Factors to Consider in Choosing a Testing Laboratory

Choosing an analytical testing laboratory or the type of test methods to confirm or identify the condition of a food or beverage product is not an easy task.


Choosing an analytical testing laboratory or the type of test methods to confirm or identify the condition of a food or beverage product is not an easy task. Decisions need to be made on the types of testing that are appropriate or necessary; whether to perform the testing oneself or to outsource; and how to ensure one’s confidence in the competence of those performing the activity and thus ultimately in knowing that the results are credible and accurate.

Notwithstanding the legal ramifications of selling an “inherently dangerous product,” a foremost aim of all food and beverage processing plants is ensuring that the products they have produced are safe for human consumption. Plant management ensures this by making certain that the raw materials coming into or produced in the plant are not contaminated or otherwise adulterated, and that their manufacturing processes do not inadvertently render a product unsafe.

Human nature being what it is, the degree of one’s willingness to accept the risk of an adverse outcome (for example, for a food plant’s inaccurate test results) is inversely proportional to the degree of rigor expended in managing the risk; and it is generally directly proportional to the potential adverse economic consequences of failing to manage the risk appropriately. All other things being equal, judgment in choosing between in-house product safety testing and outsourcing this activity should be based on the complexity of the task, the confidence in the competence of those doing the analysis, the relative risk of contamination or adulteration of the food product itself and, as mentioned above, the willingness to accept the risk of an inaccurate test result.

Photo courtesy of USDA/ARS, Scott Bauer

However, some major determinants in choosing to do some or all of the testing in-house or outsource this activity lie in the careful evaluation of your organization’s needs and goals. In-house testing requires accounting for both the direct and indirect (overhead) costs of maintaining a laboratory operation (e.g. the human and capital resources), but it also needs to account for the liability of having an inaccurate test outcome. Based on the product being manufactured, decisions need to be made on the required complexity of the methods, the availability of staff with appropriate skills, and the anticipated volume of testing. In addition, decisions on the types of testing and whether to perform it internally are also related to the relative risk category of the food product being manufactured. Products with a higher propensity toward microbial contamination or greater possibility of chemical adulteration will have to have more safety and quality testing done than ones that are lower on that scale. Many manufacturers address the latter by outsourcing any product with an adverse initial finding (e.g. presumptive positive test result) to an independent, third-party laboratory, and address the former by maintaining a small screening testing capability. This approach demonstrates due diligence as well as keeping the costs as low as feasible.
 

Choosing a third-party lab. So, if based on the issues mentioned earlier, the manufacturing plant decides to outsource some or all of its product testing, how does it decide which independent, third-party testing laboratory to choose to perform this work?

Factors to consider are location and distance from the manufacturing activity, analytical turnaround time, cost of the particular test, and reliability of results.

  • The cost of a test is certainly a factor in the decision-making process, however it should never override the other factors.
     
  • The logistics of location and distance from the manufacturing facility play a large role in deciding which testing laboratory to utilize. While perhaps still appropriate for confirmatory testing, if the testing facility is not sufficiently close, or if it does not have a courier service capability, then the efficiency of transportation is likely to render this particular laboratory inappropriate for routine testing, regardless of the other factors involved.

  • Analytical turnaround time is another significant factor in the decision tree, especially in the food industry where many products are highly perishable. It does the food producer no good if the time it takes to obtain the results is not consistent with the allowable product hold time. Thus, this should be a major factor in deciding on a laboratory.

  • The quality and reliability of test results is probably the hardest to judge because it is the least empirical. However, there are a number of approaches that can be taken, each of which has its pros and cons:

1. Have a prospective laboratory complete a questionnaire (self audit). This would be useful in obtaining categorical data, such as turnaround time, frequency of repeated or redone work, and success rate in PT activities, but it would not give a good sense of qualitative factors, such as competence of the staff or how errors are mitigated. If the questions are not worded properly it could also be prone to the garbage-in/garbage-out syndrome.

2. Conduct a quality (supplier) audit. This would cover areas deemed important by the laboratory, and typically would involve the plant staff evaluating the laboratory’s method performance, results verification, records management, and corrective action processes. The producer usually prioritizes these visits based on the importance and criticality of the supplier and/or product, with those of lower criticality often assessed, instead, through a self-audit or high-level review. The down side is that this activity requires a significant expenditure of dedicated resources and some degree of specialized training.

3. Rely on testing results evidenced through proficiency testing or blind samples. This provides the manufacturing plant with real-life information on the laboratory’s ability to obtain generally acceptable results. The down side is that the available proficiency testing programs may not offer fully relevant matrices or analytes reflective of the producer’s product. In addition, the time and effort to set up blind check samples may be costly and time consuming, and the results may not have sufficient statistical validity.

4. Rely on laboratory recognition by conformity assessment bodies. Focusing on recognition of a quality control testing laboratory as assessed against internationally recognized quality system standards by an accreditation or certification body combines the components of all the activities mentioned above into one process. While, unlike an accreditation body, a registrar cannot confirm the technical competence of a laboratory, both systems look at the laboratory’s integrated structure of responsibilities, activities, resources, and events so as to ensure that the laboratory is capable of meeting the international quality requirements. In essence, both look at and evaluate the prospective laboratory’s commitment and compliance to its quality policies or objectives (what to do), its quality and technical procedures (how to do it, in the cases of accreditation bodies) and evidence (records) of implementation. The negative aspect of this is the cost and time incurred by the laboratory involved in getting these recognitions, but an additional benefit to the producer is the defensibility of the testing result and the reduction of incurred legal liability.


With so many options and variables, it is not easy for a food and beverage processing plant to decide the best manner of obtaining quality and safety testing data on its products. There are many aspects to consider, each of which may fit the manufacturer’s needs to a different degree and may change as circumstances dictate. But because the choice has considerable impact on the safety of the food product and its fitness for human consumption, it is critical that all things are considered, as the final choices made on testing can have significant ramifications.

February 2012
Explore the February 2012 Issue

Check out more from this issue and find your next story to read.