You don’t need to do design verification testing.

That’s right, I said it. The FDA regulations (21 CFR 820.30) do not say anything about testing for design verification. Nor does FDA’s definition of design verification:

 820.30(f) Design verification:

Each manufacturer shall establish and maintain procedures for verifying the device design.

Design verification shall confirm that the design output meets the design input requirements.

The results of the design verification, including identification of the design, method(s), the date, and the individual(s) performing the verification, shall be documented in the Design History File.

820.3(aa) Verification means confirmation by examination and provision of objective evidence that specified requirements have been fulfilled.

I don’t see the word test in those clauses above. But why not? For most of you when you hear the term design verification, your mind immediately goes to a test lab and sample sizes. Are you wrong? No. But I think the old, and still very relevant FDA Guidance on Design Controls says it best:

Any approach which establishes conformance with a design input requirement is an acceptable means of verifying the design with respect to that requirement.

I love this section of the guidance, because it really drives home the concept that design input requirements are the key to good design controls. Once you have appropriate requirements, you will naturally do the work required to verify those requirements (and then later validate). If you create a robust set of design input requirements that do not require testing, then you don’t need to do any verification testing. In this case the verification would be carried out by analysis and inspection of the design.

Analysis is a very broad bucket and can include activities like materials analysis to identify the strength of a component. For example, your analysis might show that a handle can withstand a weight of 50 kg, but from your design input requirement, the handle only needs to withstand 10 kg. Assuming that the user risk related to this requirement is low, it could be appropriate to just use this analysis to verify the requirement, since the calculated strength is five times larger than the expected load. This of course assumes that the design input requirement is correct as well…

However (I’m sure you saw a big however coming), it would be very difficult to have a medical device with design inputs that didn’t require testing. I could potentially see a very simple Class 1, very low risk device, that is similar to other products on the market. But in the industry, those types of products are very few and far between.

But at this point I need to make it clear that verification testing is the highest and most comprehensive level of verification. In honor of the late Sinéad O’Connor, for design verification testing, Nothing Compares 2 U.

Realistically, you are never going to have a design that does not need some testing. There are often standards that you will need to adhere to that involve testing and you are rarely going to have a 5X safety factor, like in the example above. Therefore, you’ll have to test to have high confidence of the design meeting the requirement.

That final functional design verification testing is really the summation of all the design verification activities that were completed on the device. This is going to include analysis, inspection and characterization testing (this has other names like engineering studies, prototype testing, etc…). Those activities essentially hold up the final design verification testing. For example, analysis is required to determine the worst-case configuration of the product and inspection is utilized to confirm that the design meets the requirements of applicable standards. Lastly, characterization testing is done early to potentially be an input into the design requirements, and gain insight on how the device is performing.

If you take away one of the legs of that stool, the final verification testing is not going to be very stable. But if the team did a good job on the lower-level verification activities, it would better support the entire verification effort, and give high confidence in device performance.

The big takeaway is understanding that testing is not the only way to verify your design but is really the summation of all the design related verification activities. And Nothing Compares 2 Verification Testing.

Ref: Nothing Compares 2 U.  1990.  Written by Prince and performed by Sinéad O’Connor.  Chrysalis.