Multidisciplinary panel finds little evidence to support dietary interventions as a treatment for autism.

Click here to listen to the audio.

By Dennis Miller, BHC Senior Writer

It has long been suspected that autistic children have a higher incidence of gastrointestinal disorders, but until now, there’s no been no uniform set of diagnostic and treatment guidelines to help medical professionals and parents manage the problems. With that in mind, Timothy Buie, M.D., a pediatric gastroenterologist from Harvard University and Mass General Hospital, last year convened and chaired a multidisciplinary panel to examine the evidence and reach a state-of-the-science consensus on best practices. The group has now published their conclusions as the field’s first-ever comprehensive set of guidelines for evaluating and treating gastrointestinal disorders among children with autism.

Dr. Buie says the unique multidisciplinary makeup of Massachusetts General Hospital’s (MGH) clinics helped make such an effort possible. “We at Mass. General have a very nice group of specialty providers who make up the developmental clinic, which is the Lurie Family Center for Autism and LADDERS (Learning and Developmental Disabilities Evaluation & Rehabilitation Services). And at our institution we are very fortunate that a variety of specialists are in the autism clinic, which is very atypical, I think, in the country,” he explains. “There aren’t a lot of places in the country that have that kind of opportunity. And it has made us, as a location, have a different feeling about children with autism. We look at these children as having an underlying medical phenomenon accounting for their autism, but also that they may well have an associated phenomenon going on as a result of that.”

A careful review of the literature

The panel included 28 experts across 12 scientific disciplines, including child psychiatry, pediatric allergy, pediatric gastroenterology and pediatric nutrition. The team did an extensive review of the literature on gastrointestinal disorders among children with autism spectrum disorder (ASD), and developed a list of 23 consensus statements intended to offer some clarity to clinicians and parents. Along with that they developed a consensus set of recommendations for evaluation and treatment. The two works were published in the January 4th edition of Pediatrics, and are available here for download as PDFs:

• Evaluation, Diagnosis and Treatment of Gastrointestinal Disorders in Individuals with ASDs 
• Recommendations for Evaluation and Treatment of Common Gastrointestinal Problems in Children With ASDs

The process of distilling the extensive body of available research down into a clear set of conclusions and treatment guidelines was painstaking and laborious but democratic, with information making it in to the final report only if a significant number of panelists agreed. “We brought sleep specialists and immunologists and gastroenterologists from a variety of locations, including locations around the world, and asked them to review particular documents in the literature to address particular questions,” says Dr. Buie. “And we broke those groups into small groups, including nutritional input, chronic gastrointestinal inflammation, behavioral issues, etc. And small subgroups read the literature and scored it as it is available, to say, ‘Is this a valuable paper? Is there not enough information from that paper?’ etc. And then we brought all of that information into a large core group of people and literally voted on the papers. We decided how well we felt the data was qualified and tried to make our best judgments from that.

“Then we used a voting system where we made statements. We came out with statements or conclusions about the literature and the way that those statements had to be qualified. We used a nominal technique so that all of us participated in the wording of the statements and everyone in the room had to agree with the statement or we wouldn’t post it in the paper. So in the end, what we had were 27 specialists who were all willing to agree with the statements in this consensus paper about particular issues, including diet, behavioral issues, and how gastroenterologists would evaluate these children.”

Media spins the wrong focus

The release of the group’s report a few weeks ago prompted a flurry of media reports that the panel had determined that dietary interventions are ineffective for ASD patients. Such special dietary approaches have in recent years become somewhat an article of faith among parents, advocacy groups and some clinicians, who speculate that there's a digestive problem specific to autism called "leaky gut" or "autistic enterocolitis." The theory has it origins in a now-discredited 1998 study by British physician Dr. Andrew Wakefield.

Dr. Buie says the panel did not, in fact, conclusively rule out the effectiveness of such diets, but found only that there is little evidence for them in the published medical literature. “The intent of this paper was to try to put a context around the question of dietary interventions,” he says. “And when we reviewed the literature of diet intervention studies, we were not able to find enough information based on those intervention studies that supported the idea that we should imply diets into everyone with autism as if we think there’s going to be a general treatment for all of those children.”

He’s careful to point out that the panel did not squarely recommend against such diets, but only that more research is needed to prove their effectiveness, and that until such proof is available, clinicians and parents should base treatment decisions on the best evidence. “I honestly don’t have a lot of difficulty when a mom stands up and says, ‘My child got better with an intervention,’” he explains. “ The bigger problem I have with those statements is that they don’t — or shouldn’t, in my opinion — set the proper agenda for how we go forward with care. If that response happened with that individual child, I’m really pleased. I’m glad that that was a helpful intervention, but it doesn’t help me and my ability to manage this community of children. And I think it’s not always the most responsible pathway to suggest that because it worked for me, everyone should try this diet because it might work for you.”

An emphasis on alertness

Dr. Buie hopes the panel’s work prompts a greater emphasis on alertness for GI problems among autistic children, as the findings suggest many behavioral problems could very well have their roots in undiagnosed gastrointestinal distress. The trouble, he says, is that kids with autism are often unable to communicate their discomfort clearly, and if clinicians aren’t on the alert for such problems, the child may not get the treatment he or she needs. The discomfort itself may then become the root cause of unruly behavior.

“One of the risks in children with autism is that they don’t communicate very well,” he says. “Children, for instance, who might have self-injurious behaviors or aggression who are not obviously vomiting might be having gastrointestinal pain or other medical pain that’s accounting for their behavior. But they’re not going to be identified by any standard gastroenterologist or pediatrician as being recognized as having that underlying medical condition unless you think about it.”

As to the question of whether there is some intrinsic connection between autism and gastrointestinal disorders, the panel was unable to find conclusive evidence of a link. But the research does seem to point to a higher incidence of such problems among autistic children. Exactly why that may be remains somewhat of a medical mystery. “We see a relatively high frequency of GI symptoms in children with autism, and the comparative groups for the most part show a lower frequency,” Dr. Buie says. “So we’re worried that this is a little bit more common in this population and we don’t have a very good understanding of why that is.”

An urgent call for more research

For that reason, one of the panel’s strongest recommendations is that research efforts be strengthened. “I hope that if we don’t make conclusions that families agree with, they understand that most of the things that we couldn’t make conclusions about [were] because there was insufficient information,” he explains. “Most of the data out there is not excluding the possibility of a dietary intervention being helpful, etc. It was because we could not find enough evidence at this point, and all through the paper we say, ‘We need more help. We need more qualified information before we can make some of these conclusions.’”

To listen to our full interview with Dr. Buie, click on the audio icon above. The following is a complete edited transcript.

BHC: Tell us a bit about this consensus report that you and your colleagues developed — why you decided it was necessary; how you went about developing it; and the significance of the guidelines for medical professionals treating children with autism.

TB: This consensus group came together really as a sort of a dream. We at Mass General have a very nice group of specialty providers who make up the developmental clinic, which is the Lurie Family Center for Autism and LADDERS (Learning and Developmental Disabilities Evaluation & Rehabilitation Services). And at our institution we are very fortunate that a variety of specialists are in the autism clinic, which is very atypical, I think, in the country. We have neurologists; we have developmental pediatricians; we have psychopharmacology treaters; we have a gastroenterology group which sees children in the clinic; we have genetics input; and we have sleep specialists who are seeing these children. So we’ve put together a band of doctors who come together and are in the same hallway at the same time seeing children with autism.

And the feeling at our institution is that this is unique. There aren’t a lot of places in the country that have that kind of opportunity. And it has made us, as a location, have a different feeling about children with autism. We look at these children as having an underlying medical phenomenon accounting for their autism, but also that they may well have an associated phenomenon going on as a result of that.

For instance, in gastroenterology, we think that probably about 50 to 75 percent of the children that come to the autism clinic will have gastrointestinal problems that require evaluation and care.

So I tried, then, with the group of doctors that we work with, to bring together a group of experts from other fields and other locations who might be able to help us look at what was known in the literature at the time, and review that information so that we could come out with some guidelines for people who don’t have a clinic setting like us — who don’t have the opportunity to refer to a multidisciplinary type program. So they can go to their pediatricians and say, “Here’s what the state of the art is. Here’s the best understanding of what’s happening in medicine right now, and I’m worried about these certain things — here’s what this panel of people recommend.”

We were able to put together a group of specialists with the help of Easter Seals of Oregon and several other funders, including the Autism Society of America and the Autism Research Institute, among others. They funded the bringing-together of this group and the construction of a paper as an outcome.

And our thought was if we did this we would put together a variety of specialists. Again, we brought sleep specialists and immunologists and gastroenterologists from a variety of locations, including locations around the world, and asked them to review particular documents in the literature to address particular questions.

And so we broke those groups into small groups, including nutritional input, chronic gastrointestinal inflammation, behavioral issues, etc. And small subgroups read the literature, they scored the literature as it is available, to say, “Is this a valuable paper? Is there not enough information from that paper?” etc. And then we brought all of that information into a large core group of people and literally voted on the papers. We decided how well we felt the data was qualified and tried to make our best judgments from that.

Then we used a voting system where we made statements, where we came out with statements or conclusions about the literature and the way that those statements had to be qualified. We used a nominal technique so that all of us participated in the wording of the statements and everyone in the room had to agree with the statement or we wouldn’t post it in the paper.

So in the end, what we had was 27 specialists who were all willing to agree with the statements that came out in this consensus paper about particular issues, including diet, behavioral issues, and how gastroenterologists would evaluate these children.

BHC: Dr. Buie, are children with autism especially susceptible to gastrointestinal disorders, or is the problem more that they’re less able to report their symptoms to parents and clinicians and they just go on worsening?

TB: I’m not so sure about the answer to that. The studies that have tried to look at the frequency of gastrointestinal problems in children with autism suggest that there seems to be a higher frequency of gastrointestinal problems compared to whatever comparison group is listed in the paper. For instance, a paper by Dr. Valicenti reported that in her children with autism, there was a frequency of gastrointestinal problems of about 70 percent. But when she had comparison groups of children with other developmental disabilities, she saw that that number dropped to about 40 percent, and in children who were not affected with autism, there was a frequency of gastrointestinal complaint of somewhere around 28 percent.

Most of the GI papers looking at the prevalence find similar frequencies. We see a relatively high frequency of GI symptoms in children with autism, and the comparative groups for the most part show a lower frequency. So we’re worried that this is a little bit more common in this population and we don’t have a very good understanding of why that is.

I think your second question is really important, because one of the risks in children with autism is that they don’t communicate very well. And so, one possibility is that we may actually over-interpret problems. A child who is crying while on the toilet we may interpret it as having pain while bowel movements are happening, when it may not be that, but we’re tied to that. Or a child who hasn’t toilet-trained because he’s got developmental disabilities might be viewed as having chronic loose stools, when if he were toilet-trained, you wouldn’t know about it and it wouldn’t be a problem.

So I think there’s the risk in over-reporting in that community, but we also know that there’s the equal risk of under-reporting. Children, for instance, who might have self-injurious behaviors or aggression who are not obviously vomiting might be having gastrointestinal pain or other medical pain that’s accounting for their behavior. But they’re not going to be identified by any standard gastroenterologist or pediatrician as being recognized as [having] that underlying medical condition unless you think about it.

BHC: Dr. Buie, the media seems to have zeroed in on the report’s finding that there’s little proof for a link between food sensitivities and autism, nor is there much evidence of the effectiveness of special diets for it. Talk about that aspect a bit and maybe how the media kind of got the wrong focus.

TB: I think many people want to take a position in this regard and say, “Here’s another paper that suggests against what the popular ‘mommy’ theories are,” or people who have media access have often proposed that diets have been helpful, or that biomedical interventions or vitamins or other nutrient supplements are helpful.

I honestly don’t have a lot of difficulty when a mom stands up and says, “My child got better with an intervention.” The bigger problem I have with those statements is that they don’t — or shouldn’t — in my opinion, set the proper agenda for how we go forward with care. If that response happened with that individual child, I’m really pleased. I’m glad that that was a helpful intervention, but it doesn’t help me and my ability to manage this community of children. And I think it’s not always the most responsible pathway to suggest that because it worked for me, everyone should try this diet because it might work for you.

I think what the intent of this paper was, was to try to put a context around the question of dietary interventions. And when we reviewed the literature of diet intervention studies, we were not able to find enough information based on those intervention studies that supported the idea that we should imply diets into everyone with autism as if we think there’s going to be a general treatment for all of those children.

But we spent a lot more time honestly looking at the potential mechanisms by which foods might cause problems. For instance, if a child has lactose intolerance, they may get gas or distention and may have an awful lot of distress related to that lactose intolerance that might cause them to be fidgety, or pacing behaviors might be more prominent. And so we certainly want to suggest that people think about those underlying medical mechanisms and be more focused on how to get to that through an evaluation.

Food allergy we know happens in a patient population. We know that somewhere around 5 percent of children in the general pediatric population have a food allergy. We at least need to expect that if that’s true in the general pediatric population, there will be a subgroup of children with autism who also have evidence of food allergy, and if we recognize it, they too might do better.

We can’t take that next step — I don’t think we should take that next step — to say that because they have that condition, it’s the reason for their autism or that it’s necessarily the reason for their neurological dysfunction. We’ve tended to take the other position and say, “Look, if you’ve got autism and you’ve got sensory processing issues or you’re unable to communicate your distress, you may well have behaviors that result from that food sensitivity. It’s not the cause of your autism but it surely may exacerbate your autism behaviors.”

BHC: How do you and the other members of the panel hope to see the results of your consensus reports used in the field?

TB: I’m very hopeful that parents or pediatricians who read this (and you can get this through open access if you go to the American Academy of Pediatrics website and actually go to their journal section — it’s in the January supplement of Pediatrics— and you can lift it in full text [Ed.Note: Both PDFs are also available above]) say, “Well, there’s a pretty good group of people here who are, I think, working at legitimate academic centers, who are really willing to believe in the possibility that these children have underlying medical issues. And here is a pathway for how I can do some work-up of those children to figure out if my child, the child I’m caring for, has these problems.”

At least we want people to be aware of the risks — that these children have underlying medical issues and we shouldn’t ignore them. And the very first statement that we make in the evaluation paper is to say that children with autism at a minimum merit an equal evaluation as we would children who do not have autism to look for these gastrointestinal problems. We can’t limit access to these children because they’re disruptive in our waiting rooms or they’re unable to be settled in the examination room. We still need to figure out what ways to properly evaluate these children. And if we’re going to do it for children without autism, we need to do it for this group of kids as well.

And so I’m hopeful that for the pediatrician, it gives them some guidelines about how to think of these problems. And we hope we’re giving new suggestions to particular behaviors that people might not have thought about that may be accounting for an underlying gastrointestinal problem, such as tapping on the chest or pressure-seeking behaviors where children are actually looking to press their abdomen against the desk because they’re trying to get relief. Some of those guidelines I hope are going to be useful to people.

We also wrote this paper with the intention that parents can read it, and that they can look through the paper and recognize their own children and see if some of those behaviors resonate with them. And that they can highlight those things in the paper and take it to their doctor and say, “These are the things I am worried about, and this group of doctors has suggested that we consider these underlying medical phenomenon.”

BHC: You just touched briefly on the subject of my next question which is: What are some of the signs of a gastrointestinal disorder that clinicians and parents of children with autism frequently miss?

TB: We’re worried about things that are not obvious. I think sleep disorder is a major problem in children with autism. And we know that certainly these are children that are very stressed and stress may keep them awake. They may not have a great availability to a full night’s sleep, and if you don’t get that, you don’t perform very well the next day.

And so, we’re certainly concerned that we do the steps that we can do to recognize the underlying medical conditions that may affect sleep, and certainly acid reflux is one of those. And we want to consider that child who’s got night awakening or who’s very difficult to put down to sleep, as if lying down flat may be somewhat of a difficulty for some of those children — [that] we at least think about that underlying condition.

I think a lot of the problem behaviors that people talk about and have written off as an autism behavior, like hyper-stimulatory behaviors or pacing, etc., may be exacerbated by distress and discomfort, and it’s not to say that I want to discount this and say that every kid that has those types of behaviors which we know are fairly common in this community. We’re always going to find some underlying medical issues, but we ought to at least think about them in the context of what else is going on for those children.

And I think so often we don’t — we ask parents all the time, “How often does your child have a bowel movement?” Once they’re toilet-trained, a lot of parents don’t know. And a lot of parents stop thinking about those kinds of things. And so, we really ask parents to go back and really try to catalogue what’s happening for those kids. How long are they spending in the bathroom? Do they seem to be having a difficult time? Maybe kids go in and make the effort to have a bowel movement and they don’t succeed, and that’s a very classic irritable bowel type symptom, but a child might not be able to report that they’re incompletely evacuating or having difficulty evacuating, so they’re never going to report to us the symptoms that would characterize them as irritable bowel. We’ve got to be the observer to identify those issues.

BHC: Which neatly enough brings me to the subject of my next question. The report seems to call for a higher degree of communication and care coordination between behavioral health clinicians and physicians. Do you think this has been somewhat of a problem to date?

TB: I don’t think it’s a problem. I think so many of the children that I see are sent to us by very good observers. And so many of these children are spending an hour or two hours or more with their occupational therapist or their behavioral therapist that I think very often they’re the best observer of these features. And so I think the most important thing is just feeling confident that we’ve made those care providers aware of these things to be watching for.

I think there are sometimes problems with communication and I think one of the biggest problems that happens is that when you look at some of the studies, there is somewhat of a disconnect from what parents report compared to what researchers seem to report in their ultimate conclusions. And I think parents are very frustrated by that. They think providers are discounting their observations.

I think that’s not true. I think providers are trying again to put a context around those observations. And I think the behavioral professionals are awfully good about saying, “Wow, I lost this kid for half an hour and why I lost him was he seemed to have to go across the room and play on the therapy ball for a period of time and I just could not re-engage that child”. Those kinds of examples are the kids that we say, “Wait a minute, come back. Let’s think about that child. I really want to do a good bowel history. I want to think about whether he’s constipated. I want to think about whether he’s got something else going on.” And that’s information that’s going to come to us from the therapist.

We’re lucky in our program because our therapists are actually, as a part of our clinic, participating in the care of the same kid. And so I think we’re uniquely lucky to have really observant therapists. And they’ll walk a child right over to us and say, “Hey, I really want you to think about this problem for this child.”

BHC: Dr. Buie, anything we haven’t touched on here so far that you’d like to mention in closing?

TB: I hope we’ve covered it and I think that I would strongly ask for families to read this paper, because, in a way, I think it’s our intention that it becomes a bill of rights for children with autism. And that at a very minimum we have to think about how we pursue medical evaluations, not only for gastrointestinal problems, but for all medical problems.

We’ve really tried to say, “Here’s how we should care for children.” And I hope that if we don’t make conclusions that families agree with, they understand that most of the things that we couldn’t make conclusions about, we made conclusions about because there was insufficient information. Most of the data out there is not excluding the possibility of a dietary intervention being helpful, etc. It was because we could not find enough evidence at this point, and all through the paper we say, “We need more help. We need more qualified information before we can make some of these conclusions.”

Comments (0)

Subscribe to this comment's feed