I am circling back on this question I posted a while back.

We are in the process of developing our design controls and have a few questions. We use electronic lab notebooks, which are controlled, for our design work.
The experiments that we intend to run are approved first before we start the work, with the acceptance criteria stated. The experiments we run have approved work instructions for each part of the process. The tests are then run, with the results documented in the same lab notebook that was approved previously. If the experiment does not meet the acceptance criteria, we document the next steps in the lab notebook, and send it out for approval. If the experiment passes all the pre-defined acceptance criteria, the lab notebook is reviewed/approved and the information from the lab notebook is entered into what we call a final release report ( think of it as a test summary report). This final report is then approved and reviewed. This is used as our evidence that we met the verification of the product design. For validation, we rely on our end user to proved feedback that the product meets their intended need.

During the initial design process we get all the user and product requirements that is needed to determine what the product requirements should be. So it seems we have our initial product requirements determined, we determine what testing is needed to meet the requirements, execute the verification tests, then send the sample to the end use for confirmation. During each phase of development we have a design review to ensure all functions are in agreement at various stages.

From the description above, does this sound like we are on the right track.

What I am trying to answer, I guess, does the documentation stored in the lab notebooks, and the final release report serve as my verification that the product meets the design inputs (product requirements)? Or do I need to take this further and have to create an number of test plans/protocols, which are already included in the lab notebooks, a have these as separate documents?

Any advice is always appreciated.
 

yodon

Leader
Super Moderator
Here's what the standard says for Design Verification (ignoring Design Validation since you didn't explicitly mention it):

The organization shall document verification plans that include methods, acceptance criteria and, as appropriate, statistical techniques with rationale for sample size.

If the intended use requires that the medical device be connected to, or have an interface with, other medical device(s), verification shall include confirmation that the design outputs meet design inputs when so connected or interfaced.


One might be able to argue that the methods and acceptance criteria are addressed with your approach. I don't see, however, what you do for using relevant sample sizes.

It's also not clear if your device connects to or interfaces with other devices.

I don't see how you demonstrate that the requirements are traced to the tests where they are verified. It would seem that would be rather challenging in the approach you describe.

There is another consideration that may not be addressed, risk management (ISO 14971). How do risk controls get verified for implementation and assessed for effectiveness?

The approach you describe is a bit out of the ordinary from what I've seen. Not saying it can't work but when reviewers see something they're not used to, they tend to want to do a deeper dive.

Also, the information collected for design verification will be provided to regulatory reviewers so that means you'd be surrendering the entire set of lab notebooks and if there's not really good indexing, they may reject it outright. Personally, I would see this approach as high-risk.
 

duff999

Quite Involved in Discussions
Thank you for your reply Yodon.

It seems from reading your response I could defend the verification process with the lab notebook, however I do not include statistical rationale. If my product requirements document states that I need at minimum of 2 lots produced to pass all tests, is this acceptable?

I plan to have a separate risk document to address 14971.

For the trace matrix, is it acceptable to trace testing back to product requirements using another separate document, this includes verification and validation testing.

I appreciate the feedback.
 
Also from ISO 13485:
Records of the results and conclusions of the verification and necessary actions shall be maintained
(see 4.2.4 and 4.2.5).
The reference to section 4.2.4 indicates that you need to have your reports or summary document undergo your process for controlling quality system documents. Is your summary document a controlled quality system document?

If my product requirements document states that I need at minimum of 2 lots produced to pass all tests, is this acceptable?
Sample sizes are based on the associated risk.
I plan to have a separate risk document to address 14971.
More than one document will be needed. Let us know if you need help.
 
Top Bottom