Consumers & Manufacturers: What you must know about gluten test results using the Nima Sensor

Consumers & Manufacturers: What you must know about gluten test results using the Nima Sensor

In Honor of Celiac Disease Awareness Month, Gluten Free Watchdog is writing a series of articles (the goal is one per day during the month of May) related to the gluten-free diet–currently the ONLY treatment for celiac disease.

Post (#30)…

At Gluten Free Watchdog we have been testing a wide variety of products with the Nima Sensor. It is very difficult to put results into the proper context due to the lack of a published validation report on this device.

Four emerging themes from the approximately 50 products tested: 

  1. Five products testing low gluten (four labeled gluten-free, including one certified gluten-free by GFCO) with the Nima tested below the limit of detection of 1 part per million when tested with the R5 ELISA.
  • Comment: These results may be true false positives (meaning there is no gluten in the sample) OR this device may have an exceedingly low limit of detection (i.e., close to zero).
  1. Barley grain and barley flour tested low gluten.
  • Comment: The antibody used in this device may have a low cross-reactivity to barley.
  1. Recalled gluten-free Cheerios, regular Quaker Oats, and gluten-free rice crackers placed on top of wheat-based bread crumbs all tested smile.
  • Comment: The sampling methodology for this device (i.e., testing a pea-size amount from a non-homogenized sample) is not sufficient to find gluten in samples when gluten is not evenly distributed.
  1. Diluting mustard containing wheat flour with water (as recommended by the Nima website for brightly/intensely colored foods) changed the test result from high gluten to low gluten.
  • Comment: Diluting a sample with water decreases the part per million of gluten in the sample.

The table of test results is available publicly at

Tomorrow’s Post: Thank you to the Gluten Free Watchdog community

Share this post

Comments (17)

  • Doug Reply

    The usefulness of the sensor depends on the criteria one uses on whether or not to eat. For the two situations below, assume the Nima is used according to the manufacturer’s recommendations. Also, keep in mind that as a celiac I think false positives are an inconvenience, but false negatives are a disaster.

    Situation 1. Assume I can’t avoid going to a new restaurant that offers both gluten and gluten-free food. In this situation I choose the restaurant carefully by reputation. Additionally, I typically go through the typical questions of the wait staff and chef. Before NIMA, I’d have to make a final decision on whether or not to eat solely based on the reputation and the answers to the questions. With Nima, I still go through the questions and make my initial decision using the same decision standards. However, the final decision is made once the results of the Nima test are known. Using the test in that manner does not increase the likelihood that I will eat gluten; it only reduces that likelihood. Thus, it adds significant additional decision support.

    Situation 2. Assume I go to a tried and true restaurant that offers both gluten and gluten free food. Rather than simply assume that the food is safe based on a long history with the restaurant and personal relationships with the chef and staff, I can test a portion to gain additional decision support.

    When NIMA is used to solely to reduce the likelihood that gluten will be ingested, it is a useful tool. When it substitutes for other decision support tools (e.g. reputation, Q&A), it does not.

    May 30, 2017 at 2:27 pm
    • Tricia Thompson Reply

      Hi Doug, I appreciate your comments but I do not agree. Testing a pea-size amount of a meal from a non-homogenized sample with the potential for spotty cross-contact with gluten and getting a smiley face does not decrease the likelihood that the meal contains gluten. Would you feel the same way if the person doing the testing had an allergy to peanuts and was using a Nima to test for peanut protein and they were testing a brownie made in a kitchen that also made brownies containing peanuts? This particular scenario keeps me awake at night. Please watch the video

      May 30, 2017 at 3:20 pm
      • Doug Reply

        I appreciate your position, but without real world experience it seems theoretical in nature. Do you have real world usage that says otherwise? Additionally, I question the absolutism in saying Nima does not decrease risk. There is a non-zero likelihood that the pea size portion does contain gluten. The non-zero likelihood means risk is reduced. Finally, the peanut allergy analogy is questionable from a risk perspective. Risk is likelihood and impact. A single exposure to peanuts is more likely to be deadly to someone with a peanut allergy than a single exposure to gluten for a celiac. With greater impact and assuming similar likelihood, the only way to equalize residual risk is through greater risk mitigation mechanisms for the peanut allergy. Perhaps we will have to agree to disagree.

        May 30, 2017 at 4:17 pm
        • Tricia Thompson Reply

          Hi Doug, It is actually common sense that says otherwise. But yes, there is also real world experience. Here is another video you may find useful In it I test a box of “gluten-free” Cheerios that are part of the recall for wheat flour contamination. As mentioned in the video, a pea size amount of Cheerios is less than one Cheerio. A pea-size amount of regular Quaker oats also are tested. While it is possible that spotty contamination may be found in that portion of one Cheerio or the few flakes of Quaker oats, it is not likely given the nature of contamination in oats (quite heterogeneous–this is why the new recommendations are to homogenize a 200 gram portion of oats for testing and then test multiple 1 gram extractions). If you are interested in the statistics of this, please see the excellent studies published by Pepsico/Quaker in the peer reviewed literature

          May 30, 2017 at 4:42 pm
          • Doug

            I’m having trouble fitting your response into the real world scenario I posited in my first message: Assume someone with celiac has to eat in a restaurant that serves both gluten and gluten free food. A decision to eat can be made based (1) solely on scrutinizing the staff and the restaurant, or (2) scrutinizing the staff and restaurant for an initial decision, and if the initial decision is to eat, testing with Nima and not eating if there is a positive result. In which of the two is the person with celiac most safe? It appears that your answer is that since the Nima might have a false negative, the celiac is most safe not using the Nima. That position doesn’t make sense because the Nima decision does not overrule a decision not to eat based on the scrutinization criteria. Using the Nima in that fashion will never increase risk. But your position also leads to one other possibility: People with celiac should not eat in the restaurant at all. Which is your position — make a decision based solely on scrutinization, or don’t eat at the restaurant at all?

            June 3, 2017 at 11:26 am
    • Adrian Reply

      I think it all boils down to whether or not you trust the answer that the device gives. Testing data I have seen suggests that the device is prone to false positive results. While this is only an inconvenience for the individual it has a knock on effect for the industry. The cloud based nature of the device means that all users can access other peoples results and so choose a restaurant and free from food accordingly. If the number of false positives is disproportionate then this could eventually lead to the reduction of choice for the allergic/coeliac consumer as previous “safe” foods are removed from the market and restaurants close due to a bad reputation.
      We are yet to see a full validation report for the device, something that should have been in place before the device was launched on to the market. It does give the impression that early adopters are being treated as guinea pigs, something that is extremely concerning when NIMA turn their attention to allergenic foods

      May 30, 2017 at 7:16 pm
  • Kathy Reply

    I think the Nima will be useful to confirm that you received the gluten free version of something. I would not use it to avoid cross contamination. I was just traveling this weekend and ate at a restaurant that had a regular menu and a gluten free menu. I ordered a dish subbing gluten free pasta for regular pasta after asking the usual questions. When it came out, I was a bit skeptical because I had never seen gluten free angel hair style before. The server double checked and it was the gluten free kind. I was still nervous eating it. The Nima would have been able to tell me if they accidentally gave me regular pasta or not. I would not rely on the Nima to tell me if they used the wrong boiling water though.

    The restaurant the night before sent me fish that looked like it had bread crumbs on it. I had ordered it GF, pan fried. I didn’t know if they had GF bread crumbs. I thought it was going to come plain just seared. I called the waiter back over and sure enough, they had messed up and had to remake it for me. Luckily I hadn’t eaten any. The Nima would have told me those were regular breadcrumbs.

    May 30, 2017 at 5:50 pm
    • Tricia Thompson Reply

      Kathy, I completely agree. When gluten is homogeneous–evenly spread–throughout a product the Nima is useful (assuming of course that issues with false positives, etc can be resolved).

      May 30, 2017 at 5:59 pm
  • Terry Fromm Reply

    Trisha, I also use my Nima sensor when I must eat out or go to family potlucks, etc. Most gluten eating non-Celiacs do not understand gluten or contamination. Prior to my Nima, my mantra was, “if in doubt, do without”. With my Nima, I have some comfort in knowing the food is okay or not. I honestly have not been poisoned since I have used my Nima senser. The original instructions was to stir soups/sauces, take sample from several parts and/or rub the sample around the plate. Until there are better tools available to the Celiac/gluten sensitve lay persons, this is all we have for now. I don’t belive the Nima was ever intended to be compared with the professional testing sources. Also, please note the Nima device firmware has been updated and no longer reflect’s low gluten.

    May 30, 2017 at 7:02 pm
    • Tricia Thompson Reply

      Terry, It is my hope that independently reviewing this device and sharing the data with the Nima team will help improve it. The testing data has also been shared with the group that will be developing a community guidance document for consumer devices for the analysis of food allergens and gluten. If anyone reading these comments is an expert in food allergen analysis there is a call for contribution to the development of this document. For more information please see

      “At the AOAC mid-year meeting, members of the community came together to discuss emerging consumer devices for the analysis of food allergens and gluten. It was noted that in several cases results obtained by consumer devices and by analytical tests used routinely by food industry showed discrepancies. This can, in case of a false positive analysis by the consumer device, lead to brand damage of the producer, and, in case of a false negative, have severe health consequences for affected consumers. It was therefore suggested to develop a guidance document for consumer devices, which addresses the need for performance testing of any such devices according to set standards by the intended user groups and laboratory experts. Results would need to be compared with standard industrial tests.”

      May 30, 2017 at 7:53 pm
  • Shirley @ gluten free easily (gfe) Reply

    Trish, I so appreciate this post and all the work you did in getting these results and summarizing them here to help keep our gluten-free community safe. I find it pretty unbelievable that this tool has not been fully vetted, so to speak. I only wish that the Nima Sensor folks had done the same amount of work comparing Nima’s testing results to independent testing results beforehand. However, if they had, its clear that it would not be on the market at this time.

    I am truly baffled at folks who are continuing to use this testing tool given all the valid concerns you have shared. It’s tempting to re-state all your key points, but they’re there in black and white. Again, I don’t get how folks can ignore them.

    As far as the comment stating that Nima Sensor has been updated to eliminate the “low gluten” reading, that information is still up on their website.

    Thank you again for all your work in this area despite the resistance of some to clear facts.

    May 31, 2017 at 2:41 am
    • Tricia Thompson Reply

      Thanks so much for commenting, Shirley! From what I understand the Nima team has done considerable testing but the validation report has not yet been published. When it is, the scientific community will be able to assess the data and conclusions made by the Nima team. The identification of the antibody used in this device, the extraction solution, the limit of detection for wheat, barley, and rye, etc will help put the test results we are seeing at Gluten Free Watchdog into context. As Adrian states above, this validation report should have been released prior to the release of this device.

      May 31, 2017 at 10:40 am
      • Shirley @ gluten free easily (gfe) Reply

        You’re welcome, Trish. Because the validation report was not released prior to the Nima Sensor’s release, one can only assume that was intentional. That makes me think that they were well aware that there were concerns with the accuracy of their tool. That’s my take anyway. If they never release a validation report or if when they release a validation report, your testing of products with the Nima yields the same results as the independent testing, then we’ll know that they purposely waited until “they got it right.” The fact that so many in the gf community are using this sensor, many of them encouraged to do so by well-known spokespeople in our community, truly disturbs me. Getting gluten unknowingly because you’re trusting a tool which is giving you invalid results can lead to continued intestinal damage and the triggering of other health conditions and autoimmune diseases as we well know. It’s ironic and sad that so many are using this sensor to stay safe while it’s putting their health in jeopardy.

        May 31, 2017 at 5:17 pm
  • Al Reply

    Testing for the presence of gluten is not a simple matter; there are many variables to juggle. Can the Nima can account for all those variables? Perhaps – in time.

    Terry said “ I don’t belive the Nima was ever intended to be compared with the professional testing sources.” I feel testing devices should be compared to the gold standard of validated, professional testing. If isn’t, how are we to judge its accuracy? Should we accept less accurate results simply because it’s a consumer device? I don’t think so.

    There has been a fair amount of hype surrounding the Nima. Bloggers, enthusiastic users/owners, GF websites, GF magazines – all of them have been touting the wonders of this device. I ask myself – what credentials do these people have to vouch for the reliability of this device?

    If I’m going to believe anyone – it’s going to be a non-biased individual with a scientific background. Someone who uses evidence-based practices to form their opinions, and an expert in detecting the presence of gluten. It would be prudent to be open to the analysis/comments/suggestions of Gluten-Free Watchdog.

    June 1, 2017 at 1:07 am
    • Tricia Thompson Reply

      Thanks so much for commenting, Al. I will update the community once a validation report is published and the AOAC group releases a guidance document.

      June 1, 2017 at 1:19 pm
      • Al Reply

        You’re welcome. AOAC’s response will be interesting. Thanks!

        June 1, 2017 at 3:46 pm
  • Peggy Reply

    I have used GlutenEase with good results when unknowingly having gluten. (I am gluten-intolerant.) I recently found “Gluten Assist” by me and my naturals at Walmart. Do you know if it would help with gluten like GlutenEase does? Ingredients in Gluten Assist: Vegetarian Enzyme Blend – 143 mg. Tolerase G – 63,800 PPI; Amylase – 4,000 DU.
    Ingredients in GlutenEase: DPP-IV – 1,000DPPU; Amylase Thera-blend – 15,000DU; Protease Thera-blend (G1) – 95,000 HUT; Glucoamylase – 15 AGU.
    Thanks you for any evaluation you can offer.

    June 13, 2017 at 3:12 pm

Leave a Reply

Your email address will not be published. Required fields are marked *