Search the Elsmar Cove!
**Search ALL of Elsmar.com** with DuckDuckGo including content not in the forum - Search results with No ads.

FDA News Harmonizing and Modernizing Regulation of Medical Device Quality Systems

kj_wilson

Optimistic Global Curmudgeon
#11
In the US, QMS review and consideration occurs during pre-approval/clearance review (this must be done by FDA/HHS in the USA), during routine post-marketing inspections (MDSAP is great here!! Gotta be pre-approved by FDA, tho), and during for-cause inspections (these must be done by FDA/HHS/FTC).

FDA does assess ALL pre-approval QMS for Class II/III device submissions, just in a risk-based, least-burdensome fashion, and only to the degree they feel necessary to assure the claimed risk/benefit. If you are pushing a low-to-moderate risk 510(k) type device through the clearance process, your submission must include an attestment that you are in compliance with 21 CFR 820 (yes, the US government extends a lot of trust to manufacturers at this stage; our EU regulatory colleagues require more formal evidence at this point); the FDA reviewers look at the design control and IDE (if required) information, primarily, to deem if there is adequate assurance that a functioning QMS is possible/probable. This is where, if you claim compliance to global standards, you include your third-party certificates of compliance to said standards; this provides FDA review greater assurance that things are as you claim. Once the device is cleared, annual reports, Medical Device Reporting System Info, and routine inspections serve as primary FDA inputs for QMS assessment.

De Novo Class II (in the USA categorization) device submissions have their QMS come under additional scrutiny in the review of Special Controls, and if deemed necessary, pre-approval (classification) inspections.

In EU and countries where the GHTF/IMDRF described MDSAP is accepted fully, the pre-approval QMS assurance can be (is encouraged to be) by a MDSAP report (certificates of compliance to other required/claimed standards should be supplied at this point as well).

By the way, the "Act" in the USA mandates FDA regulatory authority over all medical devices, whether or not they are "marketed", "approved", "cleared", or just used on people in the US.
 

kj_wilson

Optimistic Global Curmudgeon
#12
Yes, but why has it done this? This is one answer from an RA colleague:

The cost of the rapidly changing standards AND their TIRS is ridiculous. Even with subscription services. And then there is ISO 13485 pointing to control of External Documents, with expectations you purchase all of these to sit unused as "references". It is a racket. AAMI, ANSI, IEC, ISO, etc.

I have heard similar opinions expressed by many individuals within the industry, including some here in the Cove. My MBA side concurs. The purpose of standards is to benefit various players financially. If they have any other benefits, those are unintended side effects.
I big part of me concurs, in the era of "open-source" and essentially free transmission, the cost of obtaining and maintaining all of the (actually useful) standards from the plethora of organizations seems, to be blunt, profiteering. If you look at the way software development standards are currently used/distributed/enforced, I think you'll see a more business efficient model.

That said, the old pragmatist considers the current state of standards management the "cost of doing business" (grumble, grumble, grumble...)
 

kj_wilson

Optimistic Global Curmudgeon
#13
So perhaps you can tell me. Were design controls intentionally buried in the QSR (which was presented as an upgrade to the GMP regulation) to assure that ODE would never look at them (or perhaps to give ODE an excuse not to), or was this just a horrible, horrible mistake?
The vagueness of the design controls section of the QSR (and to a certain/significant degree, ISO 13485) is due to the fact that no single set of instructions (description of required controls) adequately covers the extremely diverse range of medical devices. Unlike drugs (and most biologics), it is very difficult to put a fine point on just what a design control system should include. IVDs exemplify this: blood pH measuring controls will be inadequate for modern Next-Gen gene sequencing based devices, and so on.

Over the years, FDA has issued several guidances on Design Control (and is a frequent presenter at industry meeting on this matter) in an attempt to assist device manufactures navigate thru the process.
 

Watchcat

Quite Involved in Discussions
#14
The vagueness of the design controls section of the QSR.
I've never thought the design controls section was at all vague, nor have I ever heard anyone in the industry describe it as such. In any case, that's not my point. My point is that ODE seems never to have read them, much less understood them, much less lifted a finger to implement them. (And I have heard plenty of people in the industry say this.)

Over the years, FDA has issued several guidances on Design Control (and is a frequent presenter at industry meeting on this matter) in an attempt to assist device manufactures navigate thru the process.
Yeah, those I would call vague, and I think they reflect FDA's lack of understanding of design controls. Plus, another ironic example of FDA trying to explain to industry how to do industry's job, when FDA doesn't do that job itself (e.g., design and develop medical devices) and therefore is not in a good position to be telling anyone else how to do it.[/QUOTE]
 
Last edited:

Watchcat

Quite Involved in Discussions
#15
FDA does assess ALL pre-approval QMS for Class II/III device submissions,
Whoa. Not on my planet. I spoke with several colleagues who routinely work with 510(k) devices. Not on their planets, either. They confirmed that QMS information isn't included in 510(k) submissions, and it seems clear that CDRH barely has the resources to do Class III pre-approval and routine inspections.

They speculated that you might be in some specialized area in which FDA does assess the QMS for Class II devices (one mentioned additive manufacturing), or perhaps confusing the manufacturing information that is sometimes included in a 510(k) with the QMS. Otherwise, they were at a loss.

Alternatively, you might be in Minnesota...
 
Last edited:
Top Bottom