Differences - Nonconformances vs CAPA, Corrections vs Corrective Actions

G

gporter

Disclaimer: My experience with medical devices is practically nil. Having said that, I've never worked in a system where CA isn't a normal part of the process in dealing with NC material. I'm still not clear on the motivation for creating borders and territories. We get too hung up on what things are called and lose sight of what actually needs to be done. I'm often reminded of the two guys being eaten by a carnivorous reptile while arguing about whether it's an alligator or a crocodile. If something is broken, fix it.
While I agree with the spirit of your sentiment, I've been bit in the past in my medical device experience. Even logic can get you in trouble with an auditor if you have a procedure that essentially says "don't be logical" ;-)
 

Jim Wynne

Leader
Admin
While I agree with the spirit of your sentiment, I've been bit in the past in my medical device experience. Even logic can get you in trouble with an auditor if you have a procedure that essentially says "don't be logical" ;-)
You didn't specify before that you're following an illogical path in order to appease benighted auditors. I suppose that if I had more experience in medical devices that I might be in a better position to feel your pain, but I have trouble imagining a situation where I would allow a CB auditor to throw monkey wrenches into a system that (a) meets the requirements of the standard and (b) works well for my company
 

Sidney Vianna

Post Responsibly
Leader
Admin
I'd like to understand your response to that
High performing quality systems have to be well designed. In my experience, a well designed QMS should have trigger points that are seamlessly embedded in the business processes to:

first - identify any existing and potential source of nonconformity emanating from product, processes, behavior, system and external sources
secondly - determine, based on severity, frequency, risks, criticality and costs if the organization should invest time, money and effort in preventing recurrence.
thirdly - for minor instances of nonconformity, perform appropriate correction
fourthly - for instances where recurrence should be prevented, perform robust root cause analysis and corrective action.

Having "multiple" corrective actions processes/systems in a QMS seems counterproductive to me, as creates unnecessary complexity and leads to confusion. In other words, in my opinion, an organization should have a single process for corrective actions, irrespective of the source that led to the serious problem.
 

Ronen E

Problem Solver
Moderator
... if I had more experience in medical devices ... I have trouble imagining a situation where I would allow a CB auditor to throw monkey wrenches into a system that (a) meets the requirements of the standard ...
Well, I have some experience in medical devices... and the above happens occasionally (apparently more and more lately), because:
1. There isn't always agreement on what the written (standard/regulatory) requirements actually are, and
2. it's not simple or easy (sometimes hardly possible) to change your certifying body.
 

Ronen E

Problem Solver
Moderator
if the risk does not reach that high level, is it appropriate to perform corrective actions in your NMR system?
IMO, no. If the risk doesn't cross the threshold, it doesn't warrant a CA, anywhere! A proper CA is resource hungry and should not be dispensed where not necessary.
if we chose our words properly, "corrective action"-type activities could happen in an NMR, but called something different as to not step on the toes of formal Corrective Action.
Why would you want to "choose your words properly"?... "Say what you do, do what you say" (as they say :)) and if you think something is the right thing to do say it loud and clear, and if you think something is wrong eliminate it completely and thoroughly.
 
G

gporter

IMO, no. If the risk doesn't cross the threshold, it doesn't warrant a CA, anywhere! A proper CA is resource hungry and should not be dispensed where not necessary.

Why would you want to "choose your words properly"?... "Say what you do, do what you say" (as they say :)) and if you think something is the right thing to do say it loud and clear, and if you think something is wrong eliminate it completely and thoroughly.
Great points, thanks Ronen. I agree with your "say what you do, do what you say" point. I guess it was a way of showing we may take an action that goes beyond "correction" but not quite "corrective action". We could explain in our NMR procedure that its appropriate to take actions to address "proximate causes" of issues within an NMR. For example, maybe we address the cause of a "2 Whys" exercise (proximate cause) instead of a "5 Whys" (root cause), if that makes any sense...

And to be honest, I'm starting to play devil's advocate because I do agree with most of the comments on here about having "corrective actions" only in the CAPA system.
 

Mortalis

Involved In Discussions
Pardon me for resurrecting an old discussion.
Confusion can be an issue (especially in my current Company) when you throw an eQMS into the mix. For whatever reason the programmers of many of these software packages will label certain gating activities incorrectly. Whether this is laziness on their part or cost to produce the program or simply a misunderstanding of the regulations, it causes agita because the user generally cannot change those gating labels.
I bring this up because I am currently battling this situation. The inclusion of process gates called "Investigation" and "Verification" in the Nonconforming material process leads toward misunderstanding as technically they are not in the nonconforming material process. They are a part of the Corrective Action process.
The nonconforming material requirement in the standards is in place to keep true non compliant parts and pieces from reaching the Customer. Those requirements are to detect, segregate, evaluate and disposition (which has it's own set of requirements). We get hung up with the investigation in that it becomes way too in depth (for NCM) and verification gets hung up waiting for something to happen when it may take months to come back around. As a result we end up with NCRs that become months old. Yes, I am talking 100-200 days old. We get nailed by auditors all the time for not completing them "in a timely manner".
I continually have to remind the upper management as well as the Quality department of the requirements of the standard and that we have separate processes for a reason. NCR process for keeping the manufacturing process moving but not advancing bad parts (Acute issues) and the CA process for fixing systemic (chronic) issues that may continue to take part in the result of nonconforming material, among other such issues. But, too many times I see companies wanting the NCR process to have the outcomes of the CA process without having to perform a CA.
Al Rosen is absolutely correct when he stated, "Analysis of NMR data may result in escalation to a Corrective Action in your CAPA system. "
And Somashekar stated it very succinctly, "NMR .. Nonconforming Material report
is for the identification and control of the nonconforming material to prevent its unintended use or delivery. as in 8.3.2
CAR .. Corrective Action Report
is for the action to eliminate the cause of nonconformity in order to prevent recurrence. as in 8.5.2
"
 

QHawk

Starting to get Involved
We get nailed by auditors all the time for not completing them "in a timely manner".
Yes, "without undue delay".

Tick tock, tick tock it's all about the clock. All MedDev CAPA systems I have seen use "Time to Phase" as a metric. That's fine if all CAPAs are created equal. Risk assessment and scope tell us a different story. The last step of your last phase (Implementation?) prior to Effectiveness Verification (EV) should be to establish criteria for EV. The ONLY metric for EV is effectiveness. EV is usually dependent on the reporting cadence of the supporting data. Could be daily, weekly, monthly, etc.
 

Hi_Its_Matt

Involved In Discussions
@Mortalis Without knowing the details of the eQMS or its built-in process flow, I would think that "Investigation" is synonymous with "evaluate" (as in: detect, segregate, evaluate, and disposition non-conforming material). It's not a detailed investigation of the cause of the nonconforming material, but rather an investigation/evaluation of the material itself, in order to determine how to disposition it (rework, scrap, accept with concession, return to vendor, etc).

Similarly, depending on where "verification" is at in the built-in workflow, the software designer likely meant for it to be mean "verifying that the nonconforming material is indeed nonconforming," or it could mean "verifying that the material was dispositioned appropriately."

Your Nonconforming Material SOP should specify what information is to be captured in these fields.

I do not believe that this verification process-gate should be interpreted as "verifying that the actions taken to address the nonconforming material have been effective." This type of Verification activity should be reserved for your corrective action system.

Your Nonconforming Material SOP should also give guidance on when a detailed analysis of the cause(s) of nonconforming material is warranted, and when actions should be taken to address those causes. THAT is when verification of the effectiveness of actions comes into play.
 
Top Bottom