Year	Author	Study Subjects	Controls	Type of Intervention	Study Design	Outcome Measures	Results
1988	Gillberg	29 children with autism in the age range 2 - 13 years.	8 normal children, 3 children with cerebral palsy and 1 with Hallevorden-Spatz disease	Analysis of endophins in cerebrospinal fluid	Controlled study	Levels of endorphins	Produced evidence of elevated levels of endorphin like substances in cerebro-spinal fluid of some people with autism. In particular elevated levels appeared in those who appeared to feel pain less than normal and who exhibited self-injurious behaviour. p < 0.05 endorphin fraction I, p < 0.01 endorphin fraction II
1990	Knivsberg et al	Fifteen autistic people aged 6 to 22.	None	Gluten free casein free diet for one year.	Single blind study	Behaviour measured on Haracopos and Kelstrup scale and urine samples were taken. Information on autistic behaviour was taken from parents before diet and after one year.	Significant reduction of peptide levels (p < 0.001). Significant increase in language (p < 0.005), social interaction (p < 0.005), structural ability (p < 0.005) and sensory motor ability (p < 0.005).There was a reduction in epileptic seizures for four children. After four years there was still a significant increase p < 0.005) in linguistic skills. Children who stopped diet after four years had substantial reductions in test scores.
1990	Shattock et al	25 adults with autism	20 adults	Early morning urine samples taken	Controlled study	Measured peptides in urine.	Those with autism showed different levels of peptides, but not significantly.
1991	Sponheim	Seven children with Asperger's Syndrome and five children with autism and mild mental retardation.	Six mentally retarded children.	One had gluten and casein free diet and the other six had gluten or casein reduced diets followed by a diet with gluten for at least three months.	Open study	Visual Analogue Scale and Real Life Rating Scale	Groups were of different ages, received different interventions, no separate controls. No improvement noted in six, but the other noticed considerable improvement.
1995	Lucarelli et al	36 autistic children aged 8 to 13.	20 non-ASD children.	Children had a casein free diet but were given casein or a placebo in capsules for 8 weeks.	Double blind, no randomisation	Behaviour Summarized Evaluation, Student's t-test and blood tests	Significantly (p < 0.001) higher levels of IgA specific antibodies to casein, lactalbumin, B-lactoglobulin and ovalbumin and high levels of IgM and IgG for casein and IgM for lactalbumin. Significant (p < 0.05) improvements in autistic isolation, verbal communication, emotional responses, concentration, perception and cognition.
1998	Knivsberg et al	Twenty children	Yes	Gluten and casein free diet for one year	Single blind study	Tests and interviews	Significant reduction in autistic traits in diet group (mean of 12.5 to mean of 5.6) but no p numbers given.
1999	Cade et al	128 boys with autism (age 3.5 - 16 years) and 22 girls with autism (age 3.5 - 14 years).	83 men with schizophrenia and 37 women with schizophrenia.	Gluten free casein free diet for one year.	Controlled study	IgA in blood. Parental, physician and teachers' assessment.	Significant improvement in social isolation (p < 0.001), eye control (p < 0.001), mutism (p < 0.001), learning skills (p < 0.001), hyperactivity (p < 0.001), stereotypical activities (p < 0.001), hygiene (p < 0.001), panic attacks (p < 0.001), self motivation (p < 0.001).
1999	Whiteley et al	22 children with autism.	Controls did not have autistic spectrum disorder	A non-randomised controlled trial of gluten free diet for at least five months.	Non-randomised controlled trial.	Behaviour was measured on Behaviour Summarized Evaluation scale with cognitive ability measured on six sub tests from Kaufman's Assessment Battery for Children. Urine analyses were done.	Improvements were noted in social and communication skills, motor ability, attention, perception and cognition. One child had a major epileptic fit when gluten was reintroduced.
1999	Whiteley et al	31 children with autism (aged 29 - 178 months)	Six children with autism with no intervention	Gluten free diet	Controlled study	Behaviour was measured on Behaviour Summarized Evaluation scale with cognitive ability measured on six sub tests from Kaufman's Assessment Battery for Children. Urine analyses were done.	Signficant increase in scores on four out of six subtests. No figures given.
2002	Cornish	8 children with autism on gluten free casein free diet	29 children with autism not on gluten free casein free diet	Detailed dietary information and a food diary were taken to ascertain whether levels of nutrients were adequate	Non random postal survey	Postal survey	Nutrient intakes fell below the Lower Reference Nutrient Intake in 12 children (32%) for zinc, calcium, iron, vitamin A, vitamin B12 and riboflavin in the non diet group and four children (32%) for zinc and calcium in the diet group. Fruit and vegetable intakes were higher and cereal, bread and potato consumption were lower in those children using the diet. Also, the families of those on the diet spent a lot more money and time on food and they felt socially isolated.
2002	Knivsberg et al	10 children with autism	10 children with autism	Gluten free casein free diet for one year.	Single blind, randomised trial	Autistic traits (measured by DIPAB), Cognitive skills (measured by Leiter International Performance Scale), Linguistic ability (measured by ITPA and Reynells Spraktest, Motor Ability (measured by Movement Assessment Batterry for Children).	Significant reduction of resistance to communication (p < 0.004). Significant reduction of social isolation (p < 0.005). Significant reduction in strange behaviour (p < 0.007). Significant decrease in autistic behaviour (p < 0.005).
2002	Vojdani et al	40 children with autism aged 3 - 12.	40 healthy chidren.	Using ELISA assays, sera was analysed for the presence of IgG, IgM and IgA antibodies against gliadin and cerebellar peptides.	Controlled study	Levels of antibodies.	For all 12 antigens, the differences between mean+- SD of control sera and autistic children's sera were highly significant (p < 0.001). This suggests that bacterial infections and milk antigens may modulate autoimmune responses in autism.
2003	Arnold et al	10 children with autism on a gluten and casein restricted diet.	26 children with developmental delays, 26 children with autism on a normal diet.	Blood samples were taken	Controlled study	Levels of amino acids in blood were measured.	Findings suggest that children with autism are at high risk for amino acid deficiencies, (p < 0.001). Those on a GFCF diet were more likely than those on normal diet to have a deficiency.
2004	Millward C et al	All reports, studies etc on the gluten free casein free diet and autism	N/A	Research	N/A	Survey	Only one, Knivsberg et al 2002, fitted reliable criteria.
2004	Vojdani et al	50 autistic children aged 3 - 14.	50 healthy people	Assessed the reactivity of sera to specific peptides from gliadin and the cerebellum.	Controlled study	Assessed the reactivity of sera to specific peptides from gliadin and the cerebellum	A significant (p < 0.003) percentage of autism patients showed elevations in antibodies against the dietary protein, gliadin, and cerebellar peptides simultaneously.
2005	Rimland and Edelson	1446 children and adults	Autistic children using other interventions	Children already on a gluten free casein free diet.	Survey	Parents' opinions	3% of the children got worse on this diet, 32% found it had no effect and 65% improved.
2006	Elder et al	15 children aged 2 - 16 years with autistic spectrum disorder.	Yes	Children were put on gluten free casein free diet for 12 weeks.	Randomized, double blind repeated measures crossover design.	Autistic symptoms were recorded by parents based on the Childhood Autism Rating Scale and Autism Diagnostic Interview and urinary peptide levels were collected.	There were no statistically significant findings but several parents noticed improvements.