From Our 2010 Archives
'Autism Diet' May Not Improve Symptoms
Study Casts Doubt on Effectiveness of Casein-Free and Gluten-Free Diets
Reviewed By Laura J. Martin, MD
May 19, 2010 --The so-called autism diet -- a gluten-free, casein-free eating plan -- does not appear to improve the symptoms of children with the group of neurodevelopmental disorders known as autism spectrum disorder or ASD, according to a new study.
About one in 110 U.S. children has ASD, which includes classic autism as well as Asperger's syndrome and other forms marked by difficulties in social interaction and communication. Bowel problems are also common in ASD children.
The use of the so-called autism diet has become popular, with up to 27% of parents reporting its use and anecdotal reports praising it. It eliminates the proteins gluten and casein.
But in the small group of children studied, "we did not see a demonstrable improvement," says the study researcher Susan Hyman, MD, associate professor of pediatrics, Golisano Children's Hospital at the University of Rochester Medical Center in Rochester, N.Y. She is due to present the findings Saturday at the International Meeting for Autism Research in Philadelphia.
The new study results follow another report, published in the summer edition of Research in Autism Spectrum Disorders, concluding that 14 published studies of the gluten-free, casein-free diet did not find it useful.
Putting the Autism Diet to the Test
Hyman and her colleagues enrolled 22 children with ASD, all between 2 and 1/2 and 5 and 1/2 years old. After dropouts, 14 finished the 18-week study. None of these 14 participants had wheat or milk allergies, celiac disease (in which the small intestine is damaged from eating gluten), or iron deficiency.
All the children participated in at least 10 hours a week of early intensive behavior intervention to make the group as similar as possible. All children were put on a strict gluten-free, casein-free diet.
The autism diet has become popular based on a theory that some children have insufficient enzyme activity in their gastrointestinal tracts, resulting in incomplete digestion of casein, a protein found in milk and other dairy products, and gluten, a protein found in wheat, barley, and other grains. The incomplete breakdown is what leads to the symptoms, advocates of the dietary approach contend.
After being on the autism diet for at least four weeks, the children were given a "challenge" snack once a week with either 20 grams of wheat flour, 20 grams of evaporated milk, both, or neither. The routine continued until each child received each snack three times over 12 weeks.
All the snacks were made to look identical, with similar taste and texture, and no one knew which snack was being given.
Parents, teachers, and researchers observed the children's behavior and symptoms before the challenges and two and 24 hours after. "We looked at various parameters of behavior before and after the challenges," Hyman says.
Included were measures of improvements in sleep problems, common in children with ASD, improvement in bowel problems, also common, as well as improvement in socializing and language.
"There was no difference with the challenge compared to the placebo," Hyman says
Anecdotal reports from parents, especially in children with autism and significant GI symptoms, may be fueling interest in the diet, Hyman says.
Despite the results, nutritional interventions and effects on autism warrant more study, Hyman says.
One of her co-researchers, Patricia Stewart, PhD, RD, a senior instructor of pediatrics at the University of Rochester Medical Center, agrees. "I think we need bigger studies that are more inclusive," she says, perhaps including children with GI issues. "There could be subpopulations that benefit."
One plus of the study is that the scientific community is focusing on whether the diet plan, long popular with parents, works, says David Mandell, ScD, associate director of the Center for Autism Research at The Children's Hospital of Philadelphia and associate professor of psychiatry and pediatrics at the University of Pennsylvania School of Medicine, Philadelphia.
"Parents have been trying what we call complementary and alternative treatment for a long time, and I think the scientific community has tended to ignore them," say Mandell, who reviewed the study results for WebMD.
Critics might point out that an 18-week study is not long enough to produce real change, he says. "Some parents would say six months to a year [is needed to notice changes]."
"This [study] is a very specific type of study, a challenge," he says. "That's very different than randomizing them to a gluten-free, casein-free diet compared to a regular diet." Such a comparison study would help, he says.
Mandell still expects parents to try the autism diet, despite the results saying it's ineffective.
If they do, he has this advice. "Approach it with even more skepticism than they would have before." Deficiencies in calcium and vitamin D can occur with the diet.
Mandell says parents should decide what specific symptoms or behavior they are trying to change with the diet and develop a way to assess that change once the diet is introduced.
That may reduce the subjective assessment that's a natural reaction to many interventions for autism, he says. "We all want to believe our kids are getting better based on what we do."
SOURCES: International Meeting for Autism Research, Philadelphia, May 20-22, 2010.