Regulatory Open Forum

 View Only
Expand all | Collapse all

Class I devices for Design Controls

  • 1.  Class I devices for Design Controls

    Posted 19-Oct-2016 13:26

    Can anyone provide any guidance on Class I medical devices requiring design controls. 

    This link says the following:

    https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?fr=820.30

    (2) The following class I devices are subject to design controls:

    (i) Devices automated with computer software; and

    (ii) The devices listed in the following chart.

    What kind of devices fit into category (i)? This is not any device encompassing software, correct?

    Any recommendations on minimal requirements for design controls for any Class I devices (across different jurisdictions) would be great. I'm curious to know what level of design controls, if any at all, other companies are applying to Class I devices.

    Thank you.

    ------------------------------
    Maham Ansari MS RAC
    Synaptive Medical
    Toronto ON

    ------------------------------


  • 2.  RE: Class I devices for Design Controls

    Posted 19-Oct-2016 21:05

    I'm not exactly sure what kind of guidance you're looking for, but two ways I can suggest to determine whether design controls are required for a class I device is to:

    1. find the product code of the device in question and see if it's GMP exempt.

    2. brute force approach. download the product classification files from http://www.fda.gov/MedicalDevices/DeviceRegulationandGuidance/Overview/ClassifyYourDevice/ucm051668.htm. Look for the 'foiclass.zip' file. Once you download the file (import to Excel), sort by 'DEVICECLASS' and then see which ones have the 'GMPEXEMPTFLAG' set to N.

    I don't know of any Class I devices with software that would be excluded from Design Controls.

    There are no "minimal requirements" for different device classes; all elements of Design Control are required. The scope of the activities may be different, however.

    ------------------------------
    Michael Zagorski RAC
    Pittsburgh PA



  • 3.  RE: Class I devices for Design Controls

    Posted 20-Oct-2016 16:34

    Michael, does the word "automated" add anything to this guidance?  Are there devices "with" software that are not "automated with" software?

    ------------------------------
    Julie Omohundro, ex-RAC (US, GS), still an MBA
    Principal Consultant
    Class Three, LLC
    Durham, North Carolina, USA
    919-544-3366 (T)
    434-964-1614 (C)
    julie@class3devices.com



  • 4.  RE: Class I devices for Design Controls

    Posted 20-Oct-2016 16:54

    In general, 'automated' and 'software' are equivalent terms, in my opinion.

    I think 'automated' is an older term that conveyed that the device contains software and its functions are automated, as opposed to requiring human action or intervention. Stand-alone software is in essence 'automation' too, because it automates, i.e. replaces human action (e.g. software collects and analyzes a thousand data points automatically, rather than by a person).

    ------------------------------
    Michael Zagorski RAC
    Pittsburgh PA



  • 5.  RE: Class I devices for Design Controls

    Posted 20-Oct-2016 16:30

    Are we talking US or EU?

    ------------------------------
    Corey Jaseph RAC
    Regulatory Affairs Manager
    Ultradent Products, Inc.
    South Jordan UT
    United States



  • 6.  RE: Class I devices for Design Controls

    Posted 20-Oct-2016 16:50

    In the US, design controls apply to Class III, Class II, any device with software, and any device on the list in 820.30(2)(ii). If your finished device contains software, then you must apply design control.

    If the finished device has software, then it is subject to design controls. FDA QSR has the minimal requirements for design control. If design control applies, then you must establish (define, document, and implement) and maintain procedures for each sub-section of 820.30. The manufacturer determines the content, as long as they meet the minimum requirements. In 820.30(g), you can exclude software validation as not applicable if the finished device does not contain software.

    For completeness sake, there are other requirements for software used in the QMS or software used in production. These apply regardless of device class.

    ------------------------------
    Dan O'Leary
    Swanzey NH
    United States



  • 7.  RE: Class I devices for Design Controls

    Posted 06-Nov-2016 01:35

    And then there is mobile app enforcement discretion, where design controls are strongly recommended...




  • 8.  RE: Class I devices for Design Controls

    Posted 07-Nov-2016 01:25

    ... And, design transfer always applies. 

    ------------------------------
    Randall Wheeland RAC
    Principal Consultant - Medical Device RA-QA
    rlwheeland@yahoo.com
    (210)978-3083
    San AntonioTX
    United States



  • 9.  RE: Class I devices for Design Controls

    Posted 07-Nov-2016 09:58

    There is one more variant ... If the Class I device is GMP exempt, then that is stronger than the part of the GMPs that defines Design Control applicability. i.e. if the device is GMP exempt then Design Controls are not mandatory. If in doubt, check with your favorite regulatory/FDA lawyer.

    Art

     






  • 10.  RE: Class I devices for Design Controls

    Posted 07-Nov-2016 10:40

    If you are responsible for regulatory compliance, then of course this matters.  Otherwise, it doesn't. 

    Design controls weren't invented by FDA, but by product developers.  In my experience, software developers were following design controls long before CDRH ever heard of them.  If a company is trying to develop a software product (standalone or otherwise) without following design controls, any regulatory requirements are the least of its worries. IMO.

    ------------------------------
    Julie Omohundro, ex-RAC (US, GS), still an MBA
    Principal Consultant
    Class Three, LLC
    Durham, North Carolina, USA
    919-544-3366 (T)
    434-964-1614 (C)
    julie@class3devices.com



  • 11.  RE: Class I devices for Design Controls

    Posted 08-Nov-2016 13:44

    Hmmm, my experience has been more that some software developers were following some design controls.prior to the QSR. I have, however, over the years, seen many, many gaps in software design controls. Among the most common

    - using incomplete or poorly written requirements

    - using direct VoC as requirements rather than translating them

    - thinking that DVal can happen "after beta launch"

    - little to no thought put into design transfer "what do you mean the service guys can't install it?"

    - not keeping trace matrix "history" or good design history - often in Agile software programs where it is deemed to be "busywork"

    At one point I even had a software developer tell me "FDA doesn't require software validation" so there is quite the mix out there.

    g-

    ------------------------------
    Ginger Glaser RAC
    Vice-President, Quality and Regulatory Affairs
    Maplewood MN
    United States



  • 12.  RE: Class I devices for Design Controls

    Posted 09-Nov-2016 01:22

    Sounds like the normal isolated challenges we've all faced over the years. Doesn't mean the industry on average did the same. 

    ------------------------------
    Randall Wheeland RAC
    Principal Consultant - Medical Device RA-QA
    rlwheeland@yahoo.com
    (210)978-3083
    San AntonioTX
    United States



  • 13.  RE: Class I devices for Design Controls

    Posted 09-Nov-2016 08:50

    I agree with the  other Ginger Glaser. I have run into more software developers who have great technical skills but don't document or dismiss the need for it. The irony is then subsequently hearing them criticize code their colleagues wrote but they cannot figure out, then have to leave in, because if they purge it, everything crashes. Then, I chuckle.

    Lately though, I have been lucky to work with some who get it. Do what you need to based on complexity and risk.

    Ginger Cantor, MBA, RAC
    Centaur Consulting
    centaurconsultingllc@gmail.com






  • 14.  RE: Class I devices for Design Controls

    Posted 10-Nov-2016 17:02

    Once upon a time, these developers inspired endless "real programmer" jokes:

    http://www.bernstein-plus-sons.com/RPDEQ.html

    I'm not sure if they are still quoted much today.

    ------------------------------
    Julie Omohundro, ex-RAC (US, GS), still an MBA
    Principal Consultant
    Class Three, LLC
    Durham, North Carolina, USA
    919-544-3366 (T)
    434-964-1614 (C)
    julie@class3devices.com



  • 15.  RE: Class I devices for Design Controls

    Posted 10-Nov-2016 17:00

    Yes, some, not all. In all areas of expertise, and in all industries, there will always be people who perform at different levels.  I should also clarify that my pre-QSR experience with software developers was not in the medical device industry, so I can't speak to what practices were followed by software developers at that time.

    ------------------------------
    Julie Omohundro, ex-RAC (US, GS), still an MBA
    Principal Consultant
    Class Three, LLC
    Durham, North Carolina, USA
    919-544-3366 (T)
    434-964-1614 (C)
    julie@class3devices.com