Background: Secretin is a gastro-intestinal hormone which has been presented as an effective treatment for autism based on anecdotal evidence.

Objectives: To determine if intravenous secretin:1. improves the core features of autism (social interaction, communication and behaviour problems); 2. improves the non-core aspects of behaviour or function such as self injurious behaviour;3. improves the quality of life of affected individuals and their carers; 4. has short term and long term effects on outcome; 5. causes harm.

Search strategy: Results of electronic searches of CENTRAL, MEDLINE, EMBASE, PsycINFO, CINAHL, ERIC, HealthStar and Sociofile (1998 - March 2005) were independently examined by two authors. Reference lists of trials and reviews were searched; experts and trialists were contacted to find unpublished studies.

Selection criteria: Randomised controlled trials of intravenous secretin comparing secretin with a placebo treatment in children or adults diagnosed with autism spectrum disorders, where at least one standardised outcome measure was reported.

Data collection and analysis: Fourteen studies met inclusion criteria. All outcome data were continuous. Where trials used cross-over designs, analysis was conducted on results from first treatment phase, allowing combined analysis with parallel design trials. Where standardised assessment tools generated scores as outcome measures, comparisons were made between means of these scores. Where baseline means were reported, differences between treatment and control were determined to assess possible bias. Where mean change from baseline was reported, this was used in preference to post-treatment scores for meta-analyses or forest plots. As meta-analysis was possible for only one outcome (Childhood Autism Rating Scale), it was impossible to use sensitivity or subgroup analyses to assess impact of study quality, clinical differences in the intervention, or clinically relevant differences between groups, such as age or presence of gastrointestinal symptoms.

Main results: Twenty-five established standardised outcome measures were reported to assess core features of autism, communication, behaviour, visio-spatial skills, affect and adverse events within fourteen included studies. No more than four studies used any one outcome measure similarly. Outcomes were reported between three and six weeks. RCTs of efficacy of secretin in autism have not shown improvements for core features of autism.

Authors' conclusions: There is no evidence that single or multiple dose intravenous secretin is effective and as such it should not currently be recommended or administered as a treatment for autism. Further experimental assessment of secretin's effectiveness for autism can only be justified if methodological problems of existing research can be overcome.