Determination of direct observations for a CCP
Were do I look to find documentation to support or determine if our current direct observation verification procedure and frequency is adequate. At present it is a minimum of once per month, direct observation of station attendant taking temperature of the product
Thank you
Tigo
Were do I look to find documentation to support or determine if our current direct observation verification procedure and frequency is adequate. At present it is a minimum of once per month, direct observation of station attendant taking temperature of the product
Thank you
Tigo
Dear Tigo,
I presume you are referring to the "verification step" in a typical HACCP system.
This typically involves a review of adequacy of recorded results, ie not just the observation of a monitoring procedure.
The choice of frequency may well depend on the type of product / process / measurement involved (??). And the associated HACCP status (??).
Rgds / Charles.C
PS - welcome to the forum !
Thank you Charles for your clarification question. I hope this further explanation helps.
In my recent FSA, it was stated that we failed to provide sufficient documentation to support our current direct observation verification procedure and frequency. This is a noncompliance under 9 CFR 417.5(a)(2) - the written HACCP plan, including decision making documents associated with the selection and development of CCP's and critical limits and documents supporting both the monitoring and verification procedures selected and the frequency of those procedures
The CCP is temperature of the meat in various ready-to-eat products being assembled. An assembler records the temperature every hour. Our QA tech, does a direct observation on the assembler taking temperatures as verification they are doing it correctly. The QA tech then records on the temperature log, her results ... time, item observed, temperature, acceptable or unacceptable results and signature. During the FSA our plan stated this process would be done once a month. I have since changed it to daily. What literature do I refer to, to get guidance to state this is a sound decision?
Perhaps a daily verification using a different "certified" thermometer than the one the operator uses to see if the results compare would add some credibility to your verification procedure.
Dear Tigo,
I presume this is a USDA audit.
I deduce you are only querying the frequency of yr direct observation procedure (my reservations to this activity as a sole verif. procedure already stated)(added later - I subsequently noticed that additional verif.reqs are listed in para. 417.4.)
Not in USA myself but there are several meat people on this forum who will probably know if there is a specific (USDA) maximum interval specified. I doubt there is a minimum. :smile:
For a typical (US) overall haccp reference can see -
http://www.fda.gov/F...06801.htm#app-g
(especially appendix G)
and see Fig.2 for example of schedule. (The frequency for corrective actions records review is unusually long IMEX)
As per above treatment, verification includes validation. I am unsure if CFR details follow same logic, probably yes.
IMEX (Codex haccp) most verification activities are regarded as a review function. Daily is often applied for record reviewing. Other routine functions are often done at longer intervals. One approach if unsure is to initially use short intervals, eg daily, then extend after a reasonable time period if no discrepancies.
Rgds / Charles.C
PS - (added) - the FSIS viewpoint on Verification/Validation (using CFR documentation) is here -
FSIS HACCP Systems Validation,Verification,2013.pdf 750.34KB 121 downloads
I noticed this text (my underlines) -
Following the 90 calendar day period of initial validation, an establishment uses its findings during the initial validation period to fully implement its system and solidify its monitoring and on-going verification procedures and frequencies. The establishment then continues on a daily basis to perform monitoring and verification activities to ensure that the HACCP system continues to be implemented properly. Establishments are required to support both the monitoring and verification procedures selected and the frequency of those procedures as part of 9 CFR 417.5(a)(2). Data gathered during initial validation, during which critical operational parameters are monitored at an intense frequency, is one source of information that can be used to support monitoring and verification procedures and frequencies (see examples in Appendix 4).
Importantly, not all critical operational parameters that are measured during initial validation are monitored on an ongoing basis after the initial validation period is over. For example, some parameters such as spatial configuration or ingredient formulation, may not change over time and therefore, do not need to be monitored. In addition, ongoing verification may include activities that were not performed as part of initial validation, because the purposes of these two processes differ.
Nonetheless, the later text, eg pgs 27-28, rather suggests that in practice verification frequencies are ultimately somewhat arbitrary. :smile:
Here is another criterion for the frequency -
It is generally recommended (or required under HACCP regulations) that HACCP verification be done at least annually, and whenever there is a significant change in the food processing and handling system. However, many HACCP verification activities (e.g., equipment calibration) may require more frequent verification. The frequency should be determined by the HACCP team after careful consideration and must be sufficient to ensure and document that the HACCP plan is being implemented properly.
EDIS - FS14300.pdf 185.5KB 73 downloads
A polite way of saying that it's arbitrary subjective. :smile:
PPS - A very readable analysis of haccp verification requirements is extracted below from the famous US Blue Book. This, i think, is primarily oriented to seafood but the basic principles should be equivalent. As noted in the text, frequencies for certain activities are regulatorily specified, for others the logic influencing the choice of frequency is discussed in some detail. But ultimately IMO the typical result is "subjective".
Blue Book - Chapter10.pdf 56.02KB 85 downloads
P3S - added later - here are two references containing verification "observation" examples of which (a) supports yr original frequency and (b) supports yr modified rate (latter due Kansas State University) -
haccp Verification (a).pdf 127.62KB 109 downloads
haccp Verification (b).pdf 1.89MB 106 downloads
Thank you
TIgo
Dear Tigo -
USDA FSIS expectation is that you have some technical. scientific, historical, and/or statistical basis for your decisions. Charles.C provided some excellent info regarding where you could have started. If you are far past the initial 90 day validation period, then you can gather current (and use historical) data for analysis. I am guessing you have 1 of 2 situations:
- You are not having issues with the monitoring process: In this case, your food safety system is adequate, but you failed to support your verification frequency decision-making. If your system is operating properly, you did yourself no favors by arbitrarily adding to the work load - how did you support the decision to go to daily...because USDA FSIS said so?
- You are having issues with the monitoring process that may or may not be being identified by the verification activity.
a. having problems not caught - then your verification frequency is not adequate
b. having problems being caught - then something is wrong with your training to prevent recurrence
Without all the background info detailed in the FSA, it is difficult to say exactly why this regulation was quoted for your specific situation. One way to get a handle on understanding verification is to go to the USDA FSIS website and search on verification and validation (separately). About a year ago, FSIS published a signficant amount of information regarding their intent to start enforcing the requirement to have supporting decision-making documentation, as well as their recommendations on what is needed. Also, go the AskFSIS section of the USDA FSIS website and search on the same keywords. If you can't find what you need, let me know and I can load some of the major items here.
One thing to realize from a USDA FSIS perspective... Verification does not require that the Verifier duplicate the monitoring activity being conducted by the Monitor. The last thing you need is for the Monitor and Verifier to generate contradictory data...