16 replies to this topic

[oxkjs1]

We are working on becoming FSSC 22000 certified and we are drafting our food safety procedures.  Many of you have working under ISO 22000 for some time. Can you help me clarify our understanding of the following specific question?

 

If a product violates a CCP limit, but the product still meets all final product specifications (liquid product) can it be released for sale to its intended use as food or feed? 

 

Or is it required to manage the product in accordance to clause 8.9.4.3 (product must be reprocessed, redirected for other use, or destroyed?)

 

The reason we ask is because clause 8.9.4.3 clearly begins with the statement:   "products that are not acceptable for release shall be:"

 

Thanks all for your support and experience!

 

Karen

[olenazh]

Hi Karen. No, I don't think you can release that product - not only because it's not meeting GFSI requirement, but it also would not be safe for public health, regardless what your CCP is (as it is a CRITICAL control point). Though it might not be a food safety reason (e.g. you're releasing a product with pH below lower limit which does not permit bacteria growth), but still it would be quality issue - which could lead to other issues (e.g. loss of revenue or reputation).

[oxkjs1]

Hi Karen. No, I don't think you can release that product - not only because it's not meeting GFSI requirement, but it also would not be safe for public health, regardless what your CCP is (as it is a CRITICAL control point). Though it might not be a food safety reason (e.g. you're releasing a product with pH below lower limit which does not permit bacteria growth), but still it would be quality issue - which could lead to other issues (e.g. loss of revenue or reputation).

 

 

Thank you so much for your reply.  Just to be sure I have explained our understanding well enough: 

 

8.9.4.2 states:

Each lot of products affected by the nonconformity shall be evaluated. 

Products affected by failure to remain within critical limits at CCPs shall not be released by shall be handled in accordance to 8.9.4.3

 

8.9.4.3 Disposition of nonconforming products

Products that are not acceptable for release shall be:

 

This seems to indicate that if a CCP limit is exceeded, the product lot can be evaluated to determine if it was affected by the CCP limit exceedance.  If the product is not affected by exceedance, then it can be released. 

 

We understand this to mean that only if the product was evaluated and found to be affected by the failure to remain within the critical limits - then 8.9.4.3 applies.

 

Thanks for bearing with me!

Karen

[olenazh]

Of course, it will be your decision after evaluation. That was just my opinion, based on experience - let's see though what other experts suggest.

[JensV]

Hi Karen,

 

In a "good" HACCP plan, when a CCP (= Critical Control point) is exceeded, you can no longer guarantee food safety.

 

So when a CCP has been exceeded, it's in my opinion impossible you can still release the concerning product as it is.

 

Edit: so I agree with olenazh, however with a CCP it should be a food safety reason.

Edited by JensV, 11 August 2022 - 02:17 PM.

[oxkjs1]

Thanks for the replies and the insight.  When we have our external audit, I will ask this specific question as well for curiosity.   In the meantime, do you agree that a good way to handle this is to advise my organization to designate a control point as an OPRP when there is strong agreement that the product could still be safe for food if the product is fully evaluated and meets all specifications?

 

As an aside:  my personal opinion is that ISO 22000 creates unnecessary confusion and work by requiring us to distinguish between an OPRP and a CCP.  I think everything should be a CCP and there should be a required protocol to always evaluate the impact if a limit was exceeded.  In some cases a standard answer may be to release the product under certain circumstances and in other cases the product must be destroyed, reprocessed or reclassified. 

[olenazh]

There were some topics of your interest before on this forum, look into these:

https://www.ifsqn.co...s-oprp-in-fssc/

https://www.ifsqn.co...-contamination/

https://www.ifsqn.co...n-ccp-prp-oprp/

https://www.ifsqn.co...-iso-220002018/

[Brothbro]

Thanks for the replies and the insight.  When we have our external audit, I will ask this specific question as well for curiosity.   In the meantime, do you agree that a good way to handle this is to advise my organization to designate a control point as an OPRP when there is strong agreement that the product could still be safe for food if the product is fully evaluated and meets all specifications?

 

As an aside:  my personal opinion is that ISO 22000 creates unnecessary confusion and work by requiring us to distinguish between an OPRP and a CCP.  I think everything should be a CCP and there should be a required protocol to always evaluate the impact if a limit was exceeded.  In some cases a standard answer may be to release the product under certain circumstances and in other cases the product must be destroyed, reprocessed or reclassified. 

 

Any advice on switching from a CCP to a OPRP would be difficult to answer without very detailed knowledge of your process, facility, and HACCP plan. Your hazard analysis would have determined that without this CCP, a food safety risk would exist uncontrolled in your product. I would suggest that moving from a CCP to OPRP shouldn't be based on "strong agreement" amongst your team, but be based on a formal and documented hazard analysis by your HACCP team. A HACCP consultant could be useful to get a fresh pair of eyes on your plan. Situations like this can make it difficult for companies to re-evaluation their plans themselves, because of the bias that leans towards clearing a particular hold vs benefitting the process.

 

Could the product be re-processed to properly apply the previously failed CCP? Sometimes CCPs are still effective even when applied out of order. For example, running product through a metal detector later than intended. Basically, I don't think this product could be released either without some kind of appropriate reprocessing because it violates your HACCP plan. Whether any appropriate reprocessing exists depends on what happened.

Edited by Brothbro, 11 August 2022 - 04:06 PM.

[oxkjs1]

Any advice on switching from a CCP to a OPRP would be difficult to answer without very detailed knowledge of your process, facility, and HACCP plan. Your hazard analysis would have determined that without this CCP, a food safety risk would exist uncontrolled in your product. I would suggest that moving from a CCP to OPRP shouldn't be based on "strong agreement" amongst your team, but be based on a formal and documented hazard analysis by your HACCP team. A HACCP consultant could be useful to get a fresh pair of eyes on your plan. Situations like this can make it difficult for companies to re-evaluation their plans themselves, because of the bias that leans towards clearing a particular hold vs benefitting the process.

 

Could the product be re-processed to properly apply the previously failed CCP? Sometimes CCPs are still effective even when applied out of order. For example, running product through a metal detector later than intended. Basically, I don't think this product could be released either without some kind of appropriate reprocessing because it violates your HACCP plan. Whether any appropriate reprocessing exists depends on what happened.

 

Thanks so much for your feedback and I feel I understand your point.   Our decision tree drives our decision to designate CCP or OPRP as it is based on many factors such as the production process step, the type of hazard, measurement or observation, the hazard's behavior in the process and final product / end-use application, and so on.

 

I meant to explain that in a situation where a control measure could be argued to meet both a CCP and an OPRP classification (using our decision tree) but the product can literally be fully evaluated to confirm that it conforms to all its specifications and there is no legitimate evidence or concern for presence of the hazard in the final product, then it could be beneficial to classify this as an OPRP.  This would allow evaluation of the product for the hazard so that the product can be released in accordance to criteria in 8.9.4.2.

 

It seems to me that if a product can be fully analyzed for the hazard - then the control measure is not a CCP regardless because one can prove that the product is safe after the critical limit was exceeded.  For example, a homogeneous liquid can be analyzed for a specific heavy metal contaminant to a ppb or ppt level.  This distinction, to me, is one of the easiest ways to differentiate between a CCPs and OPRP when there is doubt....   I can't imagine the FDA would not let me sell a product if I can prove the product is safe... so I can't imagine the standard would prohibit me from doing so?

 

However, it is much easier to demonstrate compliance to the standard's requirements for release by designating the control measure as an OPRP for a situation like this.  Otherwise you are left explaining to an auditor why you released a product that violated a CCP control limit.

 

Thanks again for your time!  Great discussion! (I think!) 

 

Karen

[Charles.C]

Hi Karen,

 

It is undeniable IMO that the introduction of OPRP by ISO has generated, so far,17 years of (continuing) argument/confusion.

 

The 2018 version of ISO22000 has attempted to reduce some of this confusion by emphasising  that both  CCPs and OPRPs are only associated with Significant (Safety) Hazards.

 

This would therefore seem to decisively  answer the basic question in your OP.

 

Without knowing the specifics, one can only speculate but it is possible that your present dilemma (eg haccp vs Specification)  may indicate the necessity for a revision of your haccp plan (and/or Specification).

[Bo16]

We have put in place the requirements if a CCP fails in our process, for example if the pressure fails cross the filtration membrane, the product must be re-filtered. All our controls have a pathway for correction/approval if there is a failure.  Yes, sometimes it is more testing, sometimes it is more processing.  In all instances a Root Cause Analysis is performed and CAPAs are put inplace/evaluated prior to QA release, by Sr. Manager only.

[oxkjs1]

We have put in place the requirements if a CCP fails in our process, for example if the pressure fails cross the filtration membrane, the product must be re-filtered. All our controls have a pathway for correction/approval if there is a failure.  Yes, sometimes it is more testing, sometimes it is more processing.  In all instances a Root Cause Analysis is performed and CAPAs are put inplace/evaluated prior to QA release, by Sr. Manager only.

 

Thanks for your reply. To be sure I am not putting words in your mouth, your statement "sometimes it is more testing" means that, in your opinion, it is possible to release a product if a CCP has failed as long as the "additional testing" can confirm conclusively that the hazard of concern is not present?

 

Good points about the additional requirements for RCA, CAPA and  QA Sr. Manager involved in the review /approval of the outcomes / product release.  Completely agree!

 

Thanks again for taking time to share your experience! 

[Charles.C]

Thanks for your reply. To be sure I am not putting words in your mouth, your statement "sometimes it is more testing" means that, in your opinion, it is possible to release a product if a CCP has failed as long as the "additional testing" can confirm conclusively that the hazard of concern is not present?

 

Good points about the additional requirements for RCA, CAPA and  QA Sr. Manager involved in the review /approval of the outcomes / product release.  Completely agree!

 

Thanks again for taking time to share your experience! 

Hi Karen,

 

This CCP topic and the related text in the ISO22000 standard, eg section 8.9,  is a (highly) detailed, generic, enunciation of the fundamental HACCP Plan (eg Codex) step relating to Corrective Actions following a sampling-based failure of a CCP's Critical Limit.

 

IMEX a HACCP Plan's "Corrective Actions" typically include/mirror the ISO's generic comments that the sampled "lot" may be re-evaluated prior to final disposition. (As an aside, the specific accept/reject criteria subsequently utilised in any re-evaluation may be contentious, eg sampling size/defect tolerances).

 

Returning to the original OP I suggest that, assuming the HACCP Plan has been "properly" validated, your Product "Specifications" are/were either incorrect/questionable or incomplete.

[MOURADTALBI]

Hi Karen, 

When critical limits at CCPs are not met, affected products shall be identified and handled as

potentially unsafe products. 

You shall take action to prevent potentially unsafe products from entering the food

chain or from releasing , unless you  can demonstrate that:

    -  the food safety hazard of concern is reduced to the defined acceptable levels;   

    -  the food safety hazard of concern will be reduced to identified acceptable levels prior to entering

the food chain or to  releasing ; or

    -  the product still meets the defined acceptable level(s) of the food safety hazard(s) of concern despite

the nonconformity.

 

You shall retain products that have been identified as potentially unsafe under your  control

until the products have been evaluated and the disposition has been determined.

 

 

 

[oxkjs1]

I really appreciate this opportunity to discuss this topic with all you experienced people.  I just want to say that while I am new to FSSC 22000 and food safety, I am not new to compliance.  I am a 20+ year veteran who spent almost my whole career in quality and EHS management in a very complex chemical manufacturing plant prior to this current project. I have a lot of experience in ensuring that systems are compliant. 

 

I am inexperienced in food systems and I am trying to understand how the community applies this certain point about releasing product that has violated a CCP limit.

 

Let's assume my entire food safety management system is compliant, thorough, defensible and in place for many years (i.e. tested/audited)

 

The language in ISO 22000 clause 8.9.4.2 states:

"Products affected by failure to remain critical limits at CCPs shall not be released, but shall be handled in accordance with 8.9.4.3"

 

What is meant by 'affected' ?   If I can demonstrate definitely and defensibly that the safety and quality of the product was truly not affected by the failure after investigating the exceedance - and assume all other appropriate and required Food Safety protocols are compliant to regulatory and FSSC requirements - can it be released??   

 

With that said:  there is always regulatory/statutory codes, standard language, legal interpretations and then there is industry practice.   

 

According to ISO 22000 -  it seems to state that only product that has violated a OPRP action criteria can be released provided that you can demonstrate one of the conditions in clauses (a) - c) apply in 8.9.4.2. 

 

So, with that said:  What is industry practice when a CCP limit is exceeded under FSSC 22000?  Is it "never" released but always either reclassified, further processed or rejected?

 

That is what I am trying to understand.  Thank you for sticking with me on this.  I really am trying to benefit from everyone's experience.

Sincerely,
Karen

Edited by oxkjs1, 19 August 2022 - 01:59 PM.

[Bo16]

Sorry, if I made it sound like testing alone would release a failed Preventive Control.  Since Preventive control are monitored during processing, the failure would trigger additional processing.  Verification testing would be re-tested after corrective actions.

[Charles.C]

Hi Karen,

 

I think  "affected" In the current context means something like "involved". The original ISO Standard was IIRC in French which resulted in a few translation "specialties". This may be another.

 

IMO it is sometimes difficult to generalise the (unavoidably generic) ISO text in a practical HACCP way. Specific Corrective Actions may vary on a case-by-case basis.

 

I suggest an alternative approach which is occasionally recommended in HACCP  texts and I have found effective in industrial practice. Instead of trying to, perhaps, over-analyse ISO's textual nuances, it is maybe preferable to avoid hypotheticals by implementing what are known as "Operational Limits", ie "Action Limits". The latter act as a pre-emptive mechanism so as to prevent process  failure of the Critical Limits/Action Criteria.

Edited by Charles.C, 19 August 2022 - 10:12 PM.
edited