[ October 8–11, 2019    McCormick Place    Chicago, IL USA ]

Newsroom                     

“Validation” is a term that has been used with steadily increasing frequency and is important to the meat and poultry industry.  It’s a word with many meanings for justifying, confirming, or proving the validity of something.  Most often and as of recent, the term has focused on and has been used regularly in the context of Food Safety.  Validation can be a paper exercise of simply gathering records, data, and documents to prove the safety of a product and/or process.  Validation can also include minimal to extensive data collection (e.g. microbiological) on current products and/or processes to help confirm or support decisions being made regarding the safety a product and/or process.  Periodically, however, these approaches aren’t sufficient to achieve validation needs and a planned scientific study is then in order.  So…what makes for a “good” validation study and what considerations should be made during the entire process of the study?  The following 10 steps may be useful to guide you through the validation study process.

 

  • Step 1: Clearly define the objective of the validation. It’s important to keep the study straightforward. Avoid confounding the results due to uncontrolled outcomes due to study complexity.  A great way to address this is to very specifically generate the research question (what are you trying to prove) and avoid temptations to have multiple questions answered in a single study (i.e. “reduction of salmonella in sausage” vs. “reduction of salmonella in cheese wieners”).
  • Step 2: The design of the experiment is important. The end goal should be clear. For example, are you trying to simply reduce pathogens in a product/process or is there a regulatory reduction requirement that must be achieved?  The design must be able to attain the study’s goal, so care and consideration of this is crucial to having a useful product.
  • Step 3: If performing a validation on a category of products, carefully select parameters to establish for testing and consider the impact “group” testing will have on results. Can a “worst case” product/process be incorporated into the study?  Have considerations been made regarding attributes such as specie type, salt concentration, inclusion of sodium nitrite, product diameter, type of process, etc?  Variations to any of these factors, if not considered and incorporated into the study design, can render a validation study unusable or significantly limit its application.
  • Step 4: Carefully select the measurements that will be captured. A thorough review of what factors are important for the food safety of a product/process and what measurable attributes are critical to applying and verifying the validation study is essential. Attributes such as pH, time, temperature, water activity, etc. are commonly built into validation study deigns and are later incorporated into HACCP plans for monitoring and verification of the studies ability to support food safety.  Keep in mind…you must be able to live with the critical parameters you identity through the validation study process.
  • Step 5: Selection of microorganisms. Options for which microorganisms to include for microbiological based study do exist including pathogenic (highest correlation to study-application expectations), indicator (low correlation to pathogenic reduction, or surrogates (helpful but not consistently well correlated). It’s almost always advised to stick with the bugs you’re concerned about and you can prove you can control…the pathogens.  ..inoculation levels should also be considered.  High levels (~8 logs) are used when showing a reduction is desired while low levels (~3 log) are used when the interest is to demonstrate growth control.
  • Step 6: Determine the number of replicates needed. Replication demonstrates repeatability and strengthens the confidence that the results found in the study are highly likely to also be expected in the actual process. Replications are repeating experiments from start to finish to infer that any normal variation that would be expected during production would also be addressed in the validation study.  “True replicates” best accomplish this (different study day, raw materials, ingredients, etc.) offer the greatest confidence to capture variation and build that into the study.  Often a statistician can offer suggestions regarding the number of replicates needed but two or three (depending on expected variation within the study is generally sufficient).
  • Step 7: Determine the sampling frequency. Carefully plan what your microbiological sampling frequency will be as you can go back and change this after the experiment is completed.
  • Step 8: What type of microbiological sampling method will be used? Thought regarding the type of sampling method should also occur. Direct plating counts how many living bacteria are present but doesn’t capture how many bacteria are injured and how many might be able to recover from their injury.  Enumeration methods are sometimes included in studies to recognize this and follow a more conservative approach.
  • Step 9: Analyzing the data and developing the final report. After the study is completed, it’s important to generate a clear, concise (but thorough) report. The report becomes the communication piece of the study and details the validity for the product/process.  It will be viewed and will need to be understood by many so clear communication is critical.  It would be advisable to follow a scientific study format for presenting the information (abstract, introduction/background, material & methods, results and discussion) to aid in developing a good report.
  • Step 10: Implement the validated study properly. After all the work has been done, the most important thing to do is implementation. This is where all the pre-planning and proper execution during the study pay off. Implementation means you are obligated to: 1) follow the parameters measured, 2) use the study for only the product or products investigated, 3) establish and/or support critical limits based on results, and 4) be prepared to explain and defend study if needed.  If implementation is thought about during Steps 1-3, impact and value of the study can be more likely realized.