10 replies to this topic

[Spiceman]

I have talked myself into a circle regarding validation (again).

 

Which programs need to be validated besides:

 

Process Controls - metal detection (our only CCP)

Sanitation - verification the cleaning is procedures are effective

Allergen Management - food contact surfaces

 

 

 

Do all GMP's? Some?

 

Calibration

Environmental Monitoring

Shelf Life testing - our shelf life is a quality issue not a food safety issue

Pest Control

Personnel Practices

Training

 

 

[Spiceman]

 Example.xlsx   13.95KB   261 downloadsHere are examples of what we are currently doing.  

 

 

[Charles.C]

I have talked myself into a circle regarding validation (again).

 

Which programs need to be validated besides:

 

Process Controls - metal detection (our only CCP)

Sanitation - verification the cleaning is procedures are effective

Allergen Management - food contact surfaces

 

 

 

Do all GMP's? Some?

 

Calibration

Environmental Monitoring

Shelf Life testing - our shelf life is a quality issue not a food safety issue

Pest Control

Personnel Practices

Training

Hi Buddha,

 

I'm not surprised. SQF periodically generate "new" Circles. "Validation" is SQF's equivalent to BRC's equally infamous "risk-based".

 

Can compare ver 8.1 to ver 9 -

 

Ver 8.1

2.5  SQF System Verification  

2.5.1  Validation & Effectiveness (Mandatory)

2.5.1.1  The methods, responsibility and criteria for ensuring the effectiveness of all applicable elements of the SQF Program shall be documented and implemented.  The methods applied shall ensure that:

i.  Good Manufacturing Practices are confirmed to ensure they achieve the required result; 

ii.  Critical food safety limits are validated, and re-validated annually;

iii.  Changes to the processes or procedures are assessed to ensure controls are still effective; and 

iv.  All applicable elements of the SQF Program are implemented and effective. 

2.5.1.2  Records of all validation activities shall be maintained.

---------------------------------------------------------------------------------------------------------------

Ver 9

2.5   SQF System Verification

2.5.1   Validation and Effectiveness (Mandatory)

2.5.1.1    The methods, responsibility, and criteria for ensuring the effectiveness of all applicable elements of the SQF Program shall be documented and implemented. The methods applied shall validate that:

i.  Good Manufacturing Practices are confirmed to ensure they achieve the required results;

ii.  Critical food safety limits are reviewed annually and re-validated or justified by regulatory standards when changes occur; and

iii.  Changes to the processes or procedures are assessed to ensure the controls are still effective.

Records of all validation activities shall be maintained.

 

 

 

The first "red" looks like a backtrack to  SQF's "Good Old Days" which, after significant auditorial confusion,  ultimately (ca 2016) led to this Statement appearing on the SQF website -

 

The term “validation” specifically applies to control limits and requires scientific analysis to demonstrate that control limits are effective. Based on advice from our stakeholders, we [have] reworded the requirement for pre-requisite programs to state that the PRPs be “confirmed" to ensure they achieve the required result rather” than “validated.” The intent is to clarify the intent of the Code so that the effectiveness of the PRP is being met.
{my" reds"]

 

I'm not sure how you validate a confirmation of an "ensurance" in a Codex context.

I anticipate that the new (ii) adjusts a requirement  which was already being politely ignored.

[kingstudruler1]

I guess i was taught a long time ago that thedefinition of validation was "proving what you are doing is effective to reach the intended purpose.  I never really thought the HACCP had exclusive rights to the term.    Although, I recognize that the HACCP validation is more critical and usually more in-depth.    

 

 

I think its a good idea to prove that all of your programs and / or gmps are "effective".   If they weren't, I would want to improve them.  I have created a table nearly identical to your example to cover this.    To me its good way to do document the methods, responsibilities, and frequencies of verification and validation / "conform to ensure effectiveness"  tasks.   I include everything / as much as I can.  

[Cathy]

I recommend validation of all programs for SQF. Do as much as you can but do not over think it. Verification - did you do what you planned to do? Validation - did it work? The "easiest" ways to validate are to review data and look for trends and to conduct internal audits.  

[Charles.C]

I recommend validation of all programs for SQF. Do as much as you can but do not over think it. Verification - did you do what you planned to do? Validation - did it work? The "easiest" ways to validate are to review data and look for trends and to conduct internal audits.  

Hi Cathy,

 

I suggest you have a look at Codex/FSIS's interpretation of Validation. Chronology for example.

[Cathy]

Agree, Charles - there is more to it.  I often start with a simple approach when people are frustrated. Then, we improve. Validation is frustrating for many as there are often disagreements as to how much information is enough. It is important to have some - to get started. 

[Charles.C]

Agree, Charles - there is more to it.  I often start with a simple approach when people are frustrated. Then, we improve. Validation is frustrating for many as there are often disagreements as to how much information is enough. It is important to have some - to get started. 

Hi Cathy,

 

The sadness as IMO well illustrated by Codex/FSIS is that this topic does not have to be a mental conundrum.

I fear that for SQF, it's simply become like a dog with a bone.

[Scampi]

SQF is like a dog with a bone with  a lot of things

 

They should be using a thesaurus when the write the code or at the very least a dictionary.  v9 is asking for label reconciliation, which is damn near impossible for most----but the guidance et all is showing they're actually looking for a VERIFICATION that the correct label is being applied to the correct product........apparently they lack a basic understanding of English  

 

According to SQF Appendix for SQF 9.0 Manufacturing, the term Validation refers to an element of verification focused on collecting and evaluating scientific and technical information to determine if the HACCP food safety (or quality) plan is properly implemented, will effectively control the hazards (CODEX).

The goals of validation are to determine the effectiveness of a food safety plan for controlling identified hazards. Validation evidence can come from two main source categories -internal and external sources.

Internal validation sources:

• Food Safety trends and data analysis from in-house data (e.g. GMP trending, Microbial trending analysis etc.)
• Internal validation studies (e.g. allergen cleaning validation, Metal Detector validation)

External validation sources

• Food Safety Critical Limits (typically from local and internationally recognized sources, e.g. government or regulatory website, industry-related guidance documents)
• Validation studies data (from industry-related documents, supplier’s validation report studies)
• Peer-reviewed scientific literature
• Mathematical modelling
•  

https://sfpmfoodcons...n-requirements/

[Cathy]

This may be a topic for a different thread. Guidance on label reconciliation indicates they really do want inventory counts of labels.

https://www.sqfi.com...iation.docx.pdf

 

Label errors have been the cause of too many recalls and so reconciliation efforts are appropriate. The code doesn't mandate perfection but it expects control. An explanation of  a system with controls sufficient for someone to notice a problem -and to investigate that problem, may be enough.  If you loose a label to waste that may not be an issue.  If a whole roll goes missing - that's another story. 

[Spiceman]

Thank you everyone for all the information. Very helpful!