Design: Randomized, double-blind repeated measures crossover design

Population, Sample, Setting, Years: The sample included 13 children from the population of children with autism (2-16 years). Data were collected in subjects' homes.

Variables: The independent variable was the GFCF diet. We evaluated the effects of this diet on the: (a) severity of autistic symptoms as measured by the Childhood Autism Rating Scale (CARS), Ecological Communication Orientation Scale (ECOS), and direct behavioral observation frequencies of child social and language behaviors, and (b) urinary peptide levels of gluten and casein.

Methods: Each child was videotaped during in-home play sessions with the primary caregiver for 15 minutes before the diet's introduction, at the end of the first 6 week period, and at the completion of the 12-week protocol. Data were entered onto lap-top computers and analyzed using an observational software package.

Findings: Initial analysis indicated that while some children demonstrated improvements in social relatedness and language, others showed minimal positive effects. Also, parental anecdotal reports of improvement were not always consistent with data. Currently, a secondary analysis is underway examining differences between children who responded to the GFCF diet with those who did not. Results of both analyses will be presented.

Conclusions: This research is the first reported double-blind clinical trial of the GFCF diet in children with autism. Although preliminary, this study demonstrates how a GFCF diet controlled clinical trial can be conducted and answers important questions related to social validity and practical demands.

Implications: Nurses are often asked to assist families in making decisions regarding alternative treatments. Results of this study provide important information for advising families.

Back to Child Psychiatry
Back to 15th International Nursing Research Congress
Sigma Theta Tau International
July 22-24, 2004