4 replies to this topic

[D-D]

Probably an old issue but I can't find other threads to the answer:

Here is our specific situation: We ensure that incoming raw materials such as chilli peppers are tested for illegal dyes (supplier provides a clean certificate or we test ourselves). If I use the Codex HACCP decision tree I would say:

Q1 Yes
Q2 Yes / Stop CCP
Q3 -
Q4 -

If I understand correctly this is not a CCP though, it is a prerequisite so I am struggling as to whether I am using the questions correctly. All this comes under the "Purchase" step of the flow diagram and the hazard identified would be "Contamination with illegal dyes" so maybe this is where I am not interpreting correctly...?

[Charles.C]

Dear D-D,

There are sooo many different ways of doing haccp.

For example many people do not use the routine Codex Q1-Q4 for ingredients but other systems do integrate ingredients with rest of the haccp development (from memory, don’t normally use decision trees myself).

Is this haccp for some particular standard or ?

The method used to guarantee the safety from illegal dyes is 50/50 debatable but probably not unacceptable although requires back up validation, eg supplier audit etc.

PRP programs do not obligatorily preclude related steps (eg ingredients receiving) from being included in the hazard analysis although may offer such an option. It basically comes back to the risk analysis.

IMEX “Purchasing” as a step is not typically included in a process hazard analysis although it may nonetheless be organised via a typical prerequisite program.

But perhaps you are following some specific haccp model ?.

Rgds / Charles.C

[D-D]

Hello Charles,

Hmmm...well it is a re-vamp of the HACCP plans ready for the upcoming BRC audit. Previously we had "HACCP Team opinion" on each step / hazard as to whether or not it was "significant" and needed further thought or not. Various auditors have told me that best practice is to do a proper Likelihood x Consequence risk assessement so I am trying that and then following up each point with the four Codex questions. It all looks very pretty as a nice colour coded spreadsheet and is not as complex as it might sound but I am just a bit stuck on the outcome of those four questions as so many can say "Yes" at Q2 but they are clearly GMP not CCP.

Thanks for your help as always.

[Charles.C]

Dear D-D,

The point about GMP / prerequisite programs is that, in general, they are supposed to make the hazards with which are associated "not reasonably likely to occur". Consequently they do not yield a significant hazard in most risk matrices. And not significant >>>> not CCP, by definition.

But note that there are 2 variations / opinions which can be found on this logic, ie -

In general, CCP determination involves identifying and characterizing the hazards, the control measures, and the processing steps where control is applied. Thus, it is imperative that the HACCP team revisit and re-evaluate the Hazard Analysis as well as the Flow Diagram.

If a significant hazard that is reasonably likely to occur is identified in the Hazard Analysis, it must be controlled somewhere in the food processing and handling system. Due to the rigorous HACCP requirements for setting appropriate CLs and monitoring, there are certain hazards in certain situations which are more appropriately controlled by prerequisite programs. This has been the subject of some debate and has led to two approaches in dealing with such hazards. While they differ philosophically, the end result of these two approaches is similar. *

Approach A: This approach is used by some HACCP auditors in the regulatory and private sector. It dictates that a hazard deemed significant in the Hazard Analysis must be controlled at a CCP. It then follows that a hazard which is, in fact, managed by a prerequisite program is considered not "reasonably likely to occur" because of the prerequisite program.

Approach B: In this approach, prevention and control of identified significant hazards can either be at a CCP or through rigorously applied prerequisite programs.

The Codex tree was i think designed specifically for process steps, not ingredients. But i may be wrong. (If you look in standard texts and on this forum you can see slightly modified (eg Q1-Q3) trees. But I think cfia adjust the appropriate answer format so as to conveniently maintain Q1-Q4, from memory),

Routine users of codex tree should know the answer to this one like the back of their hands.

Rgds / Charles.C

PS as example, refer this thread -

http://www.ifsqn.com...ndpost__p__6330

post#19 is similar to one textbook tree. It is posted in more pictorial form elsewhere on this forum.

[D-D]

The point about GMP / prerequisite programs is that, in general, they are supposed to make the hazards with which are associated "not reasonably likely to occur". Consequently they do not yield a significant hazard in most risk matrices. And not significant >>>> not CCP, by definition.

Dear Charles,

I think this is it! My risk assessment has a 1-5 score each of Likelihood and Consequence so multiplying those up, the Risk is 1-25. I only need to plug into the Codex questions the high-to-extreme risk outcomes (scoring 8-25) and just put comments next to every other entry on the list (1-6 for very low-to-moderate Risk) to indicate the GMP/PRP controls.

Many thanks.