Justification of use of non-harmonized standards for MDR conformity

[TomQA]

Hello,
Harmonised standards are used to provide manufacturers with a means to presume conformity with the requirements of the MDR (with the annexes ZA, ZB, etc.) . Whilst for the MDD almost all the standards were harmonised, it is not the case yet for the MDR with only a few harmonised standards .

Therefore, how do we justify the applicability of a non-harmonised standard to presume conformity with the GSPRs ? Do we need to build for each non-harmonised standard a justification that this standard is applicable ? I have read that it is not enough to say that they were previously harmonised with the MDD ?
If anyone has an example to share, a template or other I would be highly grateful
Thank you very much !

 

[planB]

Tom,

rather straightforward: regulators / notified bodies "love" /recognise standards - so whenever there is a published standard with your device in scope, you are more than encouraged to use it. You typically do not have to justify using it.

Examples:
EN ISO 10993-1:2020 is not (yet) harmonised under the MDR. But it contains the recognised state-of-art approach for demonstrating biological safety of medical devices with body contact.

AAMI ST72:2019: being a US-national standard, never going to be harmonised under the MDR; however, the recognised state-of-art approach for dealing with bacterial endotoxins related to medical devices, just for the fact, that in terms of standards there is nothing else that specific and elaborate around.

HTH,

 

[mihzago]

Most recent standard, even if not harmonized, is considered the state-of-the-art (SOTA). MDR expects you to apply SOTA.
Most, if not all NBs will not reject your approach to use the most recent version, especially considering the harmonization process in EU is currently a total mess.

 

[TomQA]

Hello !
Thank you for both for your responses !
Do you think I need to write a rationale to formalise and explain that the most recent standards are SOTA ?
Thanks !

 

[planB]

Do you think I need to write a rationale to formalise and explain that the most recent standards are SOTA ?

As @mihzago explained: no - that's the inherent nature of a most recent standard revision to be be more technically advanced than the superseded version.

 

[mihzago]

You shouldn't need to have one, but if you do write it, you'll have it ready in case an NB or some other auditor gives you hard time about it.

 

[TomQA]

Hi guys,
Thank you very much for your help.
May I have your feedback on the following entry present in the guidance "MDCG 2021-5 - Guidance on standardisation for medical devices" :

"the most recent versions of standards with the technical solutions they contain reflect the “state of the art”. However, due to the nonlegal status of the concept of “state of the art” and its complexity, with so many different and dynamic aspects to be taken into account, the mere compliance with the most recent version of a standard which reference is not listed in the OJEU does not automatically imply compliance with the requirements of the applicable EU legislation, if no further evidences are provided in the technical documentation of the product"

This is why I was questioning the presence of a document to provide "further evidences" for the harmonised standards.

 

[mihzago]

Thanks for sharing this.
I imagine there may be cases where an older standard is more appropriate than the latest one, or an industry does not accept the new one, but probably very rare. I don't have any examples.

I guess you have a few options:
- do nothing and wait until an NB or potentially a different auditor pushes back on your choice to use the latest standard, or;
- create a generic justification that states what I wrote earlier, that the current standard is generally considered to be SOTA and in the absence of properly functioning harmonization process that's the best choice, or;
- write a detailed analysis for each standard explaining why the latest standard is appropriate rather than the previous version, or a non-harmonized one.

 

[Ronen E]

As @mihzago explained: no - that's the inherent nature of a most recent standard revision to be be more technically advanced than the superseded version.

I respectfully disagree. Normally it takes time (as in years) for a newly published standard to be truly and broadly accepted as "more technically advanced", e.g. by industry and regulators worldwide. This is not necessarily related to the formal EU harmonization. First of all, there is usually a transition period outlined by the publishing org (e.g. ISO or IEC). Then there are sometimes essential disagreements among prominent stakeholders - especially when the new revision represents a new approach about some aspect (technical or other). The fact that a standard has successfully gone through the authoring, review and approval process by the participating parties still doesn't, in real terms, make it SoTA, even when many/prominent parties are involved in this process.

You shouldn't need to have one, but if you do write it, you'll have it ready in case an NB or some other auditor gives you hard time about it.

Again, I disagree. I know from experience that NBs and auditors/reviewers DO expect to see such rationalization, and BSI's guidance even states it explicitly. When you actually try to compose such a rationalization you realize how non-trivial this issue is - sometimes it's not at all clear which is SoTA - the latest published version or a previous long-standing one. I think it's actually a beneficial process for everyone involved, because it highlight the differences and makes you think really hard which ones matter, and how all that relates to real life and what the competition is doing, where science and technology are at, etc.

- write a detailed analysis for each standard explaining why the latest standard is appropriate rather than the previous version, or a non-harmonized one.

Yes, this is definitely what I do, and also what NBs seem to look for. "Detailed" doesn't mean it has to be lengthy. In most cases a paragraph or two will suffice, and I never wrote more than half a page for that. If you need more, it's probably a sign you're trying to justify something that is questionably justifiable.

I have read that it is not enough to say that they were previously harmonised with the MDD ?

It might not be enough as a single argument, but it's not a bad starting point. Then you can consider what happened since the publication of the last HS list for the MDD, and also any relevant gaps between the MDD and MDR.

Cheers,
Ronen

 

[TomQA]

Thank you both for your precious feedback !

Yes, this is definitely what I do, and also what NBs seem to look for. "Detailed" doesn't mean it has to be lengthy. In most cases a paragraph or two will suffice, and I never wrote more than half a page for that. If you need more, it's probably a sign you're trying to justify something that is questionably justifiable.

In your small justification do you make a mention to the corresponding GSPR ? Or even simpler ? Do you think you could share a small exemple for 1 standard please ? I would be highly highly grateful
Thank you very much !