← Back to blog

Gluten-free and casein-free diets for autism: what the research says

Gluten-free and casein-free diets for autism: what the research says

NeuroDifferent Research Digest

In one sentence

This Cochrane review found only weak evidence from two very small trials that gluten-free or casein-free diets improve autism-related behaviour, and authors called for larger, high-quality studies.

What the researchers did

Some theories propose that peptides from gluten (found in wheat and related grains) or casein (a milk protein) might affect brain function in autism. Because of this idea, many families try elimination diets that remove gluten, casein, or both.

The reviewers searched medical databases through April 2007 for randomised controlled trials of gluten-free, casein-free, or combined diets in people diagnosed with autism spectrum disorders. They planned to compare diet groups with usual eating patterns on behaviour, cognition, and social functioning.

Only two small randomised trials met the criteria, with a combined total of 35 participants. The studies used different outcome measures, so the reviewers could not pool results in a formal meta-analysis and instead described findings in narrative form.

Note: the Cochrane publication was later marked as withdrawn, but its abstract and conclusions remain available as a summary of the evidence available at the time.

What they found

  • One trial reported statistically significant improvements in overall autistic traits with the diet intervention.
  • The same research group’s work also reported improvements in social isolation and in overall ability to communicate and interact in separate analyses.
  • Ten other outcomes could not be analysed in a standard way because the data were skewed.
  • Three outcomes showed no significant difference between diet and control groups.
  • Neither trial clearly reported harms or disadvantages of the diets, such as nutritional deficiencies or stress around restricted eating — so risks may be under-documented in the published evidence.

Overall, the authors concluded that current evidence for efficacy is poor and that large, good-quality randomised trials are needed.

What this means for families and therapists

Gluten-free and casein-free diets are among the most common complementary approaches families try. Surveys often show high use, but this review suggests the scientific proof base is still thin.

Strict elimination diets can be difficult to maintain, may affect growth and nutrition if not planned carefully, and can increase mealtime stress. If a family chooses to try a diet, working with a paediatrician or dietitian helps monitor adequate calories, protein, and micronutrients.

Some children have separate medical reasons to avoid gluten (such as coeliac disease) or dairy allergy — those decisions should be guided by medical diagnosis, not only by autism-related hopes.

Limitations and what we don't know yet

With only 35 participants across two trials, positive findings may reflect chance, bias, or the specific way outcomes were measured.

Both positive trials came from the same research group, so independent replication is especially important before drawing strong conclusions.

The search end date was 2007 in the version described; families should know newer trials may exist outside this withdrawn review. Even so, major guideline groups have generally not endorsed elimination diets as proven autism treatments.


This is a plain-language summary of Gluten- and casein-free diets for autistic spectrum disorder by Millward C, Ferriter M, Calver S.J et al., Cochrane Database of Systematic Reviews (2019). Source license: CC-BY-NC-4.0.

It is not medical advice — talk to a qualified clinician before changing therapy.

← Back to blog