Document Approval Signature Order

[JPerry]

For medical devices, is there a standard or guidance that dictates the order of approval of a document?

For example, must the Author approve before the Reviewers, and only then can the Approvers sign?

It seems that that would be an ideal order. But a situation where all parties are geographically (and time zone) separated can generally make this take longer than a process where, once everyone has agreed (verbally, etc) to the final version, it is signed as early as possible by the signatories. This might mean that an Approver would sign before the Author.

Any signatory could apply their own judgement on pre-conditions; e.g. they're not going to sign until they've seen particular other signatures. But, that would be above-and-beyond the

As far as I see, ISO 13486 and the QSR only require that records be signed by the appropriate authorities - the do not impose any requirements or expectations on the order/dates of the signatures.

Am I missing something in regulations or guidance that prescribe a particular order? Or, is there some inherent danger in allowing approvals to happen in any order?

Thanks!

 

[John Broomfield]

If the document has to be signed by more than one person then it usually ordered by their rank in the organization or the technical folk before a manager.

 

[yodon]

No order is imposed nor are any particular roles required. You mention the author signing like it was a requirement; it's not. Your organization needs to determine what roles are required to approve (presuming, as @John Broomfield points out, IF the document has to be signed)

 

[outdoorsNW]

When I worked in medical a while back, the author signed first, then anyone else could sign in any order. I faintly remember there was one set of conditions (uncommon) where one person/role always signed last.

 

[Jim Wynne]

In theory at least, if there are multiple people signing (indicating approval), all it takes is one who objects and refuses to sign to send the thing back to the drawing board, so the order is mostly irrelevant. I also don't see a need for the author to sign at all, as it should be safely assumed that when the author completes the document and passes it on for approval that she has tacitly approved it. Remember that the "author" could be a person who has no authority in the matter whatsoever, and is writing because he's the only one who's not a crappy writer.

 

[Johnnymo62]

What's the point of having a reviewer? spelling, grammar, etc.?

 

[Jim Wynne]

What's the point of having a reviewer? spelling, grammar, etc.?

When a document sets out process requirements, it's usually necessary that more than one person reviews the requirements to ensure that what's been specified is (a) doable and (b) correct as to technical aspects. Of course, it's good to make sure that spelling, grammar and word usage are correct as well.

 

[Ninja]

In my experience (non-medical), we had four signatories per document:

1. The person who wrote it ('cause none of the managers were going to take time doing it) that it was what they wrote after editing by the below.
2. Production manager - affirming that production could actually understand and do what was written...and that it made sense.
3. R&D manager - affirming that R&D could actually do what was written, since they would do it a while before Xfer to production.
4. Quality - because none of the above had any sense about grammar, spelling, etc. and that it was clear enough for an internal audit later.

Didn't have to be that way...it could have been one if it worked...four is what worked for us. HTH

 

[Johnnymo62]

Silly me, I assumed the Approver would verify everything...

 

[Ninja]

Silly me, I assumed the Approver would verify everything...

LOL...I'm the CEO...you think I have a clue what those R&D folks are doing, let alone why? If I knew that, I wouldn't pay them so much.
I look at R&D Spend, 3yr average revenue from products launched less than 18mnths and calculate ROI.

You want me to approve whether or not the SiO2 should be added all at once, or added in stages throughout the mixing process?
You want me to approve whether a 3000cP nominal viscosity is adequate to avoid streaks in the coating?
...and THATS why multiple approvers is not so bad a thing...we're all good at some things, very few are good at it all...and those who are good at it all are too busy to read docs.

(and typically the absolutely WORST person to review a document is the person who wrote it...exceedingly smart folks (spikes) tend to be horrible writers).