Below is the full transcript of Episode 98 of the Autoimmune Wellness Podcast. This transcript is provided for accessibility and reference.

Title: Food Sensitivity Testing: What Tests Can and Can’t Tell You (Ep 098)

Mickey: What if a blood test could tell you exactly which foods were causing your symptoms?

You send in a sample, and a few weeks later, you get a personalized report. Green foods are fine. Yellow foods are questionable. Red foods are supposedly causing inflammation or other reactions in your body. No elimination diet, no wondering whether that headache, digestive flare, or increase in joint pain was actually connected to something you ate.

Now, this sounds pretty amazing, and if you’ve been living with autoimmune disease and trying to figure out whether food is contributing to how you feel, I completely understand why you might spend hundreds of dollars on a test like this.

But does it actually work? Because there’s a very important distinction here. A laboratory can accurately measure something without that measurement actually diagnosing the thing you want to know, and that’s exactly where the controversy around food sensitivity testing begins.

So today, we’re going to look at the most common tests being marketed for food sensitivity, particularly IgG panels, MRT, and ALCAT. We’ll talk about what they actually are measuring, whether those measurements have been validated against real food reactions, and what major allergy organizations have concluded after reviewing the evidence.

​

Mickey: Welcome back to the Autoimmune Wellness Podcast. I’m your host, Mickey Trescott. As always, this podcast is for educational purposes only and is not a substitute for individualized medical advice, diagnosis, or treatment.

In our last episode, we talked all about the types and categories of reactions a person can have to foods. We covered everything from food allergy and celiac disease to digestive intolerances, histamine, and those frustrating situations where a person experiences a reproducible reaction, but we don’t yet have a clear biological explanation for it.

Today, we’re going to ask the next logical question: Is there a test that can identify these foods for us? And I really want to approach this with some nuance. There are absolutely validated medical tests for specific food-related conditions. That’s not what we’re questioning today. But what we are talking about are the broader commercial food sensitivity panels that claim to identify dozens or even hundreds of foods that might be causing symptoms.

We’ll look into the science behind the claims, including how we determine whether a test is actually validated. We’ll also talk about some fascinating emerging research suggesting that there may be food reactions conventional allergy testing doesn’t capture, which makes this conversation even more interesting.

So let’s get into it.

[00:02:59] IgG Food Sensitivity Testing

Mickey: Let’s start with probably the most common type of food sensitivity testing, and that is IgG. These tests usually take a blood sample and then measure IgG antibodies against dozens or sometimes hundreds of different foods. You get a report showing which foods produced higher levels of IgG, and those foods might be labeled as having low, moderate, or high level of reactivity. The implication is that a high IgG response means your immune system is reacting negatively to that food and that removing it might improve symptoms or reduce inflammation.

There’s just one major problem, and that’s not what food-specific IgG has been shown to mean. IgG stands for immunoglobulin G. It’s a class of antibody produced by the immune system, and unlike IgE, which we talked about in the last episode in connection with immediate food allergy, IgG antibodies to foods are commonly found in healthy people who eat those foods.

In fact, the presence of food-specific IgG can simply be evidence that you’ve been exposed to and are consuming that food. And Ig4, which is a subtype frequently discussed in this context, makes the interpretation even more problematic. Ig4 responses to food can be associated with immune tolerance. So in allergy immunotherapy, for example, increases in allergen-specific Ig4 actually occur as tolerance develops.

So you can see the problem. A lab might actually be accurately measuring these IgG antibodies to eggs or almonds, that doesn’t mean the lab measurement is wrong. What’s unsupported is the next step, interpreting that antibody as evidence that eggs or almonds are causing your symptoms or inflammation.

And this isn’t just my interpretation of the research. Major professional allergy organizations have specifically recommended against using food-specific IgG or IgG4 testing to diagnose food allergy or intolerance. The European Academy of Allergy and Clinical Immunology published a position paper concluding that IgG4 testing should not be used for the diagnosis of food-related adverse reactions.

The Canadian Society of Allergy and Clinical Immunology has also issued a formal position against food-specific IgG testing, noting that positive results are expected in healthy adults and children and can reflect exposure and tolerance rather than hypersensitivity. And the American Academy of Allergy, Asthma, and Immunology advises against using IgG food testing to diagnose food intolerance or sensitivity.

That’s an important consensus. It doesn’t mean that somebody who takes an IgG test and discovers that they feel better after changing their diet is imagining it. We’ll talk about that later because I think that experience deserves a thoughtful explanation. But what it does mean is that the IgG result itself hasn’t been validated as evidence that the food is causing the reaction.

And this is a great example of why the word validated matters. You can have a lab test that accurately and reproducibly measures the biological marker. That is analytical validity. But then we have a different question. Has that marker been shown to reliably distinguish between foods a person clinically reacts to and foods that they tolerate? That’s clinical validity.

With commercial IgG food sensitivity testing, that’s where the claim falls apart, and there can be a real downside to getting that wrong. If your report flags 30 foods and you remove all 30, you’ve potentially made your diet much more restrictive without good evidence that those foods were actually causing a problem.

In our community, where many people may already be following some form of an elimination diet, that’s especially important. So if an IgG panel tells you that you’re highly reactive to a food, I would not consider that result alone a reason to eliminate that food. The much more useful question is, what actually happens when you eat it?

And we’ll come back to exactly how we answer that question later in the episode.

[00:07:16] MRT, ALCAT, and Other Food Sensitivity Tests

Mickey: IgG panels aren’t the only tests marketed for food sensitivity. There are also several other approaches that claim to detect reactions through changes in blood cells or other biological responses.

Let’s start with MRT, or mediator release testing. MRT starts with the idea that adverse food reactions may involve inflammatory mediators released by immune cells. A sample of your blood is exposed to individual food or chemical extracts, and the test measures changes in the relative volume of cells and plasma.

Those changes are interpreted as evidence that cells have released mediators in response to the substance. So the test isn’t directly measuring a specific inflammatory mediator and showing, for example, that a particular cytokine increased in response to tomatoes. It’s using changes in blood cell volume as an indirect marker of mediator release. That’s an important distinction. For MRT to function as a clinically useful food sensitivity test, we’d need good evidence that this laboratory response reliably predicts which foods actually trigger symptoms in people.

Ideally, we’d want independent studies comparing MRT results with meaningful clinical reference, such as a blind food challenge, and showing good reproducibility, sensitivity, and specificity. And that is exactly where the evidence is lacking. There have been many studies using MRT-guided diets, including research in people with IBS, and some have reported symptom improvement, but that isn’t really the same as validating the test itself.

If somebody receives MRT results, changes many foods at once, and then feels better, we still don’t know whether the foods classified as reactive were correctly identified by the test. To establish that, we’d need to show that people consistently react when they’re challenged with MRT-positive foods and tolerate MRT-negative foods, and that the test performs reliably when studied independently. And we just don’t currently have that level of evidence.

Then there’s ALCAT, which has been around for decades. ALCAT also exposes a person’s blood cells to individual food extracts, but it evaluates changes in the size and number of white blood cells after that exposure. Those cellular changes are then interpreted as indicating intolerance or sensitivity to that food. Again, there’s a biological-sounding rationale here. Immune cells absolutely can change when they’re activated, but that’s not enough to validate this diagnostic test.

Again, we need to know whether the particular cellular changes ALCAT measures reliably correspond to symptoms that a person might get by eating the food. And research on ALCAT has not established adequate reproducibility or clinical validity for diagnosing food intolerance or sensitivity, and it isn’t accepted as a validated diagnostic method by mainstream allergy organizations.

And then I want to talk about even other types of testing where the evidence is even weaker. Hair analysis might claim that examining maybe a little sample of your hair can identify food intolerance or sensitivity, but there is not an established physiological basis or validated method showing that the composition of your hair can identify what foods cause symptoms when you eat them.

Electrodermal testing, sometimes associated with Vega testing, measures changes in electrical resistance at the skin while different substances are introduced into the testing system. Again, controlled research has not demonstrated that this can reliably diagnose food allergy or sensitivity. And then applied kinesiology uses changes in muscle strength while a person is exposed to or holds a suspected food. There’s no validated evidence that this response can diagnose a food reaction.

You’ll also see bioresonance and similar energetic testing approaches marketed for food intolerance. They claim to detect or analyze electromagnetic frequencies associated with foods or the body, but they do not have an established, validated basis for diagnosing food reactions.

So there’s a spectrum here. MRT and ALCAT are measuring actual changes in blood samples, while some of these other approaches rely on methods that have much less biological and clinical support. But the question that I want you to ask is the same for all of them: Has this particular measurement been shown in well-designed, preferably independent studies, to reliably predict whether somebody reacts when they eat the food?

For these broad food sensitivity tests, we do not have the evidence I’d want to make that leap, and that’s especially important when the result isn’t just interesting, right? The result might lead you to remove 10, 20, or 50 foods from your diet. And before you make that decision with the kind of nutritional and quality-of-life consequence, I think the scientific bar should be much, much higher.

[00:12:31] What Does It Mean for a Test to Be Validated?

Mickey: So next, we’re going to talk about what it means for a test to be validated. We’ve used the word validated a lot, so I want to explain this. This is where you can start evaluating some of these claims for yourself. First, there’s analytical validity. Can the laboratory accurately and consistently measure the thing that it says it’s measuring?

This is important, but it’s only the beginning. The bigger question is clinical validity. Does the result actually predict what happens when a person eats the food? And conceptually, that question is pretty straightforward to test. Let’s say a blood test identifies eggs as highly reactive and rice as non-reactive.

We can compare those results with what happens when the person actually eats eggs and rice under controlled conditions. Ideally, we’d do this with a blinded oral food challenge where neither the participant, and in the strongest design, neither the participant nor the person assessing the response knows whether they’re receiving the suspected food or a placebo.

The foods are prepared so that they are difficult to tell what it is, symptoms are measured systematically, and challenges are repeated when appropriate. Then we compare those results with the blood test. Did the person react to the food the test predicted they would react to? Did they tolerate the foods the test said were safe?

If you do that in enough people, you can start calculating things like sensitivity and specificity, how often the test identifies a true reaction, how often does it correctly identify a food that’s tolerated, and how many false positives and false negatives does it produce?

We also want to know whether the test is reproducible. If we test the same person again under similar circumstances, do we get reasonably similar results? Now, conducting rigorous food challenge research isn’t trivial. It is pretty expensive. Blinding foods can be difficult. Symptoms can be subjective. They can be delayed.

And potentially allergic reactions require medical supervision. But none of that changes the fundamental point. If you are selling a test on the claim that identifies foods people react to, we should have evidence showing that the test results actually correspond with reactions to these foods, and that’s the piece that’s missing for these broad food sensitivity panels.

Then there’s that other level, the clinical utility. Does using the test actually help people make better decisions and achieve better outcomes than they would without it? That’s especially important because these tests are not free, right? Depending on the test and the number of foods, commercial food sensitivity panels can cost hundreds of dollars. They’re often paid out of pocket because insurance may not cover testing that is not considered medically indicated or validated for that purpose.

And for somebody living with autoimmune disease, that is not an insignificant amount of money. That’s money that might otherwise go towards nutrient-dense food, prescriptions, supplements that are actually indicated, maybe your medical copay, or working with a qualified provider. And then there are the downstream costs. If your test tells you to eliminate 20 or 30 foods, you might start buying specialty substitutes, spending more time and money sourcing food, or paying for additional testing and appointments based on those results.

So when we’re talking about clinical utility, I don’t think we should just be asking, “Can this test give me information?” We should ask, “Is it going to give me reliable information that changes what I do in a way that actually helps me? And is that information valuable enough to justify the cost and potential consequences?” Because a false positive isn’t harmless when it costs you money and removes nutritious foods from your diet.

[00:16:22] But What If the Test Worked for Me?

Mickey: And this brings us to what I think is one of the most important questions in this whole conversation. What if you took one of these tests, followed the results, and genuinely felt better?

Now, this is where I want to add some nuance because I know there are people listening who are thinking, “But I took one of these tests. I removed the foods it identified, and I felt so much better.” I do believe that that experience is real and important. If your digestion improved, if your headaches went away, or your autoimmune symptoms settled down a little bit after changing your diet, we don’t need to dismiss that just because the test you used isn’t well-validated.

But we do need to separate two different conclusions. “I changed my diet and felt better,” is not the same as, “This test accurately identified my food sensitivities.” Let’s say your test flags gluten, dairy, eggs, almonds, coffee, tomatoes, 15 other foods. You remove all of them at once, and six weeks later, you feel dramatically better.

That is so useful, right? Something about what you changed might be helping you, but which part? Maybe one or two of those foods really do contribute to your symptoms. Maybe you reduced alcohol or sugar or ultra-processed foods in this process. Maybe the overall composition of your diet changed, like you started cooking more meals at home, or maybe your symptoms improved for reasons that had nothing to do with food at all.

This is exactly why an elimination diet by itself doesn’t really give us the whole answer. The really informative part is reintroduction. So if you remove eggs and feel better, and then bring eggs back in while keeping everything else relatively stable, and those same symptoms return, and you can reproduce that pattern, that gives you much more useful information about eggs than simply seeing them in the red column of a lab report.

And here’s where I think we can hold two ideas at the same time. The test might not be validated, and your improvement might still be completely real. Maybe the test happened to identify a food that you genuinely don’t tolerate. That can happen. But an accurate prediction in one person doesn’t establish that the method reliably predicts reactions across more people.

This is important to me because I don’t want anyone who feels better after using one of these tests to hear the episode and think I’m telling them their experience wasn’t real. I’m saying something different. Use your experience to investigate further. Don’t let the lab report become a permanent food list without testing whether those foods actually matter to you. So if you’ve already spent the money, you’ve got one of those reports sitting in front of you, I would not panic about every food marked red, and I wouldn’t assume every food marked green is automatically safe either.

And ultimately, the question we’re trying to answer is, what happens when I actually eat this food? That brings us to the most practical part of this episode. If these panels can’t reliably give us our personalized food list, how the heck do we figure it out, right?

[00:19:28] Why Elimination and Reintroduction Remain the Gold Standard

Mickey: And if you’re familiar with AIP, you already know that elimination is only half of the process. Reintroduction is where we learn which foods actually work for each individual, and this is exactly why AIP has been designed the way that it is.

In allergy and immunology, when researchers want to establish whether a food actually causes a reaction, a food challenge is the reference standard. So for IgE-mediated food allergy, an oral food challenge is considered the gold standard when the diagnosis is uncertain. And for a number of non-IgE-mediated food reactions, improvement with elimination followed by recurrence with reintroduction or challenge is also central to figuring out whether a food is actually responsible.

AIP is based on this allergy and immunology research, and it uses a period of elimination to create a more consistent baseline. Then, instead of assuming that all the eliminated foods are problematic, we systematically bring them back and observe what happens, one food at a time, increasing amounts, careful observation during the challenge and afterward.

And when the answer isn’t clear, we can revisit that food later and look for reproducibility. That’s what turns AIP from this generic elimination diet into a personalization process, and it’s also why we don’t use food sensitivity testing as a shortcut around reintroduction. It would certainly be easier to take a blood test and let a report determine which foods come back, but AIP is designed around the question that ultimately matters: What happens when you actually eat the food?

Those answers can be more nuanced than reactive or non-reactive. Maybe you tolerate a food perfectly. Maybe you tolerate a small amount but not large serving. Maybe frequency matters. Maybe preparation matters. And sometimes you get the best possible result. You reintroduce a food, nothing happens, and it goes right back into your diet. That is a win. Because the purpose of AIP was never to prove that all these foods that we eliminate during elimination are harmful. The purpose is to set up the reintroduction experiment.

If you eliminate 20 foods and feel better, you have not demonstrated that you react to 20 foods. Reintroduction is how you begin sorting out which foods actually matter, and which ones never needed to remain restricted in the first place. Of course, home reintroduction is messier than a controlled research challenge. Autoimmune symptoms fluctuate. You’ve got your sleep, your stress, infections, hormones, medications, exercise routines, your disease activity.

All of this is going to complicate the picture. That’s why we’re looking for meaningful, reproducible patterns, rather than treating every single symptom after a meal as a proof of a reaction. And if you want a deep dive into exactly how to conduct reintroductions, including timing, amounts, what symptoms to track, and what to do when the result isn’t clear, make sure to go back to episode 54, where I take a deep dive through the entire AIP reintroduction process.

For our purposes today, the important point is this: AIP is designed to use elimination to create the conditions for a meaningful experiment, and reintroduction to build the broadest personalized diet that works for you. That’s not a step I want to replace with a shortcut that has not been shown to give us better information.

[00:23:10] Emerging Research Into Food Reactions We Can’t Yet Test For

Mickey: I want to end the science portion of this episode with some research that I find really fascinating because it highlights an important point. Just because our current tests can’t identify a food reaction doesn’t necessarily mean the reaction isn’t biologically real. A research group in Germany has been investigating this in people with irritable bowel syndrome who reported food-related symptoms but didn’t have an explanation from conventional testing, and I’m going to link these studies in the show notes if you want to take a look.

These weren’t simply people who had never been evaluated. The researchers were specifically interested in patients whose food reactions were not being explained by the tools available. They’d undergone evaluation for recognized food allergies and intolerances, including all of the conventional allergy testing, without identifying a clear explanation for their symptoms.

So the researchers asked a really interesting question. What if instead of looking for a marker in the blood, we watched what actually happens in the intestine when it is exposed to food? They used a technology called confocal laser endomicroscopy, or CLE. During an upper endoscopy, they applied tiny amounts of foods, including wheat, milk, soy, and yeast, directly to the lining of the small intestine.

CLE essentially allowed them to use a microscopic camera to watch the intestinal barrier respond in real time. And in a subset of these patients, something happened almost immediately. Within minutes, researchers could actually see the changes in the intestinal lining. They observed widening of the spaces between those intestinal cells, increased movement of fluid through the barrier, and other microscopic changes suggesting a disruption of that normal barrier function.

But here’s the part I find especially interesting. The microscopic reaction happened within minutes, but the symptoms didn’t necessarily happen that quickly. Some people developed recognizable symptoms hours later, and reactions could persist for days after eating the food. That’s particularly relevant when we’re talking about the kinds of delayed reactions that can be so difficult to identify through everyday observation.

And importantly, the researchers didn’t stop with, “We saw something under the microscope, therefore this food is a problem.” They also investigated whether the foods producing these visual responses corresponded with what happened when people actually consumed the foods, using food challenges and dietary intervention to test the clinical relevance of what they were seeing in that microscopic camera.

That’s exactly the kind of validation we’ve been talking about throughout this episode. Does the test line up with the actual response to eating the food? And the results suggested that the microscopic response was identifying something clinically meaningful. When foods associated with these responses were removed, symptoms improved for many people.

And I think that these studies are important because they give us evidence that some people can have measurable biological responses to foods despite negative conventional allergy or food sensitivity testing, and that gives us a much more nuanced way of thinking about this whole subject. I find this really encouraging because instead of assuming that every unexplained food reaction is imaginary or pretending that we already have a test for something that we don’t yet understand, we can keep doing research needed to figure out what is actually happening.

[00:26:52] Reframing Food Sensitivity Testing

Mickey: So where does this leave all of us? I think we need to be able to hold two things at the same time. Food reactions that we don’t yet have a good test for can be real, and that doesn’t mean every test marketed to detect those reactions actually works. And I want to acknowledge why these tests are so appealing in the first place.

AIP is hard. Elimination takes planning, cooking, shopping, reading labels, navigating restaurants and social situations, and changing your routines that may have been part of your life for years. And then you get to reintroduction, which in some ways can be even harder.

During elimination, at least you have this structure, the set of rules. Reintroduction asks you to become an observer and a decision-maker. Was that symptom actually a reaction? Was it the food, or did I sleep terribly last night? Was that serving too big? Should I try it again? How long should I wait? What if I tolerate a food sometimes but not others?

Of course, it would be nice to skip all of that. Wouldn’t it be amazing to send off a blood sample and get back a piece of paper that says, “Here are your foods. These ones are safe, and these ones aren’t”? And I think there’s something else we’re looking for in that piece of paper besides the convenience, validation. When you’ve spent years trying to understand symptoms that can be difficult to explain, having a lab report say, “Yes, something is happening here,” can feel really powerful.

So if you have already spent hundreds or even thousands of dollars on food sensitivity testing, I don’t want you to finish this episode feeling embarrassed or like you should’ve known better. The wellness industry is this behemoth, and the marketing of these tests can be incredibly convincing, and importantly, these tests haven’t just been marketed to us, consumers.

They have also been marketed directly to health and wellness practitioners, often using sophisticated, scientific language about antibodies and immune cells and inflammatory mediators and personalized medicine. A practitioner can look at that and think, “This is probably measuring something that is important and can help my patients.”

But this brings us back to one of the most important lessons from this episode. Just because we can measure something doesn’t mean that measurement is clinically useful. We can measure food-specific IgG. We can measure those changes in blood cells. We can generate incredibly detailed lab reports, but the scientific responsibility doesn’t end with showing that something can be measured, and I think that if a company is going to be marketing that measurement as a way to identify foods that are causing symptoms, the burden is on them to demonstrate that claim with good clinical validation studies.

So if a practitioner recommended one of these tests to you, or you encountered one marketed as scientifically advanced and decided to try it, I don’t think the lesson here is that you should’ve known better. And if you took one of these tests, changed your diet, and genuinely felt better, that experience still matters. Maybe you learned something important about your body. Maybe one or more of the foods on that report really actually doesn’t work well for you.

And if you’re still avoiding a long list of foods because a report told you to many years ago, maybe this is an opportunity to revisit some of them safely, systematically, and with professional support when appropriate. Because reintroduction is hard, but it is also so worthwhile.

Every successful reintroduction can mean another source of nutrients, more dietary variety, more options in the grocery store, and an easier time eating with your family, more flexibility when you go travel or you go to a restaurant, maybe some favorite recipes coming back into your life, less expense, and less mental energy devoted to managing food.

I say this so many times, but it really is this important. The goal is to gather enough good information to build the broadest, most nourishing, most sustainable diet that works for your body and your life. So I would love for science to eventually give us better tools to make that process easier, and the emerging research we talked about today makes me hopeful that we are going to understand much more about these reactions in the future.

But right now, there isn’t a piece of paper that can replace the work of learning your own response to food. That work can be frustrating, it can be slow, and it can be really nuanced, but it’s where the personalization and the magic happens, and that is exactly what AIP was designed to help you do.

[00:31:33] Wrap-Up and Closing

Mickey: So that is it for today’s episode.

If this episode helped you think differently about food sensitivity testing, I hope you will share it with somebody who might find it useful, especially someone staring at a long list of red foods and wondering what to do next. You can find more resources on AIP, elimination, and reintroduction at my website, theautoimmuneprotocol.com. And be sure to check out that episode 52 of the podcast for a deeper dive into the reintroduction process.

Thanks so much for listening, and I will see you next time on the Autoimmune Wellness Podcast.