If an IVD device QC panel (or any other medical device component or finished device) fails to meet final release criteria, then, as applicable, ISO 13485:2016 (as amended; hereinafter "ISO 13485") and also the U.S. FDA's Quality System Regulation both deem that such a failure represents "nonconforming product" which shall be processed in accordance with ISO 13485 clause 8.3 ("Control of nonconforming product") and in the case of the FDA, pursuant to 21 CFR section 820.90 ("Nonconforming Product"). As others have already said, that generally means that the nonconforming product must be identified, documented, controlled/segregated/dispositioned to prevent unintended use, evaluated, and, where necessary commensurate with risk, investigated.
At this stage, "CAPA" hasn't yet happened. As an aside, the GHTF rightfully reminds us that "CAPA" isn't the best term to use since each "CAPA" will either be CA or PA, but not both. Nonetheless, the governing authorities (e.g., U.S. FDA, European Notified Bodies, ISO registrars, etc.) nonetheless relentlessly, pervasively, and formally continue to use the acronym "CAPA". So be sure you are ready to cooperate with them regarding their preferred terminology rather than fighting.
Remember that CAPA is different than nonconformity. Ultimately, for example, the U.S. FDA has said that, "...every nonconforming issue...is not a CAPA...".
Now, if your nonconforming product evaluation determines the need for "investigation" (i.e., root cause investigation) and CAPA, then it is at this point that the separate CAPA mechanism is to be initiated (see ISO 13485 clause 8.5.2 or 8.5.3 as applicable, and/or FDA's 21 CFR section 820.100). This evaluation of the need for investigation and CAPA needs to, at the very least, be a risk-based evaluation.
Regarding the risk-based evaluation:
- Don't let yourself be discouraged from the proper use of FMEA during risk analysis if that's what your current process leans on. For example, ISO/TR 24971:2020 Annex B "Techniques that support risk analysis" reminds us that FMEA is indeed a technique (among others) that can be leveraged to support the risk analysis. There are a variety of risk analysis techniques that can be employed to characterize the risk of the nonconforming product. While ISO 14971 risk analysis indeed ultimately requires assessment of risk under fault conditions in addition to when there is no fault condition, FMEA absolutely remains an analytical technique that can be employed to support a risk analysis when there is a fault condition (such as nonconforming product). Now, FMEA, if employed, does indeed need to be extrapolated where appropriate to derive the ultimate harm (i.e., damage/injury), if any, to patients and users if FMEA is used as part of ISO 14971 risk analysis. FMEA can be quite useful for elucidating/linking the ISO 14971 sequence(s) of events that can lead to harm. Yet FMEA is still also useful, where relevant, and thus should not be banned, for gaining process-centric or product-centric risks that might also trigger the need for CAPA.
Regarding the root cause investigation:
- My resource library is full of longstanding, sound, and expert basis clearly maintaining the need for identification of the "root" cause(s) during CAPA investigations. A few paramount examples repeatedly establishing the "root" cause concept are ISO 9000:2015 (QMS vocabulary and fundamentals), Juran's Quality Handbook, ISO TC 210's ISO 13485:2016 Practical Guide, FDA's past and current CAPA trainings for industry, and others. Indeed, I'd say that no CAPA fundamental is more crucial to effective CAPA than finding and eliminating the "root" cause(s). Dan and Mark, can you elaborate further on your withdrawal from the notion of "root" cause(s)? If this is just a semantical issue, then I'd advise not abandoning the "root" cause vernacular. It unequivocally remains true that the intent of the longstanding "root" cause concept still needs to be carefully addressed to assure proper CAPA. When ISO 13485 demands that we, "eliminate the cause of nonconformities", "the" cause means the root cause(s). That emphasizes the longstanding principle that, if only superficial / symptomatic cause(s) are addressed rather than the "root" cause(s), then the CAPA will fail. It is the "root" cause(s) that ISO 13485 and FDA intend to be eliminated when implementing CAPA.
------------------------------
Kevin Randall, ASQ CQA, RAC (U.S., Europe, Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright 2023 by ComplianceAcuity, Inc. All rights reserved.
------------------------------
Original Message:
Sent: 13-Jul-2023 15:27
From: Dan O'Leary
Subject: QC Panel non-conformance
This is a two step process. The first step is control and disposition of nonconforming product under 820.90. The second step is corrective action under 820.100.
Under 820.90 you need to identify, document, evaluate, segregate, and disposition the nonconforming product.
Documentation means that you will write an NC. Evaluate includes deciding whether you need an investigation. Disposition means deciding what to do wither nonconforming material. Disposition could include correction, use-as-is, or scrap. The role of the NC is to document your decisions for each of these three areas.
If the evaluation decision is to investigate, that leads to a corrective action under 820.100. (Note: there is no such thing as a CAPA.) The investigation will, hopefully lead to the cause (Note: there is no such thing as a root cause) and action to prevent recurrence.
The problem is that your question convolves two steps – controlling the non-conformance and taking corrective action.
Many companies have defined paradigms for corrective action to help determine the cause. They don't always work.
The good news is that you don't think you can determine the cause, then in the evaluation step, determine that corrective action (investigation) is not necessary. However, I would open a corrective action, conduct the investigation, and document that the methods did not determine the cause. This puts you on better grounds for an external audit or an FDA Investigation.<o:p></o:p>
------------------------------
Dan O'Leary CQA, CQE
Swanzey NH
United States
Original Message:
Sent: 13-Jul-2023 07:06
From: Anonymous Member
Subject: QC Panel non-conformance
This message was posted by a user wishing to remain anonymous
Hi every one,
Our QC panel has failed to meet final release criteria. QC team has conducted a top level FMEA but they could not identify the root cause. what is the point of opening NC if we are unable to to determine the root cause? do we need to escalate this to CAPA? IF YES, how CAPA can help to identify then prevent this ?
Thanks.