Gluten- and casein-free diets and autism: the Hyman results (at last)

"Although these findings must be interpreted with caution because of the small sample size, the study does not provide evidence to support general use of the GFCF [gluten-free/casein-free] diet."

So said the results of the study finally published by Susan Hyman and colleagues [1] detailing the effects (or not) of a small (n=14) "double-blind, placebo-controlled challenge study" of the use of a diet devoid of gluten and casein for young children diagnosed with an autism spectrum disorder (ASD). If you really want some background history to this often controversial area of autism research, look no further than some of my past musings on this blog (see here and see here) or if you wish, in peer-reviewed form [2].

I say 'finally' in that previous sentence about publication because there has been a considerable degree of waiting for these results to appear in complete peer-reviewed form given that the trial was initially registered in 2004 and things were all supposed to have been wrapped up in 2009 (see here for the ClinicalTrials.gov entry). Some people with their eyes and ears to the autism research grapevine will have probably heard about some of the whys and wherefores of the delay in publishing these results, but I'm not going to get too involved in that here.

Unfortunately the paper is not open-access at the present time but I'll give you a summary of some of the mechanics and findings:

• Some 66 children were initially assessed for eligibility. This was whittled down to 22 kids taking into account those who declined to participate and those who "did not meet inclusion criteria." That inclusion criteria by the way "required children to be enrolled in a comprehensive applied behavior analysis (ABA) intervention program from one of two community agencies" as well as excluding those where seizures were part of clinical presentation and/or the "presence of a chronic illness in addition to ASD that required medical management, celiac disease, documented food allergy to wheat or milk, nutritional compromise such as iron deficiency that required treatment, and family inability to complete rating scales and assessments in English."
• The study design was interesting. It included an implementation phase whereby a GFCF was put in place over the course of 2 weeks (baseline) and maintained for at least 4 weeks. Then came the challenge phase which consisted of weekly challenges to the diet for 12 weeks including one of the following: "foods that contained gluten only, casein only, both gluten and casein, or neither (placebo)." Finally, there was a maintenance period where families were free to "maintain, modify, or abandon the GFCF diet in this phase."
• Various assessments were carried out throughout the study phases covering areas of "physiologic functioning, challenging behaviors (not specific to ASD), and behaviors associated with ASD."
• Results: well, data for 14 of the 22 children were analysed as a function of attrition and or other factors such as "laboratory exclusion criteria." First and foremost we are told: "No serious adverse events were reported during the trial." First, do no harm and all that. When looking at sleep quality and quantity, stool frequency and type, a measure of ADHD (attention-deficit hyperactivity disorder) and a measure of behaviours associated with autism (the Ritvo-Freeman Real Life Rating Scales), authors reported no significant effect following dietary challenges. In other words: "experimental challenges [to the GFCF diet] were not reliably associated with more frequent ASD behaviors." That being said: "All of the families elected to continue the diet for the 12 weeks after completion of the challenges."
• There are some significant strengths to this data based on close monitoring of adherence to the diet, controlling the consumption of gluten and casein levels in challenge snacks and the use of ABA in terms of "stable, consistent educational and behavioral services."
• Likewise however, there are some notable issues associated with the study and the findings outside of the small participant group, not least the emphasis on 'dietary challenge' and importantly the fact that researchers "excluded children who had known gastrointestinal disorders, who might have been more likely to respond positively to dietary restriction." This last point ties in well with other literature in this area [3]. I might add that "individualized supplementation was added for a few participants when deemed necessary by the study dietitian to address low intake of iron, calcium, or vitamin D." Interesting (see here).
• The authors conclude that their study "does not provide evidence to support general use of the GFCF diet" with caveats. 

I can imagine that the Hyman results are probably going to generate some interesting discussions depending on your view of a GFCF diet for autism. I have a professional interest in this topic given some of my research history in this small part of the autism research arena [4] but have tried to stay as objective as possible as per other entries on this topic (see here).

Despite any potential bias I might have, I do still think there is more to see in this area of diet and autism. I've talked before about the idea that the diagnosis of autism is by no means protective against other conditions/labels appearing including those related to issues with gluten for example (see here). This similarly applied to milk also (see here). Some of other peer-reviewed research that I've also been a part of has hinted that there may be 'best responders' to this type of intervention [5] (see here for more discussion); something which ties in well with the concept of plurality and autism (see here).

That caveat about the Hyman study excluding children with known gastrointestinal (GI) disease is also worth re-iterating. Again, this takes autism research into some controversial areas (see here) but as per recent data, both functional (see here) and pathological bowel disorders (see here) do seem to over-represented in cases of autism, so one might see this as an area ripe for further dietary investigations. Indeed, one assumes we might see if there is anything in such an association as and when the Harland Winter trial sees the peer-reviewed light of day (hopefully quite soon).

----------

[1] Hyman SL. et al. The Gluten-Free/Casein-Free Diet: A Double-Blind Challenge Trial in Children with Autism. Journal of Autism and Developmental Disorders. 2015. Sept 5.

[2] Whiteley P. et al. Gluten- and casein-free dietary intervention for autism spectrum conditions. Front Hum Neurosci. 2013 Jan 4;6:344.

[3] Genuis SJ. & Bouchard TP. Celiac disease presenting as autism. J Child Neurol. 2010 Jan;25(1):114-9.

[4] Whiteley P. et al. The ScanBrit randomised, controlled, single-blind study of a gluten- and casein-free dietary intervention for children with autism spectrum disorders. Nutr Neurosci. 2010 Apr;13(2):87-100.

[5] Pedersen L. et al. Data mining the ScanBrit study of a gluten- and casein-free dietary intervention for children with autism spectrum disorders: behavioural and psychometric measures of dietary response. Nutr Neurosci. 2014 Sep;17(5):207-13.

----------

Hyman, S., Stewart, P., Foley, J., Cain, U., Peck, R., Morris, D., Wang, H., & Smith, T. (2015). The Gluten-Free/Casein-Free Diet: A Double-Blind Challenge Trial in Children with Autism Journal of Autism and Developmental Disorders DOI: 10.1007/s10803-015-2564-9