Study: 8% of Gluten-Free Products Test Over 20ppm, and 15% of "Gluten-Free" Products Certified by GFCO Contain Gluten at Over 10ppm

[trents]

I'm beginning to wonder if Mom's of America as well as some in our forum community (including me) went off half-cocked in this matter. 

[Scott Adams]

I'm not convinced that is the case, taking a few more tests of each lot still would not rule out a hot spot in a sample that reached Moms Across America. I will try to reach out to their lab for a comment. While false positive results can occur with the Romer Labs AgraStrip Gluten G12 test, they are generally not very common if proper testing protocols are followed. The test is considered reliable and is extensively used in the food industry to ensure products meet gluten-free standards. Proper handling, sample preparation, and adherence to testing protocols are key to minimizing the risk of false positives. Health Research Institute Labs carried out each test in duplicate, to ensure accuracy and reproducibility of the results. 

[Scott Adams]

PS - This part of that statement concerns me: "We asked the manufacturers to run multiple tests on their retained samples from the lots named in the MAA report..."

So they are relying solely on the companies in question to perform and report follow up testing results on those samples. To me it seems like a 3rd party lab working for the GIG should be doing all follow up tests on those lot samples. So far, we've heard only denials from the food companies in question, and each has a vested financial interest in creating plausible deniability (I'm not claiming here that they are doing this, but just pointing out that the possibility exists, and I believe that an independent lab should be handling this follow up testing on behalf of the GFCO).

[RMJ]

On 6/13/2024 at 11:13 AM, Scott Adams said:

I contacted Moms Across America and found out that the lab used in this study was Health Research Institute Labs - https://hrilabs.org, and they used the Romer Agrastrip Gluten G12 test, which is highly accurate at detecting the 33-mer peptide of gliadin down to 4ppm. The scientists at HRI carried out the tests in duplicate to ensure accuracy and reproducibility of the results, so I trust that their results are accurate. 

I have also reached out to each company in the study for comment that had too high gluten measurements per the FDA or GFCO, and so far none have replied.

This will be covered in a future article.

I just saw this comment, sorry to be a week late.  The AgraStrip test is qualitative (yes/no gluten), not quantitative.  Yet the HRI results are reported to two decimal places, not as yes/no.  Might you mean the Romer AgraQuant test? 

Romer AgraStrip

Romer AgraQuant

[Scott Adams]

They probably meant that, but I cut and pasted their response. It is likely that they made an error and it was the AgraQuant.

Here is Trader Joe's reply to our inquiry about the Moms Across America study:

Quote

Thank you for reaching out. While we can’t speak to the specific tests they performed on any product, we can share with you that our Gluten Free Almost Everything Bagels are made in a gluten free bakery, and each “lot” of bagels is tested to ensure that it meets our gluten free standard. Our standard is aligned with GFCO at less than 10 ppm, which is more stringent than the FDA standard of 20 ppm.

If you can share the lot code on the Trader Joe’s Gluten Free Almost Everything Bagels that were tested, we can contact our supplier to verify the test results for that specific lot. Again, every lot is tested, and the product will only be sold at Trader Joe’s if those tests verify that it meets our standards.

We approach product quality and integrity seriously. When there is an issue, we take action quickly, aggressively investigating potential problems if there is any doubt about its safety or quality.

Thank you,
Nakia Rohde
Public Relations Manager

I replied to inform Nakia that there is no lot number included in the study, and she replied back with:

Quote

Thanks for sharing these details. We will share these results with our supplier, who, as indicated in my earlier email, is an entirely gluten free bakery facility. Again, without a lot code, neither we nor our supplier can compare results against any of the results of the regular testing done on this product. And again, every lot of product that is delivered to Trader Joe’s tests at a level below 10ppm.

 

[RMJ]

Too many errors from HRI.  Vanilla cookies listed on certificate of analysis when chocolate chip was provided for testing, item descriptions not detailed enough (for example: can’t tell what type of King Arthur gluten-free flour was tested), telling you they used a test that canNOT give the quantitative results they reported, and not all items on certificate of analysis include lot numbers (were items removed from packaging prior to being sent to HRI)?  

IF they used the AgraQuant test the LOQ on the HRI certificate (5ppm) doesn’t match the AgraQuant LOQ (4ppm)

I used to work in pharmaceutical Quality Assurance and audit laboratories that provided testing to FDA regulated companies.  If I found this many errors I’d be telling the company to find a different laboratory.

[Scott Adams]

I agree that the lab seems to have made some errors, and am writing up some questions for them and will post their reply. I don't believe that the errors mentioned would negate their findings, but they do deserve clarification.

[Scott Adams]

On 6/20/2024 at 6:35 PM, RMJ said:

Too many errors from HRI.  Vanilla cookies listed on certificate of analysis when chocolate chip was provided for testing, item descriptions not detailed enough (for example: can’t tell what type of King Arthur gluten-free flour was tested), telling you they used a test that canNOT give the quantitative results they reported, and not all items on certificate of analysis include lot numbers (were items removed from packaging prior to being sent to HRI)?  

IF they used the AgraQuant test the LOQ on the HRI certificate (5ppm) doesn’t match the AgraQuant LOQ (4ppm)

I used to work in pharmaceutical Quality Assurance and audit laboratories that provided testing to FDA regulated companies.  If I found this many errors I’d be telling the company to find a different laboratory.

I got back responses to the following questions:

1) In an earlier question for you about what test was used, you answered: "Health Research Institute used a special test kit, the Romer Agrastrip Gluten G12 test," however this is not a quantitative test, but is a yes/no, positive/negative test. Did you instead mean that the test used was the Romer AgraQuant, which is a quantitative test?:
https://www.romerlabs.com/en/romerlabs/msds/filedownload/file_id/2CB4A0D8F0C44E3ABA22961AB61AD3A8/ 

Dr. John Fagan, Chief Scientist:

Quote

We used the AgraQuant Gluten G12 test. I mistakenly stated “Agristrip” in the previous communication. We used the quantitative test. We tested, first, the undiluted extract in duplicate. Those tests exceeded the calibration curve. We then made two dilutions of the extract and tested those in duplicate. Duplicates matched in both cases and the two dilutions were quantitatively consistent. That was the value that we reported.

2) On the testing certificate you list the item tested as Made Good Vanilla Cookies (Lot/Exp: 1023; 08052024 - with 56.10 gliadin detected), but Moms Across America has indicated that the actual product tested was Made Good Soft Baked Double Chocolate Cookies. How could such an error take place in your laboratory, and why hasn't it been corrected on your report?

Dr. John Fagan, Chief Scientist:

Quote

We were provided with a spreadsheet list of sample names and descriptions, in which that sample was initially listed as Made Good Vanilla Cookies. Then our team used that list as it was provided without properly checking. After we reported the results we were cross-checking the results with Moms Across America and they noted the discrepancy and we corrected our records and asked them to correct theirs. The sample tested was certainly Made Good Soft Baked Double Chocolate Cookies. Zen: Maybe the way to address this is: “There was mis-communication between the HRI and MAA teams. This has now been corrected."

3) The item descriptions are vague, for example we cannot tell which King Arthur Flour was tested, as they make several different types. Why weren't the full product names used in the descriptions for more clarity?

Dr. John Fagan, Chief Scientist:

Quote

This is being corrected. This information will be provided shortly. 

4) Were the samples tested removed from their packaging before they reached your lab, or did your lab receive all samples in their original packaging?

Dr. John Fagan, Chief Scientist:

Quote

We do not test consumer products unless they are provided in the original packaging sealed and clearly untampered. We check this carefully and we maintain traceability throughout the testing process in the lab.

5) Why are so many items missing lot numbers and expiration dates? For example, the item with the highest detected amount of gluten was Trader Joe's Everything Bagels, and I have a bag of those and the lot and expiration info is contained on the small plastic clip that holds the bag closed, yet this is not included in your report.

Dr. John Fagan, Chief Scientist:

Quote

We are now providing lot numbers for all of the samples. Only best-by dates are present on some products. Those are now included in the Certificate of Analysis.

6) If you used the AgraQuant test the LOQ on the HRI certificate (5ppm) and it doesn’t match the AgraQuant LOQ (4ppm). Can you explain this?

Dr. John Fagan, Chief Scientist:

Quote

This is the result of miscommunication within our team. We are correcting the Certificate of Analysis to state that the LOQ of the AgriQuant test is 4 ppm.

I agree we all need to be more rigorous in future. This testing and report experienced a lot more scrutiny because there are regulations around gluten-free, and because it is a recognized health-sensitive issue. In future, we will apply consistently what we have learned here. And, Zen, I think there is an interesting learning here, namely that MAA can garner more attention testing sensitive issues like gluten-free. 

We are sending the revised Certificate of Analysis in a separate email. We are completing that revision right now and will have it to you shortly.

Here is a link to their revised certificate:

https://www.celiac.com/images/Certificate_of_Analysis.pdf 

[RMJ]

Thanks for the update! It’s great that they did have the more detailed descriptions and lot numbers, and provided a revised certificate.  However, if they had a decent Quality department, the original certificate never would have been issued. It’s hard for me to believe that they’re ISO 17025 certified with that poor a certificate.

[Scott Adams]

After finally receiving the lot number from Trader Joe's for the Everything Bagels tested by Moms Across America for their study, I received this response from Trader Joe's yesterday:

Quote

Thanks for sharing the lot code information. We have confirmed the code with our supplier – connecting it to a specific production batch.  The results of testing on this lot/batch of Trader Joe’s Gluten Free Almost Everything Bagels are in compliance with the FDA requirements for “gluten-free” labeling and meet the GFCO standard (which is also Trader Joe’s standard) – below 10ppm.

Best,
Tara

Tara J Miller
Vice President, Marketing

So basically all companies we've contacted about high gluten results that were found in the study are circling their wagons and claiming that the batch in question is normal and gluten-free. Can both sides be right? Yes, especially if they have issues with hot spots within a batch, which would be the best way to explain the conflicting results.

[trents]

If "hotspots" are the explanation, what can food companies do to address the problem? Even if they tested every batch, we're talking about variations within batches, not between one batch and another, correct? What is reasonable to expect of them?

[RMJ]

It would be very interesting to know how the duplicate samples were prepared, and how close the numbers were. If genuine duplicates with hot spots, I’d expect the two numbers could be quite different.

[Scott Adams]

A class action lawsuit was filed last Friday against Trader Joe's, and the danger here is that some companies may discontinue their gluten-free products, or smaller companies may go out of business due to such lawsuits:

 

[placey0918]

Oh my gosh! I was super sick after eating a Trader Joe's  gluten-free muffin. I emailed them about it. Asked them about their gluten-free certification since I realized I didn't see any stamp and was wondering how they can claim its gluten-free without it.  They responded by stating they actual use a local gluten-free bakery that are certified . Odd, since that companies name is not on the package and no certification mentioned.  So they passed my email on to them.

That company is located here in WA. They emailed me, and apologized and stated they are not sure how I could be sick because their bakery is 100% gluten-free. I responded by asking about their certification process and asked what equipment do they use to test pmm, because I noticed on their website  it was mentioned using the Nim device and that's not a commercial grade testing device. I also asked why their companies name is not listed on the Trader Joe’s product with their certification information and  contact information. 

Crickets, they never responded to those questions and sent me a Trader Joe's gift card in the amount of the muffins.  

Once I learned of the company, it happens to be one I haven't  trusted  even though it states it is certifiied gluten-free. I was very ill for wks after I had a cupcake from this company.  I should have known since the certified stamp does not look anything like the usual 2 types we see on legit products. 

I was and still am planning on filing a complaint with the Health Dept. and our L&I dept. For both companies. So imagine  my shock when I opened my email and saw this article.  Wow!

[trents]

8 minutes ago, placey0918 said:

Oh my gosh! I was super sick after eating a Trader Joe's  gluten-free muffin. I emailed them about it. Asked them about their gluten-free certification since I realized I didn't see any stamp and was wondering how they can claim its gluten-free without it.  They responded by stating they actual use a local gluten-free bakery that are certified . Odd, since that companies name is not on the package and no certification mentioned.  So they passed my email on to them.

That company is located here in WA. They emailed me, and apologized and stated they are not sure how I could be sick because their bakery is 100% gluten-free. I responded by asking about their certification process and asked what equipment do they use to test pmm, because I noticed on their website  it was mentioned using the Nim device and that's not a commercial grade testing device. I also asked why their companies name is not listed on the Trader Joe’s product with their certification information and  contact information. 

Crickets, they never responded to those questions and sent me a Trader Joe's gift card in the amount of the muffins.  

Once I learned of the company, it happens to be one I haven't  trusted  even though it states it is certifiied gluten-free. I was very ill for wks after I had a cupcake from this company.  I should have known since the certified stamp does not look anything like the usual 2 types we see on legit products. 

I was and still am planning on filing a complaint with the Health Dept. and our L&I dept. For both companies. So imagine  my shock when I opened my email and saw this article.  Wow!

Products labeled "Gluten Free" but not "Certified Gluten Free" do not require a certification "stamp". The "Gluten Free" standard is less than 20ppm (the FDA standard) whereas the "Certified Gluten Free" standard is less than 10ppm. Was the Trader Joe muffin in question labeled as gluten free or certified gluten free?

[Scott Adams]

I would only add that the article this thread is based on also found that 8% of products sampled which were certified gluten-free did not meet thee required certification standards, so it raises the question of just how much protection such a certification actually affords anyone.

[Patty harrigan]

It’s horrifying to me. I pay the extra for that peace of mind. They certainly need to tighten up their inspection protocols.

[placey0918]

18 hours ago, trents said:

Products labeled "Gluten Free" but not "Certified Gluten Free" do not require a certification "stamp". The "Gluten Free" standard is less than 20ppm (the FDA standard) whereas the "Certified Gluten Free" standard is less than 10ppm. Was the Trader Joe muffin in question labeled as gluten free or certified gluten free?

The Trader Joe's was not labeled certified,  BUT they said it came from a company that claims to be certified.  Now that company has stated on their website they certified and has a circle image on packaging that says ' certified gluten free ', it doesn't look like any other gluten-free stamp I've seen and when I asked about their certification process they never replied back.  Something doesn't sit right with me about that.  Wouldn't you want to share that info? 

Thanks for the clarification. Hopefully the health Department can shed some light on this for me. 

 

 

 

 

[Scott Adams]

As this earlier post of mine mentions, Trader Joe's uses the 10ppm standard:

 

[AlyO]

Hi everyone, have there been any further updates on this topic? Anything more from GFCO? It seems like 3rd party testing is required at the very least. 

Many of these products I have regularly fed my little one.  

Thank you Scott for your work on this. 

[Scott Adams]

We've done a detailed analysis on the Moms Across America study and found that 15% of GFCO Certified "Gluten-Free" products that were randomly tested by them were found to contain gluten at over 10ppm.

We'll comment more on this later, but we updated the topic title to include this new revised amount (it said 10% before).

32 out of 46 products were certified gluten-free by the GFCO, and 5 were tested and found to contain more than 10ppm, or 15% of them.

[Scott Adams]

We published a new article today here:

 

[Lotte18]

Scott, As always thanks for providing our community with this much needed information.  Like so many, I just could not figure out why I would suddenly feel intestinal pain--everything thing I ate was "Certified gluten-free."  After reading about Simple Foods I threw out their crackers and cookies--products that I was eating on a regular basis.  Their Almond Flour cookies claim to be grain free and gluten free but are "made on a line that also processes tree nuts and WHEAT." After the Mom's Across America test results I believe that this was the source of the problem.  While I'm grateful for companies that provide food for our community, I would remind them that we represent most, if not all of their business.  If I didn't have celiac I'd never buy any of this stuff!  If they want to survive they really have to turn this around.  

[RMJ]

Here is an update from Gluten Free Watchdog, indicating that the high results on at least one item may be false positive.  I am a scientist and used to develop ELISAs, the type of test used.  Non-specific binding such as is described can be a real problem with these tests, so this is a scientifically reasonable explanation.

possibly false positive

[trents]

Thanks for this input and the link, RMJ. As I said some time ago in this discussion, I fear the celiac community may have gone off half-cocked in their reaction to these claims. Having said that, I still think random, unannounced drop in testing added to the mix of protocols would be a good idea.