To control the essential design outputs


Hi all,
I got a question from the notified body, asking what are the EDOs (essential design outputs) of my product and the evidence that these outputs are in control. The EDOs should be traced to min. 1 EPR (essential performance requirement). I intend to do the follwing steps:
* Identify the EPRs from the product requirements (PRs) by mean of the CTQ (the CTQ is a 6-sigma methodology. Therefore, I need to implement the design control elments in this). As EPRs are a part of the PRs --> the EPRs will automatically be verified.
* Identify the EDOs by mean of the dFMEA and pFMEA (with high risk evaluation values).
* Trace the EDOs back to the EPRs.
* When conducting the process validation, use EPRs and their specifications as the minimum requirements of the product that need for the process validation (in OQ and PQ).
* Include the EPRs and their specifications in the release control requirement specification.
All this will be included in the Design Transfer process.
Is it an appropriate way to show that the EDOs are in control? Are there anyone, who got the same question from the authorities and what did you provide to them as the evidence?
Thank you very much for your help.


Quite Involved in Discussions
From the 13485 standard; emphasis added.
7.3.4 Design and development outputs
. . .
d) specify the characteristics of the product that are essential for its safe and proper use.

My take. Specifications (a design output document) can have lots of requirements; some are more (or less) important than others. EDO (essential design output) is how orgs identify the vital few. Essential to safety and essential to function is a common way of looking at these.

Edit: OP is on the right track by weaving this into the risk mgmt process.

Edit 2: When done well, all downstream production and process controls (including supplier controls) link to this to ensure a safe and effective device.
Last edited:


Involved In Discussions
I think it's important to note that only looking at "high risk" items may be insufficient. Something that may present an unacceptable risk is obviously important, but you may have design outputs critical to the device being suitable for its intended use.

A simple example might be a scalpel. I'm oversimplifying, but if your scalpel is totally blunt, it might not present an unacceptable risk - it can't accidentally cut anybody and the user might just get a new one. However, if it's blunt, the scalpel is totally useless for its intended purpose - cutting things. In this sense, being sharp enough could be a critical design output even if it isn't given a "high" risk when absent.

So, be careful that risk assessment isn't your only tool for analysis here. Do some design analysis to decide what is required to be both safe AND effective for the intended use.
Top Bottom