As the founder and publisher of both the website “Gluten-Free Watchdog” and the blog “Gluten-Free Dietitian,” Tricia Thompson, MS, RD, saw a need for extensively testing products labeled gluten-free to ensure that consumers who medically require a strict gluten-free diet can eat safely. With this background, Thompson also identified the urgent need to determine the level of compliance with the FDA’s gluten-free ruling. So, she partnered with fellow researcher from the Celiac Disease Center at Columbia University, Suzanne Simpson, RD, LD, to retrospectively analyze Gluten Free Watchdog’s testing of 158 products bearing a gluten-free claim. Forty six of the products were certified gluten-free by a third party organization.
The researchers collected three samples of most products, which were all tested twice using either the sandwich ELISA method, or a combination of the sandwich ELISA and the competitive ELISA methods in those cases where the product contained fermented or hydrolyzed ingredients. The researchers found that 95% of the foods they tested were below the 20ppm gluten threshold, and 87% were below 5ppm gluten. However, two certified gluten-free products tested over 10 ppm, but below 20 ppm, which are not in compliance with the FDA’s ruling. This is due to the fact that the FDA holds certified products to the certifying body’s standards as opposed to the less than 20 ppm threshold.
A total of 8 products in this study tested at or above 20 ppm. Two (4.3%) products with a gluten-free certification tested above the 20ppm gluten threshold, which is surprising given the fact that the certifying agencies require products to meet a threshold lower than the FDA’s – usually under 5-10ppm. Additionally, six (5.4%) products labeled gluten-free but not certified gluten-free tested at or above the 20 ppm threshold.
Since this is such a hot button topic within our community, Tricia Thompson and Suzanne Simpson agreed to answer some further questions about this.
What’s the take home message, or the overall result, or your study?
Tricia: The vast majority of the products tested below the lower limit of quantification of 5 parts per million of gluten using the sandwich R5 ELISA Mendez Method. There were a handful of products both certified and labeled gluten-free that tested at or above 20 ppm gluten so there remains room for improvement. Consumers want to feel confident that ALL labeled gluten-free foods test below 20 ppm. They also want to feel confident that ALL certified gluten-free foods test below the more stringent gluten levels set by a certifying agency.
Suzanne: As Tricia mentioned most of the products tested, tested at less than 20 ppm. However, some tested above that limit and this is very disappointing. The claim “gluten-free” should be a trustworthy claim meaning 100% of products test at <20ppm. In light of the results of this study and other studies manufacturers should develop routine testing procedures to ensure that, when they claim “gluten-free,” they meet all of the criteria of that claim.
Why did you test each product in the study twice?
Tricia: Products are tested in duplicate for a couple reasons. We want to make sure the sample tested is reasonably homogenized, meaning that any contaminant (gluten) is “evenly” distributed within a sample. Sometimes this is tricky to accomplish. We also test samples in duplicate to guard against laboratory error. Manufacturers should also follow this protocol when having products tested at a lab—either third party or in-house.
Some of the products had to be tested using the competitive ELISA method in addition to the sandwich ELISA. Why was this necessary? In other words, can you explain the differences between the two tests?
Tricia: A sandwich ELISA is used to test intact or relatively intact gluten protein. When gluten protein is hydrolyzed—meaning it is broken apart into fragments—the sandwich may not be able to detect all the fragments. It is complicated, but in short, the sandwich R5 ELISA requires two epitopes or antibody binding cites. Two epitopes may not be available when gluten protein is hydrolyzed so the sandwich will not detect these fragments. A competitive ELISA is used to detect protein fragments. The competitive R5 ELISA requires only one epitope. At Gluten Free Watchdog we use the competitive ELISA along with the sandwich ELISA when there are concerns that a product may be contaminated with hydrolyzed gluten.
Suzanne: The Sandwich ELISA is a very useful test but for the reasons Tricia outlined, the competitive ELISA is more appropriate for certain foods/products that are hydrolyzed.
How much do you think consumers rely on gluten-free certification for their shopping choices? Do you think your results will have any impact on this?
Tricia: Subscribers to Gluten Free Watchdog tend to be very concerned about cross-contact and they are looking for products that test as low as possible for gluten. Some of them rely on gluten-free certification to help guide their choices. It is my hope that this study causes every manufacturer—certified or not—to take a look at their testing protocols and make sure they are doing everything possible to ensure the veracity of the claims they are making on product labels.
Suzanne: Many patients look for the certification logo on foods as it gives them confidence that the food is gluten-free. The results of this study could have an impact on whether consumers will go out of their way to buy foods that are certified gluten-free.
How do your results compare to the FDA’s recent study about gluten levels in foods with and without gluten-free claims?
Tricia: The FDA study found approximately 1% of labeled gluten-free foods to contain above 20 ppm gluten.
In your study, you mention that it has been estimated that, by 2017, gluten-free offerings will become a $6.6 billion market. Other current references have even shown the gluten-free market to already be at $20 billion. Do you think that, as the gluten-free market grows, we will see a greater percentage of products that are not in compliance with the FDA’s ruling?
Tricia: It depends. If manufacturers who do not understand the gluten-free labeling rule start making gluten-free claims this may happen, but in our experience at Gluten Free Watchdog the percentage of foods testing at or above 20 ppm gluten has held fairly steady. Now that the gluten-free labeling rule is being enforced it would be nice to think that all foods will be in compliance.
Suzanne: I agree with Tricia. Three studies have been published recently testing foods for gluten content. I am hopeful that manufacturers will be made aware of these studies and take the gluten-free claim very seriously, doing regular testing of foods for gluten content and having strict protocol in place to prevent cross-contact.
You list a few ways that consumers can increase their confidence in a product, which includes “reading the ingredients list and contacting manufacturers to as about the gluten-testing protocols” (p. 2). On the flip side, what should consumers do if they think they have been “glutened” by a product?
Tricia: Consumers who feel they have been glutened or who come across a product they feel is labeled incorrectly (e.g., labeled gluten-free yet contains malt) can contact an FDA consumer complaint coordinator in their area http://www.fda.gov/Safety/ReportaProblem/ConsumerComplaintCoordinators/default.htm. They can also contact the Center for Food Safety and Applied Nutrition’s Adverse Event Reporting System. The phone number is 240-402-2405. In addition, if a consumer comes across a product they feel is mislabeled, Gluten Free Watchdog will also contact manufacturers and FDA as part of a group complaint. Consumers should take photos of the product, including the product name, ingredients list, gluten-free claim, lot number, and UPC code. They can send these photos to me at firstname.lastname@example.org.
Suzanne: In addition to the steps outlined by Tricia on reporting the food to the FDA and Gluten Free Watchdog, consumers should individually contact the manufacturer of the food to inquire about their testing procedures and manufacturing practices. The company should be alerted immediately if a product they are selling could be contaminated.
Based off of your test results, is there a difference in manufacturers’ compliance of certified foods versus non-certified foods?
Tricia: Based on our testing data, there is a small difference in compliance with the FDA’s gluten-free rule of less than 20 ppm gluten.
We know that legumes and naturally gluten-free grains are more likely to come into cross-contact with gluten-containing grains during the growing, milling and manufacturing processes. Were there certain food categories that had higher amounts of products that tested at or above 20 ppm?
Tricia: Not really. Food testing at or above 20 ppm gluten included a hot beverage, bread product, bread crumbs, crisp bread, cookie, hot cereal, spice, and tortilla.
Suzanne: I tested a variety of legumes a few years ago for gluten content and none of them contained greater than 20ppm. I also spoke with bean growers and despite the crop rotation regimens utilized in growing legumes, it sounded like, due to the difference in size, the risk of cross-contact was low.
You mention in the final summary of your study that the ultimate goal is to have 100% of gluten-free manufacturers in compliance with the FDA’s ruling. In your opinion, what types of oversight should manufacturers of gluten-free products have in order to reach this level of compliance?
Tricia: Manufacturers, including those who are certified, must have a rigorous testing protocol in place. Products should be tested at least periodically by a third party testing facility using a fully validated assay. In my opinion, manufacturers should use the assay FDA has indicated they will use—the Ridascreen Gliadin R5 ELISA (R7001) Mendez Method manufactured by R-Biopharm. If they choose not to use this assay, they should make sure the assay they do use is validated against this assay. Manufacturers must make sure they know what they are doing when it comes to testing. And, finally, manufacturers should be fully transparent regarding their testing protocols. They should not be telling consumers that testing is proprietary information.
Thompson, T., & Simpson, S. (2014). A comparison of gluten levels in labeled gluten-free and certified gluten-free foods sold in the United States. European journal of clinical nutrition.