When a Mom Says Something Works: The GFCF Diet



My six-year old daughter Joy loves Baby Einstein videos, and has found them mesmerizing since infancy. I’ve heard them so often that I practically know them by heart, including the promotional material at the end of the VHS tapes. In one of the self-advertising sequences, Julie Aigner-Clark, creator of Baby Einstein, is heard to exclaim, “As moms, we’re all looking for help … and if a mom tells you, ‘Try this, it works,’ you automatically try it if you’re a mom!”

She wasn’t talking about alternative therapies for autism. But as the mother of a child on the autism spectrum, I hear the echoes.

One place I heard such reverberations was in a Time magazine article on Jenny McCarthy and autism1, in the March 8, 2010 issue. Actress and former Playboy Playmate Jenny McCarthy, whose son was diagnosed with autism in 2005, has become the celebrity-mom face of the unproven-therapies-for-autism movement. The article’s author somewhat wryly notes, “…it is hard to find a controversial, novel, or alternative treatment that McCarthy doesn’t say has some merit.” The article sums up McCarthy’s autism treatment advice in these words: “‘Try everything,’ says McCarthy. ‘Hope is the only thing that will get us up in the morning.'”

The second part of McCarthy’s statement resonates with me. Not the first part.

I had an extraordinary opportunity in early 2010 to examine controversial “try-everything” therapies, as part of the MCH-LEND program, which stands for Maternal and Child Health – Leadership Education in Neurodevelopmental Disabilities2. This interdisciplinary leadership training program at select universities offers graduate students and family members the chance to engage with clinical and advocacy and public health issues around developmental disability, and to develop leadership skills to bring to bear in their professions and communities. One of our LEND seminar assignments involved small-group presentations on alternative therapies, for autism and beyond.

My group consisted of a nutritionist/mom, a psychology grad student, and me (librarian/mom). We were assigned to examine the gluten-free, casein-free (GFCF) diet for autism. The point of the assignment was to become familiar with the claims of the intervention, and learn how to evaluate non-standard therapies in general. We were to examine a substantial list of questions for our presentation, including:

  • What exactly is the claim made by this intervention? For whom is it intended?
  • Is the claim consistent with current knowledge, i.e., does the rationale for its claimed effectiveness make sense in terms of already-existing scientific knowledge?
  • What is the evidence base for this intervention?
  • Have results been described in peer-reviewed journals, or only the lay press?
  • Are there any known risks for using this therapy?
  • Are there position statements regarding this therapy, issued by any professional organizations?
  • Are there conflicts of interest for those evaluating the claims?
  • How widely used is the intervention?
  • How much does it cost in time, energy, and money, for families?
  • What is the attraction for families to try this therapy?3

Here’s what we found: The gluten-free, casein-free diet is an elimination diet. To be eliminated from the menu are gluten, a protein found in grains (wheat, barley, rye, and some oats) and casein, a protein found in animal milk. Elimination diets have been around for a long time, and in general they rest on a sound premise: if you are sensitive or allergic to something in your diet, stop eating it! The gluten-free diet is the one best approach for celiac disease; if you have a demonstrated sensitivity or allergy to milk, dairy-free is the way to go. It’s not an easy regimen to follow, but in those situations, it’s known to be effective.

The “alternative” part of this therapy is the claim that GFCF can cure or ameliorate autism spectrum disorder (ASD) — in up to 90% of cases, according to the organization Talk About Curing Autism (TACA). The controversial underlying theory is this: People with ASDs are said have a predisposition to a gastrointestinal condition known as “leaky gut” or “autistic enterocolitis.” The condition is supposedly triggered by an environmental insult — some say vaccines, some suspect other sources of toxicity. When the condition is triggered, the hypothesis goes, you wind up with an incomplete breakdown of gluten and casein which escape through the leaky gut. The protein molecules/polypeptides travel through the bloodstream and attach to opiate receptors in the brain, leading to autism or magnifying the symptoms of autism.

According to this scenario, if you remove the gluten and the casein from the diet, the gut will heal, the opiate reaction is halted, and the autistic symptoms are alleviated — generally in a couple of months of strict adherence to the diet, sometimes taking up to a year.

GFCF is not an easy diet to follow, especially since people on the autism spectrum may already be eating a self-restricted range of foods. The diet requires constant vigilance on the part of parents, such as careful ingredient checking and diet balancing and making sure that gluten and casein don’t slip in from unexpected sources: a playmate’s cracker snagged from the floor, a bit of standard Play-doh, etc. Pre-fab GFCF foods can be very expensive. If you want to do the diet less expensively, you’ll spend a lot of time in preparing foods from scratch. Recipe-sharing online can ease that burden; there’s even a webpage from TACA that outlines how to do the GFCF diet on food-stamps.

There have been related scientific studies from a number of angles: measuring opioid peptides in the urine, measuring intestinal permeability, studying the behavioral outcomes of people with ASD who go on the diet. In all of those areas, research is relatively thin on the ground, and the results are not definitive. There are even conflicting studies as to whether people with autism have more gastrointestinal (GI) issues in general than people without, though we do know that GI issues that occur in typically-developing folks do occur in folks with autism as well. One 2006 study found GI issues in 70% of their study group with autism, much higher than in their non-autistic comparison group4; another published in 2009 found little difference other than in GI issues that could be caused by behavior (constipation and food selectivity)5.

In a happy circumstance for our presentation, a consensus paper6 came out in Pediatrics, a high-profile peer-reviewed journal, right at the start of the semester. A whole slew of GI experts had gotten together, weighed existing evidence, made consensus statements regarding GI issues and ASD, and recommended a lot of new research. Here are several of the consensus statements most pertinent to GFCF (there were 23 statements in all):

  • Individuals with ASDs who present with gastrointestinal symptoms warrant a thorough evaluation, as would be undertaken for individuals without ASDs who have the same symptoms or signs.
  • The prevalence of gastrointestinal abnormalities in individuals with ASDs is incompletely understood.
  • The existence of a gastrointestinal disturbance specific to persons with ASDs (eg, “autistic enterocolitis”) has not been established.
  • Available research data do not support the use of a casein-free diet, a gluten-free diet, or combined gluten-free, casein-free (GFCF) diet as a primary treatment for individuals with ASDs.

Interestingly, this consensus paper received some acclaim from both skeptics and proponents of the GFCF diet as a treatment for autism. Skeptics said (and I paraphrase immensely), “Look, here it is in black and white — current research data do not recommend this treatment!” Proponents said, “Look at all this research they’re recommending — we’re finally being taken seriously!”

This kind of recommended research does not happen overnight. One pilot study of the GFCF diet for autism at the University of Rochester, approved as a clinical trial since 2004, just announced results in the spring of 20107. A group of children with autism were put on a carefully-controlled GFCF diet. After four weeks, half of the group received “challenge snacks” containing gluten and casein, while the other half received GFCF snacks. Neither the families nor the evaluators knew which children received which snacks. After three sets of challenge snacks across 12 weeks, the results showed no difference between the challenge-snack children and those who remained strictly GFCF. In the words of Susan Hyman, the lead researcher on the study, “It would have been wonderful for children with autism and their families if we found that the GFCF diet could really help, but this small study didn’t show significant benefits.”

A previous pilot study published in 20068 took a different approach, putting only half the participants on the diet but providing all foods for both groups, such that the participants and parents did not know whether or not they were GFCF. The study did not find any significant effects on either behavior or urinary peptide levels. However — and here’s the really interesting bit — the parental reports and the researcher reports differed markedly. Parents thought they were seeing improvements that the researchers could not document. And nine of the families decided to keep going with the GFCF diet even when they were told that the researchers’ observations didn’t support it.

To me, this screams “placebo effect.” People see what they want to see. No wonder the claims of “try this, it works!” are so high. I’m sure the parents in the study truly believed that they were seeing benefits that the researchers didn’t catch. But that’s one of the core reasons that we need science to help us distinguish between real effects and rose-colored glasses. As physicist Richard Feynman described the scientific endeavor, “The first principle is that you must not fool yourself — and you are the easiest person to fool.”

My husband and I have not tried the GFCF diet with our Baby-Einstein-loving daughter, because we have not seen anything that would make us suspect a dietary sensitivity if she weren’t on the autism spectrum.

Many, many families are trying it, however. This therapy seems extremely attractive to parents, and I think there are a number of reasons. There are powerful testimonies out there, and high-profile proponents, from Jenny McCarthy to the autistic author and animal scientist Temple Grandin (who takes a much more nuanced approach than McCarthy). The diet has a reputation of harmlessness, though it can be nutritionally risky if not done carefully, especially for people whose diet is self-limited in the first place. There’s definitely a “why not try it, what can it hurt?” vibe out there.

Besides all that, dietary changes are a familiar kind of intervention. Who hasn’t gone on a diet at some point or another? And it’s something that’s possible to attempt without a prescription. There’s plenty of do-it-yourself GFCF advice available, on the web and in books and in support groups.

There is value in hope. There is value in parents feeling empowered and feeling that they are helping their child. And the GFCF diet obviously works for some people with autism, as it works for some people without. Some people with ASDs do have dietary sensitivities, just like their typically-developing counterparts. When you address those sensitivities, people feel better. When people feel better, they behave better, learn better, and interact better.

I’m open to the thought that there may be more food sensitivities than we currently understand. But the state of the art in 2010 is clearly presented both in the Pediatrics consensus paper and in a new systematic review in the July 2010 issue of Research in Autism Spectrum Disorders9: right now there is not enough evidence to recommend the GFCF diet as a primary treatment for autism, and there’s no current proof of the existence of autistic enterocolitis. My personal prediction is that the evidence against the leaky-gut/autistic enterocolitis hypothesis is going to start to pile up. Autistic enterocolitis as a concept is under a shadow already, in that its originator Andrew Wakefield has been de-licensed by the U.K. General Medical Council, and his 1998 paper10 that fueled much of the speculation has been retracted by the Lancet.

The outsized role of parental enthusiasm in promoting the GFCF diet, as with other non-evidence-based interventions for autism, makes for unfortunate peer pressure situations. Parents are being made to feel guilty if they agree with the current scientific consensus and choose not to try the diet for their children with autism. Or, if they try the diet and don’t see any benefits and quit, they may well be told that they didn’t try long enough or that they must not have been doing it right. It can make for a lot of unnecessary self-doubt and hard feelings.

A parent’s recommendation may be enough to cause one to “automatically try” a new toothpaste or a twenty-dollar Baby Einstein DVD. When it comes to autism treatments, though, there are many more questions to ask. The list from the LEND assignment is one fine place to start. As a parent, I found it empowering to have access to the questions being used to educate the up-and-coming professionals! If conflicting claims online feel too overwhelming, consider taking your questions to the professionals in your child’s life.

When a mom says something works for her child with autism, congratulate her — but don’t let your search stop there, as you work to sort out the snake oil from the truly useful, and make your well-informed decisions on what is really worth trying.


  1. Greenfield, Karl Taro, The Autism Debate: Who’s Afraid of Jenny McCarthy? Time 8 March 2010; 175(9):40.


  2. http://www.aucd.org/template/page.cfm?id=473
  3. Assignment questions courtesy of MCH-LEND, University of Wisconsin-Madison, Waisman Center. Used with permission.
  4. Valicenti-McDermott M et al., Frequency of gastrointestinal symptoms in children with autistic spectrum disorders and association with family history of autoimmune disease. J Dev Behav Pediatr 2006;27(2 suppl):S128 –S136.


  5. Ibrahim S et al., Incidence of Gastrointestinal Symptoms in Children with Autism: A Population-Based Study. Pediatrics 2009;124;680-6.


  6. Buie T et al., Evaluation, Diagnosis, and Treatment of Gastrointestinal Disorders in Individuals With ASDs: A Consensus Report, Pediatrics 2010;125;S1-S18.


  7. Hyman S et al., The Gluten Free and Casein Free (GFCF) Diet: A Double Blind, Placebo Controlled Challenge Study, International Meeting for Autism Research, Philadelphia, PA, May 22, 2010.


  8. Elder JH et al., The gluten-free, casein-free diet in autism: results of a preliminary double blind clinical trial, J Autism Dev Disord 2006 Apr;36(3):413-20.


  9. Mulloy A et al., Gluten-free and casein-free diets in the treatment of autism spectrum disorders: A systematic review. Res Autism Spectr Disord 2010 Jul; 4(3):328-39.


  10. Wakefield AJ et al., Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children, Lancet 1998 Feb 28;351(9103):637-41. Retraction in Lancet 2010 Feb 6;375(9713):445.


A version of this essay appeared on elvis-sightings.blogspot.com in February 2010.