Change Control for Test Samples

[ga2qa23]

Howdy We're a medical device company that will soon transition from R&D to testing of our medical device components/assemblies. The FDA requires that our test samples are "representative of the final product". But we have a very expensive circuit board, sealed inside of our medical device, and it's not actually relevant to most of our testing (biocompatibility, cleaning, sterilization).

So we will make a separate "non-functional" model that is the same shell, but no guts inside (no expensive circuit board) to use for these kinds of testing. Our real/functional medical device goes through our change control process and is assigned an document ID number and a version number. So should this fake "non-functional" model also go through our change control process to get a separate document ID number and version number?

In my mind, we could either A) put the "nonfunctional" model through our change control too, or B) don't do that but instead write any explanation of differences in the separate test protocol documents for biocompatibility testing, cleaning testing, sterilization testing, etc.

Any input would be appreciated. Thank you!

 

[indubioush]

Either way is fine, but here are some "shoulds" for you:

You should have an approved test protocol that justifies the use of these shell-only assemblies and describes how they are different from the clinical device. The protocol should be approved prior to building these units.
The build plan for these units should be approved to ensure the build will be in line with the protocol requirements.
You should have a complete device history record for these units that meets the applicable DHR requirements in 21CFR820.

An additional consideration: Prior to performing design verification testing, develop a test plan and try to use devices for more than one test where possible. This little bit of planning can save a lot of money.

 

[ga2qa23]

I understand what you mean. You're basically saying that I have two options:

• Either approve the "shell-only assembly" as its own print with a document ID number and version number. Then save a DHF folder for it. Then reference it in the test protocol;
• Or instead I save the "shell-only assembly" print in the Appendix of the test protocol. Then in the resulting test report, I'd save the DHR documents as an Appendix in the test report.

 

[indubioush]

I understand what you mean. You're basically saying that I have two options:

• Either approve the "shell-only assembly" as its own print with a document ID number and version number. Then save a DHF folder for it. Then reference it in the test protocol;
• Or instead I save the "shell-only assembly" print in the Appendix of the test protocol. Then in the resulting test report, I'd save the DHR documents as an Appendix in the test report.

Actually no, that is not what I am saying. You can go about documenting the modified assembly in many different ways. I am just saying that you need to document it and justify its use. The modified assembly does not need a design history file. It is not a device. It represents a device. You need to be very clear on how it represents the device in the correct way in order to show that the results of your design verification test are applicable to the actual device.