It isn’t just oats that have gluten cross contact issues

We’ve known for well over a decade that standard oats are highly likely to arrive at a mill from the farm containing errant wheat, barley, and rye grain. But we also have a problem with errant gluten-containing grain showing up in other naturally gluten-free grains, seeds, and legumes, including millet grain and dried lentils.

This is happening in products labeled “gluten-free” as well as those labeled “certified gluten-free.” And it doesn’t matter if the food manufacturing plant is dedicated gluten-free. Why? Because these grains, seeds, and flours may be coming into contact with wheat, barley, and rye in the field, during harvest, during storage, and during transport. A grain may not become any “dirtier” in a dedicated facility but it isn’t going to become any cleaner either.

What is being done to address this problem?

Some manufacturers and certification organizations are taking steps to decrease the likelihood of an errant wheat, barley, or rye grain showing up in a gluten-free product. For example, according to information sent to Gluten Free Watchdog from a manufacturer certified by GFCO:

“Beginning January 1, 2019, GFCO is adding an additional requirement for verification that whole, intact grain, seed, bean, pulse and legume products meet the 10 ppm gluten threshold. For any whole, intact grain, seed, bean, pulse or legume, the manufacturer/processor must use an appropriate sampling method and visual inspection to show that the material contains less than 0.25 gluten-containing grains (wheat, rye or barley) per kilogram, in addition to the antibody-based testing they are currently doing.”

How did GFCO arrive at this threshold?

In the GFCO article, “The Use of Visual Examination for Determining the Presence of Gluten-Containing Grains in Gluten Free Oats and
 Other Grains, Seeds, Beans, Pulses, and Legumes” by Allred et al. published in the Journal of AOAC International, researchers write:

“Based on these assumptions, in the worst-case scenario, one contaminating GCG would contain 10.5 mg protein, 9.45 mg of which would be gluten. Therefore, to stay below the GFCO threshold of 10 mg/kg (10 ppm) gluten, 1 GCG/kg would be the highest level of gluten contamination that would be acceptable to consumers. In order to reduce the risk of a consumer purchasing a grain product with 1 or more GCGs/kg to <1%, GFCO recommends a threshold of 0.25 GCG/kg for visual examination.”

Is this threshold low enough?

• It is a step in the right direction for GFCO.
• As the paper states this threshold would result in 1 gluten-containing grain (GCG) in every 100 (40 gram) servings.
• Compare this to the protocol followed by Quaker/Pepsico for their gluten-free oats (1 gluten-containing grain in every 1,000 (40 gram) servings):
  • Described as “an attribute-based sampling plan requiring zero GCGs (i.e., zero GCG assumed equal to <20 ppm gluten in a serving here) in 3000 servings (40 g) of product in order to pass, with this being evaluated on a production lot basis. This approach is designed to detect a 1 in 1000 serving noncompliance rate with 95% confidence.”
    • Source: Manufacture of Gluten Free Oats for Gluten Compliance at the Serving Level: An Attainable Industrial Standard (Fritz and Chen [Pepsico] Letter to the Editor, Journal of AOAC InternatIonal, epub 2018).  This article is available for free at https://www.ingentaconnect.com/content/aoac/jaoac/pre-prints/content-jaoac_180100#

A concerning aspect of the GFCO article

To determine whether the threshold of 0.25 GCG/kg for visual examination was achievable for suppliers, GFCO evaluated oats from two gluten-free oat-processing plants—Cream Hill Estates* (purity protocol oats) and Grain Millers (sorted oats).

The article includes a table (see table 3 at this link https://www.gluten.org/wp-content/uploads/2017/12/Oat-Study.pdf). Data from Cream Hill and Grain Millers is presented side by side** and includes information on the number of samples, the
 total weight of product analyzed, the number of wheat grains detected, the
number of rye grains detected, the number 
of barley grains detected, and the total number of gluten grains 
detected.

It would not be surprising if readers looking at the data come away with the conclusion that in general sorted oats from Grain Millers contain fewer wheat, barley, and rye seeds than purity protocol oats from Cream Hill. However, important information about the data is missing from the article that prevents the data from being put into context.

Based on correspondence with Cream Hill (with permission granted to post the information):

• The data included in the article for Cream Hill is for 7 growing seasons 2004 to 2010 as this is when their seed lab testing was done (we do not know the number of growing seasons or the specific years for the data from Grain Millers but sorted “gluten-free” oats hit the market years after purity protocol oats in approximately 2013).
  • In other words, the data from Cream Hill includes the beginnings of the purity protocol.
  • The number of seeds/1,000 grams in 2004 was more than double the amount found in the next highest year.
  • If the authors do not want the data between the two suppliers compared (as has been stated) then the data should not be presented in the same table and additional contextual information should be provided.
• No rye seeds were found in Cream Hill oat seeds as reported in the GFCO article.
  • Instead the seed data is for barley.

*Cream Hill Estates is no longer in business.

**Despite the data being presented side by side in table 3, the article stresses, “This study does not serve as a validation for either the Purity Protocol or the mechanical sorting method of producing gluten free grains, but rather demonstrates that achieving the proposed threshold is possible under both systems. Because the starting material can vary widely from season to season and even truckload to truckload, no method for generating gluten free grains, pulses, seeds, beans, or legumes can ever be considered validated, and examination of each lot from beginning to end is necessary.”

At Gluten Free Watchdog we continue to support the use of oats produced under a gluten-free purity protocol (list available at https://www.glutenfreewatchdog.org/news/oats-produced-under-a-gluten-free-purity-protocol-listing-of-suppliers-and-manufacturers/. We are supportive of the use of Quaker gluten-free oats due to the transparency of Quaker and their extensive and detailed protocol. At this time, we are not supportive of the use of any other mechanically and/or optically sorted “gluten-free” oat product.

If you have any questions, please do not hesitate to contact me.