15 replies to this topic

[matthewcc]

Hello all, I have a question about a critical control point (CCP)/preventive control that was missed during a packaging run.

 

To be clear, this was a metal detection CCP, and two of the three test pieces were rejected properly. The third was not. In lieu of repairing the metal detector, confirming all three test pieces were rejected, and re-running all product through the metal detector, we inspected screws and bolts on the packaging line to see whether any were missing. Seeing none missing, we released the product.

 

1. Is it acceptable to handle a missed CCP like this? That is, is it acceptable that, in place of repairing the metal detector, confirming all three test pieces were rejected, and re-running all product through the metal detector, we had a different way of determining that no metal pieces were in the product?

 

2. If it is acceptable, then don't we need some comprehensive list of all screws we need to check for, and verify that's done, instead of simply saying we examined the machines and saw no screws missing. In other words, how detailed do we need to get?

 

I'm wondering how alarmed I should be that this CCP was missed and we released product anyway with this documented check.

 

We are regulated in the United States under 21 CFR Part 111 and 117 and we have SQF and NSF GMP certifications. 

 

Thank you,

Matthew

[SQFconsultant]

is that the procedure that your written documentation on metal detecting calls for?

 

But, then, what happens every time you run a check on the test piece and it fails to trip - do you have to do all this stuff instead of what the standards call for?

 

As an Auditor if i saw this as the correction, I' compare what was done vs. what the written documentation says is to be done and act accordingly - something tells me this would not be an acceptable thng considering that metal may possibly be introduced - then making sure all the bolts and screws are in place would not matter.

[matthewcc]

No, this is not what our procedure and written documentation on metal detecting call for.

[kconf]

I am lost. Where does checking nuts and screws come from?

[kconf]

And what was the third piece for? SS, Fe, NFe? 

[matthewcc]

Checking the screws and bolts on the packaging line was to verify that none were missing in lieu of meeting the CCP requirements; it was to verify that there were no metal pieces in the product. It was an attempt to perform the same function as the metal detector to verify that there was no foreign object metal in the product bottles.

 

I am lost. Where does checking nuts and screws come from?

[matthewcc]

The blue ferrous card 1.0 mm Fe passed through undetected.

 

And what was the third piece for? SS, Fe, NFe? 

[kconf]

IDK man. You are in trouble zone. 

[GMO]

Not acceptable.  Not acceptable at all.  But now I will get into split personality mode. 

 

In the heat of the moment, that was a stupid thing to do by whomever did it.  A metal detector doesn't just check for missing nuts and bolts but metal on metal swarf, damaged engineer equipment, something that has come in from an ingredient, deliberate contamination with metal, etc etc.

 

So far so not good.

 

BUT, you do have an upside.  The source of ferrous metal in your product, if it's there is unlikely to be your factory.  Most of your factory is stainless steel right?  So while it's not acceptable IN ANY WAY, the risk of harming someone is pretty low unless you have an ingredient which is possible to have some metal in there (e.g. root vegetables sometimes have bits picked up in the field but let's face it, stones will do as much harm.)

 

Does all of that make it right?  Nope, not even slightly.  And right now, I'd be having some bloody firm discussions with senior leaders on rechecking it (if it can be put back on hold) or even recalling it...  

But honestly, behind the closed doors of my mind, I'm not all that worried about that incident because if there is metal in your product then most metals you have will be detectable, what I'd be worried about is how it happened that the easiest to detect metal wasn't detected and what would happen if you don't call it back next time.  Lastly it's worth considering what it would do to your credibility if you don't kick up a mahooosive fuss about it. I've been in similar situations to this (not failing to check a CCP but leaving a product out where there was an incident I felt to be low risk), no explanation I gave satisfied people that it was low risk. We had zero complaints but still people felt I'd made the wrong decision and in some ways they're not wrong. I'd ruined my image in their eyes.

 

So with all of that, be careful of your own reputation and risk of regulatory issues but I don't genuinely believe there is huge food safety risk.

Edited by GMO, Yesterday, 06:25 PM.

[SHQuality]

1. By checking the screws and bolts on the packaging line, all you have proven is that the packaging line did not have any missing pieces that could cause a product contamination.

 

The problem is that it was one of the test pieces that was not properly ejected. That means the metal detector was not working properly and therefore the machine was unable to properly dispose of all metal contamination if it happened to be caused by a contamination on the raw materials before they entered your packaging process or if there was a failure of food defence that resulted in intentional contamination or anything else that was not missing screws and bolts.

 

This was definitely not the right course of action. The machine should've been stopped and all products produced should've been rechecked after the machine was repaired and shown to correctly eject all test pieces.

 

2. You need to have a list that explains in some detail how a check should be performed. I've seen instances of people claiming that a cleanliness check on a machine was performed, only to later be told that an operator merely checked it off the list without checking properly and I've been told a piece of metal actually came from a supplier's own machine from a place they previously had failed to check. You need a checklist that is regularly verified and validated to ensure the check happens the same way every time, no matter who performs the check.

---

I would suggest you take immediate action to repair that metal detector and run all the products that were produced while the machine was defective through a detector again to be on the safe side.

[GMO]

 

I would suggest you take immediate action to repair that metal detector and run all the products that were produced while the machine was defective through a detector again to be on the safe side.

 

I don't know if I'm making assumptions but I get the feeling that things are no longer on site...

[SHQuality]

I don't know if I'm making assumptions but I get the feeling that things are no longer on site...

So then they have to decide whether to recall or take the risk of running the business into the ground if a problem is found and it turns out they didn't do anything about it in time. 

 

As I'm reading it, they're still producing with a faulty detector so at least some of the product would still be onsite.

[GMO]

So then they have to decide whether to recall or take the risk of running the business into the ground if a problem is found and it turns out they didn't do anything about it in time. 

 

As I'm reading it, they're still producing with a faulty detector so at least some of the product would still be onsite.

 

Yep.  That's the risk.  ANYTHING found and it will be a serious regulatory risk so while I genuinely think the actual risk of something getting to a consumer is really low, for that reason alone, on balance, I'd recall.  Depends on the product as well.  Some suppliers stall for so long that the product goes out of date.  We know you're doing it and we see you guys...

 

Not sure on the latter, perhaps Matthew you could elaborate?

[matthewcc]

The metal detector was fixed immediately after the lot, and there is a more sensitive metal detection that occurs in the process prior to the packaging.

 

Yep.  That's the risk.  ANYTHING found and it will be a serious regulatory risk so while I genuinely think the actual risk of something getting to a consumer is really low, for that reason alone, on balance, I'd recall.  Depends on the product as well.  Some suppliers stall for so long that the product goes out of date.  We know you're doing it and we see you guys...

 

Not sure on the latter, perhaps Matthew you could elaborate?

[SHQuality]

 

The metal detector was fixed immediately after the lot, and there is a more sensitive metal detection that occurs in the process prior to the packaging.

 

Now that is completely new information.

 

If you already know the product was good when it went through the first metal detector, then it would be much more acceptable to check the machine for missing pieces in lieu of checking those packs again; still not great, but at least defensible.

[jcieslowski]

The more I read of this thread, the more questions I have.

 

1. Why do you have two metal detectors in a row?

2. What does your CCP violation procedure say?  Did you do that?  "Say what you do, do what you say, prove it" = fundamental of a food safety program 101.

3. If the metal detector was fixed immediately to detect the ferrous metal, then why didn't you just re-run the product through said fixed metal detector?

 

All these questions are sort of just information because we don't know your setup.  But I can say that in no way would I ever say it's ok to release product that did not 'pass' CCPS.  Your CCP was non-functioning and thus, did not 'pass' your product.  That's a no go for me.