Dietary interventions for developmental and learning disabilities have been controversial for decades. When a child presents with a metabolic disorder, it often follows that diet may be a critically important variable. Certain metabolic disorders, like phenylketonuria, are effectively managed by diet. Feingold (1975) first proposed that eliminating certain substances like preservatives would result in a child with attentional problems, becoming more focused and less hyperactive. These claims did not turn out to be valid for 95% of the children treated with diet (NIH, 1996) and those who were helped tended to have identifiable food allergies. Evidence suggests that ADHD is a neurological condition but it is a complex disorder with little solid information to link its cause to diet. We do know that ADHD is unlikely to be caused by food allergies, excessive sugar intake, or other variables that do not affect an individual's neurological make-up (NIH, 1996).

Although there was no solid evidence that Feingold's approach was effective, this strategy has also been applied to autism and other disabilities. Other diets have been developed with a variety of hypotheses that also lack empirical evidence. However, unless a child has a food allergy/intolerance or metabolic condition it is unlikely that dietary changes will affect their disability. Currently, the most popular diet promoted as a cure for autism involves restricting casein and gluten from the child's diet. The unfounded assumption behind recommending that casein and gluten be restricted is that the processing of these substances causes many of the symptoms of autism. Beside the fact that most of the evidence provided as support for casein/gluten theory of autism comes from individuals involved in selling these products or diet-related advice, the lack of scientific acceptance of casein and gluten processing difficulties as a cause of autism is one good reason to be wary of this approach.

Though children with ASD seem no more likely to have gastrointestinal difficulties (see my previous blog post), including food allergies, than typically developing children, the fact is that both children with ASD and those who are typically developing can have this problem. It is also the case that children are more likely to have food allergies than adults. So food allergies are clearly an appropriate concern. In fact, at times they can involve severe reactions that are life threatening. Food allergies occur because the body develops antibodies specific to the allergen in the offending food. The next time that this substance is ingested, the antibodies then trigger the release of histamine and an allergic reaction occurs. Most children with a significant food allergy appear sickly. They have symptoms that are typical of an allergic reaction, may vomit frequently, and often have abnormal bowel movements. Shortly after the food the child has an allergy to is ingested, symptoms (e.g., diarrhea, tingling/swelling in the mouth area, difficulty breathing, pronounced rashes especially on the face or torso) appear. Most children outgrow their allergies although it is thought that allergies to nuts (and peanuts), fish, and shellfish are likely life-long allergies.

For over 17 years I have worked with children with feeding difficulties with and without ASDs and one frequently encountered cause of a child's feeding problem is food allergy. If a child has a documented food allergy then removing the allergen(s) from the child's diet is one necessary component of resolving the problem. However, determining that the child has a food allergy is a critical first step. If a child is suspected of having an allergy because of symptoms, then there are two tests commonly accepted as valid means of diagnosing food allergy. These are the skin prick and RAST (radioallergosorbent) tests. So long as the allergic reaction is not severe, these tests can then combined with information gathered by alternately exposing or restricting access to the suspected allergen and determining whether symptoms respectively appear and dissipate to confirm test results. For the child with a food allergy, the benefit of removing the allergen will be apparent in that the overt symptoms caused by ingestion of the offending substance will dissipate but if the child has a disability as well, the disability will persist. The child who is no longer in discomfort will also likely be more receptive to instruction and have a more positive affect.

So, why not restrict access to casein and gluten just to make sure you are doing everything you can for your child? First, the lack of scientific evidence that casein and gluten cause autism is concerning. This combined with the evidence that gastrointestinal (GI) difficulties and autism are not linked, should cause you to question the plausibility of dietary intervention for this disorder. Then again, most significant in my mind is the fact that children with ASD are prone to selective eating (Ahearn et al., 2001). During our study of the eating habits of children with an ASD, my colleagues and I provided children with a variety of foods across six meals spread out over at least two weeks. Over half of the children we observed presented with some form of selective eating. Though we saw selectivity for various food groups, the food group that our selective eaters were most likely to prefer was starch. Gluten is a constituent of many starches and restricting access to this substance can lead to eliminating the only foods that a child regularly eats.

The course of action that I have followed as a clinician treating selective eating has been to attempt to expand the variety of foods a child eats. I've worked with children who ate only one food and would not eat unless that food was prepared in a specific manner (e.g., macaroni and cheese made with whole milk and butter served straight out of the pan). Going from such selective acceptance of food to eating other foods often takes quite a long time. The only situations in which I would recommend restricting access to food would be if the child had a documented food allergy or excessively consumed food. Several parents have reported to me that they have tried the casein/gluten-free diet and their child would not eat any of the foods presented to them. To be fair, I've also had a number of parents swear to me that this diet was very helpful but none of them answered yes when I asked them if the diet cured their child of autism. It is my opinion that children with an ASD will not benefit from dietary restrictions of any kind unless they also have a food allergy or intolerance. Furthermore, children who are selective eaters are likely to become more selective and may stop eating when their diet is radically restricted. Because children with ASDs are prone to selective eating it is probably a good idea to attempt to expand their diets rather than restrict them.

 Ahearn, W.H., Castine, T., Nault, K., & Green, G. (2001). An assessment of food acceptance in children with autism or pervasive developmental disorder - not otherwise specified. Journal of Autism and Developmental Disorders, 31, 505-512.

Feingold, B.F. (1975). Why your child is hyperactive. New York: Random House.

National Institutes of Health. (1996). NIH Publication No. 96-3572, printed 1994, reprinted 1996. Booklet. 44p. Retrieved September, 1997 from

For more sound information about food allergies, go to the National Library of Medicine's food allergy site at:

Recent Posts in A Radical Behaviorist

Time to throw out the DSM?

Diagnoses of ASDs are better informed by the ADOS than a clinician's experience.

Another autism tragedy

She killed her children with autism because she wanted "normal kids."

Requests: Verbal or not?

Verbal interaction requires that one be a speaker AND a listener.

Torture or treatment?

Skinner felt punishment was the cause of many ills in society.

Repetitive behavior: To redirect or not to redirect?

Everyone should be allowed to eat donuts and take a nap.

CNVs: The True Cause of Autism?

It seems as if it really isn't the vaccines that cause autism.