When does an engineering change become a design change?

[Jon Griver]

Dear Forum,

Manufacturing companies generally control engineering changes using ECO forms or similar mechanisms.

It is not clear to me how it should be decided whether an engineering change should be managed through the 'full' Design Control process, i.e. including design planning, design input, etc. etc. Should the design control documentation be reviewed during every ECO?

What about the situation where a product was developed before Design Control was implemented, and all changes since then have been managed by ECO's, without design control documentation ever being prepared.

And finally, does anyone know what would be the FDA's approach to this issue?

Jon Griver

 

[Al Rosen]

Manufacturing companies generally control engineering changes using ECO forms or similar mechanisms.

It is not clear to me how it should be decided whether an engineering change should be managed through the 'full' Design Control process, i.e. including design planning, design input, etc. etc. Should the design control documentation be reviewed during every ECO?

I believe that all changes should be reviewed to determuine the extent of the controls required. The FDA regulations say that you have to perform verification and validation "where appropriate". If you do not perform verification or validation, they require documented justification for not doing it. As part of an inspection they will review engineering changes, especially software changes. I had the pleasure of an inspection last year.

What about the situation where a product was developed before Design Control was implemented, and all changes since then have been managed by ECO's, without design control documentation ever being prepared.

And finally, does anyone know what would be the FDA's approach to this issue?

Even though the device was developed prior to the design control regulations, when the regulations went into effect, you were required to institute the controls. Simply stated, they will issue your company a warning letter. You can search some of the warning letters issued, at this web site:
http://www.fda.gov/foi/warning.htm

To avoid the warning letter, you will need to implement corrective action prior to an inspection. If not, then at least before the end of the inspection. Letting it go until the inspection, the finding will be documented on the 483 and might still make it to a warning letter depending on what else they uncover.

 

[Wes Bucey]

Dear Forum,

Manufacturing companies generally control engineering changes using ECO forms or similar mechanisms.

It is not clear to me how it should be decided whether an engineering change should be managed through the 'full' Design Control process, i.e. including design planning, design input, etc. etc. Should the design control documentation be reviewed during every ECO?

What about the situation where a product was developed before Design Control was implemented, and all changes since then have been managed by ECO's, without design control documentation ever being prepared.

And finally, does anyone know what would be the FDA's approach to this issue?

Jon Griver

All our Engineering Change Orders (to pass muster with the FAA) go through a process to determine the "impact" of the change. I redacted a computer form (to remove company ID) and printed it to show the checklist we consider for the process. Some changes have little or no impact and are dealt with accordingly. Others may have great impact and go through a process similar to a new design.

A non-impact change might be "changing fasteners in a non-user-accessible portion of a device from phillips head to star drive to accommodate assembly machines." It is important to note there is a point where the change is so drastic, it is no longer viable to consider it a change, but a new design, effectively obsoleting the original design.

I hope this helps you put your own changes into proper perspective and to help you form a "justification" when you do not go through a new verification and validation of the revised design.

Any questions about the form can be addressed in this thread.

 

[Al Rosen]

Other aspects I did not mention in my previous post involve review to determine what effect the change has to safety and whether a risk analysis should be updated as well as a determination of the need for a Premarket notification, 510(k) to be submited.

See this post for the FDA design control guidance document
http://elsmar.com/Forums/showpost.php?p=72191&postcount=44

and, see the guidance document I've attached regarding 510(k)s for a change to a device.

 

[Jon Griver]

Thank you Al and Wes for your replies.

I would like to summarize what I have learned from your replies and other research I have done.

I previously regarded the design change mechanism and the ECO mechanism as parallel routes to implement changes, where design change is for 'big' changes and ECO is for 'small' changes.

My viewpoint now is that ALL changes have to be evaluated with regard to the design change mechanism, though not all elements of design change will be applicable to all changes. The ECO mechanism is actually a document control procedure, basically describing how to fill out the form, and does not really relate to the technical content of the change.

I would welcome any comments on this.

Regards,


Jon

 

[Wes Bucey]

Thank you Al and Wes for your replies.

I would like to summarize what I have learned from your replies and other research I have done.

I previously regarded the design change mechanism and the ECO mechanism as parallel routes to implement changes, where design change is for 'big' changes and ECO is for 'small' changes.

My viewpoint now is that ALL changes have to be evaluated with regard to the design change mechanism, though not all elements of design change will be applicable to all changes. The ECO mechanism is actually a document control procedure, basically describing how to fill out the form, and does not really relate to the technical content of the change.

I would welcome any comments on this.

Regards,


Jon

By Jove! I think he's got it!

Excellent synopsis, Jon! Especially the point about document control. Google "configuration management" for more detailed discussions of a subset of document control which may discuss the interchangeability of various versions of documents and products. (as in my example of merely changing drive heads on fasteners while other aspects - thread, length, material - remain the same)

 

[Al Dyer]

I know in the auto arena that design responsibility comes into play when an entiry has the authority to change the design record.

Is there a similar activity in place in the FDA/Medical field?

Al...

 

[Al Rosen]

I know in the auto arena that design responsibility comes into play when an entiry has the authority to change the design record.

Is there a similar activity in place in the FDA/Medical field?

Al...

Al:

In medical devices that would be the entity that holds the FDA approval for the device.