Raw material specification vs risk analysis (meat products)

Hi,

We have recently had our initial FSSC audit, and passed with a few minors. 

One of the minors is confusing me, I would like to hear your opinion on this:

Our raw materials are frozen slaughter byproducts which are obtained from EU approved slaughterhouses, most of them are IFS/BRC certified. We do not receive productspecifications from all of them, so in some cases we have them sign our own raw material specification. In this specification I have included "salmonella absent/25g" as specific microbiological criterium, also to the criterium that the raw material needs to be derived from animals that are slaughtered in EU approved slaughterhouse, and comply to the 853/2004, etc. etc. This specific specification is for turkey skins and in the 2073/2005 I can not find any other criterium for pathogens for these products. 

In my risk analysis I have mentioned the theoretical possibility of presence of pathogens as salmonella, listeria, E. coli,. Staph. Aureus, etc. Since we have a very effective germ reducing process, presence of these are no high risk, (unless ofcourse there are such high numbers the toxine would be a risk, but there are other measurements to prevent that) 

The auditor dit not accept the specification because it only mentioned the absence of salmonella and therefor it did not match the risk analysis. 

When I look at the microbiological requirements for fresh meat (which is going to have a germ-reducing step) in the 853 or 2073 I can only find requirements for the absence of salmonella. She thinks I need to include also limits for the presence of other pathogens. But what limits do I need to include, I have examples of specifications for raw materials from other suppliers, but they are all different, also do not mention all pathogens and since there are no specific limits mentioned in legislation I have no clue what to include in the specification. It is not possible to require "absence of all pathogens" because of the material (frozen slaughter by products), and we do not monitor the microbiological quality of the incoming raw materials (I do ask for monitoring reports from our suppliers). She did not ask for any specification of other suppliers, so I do not know what her opinion was on those.

How should I solve this one... 

Thank you in advance!

Raw chicken, turkey and other poultry hazards - Salmonella and Campylobacter per USDA guidelines - is it a simple as "the specs did not match the risk analysis" so instead of adding other pathogens to your spec sheet, you reduce the pathogens on your risk analysis for poultry?

Here is a, I believe, a USDA approved meat haccp plan -

 rawNotGround_haz_analysis 05-04-12.doc   83.5KB   111 downloads

Can compare the hazards.

The specification IMO should match, or approx. match, the haccp listed micro. hazards.

IMHO yr micro. specification should be expanded somewhat and the auditor's expectations should be reduced.

Ideally an agreed spec. should reflect routinely achievable data. In practice, some optimism may be occasionally observed.

Personally for my haccp plans i added a preface sheet listing a number of typical BCP hazards for raw inputs, packaging etc

I then just entered a couple of examples from the list in the hazard analysis's initial stages.

PS- I do agree that most published specifications seem to make little mention of pathogenic E.coli or L.mono.

Thank you both for your input! 

I suggested that immediately, that I'd change the risk analysis then (you can have it as you want it ;-)), but she would not accept that as a solution :-)  

I can agree on inserting campylobacter for the poultry, but if I check the 2073/2005 (we are in Europe), there are only limits for carcasses for broilers, which is something different then our products. So then I still do not know what to require from our suppliers.

Maybe it's comparable to acrylamide, this can be a risk in some products, you probably would mention it in the HACCP analysis, but there are no legal limits for it, what would you ask for on your specification?

Hi Patgear,

This is a US oriented follow-up to the OP. Yr location unknown but details in OP seemed more EU-centric so relevance not necessarily same but may well be.

The reason for focus on Salmonella in beef safety related specs possibly derives from FSIS's issuance (in Pathogen Code 1996) of a "Standard" for Salmonella despite their recognition of potential, additional, micro.hazards, eg pathogenic E.coli. The basis of decision is discussed in the Code and in part relates to the prediction that recommended safety precautions would be effective against other relevant pathogens.

FSIS's evaluation of micro. hazards in beef is summarised here (1999)

 meat1 - FSIS,hazards in meat,poultry,1999.pdf   81.9KB   63 downloads

Since that time various (complicating) micro. developments have occurred with respect to FSIS/(beef/pork/poultry) industry which are detailed in this 2018 document,  -

 meat2 - GAO,USDA, pathogens in meat,poultry,2018.pdf   684.29KB   73 downloads

This extract from above is illustrative -

In addition to regulating meat and poultry sold in commerce, the U.S.Department of Agriculture (USDA) also purchases food and, in some cases, has additional food safety requirements for food it purchases. USDA’s Agricultural Marketing Service (AMS) purchases beef and other food for various federal nutrition assistance programs, including the National School Lunch Program. 1  USDA provides this food to states in support of about 100,000 public and private nonprofit schools that provide lunches to about 30 million children.

 

Ground beef is a staple of school menus. For example, according to AMS officials, during fiscal year 2016, the agency purchased more than 110 million pounds of raw beef, over 90 percent of which was delivered to the National School Lunch Program. Further, according to AMS officials, about 41 million pounds (37 percent) were delivered raw while the rest was delivered to a federally inspected processing facility for cooking prior to delivery to school lunch program agencies. Beef to be delivered raw to the National School Lunch program is tested for pathogens (Salmonella and Shiga-toxin-producing E. coli, two pathogens that can cause foodborne illness in humans) and certain microorganisms such as aerobic plate count bacteria, coliform bacteria, and generic E. coli that serve as indicators of the effectiveness of slaughter and processing plants’ process controls to limit pathogens. According to AMS officials, these indicator microorganisms indicate the quality of the food safety controls at the plant. 

 

For raw beef products that AMS considers for purchase for its programs, the agency rejects any beef that tests positive for Salmonella, a pathogen that can cause foodborne illness in humans. According to AMS officials, this requirement that beef purchased for these programs not test positive for Salmonella differs from the regulatory standard for beef inspected by USDA’s Food Safety and Inspection Service (FSIS). Further, according to AMS officials, AMS set this requirement because raw beef was considered the product with the most risk for recipients and enough plants were able to meet the requirement. AMS officials said that as a purchaser for various federal nutrition assistance programs, the agency has discretion to set requirements for qualified suppliers, and plants can choose whether to become qualified suppliers.

 

Most Product micro. specs include both safety/non-safety items so that inclusion of Salmonella, Pathogenic E.coli, APC, generic E.coli seems a reasonable mix.

With respect to limits,  2 ideas here (many other permutations exist) -

 meat3 - micro.meat.PNG   29.11KB   2 downloads

 meat4 - micro.meat.PNG   68.81KB   2 downloads

Thank you Charles, this is very interesting information. Your example of limits will be very helpful. 

I am going to study it and will try to compose a reasonable spec from this.     

Here is the link to USDA limits for pathogens associated with poultry  

http://www.inspectio...8/1396291666107

http://www.inspectio...5/1371663913885

http://www.inspectio...6/1371152627120

It also gives the testing frequency that the slaughterhouses should be using, they should give you something concrete to work with

We have spec sheets on files from our turkey supplier stating:

We list salmonella as the significant risk associated with the raw material and subsequent steps in the process (cook and chill) reduce/and or eliminate the hazard in the final product. 

  You have a very effective germ reducing process, presence of these are no high risk

 She thinks I need to include also limits for the presence of other pathogens.  

 Her request does not makes sense to me.

I'm thinking your auditor has zero experience is meat, or understands the slaughterhouse requirements.

If your suppliers are meeting USDA limits, then your spec should simply state what the USDA demands of them and that your process will take care of the rest (so you could simply take the USDA protocol and limits and insert them into your plan)???

E coli and camploybactor should be listed, however, as in reality very few carcasses actually get tested and they are representative of the LOT, which basically means, you could have products in your facility that are well above what USDA allows.

Does that help?  (5+ years in poultry slaughter in Canada exporting to the US)

I suppose it all depends on the context, ie the  purpose of the obtainance of a "specification". To use for ? Perhaps it's obvious in the slaughterhouse business ?

Is it only a "safety" specification ?

is it supposed to be a "narrow"  "haccp" related product "specification" or a general "Product Specification" ? (I assumed the latter).

eg Post 7 - Great links but no mention of  "specification".

@Scampi - you are correct about Campylobacter as indicated in file "meat1". I missed the "turkey" and assumed "meat" = beef. If poultry, logically/apparently also no need for pathogenic E.coli.

@ lynne - does the raw product usually comply with APC max 100,000 cfu/gram ? Just curious, sounds remarkably excellent, eg -

 meat-poultry micro.PNG   32.49KB   1 downloads

https://www.fsis.usd...icken_Parts.pdf

Page 16 Table 4 you'll find the survey shows poultry majority sampling that max or better

for the links I provided  (all pathogens are tested on a schedule..........the samples either pass or fail and there is a certain # of carcasses allowed to have a pathogen detected depending on the pathogens and frequency within the sample set)

Ecoli ---USDA uses a "moving" window approach of statistical control (scroll to section T.2.8.1 The Moving Window Approach Used by the USDA)

Salmonella; Table U.2.2

USDA Performance Standards for Salmonella by class of product
Sources: Code of Federal Regulations, Ch. 9, 310.25 (7), Jan. 1st 2012; and FSIS Notice 54-12 (09/11/ 2012)

Table U-1.1:USDA Performance Standards for Campylobacter in chilled carcasses of young chickens and turkeys  1.3 The establishment will meet Campylobacter standards if the results show no more than eight (8) positive samples in a 51-sample set for young chickens; and three (3) positive samples in a 56-sample set for young turkeys. The standards for both organisms are indicated in the table below:

So it's not really Yea or Nea..........it's whether or not the slaughter house met the requirements for EACH sample set

https://www.fsis.usd...icken_Parts.pdf

 

Page 16 Table 4 you'll find the survey shows poultry majority sampling that max or better

Hi Lynne,

Thks for file.

There is difficulty in interpreting/comparing data with "specifications" since -

(1) 3 different units/techniques are potentially being involved - cfu/gram product, / cm² surface, /ml rinsate. Possible correlations seem to lack a consensus.

(2) data for skin-on, skin-off, all varieties is lumped together

(3) arithmetic mean is astronomic but "distorted" (not so unusual for APC data)

As a side-note, from a "specification" POV, the data shown for Salmonella/Campylobacter  illustrate the degree of "optimism" mentioned in Post 3.

JFI  one can compare the opinion on the US Chicken/Pathogen Scenario in these 2 documents dated 2010 and 2018 respectively -

https://www.foodsafe...in-raw-poultry/

https://www.wired.co...than-you-think/

Hi Lynne,

 

Thks for file.

 

There is difficulty in interpreting/comparing data with "specifications" since -

 

(1) 3 different units/techniques are potentially being involved - cfu/gram product, / cm² surface, /ml rinsate. Possible correlations seem to lack a consensus.

(2) data for skin-on, skin-off, all varieties is lumped together

(3) arithmetic mean is astronomic but "distorted" (not so unusual for APC data)

 

As a side-note, from a "specification" POV, the data shown for Salmonella/Campylobacter  illustrate the degree of "optimism" mentioned in Post 3.

Thanks all for the input again!

This is also what I'm struggling with, the data is mostly about carcasses or parts, but eventually we receive only the skins and it is a challenge to set limits which actually can and will be met. 

Below an example of microbiological criteria in a specification from one of EU's largest supplier of turkey meat/skins. It's more about processing hygiene than the product when it comes to slaughterhouses I guess, all fresh meat/parts will be further processed and if not there are specific standards for it. I also have a specification from a rind supplier who states that they cannot guarantee absence of salmonella or listeria. They have limits for TVC 100.000 cfu and Entero 1000 cfu.