Discussion in 'ISO 13485 and ISO 14969 – Medical Devices QMS' started by Marcelo Antunes, Aug 25, 2015.
But maybe my sample experience was not big enough
I respect and appreciate your experience, but I have seen such compelling arguments. In some of these cases it would have been difficult to establish safety through theoretical analysis, and extremely expensive / lengthy to do so via lab testing, so it made much more sense to rely on field data.
There's a lot of variety out there.
Okay, but was an analysis made of this? I'm saying this because what I usually see is people directly going to the field data ONLY route, without any analysis, with a simplistic argument that "it has always never caused harm, so it's safe". This is what I mentioned that makes no sense at all. Worse, most of the time a decent rationale could be provided without too much work, simply because some analysis/tests/etc have been made in the past, but people did not notice that or that they already have risks controls in place.
I think we have got 2 issues mixed up here, I was talking about retrospective validation of software used in the manufacturing process on in support of the QMS.
With regard for safety and effectiveness of devices in the market we need to review all of the available data from pre and post market sources so it can be declared as safe and effective when first placed on the market and during its life time.
As Marcelo says this data needs to be analysed, to understand what is actually happening. I have seen areas where it has been reported that there is a problem with a device as one user has had several failure in the same way, and it sound like this had all occurred recently. On investigating this statement is true but these failures occurred over a 4 to 5 year period. The user has had used the device multiple times without a problem but only remembered the problems. So you need to be careful of the reverse of saying we have had minimal problems with a device so it is safe, but on looking at the data you have lots of single reports for the same issue from multiple places.
I think further discussions along these lines would need a different thread under post market follow up, not under the impact of the changes to ISO 13485
Yes, we went a little off-topic because Ronen commented on part of a comment of mine, however, if you think about, the discussion is in fact the same - how to use historical data to support a claim.
With the above in mind, here is a basic flowchart I use to validate claims.
So ISO 13485 was approved, we will have ISO 13485:2016 published in a while!
I am getting pressure to update our system to 13485. I think it would make sense to wait for the newest revision to be published. Any chance you could give me a very rough idea when it will be published? Thanks in advance!
Probably in February.
ISO 13485:2016 is now released...http://www.iso.org/iso/home/store/catalogue_tc/catalogue_detail.htm?csnumber=59752 and you can preview it.
Is the eventual intent to have this new revision be an internationally recognized standard in terms of meeting national regulatory requirements?
Are CB's now going to review all risk-assessments for each process, and scrutinize software validations? Will they have authority to reject a manufacturer's risk-assessment conclusions? If so, wouldn't this entail a whole new level of CB qualification?
Note that also the european version are available from the different normalization entities as BSI (first to publish)
Is there any notable difference between the (BS) EN version from shop.bsigroup.com versus the one from the iso.org shop?
I'd like to just purchase a single copy...and the EN version (from the BSI shop) is nearly twice the price of the one available at iso.org!
Is there any reason to purchase the EN version over the ISO version?
The (BS) EN version has the ‘Z’ annexes which do not appear in the ISO version. These indicate the standard’s relationships to the essential requirements of the three EU medical device directives.
Worth the extra price-tag? I'm actually curious how this relationship is documented...
The essential requirements (Annex I) of the Medical Device Directive 93/42/EEC, for example, does not explicitly have requirements for a quality system: only statements with phrases such as "must be designed and manufactured in such a way...". Is the relationship pretty much just an index to these requirements? ...or is there additional information that could be useful?
Just a reminder that until EN ISO 13485:2016 is officially harmonised (published as such in the OJEU) this aspect is rather meaningless - except as a heads-up.
The Z annexs are available elsewhere as part of the "previews" given by some standards authorities. BSi have 6 pages in their preview - the Danish standards authority shows 35! which includes all of the Z annexs. Link below:
Fantastic! Thanks for the link! Very helpful.
It's interesting that the comparison is not actually against the essential requirements (Annex I), but rather the regulatory pathways outlined in the other Annexes (II, II, IV & V).
This is because ISO 13485 is used primarily not for demonstrating conformity with the ER, but rather with requirements under various conformity assessment routes.
Annexes Z were carefully revised by ISO TC 210 WG 1 before the FDIS, and from what I quickly saw, the EU Commission reviewers seemed to agree with the proposals (but I may need to check more carefully or ask the right people).
Separate names with a comma.