Are Food Intolerance Tests Accurate?

In order for a test to produce actionable and reliable information, the methodology must be plausible, it should be independently verified, and the results should be consistent and reproducible.  These are really all fundamental aspects of the scientific method in general, and specifically, it’s how clinical diagnostic tests are validated.  If a test doesn’t meet these criteria, then we really can’t rely on the information that it gives us.

In this episode, we cover:

00:32  Upcoming events in Pennsylvania and the UK
04:45  Clinician training program update
13:06  Is food testing methodology plausible?
20:10  What determines testing accuracy?

Podcast: Play in new window | Download

Subscribe: RSS

Chris Kresser:  Hey, everyone, it’s Chris Kresser, co-director of the California Center for Functional Medicine and creator of ChrisKresser.com.  Welcome to another episode of Revolution Health Radio.

Upcoming Events in Pennsylvania and the UK

Before we dive into today’s episode, I want to make a couple of announcements.  The first is that I have two great events coming up.  The first is a seminar in Pennsylvania, about an hour outside of Philadelphia, on Saturday, September 19, from 10 a.m. to 4 p.m., and this is geared more towards healthcare practitioners and clinicians, though everybody is welcome.  I’m going to discuss new trends that will define the future of medicine, the crucial role of the exposome — which I’ve talked about on the blog before; that’s the sum total of all environmental exposures that we encounter from the moment of our conception to the moment of our death — so the role of the exposome in human health and disease, a comprehensive systems model of functional medicine, how to structure and layer treatment for maximum efficacy, how to customize nutrition plans for individual needs, and of course, a Q&A where you can ask me really anything you want on any topic!  We’re going to have a delicious paleo lunch of sliced roast Berkshire pork loin with huckleberries, fresh lavender oil, roasted seasonal vegetables from Harvest Lane Farm, and seasonal fresh fruit also from Harvest Lane Farm.  The cost is $125 for the day, including lunch, and we’ve already had quite a strong response, so make sure to get tickets in advance.  You can do that by going to ChrisKresser.com/PA, PA for Pennsylvania.

The next two events are in London in October.  This is my first time speaking anywhere in Europe and in the UK and probably will be my last time for the foreseeable future.  I don’t have any other trips to Europe or the UK planned in the next few years.  The first event is on Saturday, October 31, Halloween, from 10 a.m. to 4 p.m. at King’s College, and lunch will be provided.  Don’t have a menu for that one yet.  This one is geared more to the general public, and these days when I speak to the public, it’s almost entirely driven by the interests and needs in the room.  I started doing this on my book tour and found it was way more effective than coming in with a pre-planned agenda.  Since there’s often such a wide range of experience and knowledge levels in the audience and desires in terms of what people want to learn about, I’ve found that being spontaneous and just almost doing a Q&A for the entire period but extending topics of interest out and speaking on them in further detail is a really great way to do these events, and I’ve gotten a lot of great feedback on them since I started doing it that way.

The second event is the next day on Sunday, November 1, also from 10 to 4.  We don’t have a venue for this yet, but I imagine it’ll be somewhere in the vicinity of King’s College, definitely in London, and this is oriented more towards practitioners and is much like the Pennsylvania event that I just mentioned.

Some of you might be aware that Darryl Edwards is putting on his paleo event this same weekend, and this is just really unfortunate timing.  I’d chosen these days over a year in advance, and it wasn’t possible for me to change them once Darryl announced that he was doing his event on this weekend a few weeks ago, and apparently those were the only dates that worked for him as well, so sorry about that.  That’s something that neither of us really had control over.

For more info on the public event in London, you can see ChrisKresser.com/UKpublic, and for the practitioner, ChrisKresser.com/UKclinician.  I really hope to see some of you there.  I know there are a lot of listeners in Europe and the UK, and I hope you can make it down, especially since this will be probably my first and only trip to Europe for the foreseeable future.

Clinician Training Program Update

The last announcement is about my clinician training program.  Those of you who are on the early-interest list — not my general email list, but who signed up for notification about the clinician training program — will have already received an email this Tuesday announcing that pre-registration is open, but I know that some of you who listen to the podcast aren’t on any of those email lists, so I wanted to at least make an announcement here.  This training program is really the product of over 10 years of blood, sweat, and tears, and in some ways it feels like my life’s work and what I’m really here to do.  It’s the first and only training to combine all three elements that I believe people need to be a successful practitioner here in the 21st century — functional medicine and ancestral or evolutionary perspective and practice management.  There are certain training programs out there that have pieces — some for functional medicine, some have an ancestral or nutritional perspective, some do a little bit of practice management — but this is the first and only training that combines all three of those, and I’m really excited about it.  I think you will be too.

One thing to know is that there are over 3500 people now on the early-interest list, and I can only take about 200 people for this first round, so we expect it to sell out quickly, and because of that, we’re doing this pre-registration period where we’re giving people a chance to put a deposit down to secure their spot.  So if you’re interested in that, head over to KresserInstitute.com to learn more about the training and guarantee your seat.

Chris Kresser:  Hey, everyone. If you’re like most people I know, you don’t have the time — or the desire or training, for that matter — to sift through the hundreds of medical studies that are published each month. But whether you’re struggling with a chronic health problem or just want to maintain your health and live as long as possible, it’s important to stay informed about the latest research and how it might affect your well-being. And let’s face it — it’s easy to get overwhelmed by all the conflicting stories in the media. You need a source of unbiased information that you can rely on, one that’s easy to understand without too much jargon and full of practical tips that you can put into action. That’s where my website, ChrisKresser.com, comes in. You’ll find in-depth, evidence-based articles on a wide range of health and nutrition topics each week. You’ll also find over 10 free eBooks on subjects like effortless weight loss, healing your gut, thyroid disorders and natural skincare, as well as online classes and programs on optimizing nutrition for fertility and pregnancy and lowering your cholesterol without drugs. What you won’t find is any third-party advertising, corporate influence, or hype-driven, headline-grabbing stories with a lot of drama, but no substance. You just get the real inside scoop on the health issues that matter to you most.

So head over to ChrisKresser.com and check it out, and when you sign up for my email newsletter, you’ll get instant access to my award-winning 63-page eBook, 9 Steps to Perfect Health. It’s like an instruction manual for preventing and reversing disease and living a long and healthy life.

OK, now let’s get on to the show!

This week we have a great question from David about food intolerance testing, and I think it’s going to be of interest to a lot of you, so let’s give it a listen.

Question from David:  Hi, Chris. I recently had a Carroll Food Intolerance Assessment with my naturopath that showed I’m intolerant to fruit.  While I generally don’t eat much fruit to begin with, maybe a banana or some berries a few times a week, avocado and coconut are staples in my diet, and I generally seem to thrive with these healthy fats that also happen to be fruits.  “Intolerant to fruit” seems really vague, so I am hoping to get more information about what might be causing this.  Also, how accurate are food intolerance tests in general and the Carroll Assessment in specific?  I should mention that I have Crohn’s disease that is currently in remission.  I’ve noticed flare-ups with foods high in insoluble fiber, like broccoli and asparagus, but haven’t noticed a reaction to fruit.  I appreciate any thoughts you may have.  Thanks.

Chris Kresser:  OK, thanks, David, for sending that in.  Please do keep sending your questions.  As you know, we’ve recently changed the format of the show and will be doing more interviews and guest hosting, but I’m still going to answer questions, so please do keep sending them in.

I think I’d like to start with a general answer to any question about whether a test is worthwhile, and then we’ll discuss Carroll and other food intolerance tests and their validity or lack thereof.

In order for a test to produce actionable and reliable information, the methodology must be plausible, it should be independently verified, and the results should be consistent and reproducible.  By ‘plausible methodology,’ I mean that the reasoning behind the test should be consistent with what’s currently understood about human physiology and the specific disease processes that the test is designed to identify.  By ‘independently verified,’ I mean that the results should have been published in a peer-reviewed journal and replicated by other independent research groups.  This is a key aspect of the scientific method, of course.  By ‘consistent and reproducible,’ I mean that if the test is run on the same sample multiple times, whether we’re talking about a blood sample or a saliva sample or what have you, it will produce the same or very similar results within a certain range of tolerance.  These are really all fundamental aspects of the scientific method in general, and specifically, it’s how clinical diagnostic tests are validated.  If a test doesn’t meet these criteria, then we really can’t rely on the information that it gives us.

With all that in mind, let’s talk about the Carroll Method that David asked about and see how it measures up.  This was created by Dr. Otis Carroll, who was a naturopathic physician practicing in the 1920s and ‘30s, I believe.  It’s pretty hard to find any reliable information about the methodology of this test, at least from just searching Google, but what I’ve seen makes me pretty doubtful.  One website said, “The food intolerance evaluation utilizes a person’s blood to determine enzyme compatibility with particular food groups or combinations of foods.”  And these include eight major food sensitivities — dairy, fruit, meat, eggs, sugar, potatoes and grains, as well as salt.  And they go on to say, “The majority of patients we treat have one main food intolerance that they need to avoid completely in their diet, as well as one combination of foods that they must keep separate in their diet by 4 to 8 hours.”

Then from another site, “This is a blood chemistry analysis of the cell’s response to different foods, food groups, and food combinations.  The analysis involves electrical responses of blood cells to food, related to Voll testing.”  The Voll test uses an electrodermal diagnostic device invented in 1958 by Reinhold Voll, a German MD who claimed that skin conductance measurements were affected by allergic reactions to foods, and again, as far as I can tell, there’s no peer-reviewed evidence to support Voll testing, and it’s not reproducible.

Now, finally from the Carroll Institute website, which offers training in the Carroll Method of food intolerance testing, they said, “Foods are otherwise consumed, which cause toxemia of the body.  This toxemia is the cause of all disease.”  And then they go on to say that, “You have the option to purchase the equipment so you can utilize this procedure in your own clinic,” so apparently testing is done in the clinic, which is another red flag in my mind if it’s not a validated method.

OK, so that’s what we know about Carroll testing.  Let’s see how it stacks up to the criteria.

Is Food Testing Methodology Plausible?

Is the methodology plausible?  It’s hard to say for sure since there’s little written about it, but I’ve never seen any published research suggesting that using a person’s blood to determine enzyme compatibility with particular food groups would make sense, given what we know about how humans digest and absorb food.  There’s also no mechanistic evidence that I’m aware of that supports electrical responses of blood cells to food or any clear reason why foods would be grouped into the eight categories that they’ve used.  They seem completely arbitrary.  I don’t know why someone would react to all fruit or all meat or any of the other categories specifically that they’ve called out.

The Carroll Method has not been independently verified.  I’m not aware of any peer-reviewed studies supporting this method performed by any group, and obviously if no peer-reviewed study has been performed, it certainly hasn’t been verified by an independent group.

I’m not sure whether it’s consistent or reproducible because these tests are apparently done in the practitioner’s office and there are no studies showing results from split samples where the same patient’s blood was labeled anonymously and run twice.  So I feel pretty safe in saying that the Carroll Method doesn’t meet the standards of modern clinical science for determining the reliability of a diagnostic test.

Let’s move on to another form of food intolerance testing, which is known as ALCAT or MRT testing.  In the scientific community, this is referred to as cytotoxicity testing, and it was popularized in the late 1950s.  It’s riddled with problems, unfortunately, especially at the beginning, and it was eventually banned in 1985 in most states, including California.  Today there’s an improved version that’s still in use under the names of ALCAT, which stands for antigen leukocyte cellular antibody test, and MRT, which stands for mediator release testing, and LRA, which stands for lymphocyte response assay.

This whole category of testing, cytotoxic testing, involves placing a drop of the patient’s blood onto a plate that’s coated with a liquid or dried food extract.  After about 10 minutes, the plate is examined under a microscope and again at 30-minute intervals for up to 2 hours, depending on the test.  Then a technician looks for changes in the structure and shape of the white blood cell, and if there are changes in the size or the rounding or the inactivity of the cell, or if it disintegrates completely, a positive result is indicated.

It might make sense on the surface for someone who’s not familiar with this kind of testing, but there are actually numerous problems with cytotoxic food intolerance testing.  First is that the methodology has never been proven.  It’s never been shown, to my knowledge, that measuring changes in the size of white blood cells, which are supposedly caused by food chemicals, is clinically relevant related to intolerance of a given food.  In other words, if a food chemical changes the size of a white blood cell, we have to show that that is actually related to food intolerance for this test to be valid, and as far as I know, that has never been demonstrated.  Another thing is that there are many non-food ingredients in the food extracts that are typically used in cytotoxic testing that could potentially cause false positives or false negatives.

There’s a noticeable lack of peer-reviewed articles in the literature on cytotoxic testing.  The only articles that have been published and supportive of this kind of testing have been in magazines or non-peer-reviewed journals that are marketed to healthcare practitioners and even patients.

Third, cytotoxic testing is not consistent and reproducible.  The results of a single sample can be read differently from technician to technician at the lab.  This particular method is heavily reliant on the interpretation of the technician, so that can vary from person to person, and different labs produce different results, even for the same sample.  In fact, according to one recent review of this type of testing in the peer-reviewed literature, the authors concluded that, “Well-designed clinical trials should be published before patients are subjected to expensive testing, like ALCAT and MRT, that offer little evidence of effectiveness.”  So I don’t think that ALCAT and MRT tests are a valid way of testing for food intolerance.

OK, moving on, we can talk a little bit about IgG and IgA testing.  This type of testing measures the production of IgG and IgA antibodies to food antigens.  Unlike the Carroll Method and ALCAT and MRT testing, which we’ve already discussed, I do believe that IgG and IgA testing at least has the potential for meeting the criteria I mentioned in the introduction for establishing a reliable test.  However, whether this criteria is actually met depends a lot on the specific lab and the procedures that they use.  But let’s talk about general first.

The methodology is plausible.  Elevated IgG antibodies against food and tissue antigens indicate a failure in immune tolerance, and it’s well established that the first antibody produced after a breakdown in immune tolerance to food antigens is IgA.

Second, IgG and IgA food intolerance testing has been published in peer-reviewed literature and independently verified by more than one research group.

And as for the third question, whether IgG and IgA results are consistent and reproducible, that one’s definitely harder to answer.  Clearly in many or, I would say, even most cases, the answer is no.  There was a study back in 1998 that looked at this, and they drew three samples from one patient on the same day, sent it to three different labs, and two out of three of the labs had numerical variances ranging between 49% and 73%, which is well outside the acceptable standard of 20%.  Even worse, the clinical interpretation of whether the patient should avoid particular foods varied from 7% to 59%, which is really bad.  On the other hand, a later study did find IgG testing to be reproducible using different labs and samples and showed results that were between 82% and 95% identical, which is well within the acceptable standard.

What Determines Testing Accuracy?

So why the discrepancy?  Well, there are several factors that determine accuracy and reproducibility of this kind of testing.  One is the specimen type, so are they testing serum or saliva or some other fluid?  Another is antigen purity.  Several foods, like cabbage and ginger, contain enzymes that can react nonspecifically with the reagents used in IgG and IgA testing, and unless these proteins are purified before testing, you’ll get false positive results.

Another factor is the in-house testing process, which differs from lab to lab.  For example, does the lab run a side-by-side duplicate of the test to ensure that their own results are reproducible?  And if they don’t do that, the results will not necessarily be reliable and consistent.

Another factor is whether the lab tests raw or cooked antigens or both.  The vast majority of us eat some foods cooked and some raw.  You might eat raw cucumber, tomatoes or lettuce or fruits, but you probably don’t eat raw pork or chicken or potatoes or rice.  The problem is that cooking foods changes the structure of the proteins, which in turn changes their antigenicity or their capacity for provoking an immune response, so to get an accurate assessment of intolerance, the antigen that’s tested should be representative of what a patient is actually eating.  In other words, we should be testing specifically for raw cucumber antigen and raw tomato antigen, but we should be testing for cooked pork antigen and cooked chicken antigen.  The problem is that virtually all labs that I’m aware of, with one exception, which we’ll talk about in a second, only test antigens from raw foods, and this just doesn’t work!  Studies have shown that some people can react to roasted peanuts, for example, but not raw peanuts, so that would be a false negative.  And in that same study, while some people reacted to raw meat, they didn’t react to cooked meat, so that would also be a false test result.

There are other issues to consider, but this podcast is already getting a little long, and I think you have a good general idea by now, so I’m going to move on and I’m going to tell you what lab we use in our clinic that meets all of these criteria that I’ve laid out.  It’s the only lab that I know of at this point that’s doing everything right, including properly purifying the antigens, running side-by-side duplicates, and testing both raw and cooked antigens that are appropriate for the various food types that we eat, and that lab is Cyrex Labs.  I’ve talked about them before.  The scientific director is Dr. Aristo Vojdani, who is a highly respected immunologist who has published many peer-reviewed papers in the scientific literature.

All of this said, and despite the trust that I have in Cyrex and their methodology, I do still have some reservations about even this very high level IgG and IgA food intolerance testing.  Number one is that I think we need more research linking IgG food reactions to specific clinical manifestations.  We have studies showing that IgG testing is linked to a breach in oral tolerance and/or disturbances at different steps in the path to oral tolerance that are known to result in food hypersensitivity, but the direct clinical evidence that links food IgG testing to signs and symptoms is still a little bit murky.

So because I’m still not 100% certain about these tests, here’s how we tend to use them in my practice.  With the Cyrex Array 3, which is their gluten/wheat sensitivity panel, definitely the most advanced test of that kind available, if a patient is already 100% gluten free and doesn’t have any interest in eating gluten, I don’t actually think there’s any need to run that test, and it probably won’t be useful anyway because if they’re not eating gluten, they’re not going to be producing antibodies to it.  However, if a patient is still eating some gluten or they would like to be able to occasionally eat gluten if they go out to a restaurant or something like that as part of their 80/20 rule, we will run Cyrex 3 and use it as a means of determining reactivity.  If they have positive markers on that test, given what we know about gluten intolerance and the links between it and many other clinical problems, diseases and health conditions, I will usually advise strict removal of gluten for life.  Now, having said that, there is some research that does suggest it may be possible to restore oral tolerance to gluten, and I’ll discuss this in a future episode, so in some cases, it may be worth retesting after you’ve addressed those issues to see if there’s still antibody production happening.

For patients that are positive to Cyrex 3, we’ll often run Cyrex 4 to see if they have intolerance of other food proteins, and according to Cyrex’s internal research and some peer-reviewed research, people who either have celiac or non-celiac gluten intolerance are far, far more likely to react to dairy proteins and proteins in some of these other foods like coffee and alternative grains and yeast, etc.  So we will do Array 4 to see if they have intolerance of these other proteins, but in this case, we will often do an elimination provocation protocol if they want to try bringing some of these foods back in to see if they react after we have addressed their gut health and some other underlying issues that can lead to food intolerances.  So we don’t necessarily believe that the positive results that come up on Cyrex Array 4 mean that the patient is going to have to avoid those foods for the rest of their life, but we will address those underlying problems and then have the patient retest in the future.

Now, Cyrex Array 10 is their newest test, and this looks at intolerances to many other foods, not just wheat and gluten on Cyrex Array 3 and then the cross-reactive proteins on Cyrex Array 4.  Array 10 looks at all kinds of other foods, fruits and vegetables and meats and things that people typically eat.  And we usually only order this if we’ve addressed gut issues and other underlying issues but the patient is still having problems that seem to be related to food.  I think that it’s better to treat the underlying causes of food intolerances whenever possible than it is to tell someone to remove 25 to 50 foods from their diet forever.  But having said that, if the patient is struggling to deal with some of these things and just wants to feel better, there’s no reason not to run the Cyrex Array 10, have the patient remove some of those foods that are positive from their diet and start feeling better and reduce the immune reactivity while you’re addressing the underlying problems, which hopefully down the line will enable them to eat some or most or even all of the foods that they’re showing reactivity to in the future, because the reactivity is typically caused by a breach of the gut barrier or an infection or bacterial overgrowth in the small intestine or dysbiosis or any of the things that compromise gut health and oral tolerance in general, and if that’s the case, then it makes sense that addressing those things could lead to resolution of those intolerances in the future.

OK, so I hope this was helpful.  This is an area where there’s a lot of misinformation and misunderstanding out there, and unfortunately there just haven’t been very many reliable tests available, not many labs that are actually run by people who have the scientific background and understanding in this area to create a legitimate and reliable assessment, and Dr. Vojdani at Cyrex Labs definitely fits that bill.  I believe he even invented IgG food intolerance testing.  I think that’s accurate.  So he’s been involved in this for a very long time.  And Cyrex, if you go to their website, they have some great information about their tests.  They have some free webinars for clinicians, where you can learn more about the technology, and if you are a clinician or a healthcare practitioner, I would definitely recommend doing that.

OK, that’s the end of this episode of Revolution Health Radio.  If you appreciate the show and want to help me create a healthier and happier world, please head over to iTunes and leave us a review.  They really do make a difference because more reviews means a higher ranking, and a higher ranking means more people listening.

If you’d like to ask a question for me to answer on a future episode, you can do that at ChrisKresser.com/PodcastQuestion.  You can also leave a suggestion for someone you’d like me to interview there.

If you’re on social media, follow me at Twitter.com/ChrisKresser or Facebook.com/ChrisKresserLAc.  I post a lot of articles and research that I do throughout the week there that never makes it to the blog or podcast, so it’s a great way to stay abreast of the latest developments.

And if you’re on the East Coast, don’t forget about the seminar in Pennsylvania on Saturday, September 19, for 10 a.m. to 4 p.m.  You can get more info on that at ChrisKresser.com/PA.  And if you’re in the UK or Europe and you’d like to come see me speak, that’s Saturday, October 31, from 10 a.m. to 4 p.m. for the general public and Sunday, November 1, from 10 a.m. to 4 p.m. for clinicians.  For more information, ChrisKresser.com/UKpublic for the general public and ChrisKresser.com/UKclinician for the clinician event.

Thanks again for listening, and I’ll talk to you next time.