Standard Plate Count Reporting
What is the standard/generally accepted practice for reporting standard plate counts (SPC) on a Certificate of Analysis (COA) when you have multiple results for the product being represented by the COA? The max or the average (or something else)?
For reference, I am in the dairy industry.
Depends on how you're defining your own micro limits here. If your microbiological limits are based on an average (e.g. a n-c-m limit) then the average might make more sense. Alternatively, if your spec has a limit of say 100,000/g of ANY sample, I would report your highest value observed.
We used to just provide all test values on the CoA, gives you credit for all the testing you actually do, but can put you in a weird position for something like APC where you might release on risk assessment regardless if it exceeded your limit on one sample.
I agree with Furfarmandfork. Might as well give yourself credit and put down all the test results that you got. If i am receiving an ingredient, I would rather know the worst test result.
Out of curiosity/caution, what is -
(1) the customer/regulatory specification (if any) ?
(2) the range/average of the data. Sometimes former is astronomic !.
Out of curiosity/caution, what is -
(1) the customer/regulatory specification (if any) ?
(2) the range/average of the data. Sometimes former is astronomic !.
1) Our standard spec is <10,000 cfu/g.
2) We are typically under 1,000 cfu/g.
My question comes from my personal agreement with the previous commenters: from the customer's perspective, our worst case is the most useful/important information, though our current COA format doesn't allow for the reporting of multiple results per lot. My direction is to report the average, and to consider values that are below the testing limit (<100) to be zero in calculating that average. I had hoped to find some sort of information regarding the common practice, but haven't really been successful thus far.
1) Our standard spec is <10,000 cfu/g.
2) We are typically under 1,000 cfu/g.
My question comes from my personal agreement with the previous commenters: from the customer's perspective, our worst case is the most useful/important information, though our current COA format doesn't allow for the reporting of multiple results per lot. My direction is to report the average, and to consider values that are below the testing limit (<100) to be zero in calculating that average. I had hoped to find some sort of information regarding the common practice, but haven't really been successful thus far.
IMEX the sample size is usually built in to the seller/buyer's contractual agreement. i assume the spec you quote is mutually agreed.
It might relate to the product but If all the APC results are <1000cfu/g it likely not matters much to yr customer how you present them ?.
Results for other micro. parameters might have a/more significance.
Personally i refrain from donating excess information unless it's agreed on so i would probably calculate an average including the 100's since they presumably don't make much difference to the answer anyway ?, eg 900,0,0 and 900,100,100 = pretty much the same average allowing for micro. accuracy.
PS - <100cfu/g presumably means no colonies on 10(-2) plates ?. Maybe should do 10(-1) plates as well unless not practical.
Currently in dairy and been in dairy for 15 years or so. I've always reported micro as the max limit under the regulatory guidance.
For example, pasteurized milk (California limit is <15,000 CFU/mL for SPC). Keep in mind there is a 2 out of 3 for the sampling and checks by regulatory. I'm assuming you are Grade A plant? If not, then don't mind me...
As someone else said, it is what your specification allows. You may consider reporting both average and max. Up to you...no specific standard on anything for any food group as it relates to Certificates of Analysis.