13 replies to this topic

[skibum]

Good morning all,

 

I've been reading this forum for some time, and have been able to find a ton of really useful information and experiences. I have kind of a funny situation going on, and am pretty conflicted about how to approach it. Any advice or insight is appreciated.

 

We manufacture frozen cookie dough. This is sent to stores, baked, and sold to customers. The issue is that the baking does not constitute a true kill step, as the internal temperature does not rise above 130F.  On some cookies, with a denser filling, this temperature is as low as 100F.

 

To reformulate the cookies, or change the baking profile to hit a time/temp that would kill vegetative pathogens, would mean making an entirely different product.  So, I've tried to mitigate the risk in other ways:

 

-We use only heat treated flour and pasteurized eggs. We have a strong supplier approval program, with COAs for all lots, letters of guarantee, have conducted supplier audits, etc. We have also done raw material testing to confirm vendor COAs.

 

-We have a robust EM program and test zone 1 for indicators (EB,APC,Coliforms,Y+M) weekly, zone 2 for indicators and pathogens (Salmonella, Listeria, E. Coli, Staph)  2x/month, zones 3 and 4 for pathogens 2x/month. No positive pathogen results, and indicators are very low.

 

-Processing time is limited, and the batched product is never held unrefrigerated for greater than 2hrs.  It is molded into pucks, and frozen right away.  

 

-aW of finished product is <0.75, so not supportive of vegetative pathogen growth. Product is stored frozen before being baked in stores

 

-We have a positive release program, and test every batch of finished product for APC (typically less than 500 cfu/g, never over 5,000), Coliforms (always under 50, typically under 10 cfu/g), E. Coli (always negative), and Salmonella (always negative). 

 

Basically, I've been approaching it more like an RTE cookie dough, and less like a baked product.  I figure if we can always keep the micro levels low by using clean ingredients and having strong sanitation and processing controls, then maybe it would be acceptable to justify not having a kill step as a CCP? I know from reading this forum that many have justified having baking as a QCP, because the product would otherwise not be saleable, if it did not reach the proper baking time/temperature for a kill step. But that is not the case here; we know the product will not reach the proper temperatures to kill pathogens. 

 

What do you guys think? Would you ever feel comfortable with a situation like this? Do you think it is possible to ensure food safety with alternative controls for this type of product? And finally, would an (SQF) auditor be likely to accept a HACCP plan and the associated risk assessments / validations that follow this logic? Or do you think they would just say 'no kill step, no pass'?

 

Please let me know if I can provide any more info. Very interested to hear all of your thoughts! Thanks!  

 

-Processing time is limited, and the batched product is never held unrefrigerated for greater than 2hrs.

Edited by skibum, 27 June 2023 - 03:38 PM.

[Scampi]

I think you have a beautiful plan in place.  It all makes great sense to me

 

This allows you to keep your finished good the way you want it AND maintain safety

 

The kill step has already been handled for the flour and eggs-------you just need to ensure you have tight PROCESSING controls (time/ambient temp/storage temps etc) to ensure the ingredients stay wholesome prior to use

[Charles.C]

Good morning all,

 

I've been reading this forum for some time, and have been able to find a ton of really useful information and experiences. I have kind of a funny situation going on, and am pretty conflicted about how to approach it. Any advice or insight is appreciated.

 

We manufacture frozen cookie dough. This is sent to stores, baked, and sold to customers. The issue is that the baking does not constitute a true kill step, as the internal temperature does not rise above 130F.  On some cookies, with a denser filling, this temperature is as low as 100F.

 

To reformulate the cookies, or change the baking profile to hit a time/temp that would kill vegetative pathogens, would mean making an entirely different product.  So, I've tried to mitigate the risk in other ways:

 

-We use only heat treated flour and pasteurized eggs. We have a strong supplier approval program, with COAs for all lots, letters of guarantee, have conducted supplier audits, etc. We have also done raw material testing to confirm vendor COAs.

 

-We have a robust EM program and test zone 1 for indicators (EB,APC,Coliforms,Y+M) weekly, zone 2 for indicators and pathogens (Salmonella, Listeria, E. Coli, Staph)  2x/month, zones 3 and 4 for pathogens 2x/month. No positive pathogen results, and indicators are very low.

 

-Processing time is limited, and the batched product is never held unrefrigerated for greater than 2hrs.  It is molded into pucks, and frozen right away.  

 

-aW of finished product is <0.75, so not supportive of vegetative pathogen growth. Product is stored frozen before being baked in stores

 

-We have a positive release program, and test every batch of finished product for APC (typically less than 500 cfu/g, never over 5,000), Coliforms (always under 50, typically under 10 cfu/g), E. Coli (always negative), and Salmonella (always negative). 

 

Basically, I've been approaching it more like an RTE cookie dough, and less like a baked product.  I figure if we can always keep the micro levels low by using clean ingredients and having strong sanitation and processing controls, then maybe it would be acceptable to justify not having a kill step as a CCP? I know from reading this forum that many have justified having baking as a QCP, because the product would otherwise not be saleable, if it did not reach the proper baking time/temperature for a kill step. But that is not the case here; we know the product will not reach the proper temperatures to kill pathogens. 

 

What do you guys think? Would you ever feel comfortable with a situation like this? Do you think it is possible to ensure food safety with alternative controls for this type of product? And finally, would an (SQF) auditor be likely to accept a HACCP plan and the associated risk assessments / validations that follow this logic? Or do you think they would just say 'no kill step, no pass'?

 

Please let me know if I can provide any more info. Very interested to hear all of your thoughts! Thanks!  

 

-Processing time is limited, and the batched product is never held unrefrigerated for greater than 2hrs.

Hi skibum,

 

Is this sold as a RTE product ? (I anticipate Yes which auto-raises an Auditor's level of rigorousness. If No, it's a rather different ball-game).

Is there any manual handling ?

Not my Product area but I predict that an auditor will look at yr Process and (esp.of course the absence of a "kill" step) compare it to the methodology/Risk Assessment  being used by other processors for a similar product,

 

So how does it compare ?

 

PS- need some actual indicator data/sampling procedure to comment on env. micro controls.

PPS - I'm surprised you don't test finished product for L.mono, S.aureus COP ?

[kfromNE]

A true kill step - you have to consider time and temperature into play not just temperature. Many microorganisms can be killed at lower temperatures if left there for a period of time.

You need to think about your oven temperature as well. What do you tell customers to cook items at. 

[skibum]

Wow, after reading this forum for so long, it feels pretty cool having Scampi and Charles C. reply to my post!  You two have indirectly answered a ton of my questions over the last few months. I really appreciate your expertise and willingness to help others!

 

Hi skibum,

 

Is this sold as a RTE product ? (I anticipate Yes which auto-raises an Auditor's level of rigorousness. If No, it's a rather different ball-game).

Is there any manual handling ?

Not my Product area but I predict that an auditor will look at yr Process and (esp.of course the absence of a "kill" step) compare it to the methodology/Risk Assessment  being used by other processors for a similar product,

 

So how does it compare ?

 

PS- need some actual indicator data/sampling procedure to comment on env. micro controls.

PPS - I'm surprised you don't test finished product for L.mono, S.aureus COP ?

 

Charles,

 

No, it is not sold as RTE. We make cookie dough, freeze it, and sell it to a customer who bakes and sells it in their store. We run cook tests on the product per customer request, on the same oven and settings as they use in stores. This is how I know that the baking does not reach time/temp requirements for a kill step. Its really not close.

 

There is very minimal manual handling. Everything is automated, except for the tray-up process, where freshly made dough pucks are placed by hand onto a tray for freezing. I anticipate this element being removed in the next few months, as we invest in automatic conveyors and a cyro freeze tunnel.

 

I agree that an auditor would likely try to draw comparisons to similar products and their risk assessments -- that's what has me concerned, as I haven't heard of or read about other instances of a 'baked' product which, by design, doesn't reach temperatures necessary to kill pathogens. This led to a little out of the box thinking, which is why I started drawing comparisons to RTE dough. However, there doesn't seem to be a ton of information available about risk assessments used for this product type.

 

Not sure exactly what info about the EM program would be helpful. I use sponge swabs, swab a 10x10cm area. Food contact is tested for indicators and results are generally very low (<1000 APC, <10 coliform, <10 entero, <500 Y+M all cfu/g). Food contact adjacent is tested for the same, plus pathogens (always negative), and the levels are similar.

 

We do test for S. Aureus on finished goods, left that off the list accidentally. I had not considered testing for L. mono - it didn't seem like a risk for this product type.

 

Thanks again for the responses!

[skibum]

A true kill step - you have to consider time and temperature into play not just temperature. Many microorganisms can be killed at lower temperatures if left there for a period of time.

You need to think about your oven temperature as well. What do you tell customers to cook items at. 

 

Right - the combination of time and temperature is nowhere close to sufficient to function as a kill step. Baking is around 300 for around 15 minutes - as mentioned, this results in a core temp well below that required to kill pathogens

[Charles.C]

Wow, after reading this forum for so long, it feels pretty cool having Scampi and Charles C. reply to my post!  You two have indirectly answered a ton of my questions over the last few months. I really appreciate your expertise and willingness to help others!

 

 

Charles,

 

No, it is not sold as RTE. We make cookie dough, freeze it, and sell it to a customer who bakes and sells it in their store. We run cook tests on the product per customer request, on the same oven and settings as they use in stores. This is how I know that the baking does not reach time/temp requirements for a kill step. Its really not close.

 

There is very minimal manual handling. Everything is automated, except for the tray-up process, where freshly made dough pucks are placed by hand onto a tray for freezing. I anticipate this element being removed in the next few months, as we invest in automatic conveyors and a cyro freeze tunnel.

 

I agree that an auditor would likely try to draw comparisons to similar products and their risk assessments -- that's what has me concerned, as I haven't heard of or read about other instances of a 'baked' product which, by design, doesn't reach temperatures necessary to kill pathogens. This led to a little out of the box thinking, which is why I started drawing comparisons to RTE dough. However, there doesn't seem to be a ton of information available about risk assessments used for this product type.

 

Not sure exactly what info about the EM program would be helpful. I use sponge swabs, swab a 10x10cm area. Food contact is tested for indicators and results are generally very low (<1000 APC, <10 coliform, <10 entero, <500 Y+M all cfu/g). Food contact adjacent is tested for the same, plus pathogens (always negative), and the levels are similar.

 

We do test for S. Aureus on finished goods, left that off the list accidentally. I had not considered testing for L. mono - it didn't seem like a risk for this product type.

 

Thanks again for the responses!

Hi skibum,

 

TBH if the product is not marketed as RTE, I would anticipate that from a HACCP POV, this is a low risk product  since I presume any pathogens which might be present will presumably be eliminated at the next stage.

Of course, if the item is promoted as having microbial quality equivalent to an RTE "baked" end-product, the practical situation might be different.

Thks yr env.results, i deduce that all the numbers quoted are equivalent to "nil detections" and seem exemplary although the detection level for Y&M looks curiously high.

[skibum]

Hi skibum,

 

TBH if the product is not marketed as RTE, I would anticipate that from a HACCP POV, this is a low risk product  since I presume any pathogens which might be present will presumably be eliminated at the next stage.

Of course, if the item is promoted as having microbial quality equivalent to an RTE "baked" end-product, the practical situation might be different.

Thks yr env.results, i deduce that all the numbers quoted are equivalent to "nil detections" and seem exemplary although the detection level for Y&M looks curiously high.

 

Hi Charles,

 

Not marketed RTE. 

 

Detection for APC and Y+M are both 100 cfu/g. You're right; for food-contact areas, our results are pretty much always right there at the minimum level for detection. We see minor fluctuations in these 'general' microbial indicators in zone 2, but very rarely above the numbers quoted. These are adressed immediately with additional cleaning/training, and modifications to the master sanitation schedule. We do ATP swabbing + visual inspection at pre-op, which I imagine is catching most OOS micro conditions before EM swabbing takes place.

 

I think that would be the normal presumption for baked products - but in our case, we happen to know that they will not be eliminated.  In fact, my thinking was that the customer's 'cooking' step is much closer to an 'incubation' than a kill step     

 

This is all complicated by the fact that we are in discussions to begin baking some cookies here, on-site, for direct shipping to consumers.  This would obviously have all kinds of implications for our food safety plan (not to mention facility registrations, etc) So I'm kind of pre-gaming for our new HACCP plan if this process materializes. It would then go from "we know that our customer isn't baking them enough to kill pathogens" to "we don't have a kill step". 

 

Thanks again, Charles.

Detection for APC and Y+M are both 100 cfu/g. We see minor fluctuations in these 'general' microbial indicators. Very rarely above the numbers quoted. We do ATP swabbing + visual inspection at pre-op, which I imagine is catching most OOS micro conditions before EM swabbing takes place.

[Charles.C]

Hi skibum,

 

Apologies that just realised I misunderstood yr OP in that the baking step is apparently insufficient to justify elimination of relevant pathogens. Why is this cookie process implemented in this way ? "Super Cookies" ?

 

So the contractual micro requirement applied to yr product presumably becomes critical, ie effectively RTE. Sounds a bit like Russian Roulette.

 

Suggests you need to be doing some intensive environmental sampling and on employees if manual handling (ie S.aureus COP).

The APC detection level  in Post 8 is hopefully a typo, ie too high.

 

L.mono is the classic env. hazard, If the baking step will not kill it then be prepared !

[skibum]

Yes Charles, that's my reason for writing this post. 

 

The customer bakes the cookies. This baking does not hit time/temp to eliminate pathogens.

 

I can't speak to the design of the process. I imagine they bought the formulas and equipment because they liked the product, and no one ever considered microbial growth potential until they contracted us to produce.

[Charles.C]

Yes Charles, that's my reason for writing this post. 

 

The customer bakes the cookies. This baking does not hit time/temp to eliminate pathogens.

 

I can't speak to the design of the process. I imagine they bought the formulas and equipment because they liked the product, and no one ever considered microbial growth potential until they contracted us to produce.

Definitely Russian Roulette. Assuming the cookie process is FDA controlled, must  have been some interesting discussions.

I would like to see the cookies haccp plan inasmuch as many baking plans claim that not possible to make a visually saleable product at temperatures below that which will eliminate pathogens (as per yr OP).

 

PS - you should also be testing raw materials etc for B.cereus

PPS - Basically it sounds like a lot of Process Validations need to be seen as far as the cookie process is concerned,  but as I understand yr responsibility currently ends at the dough production. The dough haccp plan will presumably become high risk so will attract increased auditorial scrutiny..As to SQF auditorial feasibility, I guess it depends on yr capability to validate/verify the associated micro.specification etc. Maybe think Sushi ? If you are the innovator, can anticipate some headaches IMO.

 

P3S - I guess you are not exactly the innovator, eg -

 

https://www.houstonm...w-cookie-dough/

 

https://www.verybest...kie-dough-tips/

Edited by Charles.C, 28 June 2023 - 06:39 PM.
added

[skibum]

Definitely Russian Roulette. Assuming the cookie process is FDA controlled, must  have been some interesting discussions.

I would like to see the cookies haccp plan inasmuch as many baking plans claim that not possible to make a visually saleable product at temperatures below that which will eliminate pathogens (as per yr OP).

 

PS - you should also be testing raw materials etc for B.cereus

PPS - Basically it sounds like a lot of Process Validations need to be seen as far as the cookie process is concerned,  but as I understand yr responsibility currently ends at the dough production. The dough haccp plan will presumably become high risk so will attract increased auditorial scrutiny..As to SQF auditorial feasibility, I guess it depends on yr capability to validate/verify the associated micro.specification etc. Maybe think Sushi ? If you are the innovator, can anticipate some headaches IMO.

 

P3S - I guess you are not exactly the innovator, eg -

 

https://www.houstonm...w-cookie-dough/

 

https://www.verybest...kie-dough-tips/

 

Thanks Charles.

 

I very much doubt that the retailer has any haccp plan in place.

 

Thank you for the note on B. Cereus. I have been leaning in their direction based on reading here. Will look into how to implement.

 

I think you're right that, currently, our responsibility more-or-less ends when our product leaves the facility. And I'm very confident that we are sending out safe food. I do wonder about things like labeling - we have nothing on the box that says "cook thoroughly" or similar..and I wouldn't want to put it on there, knowing our customer's process. 

 

What do you mean exactly by, 'the dough haccp plan will presumably become high risk'?  And could you elaborate on the comment about thinking Sushi?

 

Definitely not the innovator. And, in my role, would much rather be making a more 'traditional' product.  "My thought process was basically, if it's possible to make a safe RTE cookie dough, it must be possible to make a safe "par-cooked" cookie, right?" And I've been going down that path since.  I couldn't really find any examples of baked products that don't reach t/T requirements for a kill step - although I'm sure they exist.

 

Thanks again 

 

Edit: I just saw your edit to your previous post. 

 

There are not contractual micro specifications, we have set ours internally.  What are you comparing to russian roulette? Sorry, missing the meaning there. I feel like our EM program is pretty intensive. We test zones 1 and 2 for staph quite frequently, and test every lot of finished good.

 

APC detection level of 100 cfu/g is too high? I can't recall seeing a lab report anything lower..

 

I will have to re-evaluate testing for L. mono in finished product.  This is something I've always avoided, and based on aW + sanitation controls, EM, and supplier COAs, may have incorrectly assigned a lower risk than it deserves...

Edited by skibum, 28 June 2023 - 09:44 PM.

[Charles.C]

Thanks Charles.

 

I very much doubt that the retailer has any haccp plan in place.

 

Thank you for the note on B. Cereus. I have been leaning in their direction based on reading here. Will look into how to implement.

 

I think you're right that, currently, our responsibility more-or-less ends when our product leaves the facility. And I'm very confident that we are sending out safe food. I do wonder about things like labeling - we have nothing on the box that says "cook thoroughly" or similar..and I wouldn't want to put it on there, knowing our customer's process. 

 

What do you mean exactly by, 'the dough haccp plan will presumably become high risk'?  And could you elaborate on the comment about thinking Sushi?

 

Definitely not the innovator. And, in my role, would much rather be making a more 'traditional' product.  "My thought process was basically, if it's possible to make a safe RTE cookie dough, it must be possible to make a safe "par-cooked" cookie, right?" And I've been going down that path since.  I couldn't really find any examples of baked products that don't reach t/T requirements for a kill step - although I'm sure they exist.

 

Thanks again 

 

Edit: I just saw your edit to your previous post. 

 

There are not contractual micro specifications, we have set ours internally.  What are you comparing to russian roulette? Sorry, missing the meaning there. I feel like our EM program is pretty intensive. We test zones 1 and 2 for staph quite frequently, and test every lot of finished good.

 

APC detection level of 100 cfu/g is too high? I can't recall seeing a lab report anything lower..

 

I will have to re-evaluate testing for L. mono in finished product.  This is something I've always avoided, and based on aW + sanitation controls, EM, and supplier COAs, may have incorrectly assigned a lower risk than it deserves...

Hi skibum,

 

A lot of queries above.

 

TBH I don't quite understand yr set-up. IMEX all the finished products involved (raw/RTE frozen seafood) had to have mutually agreed/documented specifications between buyer and receiver. Otherwise no Production. Additionally this was required for BRC.

 

As I understand yr process is implicitly designed to produce RTE dough although this is an internal specification only (would present SQF/Regulatory inspectors somewhat of an auditorial HACCP conundrum IMO).. Unless yr sampling is of, say FDA-L.Mono/IICMSF magnitude, it is impossible to ensure yr finished product has a low probability of including pathogens like Salmonella if present at low oncentrations. The best you can do is to validate yr suppliers claims/HACCP plans etc and similarly for yr own Process/Environment.

 

You seem to be trying to adapt  yr product to protect  the responsibility of yr receiver to sell safe food despite yr understanding  that the latter's process is probably unvalidatable (eg as per AIB baking criteria) [and therefore illegal ??] . IMO yr protocol  is illogical since if the receiver's product is ultimately involved in a health-related incident due, for example, Salmonella, the blame will go straight back to your "RTE Process". This as what I meant by Russian Roulette.

I deduce yr own dough haccp plan is effectively designed to produce RTE raw product. This is "analogous" to sushi albeit yr product is only designed to be at an intermediate stage.. The haccp status of yr finished dough would presumably be Low Risk if known to be further "baked" via a validated baking process containiing a "killing" step or possessed equivalent, labeled baking instructions  but as I understand this is categorically not the case so the risk becomes elevated.

 

IMO you  need to precisely (haccp/spec) define what yr product actually is, ie RTE or NRTE and/or  label it appropriately although, as I understand,  such labeling is strangely not mandated by FDA, unlike USDA.

 

Note that Env. quantitative micro data  typically has units of cfu/cm², not cfu/g. I'm unsure what yr quoted numbers are referring to ?.

 

Hope the above makes sense, seems to me you have a rather complicated situation. I deduce no Regulatory inspections are involved.

 

PS - just one additional comment - if the bakery under discussion were obliged to comply to FSMA, it appears from a perusal of FSMA requirements that, the process would necessarily contain a killing -step.

Edited by Charles.C, 30 June 2023 - 07:20 AM.
edited/added

[tahoeskier]

Be sure you have validated the baking instructions. You also need to insure that the cooking instructions are on the label as well as that the product is raw and must be cooked.