The gluten-free, casein-free (GFCF) diet is an alternative intervention that has been extensively used with children with autism spectrum disorders (ASD). The GFCF diet involves the elimination of all foods containing gluten or casein (proteins found in wheat or barley products and dairy products, respectively). The intervention’s popularity is largely due to the ease of implementing it into the child’s (and family’s) day-to-day routine, the fact that it is relatively inexpensive, and the ability to implement it in conjunction with other interventions. However, given the use of gluten in an enormous array of food products as a binding agent, the GFCF diet requires extensive planning and organisation to ensure that the child’s long-term nutritional needs are met.

The two most prominent theories behind the use of the GFCF diet are the ‘opioid excess theory’, and distress due to gastrointestinal symptoms. The opioid excess theory proposes that children with ASD have increased gut membrane permeability, which allows the opioids produced during the metabolism of gluten and casein to pass more easily into the blood stream, leading to an excess of opioids in the brain and resulting in ASD symptoms such as repetitive behaviours, sleep disturbances and hyperactivity. Alternatively, it has been suggested that gastrointestinal symptoms caused by gluten sensitivity or celiac disease may explain the presence of some ASD behaviours – children with ASD often have communication or language difficulties, making it hard for them to express their discomfort verbally, and instead they resort to self-injury or tantrums, and may appear hyperactive. In support of this, children with ASD are reportedly more likely to experience gastrointestinal symptoms, such as bloating, constipation or diarrhea, than siblings and other children without the disorder. Both theories therefore suggest that gluten should be eliminated from the diet of children with ASD to improve cognitive and behavioural symptoms.

Despite the popularity and anecdotal success of this diet, there is limited evidence of its effectiveness (Hurwitz). Research involving the ‘gold standard’ experimental design - double-blind, randomised controlled trials (RCTs) where participants are unaware of whether they have been randomly allocated to either a GFCF diet or a diet containing gluten and casein - is lacking, and the findings of existing studies have been mixed.

In a double-blind RCT by Hyman and colleagues in 2015, 14 children with autism (aged 3-5) were placed on a GFCF diet, and after 6 weeks were given ‘dietary challenges’ (weekly snacks containing either gluten, casein, gluten plus casein, or neither). After 12 weeks of challenges, children were then monitored for an additional 12 weeks. The researchers reported that the dietary challenges did not significantly impact sleep quality, hyperactivity or ASD behaviours. However, the results of this study are limited due to the small number of participants. In a similar study in 2011, Johnson and colleagues found that there were no significant differences in anxious behaviours, sleep quality, attention, affect or ADHD symptoms between children with ASD placed on a 3 month GFCF diet compared to those placed on a 3 month placebo diet. Again, the small number of participants in their study (8 and 14 children in each group, respectively) is a significant limitation.

In a 2013 study by Pennesi and Klein, of 387 parents of children with ASD, 92.7% of children were reported to have gastrointestinal symptoms, and 93.4% of children were reported to have allergy symptoms (e.g. body rash). In contrast to the RCT studies, parents who completely eliminated gluten and/or casein from their child’s diet reported a significantly greater improvement in ASD behaviours (e.g. reduced tantrums, hyperactivity, and self-stimulatory behaviours) and social behaviours (e.g. increased eye contact, social responsiveness, language production, and imaginative play) and physiological symptoms (e.g. body rash, constipation, diarrhea), compared to parents who only partially eliminated gluten and/or casein. These improvements were greater for children who had prior gastrointestinal and/or allergy symptoms, compared to those who did not. However, conclusions drawn from Pennesi and Klein’s study are limited by the fact that the study was based on parental reports. Because of this, issues such as under-reporting and over-reporting of symptoms, as well as potential placebo effects (parents were aware of what their child was consuming) cannot be ruled out. Additionally, it is not clear whether these children were undergoing a behavioural intervention simultaneously.

These findings seem to suggest that GFCF diets may be effective in a subset of ASD children with gastrointestinal symptoms. ASD children diagnosed with celiac disease may also benefit from a GFCF diet. However, due to the limited number of well-controlled studies with adequate numbers of participants, much more research is needed before the GFCF diet can be recommended as an alternative treatment for these children. We advise that you speak to a clinical psychologist about the possibility of implementing this intervention to reduce ASD behaviours. Additionally, it is always important to speak to a nutritionist before changing your or your child’s diet, to ensure that all nutritional needs are met.