Jump to content

  • Quick Navigation
Photo

7.4.2.3 - ISO 22000

Share this

  • You cannot start a new topic
  • Please log in to reply
24 replies to this topic
- - - - -

Zeeshan

    Grade - SIFSQN

  • IFSQN Senior
  • 499 posts
  • 224 thanks
25
Excellent

  • Pakistan
    Pakistan
  • Gender:Male
  • Location:Karachi, Pakistan
  • Interests:QMS, TQM, FSMS, HMS (Halal Management System), IMS (Integrated Management System), Training Programs Management, Performance Management

Posted 12 November 2010 - 07:30 AM

Dear all!

7.4.2.3 states that for each of the food safety hazards identified, the acceptable level of the food safety hazard in the end product shall be determined. Our products are commercially sterilized (retort) ready-to-eat meals. Whether we should state "zero acceptable level" for biological hazard in the end product during our hazard analysis working. What about microbial load related to raw material? Where and how should we address and justify that microbial load? Definitely during raw material inspection we should have some criteria/standard values. Whether standard requires to justify those values by referring to some external documents or in-house experimentation etc.?

Regards:
M.Zeeshan



Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

  • Earth
    Earth
  • Gender:Male
  • Interests:SF
    TV
    Movies

Posted 13 November 2010 - 11:58 AM

Dear Zeeshan,

The acceptable level you consider as appropriate to an RTE food (ie being safe to consume) is required to be validated with respect to your selected control measure(s) as per section 8.2.

Zero cfu/g of any pathogenic species such as salmonella is usually quite difficult to explicitly validate IMEX. :smile:

I presume yr retort process must be designed and should be already validated as giving a safe microbiological end product. Probably someone will have already provided actual data to support that.(typically something like a 12D process).

Do you have iso 22004 ?, this has a large section on this topic.
Another possible choice may be as specified by yr local regulatory requirements (assuming they exist)(see 22004).

I am interested as to what other users of ISO 22000 state also, and additionally if the end-product is not RTE, eg raw semi-finished. :smile:

Rgds / Charles.C


Kind Regards,

 

Charles.C


Thanked by 1 Member:

S.U.Siddiqui

    Grade - Active

  • IFSQN Associate
  • 24 posts
  • 5 thanks
0
Neutral

  • Pakistan
    Pakistan
  • Location:Pakistan

Posted 15 November 2010 - 04:51 AM

Dear Zeeshan,

Nice to see you as an active member. IMO when you say that your product is COMMERCIALLY STERILIZED then you have to look for the definition of commercial sterility? As far as I know, commercial sterility refers to absence of VIABLE (growable) microbes i.e. all living bacteria are destroyed (not spores). Hence you can say zero viable bacteria as the acceptable level in your end product.

As far as your raw material is concerned, you can NEVER use microbiology testing as the acceptability or unacceptability criteria. Reason is very obvious:
a) The results are not timely and it would take at least 12 hours to confirm the microbial safety of the received material. No one would wait that much in the receiving area before production.
b) Due to non-timely results they are not genuine i.e. You get the results in 12 hours and the microbial population of the raw material has changed in 12 hours.
c) You are OBVIOUSLY going to process that material to reduce the microbial load so what's the point in testing microbial criteria?

What I suggest for raw material that you do the process (washing with any food grade chemical like vegisan, in case of vegetables) and then validate the effectiveness of that chemical. This would ensure that WHATEVER the microbial count of the RAW food was you treated it in such a way that it reduced to acceptable level. In other cases, e.g. meat, pulses, rice etc. zero count in end product is enough justification for the auditor to accept that your food is safe.

Hope this would help.

Regards
Siraj



Thanked by 2 Members:

Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

  • Earth
    Earth
  • Gender:Male
  • Interests:SF
    TV
    Movies

Posted 15 November 2010 - 10:46 AM

Dear S.U.Siddiqui,

Nice to hear from you again also and thks yr input (Sorry to preempt yr answer Zeeshan). :smile:

Hence you can say zero viable bacteria as the acceptable level in your end product.


I admire your creativity. :thumbup: Can you validate yr above suggestion with a link somewhere? I hv never seen this format previously. "Commercially sterile" is certainly a popular phrase, especially for lab. people, but I hv not previously seen this used as a viable "acceptable level" in ISO 22000 publications ?. Actually, I had thought that a 12D treatment was supposed to "destroy" spores also [up to the targeted degree anyway] ?.

a) The results are not timely and it would take at least 12 hours to confirm the microbial safety of the received material. No one would wait that much in the receiving area before production.


I agree with yr comment however some people buy frozen raw material. Zeeshan no idea.

vegisan


The supplier website i found stated -

This product provides outstanding removal of micro-organisms.


Do you have any more quantitative knowledge of its ability?, I believe most "sanitisers" cannot be validated as consistently achieving better than 2Log reduction, so it then depends on yr raw material input level although I know there are some super-exceptions of which this may be one. Of course it also depends on yr opinion as to minimum "acceptable levels" again.

It seeemed to me that for products like RTE fresh vegetables with no "elimination-power" process stage, a statement something like the typical end product spec. "pathogen X Not detected in 5x25g" might suffice from a legislatory point of view, but this would presumably not be appropriate for cooked / retorted foods. Or ?

Rgds / Charles.C


Added - After a little more thought, “zero tolerance” is probably the most offically corrrect terminology for RTE cooked foods with respect to pathogens like Salmonella, E.coli O157 with other texts appropriate to the species. Such statements are typically practically interpreted in the style mentioned in my previous post (the exact form may depend on the local situation) and can be readily validated by reference to official literature ( again, a range of official numerical interpretations exists, sometimes even within the same country). I am not sure regarding any equivalent term for canned / retorted products (not my area). Perhaps S.U. Siddiqui is right in which case it would be nice if one of the, hopefully, many ISO 22000 certified producers of retorted products reading this thread could confirm and additionally offer a link somewhere.

It should be noted that ISO 22004 also offers the possibility of using defining routes via concepts like FSO. The difficulty with these is that they are likely to ultimately require legally unacceptable statements, ie there is simply no legally acceptable numerical level for Salmonella in RTE cooked products > zero.

Kind Regards,

 

Charles.C


Thanked by 1 Member:

Tony-C

    Grade - FIFSQN

  • IFSQN Fellow
  • 4,223 posts
  • 1288 thanks
608
Excellent

  • United Kingdom
    United Kingdom
  • Gender:Male
  • Location:World
  • Interests:My main interests are sports particularly football, pool, scuba diving, skiing and ten pin bowling.

Posted 16 November 2010 - 02:06 PM

Dear Zeeshan,

Nice to see you as an active member. IMO when you say that your product is COMMERCIALLY STERILIZED then you have to look for the definition of commercial sterility? As far as I know, commercial sterility refers to absence of VIABLE (growable) microbes i.e. all living bacteria are destroyed (not spores). Hence you can say zero viable bacteria as the acceptable level in your end product.

As far as your raw material is concerned, you can NEVER use microbiology testing as the acceptability or unacceptability criteria. Reason is very obvious:
a) The results are not timely and it would take at least 12 hours to confirm the microbial safety of the received material. No one would wait that much in the receiving area before production.
b) Due to non-timely results they are not genuine i.e. You get the results in 12 hours and the microbial population of the raw material has changed in 12 hours.
c) You are OBVIOUSLY going to process that material to reduce the microbial load so what's the point in testing microbial criteria?

Hope this would help.

Regards
Siraj


Have to disagree with a fair few of these points. Firstly commercial sterility will inlcude spores if they are capable of growing in the product.

Attached File  USDA Commercial Sterility.jpg   97.78KB   60 downloads

Secondly it is quite common to hold and release ingredients prior to processing, especially when microbiological growth can be controlled.

Thirdly testing for spores can verify that your ingredient mix is within the acceptable criteria that you have validated your sterilisation process against. It can also identify poor quality ingredients.

Attached File  USDA Standard.jpg   95.13KB   49 downloads

Regards,

Tony


Thanked by 1 Member:

Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

  • Earth
    Earth
  • Gender:Male
  • Interests:SF
    TV
    Movies

Posted 16 November 2010 - 06:01 PM

Dear Tony,

Interesting comments. :smile:

Are you saying that you think there is no appropriate answer in the ISO “acceptable level” format or that details of Zeeshan’s product / system are necessary to analyse any further.? Or both ?

I suppose from the ISO 22000 perspective, it is strictly only safety which is relevant, even if spoilage is also a factor in the heat process calculation for some (commercially sterile) scenarios.

After a little googling, it did seem to me that for low acid foods, an answer could be formulated in terms of (probability) C.botulinum < XYZ / 12D cook, aka a commercially sterile product (FSIS) but I’m only speculating.

One would think that the processor would normally hv to sort / validate all these basic factors out / be approved before entering the market (maybe not yet).?

BTW, the first attachment contains an obvious temp. typo.

Rgds / Charles.C


Kind Regards,

 

Charles.C


Tony-C

    Grade - FIFSQN

  • IFSQN Fellow
  • 4,223 posts
  • 1288 thanks
608
Excellent

  • United Kingdom
    United Kingdom
  • Gender:Male
  • Location:World
  • Interests:My main interests are sports particularly football, pool, scuba diving, skiing and ten pin bowling.

Posted 16 November 2010 - 08:10 PM

Hi Charles,

Interesting comments.

:thumbup: Thank you

Are you saying that you think there is no appropriate answer in the ISO “acceptable level” format or that details of Zeeshan’s product / system are necessary to analyse any further.? Or both ?

Not sure why it is not clear what I'm saying, read again and let me know.

I suppose from the ISO 22000 perspective, it is strictly only safety which is relevant, even if spoilage is also a factor in the heat process calculation for some (commercially sterile) scenarios.

True but the comments were why would you bother checking raw ingredients and as an add on to the point that if spore levels were excessive that it could be a problem.

After a little googling, it did seem to me that for low acid foods, an answer could be formulated in terms of (probability) C.botulinum < XYZ / 12D cook, aka a commercially sterile product (FSIS) but I’m only speculating.

Speculation and Googling seems like a double gamble to me !

One would think that the processor would normally hv to sort / validate all these basic factors out / be approved before entering the market (maybe not yet).?

:thumbup: I would think so too.

BTW, the first attachment contains an obvious temp. typo.

:rolleyes: Yes from the USDA not me.

Regards,

Tony

Edited by Tony-C, 16 November 2010 - 08:11 PM.


Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

  • Earth
    Earth
  • Gender:Male
  • Interests:SF
    TV
    Movies

Posted 16 November 2010 - 10:21 PM

Dear Tony,

Not sure why it is not clear what I'm saying, read again and let me know


Probably inadequate knowledge base on my side. I daresay Zeeshan will understand better. :smile:

Rgds / Charles.C

Kind Regards,

 

Charles.C


Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

  • Earth
    Earth
  • Gender:Male
  • Interests:SF
    TV
    Movies

Posted 17 November 2010 - 03:08 PM

Dear Zeeshan,

Here is an extract taken from a French study on production of pasteurised tomato ketchup analysed as per ISO 22000.
The target species / numbers will surely not conform to your own case but it may hopefully give you an idea of a possible format.It is quite likely IMO that you may have a regulatory requirement.
I think most of the meanings in English are fairly obvious (the sections under CCP are from left to right - Control Measure, Acceptable level, Critical Limit [I guess] .... . If any problem, the Google translator is quite good for short phrases as here.)

Attached File  pasteurisation ketchup.png   30.84KB   91 downloads

I should repeat that pasteurisation / sterilisation not my field plus your "ready meal" is undefined so far so the above is just an example.

Rgds / Charles.C


Kind Regards,

 

Charles.C


Thanked by 1 Member:

HPG

    Grade - MIFSQN

  • IFSQN Member
  • 99 posts
  • 30 thanks
1
Neutral

  • Indonesia
    Indonesia
  • Gender:Male
  • Location:Indonesia

Posted 24 August 2011 - 03:13 AM

Dear all,

If we have already done hazard identification, but some of identified hazard from raw material (ie. Listeria) do not have acceptable level in end product (based on regulatory), what should we do?
Do we still need to include those hazards in hazard identification?


Regards,
Hadi



Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

  • Earth
    Earth
  • Gender:Male
  • Interests:SF
    TV
    Movies

Posted 24 August 2011 - 06:18 AM

Dear all,

If we have already done hazard identification, but some of identified hazard from raw material (ie. Listeria) do not have acceptable level in end product (based on regulatory), what should we do?
Do we still need to include those hazards in hazard identification?


Regards,
Hadi

Dear hadi,

I presume you are referring to a regulatory "zero tolerance (ZT)" hazard.
If so, ISO 22004, sec.7.4.2 answers yr last question by "yes".

For yr previous question, it may depend on the product/process/hazard/regulatory statement.The simplest input for the end product "acceptable level" is obviously given by the regulatory definition. (Assuming that the latter is not nonsense :smile:) . However a process objective (eg 5D) also seems to be an ISO-acceptable criterion in some cases (perhaps only for a regulatory requirement[?]) as per post #9 (in general, these tend to be a little more debatable for a specific case since the final level for the hazard will depend on the starting concentration.)(But one can equally argue regarding sampling limitations for regulatory product [un]detectability requirements).

You may need to specify yr situation for a more meaningful answer.

Some regulatory, product-based, examples I hv seen for ZT microbiological pathogens (not all ISO 22000) -

Maximum level of X is zero.
X is absent in 25g.
X is absent (n=5, c=0, sample size 25g)
X is absent in 5x25g samples.
The last 3 but "absent" replaced by "not detected"
5D, 6D respectively.

No doubt there are many others, i daresay ISO accepts all of them if validatable (somewhere). :rolleyes:

Rgds / Charles.C

Kind Regards,

 

Charles.C


HPG

    Grade - MIFSQN

  • IFSQN Member
  • 99 posts
  • 30 thanks
1
Neutral

  • Indonesia
    Indonesia
  • Gender:Male
  • Location:Indonesia

Posted 24 August 2011 - 06:48 AM

Dear Charles,
Thanks for your reply.

Yesterday, our company was audited for ISO 22000. The products are biscuits and wafers
The auditor stated that every hazard must be identified which I do agree.
In our regulatory (indonesia), for biscuit (end product), L. monocytogenes isn't one of microbiological hazard.
But, in our raw material (eg. cheese), one of identified microbiological hazard is L. mono.

I know that I should include L. mono in hazard identification. But, how about acceptable level?
Should I find another reference, for example RTE food?

And I want to ask another finding.
We include 7.4.2.3 (acceptance level in end product) in our raw material hazard identification, but the auditor didn't agree.
He said that we should use acceptable level from regulatory for each raw material.
For example :
S. aureus in end product : 100 cfu / g
S. aureus in raw material (for example : shortening) : 50 cfu/g
I wrote 100 cfu / g for S. aureus in "acceptable level in end product" column, but like I stated before, he said that we should wrote 50 cfu/g
based on raw material.
Which one do you think is the right one?

I attached the raw material hazard identification that we used.

Regards,
Hadi

Attached Files



Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

  • Earth
    Earth
  • Gender:Male
  • Interests:SF
    TV
    Movies

Posted 24 August 2011 - 07:04 AM

Dear HPG,

Regarding L.mono, yr question obviously relates to the process as well, eg after addition of the cheese, is there a process step which should / will "eliminate" any introduced L.mono ? Or is it added at the end (I'm guessing yes) ?

And what is the L.mono spec. for the cheese ? Absent ? :smile:
End product indefinitely shelf stable ? If so this probably reduces yr options to "one"
My guess will also not be too difficult to find another official regulation for such products (not my area) which inc. L.mono.

added -

from POV of EC, the decision (between "absent" and max 100cfu/g) seems to (logically) depend on whether the product is capable of supporting growth during it's shelf life and if so, how (validatably) much ?

eg - Attached File  EC - Reg1441_2007(1).pdf   96.96KB   103 downloads

Still reading the rest. :smile:

added Re S.aureus - I presume we are still talking about the biscuits, ie the next step for the end product is direct consumption.
If so, I suggest you refer to ISO 22004, sec.7.4.2, para.2. seems fairly definitive as per yr current procedure.
Do you hv ISO 22004 ? :biggrin:

Thks for the neat attachment, the method for use in 7.4.4 to determine CCP/OPRP etc looks a little vague. :smile: (did you give up the consultant's model discussed in an earlier thread ?)

Rgds / Charles.C


Kind Regards,

 

Charles.C


Thanked by 1 Member:
HPG

HPG

    Grade - MIFSQN

  • IFSQN Member
  • 99 posts
  • 30 thanks
1
Neutral

  • Indonesia
    Indonesia
  • Gender:Male
  • Location:Indonesia

Posted 24 August 2011 - 07:41 AM

Dear HPG,

Regarding L.mono, yr question obviously relates to the process as well, eg after addition of the cheese, is there a process step which should / will "eliminate" any introduced L.mono ? Or is it added at the end ?

And what is the L.mono spec. for the cheese ? Absent ? :smile:
End product indefinitely shelf stable ? If so this probably reduces yr options to "one"
My guess will also not be too difficult to find another official regulation for such products (not my area) which inc. L.mono.

Still reading the rest. :smile:

added Re S.aureus - I presume we are still talking about the biscuits, ie the next step for the end product is direct consumption.
If so, I suggest you refer to ISO 22004, sec.7.4.2, para.2. seems fairly definitive as per yr current procedure.
Do you hv ISO 22004 ? :biggrin:

Rgds / Charles.C


Dear Charles,
Thanks for your reply.

There is a step that can / should eliminate them : Baking with temperature more than 100 degree C
For cheese, L. mono spec is absent / 25g

Yes, I have ISO 22004. and I've already read it. And it's said that acceptable level means level in the end product
That's why I a little bit confused with the auditor's statement yesterday. (but yesterday I havent't read it, so I can't counter his statement :biggrin:)
Do you think should I ask him again (by phone, definitely) ? just to make sure the correct answer. :smile:

Regards,
Hadi


Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

  • Earth
    Earth
  • Gender:Male
  • Interests:SF
    TV
    Movies

Posted 24 August 2011 - 08:44 AM

Dear HPG,

Do you think should I ask him again (by phone, definitely) ? just to make sure the correct answer


Have to say that my first reaction to yr query was surprise that an auditor would be incorrect over such an item. How was the rest of the audit ??

I guess the option you refer rather depends on the anticipated overall result of the audit and the "atmosphere", eg friendly or tempestuous (I hv had both :smile: .)

If no specific objections as above and local protocol does not prohibit (?), I guess there should be no reason not to contact in view of some "claimed" uncertainty of understanding over the issue thereby getting a "polite" opportunity to refer to supporting references, etc.
(After all, you are paying for this privilege!)

Re L.mono, see my added attachment / comments previous post. As per yr comment, should be un-detectable in end product (ever checked? :smile: ).
(in above ref., see table paras 1.1 - 1.3)
I found the ref. via a more readable guide -
Attached File  RTE PHLS food guidelines.pdf   998.98KB   126 downloads

After yr baking info. the temptation is to select "absence" and avoid any chance of argument (assuming yr results are supportive!). But sometimes it is preferable to be conservative after considering the process / environment / shelf life / consumer type, the snag is that validation is then involved.

A search for cheese biscuits is recommended :smile: .

added - i hv one query myself - for the permanently zero tolerant species like salmonella, what statement did you select to put in the "acceptable level" column (presumably accepted by the auditor)?

Rgds / Charles.C

Kind Regards,

 

Charles.C


Thanked by 1 Member:
HPG

HPG

    Grade - MIFSQN

  • IFSQN Member
  • 99 posts
  • 30 thanks
1
Neutral

  • Indonesia
    Indonesia
  • Gender:Male
  • Location:Indonesia

Posted 25 August 2011 - 02:45 AM

Dear HPG,



Have to say that my first reaction to yr query was surprise that an auditor would be incorrect over such an item. How was the rest of the audit ??

I guess the option you refer rather depends on the anticipated overall result of the audit and the "atmosphere", eg friendly or tempestuous (I hv had both :smile: .)

If no specific objections as above and local protocol does not prohibit (?), I guess there should be no reason not to contact in view of some "claimed" uncertainty of understanding over the issue thereby getting a "polite" opportunity to refer to supporting references, etc.
(After all, you are paying for this privilege!)

Re L.mono, see my added attachment / comments previous post. As per yr comment, should be un-detectable in end product (ever checked? :smile: ).
(in above ref., see table paras 1.1 - 1.3)
I found the ref. via a more readable guide -
Attached File  RTE PHLS food guidelines.pdf   998.98KB   126 downloads

After yr baking info. the temptation is to select "absence" and avoid any chance of argument (assuming yr results are supportive!). But sometimes it is preferable to be conservative after considering the process / environment / shelf life / consumer type, the snag is that validation is then involved.

A search for cheese biscuits is recommended :smile: .

added - i hv one query myself - for the permanently zero tolerant species like salmonella, what statement did you select to put in the "acceptable level" column (presumably accepted by the auditor)?

Rgds / Charles.C


Dear Charles,
Thanks for your reply. It's very helpfull

Well, the audit actually was rather fine / good. just had 1 minor and some Potential NC :smile:
And after I received the audit report, I didn't see finding about the problem, maybe he combine with another finding :biggrin:

We never checked L. mono since cheese isn't our main ingredients and not all products used it.
Maybe next time, we'll consider to check it.
For Salmonella, definitely negatif / 25g for acceptable level, just like local authority guidance.

I want to ask another question, maybe has a litlle connection with the terms.
Is there any guidance for likely hood guidance for risk assessment?
I'll try to search in this forum, but if you can give a quick result, I'll be thankfull :oops:

Regards,
Hadi


Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

  • Earth
    Earth
  • Gender:Male
  • Interests:SF
    TV
    Movies

Posted 25 August 2011 - 04:54 AM

Dear HPG,

Is there any guidance for likely hood guidance for risk assessment?


I assume hood = good.

Yr query is a little wide scope. :smile:

I guess you are maybe asking for a useful reference for risk assessment to be able to be listed for audit purposes.? Or are you asking for a good "thread" on this forum link-wise ?

Either way, it depends on what you want the reference to demonstrate, eg -

(1) general review type reference illustrating design of risk matrices, examples not necessarily food related, but result suitable for haccp ?
(2) as above but must be food oriented.?
(3) general risk assessment procedure like used for random requests in a brc standard ? These tend to be more template-oriented with/without matrices, ie less "deep", but not always ?
(4) published evidence to support a specific case, eg yr own risk matrix ? (obviously needs further input)
(5) links / forum threads to quantitative risk assessment (eg MRAs)?
(6) otherwise ? :smile:

Can suggest any / all the above but quantity of effort (time) varies. :biggrin:

Rgds / Charles.C

PS - regarding cheese biscuit, does it hv a shelf-life ? (not my favorite food, cheese always tastes rather synthetic [like the Aluminium "cream", just speculating :smile: ]) . I haven't checked the literature but I presume that the cheese component, even after baking, retains ability to support growth (?). If not, then the decision for L.mono is presumably simplified, ie "absent etc", same as (presumably) automatic case for a plain dry biscuit. Assumes the packing environment is appropriate of course.

PPS - Of course, anybody else is (very) welcome to contribute to this thread. Especially cheese biscuit manufacturers. ;)

Kind Regards,

 

Charles.C


Thanked by 1 Member:
HPG

HPG

    Grade - MIFSQN

  • IFSQN Member
  • 99 posts
  • 30 thanks
1
Neutral

  • Indonesia
    Indonesia
  • Gender:Male
  • Location:Indonesia

Posted 25 August 2011 - 06:28 AM

Dear Charles,
Likely hood = likelihood (my bad :oops:)

I guess you are maybe asking for a useful reference for risk assessment to be able to be listed for audit purposes.? Or are you asking for a good "thread" on this forum link-wise ?

Yes, I'm asking a useful reference for risk assessment (got one finding for this matter).
Well, If there are any thread about this one, I think it's a good start.
I want to know how to choose / make desicion about likelihood (and severity) for risk assessment.

Shelf life for our products are 12 months.
Just want to let you know, that our product isn't cheese biscuit but cheese is one of the ingredient and added for a small amount. :biggrin:

Regards,
Hadi

Edited by HPG, 25 August 2011 - 06:34 AM.


Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

  • Earth
    Earth
  • Gender:Male
  • Interests:SF
    TV
    Movies

Posted 25 August 2011 - 08:46 AM

Dear HPG,

I’m happy that the audit seems to hv gone well. :thumbup:

From yr comments, the auditor may hv had a re-think/sought assistance.

Risk Assessment

Unfortunately, this is not a small topic, more like book-size. I hv limited my comments to traditional haccp-type risk matrices. These are my opinions only and (some) other people will probably not agree. I hv attached a few of the documents which I hv found helpful. First 2 are non-food but the content / explanations / examples are equally useful IMEX as far as understanding the basics. HACCP manual is quite recent (2009). There are also some very good forum threads here on risk matrix (one with over 50 posts I think).

The smallest size (3x3) is easiest/quickest to use but lacks flexibility where small differences in “degrees” of risk are involved. The larger ones (eg 5x5) are more “manipulatable” but can use up a lot of time initially.
The choice of cut-off for “significant / non significant” hazards is subjective, you can find examples for all kinds of methods as in my attachments.

Regarding “likelihood”, this is (crudely) an estimate of yr opinion as to the probability that something can go wrong with the existing process and generate a safety risk in the end product; more specifically, a particular food related hazard can occur, eg salmonella can survive a cooking process and subsequent steps.
The Americans choose to not consider a hazard as significant unless there is a “reasonable” likelihood that it will occur. Some other experts consider that a hazard of catastrophic severity is significant even if the likelihood is very low (eg nuclear meltdown). Others use only a “high” probability of occurrence to define a CCP. IMEX, from an auditor’s viewpoint, none are wrong if can be validated (assuming the standard leaves the decision procedure “open”).

This is why a debate exists over baking as a CCP, the likelihood is in a grey area since users believe that a defective heat treatment will auto-generate an unusable end product, regardless of safety. Others use the Codex tree and interpret the “design” question in a general/pragmatic way (elimination) so a CCP is achieved. i suppose you either "go with the flow" or be prepared to defend yr position.

And this "subjectivity" can be a problem with some commercial products, they may be good, but they do not justify their methodology.

Attached File  A guide for risk managers.pdf   596.61KB   96 downloads
(pgs 16-19 and examples at end [pgs 38 +] are most useful)

Attached File  A risk matrix for Risk Managers.pdf   189.75KB   86 downloads
(very readable risk discussion, focussed on one matrix)

Attached File  A HACCP manual (combat).pdf   792.5KB   107 downloads
(see especially 13.2.3, 14.2.2 [latter interprets the risk matrix])

Attached File  A guide to assessing risk using a risk matrix approach.pdf   51.79KB   89 downloads
(just for comparison, workplace type application, detailed but very unsimple IMO :smile: )

Re- biscuit, I hv seen mention in the literature to cheese flavoured crackers, cheese sandwich biscuits (central filling), pure crackers with cheese combos. I daresay there are more. Not quite sure what you have. I presume the cheese discussed was pre-pasteurised and has its own shelf life. I guess the specification of the cheese / application / final shelf life will define its significance to the end product. Interesting scenario but conclusion not so obvious (to me anyway).

Best Regards / Charles.C

PS feel free to revert any queries on risk etc


Kind Regards,

 

Charles.C


Thanked by 1 Member:
HPG

HPG

    Grade - MIFSQN

  • IFSQN Member
  • 99 posts
  • 30 thanks
1
Neutral

  • Indonesia
    Indonesia
  • Gender:Male
  • Location:Indonesia

Posted 26 August 2011 - 02:29 AM

Dear HPG,

I’m happy that the audit seems to hv gone well. :thumbup:

From yr comments, the auditor may hv had a re-think/sought assistance.

Risk Assessment

Unfortunately, this is not a small topic, more like book-size. I hv limited my comments to traditional haccp-type risk matrices. These are my opinions only and (some) other people will probably not agree. I hv attached a few of the documents which I hv found helpful. First 2 are non-food but the content / explanations / examples are equally useful IMEX as far as understanding the basics. HACCP manual is quite recent (2009). There are also some very good forum threads here on risk matrix (one with over 50 posts I think).

The smallest size (3x3) is easiest/quickest to use but lacks flexibility where small differences in “degrees” of risk are involved. The larger ones (eg 5x5) are more “manipulatable” but can use up a lot of time initially.
The choice of cut-off for “significant / non significant” hazards is subjective, you can find examples for all kinds of methods as in my attachments.

Regarding “likelihood”, this is (crudely) an estimate of yr opinion as to the probability that something can go wrong with the existing process and generate a safety risk in the end product; more specifically, a particular food related hazard can occur, eg salmonella can survive a cooking process and subsequent steps.
The Americans choose to not consider a hazard as significant unless there is a “reasonable” likelihood that it will occur. Some other experts consider that a hazard of catastrophic severity is significant even if the likelihood is very low (eg nuclear meltdown). Others use only a “high” probability of occurrence to define a CCP. IMEX, from an auditor’s viewpoint, none are wrong if can be validated (assuming the standard leaves the decision procedure “open”).

This is why a debate exists over baking as a CCP, the likelihood is in a grey area since users believe that a defective heat treatment will auto-generate an unusable end product, regardless of safety. Others use the Codex tree and interpret the “design” question in a general/pragmatic way (elimination) so a CCP is achieved. i suppose you either "go with the flow" or be prepared to defend yr position.

And this "subjectivity" can be a problem with some commercial products, they may be good, but they do not justify their methodology.

Attached File  A guide for risk managers.pdf   596.61KB   96 downloads
(pgs 16-19 and examples at end [pgs 38 +] are most useful)

Attached File  A risk matrix for Risk Managers.pdf   189.75KB   86 downloads
(very readable risk discussion, focussed on one matrix)

Attached File  A HACCP manual (combat).pdf   792.5KB   107 downloads
(see especially 13.2.3, 14.2.2 [latter interprets the risk matrix])

Attached File  A guide to assessing risk using a risk matrix approach.pdf   51.79KB   89 downloads
(just for comparison, workplace type application, detailed but very unsimple IMO :smile: )

Re- biscuit, I hv seen mention in the literature to cheese flavoured crackers, cheese sandwich biscuits (central filling), pure crackers with cheese combos. I daresay there are more. Not quite sure what you have. I presume the cheese discussed was pre-pasteurised and has its own shelf life. I guess the specification of the cheese / application / final shelf life will define its significance to the end product. Interesting scenario but conclusion not so obvious (to me anyway).

Best Regards / Charles.C

PS feel free to revert any queries on risk etc


Dear Charles,
Thanks for your fast reply.

I do agree that risk assessment isn't a simple task, need more reliable guidance.
Actually, I've already set about likelihood and severity for our risk assessment.
but the auditor suggest that we used guidance, or if we can't find one, we used historical experience from our company (for likelihood)

Since you mention it (baking as CCP), our auditors have asked it during audit. because we set baking as OPRP.
We give them an explanation and they seem to understand about it. :thumbup:

And thank you for the attachments. I'll try to read and digest them since I'll have a holiday for 8 days :happydance: (yay!!)


Regards,
Hadi


Thanked by 1 Member:

Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

  • Earth
    Earth
  • Gender:Male
  • Interests:SF
    TV
    Movies

Posted 26 August 2011 - 06:37 AM

Dear HPG,

Since you mention it (baking as CCP), our auditors have asked it during audit. because we set baking as OPRP.
We give them an explanation and they seem to understand about it

Indeed, the idea of classifying "grey" control measures like baking as part of OPRP programs did surface within one of the early drafts of ISO 22000 but subsequently vanished. Another example of pragmatism maybe ? :biggrin: Bonus point to you I think. :clap:

Rgds / Charles.C

Kind Regards,

 

Charles.C


Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

  • Earth
    Earth
  • Gender:Male
  • Interests:SF
    TV
    Movies

Posted 26 August 2011 - 09:13 AM

Dear HPG,

At the risk of spoiling yr holiday, there is another aspect to this discussion in the context of yr interest in “likelihood”.

As you hv stated, the baking temperature is a long way above the typical elimination values for vegetative pathogens eg a salmonella hazard. Plus the product intrinsic parameters (moisture / preservatives?) probably preclude any further growth of any surviving spores (?)within shelf life. Plus it is assumed hygiene is satisfactory.

I also daresay you hv been making this product for years with no microbiological problems.

It is therefore arguable that there is a very low probability of any failure of the baking process, ie low chance of salmonella survival. This "very low/low" result for the likelihood will for many standard risk matrices yield a non-significant hazard. In other words, neither a CCP or OPRP!

Somehow, I doubt that yr auditor would be less readily satisfied with this result but why not ?

(a )One alternative if pragmatically determined on a CCP or OPRP is to select another risk matrix of course.
(b) Or (rapidly) re-evaluate and decide that you are being too trusting of yr baking system.
© Or look for other epidemiological support for a higher likelihood (after all, most traditional HACCP texts seem to have a CCP?).

I expect most people do (b) of course. :biggrin:

Rgds / Charles.C


Edited by Charles.C, 26 August 2011 - 09:42 AM.
deleted (d), not practical for ISO 22000

Kind Regards,

 

Charles.C


Thanked by 1 Member:
HPG

HPG

    Grade - MIFSQN

  • IFSQN Member
  • 99 posts
  • 30 thanks
1
Neutral

  • Indonesia
    Indonesia
  • Gender:Male
  • Location:Indonesia

Posted 26 August 2011 - 04:43 PM

Dear Charles,
Thanks for your input. holiday or not, I always welcome any new knowledge :thumbup:

Dear HPG,

At the risk of spoiling yr holiday, there is another aspect to this discussion in the context of yr interest in “likelihood”.

As you hv stated, the baking temperature is a long way above the typical elimination values for vegetative pathogens eg a salmonella hazard. Plus the product intrinsic parameters (moisture / preservatives?) probably preclude any further growth of any surviving spores (?)within shelf life. Plus it is assumed hygiene is satisfactory.

I also daresay you hv been making this product for years with no microbiological problems.

It is therefore arguable that there is a very low probability of any failure of the baking process, ie low chance of salmonella survival. This "very low/low" result for the likelihood will for many standard risk matrices yield a non-significant hazard. In other words, neither a CCP or OPRP!

Somehow, I doubt that yr auditor would be less readily satisfied with this result but why not ?

(a )One alternative if pragmatically determined on a CCP or OPRP is to select another risk matrix of course.
(b) Or (rapidly) re-evaluate and decide that you are being too trusting of yr baking system.
© Or look for other epidemiological support for a higher likelihood (after all, most traditional HACCP texts seem to have a CCP?).

I expect most people do (b) of course. :biggrin:

Rgds / Charles.C


I don't think that the auditor will be satisfied if we say that, e.g. Salmonella sp., isn't a significant hazard. :thumbdown:

Btw, I've read (fast reading, of course) all the attachment. I see that a lot of guidance usually use 5x5 matrix.
Do you think using 5x5 more reliable than 3x3 or ?

Regards,
Hadi


Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

  • Earth
    Earth
  • Gender:Male
  • Interests:SF
    TV
    Movies

Posted 26 August 2011 - 09:50 PM

Dear HPG,

Btw, I've read (fast reading, of course) all the attachment. I see that a lot of guidance usually use 5x5 matrix.
Do you think using 5x5 more reliable than 3x3 or ?

Another subjective topic :smile: .

IMEX, the advantage of the 5x5 matrix compared to 3x3 is that it provides a wider band of options to (a) position the likelihood and severity parameters, (b) define the area where the hazard risk is categorised as significant / non-significant. For most hazards, this increased flexibility makes no practical difference but there are some cases where it can be useful, for example to differentiate between hazards considered to be of similar but “significantly” unequal risks and particularly if a critical/non-critical decision is involved.
One disadvantage of the 5x5 IMO is that it tends to generally require more thought to use (for me anyway).

The choice may also depend on whether you use a decision tree for the final CCP decision or not. This is optional in traditional HACCP but unavoidable (or by some other related method) for categorisation of CCP/OPRP in ISO 22000. ( Although I hv seen a few published examples solely using a risk matrix and apparently auditor-acceptable, surprising IMO).

Other posters may have other reasons for matrix size preference also.?

Rgds / Charles.C

Kind Regards,

 

Charles.C


Thanked by 1 Member:
HPG

Oraine

    Grade - Active

  • Newbie
  • 10 posts
  • 2 thanks
1
Neutral

  • Jamaica
    Jamaica
  • Gender:Male

Posted 17 May 2014 - 05:33 PM

Not even sure if I am using this thing right, but I just want tp thank everyone for the contribution I learnt a lot.





Share this

0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users