How to validate Regulations and Standards？

[nannan]

Firstly, I apologize for my poor English.
Hi, could anyone tell us how to establish validation procedure for regulations and standards.
If there were user needs that the device is lightweight, we would write that the device must be X lbs +/- Y lbs and within Z inches in height as design input. Then, I would identify them on drawings or specifications as design verification. Finally, the resulting product would be tested in the actual or simulated use environment as a part of validation. It’s easy for us to understand this process but when it comes to “Regulations and Standards, we’ve been confused.
So far, that my opinion:
1.We would write that the device must comply with XYZ Regulations and Standards as design input. Design output is specifications.
2As Early stage of design verification, we would just identify that specifications state that the device must comply with XYZ Regulations and Standards (example: Must evaluate biocompatibility complying with ISO 109930-1).
3.End stage of design output would be biocompatibility test results.
4.AS end of design verification, we would just identify that biocompatibility test result.
5. We don't know how the resulting product validation

I know my recognition is wrong so, could anyone tell us correct procedure for design the verification/validation.

 

[planB]

nannan,

for your specific example ISO 10993-1:2018, you may walk through the flowchart and referenced sections provided in Figure 1. Would this flowchart address your verification/validation questions at least with regards to biocompatibility?

HTH,

 

[yodon]

For standards, you're really just wanting to demonstrate compliance (ignore "verification" and "validation" for now). I think all of the standards describe the means by which compliance is demonstrated. NBs will generally be looking for evidence that compliance was assessed by an accredited lab. There's nothing stopping you from doing the testing yourself but that will delay the review since they won't "trust" your data whereas if you had a 10993 (or 60601-1) reports from an accredited lab then they will accept those without much review.

 

[nannan]

For standards, you're really just wanting to demonstrate compliance (ignore "verification" and "validation" for now). I think all of the standards describe the means by which compliance is demonstrated. NBs will generally be looking for evidence that compliance was assessed by an accredited lab. There's nothing stopping you from doing the testing yourself but that will delay the review since they won't "trust" your data whereas if you had a 10993 (or 60601-1) reports from an accredited lab then they will accept those without much review.

thank you for your replay.
Do you mean that standards (not regulations) are just methods of the design verification? Sorry I can't understand why it can be ignored verification and validation.

 

[nannan]

Thank you for all your support, but I think my strategies from wrong premises, so could you see following figure to clarify my understanding?
The text shown in blue is the intended question also I don't know specific things that apply to "????????".

 

[nannan]

nannan,

for your specific example ISO 10993-1:2018, you may walk through the flowchart and referenced sections provided in Figure 1. Would this flowchart address your verification/validation questions at least with regards to biocompatibility?

HTH,

thank your for your support.
May be a little different from my intention
Could you see my new post?

 

[Tidge]

I would NOT try to associate "Regulations" with "User Needs". Fundamentally: Compliance to Regulations is only required to market devices.

It is true that regulations (and standards) are often established because they are motivated for patient/user safety, but user needs are independent of regulations/standards. A design team should have a regulatory/market strategy for devices, but don't confuse this with user needs.

The model shown has many peculiarities. I would destroy it (Design Inputs exist both before and after Design Outputs?)

 

[Ed Panek]

User Needs drive Design Inputs. Generally, regulators accept solid "bench " testing to demonstrate fulfillment of user needs but sometimes its not complete. For example, if your device is to be used on human skin it will require some skin irritation mitigation. This would involve ISO 10993 and its subchapters. If you have a test report from a lab conducting ISO 10993 testing on your device along with a Literature Review on the materials you use it closes many questions from the due diligence perspective.

 

[yodon]

Sorry I can't understand why it can be ignored verification and validation.

My mistake, I meant ignore the "terms" verification and validation for now.

We frequently have risk controls that say the risk is mitigated by design in accordance with some standard. Then the test lab demonstrates compliance so in effect, the compliance testing is often associated with some of the verification. We don't, though, articulate all the requirements from a standard as design inputs (or cite the standard as a user need). The point I intended to make was to not try to put a label on compliance testing.

I also agree with @Tidge in that your model is a bit off. You won't "ask a Dr. if the device is lightweight." There are a lot of nuances there that would take a wall of text to post. If you're trying to do V&V on your own, you might want to get some more experienced help.

 

[Tagin]

We frequently have risk controls that say the risk is mitigated by design in accordance with some standard. Then the test lab demonstrates compliance so in effect, the compliance testing is often associated with some of the verification. We don't, though, articulate all the requirements from a standard as design inputs (or cite the standard as a user need). The point I intended to make was to not try to put a label on compliance testing.

Would it make sense to think of compliance with standards/regulations as 'design constraints'? Or, is it more straightforward to think in terms of two sets (or classes) of requirements: 1) user reqs., and 2) regulatory reqs.?