High likelihood of failure of antibiotic rapid test kit as control measure
Hi, I got a technical question after checking someones hazard risk assessment and control measure determination. The person identified antibiotic residue in milk as a chemical hazard, his control measure is, checking the antibiotic residue in milk using rapid test kit for screening. While calculating the score for determining CCP/OPRP, he showed that there is high probability of failure of control measure (Score 3). Cumulatively, it made the total score above the threshold value to determine the CCP.
Now my question is, in this case, high likelihood of failure means the failure of the test kit to determine the presence of antibiotic residue (Please correct me if I am wrong). Then, how is this CCP being controlled? Does it make any sence really? To me, if his test kit is not reliable enough, how he is controlling that CCP?
Honestly, I am bit confused with the term likelihood of failure. Which failure it actually means? I would be happy for the clarification.
Hi, in my opinion, high likelihood of failure relates to controls at the farms not their testing equipment, the testing equipment is there to prevent or minimise the risk of the hazard getting undetected. It looks like their must have had a high number of antibiotics failures within their milk suppliers and therefore concluded the milk intake step(antibiotics testing) a ccp to minimise the risk.
Hi AgaN
Thanks for your answer. But, I don't think so, because, the likelihood of occurrence (L) due to failure in farm level is already covered during significance (S x L) determination. Later, the likelihood of failure should relate to the control measure that is in place, and it is required to determine the categorization of control measure (CCP/OPRP). For example, the milk pasteurization is a control measure to kill the pathogenic bacteria and is considered as CCP. In this case, there is possibility of the failure of pasteurization process and it can be determined by monitoring the temperature and time. However, this logic does not go with the above mentioned example of antibiotic residue determination.
Hi AgaN
Thanks for your answer. But, I don't think so, because, the likelihood of occurrence (L) due to failure in farm level is already covered during significance (S x L) determination. Later, the likelihood of failure should relate to the control measure that is in place, and it is required to determine the categorization of control measure (CCP/OPRP). For example, the milk pasteurization is a control measure to kill the pathogenic bacteria and is considered as CCP. In this case, there is possibility of the failure of pasteurization process and it can be determined by monitoring the temperature and time. However, this logic does not go with the above mentioned example of antibiotic residue determination.
Hi Mazid,
Scoring/decision system is not given but do you appreciate that if the control measure has a high probability of failure, this increases the likelihood of control measure being associated with a CCP ?. It's a kind of inverse logic.
I'd be reaching out to the department who overseas dairy production to fully understand what the controls are
No, a rapid test isn't sufficient to measure residue
Here, each lot of incoming milk is sampled at the dairy, in the lab (not rapid tests) before the milk truck empties into the silos. If the lot doesn't pass, it doesn't go to the silo.
Sounds to me like whomever wrote the CCP controls, didn't do the research into the controls that exists ahead of the diary plant
I'd be reaching out to the department who overseas dairy production to fully understand what the controls are
No, a rapid test isn't sufficient to measure residue
Here, each lot of incoming milk is sampled at the dairy, in the lab (not rapid tests) before the milk truck empties into the silos. If the lot doesn't pass, it doesn't go to the silo.
Sounds to me like whomever wrote the CCP controls, didn't do the research into the controls that exists ahead of the diary plant
Hi Scampi,
I'm sure your procedure is viable however you can also find the OP's screening routine/control measure (assuming rapid test is in lab) at reception in some textbooks and haccp plans. It's debatable for ISO22000 whether to make it a CCP or PRP though. I'm not sure how fast is "rapid" though. :smile:
If the Netherlands have truly reduced the antibiotic use in dairy in the last 10 years, than the likelihood should reflect that----that's where you can get into disputes using historical data alone
The antibiotics are not to be on animals that are healthy, and most progressive milking systems would automatically dump the milk from a cow that is currently medicated. I'm questioning the rationale in the hazard analysis, not the process
If the Netherlands have truly reduced the antibiotic use in dairy in the last 10 years, than the likelihood should reflect that----that's where you can get into disputes using historical data alone
The antibiotics are not to be on animals that are healthy, and most progressive milking systems would automatically dump the milk from a cow that is currently medicated. I'm questioning the rationale in the hazard analysis, not the process
Hi Scampi,
Maybe it's Regulatory in the Netherlands. (No idea).
Hi Mazid,
Scoring/decision system is not given but do you appreciate that if the control measure has a high probability of failure, this increases the likelihood of control measure being associated with a CCP ?. It's a kind of inverse logic.
Hi Charles, thanks for your reply. This is exactly my point. Some people made calculation to categorize the control measure as CCP/OPRP using numbering/scoring system of the criteria that are mentioned in ISO 22000 clause 8.5.2.4.1 a & b. Here, high likelihood of failure of the control measure is considered as score 3 and multiplying it increases the value to account the control measure as CCP. I am posting here the spreadsheet
Hi Charles, thanks for your reply. This is exactly my point. Some people made calculation to categorize the control measure as CCP/OPRP using numbering/scoring system of the criteria that are mentioned in ISO 22000 clause 8.5.2.4.1 a & b. Here, high likelihood of failure of the control measure is considered as score 3 and multiplying it increases the value to account the control measure as CCP. I am posting here the spreadsheet
Hi Mazid,
This is the method I previously posted for iso22000-2005.
I developed similar scoring system for iso22000-2018 here -
https://www.ifsqn.co...18/#entry138153
You can see high probability of failure maximises chance of CCP in both excels ??
I agree with you that initially the logic seems opposite to commonsense. The idea is that since the control may fail it becomes even more important (aka "critical") to monitor etc as closely as possible which justifies a CCP relative to OPRP, (ie if it never failed, would hardly ever need monitoring or to be designated as a CCP.) (One might argue why would there still be a significant hazard but that's another story).
Hi Mazid,
If you prefer a (maybe) simpler option can have a look at the decision table in sheet 2 of the excel in this post -
https://www.ifsqn.co...ls/#entry192739
Above involves less mental stress since it by-passes some of the minutiae of the iso22000 standard however it closely follows a decision tree (Coca-Cola) which afaik was well-recognised for the original iso22000-2005. (Effective differences in this topic in 2018 version are quite marginal except notably the necessity for measurable critical limit/action criteria for OPRP).