Must Read Study Courtesy of Quaker on Testing Oats for Gluten
Gluten-containing grains skew gluten assessment in oats due to sample grind non-homogeneity. Fritz RD, Chen Y, Contreras V. Food Chemistry. February 2017;216:170-175. Published online ahead of print August 12, 2016.
Disclosure: The authors declare that they have no competing interests. Ronald D. Fritz, Yumin Chen, and Veronica Contreras are salaried employees of PepsiCo Inc. or Quaker Foods and Snacks (QFS), a subsidiary of PepsiCo, Inc., which funded this research. QFS has a commercial interest in gluten-free foods.
What follows is a summary of the above-mentioned article. If you are a testing laboratory, a supplier of oats, or a gluten-free manufacturer it is highly recommended that you read the article in full. It is available at http://www.sciencedirect.com/science/article/pii/S0308814616312614
Note: This write-up was written by Tricia Thompson, Founder of Gluten Free Watchdog and shared with Quaker prior to posting to ensure accuracy.
Why this Article (and not just this summary) is a Must Read
Based on the 5½ years Gluten Free Watchdog has commissioned testing through an independent laboratory we have seen first hand that it is sometimes very difficult to evenly distribute gluten within a food sample. This may occur with grains such as oats or in multi-ingredient grain products when only one ingredient is contaminated.
As a consequence of gluten not being distributed evenly within a product, samples extracted for testing may have different results. On some occasions additional sample grinding leads to more homogeneous results but often it does not. This has led to a GFWD hypothesis that finding a relatively low level of gluten in a sample (i.e., above 5 parts per million but below 20 parts per million) may be suggestive of a larger gluten problem within the sample as a whole.
The article “Gluten-containing grains skew gluten assessment in oats due to sample grind non-homogeneity” recently published in Food Chemistry demonstrates just how risky it is to assume that a single test result between 5 and 20 ppm of gluten from a sample of oats is indicative of the gluten content of the oat sample as a whole.
Oats are easily contaminated with kernels of wheat, barley, and rye. These kernels are not evenly distributed within a sample of oats and are difficult to homogenize within a sample of oats for accurate gluten testing.
In market survey: Commercially available gluten-free oatmeal was assessed for gluten via R-biopharm’s Ridascreen Gliadin R5 ELISA by an independent laboratory. Of the 636 samples (45 grams for a packet of oatmeal and 50 grams taken from a canister of oatmeal) tested, 10 sub-samples (based on “homogenizing” the entire sample and testing a single 0.25 gram extraction) tested at or above 20 ppm of gluten. Twenty additional sub-samples tested above the lower limit of quantification for gluten of 5 ppm but below 20 ppm of gluten. The researchers designated these samples as “positive yet compliant” meaning that they contained quantifiable gluten but the level of gluten was below the FDA’s threshold for gluten-free labeling of 20 ppm of gluten.
The lab conducted additional testing on the 20 samples of oatmeal initially testing between 5 ppm and less than 20 ppm gluten. A total of 11 (0.25 gram) extractions were tested from each of these individual samples of oatmeal. At least 9 (and as many as 11) of these 20 samples contained on average above 20 ppm of gluten based on testing multiple extractions from each sample. This despite the fact that the first extraction tested contained less than 20 ppm of gluten.
Gluten distribution in oats spiked with wheat: One kernel of wheat was added to 12 (50 gram) samples of pure gluten-free oat groats. Each sample was ground by an independent lab to homogenize (evenly distribute) the gluten. Each ground sample was then tested in triplicate. What remained of each 50-gram sample was sent back to the researchers. PepsiCo’s/Quaker’s analytical team tested the entire amount of sample in 0.25-gram extractions (approximately 2,300 analysis were performed).
Results of testing were highly heterogeneous (variable) and were skewed right meaning that gluten contamination remained concentrated within a relatively small number of 0.25-gram extractions. Study authors write, “with highly skewed distributions like this, the determination of gluten content in oat groats via a single 0.25-g sample test becomes error prone. This is because a few of the 0.25-g samples possess large amounts of gluten while others have received just a fraction of it. This leads to the potential for misdiagnosis (i.e., concluding either a sample average
is <20 ppm when it is not, or that all possible test results are <20 ppm when they are not, depending on one’s interpretation of the 20 ppm regulatory threshold).”
Study authors used the testing data to develop probabilities for various true average gluten levels that a single 0.25-gram test result reads less than 20 ppm. They found that if the true average gluten level in a 50-gram sample is greater than 60 ppm gluten then the chance of getting a false negative result is less than or equal to 5 percent. If the true average gluten level is between 20 and 60 ppm of gluten the probability of a false negative result was about 60% for a true gluten average of 20 ppm and about 10% for a true gluten average of 55 ppm.
Study authors write, “Our research has looked into the use of a single 0.25 g test amount to assess the gluten content of ground groats when a gluten-containing kernel is in a sample. The results indicate that a homogenous grind is difficult to attain and that resultant 0.25 g test results tend to be log-normally distributed. It appears this phenomenon is at play in finished goods as well (i.e., where whole grains have been cut and flaked), as our repetitive tests on ‘gluten positive’ servings from our ‘in-market survey’ suggest. The log-normal distribution of gluten outcomes complicates the assessment task since a single observation (or even a number of them) may not accurately represent the rest of the sample (since a substantial range of outcomes is inherent in skewed distributions like this.). Consequently, conventional use of a single 0.25 g test should be treated with caution, particularly when a positive compliant gluten reading has been obtained with that first reading.”
Gluten Free Watchdog Questions for the Authors
- GFWD is of the opinion that when gluten is detected in amounts between 5 and 20-ppm gluten in difficult to homogenize samples (e.g., naturally gluten-free grains, multi-ingredients products when only one ingredient is contaminated) there are often higher levels of gluten to be found in the sample. Your research seems to support this hypothesis. Based on your research, do you have any recommendations on how to proceed with additional testing when a positive yet compliant result is found in any given food product?
Response: We recommend the “mass extraction” method which we shared previously (and is detailed on your site). GFWD Note: The mass extraction procedure is detailed HERE.
- As you know R-biopharm recently updated instructions for testing oat samples. It is now recommended that labs homogenize 200 grams of sample and test a 1-gram extraction. How do you think these new recommendations will increase the accuracy of a single test result in terms of accurately reflecting a true average gluten level?
Response: We are working through a second paper focused on sampling. We are unable to share it while it’s under review, but are hopeful that it will be publically available soon.
Thank you to the PepsiCo/Quaker team for your research—for choosing to submit it to a peer-review journal (and thus open it up for scientific vetting) rather than simply publishing it internally. You have made an important contribution to our understanding of the limitations associated with testing difficult-to-homogenize food samples. Here’s to the development of better methods of homogenization and sampling to improve the likelihood that test results from a limited number of samples are representative of the entire sample.