Natural Products Insider is part of the Informa Markets Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Hand holding lens.

Selecting a contract lab to ensure valid lab results

Verifying a contract lab first requires transparency on the methods and processes that affect its reliability of results.

The stuff that sinks businesses—like product recalls, warning letters and class action suits—are most often related to insufficient product testing. Using the wrong method, or not testing at all, has taken steam out of many in the food and supplement industry.

Let’s admit it: There are a lot of problems with testing. It costs time and money, which exponentially increased when a product fails the test. Ignorance is bliss, until it causes a crisis.

Focusing on supply chain preventive controls under the Food Safety Modernization Act (FSMA) is great, but the devil is in the details. A brand can visit its suppliers all it wants, but if it is not reviewing and ensuring the validity of results for the substances that pose a risk to a product, it hasn’t covered all the bases.

For the average consumer products firm, it’s not just about pathogens and allergens, but also about substantiating label and marketing claims. The past decade has shown that testing-validity issues are central in many expensive actions related to product and ingredient identity, as well as “natural” and nutrient content claims for food. Paper certifications like organic and non-GMO (genetically modified organism) are slowly adopting testing, too.

Contract labs are like any other supplier and should be verified with the same level of rigor. Contract labs who are for-profit have a business to run. Their ability to obtain a correct answer depends on the employment of numerous incremental controls that help to ensure the probability of their correctness. Most, if not all, of these controls cost money or time. Method verification and validation, running duplicate samples versus single runs, performing calibration curves weekly or monthly instead of daily, and using certified reference materials are all steps that impact both the profitability of a lab and the reliability of results—except in opposite ways. So, while many responsible labs are doing their best to achieve accurate results, remember that every for-profit contract lab is working to strike the balance of profitability and reliability on a daily basis.

Certifications come second. Methods, references and processes should be evaluated first.

Certifications are great, but don’t often cover the method a brand needs to use. Beyond certifications like ISO 17025, verifying the lab first requires transparency on the methods and processes that affect its reliability of results. Some of the top questions include:

1. Is the method compendial (i.e., peer-reviewed and approved by experts) or otherwise validated? FDA, AOAC or USP are common compendial-source methods listed on specification. (Compendial methods should have a unique method number or reference a particular document). If the lab has validated a method, what were some of the key elements of the validation? Has the method been validated in multiple independent labs?

2. Does the method fit the matrix or type of a sample? A method validated for an herb may not also be valid for an extract of that herb. Validity to a particular matrix is an issue that’s often overlooked. Ask the lab if it foresees any issues with interference from the composition of the sample, because it may not be considering this issue before testing the sample.

3. How long has the lab been running the method, and how many samples are tested weekly with the method? Expertise in testing is often method-specific, and proficiency comes through repetition. Consider it a red flag if a lab says it only uses this method occasionally. It’s typically expensive and time consuming to perform all the necessary instrument calibration and verification for just one or two samples, so this is where shortcuts may be taken.

4. What calibrations and quality control (QC) procedures are employed? How are circumstances that compromise reliability of results—like incomplete extraction or signal suppression in chromatography—addressed through ongoing calibration of the instrument? Examples may include running an instrument blank every few samples to determine no carryover, a daily calibration curve with freshly-made reference standards, and weekly or monthly control samples.

5. What are the reference materials? Information like source, purity and method of verification should be the “standard” for reference material certificates of analysis but are sometimes not. Also, the purity of a reference standard may have been determined at date of manufacture, but not re-verified at the time of sale, after the material may have degraded in storage. Without verified references—the critical factor on which results are based—a test method is a house of cards waiting for the next stiff breeze.

6. What is the analytical range or limit of quantification (LOQ), and how was the range determined? Methods are only applicable within a certain range of concentration. Some of the most sensitive and specific methods that go down to parts-per-trillion are useless or require modification at higher concentrations. Likewise, excellent precision and accuracy needs to be demonstrated at the lowest LOQ, which requires that the LOQ was properly determined.

7. What is the expected within-lab variation of the method? This is a/the central metric for method precision, usually in the form of percent of RSD (repeatability standard deviation) for numerical results. Alternatively, this number can be measured as a false rejection or acceptance rate for binary methods like identity or pathogens. The more of a substance there is, the more precise the method should be.

8. Is this method included in the scope of a lab certification? ISO 17025 certification, the most common lab-proficiency certification standard in the United States, is something that many labs hold and promote. What many labs fail to disclose are the specific methods under the scope. If a method wasn’t in the scope, then it wasn’t reviewed and is not certified.

9. Does the test report include the method reference, as well as correct units and significant figures, and other relevant information?

10. How are disputed and out-of-specification (OOS) results dealt with? Testing is clear sailing until an unexpected or OOS result emerges. Does the lab automatically re-run samples that are OOS? What are the steps they take in an OOS investigation?

A brand must demand transparency from its lab, but it must also share information with the lab. If the brand doesn’t provide information on the composition of the sample, expected range of the analyte and, in some cases, how it was processed, then labs must take a leap of faith on selecting or approving the appropriate method.

And method validity and reliability abhor leaps of faith. The one takeaway to remember about ensuring valid results: It takes two to make a thing go right.

Blake Ebersole has led a number of botanical quality initiatives and formed collaborations with dozens of universities and research centers. As president of NaturPro Scientific, Ebersole established quality compliance and product development services for supplements and ingredients such as ID Verified™. Follow him on Twitter at @NaturalBlake.

Contract Labs at SupplySide West

Learn more about selecting a contract lab from Blake Ebersole during the “Effectively Partnering with a Contract Laboratory” Workshop on Wednesday, Nov. 7, at SupplySide West in Las Vegas. Underwritten by American Testing Lab, Eurofins, Micro Quality Labs and New Jersey Laboratories. 


Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.