Infallibility of QA department?

I'm interested to know your experience and practices with respect to Non-Conformance reporting for incidents involving or caused by QA personnel.

As we know, everyone makes mistakes, including QA personnel. The scenarios in which individuals in QA departments may find themselves on the wrong side of procedures are many and varied, but let me just fabricate a couple of examples:

1) A QA associate executes a procedure on which s/he has not been trained
2) A QA associate makes a material modification to a Quality System document after final signatures have been obtained (paper-based QMS)
3) A QA associate omits a required step in the execution a Change-Control process, for example a risk management step, and yet proceeds to close the file and approve the implementation of the change.

My question may seem quite odd to some of you, but it is this: Does one follow standard nonconformance procedures in such instances? Are QA errors just like errors in any other part of an operation, or is it the case that one 'never files a nonconformance report on activities of the QA department'?

On one side of this question is an assertion I've received that, 'nonconformances on QA can't exist, because QA is the basis of the Quality system.' On the other side of this is a point of view that QA personnel and systems are as fallible as those of any other department or function, and that QA practice is tremendously benefitted by the honest calling-out and examination of errors as part of a continuous improvement environment.

What is your practice? Do you file Nonconformances on QA errors? Do regulators like to see QA departments that are open and honest about their own fallibility, while striving to be better, or are QA nonconformances merely a sign of weakness to be kept behind a veil?

 

Good day @Malius , and welcome to the site.

Is there a reason that you ask? Would you expect "QA" department/nonconformances to be managed differently?

In all organizations I was a part of, the title/department had no bearing. (i.e. ....quality department or senior leadership ...all identified nonconformances were managed accordingly.

Hope this helps.
Be well.

 

Hi John,

Thanks for your response! Yes, there is a reason I ask, but I don't want to deviate the thread too far from the original question. I, personally, would not expect any bias toward or against the recognition of nonconformances anywhere in the organization, but it will really help me to know what practice is like in organizations other than ours. I really hope I can get insights from others, which might help inform an effort to evolve the 'Quality Culture' in our own.

m

 

Hi Jennifer!

Thanks for your reply. Much appreciated. I don't want to put words in your mouth, so just to be clear, you are saying that there should be no difference in the standards applied to QA and other operational departments, and that nonconformance reporting should be entirely unbiased by the department in which an error occurs. Is that right?

It seems to me that you view nonconformances as a positive from the point of view of employees who might be working in an environment where processes can fail. That makes sense to me in that nonconformance reporting is the first step in making improvements.

I am personally of the perspective that for QA to refuse nonconformance reporting involving QA processes, it ensures that opportunities for improvement are lost. This also has, frankly, the secondary effects of frustrating improvements in other areas of the organization where processes interact with QA.

Lots of words, but what I really want to know is: Is it normal for QA to explicitly hold itself to be infallible and to refuse nonconformance reporting? Is there any regulatory guidance that makes it clear that under ISO9001 or ISO13485, QA nonconformances should be filed?

If I were in charge of QA, I would encourage nonconformance reporting to ensure transparency and to foster a robust program of continuous improvement.

Best,
Malius

 

Hi Andy,

I feel that there may be 'less than meets the eye' to my question original question. I appreciate your concern that there may be a deeper reason I am pursuing this, and I think that you are trying to determine whether the QMS or our processes are sufficiently flawed that our QA professionals have to struggle to work within them.

As a matter of fact, we have a quarter-century-old, mature and custom-made QMS that is generally quite functional (with a few warts, sure). QA, and other divisions of the business, are generally quite comfortable with it. It has a mature change-control process and is kept up to date, and it is fully compliant with both ISO9001 and ISO13485. We work under both.

I'm not talking about people doing good work within the bounds of a bad quality system, or people 'doing the right thing' in a way that's nominally noncompliant. I'm just talking about mistakes. Furthermore, in my initial message in this thread, I acknowledged that neither people nor a QMS should be held to be perfect. Quite the opposite. Humans and human endeavours are error-prone.

A nonconformance report, in my mind, is a way of acknowledging this reality, of exposing its impact and implications, of assessing the attendant risks and devising risk mitigations, and of making improvements as required to make the QMS and underlying processes better -- to best decrease the risk of recurrence.

Our nonconformance process runs well in exactly this way. Every department files, investigates, and responds to nonconformances as a regular part of our practice. These run to around 250 every year in an organization of 80 people. They run the gamut, from products scrapped due to a manufacturing error, QC assays going wrong because of technician error, R&D making a calculation error causing a production problem, and problems with facilities, environmental controls, and equipment.

The one thing you will never find among our hundreds of routine nonconformances is any nonconformance in which an error was made by a QA person, or in the course of a QA process, because these are not permitted by our unwritten rule.

What kind of nonconformity, you ask? Well, today's example happens to be that QA released a product where a signature was required acknowledging QC acceptance, but where QC had not in fact accepted the product. The product did not meet specification, and QC therefore did not sign off, but QA released the product anyway. I can't tell you why, and I'm not sure the 'why' matters. I would expect that 'normal' practice would begin with a nonconformance filing and investigation, and the 'why' would become apparent. Maybe QA simply missed the lack of signature. Maybe there was some sort of management override (undocumented -- and therefore needing documentation). But in our system, we can't even file the acknowledgement of the nonconformance.

Other examples: QA releasing a label despite a significant error in its contents; QA miscounting labels, creating label reconciliation problems; QA altering a significant conclusion of a fully-executed document before applying QA signature (and not informing the approval committee or having the document re-executed); QA failing to keep up to date on required training, or executing procedures without training.

It's garden variety stuff, mostly, and my question is whether it is bog-standard practice among QA professionals to disallow the filing of nonconformances involving QA.

m

 

Malius:

In a utopia everyone would follow the rules/procedures and there would be no need for a non-conformance. However, I have never worked in a utopia so I cannot relate. I have however, experienced similar situations similar to what you described and my perspective is this. Regardless of what department errors occur in those areas have a responsibility to correct them and function within the limits of the QMS. If associates are knowingly and willingly violating the process maybe it is a personnel issue and not a system issue and should be dealt with accordingly. The often unpleasant part of this is that it sounds like it needs to be confronted in a management group setting and concerns need to be aired. It is usually an uncomfortable situation but until the instances in question are confronted there is little chance things will be improved. Many times in my experience what we think is a QMS issue is really a management issue and those issues need to be managed.

Jennifer Kirley has it right. A process is a process and each process needs to be held accountable.