Food Sensitivity Test Diets vs Low FODMAP Diets

Episode Intro:

Welcome to Dr. Ruscio Radio, providing practical and science-based solutions to feeling your best. To stay up to date on the latest topics, as well as all of our prior episodes, make sure to subscribe in your podcast player. For weekly updates, visit DrRuscio.com. The following discussion is for educational purposes only and is not intended to diagnose or treat any disease. Please do not apply any of this information without first speaking with your doctor. Now, let’s head to the show.

DrMichaelRuscio:

Hey everyone. Today I’m joined by Dr. Robert Abbott, the research director at our clinic, and we review a study that purported to show a diet that was guided by IgG testing was superior to a low FODMAP diet. And at first glance, that could seem like a quite impactful study, so it was worth a full read. Upon further investigation – while being fully open-minded – you see that there are a number of errors in this study. In just one example, they had a placebo group and they didn’t use the placebo group in their analysis. You can compare across groups or within groups, and the whole reason why you have a treatment group versus a control group is so that you factor out the impact of placebo and can be left with what the actual impact of the intervention is.

DrMR:

And in this study, even though they had multiple groups – a placebo group – they didn’t include that group in the analysis. They didn’t do cross group comparison. There are a few other methodological flaws that we go into. We get into what the fundamentals are of good scientific methodology. If you want the take home, the take home is this study was published that purported that an IgG food allergy test was superior to a low FODMAP diet or to a control diet. And unfortunately, this paper is so littered with methodological errors that I don’t think that conclusion should stand, and we should not take that away from this paper. Now, I realize that could throw up a flag that this is a biased statement because of our current position.

DrMR:

And I would argue no because we gave this paper an honest review, but some of the errors in the setup of this study are just so egregious. There’s no way we can inform this or use this to inform clinical practice. Just because an abstract has a conclusion, doesn’t mean it’s always a quality conclusion – hence different levels of evidence… different quality of evidence. So, we go into more detail about that. And we also juxtapose with the studies that we’re doing at the office – how we’ve done things differently and how it can be more difficult to do science well. So, just because something is published, doesn’t mean it’s a good finding and doesn’t mean it may be a valuable or meritful enough finding to change what you’re doing in clinical practice; or if you’re a healthcare consumer, what tests you do or do not ask your healthcare provider to run. We will now go to the unpacking of all this with Dr. Rob from the office.

DrMR:

Hey everyone. Welcome back to another episode of Dr. Ruscio Radio. This is Dr. Ruscio and I am back with Dr. Robert Abbott, who is now the Director of Research at the clinic, which I’m really excited about. He has been doing much more with the FFMR – The Future of Functional Medicine Review. We’ve really been trying to both pioneer clinical research in terms of what we’re studying in an ongoing fashion at the clinic, and we’re doing some right now as we speak — ongoing data collection and analysis. We’re also trying to progressively build the FFMR (or The Future of Functional Medicine Review) into a database – not just a monthly newsletter with case studies and research study reviews, but doing reviews on a topic like Blastocystis hominis and how we should be thinking about that in terms of its pathogenicity and the best method for testing and what have you; to make this a database that is going to provide practice guidelines and algorithms like our thyroid algorithm.

DrMR:

So, as he’s been stepping more into that role, we’ve been working together to try to bring more bright minds together to change healthcare, reform the field and move the field of functional natural medicine in different direction. Between Rob, Gavin and I, if you drop a study into the research sea, we’re like three sharks. What was great is this study was flagged as Gavin was going through research feed, and I saw it and Gavin had some comments, and then I reached out to Rob. And what I was going to say was, “Hey, we should summarize this study in The Future of Functional Medicine Review… make it one of our feature study reviews.” And Rob said, “Oh, I already saw the notice from Gavin and I’ve already done a comprehensive review. Here are my notes.” I was just like – “Great!” It just made me so proud of how we’re all functioning together really effectively. I’m just so excited to jump in on this, but welcome back to the show, Rob. Great to have you here.

DrRobertAbbott:

Thank you for that introduction. It’s always great. I was just reflecting – I think literally today might be five years from when I was out. It was early November 2016 when I met you in Walnut Creek in person. It’s a fun journey.

DrMR:

Time flies. A lot has happened between then and now. We’ll go into the details here in a second, but I’ll just hit a few high point remarks was entitled ‘Igg Food Antibody Guided Elimination-Rotation Diet Was More Effective than FODMAP Diet and Control Diet in the Treatment of Women with Mixed IBS-Results from an Open Label Study.’ That title may sound a little bit confusing and I should maybe flag that, to some extent (this is isn’t an absolute truth), but the more confusing a title and the more confusing the abstract, oftentimes the less clear the finding. Before I bias people’s opinion, this study essentially found (at least when you read the abstract) that IgG food allergy testing, guided diet recommendations was more effective than low FODMAP.

DrMR:

And when I read just the abstract, the first thing that shot through my head was, “Hmm, maybe I should be rethinking how we should start IgG testing.” We’re not using it right now because of our review of the evidence thus far, but I’m always very open-minded. So the first thing that shot through my head is, “Hmm. If this data is valid, then maybe we should rethink how we’re using this testing…” but you read a little further and you see some facets of the argument that seemed to be a bit convoluted and unclear. I’ll let Rob take it away because he went into much more detail. The thing I just want to flag is when I reached out to Gavin and I said, “What do you think?” He said, “I think this study had many methodological flaws.”

DrMR:

My reply to him was — Do you think you’re carrying a bias into this in terms of a pre-existing opinion? I just want to praise both you guys because when I posed the question to you – “Hey, let’s make sure we’re not bringing a preconceived conclusion to this study and unjustifiably critiquing it because of that?” – you both replied, “Yep. That was the first thing I asked myself and I tried to be aware of my own bias and make sure I was giving this study its fair shake” which I think is just fantastic in that nobody on our team is the “No! It’s clear that we’ve already reviewed some of this data and there’s no way this is valid!” Instead it’s “Nope. It’s a great question. The first thing I was doing was trying to catch myself falling into a confirmation bias.”

DrMR:

I just want to praise both you and Gavin for that. Maybe this is a good point for me to turn it over to you because I know there’s a number of things that you went into more detail regarding. Since you spent so much time really going through this in a methodical fashion, why don’t you take us through the salient points and how you looked at them. There are a few that I think just make such important reminders for clinicians in terms of how you can massage numbers to lead to a conclusion that’s kind of falsified. Upon further investigation, that pops out at you, but it may not be so obvious to everyone. Rob, I’d love for you to walk us through this.

The Importance of Conceptual Frameworks

DrRA:

Thank you, Michael. I want to start fairly high level with this. We talked about this before in some of our podcasts – we really want to teach conceptual frameworks and ways to think about, in this case, scientific inquiry or scientific topics. A great example of this came up in my clinic this week. I have nurse practitioner students and we had a patient who had an identified condition on their labs. And we used that as a starting place to say – what should we do next for testing? And I didn’t have those algorithms memorized and yes, we could look them up in an up to date as an augmentation to our clinical observation, but I could apply aspects of an understanding of physiology to then determine – what should I be seeing? So, I wasn’t memorizing the downstream clinical or lab findings. I understood the physiology such that I could work through, and the engineers are like, duh – that’s engineering.

DrRA:

You have a framework and you work from there. The importance of conceptual frameworks and having a core set of knowledge with which to operate for me, just time and again, shows up in this medical space as the way we should be teaching and the way we should be thinking about things. Oftentimes, we get way too reductionistic. We were laughing before this on studies that just came out from the major annual College of Gastroenterology meeting and it just pointed out how reductionistic and in the weeds we are getting. I really want to highlight here in the beginning that big picture conceptual framework. What we’re talking about here is yes, one individual scientific study, but I want to give you guys the framework of how I, and the clinic, looks at studies or papers.

DrRA:

When I’m looking at a published paper – a peer reviewed paper – I really see that that paper should be trying to achieve one of three primary goals. One is that it should be trying to encourage us to think about a specific clinical matter or perhaps a clinical approach in a different way. So, this is something that’s already established – perhaps a clinical therapeutic or just a clinical condition. And it’s encouraging thinking about it in a different way. And oftentimes this can be done with a broader thesis review-based paper or even a case report or case series, but trying to encourage different different thoughts. The second is a study or paper that’s trying to increase our certainty regarding the efficacy or lack thereof of a specific clinical approach or therapy, as well as its potential harms.

DrRA:

And we’re most familiar with the randomized control trial as the intervention to try to help with answering that question or achieving that goal. So that’s a second big category. And then third I see, which is similar to the first, but slightly different, is actually introducing a novel treatment or clinical paradigm for further exploration. And so this very commonly can be done in a case report that presented some novel therapy or a larger thesis paper. And again, it’s similar to that first one, but actually introduces greater novelty. And so when I look at a paper, it should be trying to achieve one of those three goals or fall into one of those three categories, whether it was an experimental design or simply compounded case series, it should be trying to answer or achieve one of those three goals. And the way that we as a community have sought to do that is we’ve established a methodology – or what I call the scientific language – for sharing those ideas.

DrRA:

In some cases it’s very regimented, structured experiments and others, it’s simply a form of documentation of the scientific method or of the scientific process. And so when we’re reading papers, we’re very familiar with sections – an abstract, an introduction, a method section. So there’s this core methodology, this scientific language, that the field has adopted to help us better understand each other’s ideas and how clinical experiments were conducted. The same as we have different English and other spoken languages. It’s kind of like an agreed upon language. And so when I first look at a paper and I’m looking at those goals, I want to ask – does this study/this paper have one of these previously identified goals? And if so, is it clear to me in the abstract or introduction how this paper is actually adding value to the scientific community – or medical scientific community, in this case. When I was in high school, English professors called it the ‘so what?’ question… why should I care? Why is this paper worth anything? Most of the literature sadly fails this question. The paper from the beginning is not clear on its goals. And the question or the questions they’re asking are not clinically relevant, such that no matter what the outcome (say, of an experimental study) was going to be, it was never going to provide any real value, at least in my opinion.

DrMR:

Just to echo that really quick, that clarity piece is important because one of the things I found (and it’s not an absolute trend), is there’s a fairly strong signal here that the more convoluted the abstract is, usually either the researchers don’t understand what they’re trying to ask (and they’re just looking at numbers and coming to conclusions), or they’re trying to hide their findings and massage the data to make it look like they’re finding something novel when they actually haven’t. If you found something of value, you usually are pretty excited and want to say – “Hey, look at what we found. This means we should do something different because this is better.” And when researchers don’t have that, oftentimes you get this convoluted, very long winded or confusing narrative. I just want to flag and echo that because I think that’s a really key insight that practitioners should be on the lookout for when analyzing an abstract or analyzing a paper.

The Divide Between Researchers & Clinicians

DrRA:

Great, great point. And I think this highlights the divide between researchers and clinicians. We have such a pressure in academia to simply publish almost for the sake of publishing that what actually gets published is of little value to the community as a whole and certainly to the clinician. And you see this in academia – those are requirements, depending on which program someone might be in. They just do useless studies and some of that may be driven by a lack of clinical awareness or connection to the clinical world to guide the study design, but these things (a lot of times) are set up for failure. Like this study was never going to add value. And I think there’s a lot of pressure to simply publish. Then, clinicians are very much focused in the clinical world.

DrRA:

Certainly in the integrative community, people are removed from academia for chosen good reasons perhaps, but they’re not publishing, and that’s frustrating because this is a way for us to share ideas, right? There are conferences… there’s papers. And even though we have a very hyper connected world, we need to talk to each other in some language to be able to share like, “Hey, what are you doing?” “Is this something I should be doing?” Not like – I did this thing and that’s what everyone should be doing now – we definitively answered it. That’s obviously extreme and not true, but we have to have some way of sharing ideas of… Hey, think about this thing a little differently… Hey, this is something we observed and maybe you should start trying it and let’s see what happens. But, the clinicians aren’t really the ones doing that, so a lot of the papers that come out aren’t clinically relevant. They’re so reductionistic – the details swamp you and you’re like – I don’t even know where to start with this.

DrMR:

Yeah. Just as one more quick example supporting this point – with the DAYTWO lab – which was the lab that promised to allow you to do a stool test and then through microbiome analysis be able to predict your glycemic response to foods and therefore personalize your diet – this is where this concept really became crystallized for me. I had been feeling this and sometimes you need an experience to take a number of observations that you’ve been holding in your head and then it all kind of clicks. That paper – the paper that was used as a basis for the DAYTWO lab – was very convoluted.

DrRA:

It was very complex.

DrMR:

Yeah. It was a struggle to get through. For providers – if you’re reading a paper and you’re saying, “I don’t get this…” Oftentimes, I used to think I was too dumb to get it. And I’ve learned – no, this is not the case. Do you need to have a requisite level of intelligence? Yes. Assuming that most clinicians have that requisite level of intelligence, if you are struggling to follow the thesis, then I think there’s some shenanigans going on. Now, we used DAYTWO in the clinic for a few months and I was actually quite excited about the prospect of this. We had one of the researchers on the podcast to discuss their findings, but what we ended up finding in the real world was this sadly delivered zero utility to our patients. It gave them zero benefit and I was quite disappointed. And then that’s when everything clicked for me, which is if you can’t state your findings clearly, if you can’t state the utility of your lab clearly, then what’s most likely going to happen (as Rob is alluding to) is when you go to put this in clinical practice, you are taking just this research fluff and it’s not going to have any meaning or any significance in a clinical setting.

Recognizing & Addressing Bias

DrRA:

I think that’s a great study to point out. When it first came out – I remember Robb Wolf talking about it – I’m like – this is fantastic. Of course, gut is connected to these systems. The idea sounded fantastic. Just like we’ll talk about with the IgG testing – the idea sounds fantastic. It just doesn’t actually pan out in the real world and how you can treat patients. And I wanted to make another point on this bias piece. So as you said, we just finished doing (a couple months ago), a pretty comprehensive review on this topic of IgG food sensitivity testing. And so clearly, if something new comes in suggesting maybe you should have a different conclusion, it’s easy to say no. We came to our conclusion and this doesn’t work. And so that biases – you don’t want to be presented with evidence against what your conclusion was.

DrRA:

And so the first step of our process is recognizing that bias. What we’re actually doing in the clinic to try to help this is with the PubMed safe search feature that Gavin is going through, we have safe searches for topics, but we’re not keeping some things out that we don’t like and letting other things in. We actually have a pretty wide funnel of studies, such that studies like this will get into that funnel and we will look at them. People can probably make the comparison to social media feeds where that is a very different funnel where you don’t even get the wide spectrum. Eventually, your funnel is just like liberal media or conservative media, and you’re not even allowed to expand your view or even address your biases because it’s already being done for you by an algorithm. So, we have actually in trying to address our biases… yes, we have topics of clinical interest in the PubMed safe search, but we’re not cherry picking things from that. We’re letting that funnel down and then we have a process of evaluation and I feel very good about that.

DrMR:

Great analogy. Yes. And thankfully, the Google algorithm hasn’t made its way over to PubMed yet. That would be a terrible disservice to clinicians if it did, but you’re right. Yeah. It’s not giving you more of whatever you click on… more of. Yeah. That’s a great analogy

DrRA:

Because the normal unfortunate way by which most people interact with the scientific and medical literature is basically going in with a conclusion – a preconceived notion – about a hypothesis. And they go to find evidence to support that. And guess what, with PubMed, you can find evidence to support whatever conclusion that you want. Just because a study is on PubMed does not mean it was done well or it adds any value. And so unfortunately, most people are doing the process in reverse. It’s not in a malicious way – that’s just the way the information gets to us – say through a podcast or a blog or whatever. So, we go looking for evidence to support our inquiry. And if we find something that doesn’t, then we just move on until we find what does. We’re not actually doing an evaluation.

DrRA:

So, with the PubMed safe search, we’re trying to let (again, within those clinical topics) all of that in. So, then we can parse out what is of importance. And while that may not be something that you, as a clinician, can do to try to change what information you are interacting with, we’re trying to do that for you with the FFMR; to perhaps just encourage you to see how you can expand and recognize what is funneling to you through an already decided algorithm and is becoming your bias and you can’t (even if you want to) expand out of that.

DrMR:

These are all fantastic points. And I should also – on the back of that – remind people that it is way harder to do it this way. It is much easier to say, “Hey, I’m guru Dr. So-and-So… let me grab a research assistant… Hey, research assistant… go find me a bunch of references that enforce these three points.” And you see a lot of that. And I think for those of us who do it this way, it’s frustrating and it should be called out because the only person who loses there is the healthcare consumer. When we do it our way, it’s much more laborious because you’re looking at the data that is pro and the data that is con and then you have to collate that data from highest level to lowest level. And then you have to look at it in juxtaposition and weigh it and evaluate it.

DrMR:

That is way harder than saying “…gotta be low carb… find the references…” Like you’re saying, Rob – you could find a bunch of references to find merit to a low carb diet and those are all true, but the context is inappropriate if you’re trying to represent it as the only solution or not be able to speak to how it should be applied or how it pertains to other diets. And this came up with Susan Blum on the podcast and I appreciate her work, but she made the comment that the evidence is clear that a vegetarian diet is the best for arthritis. I’m paraphrasing her, but it was essentially that. I pushed back and I said, “Well, Susan, that’s also not fair because a number of other diets – like elimination diets, the Mediterranean diet, the autoimmune paleo diet – haven’t really been studied. So, you’re just saying that the vegetarian diet has been studied and shown helpful for arthritis, which is true. It’s not true to say it is the best diet for arthritis because then you would have to have evidence of disproof or disfavor and that doesn’t exist.”

DrRA:

So well said. Yeah.

Sponsored Resources:

Hey everyone. I’d like to thank Athletic Greens for making this podcast possible. If you’re like me, you may sometimes struggle with getting enough nutrient dense foods. One thing I’ve really appreciated about Athletic Greens – or AG1 – is not only are they the best tasting one I’ve had to date, they also help ensure I’m getting adequate nutrition and really hitting the mark regarding nutrient density. I’ve been using them pretty much every morning as part of my morning shake. Now, one scoop of AG1 contains seventy-five vitamins, minerals and whole food sourced ingredients, including a multi-vitamin, multi-mineral, a probiotic and a green superfood blend – all in one serving. The other thing I really like about the company is Athletic Greens continues to improve this one holistic formula based upon the latest research, and they’ve produced 53 improvements in their formula over the past decade. This is pretty remarkable – a hat tip to them for that commitment. I highly recommend Athletic Greens as part of your daily routine. And right now, if you visit athleticgreens.com/ruscio, you can get a free one year supply of vitamin D and five free athletic greens (or AG1) travel packs. Again, visit athleticgreens.com/ruscio to really cover your bases for well-rounded immune support vis-à-vis AG1, and that one year supply of vitamin D.

Study Structure & Design Flaws

DrMR:

So, without getting too into the philosophical underpinnings of how we evaluate science (unless there’s any other key points you want to make), how about we give people the real high level ‘what was the finding?’ and ‘what was fallacious about the finding?’ And then we can maybe unpack some of the details.

DrRA:

Michael gave the title and I’ll read it again to sort of emphasize… reiterate… the confusion that could already begin with the title: ‘Igg Food Antibody Guided Elimination-Rotation Diet Was More Effective than FODMAP Diet and Control Diet in the Treatment of Women with Mixed IBS-Results from an Open Label Study.’ What I extracted from the paper was – what is the study’s purpose? So what about it? Why should I care about this? And what I identified as the study’s purpose was seemingly to investigate the effect of a personalized IgG antibody-based dietary intervention to a standardized low FODMAP dietary intervention, or even a control diet (we’ll have to explain what they use as a control diet) for women with mixed IBS. And so there’s our purpose. And I would say this is a reasonable clinical question.

DrRA:

They’re at least examining a topic that has some merit. I would like to know if there is a dietary intervention that I should be applying in this patient population that’ll have an increased likelihood of benefits. So they’re trying to ask a reasonable clinical question. What we’ll see is they were doomed to failure in the way their inquiry is designed; they were never going to be able to provide any real validity or meaning to their conclusion. There were great clinical questions to be answering, but the study was just designed in a very poor way to actually answer that and provide meaning to the field. Again, it was open label – this is not fully uncommon in dietary interventions and I don’t fault them for this – but it’s important to note when you’re evaluating a study, how was it organizing? It was open label. So what this essentially means is that everyone involved knew what they were doing. Now, if we’re actually going to effectively study how well an IgG-based antibody elimination diet works, you have to design in my view, a very specific study. And the reason for this is anytime, something is personalized, there is a massive therapeutic benefit, which is good. I want people to get better!

DrMR:

That could have been done. That wouldn’t have been too hard. Now you have to be careful with ethics, but I’m sure with some back and forth, they could have said, “Okay, how can we use the personalized lab results, but label them in such a way where people don’t think they’re personalized and how can we also change the label associated with the low FODMAP diet to make them think it’s personalized so as to reduce the placebo effect?’ Just to loop in something that happened with the trio-smart… SIBO being treated with probiotic triple therapy data that we’re currently collecting. And I want to thank David Ring, who is the Associate Dean of Medicine over at Dell Medical. We went back and forth on an email chain – the three of us – and we were hot to trot with our study design.

DrMR:

And he said, “Well, if you have no placebo group, the validity of your results is going to be so far less powerful.” And I was like – he’s got a good point. So what we didn’t do was say – screw it. I can’t say too much about that because we’re collecting data right now, but we had to go through three (at least) hour long meetings with our research team and our clinical team to figure out how can we guard against placebo in this interventional group. I just want to make a point that good science doesn’t happen easily. It’s not necessarily incredibly challenging, but we were willing to get perspective from different people, look for holes in our methodology and then go through the laborious process. It took us probably six weeks longer to get the study off the ground… probably more like two to three months actually if we’re talking about three meetings and probably a few weeks in between meetings and follow-up research in between in terms of how we can make this happen and then how we can make this into policy changes and systems changes at the office. I don’t mean to be overly long winded and/or self-congratulatory, but what I’m trying to point out is that could have been done in this study and the fact that it wasn’t makes me feel like the study is a little bit sloppy because you know one group is going to be more prone to placebo and you did nothing to counteract it. Also credit to you – you guys published a study, but as a group who is going through the same thing, a little bit of extra effort is going to make the validity of whatever you find carry much more weight. So, I just wanted to pepper that in really quick to give people some context in terms of how that juxtaposes to what we’re doing.

Sloppy Organization & Methodology

DrRA:

Really wonderful. And what I’ll add to that, and this may come across a little bit cynical, but if you’re going to publish, you have a duty to organize the study in a way so that it does offer meaning. And so I give these people credit for organizing, for publishing what they found, for trying to support the field, but it ends up being so sloppy that it could potentially do more harm than good because you end up stating things that are not really true. And so maybe they didn’t have the tools to do the study that needs to be done. And we’ll actually talk about one that we reviewed that’s similar to the one that needs to be done to answer this question.

DrRA:

So they maybe didn’t have the tools to do that. I don’t fault them for that, but actually doing the study in the way that they did and publishing on it, could actually hinder the field or be harmful. And so we have a duty to spend more leg work in the beginning to try to have an appropriate clinical question and an appropriate methodology so no matter what we find, we’re having something that we can stand behind and add meaning rather than something that’s sort of sloppy gets to conclusions that are not great. And it could actually be harmful to the net field. Now this isn’t a huge study, so it’s not really going to be harmful, but this study could certainly be taken by companies that offer this test and then they’re going to make a little blog about it, and nobody is going to read this study.

DrMR:

Absolutely. This could be powerful marketing fodder for a food allergy testing lab. Absolutely.

DrRA:

And this is essentially the one that I reviewed – the one primary one that I reviewed from Atkinson, et al. They had the best study design that I’ve seen to date. They did exactly what you were describing. In order to really know the effect, anyone who’s getting a dietary intervention (whether it’s based on anything or complete nonsense) has to think that it’s personalized for them. Other ways/metaphors of thinking about this (and maybe it’s hard to appreciate if you’re not a clinician), but when people feel that they’re being cared for and they’re getting something that was just curated for them, there is such an enhanced therapeutic effect.

The Issues with Novelty Bias

DrRA:

And again, it can be used for good. It could also happen in studies and disguise what was actually being done. And this, I think, is one of the huge effects that’s happening with these IgG tests that are out there. People are getting like… this is MY diet… this is me… this isn’t based off of somebody else on the internet… this is me. And it works because they’re so bought in to it being them. And so if you’re going to truly compare some actually objectively different diets with that in there, you have to make it so everyone believes they actually are getting ‘their’ diet. And again, there’s some ethical potential dilemmas there, but the Atkinson study did the best job I’ve seen at giving people a diet where they thought what they were getting was this like ‘my diet’ and they didn’t find a huge difference.

DrRA:

And so that study was done methodologically very, very well, I would say. And set up from the beginning to be very helpful. This study was not. So, when I first saw this – having already reviewed Atkinson – I thought no matter what this study finds, it is unlikely to change my position or certainly move itself in front of the line of this other larger study that was much more well done. Just because this is newer and has that novelty bias doesn’t mean this is going to no matter what it finds change my opinion and leapfrog a much well better done study that has very few methodologic issues. And so that’s another thing that we sort of see – just because it’s new doesn’t mean it should change your practice. So I read that structure and see how it’s organized, and I think no matter what they find, it’s really probably not going to change my practice. I’m going to hold it in space, but there was already a better study done before this. I’m going to use this better evidence to ultimately choose how I’m going to practice.

DrMR:

Such a great point. Novelty bias, I think, is endemic in our field. And hopefully our podcast and what we’re doing is helping people to bridle that, but new doesn’t mean better. And I think of a few clinicians who just seem to be going a thousand miles per hour. And I think it’s because every new thing that comes up, they start doing it and I can’t help, but just feel bad for them because they’re going so fast that there’s no way they’re ever going to be able to evaluate – does this thing even help? Of course, I think they’re trying to help people, but that novelty bias is definitely something to be cautious about because new doesn’t mean better. New means something could be here, but we’re pretty scrupulous and cautious. We really fact-check and analyze and then we’ll bring certain things into the clinic and do some experiments. How this is playing out now is some person – you, myself, Joe – may have something that they think they’re getting a signal from and they’ll start bringing it in the clinic and isolating it and trying to be in a minimally confounded situation where they can get a better read on it. And then once they do, we’ll all discuss it. And then others of us will start experimenting with it. And then if we get good reproducible signal, then it gets brought into our clinical model and it’s not this haphazard “here’s a new study”… “let’s all start doing this test” model that’s just changing so fast that we can never get a read on what’s helpful or not. That’s the real value for patients — doing all that analysis. The value for patients isn’t “I’m Dr. So-and-So – super high tech, cutting edge.” If the edge is cutting the wrong way, it’s not good. I hope we’re not getting too philosophical for the audience, but these things really matter a whole heck of a lot. I just try to really speak to the thought process (like you are also, Rob) behind this so both patients and providers alike have some ideas of what they should be looking for in terms of what the optimum thought process is and framework and way these things are being integrated into a clinical model.

The Three Groups & Dichotomous Variable Constructs

DrRA:

I’m with that. I can hang out in that philosophical land all day as people can probably guess by now. So with the intervention – I talked about it’s open label. Everyone knows what they’re getting. And so the three dietary interventions they studied were a standardized low FODMAP diet, this IgG personalized elimination diet, and then a control diet, which was honestly in their paper an another red flag. It was confusing for me to actually understand what the control diet was doing. It was like — if you’re having constipation, do this… if you’re having diarrhea, do this. It was advice being given by their (what they described as) attending GI physician.

DrRA:

And so what came clear to me was you have one group who is getting a reasonably formulated diet – low FODMAP. They probably think – yeah, it’s going to be helpful. One group getting a convoluted, weird diet and not really personalized care. And they’re probably thinking – there’s no way this is going to work and it’s even problem focused so I’m already expecting to have these issues and then I’m going to have to shift my diet. It almost can lean into like a nocebo effect. So, the control, if anything is getting nocebo’d on the type of advice that they’re getting. And then you have that IgG antibody group, who is feeling awesome – I’m getting MY diet and this is going to fix everything.

DrRA:

What they did was they allocated 30 individuals at 90 total people; 30 into each group. They didn’t technically randomize them. And this may seem like a minute point, but this starts to introduce issues from the beginning and what each group looks like. Rather than doing a randomization generated, they basically said 1, 2, 3… 1, 2, 3… 1, 2, 3. That could be technically considered randomization, but it’s more allocation. And so from the beginning, they already have an open label study that’s not well randomized. And then – this is just what really killed me – they’re trying to assess gastrointestinal symptoms in patients with mixed IBS and what they ended up doing was essentially using a yes/no answer scale. And so this is what we call a dichotomous variable. But what probably seems obvious is that things like bloating, diarrhea, constipation (say over a long period of time) that’s very continuous. It’s not all or nothing.

DrRA:

What was challenging in reading this is it also wasn’t clear depending on how they asked the question to the individual who would say yes. Was there a certain threshold? Like if I didn’t have it yesterday, I said, no… if I had it two days ago, I said yes. They applied a dichotomous variable construct, which will simplify your data analysis as compared to continuous, but that’s not actually congruent with what people experience and these people had IBS for multiple years. Not to be cynical, but there’s nothing that you were going to do for these people that was going to take them to a true no. So, anyone answering no is kind of a lie. Nobody was going to disappear and have zero/all symptoms gone within eight weeks. I mean, yes, that’s possible, but answering yes/no… and what’s the threshold? So, they’re getting dichotomous bad data from the beginning that is not standardized per patient. And that was a really big red flag as I’m looking at what their data was even going to be.

DrMR:

Just one more point there – we really labored over what is the best way to assess change in patient symptoms with this ongoing data collection that we’re currently undertaking. And we had some ideas about it in IBS symptoms severity scale, and a few other ideas, and Mark Pimentel was nice enough to hop on a call with us and go back and forth on rationales. I really want to credit Mark, and this is where you really see his tenure and experience in research. He said if your sample size is going to be roughly under a hundred, it’s probably not going to be powered enough to capture a signal from an IBS symptom severity questionnaire. And you should really do the sliding scale of selective symptoms using a visual analog scale.

DrMR:

Sorry audience if that’s a little bit technical, but it’s making Rob’s point. And I think it’s also making the point of methodology does matter. I want to be careful – sometimes people just pick apart methodology in this super analytical way and no one can ever tell if it’s true or not true. And they just go super down the rabbit hole to this point of irrelevance. This is relevant. We’re trying to track the thing that matters here – their symptoms – and be able to quantify: Was this a clinically impactful change? Or was it just this statistical noise? And we’re going to just do what’s easy, but it may not actually correspond to that group feels better… this group feels the same. Just one more point in terms of why it’s important to get feedback and really make sure that these different pieces of the study, these different aspects of methodology, are where they should be.

The Importance of Methodology

DrRA:

Hopefully we’ve done a good job of sharing the importance of methodology. Not getting too detailed, but some of the points I’m highlighting are – if you’re going to analyze this study, these are the big ticket things. Was it randomized? Did the people know what they were getting? What variables/measures were they using? Was it yes/ no. Or was it on a scale? On what timeline? These are some basic things that, again, seem basic, but actually can really change (as we see with this study) the relevance of the accuracy of the outcomes. And so they followed all these folks for eight weeks, which I’ll say is a reasonable timeline to see meaningful change. So, I would say that’s set up in a decent way. Some of these things that do things for two weeks, that’s not even long enough to really see any meaningful change.

Per-Protocol Analysis

DrRA:

Eight weeks for a dietary intervention – I’ll credit them for that being a reasonable timeline. And what they saw at the end was that 26 of the 30 patients in the control and 26 out of 30 in the low FODMAP group finished, so they completed the questionnaire at the end, which is actually a pretty high rate for a dietary intervention. Interestingly, only 21 of 30 in the IgG group finished, which might seem like that’s not that much, but that’s actually significantly more than the other two groups. That’s nine out of 30 instead of four out of 30. And why that’s relevant is they did what’s called a per-protocol analysis. So, they basically just looked at the people who finished the study, which again, I think that’s completely reasonable.

DrRA:

The studies that I’ve done, we did per-protocol analysis. We tried to highlight that you’re going to see different things if you actually could include everybody that didn’t finish, so I don’t fault them for that, but it’s important to note that they just looked at the people that finished and there was a substantially larger number of people in the IgG group that didn’t finish the study. That should get you to say – why didn’t they finish? And the most common reason why individuals don’t finish something is they’re usually feeling worse or no better, or it was too hard to do the thing involved so they don’t continue forward and come back for the second (or whatever) study time point to do the questionnaire. You can’t just fully generalize that, but in general, it’s not the other way around where they were feeling awesome. And then didn’t come back because they’re feeling awesome.

DrRA:

That’s much less likely. People, generally, if they’re feeling awesome and they’re in a medical setting, they want to come tell the person that helped them and say, “Hey, I’m feeling good.” Like it’s usually not in that direction. When you look at that, you can already say – there’s a bigger portion of people who didn’t finish and what if they were included? What if we had five more people so that there was 26 in each group? And what if all of them actually were doing worse? With the study, with so few people, if those people had actually been included, it’s going to significantly change how good or how bad a therapeutic may look. And again, that can be overlooked, and it can maybe get a little bit in the weeds with the numbers, but again, I’m looking at scale and saying, “Hmm, that’s interesting. Why did less people finish that?”

DrRA:

And in general, it’s because they’re not doing as well. What would happen if I assumed reasonably so they weren’t doing as well? How would that change what the results may have been in the conclusions? Those are just, again, independent of the exact numbers, ways to think about when you’re analyzing/looking at a study just trying to expand. Ultimately, we’re trying to funnel down to, again, does this study change my practice? Is it reliable? And so doing this kind of analysis, you want to find that balance between the weeds to how likely is this a valid study that I should take into account? And so that’s why it matters. The results were so interesting to me.

DrMR:

This is where it gets really good. We may have buried the lead and maybe this is just my bias with how I’m looking at this, but I think this was the most telling point in terms of how they cooked the numbers a little bit.

Testing Methodologies & Inappropriate Causative Connections

DrRA:

Yeah. So, what we saw at the end is the control group saw zero change in any marker. So not even a whiff of a placebo effect, which in dietary interventions and IBS interventions, there is always a placebo effect. If there’s some degree of a personalized intervention, there just is. It sounds very demonstrative, but there is very routinely at least some aspect of a placebo effect, and there was nothing in the control group – no therapeutic anything. And this is what I was expecting – they could potentially be this nocebo’d here. So, they saw zero change in the control group. The low FODMAP group, as we would expect based off the larger body of literature, saw some significant changes in bloating, gurgling sensations, gastric fullness, nausea, changes in stool.

DrRA:

And they also saw some trends for improving abdominal pain. And so, it kind of did what I was expecting, like what the larger evidence of data that we have for FODMAP diet would suggest. And then you look at the IgG antibody group and I called it the elixir of life (being kind of sarcastic), but everyone went to zero. It was like – everything’s gone… no one had symptoms anymore. And I’m like – wow… we need to be doing this in practice. This thing is magic. And I was like – this is fascinating. This dietary intervention seemingly just evaporated these symptoms for folks. And so then, like I did with that Atkinson study, I said, “Well, what did they actually eliminate?” And I will give the researchers credit and our research team for finding the supplementary materials, but you can see the relative percentages of what was eliminated and what you find, which wasn’t a surprise to me, is what the IgG group actually eliminated were… Guess what?… common immunogenic foods – gluten, grains, dairy, soy, nuts. And so they’re basically doing a well-formulated elimination diet. A point I was actually thinking before (that I hadn’t thought of for whatever reason when doing this and before the podcast) was a lot of things that are removed in these diets, there’s a difference between a food and like a food group and a food constellation. So, we all can agree gluten is not a food. Gluten is a protein. What is it found in? It’s found in processed garbage for the most part. Yes, you can make higher quality foods, but what that’s really representing is a whole constellation of things. Compare that to say an apple. So like IgG for apple – eliminating an apple. An apple is an apple. Gluten represents a whole host of things that can be problematic for different reasons.

DrRA:

Same thing with dairy. Dairy isn’t just one thing. So, removing dairy is way different than removing… pick some singular thing like fava beans. And soy as well. Even nuts are oftentimes eliminated on a larger scale. Grains, as well. So, what’s interesting is some of the things that get eliminated, they cover wide swaths of potentially problematic food as compared to what often happens in these (and what happened in the Atkinson study) is people get a sham diet. They do these tests and it’s like grapefruit… blueberries… stupid things that don’t make any sense, but they’re singular foods as compared to actually eliminating wide groups of foods that are likely to be an issue.

DrRA:

And so it does not surprise me that the IgG group got better because of three main things. They implemented a well-formulated elimination diet that removed those commonly problematic foods. They had this personalized effect I’ve been talking about of ‘I got the diet that was just for me.’ And then the effect of being cared for. They were getting nutritional advice. It wasn’t even just like they got a study – like an EverlyWell test – they got care from another human. You put all that together and that’s actually a pretty reasonable intervention, but it just boggles my mind that they can look at this and then get to the conclusion of… the IgG is what got us there. It’s like — no, no, no.

DrRA:

Don’t you see what got you there was you gave them a reasonable diet that eliminated problematic things, you gave them care, and then they thought this was all personalized. That’s what got you there. Independent of actually your silly yes/no analysis – that probably made the data so warped on who was actually better and who wasn’t was probably hard to really actually say – but what got you there (if we take that at face value) was not the testing. And this is the fault we’ve talked about before in testing heavy methodologies, you’re going to make inappropriate causatory connections – like the test is what then gave me the treatment and that’s what got the person better. Well, actually you could have got to the treatment in a different way, or you’re not looking at what actually was being implemented. And that’s the fault that this study makes, and a lot of other studies not just using IgG testing make, and that we as a field have to refine and really understand,

DrMR:

To juxtapose this with how the wrong interpretation of this study could play out. People could spend anywhere from $400 – $1,000 for a test and follow the diet for a few months. And then, of course we have to retest now to see if any of your allergies changed and essentially there’s going to be some change, perhaps due to lack of reliability with these tests. And then probably another three to six months later – another test. Worst case scenario, the delta here could be $3,000. Now, we also have a handout called the paleo low FODMAP diet, which covers all of this. And someone could start that weeks before they even get their test results back… already be on their way to improvement… and then go through a reintroduction and figure out their diet with more accuracy and at a fraction of the cost.

DrMR:

So, this does matter when viewed through the prism of – how could this study result in fluent clinical practice? And if you start going to the – well, we can’t advise any dietary changes unless we have a test – I would argue that’s a huge disservice to someone, because again, you could cost them $3,000 and actually make it take longer to get there and make them have this weird “Well, I’d like to have a grapefruit, but my allergy testing just says I can’t.” I don’t mean to poke fun, but I think any clinician who has done the food allergy test will tell you over half the time people say ‘I eat this food all the time and I’m fine with it.’ Don’t give me the… “Well, that’s only because you haven’t ever eliminated it. And if you did, you’d notice that you felt a lot better when you reintroduced it…” Because I’ve tried that, right?!

DrMR:

I don’t mean to poke too much fun. Sorry if I’m sounding a little condescending, but the condescension comes from the perspective of seeing some patients who are so afraid of food, and we just don’t have the data to justify that. And to your earlier point Rob – which is such an excellent one – if we’re using data irresponsibly and we’re not being circumspect enough in the analysis, we can inadvertently harm people. I just want to echo that the reason why we’re putting this into the microscope is because it leads you down that road that I said, which is serial food allergy testing (anywhere from $400 to $1,000 ) and almost for certain taking longer to get to the outcome that you’re trying to get to (if you’re even able to get there at all) as compared to what could be done with a handout for free.

DrRA:

Speaking to the collective choir of we really want to help people. Why we’re sharing this is because we see people harmed by this. And I will say today, exactly. So, I have a patient coming in next week that has had numerous tests from outside providers. Again, well-intentioned, well-meaning… had a IgG food sensitivity test, and I’m looking at it, and 75% of the things on here are things like grapefruit and green beans and just ridiculous things that are not commonly immunogenic. And it’s like – how do you eat? How do you take that and find a way of eating that has any meaning that doesn’t create fear? And like you said, then they’re going to test again in three months to see what’s changed. And I think we have to really step back. People are suffering, they have symptoms, and they’re trying to understand why. Patients and providers would love for these tests to be able to say – remove this and get better.

DrRA:

Let’s really pause here. If you do one of these tests, then maybe it actually doesn’t show gluten or dairy or eggs or nuts, or some of the commonly problematic foods. Is it really like IgG to apple… removing apple that previously wasn’t removed? Is that really going to be the thing that moves the needle? I have a really hard time believing that, even if that is accurate, and even if that did reflect something immunologically pathological. The concept of some of these tests just doesn’t hold water for me, at least what we observe. There’s just a lot of risk in doing them to create fear. Going back to the bias point – yes, we have conclusions that this testing isn’t viable, but at the same time, I would love to see evidence to say that it works and I should be doing it, and it would change my practice, like anything coming through.

DrRA:

I also have a bias that I want to see it work. I think we can probably appreciate this. My metaphors is in sports. If you’re a fan of a certain team and you go watch a game, (if you’re so fanatical or so into your team) you can’t even appreciate the skill of the other team because you’re so biased. You can’t see this other team and that they’re really good players because you’re just so angry that your team is losing or something. While yes, we have that bias, I also want to see this work. It just the evidence hasn’t been there and I don’t think it will be because of what we’re seeing.

DrRA:

Why we’re highlighting this is because we’re seeing it today. We’re seeing it in practice and seeing people harmed by it. Can you go out there, find someone on the internet who said they did this test, removed this food and they got better? Yes. We’re not denying that person got better. They just probably got better from a different route than they expect. And did someone really just remove apples? And that was the game changer? I’m going to have a hard, hard time believing that, and there is just a really high risk with these things.

DrMR:

Yeah. And again, our bias is based upon using these tests, not finding them to be very effective when used in a critical model (not a confirmation bias model) plus a comprehensive review of literature, plus a paradigm in which we’re trying to intervene as far upstream as possible; meaning healing the gut as best we can so that people can have the broadest diet and not go to this point where we’re trying to contract or constrict the diet more and more. Because of all those things, I think what we’re doing at the clinic is really the far superior method. And if at any point someone produces a piece of evidence that is adequate to cause rethinking that, we will because we’re constantly purviewing the literature and we’re adding things into our clinical model once we have requisite evidence.

DrMR:

Abdominal self-massage is a good example of that, where we’re going to be integrating more recommendations into the model for that for non-responsive constipative cohorts because there’s been pretty good evidence now that that can really improve regularity for some people.

Dr Ruscio Resources:

Hi everyone. Just a quick announcement regarding the clinic. I am happy to say that I, and we, at the clinic are now offering a free monthly support call to all current patients. This applies to any patient at the clinic, even if you’re not working with me directly. This is an opportunity to ask me and our team questions, share feedback, and get support with any challenges you may have. I will be accompanied by Dr. Joe Mather, our medical director and Morgan, our clinical health coach. We have emailed details to all of our patients, so check your inboxes. And here is Erin with the date and time of our next call. Hope to speak with you there. The next call will be Friday, January 14th @ 1:45 PM Central.

DrMR:

Rob – one other thing before it slips my mind – there was this cross group comparison bit and that’s where I thought you were going next. That was the thing that struck me as the most noteworthy. So, let’s make sure to unpack this for people because this is also quite telling.

Paired & Unpaired Statistical Analysis

DrRA:

This study had three groups. You have groups and they collected data at the beginning of their study. So, before the eight weeks of diet and at the end. And so with this, you can do two different types of analysis and physical analysis. One is you can basically have each person be their own control. You can look at them at the beginning and look at them at the end and say – did you get better? And you can do that within each of the individual intervention set – the control diet, the FODMAP diet, the IgG diet. And it’s a completely reasonable analysis. And with a study like this should be done. They did that. What they didn’t do, however, is they didn’t compare people between groups. They just compared people within their own group. And why this is a problem is it can make a specific intervention look much worse or much better than it really is.

DrRA:

And it’s why we have placebo group or control groups – that group isn’t there to just see what they did pre-post. It’s there to compare how does the control group post look to the other group post? You need to compare separate groups, not just people at the beginning and at the end. The technical terms of this can be a paired or unpaired statistical analysis, but terminology is paired being within versus unpaired being between the groups. And it’s really bad if you do a study with multiple groups, a control group or multiple arms, and you don’t do an analysis between groups.

DrMR:

It’s almost like you shouldn’t even have the groups. This to me – and I’m sorry, we should have probably led with this – is the most telling point. We know when people are even in a control group, that they’re going to see improvement in the vast majority of cases. So, that’s why by definition (like you said, Rob), this is why we have placebo controlled trials. So let’s say just some arbitrary numbers to throw out there — The group that got nothing saw a 20% improvement, the group that got treatment saw a 30% improvement. So, we know that the difference between placebo and control is 10%. That gives us the actual value of the intervention when we factor out placebo. The fact that they didn’t do this just boggles my mind. That was the most telling piece of this study. It seemed like you must just really be trying to massage the data to show a favorable outcome, because if you have the information, if you have the numbers, why not do the cross group comparisons? It only bolsters the perceived impact of the food allergy testing, but it’s kind of dishonest. I mean, I don’t want to accuse them of purposely being dishonest. Maybe there’s something that I don’t understand, but I’m at a real loss in terms of why they wouldn’t do that.

DrRA:

I’m right there with you. It could certainly get a little bit lost in the statistical weeds, but I could look at the data (it’s a small enough group) and see what was theoretically significant within one group versus another, mainly the FODMAP group and the IgG group. And if they had actually done an unpaired between group, a lot of the really meaningful findings from the IgG group wouldn’t have looked that interesting. It’s still probably because the data was collected in a weird way and it looked seemingly so fantastic. There would’ve still been some things there, but it wouldn’t have looked as gangbusters as just looking at people within the IgG group. Again, it’s a little disingenuous and if you had three groups, why didn’t you do that?

DrRA:

When I did the AIP Hashimoto’s group, we had one group. So, there is no unpaired analysis – you have one group of people – so that’s one limitation and you state that. But when you have multiple groups, you could do both, but you can’t just do the paired within group. That can make things look much better/much worse than they are. It just boggles my mind. I went over it three different times – Did they really do this? I was like – wow, that’s why these p-values are the way they are because I was trying to understand which p-value was attributed to what. And like they just did this analysis – that’s amazing and that’s how they set it up.

DrRA:

Again, the methodological flaws that we see here, yes, it could seem like we got into weeds, but these are some very key things – how the study is designed, how they’re collecting data, how it was analyzed (and not esoteric like – did you use this statistical test?).

DrMR:

These are core fundamentals of research.

DrRA:

Ultimately, again, no matter what they conclude, even if it was the other way, even if they said that IgG testing was bunk, I wouldn’t even really be able to take that takeaway from this test because it was so methodologically flawed and that’s what’s unfortunate. What’s really unfortunate is that potentially the abstract that could be taken by companies or people and read at face value and provide merit when it’s not deserved for a test. And that’s what we’re trying to prevent. And we want to give you the tools, if you have the capacity, time, and desire – both with the FFMR and this podcast – to help you do at least some aspect of this critical thinking and inquiry so we can avoid some of these pitfalls that our field and the whole field (not just traditional medicine) can fall prey to; that we’re falling prey to too much. And it’s affecting people negatively.

Episode Wrap-Up

DrMR:

Folks – for all these reasons, this is why we’re doing what we’re doing with the research that we’re doing at the clinic and the analysis of studies that we’re doing through the FFMR and the practice guidelines that we’re trying to establish as part of the FFMR and the research reviews that we’re doing there also, just so we can make sure that at the end of the day, people are being given the best clinical care. That’s what this is all after. It’s not about selling a course or a supplement or a test. It’s not the cherry picking of the data. What we really care about the burning core of what we do is being effective clinicians and everything is kind of a slave to that. The reason why we went through this study in a bit of detail was to illustrate the kind of thing that we go through for the ongoing stream of science to make sure that the things that should inform and update our clinical model do, and the noise that should not that could potentially jeopardize it, is filtered out.

DrMR:

This is choice study that obviously could have impactful ramifications (as we talked about a moment ago) which is why we wanted to showcase it. And Rob – I’m glad that you really sunk your teeth in on this. For the clinicians out there – if you want to see the really full write-up, this will be published in the FFMR. By the time this podcast airs, it’ll already be in there so you can see the more fuller analysis. Also, in one of our previous additions of the FFMR, we had the full evidence review on the IgG food allergy testing. I think that pretty much picks this bone clean, Rob. Was there anything else you wanted to leave people with as we move over to a close?

DrRA:

No, I just thank anyone who is still listening. I think we did our best to not get stuck in the weeds. I hope this was of value, and really just another insight into the process in our clinic. We want to be transparent. We recognize we all have biases and we’re trying to put systems in place to help that. And you heard multiple people looking at this study. It wasn’t just one of us looking at it deciding – oh, this is what we’re doing. We have multiple minds. And so it’s just all the things we’re trying to do to really help you and getting a behind the scenes look time and time again of what that process looks like. Hopefully that is interesting or helpful. Appreciate it – you made it this far.

DrMR:

Awesome. Fully agreed. Rob – Thank you again, my friend and I will see you at the clinic.

Outro:

Thank you for listening to Dr. Ruscio Radio today. Check us out on iTunes and leave a review. Visit DrRuscio.com to ask a question for an upcoming podcast, post comments for today’s show, and sign up to receive weekly updates.

Igg Food Antibody Guided Elimination-Rotation Diet Was More Effective than FODMAP Diet and Control Diet in the Treatment of Women with Mixed IBS-Results from an Open Label Study

• 73 female IBS-M patients randomized to:
  • Low FODMAP diet
  • IgG based elimination diet (eliminated foods that had high IgG levels)
  • Control diet recommended by gastroenterologist
• After 8 weeks:
  • IgG elimination diet saw more symptom improvement compared to low FODMAP diet
  • NO improvement in control diet
• Commentary: At first glance this study suggests superiority of IgG based elimination diets. However, after further investigation, we see some methodologic issues with this study:
  • The diets did not
    • control for fiber content: during periods of diarrhea, patients in each treatment group were told to reduce insoluble fiber and fat. Conversely, during periods of constipation, patients were told to increase dietary fiber intake (to 30-50 g/day). So, in practice, these diets had a high fluctuation of fiber intake. 
    • The low FODMAP diet was NOT truly low FODMAP for the course of the 8 weeks
    • And the IgG based diet had elements of the low FODMAP diet (during periods of diarrhea)
  • The IgG test examined over 269 foods: This would add a considerable cost to the patient. Rather, an empiric trial of an ancestral template +/- further reduction of possible food sensitivities (e.g. low histamine, low nightshade) diets could save the patient financial resources and be more practical in a clinical setting. 

 

•         Poor methodology (open labeled – placebo, Y/N survey)
•         More IgG group did not finish study – could mean high dropout/AE rate
•         Open label placebo/nocebo

o Control group had 0 change (indicated nocebo)

o LF saw improvements

o IgG group – all symptoms gone (too extreme of response – V unlikely)

• Gluten, grains, dairy, soy, nuts….

o Could have covered in PLF diet

•         Did not compare groups

o No placebo

• Athletic Greens
• Dr. Ruscio Resources