1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
Dismiss Notice
You must be a registered member in order to post messages and view/download attached files in this forum.
Click here to register.

Validation during Design & Development

Discussion in 'ISO 13485 and ISO 14969 – Medical Devices QMS' started by Nikki, Apr 28, 2016.

  1. Nikki

    Nikki Well-Known Member

    Jul 31, 2015
    Likes Received:
    Trophy Points:
    Background: We are a medical plastics compounder. We make the plastic pellets and sell them to customers who make medical devices out of them. We do not make medical devices.

    This year, I am going for ISO 13485:2003, and dropping our current 9001:2008.

    I am reviewing the Design and Development - And it states that Validation shall be performed.

    Currently - we have it documented that the validation process would be that three lots of the same formulation would be produced and fully tested to ensure consistency and meeting the customer's specifications. BUT - we only perform this process if the customer requires it. This is not something that we do for all new formulations.

    We have been compounding for 17 years. We have built up enough data over this time, where we know and understand how the polymers react under certain processes.

    A lot of the formulations ordered, are very similar, therefore we know the process already needed to provide the customer with what meets their needs.

    We are custom compounder - so we don't always make the same thing.

    I see the "Shall" but do we need to run a validation run on EVERY new formulation in the way I mentioned above?

    Could I state that quality control testing that we perform is the validation? This would take place while the first production lot / order is running. It would be completed prior to delivery as the standard states.

    You're input is greatly appreciated.

    Thank you,
  2. Ronen E

    Ronen E Well-Known Member

    Jul 31, 2015
    Likes Received:
    Trophy Points:

    ISO 13485:2003 was written with finished medical devices in mind. It's application in other situations is a result of market creep over the years. ISO 13485:2016 was intended to address this shift but I won't relate to it since you are specifically working towards the 2003 version now (which is fine for the time being).

    Given the above, there is room for interpretation in your situation. I think that an auditor applying the ISO 13485:2003 language and its available guidance to your operation in a rigid way will be acting somewhat unfair, but it can happen. I can see sveveral points where the standard's language can be read such that you don't have to validate each and every tiny formulation variation as long as it is documented and tested in an orderly and planned fashion.

    Generally speaking, design validation is expected where routine (production) verification is impossible / impractical. From the way you describe it, you have 100% verification built into your production anyway. I would make sure though that all the important characteristics of the resulting pellets are within spec for every lot, if verification is to substitute validation. If you run multiple lots in sequence and the affecting conditions are stable throughout (consider both controlled settings AND noise, such as day/night temp) then maybe just one in every X lots should be tested for the attributes that are more complicated to test. The easy ones should be tested 100% and could use as a flag that the process is shifting out of consistency.

    MarkMeer and Nikki like this.