Sampling Plans in Food Safety: A 101 Guide

Sample Plan.jpg

“Product sampling” is met with uncertainty by many food manufacturing organizations because it’s a discipline of food safety that is just that: uncertain.

There’s no magic bullet or direct set of instructions for exactly how to have an effective product sampling plan, and the ambiguity can drive food safety professionals batty.  In addition, although guidance is provided, the statistical verbiage can be overwhelming.

Product sampling is a topic we dove into during our recent course at Chestnut Labs in partnership with IDFA. Out of that, we put together a brief “101” for items to keep in mind when approaching sampling for your testing strategy.

Why Sample?

The only way to be 100 percent certain that a “lot” of food product is free from pathogens is to test the entire lot of material. That’s not possible because the product material is destroyed during the testing process, so there would be nothing remaining.  Considering that, product sampling plans pull aside portions from each lot to evaluate the properties of the food, without utilizing (destroying) all of the material.

What is a Lot?

A food lot is defined as the following by the FDA:

Sampling Plan Considerations-.jpg

Lot means a batch, or a specific identified portion of a batch, having uniform character and quality within specified limits; or, in the case of a drug product produced by continuous process, it is a specific identified amount produced in a unit of time or quantity in a manner that assures its having uniform character and quality within specified limits.

A representative sampling plan should reflect the composition of the lot from which it was drawn, and logically, a higher the number of samples will better-represent the lot.

How to Start?

In developing a sampling plan, many factors should be taken into consideration including:

  • Properties of the food
  • Production process
  • Storage conditions
  • Risks
  • Target consumers

Evaluating and understanding the above factors will help in implementing a sampling plan.  Knowing your product and processes will help in determining the organisms of concern, the type of sampling plan to employ, and number of samples to collect.

Sampling Plans

After reviewing and gathering information on the above factors, the type of sampling plan can be selected.  Three types of sampling plans are summarized below:

1. Two Class Attribute Plans

Plainly, microorganisms in the lot are either acceptable or not acceptable.  This type of plan is used for qualitative testing, such as tests performed to detect pathogens.

2. Three Class Attribute Plans

In this plan, microorganisms in the lot can be acceptable, or marginally acceptable, or unacceptable. The three class attribute plan is used for quantitative testing, such as tests for aerobic organisms, coliforms, etc.  When using a three-class plan, the number of marginally acceptable results must be defined.  If the predefined number of marginally acceptable results is exceeded, the entire lot should be rejected.  

3. Variable Sampling Plans

Variable sampling plans can also be used to accept and reject lots with marginally acceptable counts. In variable sampling, the mean and standard deviation of organism counts are used to accept or reject lots.

Some Additional Food for Thought

When sampling from any lot, here are some guidelines to follow:

  • With all sampling plans, samples should be collected randomly while avoiding bias with regards to where product is taken from.
  • When possible, submit unopened containers to the lab for testing.
  • Make sure the sample is homogeneous.
  • Deliver samples to the lab as soon as possible:
  • Try to keep frozen products frozen and refrigerated samples cool, using coolers or ice packs that are preconditioned prior to shipping.
  • Use aseptic technique:
  • Use only sterile collection bags and collection tools, and always wear gloves.

For more information about sampling and how to make sure you have an effective strategy in place, contact Chestnut Labs today.

 

Bryce WilksChestnut Labs