As (I think) Mark implied in his comment, it is important for users of ISO 13485, QS regulation, and other regulatory requirements, to be open to a big game of connect-the-dots in order to get a clear(er) picture. There is and will continue to be a lot of talk and focus on harmonization, and while that may seem like an impossible goal, I believe we are making strides to a better 'alignment' of requirements, and the FDA's decision to align with ISO 13485 is an excellent example of regulators being willing and open to this end.
A TIR is written and released because the information is valuable to the industry and there is an immediate need for it. As such, it is not subject to the same approval process as a standard. While the TIR was circulated for public review and feedback, and the comments were formally addressed, a TIR permits the inclusion of differing viewpoints on technical issues.
TIR 102 was not written for 'power users' of the QS regulation and ISO 13485, but for those who are familiar with one or the other and need to revise their QMS to incorporate and meet the other's requirements. It is a tool to allow the user to more easily identify areas of their QMS that may need more focus.
Back to
@Richard Vincins comment regarding the medical device file and device master record, I really am interested in knowing what you and others see as the differences.
------------------------------
Lena Cordie-Bancroft
Victoria MN
United States
------------------------------
Original Message:
Sent: 14-Sep-2019 08:51
From: Mark Swanson
Subject: AAMI TIR102:2019 U.S. FDA 21 CFR mapping to the applicable regulatory requirement references in ISO 13485:2016 Quality Management Systems
NOTE: This is my perspective, not those of anyone at FDA or AAMI or other members of QM01 or ISO TC210, WG1.
Dan,
I generally find it unwise to comment on a document that I have not read. We have talked a few times and I generally respect your opinion, in this case though I take exception. I believe that both you and Richard are trying to take a literal approach to something that is not done in a literal way. As we reviewed this with both the QM working group at AAMI and had open feedback on this, it seemed apparent that the literal approach was not the best way to compare requirements of the linked processes of the quality management system. We were not a set of lawyers (as has been done for the Z annexes for CEN which was the basis for TR 17223), rather this was input from users.
To outline this we outlined the approach....From the first paragraph of the TIR scope: "This document is a comparison of requirements between 21CFR820 and ANSI/AAMI/ISO 13485:2016 that demonstrates similarities, highlights differences and discusses key considerations for medical device manufacturers. It is not a word-for-word literal identification of differences, thus, the reader must be familiar with quality management system requirements along with statutory definitions to apply this report." You must understand that the requirements of ISO 13485:2016 are linked processes that depend on each other to form the quality management system. The requirements of 4.2.4 are alongside 4.2.3 and this is where the requirements are for document control. This is how the requirements have no 'significant difference'.
I know auditor (and some others) may not appreciate this approach, but hopefully users do understand this in a useful way. I know it will be quite necessary for the lawyers to have their approach in writing the FDA rule to be alongside this (and I believe this will happen before the end of the year). I think it was the delay in getting this TIR completed that delayed the FDA, not a decision by anyone at the agency to put this on hold.
I have put much work into this TIR along with several other US experts. We hope that industry finds this useful in the transition time ahead.
P.S. Also to your comment on ISO wanting to revise 13485. While there is much pressure to revise the standard to meet the high level structure of the ISO/IEC Directive and the systematic review of ISO 13485 is currently out for ballot. In my opinion, it is unlikely that the member bodies would support a revision at this time (US stating intent to harmonize, EU MDR transition, China and India changes, etc.). I think most can see that industry and regulators are on the same page in this. This has the look to me that the systematic review will result in reconfirmation for 5 years (allow the transitions to happen and stabilize). This would result in the next revision starting around 2024.
------------------------------
Cheers,
Mark Swanson, ASQ CBA, CMQ/OE, CQE ASQ, MBA
Becker MN
United States
Original Message:
Sent: 13-Sep-2019 19:46
From: Dan O'Leary
Subject: AAMI TIR102:2019 U.S. FDA 21 CFR mapping to the applicable regulatory requirement references in ISO 13485:2016 Quality Management Systems
Lena,
I had planned to concur with Richard in his appraisal of the comparison of 21 CFR 820.181, Device Master Record, and ISO 13485 Section 4.2.3, Medical Device File. However, my reasoning is not the same as Richard's and is more closely related to your question to Richard.
A few disclaimers.
I have not read the AAMI document, so I'm relying on Richard's characterization of "No significant difference".
I have not discussed my opinion with Richard. My opinions are mine.
FDA's project is, in my opinion, ill-conceived and looks to me like a solution looking for a problem. I have heard three things that bear on my opinion.
Without the support of an FDA Commissioner, the project will not move forward (there is an acting Commissioner now).
FDA had planned to release the first elements of this project in April or May of 2019, but has not. This suggests to me that the project is on hold.
ISO intends to revise 13485 before FDA could complete its work.
The phrase "No significant difference" doesn't add any value because many audit nonconformances live in the gap between significant differences and actual requirements. For me, if one document has a requirement that is not in the other, then the difference is significant, and would result in an audit nonconformance.
I prefer the approach in CEN/TR 17223:2018 which maps the requirements of the EU-MDR/IVDR Article 10 to the corresponding elements of ISO 13485:2016. The attributes are "fully covered", "partially covered", or "not covered". With partially covered, there is an explanation of the gap.
If I were to map 21 CFR 820.181 to ISO 13485:2016 Section 4.2.3 my conclusion is:
Partially Covered – ISO 13485:2016 doesn't require document control, quality assurance procedures, labeling specifications, or procedures for maintenance.
Specifically –
820.182 requires "each DMR is prepared and approved in accordance with 820.40". 4.2.3 does not have a document control requirement.
820.181(a) requires device specification while 4.2.3(b) says, "specifications for the product". I consider these to be the same.
820.181(b) requires production process specifications while 4.2.3(c) says, "specifications or procedures for manufacturing, packaging, storage, handling and distribution". I consider these to be the same.
820.181(c) requires quality assurance procedures and specifications while 4.2.3 doesn't.
820.181(d) requires labeling specifications while 4.2.3(a) requires a general description of the labeling, not the labeling specifications.
820.181(e) requires maintenance procedures and methods while 4.2.3 does not.
One could map the requirements from ISO 13485:2016, 4.2.3 to 820.181 and create a similar list of differences.
The issue for me, is that there are significant differences and they should result in audit nonconformances if a manufacturer asserted that ISO 13485:2016, 4.2.3 is the same as (or without significant differences from) 820.181. (Hint: I would expect an FDA Investigator to issue a 483.)
While not asked, I'll also comment on EN ISO 13485:2016/AC2018. In particular, ISO 13485:2016, clause 4.2.3 does not satisfy EN ISO 13485:2016/AC2018, clause 4.2.3 when applied to the MDD Annex II.
The point here is most medical device manufacturers implement MDD Annex II so there are two "two hops" from QSR to the MDD QMS.
------------------------------
Dan O'Leary CQA, CQE
Swanzey NH
United States
Original Message:
Sent: 13-Sep-2019 14:29
From: Lena Cordie-Bancroft
Subject: AAMI TIR102:2019 U.S. FDA 21 CFR mapping to the applicable regulatory requirement references in ISO 13485:2016 Quality Management Systems
Richard,
Interesting perspective! Would you mind giving examples of what you consider the differences between the MDF and the DMR?
------------------------------
Lena Cordie-Bancroft
Victoria MN
United States
Original Message:
Sent: 13-Sep-2019 06:06
From: Richard Vincins
Subject: AAMI TIR102:2019 U.S. FDA 21 CFR mapping to the applicable regulatory requirement references in ISO 13485:2016 Quality Management Systems
To be brutally honest, I did not find it helpful at all. As an example, the comparison of 21 CFR 820.181, Device Master Record, and ISO 13485 Section 4.2.3, Medical Device File. The technical report states there is "No significant difference". I can tell you in discussions over the recent couple years with FDA investigators and Notified Bodies reviewers, they view content of DMR and MDF quite differently. While there may not be "significant" difference the content can vary. Sitting across from an investigator or auditor need to be prepared to explain a DMF/MDF especially considering companies will call the file one or the other or both. I just raise this specific example because there has been many discussions (even just this last week at a client) about what constitutes a DMR and a MDF. My only hope is investigators and NB/AO/Registrar auditors are trained on this technical report so when I present the DMR to a NB/AO/Registrar, they say, "That is not your Medical Device File, please show me your MDF." And then spending an hour explaining that in fact my DMR is the MDF. However, with that said I realise this is a technical report and not a guidance document so it is not intended to explain all of our real world applications of the regulations and standards.
------------------------------
Richard Vincins RAC
Vice President Global Regulatory Affairs
Original Message:
Sent: 12-Sep-2019 10:55
From: Justin Osmond
Subject: AAMI TIR102:2019 U.S. FDA 21 CFR mapping to the applicable regulatory requirement references in ISO 13485:2016 Quality Management Systems
Thanks, everyone. Very helpful to get your impressions and experience.
------------------------------
Justin Osmond MA
Sr. Regulatory Affairs Specialist
Ottawa ON
Canada
Original Message:
Sent: 12-Sep-2019 08:47
From: Edwin Bills
Subject: AAMI TIR102:2019 U.S. FDA 21 CFR mapping to the applicable regulatory requirement references in ISO 13485:2016 Quality Management Systems
This document was developed in cooperation with FDA to identify the differences between 21 CFR 820 and ISO 13485:2016 as an input to FDA's drive to harmonize 820 with 13485.
I can be useful for for those who want to identify the areas where they may need to improve their Quality System to cover both documents.
There is a a great deal of detail in the differences, so a quality organization can become much more skilled in the two documents through studying AAMI TR 102.
------------------------------
Edwin Bills MEd, CQA, RAC, BSc, CQE, ASQ
Principal Consultant
Overland Park KS
United States
elb@edwinbillsconsultant.com
Original Message:
Sent: 11-Sep-2019 09:30
From: Justin Osmond
Subject: AAMI TIR102:2019 U.S. FDA 21 CFR mapping to the applicable regulatory requirement references in ISO 13485:2016 Quality Management Systems
Hi All,
Is everyone jumping on this? Looks pretty interesting if one is building or updating a QMS for MDSAP. I'm wondering if anyone has purchased it and would recommend.
Thank you!
------------------------------
Justin Osmond MA
Sr. Regulatory Affairs Specialist
Ottawa ON
Canada
------------------------------